Skip to main content

Full text of "Research Paper"

See other formats


H" 5 



I 



Wm 

mm 



m 



Digitized by the Internet Archive 

in 2013 



http://archive.org/details/researchpaper17nort 



i ior\A CADCCT CCDV/irc 

r ':/j- - /ye, 

3DA FOREST SERVICE 
ESEARCH PAPER NC-176 



jOVT. documents 
depository item 

IAN ?« 1980 

CLEMSON 
LIBRARY 




Regeneration and productivity of aspen 
grown on repeated short rotations 



Donald A. Perala 



3rth Central Forest Experiment Station 

>rest Service, U.S. Department of Agriculture 






North Central Forest Experiment Station 

Robert A. Hann, Director 

Forest Service - U.S. Department of Agriculture 

1992 Folwell Avenue 

St. Paul, Minnesota 55108 

Manuscript approved for publication May 2, 1979 

1979 



REGENERATION AND PRODUCTIVITY OF ASPEN 
GROWN ON REPEATED SHORT ROTATIONS 



Donald A. Perala, Silviculturist 
Grand Rapids, Minnesota 



Because of the increasing demand for wood fiber, 
considerable research has been devoted to short 
rotation, full tree utilization systems, which are 
designed to produce the maximum amount of wood 
fiber in the shortest time. Quaking aspen {Populus 
tremuloides Michx.) is a likely candidate for these 
systems because it produces good quality wood 
fiber at a young age. However, the effects that 
repeated cropping would have on the aspen's 
coppicing ability have had little attention. There- 
fore, we conducted this study to decribe the produc- 
tivity and other characteristics of repeatedly 
cropped aspen. 

STUDY AREAS 

The study areas are located within 13 km of each 
other on the Chippewa National Forest, Minneso- 
ta. The soils are well-drained Warba sandy to fine 
sandy loams and are considered good aspen sites. 
The climate is continental, averaging 20 C in July 
and 64 cm annual precipitation. The topography is 
level to gently rolling and the elevation is 400 to 
410 m. The areas had been commercially logged 
for aspen during spring and early summer 2, 4, 
and 8 years prior to study installation (table 1). 

Table 1. -Parent stand summary 





Stand 
age 


Aspen yield 


Rotation Site 
schedule (yr) index 1 


Merchantable Total 
volume 2 tree 3 


m 
8 24.4 
4 22.9 
1 21.3 


yr 

42 
31 
47 


m 3 /ha t/ha 
77 31 
55 23 
77 31 



'At 50 years. 

2 To 10.2 cm top; from sale records. 

3 0ven-dry weight, except leaves, above 15 cm stump Estimated from 
Schlaegel (1975) These figures are for merchantable trees and species 
only and probably underestimate total standing crop by at least 50 
percent 



Treatment plots were located where suckers were 
uniformly dense and unimpeded by residual over- 
story. 



METHODS 

All woody stems on five circular 810 m 2 (16 m 
radius) treatment plots were clipped and removed 
on 1 '-, 4-, and 8-year cycles beginning May 1970. 
Two plots were installed in each the 4- and 8-year 
cycles and one plot was installed in the 1-year 
cycle. Stems were clipped during the dormant sea- 
son to maximize aspen suckering (Brinkman and 
Roe 1975). Five circular 16.2 m 2 (2.27 m radius) 
sample plots were systematically clustered at 
the center of each of the five treatment plots. The 
minimum distance between sample plots and 
treatment plot boundary was 8.4 m to exclude edge 
effects on sucker stocking and growth. 

The aspen from the first clipping of the sample 
plots was used to develop stand equations for 
aerial dry weight of total trees (except leaves) and 
of wood based on stem diameter and height mea- 
surements (Perala 1973). Thereafter, these equa- 
tions were applied to yearly diameter and height 
measurements, including the cropping years. - 
Other woody stems from the sample plots were 
weighed fresh by species at each cropping but were 
not measured during intervening years. Several 
stems of each of these other species of estimated 
mean size were subsampled to determine oven-dry 
weights ( 105°C oven) at each cropping. During the 
first 2 years, aspen coppice (ramets) was recorded 
as either root suckers or stump and root collar 
(S/RC) sprouts. 



*This sucker stand was 2 years old at first clip- 
ping but was clipped annually thereafter. 

^Except for the 1-year rotation after the second 
cropping- it became more convenient to dry and 
weigh entire plot clippings. 



Curvilinear multiple regression analysis was 
performed on data from the 25 measurement plots 
to determine how stand characteristics influenced 
the sizes of both aspen S/RC sprouts and suckers 
(dependent variables). 

The independent variables were: 

X,, ortet (parent) age (years); 

X 2 , ortet stem density (1,000's/ha); 

X 3 , ortet mean weight (g/stem); 

X 4 , sucker density (1,000's/ha); 

X 5 , S/RC sprout density (1,000's/ha); and 

X 6 , mean weight of suckers or S/RC sprouts, as 
appropriate, (g/stem). 
All variables in both equations were significant at 
the 0.05 level. 3 



RESULTS AND DISCUSSION 



4 


2 


22.5 


21 


1.73 


.47 


2.45 


2.24 



.71 .21 



Table 2. — Mean annual, increments of short rota- 
tions compared to conventional rotations 

Stand age, years 8 

Site index, m 24 

MAI (wood component) t/ha/yr 1.52 
MAI, conventional, t/ha/yr 1 2.66 
Ratio, short rotation -r- 
conventional .57 

1 From Perala (1977). 

Wood specific gravity was similar to mature stands 
(Schlaegel 1975) but moisture content (mainly be- 
cause of seasonal variation) and bark percent were 
considerably higher and bark specific gravity was 
lower (table 3). Bark percent decreased as stand 
age increased but the other physical properties did 
not vary significantly with age. 



Initial Sucker Stand 
Characteristics 

If expressed as mean annual increment, the 
short rotation yields are much less than would be 
achieved on conventional (40 year) complete tree 
fiber rotations (table 2). 



^Coefficients and other statistics are not given 
here but are available from the author on request. 
Response to each variable was depicted graphically 
while holding all other variables at constant mean 
values. 



Repeated Crop Yields 

Regeneration and yield of aspen decreased as 
cropping frequency increased. Regeneration and 
yield on the 1-year rotation declined steadily 
through the fourth crop and precipitously thereaf- 
ter (fig. 1). The decline in yield through the fourth 
crop was a function of both a reduction in number 
of stems regenerated and mean stem size. After 
the fourth cropping, mean stem size remained es- 
sentially unchanged. After 7 annual croppings, 
aspen regeneration ceased. 



Table 3. — Yield and properties of young aspen sucker stands 













Total 












Stand 


Dominant 


Mean 


Basal 


Number 


fresh 


Oven dry weight 




Specific 


gravity 


age (yr) 


height 


dbh 


area 


of stems 


weight 


Total Wood 


Bark 


Moisture Bark 


Wood 


Bark 




m 


cm 


m 2 /ha 


1,000's/ha 




. ...t/ha .... 




. . .Percent . . . 






8 


6.7 


2.8 


8.99 


14.9 


36.2 


16.3 12.2 


4.18 


121 25.5 


0.391 


0.509 


4 


4.4 


1.4 


6.46 


44.0 


22.4 


9.55 6.91 


2.64 


135 27.6 


0.378 


0.451 


2 


2.3 


— 


— 


73.1 


3.41 


1.50 0.94 


0.56 


128 37.6 


0.386 


0.449 


1 1 


1.3 


— 


— 


75.2 


0.87 


0.40 0.22 


0.18 


120 44.5 


0.415 


2 



'First 1-year cropping of the original age 2 stand. 
2 Not determined. 




Table 4. — Comparison of dominant stems at end 
and beginning of successive 8-year rotations 



ANNUAL CROPPINGS (NO.) 

Figure 1. — Regeneration and oven-dry above- 
ground yield of aspen coppice (except leaves) on 
successive 1-year croppings. The first cropping 
was omitted because it was at age 2. 



The 4-year rotation tended toward the same re- 
sults — number, vigor, and total yield declined 
with each cycle (figs. 2 and 3). However, it would 
take at least several more cycles to completely 
eliminate aspen. 

The 8-year rotation produced two-thirds less 
sprouts and one-third less total yield on the second 
cycle (figs. 2 and 3), but the results are confounded 
by feeding damage by hares (Lepus americanus) 
during the winters following the fifth and sixth 
growing seasons. Hares girdled some of the small- 
er stems, which died a year or two later without 
resprouting. However, even if the hare damage 
had not occurred, the estimated total yield of the 
second cycle was still 10 percent less than the first 
cycle (Perala 1973). Because the largest stems sur- 
vived, the average size of three dominant stems 
was compared at the end of each cycle as well as 
between dominant regenerated stems (table 4). 



End of 8-year 
rotation 



Beginning of 8-year 
rotation 



Cycle 



Mean 
dbh 



Mean 
height 



Mean basal 
diameter 



Mean 
height 





cm 


m 


cm 


m 


1 


4.7 


6.7 


— 


— 


2 


4.6 


6.5 


1.1 


1.52 


3 


— 


— 


1.1 


1.43 



None of these comparisons were statistically dif- 
ferent. Furthermore, of the 10 plots, 5 had larger 
dominants at the end of the first cycle and 5 at the 
end of the second. All in all, the results for the 
8-year rotation do not suggest significant physio- 
logical impairment to aspen productivity. 

Character of Regeneration 

In contrast to mature stands which produce 
suckers almost exclusively, most of the aspen re- 
generation in all rotations were stump and root 
collar sprouts (table 5). 

Table 5. — Stump and rootcollar sprouts regener- 
ated on short rotations 

Stump and root collar sprouts 



Ortet 


Percent by 


Percent by 


Per 


age 


weight 


number 


stump 


8 


58 


57 


3.9 


4 


88 


82 


1.6 


2 


60 


63 


0.6 


1 


58 


59 


0.5 



S/RC sprouts did not differ significantly in growth 
and survival from suckers during the first 2 years 
when this data was gathered. 

S/RC sprouting in aspen has rarely been 
reported before. Maini (1968) reports they are oc- 
casionally found on aspen up to sapling size in 
Canada and Baker (1925) found that S/RC sprouts 
accounted for 9 percent of the sprout regeneration 
after mature Utah aspen were felled. In this pres- 
ent stand, the stumps were cut so low that stump 
and root collar sprouts could not be differentiated. 
The propensity to regenerate S/RC sprouts is prob- 
ably related to the strong tendency for suckers to 



120 



100 



00 

b 



80 



55 eo 

UJ 

Q 

K 40 

CO 



20 



4-YEAR 
ROTATION 



CYCLE 4 



10 






O 
QC 
O 



^ 4 



CD 

Q 

to 




CYCLE 2 
CYCLE 3 



100 


\ 




80 


\ 


8-YEAR 
ROTATION 


60 


_ CYCLE 3 \ 




40 


- 




20 


i 


\ CYCLE 1 

CYCLE 2\^ 

i i 



1 2 3 

COPPICE AGE (YEARS) 



2 4 6 

COPPICE AGE (YEARS) 



Figure 2. — Regeneration and survival of aspen coppice on 4- and 8-year rota- 
tions. Arrows indicate expected trends. 



CYCLE 1 



'/ 



4-YEAR 
ROTATION 




CYCLE 4 



20 
18 
16 
14 
12 
10 



8- 



2- 



CYCLE 1 



8-YEAR 
ROTATION 



1 




COPPICE AGE (YEARS) 

Figure 3. — Standing crop (total stem oven-dry weight) of aspen coppice on 4- and 
8-year rotations. Arrows indicate expected trends. Dashed line indicates the 
expected trend without hare damage (Perala 1973). 



regenerate in groups from localized areas about 6 
cm in length along the aspen root (Sandberg and 
Schneider 1953). Usually only single stems sur- 
vive but apparently these localized areas can 
continue to dominate sprout reproduction. 

The most striking results of the regression anal- 
ysis were that sucker and S/RC sprout mean 
weights (MW) responded in opposite sign to 4 of 
the 6 independent variables : 



Independent 




S/RC 


variable 


Sucker MW 


sprout MW 


Ortet age 


increased 


maximum 
at age 4 


Ortet number 


decreased 


increased 


Ortet mean weight 


decreased 


increased 


Sucker density 


increased 


decreased 


Stump/root collar 






sprout density 


increased 


decreased 


Other ramet 






mean weight 


increased 


increased 


R 2 


.92 


.97 


Sy.x, percent of 






arithmetic Y 


15 


12 



One exception was ortet age where sucker mean 
weight (MW) rose asymptotically to age 8 while 
S/RC sprout MW peaked at age 4, and then began 
to decline. This indicates that S/RC sprouting ca- 
pacity decreases with age and with the approach to 
full suckering capacity. 

The positive S/RC responses probably reflect a 
direct relation of general ramet-ortet vigor; i.e., 
vigor begets vigor. The negative responses may 
reflect competition. For example, S/RC sprout MW 
declined with both sucker and S/RC density while, 
in contrast, sucker MW increased with both sucker 
and S/RC density. This suggests that suckers dom- 
inate S/RC sprouts. 

Clearly, the physiology of aspen S/RC sprouts 
and root suckers differ, most importantly in the 
culmination of S/RC sprouting at about age 4, fol- 
lowed by the increasing dominance of suckers. 
This may, in turn, reflect the rate of aspen root 
system development. Because of carbohydrate de- 
pletion, the parent root system in age 4 stands may 
be less able to produce suckers than it is in age 2 
stands, and new roots for sucker production likely 
are not as extensive in age 4 as in age 8 stands. 
Thus, the total amount of roots capable of produc- 
ing suckers may be at their lowest level at about 
age 4. 



Nutrient Limitations 

Nutrients were not studied but the magnitude of 
nutrients removed can be estimated from exisitng 
data (Einspahr 1977) and compared to nutrient 
reserves for a Warba soil (Alban et al. 1978) simi- 
lar to this study. The total biomass removals in the 
1 -year rotation over 9 cycles would extract no more 
than 2 percent of the available soil nutrients (or 
total N). The estimated removals in three cycles of 
the 4-year rotation ranged from 2 percent of total 
N to 13 percent of exchangeable K. Corresponding 
values for two cycles of the 8-year rotation are 2 
percent N to 17 percent K. Thus, the relatively 
greater reductions in yield with shorter rotations 
are inversely related to nutrient removals. There- 
fore, the only reasonable conclusion is the declines 
in yield can be almost wholly ascribed to regenera- 
tive stress. 

The magnitude of nutrients removed on longer 
aspen rotations can also be estimated from exist- 
ing data. Generally, the concentration of above- 
ground nutrients excluding leaves declines with 
age (Einspahr 1977). However, when the values 
for tree nutrient concentrations are multiplied by 
mean annual biomass increment (Perala 1973), 
mean annual accumulation of nutrients approxi- 
mately parallels the accumulation of biomass and 
both culminate at nearly the same age (fig. 4). 



ro 
-C 

•v. 

00 

Uj 
ct 

2 




- 6 



9 11 13 15 17 
STAND AGE (YR) 



19 21 23 



Figure 4- Mean annual above-ground accumu- 
lation of nutrients and biomass in aspen (except 
leaves). Integrated from Einspahr (1977) and 
Perala (1973). 



Therefore, nutrient removals are largely propor- 
tional to biomass removals and little adjustment 
can be made in rotation length to minimize nutri- 
ent removals without reducing fiber yields. 

Other Species 

The yield of other woody species was not altered 
as dramatically by repeated cropping as was aspen 
(table 6). The total yield of other woody species in 
the 1-year rotation was reduced by about 60 per- 
cent at the third cropping and remained stable 
thereafter. In the first crop, these species com- 
prised about 20 percent of the total yield; by the 
seventh crop, their percentage increased to almost 
100! 

Total yield of these other species under the long- 
er rotations was stable. However, species differed 
greatly in their ability to withstand repeated 
cropping. It appears that red oak, mountain 
maple, red maple, and green alder are decreasers 
and the juneberry-cherry group, red-osier 
dogwood, and possibly bur oak are increasers. 



Hazel was persistent and maintained high produc- 
tivity. The drop in hazel yield on the second 8-year 
cycle can be almost wholly attributed to hare feed- 
ing damage. 

These other woody species added about one- 
fourth more to the total yield at the first 4- and 
8-year rotation and up to two-thirds more to the 
third crop of the 4-year rotation. 

CONCLUSIONS 

The failure of aspen to withstand rotations of 4 
years or less confirms the findings of Berry and 
Stiell (1978). Furthermore, their conclusion that 
aspen productivity cannot be sustained on rota- 
tions of up to about 10 years is also largely sub- 
stantiated by this study. Considering that (1) 
Berry and Stiell's second 8-year rotation regener- 
ated 80 percent of the biomass regenerated in the 
first cycle; (2) that my study at the beginning of the 
third cycle of the 8-year rotation regenerated 93 
percent of the second cycle regeneration biomass; 
and (3) that in both studies productivity of 8-year 



Table 6. — Yield of other hardwoods and shrubs 









Hardwoods 








Shrubs 












Betula 


Quercus 










June- 


Cornus 




Rota- 




papyri- 


macro- 


Quercus 


Acer 




Corylus 


berry, 


stolon- 






tion 


Cycle 


fera 


carpa 


rubra 


rubrum 


Other 


sp. 


cherry 1 


ifera 


Other 


Total 














. . .kg ha 


-1 










8 


1 


242 


135 


— 


— 




3,234 


362 


172 


2 


4,291 




2 


352 


669 


— 


— 


— 


2,137 


909 


155 


3 


4,229 


4 


1 


— 


100 


50 


15 


— 


1,368 


43 


3 


4 


2,053 




2 


— 


88 


37 


35 


— 


1,682 


80 


14 


4 


2,484 




3 


— 


84 


11 





— 


1,356 


107 


22 


4 


1,992 












. . .presence (x) 










kg ha" 1 


1 


1 




X 


X 


X 


( 5 ) 


X 


X 


X 


( 6 ) 


166 




2 


X 


X 


X 


X 


( 5 ) 


X 


X 


X 


( 6 ) 


147 




3 




X 


X 


X 




X 


X 


X 


( 6 ) 


56 




4 




X 


X 




( 5 ) 


X 


X 


X 


( 6 ) 


68 




5 




X 


X 




( 5 ) 


X 


X 


X 


( 6 ) 


74 




6 




X 


X 






X 


X 


X 




69 




7 




X 


X 






X 


X 


X 




71 



'Amelanchier sp., Prunus pensylvanica, P. virginiana 

2 Alnus crispa, 97 kg ha; Acer spicatum. 32; Dirca palustrus and Crataegus sp., 17. 

3 Salix sp , 6; Acer spicatum. 1 . 

'Salix sp. - 90, 276, 256; Alnus crispa - 384, 272, 155; cycles 1 - 3 in order 

5 777/a americana, Acer saccharum, Fraxinus nigra. 

6 Salix s p., Acer spicatum. 



rotations was diminished much less than in 3-, 4-, 
and 5-year rotations, it seems reasonable to con- 
clude, as did Berry and Stiell, that rotations of at 
least 15 years are unlikely to impair aspen regen- 
erative and productive capacity. Because short ro- 
tation yields are at least 25 percent less than can 
be gained under longer rotations approaching cul- 
mination of MAI, it seems unlikely that financial 
rotations would be prescribed that approach these 
limits imposed by regeneration requirements. 

Rotation length is not a factor in reducing nu- 
trient losses in cropped aspen because nutrient 
extraction is directly a function of biomass ex- 
traction (leaves not considered). This is not to say, 
however, that nutrient management is not impor- 
tant in short rotation systems. On the contrary, 
short rotations timed to coincide with maximum 
mean annual increment will maximize nutrient 
extraction. Whether fertilization will be required 
to maintain productivity under such systems still 
remains a question. 



LITERATURE CITED 

Alban, D. H., D. A. Perala, and B. E. Schlaegel. 
1978. Biomass and nutrient distribution in as- 
pen, pine, and spruce stands on the same soil 
type in Minnesota. Can. J. For. Res. 8:290-299. 

Baker, F. S. 1925. Aspen in the central Rocky 
Mountain region. U.S. Dep. Agric. For. Serv., 
Dep. Bull. 1291,47 p. 



Berry, A. B., and W. M. Stiell. 1978. Effect of 
rotation length on productivity of aspen sucker 
stands. For. Chron. 54:265-267. 

Brinkman, K. A., and E. I. Roe. 1975. Quaking 
aspen silvics and management in the Lake 
States. U.S. Dep. Agric. For. Serv., Agric. 
Handb. 486, 52 p. 

Einspahr, D. W. 1977. Aerial biomass and nutri- 
ent accumulation in quaking aspen stands of 
north central Wisconsin. Rep. 1, Project 3250-2, 
46 p. The Inst. Pap. Chem., Appleton, WI. 

Maini, J. S. 1968. Silvics and ecology ofPopulus in 
Canada. In Growth and utilization of poplars in 
Canada, p. 20-69. Dep. For. Rural Develop., For. 
Branch, Dep. Publ. 1205. 

Perala, D. A. 1973. Stand equations for estimating 
aerial biomass, net productivity, and stem sur- 
vival of young aspen suckers on good sites. Can. 
J. For. Res. 3:288-292. 

Perala, D. A. 1977. Manager's handbook for aspen 
in the north-central States. U.S. Dep. Agric. For. 
Serv., Gen. Tech. Rep. NC-36, 30 p. U.S. Dep. 
Agric. For. Serv., North Cent. For. Exp. Stn., St. 
Paul, MN. 

Sandberg, D., and A. E. Schneider. 1953. The 
regeneration of aspen by suckering. Univ. Min- 
nesota For. Notes 24, 2 p., St. Paul, MN. 

Schlaegel, B. E. 1975. Estimating aspen volume 
and weight for individual trees, diameter 
classes, or entire stands. U.S. Dep. Agric. For. 
Serv., Gen. Tech. Rep. NC-20, 16 p. U.S. Dep. 
Agric. For. Serv., North Cent. For. Exp. Stn., St. 
Paul, MN. 



<: U.S. Government Printing Office: 1979 — 668-108/131 Region No. 6 



i ior\A I-nDCCT CCD\/IPC 



Perala, Donald A. 

1979. Regeneration and productivity of aspen grown on repeated short 
rotations. U.S. Dep. Agric. For. Serv., Res. Pap. NC-176, 7 p. U.S. Dep. 
Agric. For. Serv., North Cent. For. Exp. Stn., St. Paul, MN. 

Regeneration and productivity of aspen cannot be sustained on 
rotations shorter than about 8 years. Productivity losses on short 
rotations are physiological and morphological rather than nutritional 
in nature. Stump and root collar sprouts, which are rare in mature 
stands, were more numerous than root suckers. 

OXFORD: 231:613:176.1 Populus tremuloides. KEY WORDS: Popu- 
lus tremuloides, suckers, sprouts, productivity, physical properties. 



Perala, Donald A. 

1979. Regeneration and productivity of aspen grown on repeated short 
rotations. U.S. Dep. Agric. For. Serv., Res. Pap. NC-176, 7 p. U.S. Dep. 
Agric. For. Serv., North Cent. For. Exp. Stn., St. Paul, MN. 

Regeneration and productivity of aspen cannot be sustained on 
rotations shorter than about 8 years. Productivity losses on short 
rotations are physiological and morphological rather than nutritional 
in nature. Stump and root collar sprouts, which are rare in mature 
stands, were more numerous than root suckers. 

OXFORD: 231:613:176.1 Populus tremuloides. KEY WORDS: Popu- 
lus tremuloides, suckers, sprouts, productivity, physical properties. 




Plant a tree... trees give oxygen. 



i it^r-iA rADCCT CCD\/irC 
I 

USDA FOREST SERVICE 
RESEARCH PAPER NC-177 




GOVT. DOCUMENTS 
DEPOSITORY ITEM 

JAN -4 1980 

§kfMSON 
UBRARY 



The timber 
marketing process 

in Indiana 



John C. Callahan 

John M. Toth 

Joseph T. O'Leary 








Morth Central Forest Experiment Station 
-orest Service, U.S. Department of Agriculture 



ACKNOWLEDGMENTS 

We gratefully acknowledge the cooperation of 
the landowners who provided the information for 
the study. We also thank the Indiana Department 
of Natural Resources' Division of Forestry for their 
helpful and cooperative assistance. Special thanks 
are due John Datena, State Forester; William 
Schrand, Assistant State Forester; Robert Koenig; 
Harold Bruner, Eldon R. Campen, Steve Creech, 
Mike Coggeshall, Earl M. McCleery, Jack E. Nel- 
son, Larry Owen, Charles Ratts, Gary M. Ross, 
Donald Stump, Joe Schueman, and Gregory Yapp, 
District Foresters. Their cheerful assistance was 
essential to the successful initiation and comple- 
tion of the study. The counsel and assistance of 
Drs. Jerry Sesco and David Baumgartner of the 
USDA Forest Service, North Central Forest Ex- 
periment Station, were also invaluable in initiat- 
ing the study. 



*& <•- 



t 

North Central Forest Experiment Station 

Robert A. Hann, Director 

Forest Service - U.S. Department of Agriculture 

1992 Folwell Avenue 

St. Paul, Minnesota 55108 

Manuscript approved for publication January 22, 1979 

1979 



ioi->A |-r\DCCT CCD^/IPC 



THE TIMBER MARKETING PROCESS IN INDIANA 



John C. Callahan, Professor, 

John M. Toth, Graduate Assistant, 

and Joseph T. O'Leary, Assistant Professor, 

Department of Forestry, Purdue University, 

West Lafayette, Indiana 



SALE OF PRIVATE 
WOODLAND IN INDIANA 

Although each nonindustrial, private landown- 
er in the East owns only a few acres, collectively 
this group owns most of the commercial forest land 
in the entire Country. These owners and their 
forest resources have taken on added significance 
in recent years because of the renewed and more 
vigorous interest of the public sector and environ- 
mental groups in the potentials of increased non- 
industrial private forest timber production. 

Research addressed to investment and produc- 
tivity questions pertaining to small forest land 
holdings has been extensive during the past 35 
years. A review of previous marketing studies 
revealed that woodland owners sold timber infre- 
quently and were rarely informed of current mar- 
ket prices. As a result, timber was often sold at less 
than fair market value. 

Because knowledge about the initiation of tim- 
ber sales and post-logging attitudes of owners ap- 
peared to be limited, the objectives of this study 
were: 

1. To describe the timber marketing process 
and procedures used by the Indiana nonin- 
dustrial private forest landowner when sell- 
ing timber. 

2. To determine if the economic and environ- 
mental results of the timber sale affect the 
landowner's disposition toward future wood- 
land management activities and timber 
sales. 

3. To determine the factors in the marketing 
process that influence the outcome of the sale 
from the seller's perspective. 



PROCEDURE 
Data Collection 

A personal interview with a highly-structured 
questionnaire was used to obtain information in 
four general areas: 

1. Demographic characteristics of the landown- 
er and site characteristics of the owner's 
woodland property. 

2. The negotiation process and actual timber 
sale procedure. 

3. The content of the timber sale contract. 

4. The landowner's reaction and attitudes after 
the sale, and his disposition toward future 
timber sales and management activities. 

Structured responses (check the box, circle a 
number) were requested wherever possible, al- 
though many questions, particularly those relat- 
ing to the landowner's reactions and attitudes, 
were left open-ended. Only one interviewer was 
used and a minimum of "prompting" was done 
during the interview to discourage the possibility 
of the landowner giving responses "the-inter- 
viewer-wanted-to-hear." 



Choosing the Sample 
Population 

The number and geographical distribution of 
timber sales in Indiana each year is unknown. 
Based on an annual cut of 300 million board feet 
and an average sale volume of 60,000 board feet 
(Indiana Department of Natural Resources 1978), 



there may be as many as 5,000 sales in Indiana 
each year. The names of all timber sellers, there- 
fore, could not be obtained without an intensive 
canvas. 

By contacting the District Foresters of the Indi- 
ana Department of Natural Resources (IDNR) and 
timber buyers and by consulting the Purdue Uni- 
versity Timber Marketing Bulletin, we came up 
with the names of 329 people who had sold timber 
in Indiana in the previous 18 months. We concen- 
trated on sales in the unglaciated forested area of 
southern Indiana where 70 percent of the State's 
wooded areas are located and we tried to include a 
representative sample of sales in which a profes- 
sional forester participated, sales in which the 
landowner acted alone, and sales in which an in- 
dustrial timber buyer participated. 

Because we felt the sample of landowners might 
not come from a normally distributed population, 
we applied Lilliefor's test for normality to three 
sets of collected ordinal-level data — volume of tim- 
ber sold, total price received for the timber, and the 
number of acres within the sale area. In all three 
instances, the null hypothesis of normality was 
rejected with a confidence level greater than 99 
percent. As a result, nonparametric statistical 
analysis was used in evaluating the data. Wilcox- 
on's rank-sum test (Wilcoxon 1945, Hollander and 
Wolfe 1973) and the Kruskal-Wallis test (Kruskal 
and Wallis 1952, Conover 1971) were used for this 
analysis. 



Contacting Potential 
Respondents 

Approximately 6 weeks before the survey was 
begun, all 329 potential respondents were sent a 
letter, describing the purpose of the survey and 
requesting their voluntary cooperation. Enclosed 
with the letter was a postcard that the landowner 
was requested to complete, sign, and return. The 
letter and postcard emphasized that cooperation 
was voluntary and any information given during 
the interview would be kept strictly confidential. 
If no reply was received to the initial letter within 
3 weeks, a second letter was sent along with 
another reply postcard. 

The number of people who had sold timber and 
who agreed to participate in the interview process 



was 182. Interviews were conducted during a 14- 
week period throughout central and southern In- 
diana. A total of 159 interviews was obtained. 



Interview Procedure 

During the interview, the woodland owner was 
encouraged to speak freely and discuss any and all 
aspects of the timber sale. The interviewer asked 
specific questions only when it became apparent 
that the landowner was not going to touch on a 
particular aspect of the sale without prompting. 
Most landowners were willing and even eager to 
provide opinions about their timber sale. 



Sample Population 
Characteristics 

The woodland owners interviewed for this study 
had higher incomes, better educations, and held 
more professional jobs than those forest landown- 
ers studied in previous marketing surveys (Suth- 
erland and Tubbs 1959, McClay 1963, Worley 
1960) and tha*n the average population (table 1). 
The median age of landowners was 56.5 years and 
the median tenure of ownership was 15.5 years. 
Two-thirds of the landowners interviewed were 50 
years of age or older. 

Although the average tenure of ownership was 
15.5 years, examination of the distribution of the 
tenure indicates a recent, rapid turnover in forest 
land ownership — 25 percent of the landowners 
owned their forested property for 5 years or less. 
Evidence indicates that the professional/manage- 
ment portion of the sample is the least tenured and 
the group buying much of the Indiana forested 
land. Among the landowners interviewed, 52 per- 
cent of those who are classified as professionals or 
managers have owned their land for 5 years or 
less, compared to only 21 percent of the farmers. 

Eighty-eight of the landowners live on or adja- 
cent to their forest land and many of the remaining 
owners live just a few miles from the property. 
Eighteen of the landowners (11 percent) live more 
than 25 miles from their timbered property and 
several live more than 100 miles from it. Typi- 
cally, professional/management/administrative 
people reside at a distance from their forested land 
and farmers lived on or adjacent to their woodland. 



ior->A IT/^DCTOT" CCD\/IPC 



Table 1. — Median age, tenure, and income of the landowners surveyed 



Occupation 



Farmer 

Retired 

Professional 

Management/Administrative 

Salesman 

Laborer 

Craftsman 

Housewife/Widow 

Other 

Total 

Overall Sample Median 











Annual 


Frequency 


Age 


Tenure 


income 


Number 


Percent 


Years 


Years 


Dollars 


48 


30 


54.0 


15.0 


15,000 


38 


24 


68.5 


24.0 


8,000 


23 


15 


49.0 


5.0 


25,000 


19 


12 


49.0 


6.0 


19,000 


10 


6 


53.5 


10.5 


20,000 


6 


4 


48.0 


11.0 


17,500 


5 


3 


43.0 


5.0 


12,000 


5 


3 


63.0 


15.0 


15,000 


5 


3 


49.5 


13.5 


16,500 


159 


100 









56.5 



15.5 



15,043 



RESULTS 
The Timber Marketing Process 

The usual timber sale negotiation was not a 
lengthy process. About 30 percent of the sale 
agreements were completed after just 1 day of ne- 
gotiation and most (about 55 percent) were con- 
summated within 1 week. More than 75 percent of 
the sales were completed within 30 days after the 
buyer and seller first made contact with each other 
and 95 percent were completed within 4 months. 
In 80 percent of the cases the timber was cut and 
removed from the property within a 4-month peri- 
od after the sales agreement was completed. 

Reasons for selling 

The most common response to why the landown- 
er was selling the timber was that it was mature 
and ready to be cut. The woodland owner's need for 
immediate cash was the second most common rea- 
son (17 percent). If the landowner revealed the 
specific need for the money received from the tim- 
ber sale, it was usually either to improve or up- 
grade a farming operation or to purchase a major 
durable good. 

Contrary to expectations, it was the seller who 
usually took the initiative in the marketing proc- 
ess. In the sample of sales studied, sellers took the 
initial step in 85 percent of the cases. Buyers 
sought out the seller in only 10 percent of the 



cases. In a few cases a forester triggered the sale 
through suggestions to the timber owner. 

More than 90 percent said that they didn't have 
any negotiating problems. Those who cited prob- 
lems listed distrust of the top bidder, no appraised 
value to judge the merits of submitted bids, or 
apprehension due to the few number of bids. 

In most instances (70 percent) the seller was 
seemingly not concerned about the personal char- 
acter and integrity of the buyer. Buyers were cho- 
sen solely on the basis of price in about two-thirds 
of the cases. However, in one sale out of five the 
buyer was selected on the basis of reputation, pre- 
vious dealings, or recommendations of others. In 
127 of the 159 sales, the sellers knew that the 
purchaser was licensed by the IDNR and that the 
seller was entitled to protection under State stat- 
ute. Thirty sellers didn't know whether the buyer 
was licensed or not and only 2 believed that the 
purchaser of their timber was unlicensed. 



Methods of determining price 

Competitive bidding was used to set the price for 
the timber in half of the sales. Although this would 
seem to indicate a more competitive marketing 
situation than that found in other States, it is more 
likely a reflection of the larger number of sales in 
which a professional forester was involved (63 per- 
cent) because foresters normally advise the land- 
owner to sell by competitive bid. 



The person initiating the sale process influenced 
the method of price determination. In 52 percent of 
the owner-initiated sales, the price was set by com- 
petitive bidding. However, only 13 percent of the 
prices in timber buyer-initiated sales were set by 
competitive bidding. All seven sales that were ini- 
tiated by a consulting forester were concluded by 
competitive bidding. 

Requesting bids from a large number of buyers 
did not guarantee more than one bid. Seventy-one 
sales (45 percent) were made on the basis of only 
one bid. In 22 of those sales the landowner indi- 
cated that more than one buyer had been contacted 
but only a single offer had been received. These 
sales usually involved small volumes of low-value 
timber. In 59 sales the timber buyer's offer was 
accepted by the landowner without question or 
negotiation. 

In 10 sales the timber was not sold as stumpage. 
In these cases, the landowner was paid a percent- 
age of the delivered mill price for his timber. All 
but one of the landowners was satisfied with this 
method of selling and felt that the logger had an 
incentive to seek the best possible price from all 
local mills. 

Knowledge of fair market prices 

Efficient marketing implies both the buyer and 
seller are aware of the value of the commodity 
changing ownership. Many of the landowners (54 
percent) sold their timber without knowledge of its 
current market value. Two-thirds of the landown- 
ers interviewed either had no idea or only a vague 
notion of the value of the timber they were selling. 
This deficiency appeared to be one of the major 
obstacles to effective timber marketing. 

Use of a professional forester 

One hundred sales (63 percent) were conducted 
with the assistance of either an IDNR service for- 
ester or a private forestry consultant — 78 with the 
assistance of an IDNR forester and 22 with a pri- 
vate consultant. 

Ninety-one of the 100 landowners were com- 
pletely satisfied with the performance of the pro- 
fessional forester. Of the nine people who were not 
satisfied, three used consultants, and six used 
IDNR foresters. The common complaints among 
those landowners using the IDNR forester was the 
inability of the forester to appraise their timber 



(IDNR regulations prohibit the forester from do- 
ing an appraisal) and the long waiting period nec- 
essary to obtain the State forester's services. The 
three woodland owners dissatisfied with the ser- 
vices of the private consultant had differing 
complaints related only to the conduct of their 
individual sale. 

Significant differences were noticed in the mar- 
keting process between those sales with and with- 
out a professional forester, and between those 
sales involving an IDNR forester and a private 
consultant. Chi-square tests for independence in- 
dicated landowners who used a professional for- 
ester during the sale generally: 

1. had a larger number of buyers bidding on the 
timber; 

2. used a written contract to sell the timber; 

3. sold a larger volume of timber (more than 
20,000 board feet); and 

4. sold stumpage by competitive bid as opposed 
to accepting the buyer's first offer for the 
timber. 

No significant relation was noted between the 
use of a professional forester and the landowner's 
(1) age, (2) tenure of ownership, (3) education, (4) 
annual income, (5) reason for selling, or (6) knowl- 
edge of timber market prices. 

Because the private consultant handled virtu- 
ally every detail of the sale with which he was 
involved, little variation in marketing procedures 
was noted. All 22 consultant-handled sales were 
concluded with a written contract, and 14 percent 
of the sales involving an IDNR forester were com- 
pleted with an oral agreement. Although many 
landowners complained about the extended wait 
for the services of the IDNR forester, this appar- 
ently did not cause significant numbers of them to 
utilize the services of a private consultant even 
when the stated reason for selling their timber was 
the need for immediate cash. 

The timber sale contract 

Seventy-five percent of the timber sales were 
concluded with a written contract. In about 40 
percent of the sales made under contract, the buy- 
er either provided his firm's contract form or com- 
pleted the "standard" Indiana timber sale contract 
form. The IDNR forester or consultant assisted the 
woodland owner with contract language in ap- 
proximately 45 percent of the cases. Rarely (7 



sales) did the owner acknowledge that he had pre- 
pared his own timber sale contract. 

Volumes sold and payments received 

Except for 10 sales in which the owner and log- 
ger shared on a percentage basis the sawmill's 
payments to the logger, all sales were paid for on a 
lump sum basis. Only four sellers of stumpage 
reported that there had been a payment problem. 
Forty-one of the owners did not know how much of 
their timber had been sold. Of those that did know 
how much had been sold, the average volume was 
about 60,000 board feet. The amount received for 
the sales ranged from $33 to $93,000 with an aver- 
age of $6,550. In those sales where competitive 
bidding was used to determine the price paid to the 
woodland owner, the difference in price between 
the low bid and the high bid ranged from 2 to 300 
percent and averaged 70 percent. 

In those sales where both the amount of money 
received and the volume of timber sold was avail- 
able, the average price ranged from $20 to $360 
per thousand board feet and averaged $105 per 
thousand board feet. 

Post-Sale Reactions and 
Attitudes 

Twenty-five percent of the landowners indicated 
they were dissatisfied with the condition of their 
property after logging. Reasons varied, but some 
felt they had only themselves to blame because 
they had either not obtained a written contract or 
had failed to insert adequate protection clauses in 
the contract. However, many landowners who had 
such protection clauses in their contracts were also 
dissatisfied with the resulting condition of the 
property and no correlation was found between the 
specifications of the contract and the resulting ex- 
pressed condition of the landowner's property. The 
interviewees felt it would be a waste of their time 
and money to prosecute those loggers in violation 
of the contract. 

Eighty-four percent of the landowners felt they 
had received a fair price for their timber. However, 
some of the landowners stated that they were only 
vaguely aware of timber prices and were in a poor 
position to judge a fair market value. 

Eighty-four percent of the owners said they were 
planning to sell more timber from the same prop- 



erty sometime in the future. Eleven landowners 
indicated they had plans to develop or sell the 
property and three said they would not sell any 
more timber as a direct result of the outcome of 
their most recent sale. 

To test for the effect of the landowner's knowl- 
edge of timber prices on the average price received, 
the sales were segregated into two groups: those 
where the landowner claimed to have some knowl- 
edge of his timber's value prior to the sale and 
those where the landowner admitted knowing 
nothing about timber prices. The test for an in- 
crease in average price per thousand board feet 
due to the landowner's knowledge of the timber 
market was significant at the 90 percent confi- 
dence level, but not at the 95 percent level. The 
Hodges-Lehmann estimate for the average differ- 
ence in price showed that those landowners who 
had some idea of the timber's value received $19 
more per thousand board feet than those landown- 
ers who had no idea of the timber's value. 

Wilcoxon's test was used to check if the presence 
of more than one bidder increased the average 
price received by the landowner. Again, the test 
was significant at the 90 percent confidence level 
but not at the 95 percent level. The Hodges-Leh- 
mann estimator for the average difference in price 
showed that when there were two or more bidders 
the price increased $13 per thousand board feet. 

The Kruskal-Wallis test was used to see if the 
method of price determination, reason for selling, 
or volume of timber sold had an effect on the price 
received. However, none of these factors proved 
significant. Even less significant were the effects 
of who initiated the timber sale and the use of a 
professional forester. 

Perhaps the most surprising finding of this anal- 
ysis is the lack of effect on price when a profes- 
sional forester is involved, particularly in view of 
the significance of the other variables. Those sales 
with a professional forester generally had a larger 
number of buyers bidding on the sale. Yet, al- 
though the number of buyers bidding appears to 
influence the price received, the use of a forester 
does not. The explanation for this apparent contra- 
diction may be the type of timber offered for sale by 
the two groups. The data indicate that sales with- 
out the help of a professional forester included 
higher-quality timber than those sales with the 
help of a forester. A possible explanation for this 



difference may be found in the marketing process. 
In 16 percent of the sales involving some walnut 
veneer timber, the trees to be sold were chosen 
either by the landowner or a professional forester. 
However, in those sales where the timber buyer 
was allowed to choose the trees to be cut, veneer 
quality walnut was sold in 28 percent of the cases. 
This difference, though not substantial, may ac- 
count for the lack of difference in average price 
between those sales with and without a profes- 
sional forester. If given his choice, the buyer would 
most likely pick the most valuable trees in the 
woodland and would thus be able to offer a higher 
average price per board foot for the entire amount 
of timber sold. 

Plans for Future Sales and 
Timber Management 

The reactions and attitudes of the landowner 
following the timber sale may affect his disposi- 
tion toward future timber sales and management. 
Two factors are important: ( 1 ) was the landowner 
satisfied with the price received for his timber?, 
and (2) was the owner satisfied with the condition 
of the woodland after logging? Displeasure with 
either may discourage future sales. 

Statistical tests were applied to determine if the 
average price received per thousand board feet and 
the condition of the property affected the landown- 
er's plans for future timber sales. A chi-square test 
showed a fairly significant degree of association 
between price and plans for future sales (90 per- 
cent confidence level). However, Wilcoxon's test on 
the same data was considerably less significant 
(less than 80 percent confidence). The landowner's 
perception of the post-sale condition of the prop- 
erty apparently had little or no effect on the 
landowner's decision to sell timber in the future. 
Neither the price received nor the condition of the 
property appeared to influence the landowner's 
plans for future timber management because both 
tests of these factors were not significant. 



HIGHLIGHTS 

The woodland owners interviewed were not typ- 
ical of those described in previous studies. Those 
cooperating in the study were more affluent, more 
likely to be professionally employed, more often 
lived in an urban environment, and had in most 



cases utilized the assistance of a forester in mak- 
ing the sale. In 90 percent of the sales, the seller 
initiated the marketing process. Although most 
owner's professed to be market knowledgeable, 
more than 70 percent were not familiar with or 
had only vague notions about timber prices and 
the value associated with their trees. Those who 
did have price knowledge received on the average 
$19 per thousand board feet more than those that 
didn't. 

The presence of a forester in the sale proceedings 
made it more likely that the timber would be sold 
under contract and that competitive bidding 
would be used to establish fair market value (70 
percent of forester-assisted sales). However, spir- 
ited bidding by more than three bidders tended to 
be the exception rather than the usual bidding 
situation. 

More than 90 percent of the woodland owners 
using professional foresters in the marketing 
process were completely satisfied with their per- 
formance. However, landowners utilizing IDNR 
foresters commonly complained about the regula- 
tion that prevented district foresters from ap- 
praising their timber prior to offering it for sale. 
Average prices received by landowners utilizing 
IDNR foresters were not found to be significantly 
different from the prices received by owners 
dealing directly with timber purchasers. However, 
forester-associated sales were more likely to have 
had competitive bidding (positive effect on price) 
and were more likely to contain lower quality tim- 
ber (negative effect on price). 

Fully 25 percent of the timber sellers were dis- 
satisfied after the timber had been sold and cut, 
because of the price received or because of the 
residual condition of the land and timber. How- 
ever, this did not seem to affect the owners' deci- 
sions with respect to future harvests. 

The most serious deficiencies in the marketing 
process appeared to be a lack of available timber 
price information and/or estimates of the fair mar- 
ket value of the owner's timber prior to sale. This 
was particularly evident when landowners di- 
rectly approached timber buyers and accepted 
their first offer and in those other cases where 
there were less than three bidders. The admoni- 
tion of foresters to landowners that they should 
secure at least six bids for their timber to be as- 
sured of a fair market value appears to be correct 



but realistically unobtainable. Less than 16 per- 
cent of the sellers reported that they had received 
more than three bids for their timber. 



LITERATURE CITED 

Conover, W.J. 1971. Practical nonparametric sta- 
tistics. John Wiley & Sons, Inc., New York. 

Hodges, H.L., Jr., and E.L. Lehmann. 1963. Esti- 
mates of location based on rank tests. Ann. 
Math. Statist. 34:598-611. 

Hollander, M., and D.A. Wolfe. 1973. Nonparame- 
tric statistical methods. John Wiley & Sons, 
Inc., New York. 

Indiana Department of Natural Resources. 1978. 
The L.T.B. Bulletin, Vol. VI, No. 3. Div. For. 
Dep. Nat. Resources, State of Indiana. 

Kruskal, W.H., and W.A. Wallis. 1952. Use of 
ranks in one criterion variance anlysis. J. Am. 
Statist. Assn. 47:583-621. 



Sutherland, C.F., Jr., and C.H. Tubbs. 1959. Influ- 
ence of ownership on forestry in small wood- 
lands in central Wisconsin. U.S. Dep. Agric. For. 
Serv., Stn. Pap. 77, 21 p. U.S. Dep Agric. For. 
Serv., Lakes States For. Exp. Stn., St. Paul, MN. 

Wilcoxon, F. 1945. Individual comparisons by 
ranking methods. Biometrics 1:80-83. 

McClay, T.A. 1963. Similarities among owners of 
small private forest properties in nine eastern 
localities. J. For. 59:88-92. 

Sutherland, C.F., Jr., and C.H. Tubbs. 1959. Influ- 
ence of ownership on forestry in small wood- 
lands in central Wisconsin. U.S. Dep. Agric. For. 
Serv., Stn. Pap. 77, 21 p. U.S. Dep Agric. For. 
Serv., Lakes States For. Exp. Stn., St. Paul, MN. 

Wilcoxon, F. 1945. Individual comparisons by 
ranking methods. Biometrics 1:80-83. 

Worley, D.P. 1960. The small woodland owners in 
eastern Kentucky. U.S. Dep. Agric. For. Serv., 
Tech. Pap. 175, 5 p. U.S. Dep. Agric. For serv., 
Cent. States For. Exp. Stn., Columbus, OH. 



ICHA CHDCCT CCDUIPC 



Callahan, John C, John M. Toth, and Joseph T. O'Leary. 

1979. The timber marketing process in Indiana. U.S. Dep. Agric. 
For. Serv., Res. Pap. NC-177, 7 p. U.S. Dep. Agric. For. Serv., North 
Cent. For. Exp. Stn., St. Paul, MN. 

Analyzes the timber marketing process in Indiana by examining 
the sale experience of 159 woodland owners who had recently sold 
timber. 



OXFORD: 764(772). KEY WORDS: timber prices, future timber mar- 
kets, consulting foresters, woodland owner attitudes, timber sale pro- 
cedures. 



Callahan, John O, John M. Toth, and Joseph T. O'Leary. 

1979. The timber marketing process in Indiana. U.S. Dep. Agric. 
For. Serv., Res. Pap. NC-177, 7 p. U.S. Dep. Agric. For. Serv., North 
Cent. For. Exp. Stn., St. Paul, MN. 

Analyzes the timber marketing process in Indiana by examining 
the sale experience of 159 woodland owners who had recently sold 
timber. 



OXFORD: 764(772). KEY WORDS: timber prices, future timber mar- 
kets, consulting foresters, woodland owner attitudes, timber sale pro- 
cedures. 



■fr U S GOVERNMENT PRINTING OFFICE 1979—667-858/90 



Soil is for plants.. 




not for tire tracks, 



i icha chdcct ccD\/irc 

USDA FOREST SERVICE 
RESEARCH PAPER NC-178 




GOVT. DOCUMENT! 
DEPOSITORY ITEM 

JAN V 1980 

CIEMSON 



Weights 

and dimensional 

properties of shrubs 

and small trees of 

the Great Lakes 

conifer 

forest 



Peter J. Roussopoulos and Robert M. Loomis 



North Central Forest Experiment Station 
Forest Service, U.S. Department of Agriculture 






Roussopoulos, Peter J., and Robert M. Loomis. 

1979. Weights and dimensional properties of shrubs and small trees of 
the Great Lakes conifer forest. U.S. Dep. Agric. For. Serv., Res. Pap. 
NC-178, 6 p. U.S. Dep. Agric. For. Serv., North Cent. For. Exp. Stn., 
St. Paul, MN. 

Presents equations for estimating biomass and woody size class 
distributions for shrubs and small trees (< 2.5 cm d.b.h.) of 17 north- 
eastern Minnesota species. Relations between stem diameter at 15 cm 
above ground and plant height, crown length, and stem diameter at 
ground are also given. 

OXFORD: 521.1:531:518(77). KEY WORDS: forest fuels, size classes, 
component weights, fuel modeling. 



North Central Forest Experiment Station 

Robert A. Hann, Director 

Forest Service - U.S. Department of Agriculture 

1992 Folwell Avenue 

St. Paul, Minnesota 55108 

Manuscript approved for publication August 11, 1977 

1979 



IcnA CADCCT CCDV/ir^C: 



WEIGHTS AND DIMENSIONAL PROPERTIES OF 

SHRUBS AND SMALL TREES OF THE GREAT 

LAKES CONIFER FOREST 

Peter J. Roussopoulos, formerly Associate Forest Fuels Scientist, 

East Lansing, Michigan 

(currently with the Rocky Mountain Forest and Range Experiment Station, 

Fort Collins, Colorado) 

and Robert M. Loomis, Fire Control Scientist, 

East Lansing, Michigan 



Biomass estimates are often used in deter- 
mining primary productivity of ecosystems, quan- 
tifying energy pathways and nutrient cycles, 
anticipating product yields from harvest activi- 
ties, evaluating wildlife habitats, and appraising 
forest flammability. Accordingly, biomass infor- 
mation needs and estimation methods have been 
discussed frequently in the literature of several 
disciplines. Specific information requirements 
vary substantially, though, depending on the con- 
text of the problem being considered. 

One of the most information-demanding uses is 
the assessment of wildland fire behavior potential 
(Rothermel 1972), requiring quantitative esti- 
mates of available fuel weights by condition 
(living or dead) and by size category. 

Studies reporting data suitable for fuel model- 
ing in Great Lakes conifer forests (Rowe 1959) are 
rare, especially for the unmerchantable parts of a 
forest community such as small trees and shrubs 
(Ohmann et al. 1976, Crow 1977, Telfer 1969). 
Although these reports have some value for fuel 
evaluation, they fail to estimate component 
weights by dead or live categories or by size classes 
as desired for fire behavior prediction. A recent 
study by Brown (1976) devised estimating 
equations for 25 shrubs of the northern Rocky 
Mountains. Equations were presented to estimate 
foliage and stemwood with a table of percentages 
of stemwood within specific fuel size classes for 
species groups. 

To appraise upland forest fuels and wildfire po- 
tential for the Boundary Waters Canoe Area in 
northeastern Minnesota, above ground biomass 
equations were developed for locally prominent 



shrubs and small trees. The resulting equations 
are presented herein, with primary emphasis on 
applications involving fuel modeling and fire be- 
havior prediction. 



METHODS AND ANALYSIS 

Shrubs and small trees (<2.5 cm d.b.h.) were 
collected during July and August of 1976 on the 
Kawishiwi Ranger District of the Superior Na- 
tional Forest in northeastern Minnesota (47°50'N 
and 91°45'W). Stems were cut at groundline and 
were taken to the Kawishiwi Field Laboratory for 
processing. Seventeen different species were sam- 
pled, each represented by at least 20 collected 
stems. For each sample stem, the following sample 
measurements were recorded: stem diameter at 
ground level and at 15 cm above ground level to 
the nearest 0.25 cm (measurement of diameter at 
15 cm above ground avoids the region of high stem 
taper normally found at groundline); plant height, 
and length (depth) of crown to the nearest 15 cm. 
Each plant was divided into components of foliage 
and woody parts. Dead and live woody parts were 
also separated. All woody parts 1 were further sep- 
arated into size classes by diameter: to 0.6 cm, 0.6 
to 2.5 cm, and 2.5 to 7.6 cm. These size groups 
correspond to the 1-, 10-, and 100-hour timelag 
fuels described in the National Fire Danger 
Rating System by Deeming et al. ( 1972). Each com- 
ponent group was weighed to the nearest 0.1 gram 
and its moisture content determined by sub- 
sampling and ovendrying for 24 hours at 105 C. All 
fresh weights were converted to ovendry in this 
manner. 



'Hereafter, "woody" refers to the woody parts of 
the plant; i.e., the composite of wood and bark. 



To facilitate subsequent mathematical repre- 
sentation, measured dry weights of wood attribut- 
able to the three mutually exclusive size classes 
were arithmetically combined into the inclusive 
size classes: 0-0.6 cm, 0-2.5 cm, and 0-7.6 cm. 

Regression analysis was used to relate com- 
ponent dry weights to stem diameter at 15 cm 
height. Analysis of variance and graphical analy- 
sis were used to compare regression equations for 
individual species and explore possibilities for 
grouping similar species. 

Total plant weight, foliage weight, total wood 
weight (live and dead), and live wood weight, all in 
grams dry weight (Y), were regressed with stem 
diameter (X), measured in cm at a height of 15 cm, 
using the allometric model: 

Y = aX b (1) 

Regression coefficients were estimated using the 
logarithmic transformation of equation (1). The 
"a" coefficient was adjusted for bias inherent in 
this procedure (Baskerville 1972). 

For each stem, the dry weights of all woody 
material less than 0.6 cm in diameter and all 
woody material less than 2.5 cm in diameter were 
divided by the overall weight for total wood and for 
live wood only. These ratios (Y) represent the pro- 
portional contribution of size classes to 0.6 cm 
and to 2.5 cm, inclusive, to the weight of the live 
and the total woody components. They were re- 
gressed against the stem diameter (X) at a height 
of 15 cm using the hyperbolic model: 

Y = X/(a + bX) (2) 

The regressions were performed using the follow- 
ing linearized form: 

X/Y = a + bX (3) 

In this form, the dependent variable (X/Y) is used 
only to evaluate the coefficients "a" and "b" for 
subsequent use in equation (2). 

To help evaluate the bulk density and vertical 
distribution of understory fuels, linear regression 
equations were also developed for plant height and 
crown length on the stem diameter at 15 cm in 
height. These were statistically forced through the 
origin to produce a simple ratio estimator for plant 
height and crown length. Although this approach 
may be questionable for small plants (since all 
plants less than 15 cm tall are predicted to have 
zero height and crown length), the resulting errors 
are deemed negligible within the diameter range 



of principal interest. Even for smaller stems, the j 
height and crown length measurement resolution 
(± 7.5 cm) tends to minimize the importance of 
potential underestimates. 



RESULTS AND DISCUSSION 

In all, 460 stems were collected and processed 
representing 14 deciduous species of trees and 
shrubs and three coniferous trees (table 1). For all 
species, the range of sampled stem diameters was 
0.3 to 5.1 cm at 15 cm above ground. All species 
were represented over the bulk of this interval 
except Diervilla lonicera, Lonicera canadensis, 
and Rosa acicularis. These small shrubs rarely 
attain stem diameters outside the range sampled. 
Total above ground dry weight per stem ranged 
from 1 to 2,714 grams dry weight for all species. 



Component Weights 

Regression statistics were calculated for dry 
weights of all above ground components, foliage, 
total wood, and live wood (table 1). Examination of 
the coefficients of determination (r 2 ) shows reason- 
ably good fits for all species except Diervilla loni- 
cera and Lonicera canadensis. These low r 2 values 
may be partially due to the narrow range (0.2 cm 
for Diervilla ) of sampled stem diameters compared 
to the measurement resolution (± 0.12 cm). 

Meaningful species groupings, to facilitate ag- 
gregate modeling of forest communities for broad 
fuel appraisal, were illusive. No statistically de- 
fensible groups could be found that were applica- 
ble for all four dependent variables. The three 
species groups appearing in table 1 were derived 
through graphical comparisons of the regression 
equations. Though the F-test did not fully support 
these groups, differences among the individual 
"within-group" equations were generally not 
meaningful, from a practical standpoint, over the 
expected range of stem diameters. Extreme indi- 
vidual species estimates of "total" weight varied 
about 20 percent from the group estimate for the 
combined 11 species at a common 1.6-cm base 
diameter. 

The regression equations agree quite well with 
those of Ohmann et al. (1976) except for Corylus 
cornuta, where their estimates show somewhat 






lt?r»A CODCCT CCD\/I/^C 



Table 1. — Sample size and regression coefficients 1 for estimating component dry-weights of shrubs and small 
trees (< 2.5 cm d.b.h.) 







Range 




































Stems 
collected 


of stem 
diameters 




Total 








Foliage 






All wood 


Sy-x 




Live wood 




Species 


a 


b 


r 2 


Sy-x 


a 


b 


r 2 Sy-x 


a 


b 


r 2 


a 


b 


r 2 ! 


>y-x 


Abies balsamea 


25 


0.5-3.3 


72715 


2.250 0.96 


8!) 


29.319 


2.011 


3.94 


38 


42.904 


2.404 97 


50 


41.330 


2.394 0.97 


49 


Acer rubrum 


36 


0.3-4.1 


60.367 


2 342 


94 


278 


13.082 


1.840 


.91 


25 


45.947 


2.505 


93 


374 


45.085 


2.480 


.92 


246 


Acer spicatum 


25 


0.3-4.3 


73.182 


2 259 


.95 


141 


17.305 


1 696 


89 


26 


54.779 


2.407 


93 


1 2? 


52 384 


2417 


95 


127 


Alnus spp 


28 


0.8-4 1 


63280 


2.380 


93 


164 


14.725 


1.828 


.90 


18 


48.762 


2.509 


90 


164 


48.077 


2 484 


96 


160 


Amelanchier spp 


27 


0.5-4.1 


71.534 


2.391 


93 


174 


10.478 


1.988 


83 


21 


60.997 


2.445 


94 


160 


58 333 


2.458 


91 


160 


Betula papyrifera 


23 


1.3-3.6 


76316 


2 279 


93 


73 


14 717 


1.529 


66 


17 


62.830 


2.378 


93 


75 


61.956 


2.376 


.92 


77 


Cornus rugosa 


27 


0.3-3.6 


74.114 


2.457 


96 


134 


17 131 


2 093 


93 


13 


55 886 


2.591 


96 


136 


54 629 


2.551 


95 


133 


Corylus cornuta 


36 


0.3-25 


62 819 


2 420 


89 


4fi 


12 115 


2.010 


81 


8 


50.154 


2 523 


90 


47 


49.245 


2 503 


90 


44 


Diervilla lonicera 


21 


3-0.5 


14.211 


1.217 


.45 


4 


3.082 


613 


.19 


1 


12269 


1 608 


53 


3 


9.276 


1.445 


59 


1 


Lonicera canadensis 


25 


0.3-1.0 


33.900 


1.793 


68 


5 


5.319 


1.275 


39 


2 


28 899 


1.942 


67 


4 


28.017 


2 020 


69 


4 


Picea spp. 


25 


0.5-3.3 


65 757 


2.287 


.97 


68 


36.288 


2.047 


95 


43 


28 670 


2.566 


98 


38 


27.806 


2.543 


97 


34 


Populus spp. 


27 


0.5-3.3 


46.574 


2.527 


90 


52 


10828 


2.052 


87 


19 


35 264 


2.657 


97 


41 


34.906 


2.655 


97 


41 


Prunus spp. 


25 


0.8-3.8 


68.041 


2.237 


9II 


155 


12.382 


2 024 


77 


43 


55076 


2 306 


87 


1 52 


54.235 


2.253 


86 


143 


Rosa aciculans 


23 


0.3-1.3 


83.240 


2.837 


83 


9 


22853 


2.282 


79 


3 


63 140 


3 224 


82 


9 


60 282 


3.214 


83 


8 


Salix spp. 


25 


0.5-3 


55925 


2.594 


96 


113 


12 280 


2.120 


94 


32 


43.316 


2726 


95 


96 


42.495 


2.721 


95 


95 


Sorbus americana 


24 


0.5-3.8 


44.394 


3.253 


95 


350 


8083 


2.601 


93 


11 


35.960 


3427 


96 


407 


35585 


3.425 


95 


398 


Thuja occidentalis 


38 


0.3-5 1 


68423 


1.863 


94 


86 


35.288 


1.442 


90 


36 


30 800 


2.244 


94 


63 


30.632 


2.232 


94 


59 


Abies balsamea & Picea spp 


50 


0.5-3.3 


69.167 


2.267 


.97 


73 


32 743 


2 033 


94 


48 


35.691 


2 480 


.96 


59 


34 483 


2.464 


96 


52 


Diervilla lonicera & 






































Lonicera canadensis 


46 


0.3-1 


25.879 


1.636 


60 


5 


4.340 


944 


30 


1 


22768 


1.913 


60 


4 


20.190 


1.898 


63 


4 


Eleven species 2 


303 


0.3-4.1 


62.134 


2.460 


93 


155 


12.573 


2.006 


86 


24 


48.944 


2.577 


93 


152 


47.780 


2.567 


92 


146 



'Regressions are of the form Y = aX h where Y is the component weight in grams, X is the stem diameter in centimeters measured 15 centimeters above 
ground, and a and b are regression coefficients from the table Weights are expressed in grams of total above ground material (Total), foliage (Foliage), dead 
and live woody parts (All wood), and live woody parts only (Live wood) for 17 species or genera and 3 species combinations. 

2 Acer rubrum, Acer spicatum, Alnus spp., Amelanchier spp , Betula papyrifera, Cornus rugosa, Populus spp., Corylus cornuta, Prunus spp., Salix spp., 
Sorbus americana. 



lower weights — especially at the larger stem di- 
ameters. We found this species similar to Alnus 
spp., Amelanchier spp., and Salix spp. — genera 
that Ohmann et al. combined also. Because their 
samples were collected in the same general loca- 
tion, and because they also used stem diameter 
measured at a height of 15 cm as the independent 
variable, close agreement is not surprising. Brown 
(1976) and Telfer (1969), on the other hand, used 
stem diameter at ground level. 

To facilitate comparison with the results of 
these studies, the relation between the 15-cm stem 
diameter and basal stem diameter was examined. 
Scatter diagrams suggested that ground diameter 
could be predicted from the 15-cm diameter using 
simple linear regressions. The resulting coef- 
ficients were remarkably similar for all species 
(table 2). Telfer's (1969) weight predictions for 
woody plants in eastern Canada, after diameter 
adjustment, were also in close agreement. Brown's 
(1976) equations, on the other hand, yielded lower 
weight estimates for most species, perhaps partial- 
ly due to the different environmental conditions of 
the northern Rocky Mountains. Both Brown and 



Telfer predicted greater weights for Lonicera spp. 
at larger diameters (Brown's weights were lower 
than Telfer's). Brown had the broadest diameter 
range for Lonicera (0.3 to 1.7 cm); Telfer's was 
similar to this study (0.1 to 0.7 cm). 

Woody Size Classes 

Examination of scatter diagrams revealed that 
the proportional contributions of the 0- to 0.6-cm 
and 0- to 2.5-cm (inclusive) size classes to total 
woody weight are discontinuous functions of stem 
diameter. They equal 1.0 at low stem diameters 
and fall quickly away from this value above some 
"critical stem diameter" near the upper size class 
limit. To ensure realistic size class predictions on 
both sides of this discontinuity, two measures were 
necessary. First, for each size class we found the 
diameter of the smallest stem that contained 
woody material in the next larger size class. Natu- 
rally, these values were close to the upper diame- 
ter limits — about 0.5 cm for the 0- to 0.6-cm class 
and 2.1 cm for the 0- to 2.5-cm class — and varied 
little among species. Stems with diameters below 



Table 2. — Regressions through origin (y = 
for basal stem diameter versus 



-- bx) for height and crown length, and linear regressions (y =a + bx) 
stem diameter (cm) at 15 cm above ground level 





Height (meters) 
b Sy-x 


n 


Crown length (meters) 
b Sy-x n 




Basal diameter (cm) 




Species 


a 


b 


r 2 


Sy-x 


n 


Abies balsamea 


0.7094 


0.2902 


25 


0.6455 


0.2876 


25 


0.0684 


1.1302 


0.9216 


0.2929 


25 


Acer rubrum 


1.3761 


.5851 


33 


.9522 


.6927 


33 


.0003 


1.1675 


.9649 


.2039 


36 


Acer spicatum 


1.2100 


.4989 


22 


.7443 


.4093 


22 


.1645 


1.0485 


.9499 


.2488 


25 


Alnus spp. 


1.1289 


.8339 


6 


.5331 


.9089 


6 


.1409 


1.0225 


.9592 


.1695 


28 


Amelancbier spp. 


1.3176 


.3496 


17 


.8061 


.3114 


17 


.0142 


1.1037 


.9815 


.1569 


27 


Betula papyrifera 


1.5720 


.5564 


21 


.9837 


.7265 


21 


.1713 


1 .0452 


.9376 


.1968 


23 


Cornus rugosa 


1.1728 


.6782 


29 


.6860 


.4204 


27 


.0243 


1.0828 


.9714 


.1505 


27 


Corylus cornuta 


1.5314 


.3192 


36 


.9510 


.3085 


36 


.1894 


.9226 


.9476 


.1214 


36 


Diervilla lonicera 


1.3268 


.1389 


21 


.8179 


.1520 


21 


.1062 


.8818 


.5216 


.1126 


21 


Lonicera canadensis 


1.2184 


.2402 


25 


.7488 


.1876 


25 


.0809 


.9780 


.7346 


.1188 


25 


Picea spp. 


.5772 


.1769 


25 


.5050 


.1389 


25 


.0715 


1.1241 


.9711 


.1858 


25 


Populus spp. 


1.2515 


.5219 


27 


.8136 


.6473 


27 


.1294 


1.0517 


.9643 


.1752 


27 


Prunus spp. 


1.2183 


.5750 


19 


.7943 


.3435 


19 


.1151 


1 .0676 


.9417 


.2094 


25 


Rosa acicularis 


1.4505 


.1967 


23 


.8661 


.1609 


23 


.0338 


1.0412 


.8412 


.1092 


23 


Salix spp. 


1.2747 


.5282 


17 


.7497 


.4044 


17 


.0502 


1.1730 


.9810 


.1543 


25 


Sorbus americana 


1.5018 


.4470 


24 


.9532 


.4532 


24 


.0263 


1.1373 


.9735 


.1370 


24 


Thuja occidentalis 


.6290 


.2256 


38 


.6063 


.2124 


38 


.1853 


1.0906 


.9556 


.2925 


38 


All coniferous 


.6350 


.2518 


88 


.5884 


.2441 


88 


.1293 


1.1058 


.9514 


1 .4084 


88 


All deciduous 


1.3293 


.5156 


318 


.8342 


.4923 


318 


.0434 


1.1072 


.9670 


1.0127 


372 



these values were deleted from the respective size 
class regressions. This eliminated samples from 
the "flat" section of the curve where the propor- 
tional size class contribution is 1.0 and allowed 
separate mathematical representation of the 
"flat" and "falling" curve sections. Second, the crit- 
ical stem diameter — the point separating the two 
sections — was defined from the coefficients of each 
hyperbolic regression as a /(1-b). The regression 
equation applies only to stem diameters above this 
value, which results in the following expression 
for the fractional contribution of each size class ( Y) 
in terms of stem diameter (X): 



1.0, for < X < 



(1 -b) 



("flat" section) 



Y= (4) 

7 — , uvv for— t— «£ X ("falling" section) 

(a + da) (1 — b) 

Regression coefficients were calculated for use 
with equation (4), both for all wood and for live 
wood only (table 3). For the < 0.6 cm size class 
regressions, Diervilla lonicera was the only species 
that had no samples with stem diameters more 
than 0.5 cm. Regressions were performed for all 
other species. Diervilla, Corylus cornuta, Lonicera 



canadensis, Rosa acicularis, and Sorbus ameri- 
cana were exempted from the 0.0 to 2.5 cm size 
class regression analysis because each had less 
than five sampled stems that were 2.1 cm or more 
in diameter. Good fits were obtained for most of the 
remaining species with this model. Regressions 
were also run for the three species groups used 
earlier. Again, analysis indicated the combina- 
tions to be reasonable and practical, though statis- 
tically not fully justifiable. 

Actual weight estimates for each size class can 
be obtained by multiplying the appropriate frac- 
tional weight contribution estimate (equation (4), 
table 3), times the corresponding predicted wood 
weight (equation (1), table 1). Weights of the 
0.6- to 2.5-cm and > 2.5-cm size classes, as well as 
dead wood weights may be found by subtraction. 

At small stem diameters, below about 0.5 cm, 
the entire woody component is within the 0- to 
0.6-cm size class (fig. 1). As stem diameter in- 
creases above this point, the fractional contribu- 
tion of this class drops quickly to an asymptote at 
0.14 (for the grouped 11 deciduous species), while 
the 0.6- to 2.5-cm class becomes prominent. At 
roughly 2.3 cm, material greater than 2.5 cm 
appears and the middle size class begins to fall 



ior\A criDCCT ccD\/irc 



Table 3. — Regression statistics for estimating fractional weight contributions of woody components by size 
class and condition (live or dead) for 1 7 species and 3 species combinations of northern Minnesota 
shrubs and small trees. The regression model is XIY =a + bX, where independent variable X is 
stem diameter (cm) measured 15 cm above ground level and Y is the fraction by weight attributed to 
each indicated size class. 







All woody part; 


• 0.6 cm 


Live woody parts 0.6 cm 




All woody parts 2.5 


cm 




Live woody parts 


2.5 cm 








divided by all woody parts 


divided by live woody parts 
a b r 2 Sy-x 


n 


divided by all woody parts 
a b r 2 Sy-x 


n 


divided by live woodv parts 




Species 


a 


b 


r 2 


Syx 


n 


a 


b 


r 2 


Syx 


n 


Abies balsamea 


-0.8141 


2 3989 


0.924 


0.6129 


25 


-1.0298 


2.6303 


0911 


0.7324 


25 


-4 2677 


2.8728 


940 


3290 


9 


-4 7586 


3.0867 


928 


03887 


9 


Acer rubrum 


-62520 


10 3120 


718 


59481 


33 


-7.4744 


11.5724 


734 


6.5710 


33 


-6 0540 


35985 


.805 


1.1121 


8 


-6 2718 


3 7192 


687 


1.3304 


« 


Acer spicatum 


-4.7664 


7.6075 


925 


2.1277 


23 


-5 3703 


8.5717 


.913 


2.6055 


23 


- 8 6441 


4.1621 


939 


7271 


6 


-8.8463 


42436 


.940 


7397 


6 


Alnus spp 


-4.2928 


6 9640 


854 


2.3184 


28 


-5.0621 


7.7270 


821 


2.9061 


.'H 


-3.8505 


2.8249 


946 


5092 


6 


-3 9024 


2 8765 


940 


.5463 


6 


Amelanchier spp. 


-4.0400 


68436 


918 


2.1167 


27 


-4.4118 


7.2891 


905 


2.4477 


27 


-6.4998 


4.0315 


911 


7414 




-7 0820 


4.2563 


41 II, 


8053 


6 


Betula papyrifera 


-58830 


7.7092 


915 


1.7199 


23 


-7.1140 


8 5998 


898 


2 1135 


23 


-6.0057 


3.5414 


973 


2942 


10 


-6.4097 


3 7084 


972 


.3138 


10 


Cornus rugosa 


-2.6090 


5 6040 


752 


2.1095 


24 


-32924 


6.4142 


788 


2.2932 


24 


-.4652 


1 1927 


978 


1039 


7 


- 4892 


1.2027 


976 


I092 


7 


Corylus cornuta 


-2.0501 


4.6178 


896 


.8202 


33 


-2.5036 


5 2050 


904 


8849 


33 


(') 


(') 


(') 


(') 


!') 


(') 


(') 


(') 


(') 


(') 


Diervilla lonicera 


(') 


(') 


(') 


(') 


(') 


(') 


(') 


(') 


(') 


(') 


(') 


(') 


(') 


(') 


(') 


(') 


(') 


!') 


(') 


(') 


ionicera canadensis 


-.8217 


2.6503 


.939 


1054 


15 


-.8217 


2.6503 


939 


.1054 


15 


(') 


(') 


(') 


(') 


(') 


(') 


(') 


(') 


(') 


(') 


Picea spp 


-.7873 


2.5976 


964 


4800 


25 


-.9063 


2.8078 


952 


6046 


25 


-4 0003 


2.7137 


422 


3528 


9 


-4 2958 


2 8364 


938 


.3236 


9 


Populus spp 


-4.8321 


8.2591 


841 


3.1064 


27 


-4.5801 


8.3773 


829 


3.2937 


27 


-6 3969 


3.1764 


910 


.5263 


9 


-6.4176 


3.7228 


911 


.5235 


9 


Prunus spp 


-2.0843 


5.1685 


.721 


1.6514 


34 


-2.4157 


5.8313 


.694 


2.1243 


24 


-4.7809 


3 1011 


984 


.6872 


5 


-4.7884 


3.1078 


487 


6857 


5 


Rosa acicularis 


-1.1971 


3 3862 


911 


2203 


17 


-1.1971 


3.3862 


911 


2203 


17 


(') 


(') 


(') 


(') 


(') 


(') 


(') 


(') 


(') 


(') 


Salix spp 


-4.1190 


7.4681 


760 


4 4845 


23 


-4.1282 


7.7257 


.792 


4 2844 


23 


-6 0504 


3.5769 


760 


1.1493 


•3 


-6 1529 


3 6205 


7 72 


1 1484 


9 


Sorbus americana 


-10.0310 


15.1988 


932 


2 9887 


24 


-9.9149 


15 4869 


929 


3.1181 


.'4 


(') 


(') 


(') 


(') 


(') 





(') 


(') 


(') 


(') 


Thuja occidentals 


-4.8576 


5 8264 


843 


3 0684 


36 


-6.3768 


6 9339 


797 


4 2709 


36 


8 3000 


4 3000 


940 


1 1477 


13 


-9 4581 


4 7387 


933 


1.3232 


13 


Abies balsamea S 










































Picea spp 


-.8239 


2.5127 


939 


.5789 


50 


-.9923 


2.7343 


926 


6987 


50 


-4.0207 


2.7498 


926 


.3331 


18 


-4.3792 


2 9042 


424 


3590 


18 


Diervilla lonicera & 










































Lonicera canadensis 


-.8342 


2.6645 


.943 


.0785 


26 


-.8359 


2 6636 


.944 


.0782 


26 


(') 


(') 


(') 





(') 


(') 


(') 


(') 


(') 


I 1 ) 


Eleven species 


-4.0271 


7 3193 


691 


4.1284 


296 


-4.4322 


79182 


.703 


4.3433 296 


-5.0517 


3.1742 


656 


1.1939 


72 


-5.1878 


3.2378 


650 


1.2344 


72 



'Range of diameters insufficient to perform regression. 



^ IU -_••.! 
1 | 
















Size Class (cm) 


2 i 








^^"" 2.5 


\ 
















• — — 0.6 


§ 0-8 - \ 




,."""" 


'-. 




O 1 


..< 




'. 




i 


.*' 




I 




- 

^ 


i / 

l .r 




\ 




K 


1 ." 




\ 




2 0.6- 


is 
IS 




\ 




01 


■t 






^^^^^^^ 


o 


:l 






***'*.^^^ 


Ui 0.4 - 


: 1 






jr**t, 


N 


: \ 






^r **'*»# 


W 


: \ 
\ 






^F "'"•»!, 


2 


S 






f 


I 


\ 








K : 




^ 






g 0.2- 








1- : 








1 










O = 




















111 










* n = 


i 


i 




i i 



STEM DIAMETER AT 15 CM ABOVE GROUND (cm) 



Figure 1. — Fractional size class composition (by 
weight) of total stem and branchwood component 
versus stem diameter for a group of eleven 
species. 



toward its asymptote at 0.18. Once established, 
the largest class rises throughout the range of 
sampled stem diameters. 

Also of interest for flammability appraisal is the 
"dead-to-live" ratio of stemwood. This may be 
found either by size class or for the entire stem by 
subtracting the appropriate "live woody parts" es- 
timate from the corresponding "all woody parts" 
estimate, and dividing the difference by the esti- 
mate for the live. Dead-to-live ratios are often 
more easily interpreted in terms of shrub flamma- 
bility than are actual component weights. 

Plant Height and Crown Length 

Besides the quantity and size distribution of fuel 
materials, spatial distribution or fuel arrange- 
ment also influences flammability. Knowledge of 
total heights and crown lengths of understory veg- 
etation can be helpful in modeling forest fuels for 
predicting fire behavior. Equations were devel- 
oped to predict these dimensions using 15 cm stem 
diameter as the predictor variable. Regression 
analysis using a forced O-intercept was used (table 
2). To preserve the noteworthy differences in slope 



coefficients for coniferous versus deciduous spe- 
cies, only two grouped regressions were per- 
formed. Plant heights for the conifers were 
roughly half those of deciduous plants with the 
same stem diameter. The crown length ratios for 
coniferous samples (crown length/total height) are 
characteristically 1 .5 times those of deciduous spe- 
cies. These statistical observations are confirmed 
by physical experience and seem to justify the 
chosen species combinations. 

SUMMARY 

Using regression equations presented in this 
paper, one may estimate the quantity and vertical 
distribution of understory fuels by component, live 
or dead, and wood size categories from inventories 
of easily measured plant dimensions. If only plant 
heights or only stem diameters at ground level are 
known, the measurements can be converted to 
stem diameter at 15 cm, the predictor variable for 
component weights and size class proportions. The 
estimating equations can be used with the most 
confidence within the diameter ranges sampled for 
individual species and do not apply to trees larger 
than 2.5 cm d.b.h. 



LITERATURE CITED 

Baskerville, G. L. 1972. Use of logarithmic regres- 
sions in the estimation of plant biomass. Can. J. 
For. Res. 2:49-53. 



Brown, J. K. 1976. Estimating shrub biomass from 
basal stem diameters. Can. J. For. Res. 6:153- 
158. 

Crow, Thomas R. 1977. Biomass and production 
regressions for trees and woody shrubs common 
to the Enterprise Forest. In The Enterprise, 
Wisconsin, Radiation Forest — Radioecological 
Studies. J. Zavitkovski, ed. p. 63-67. Tech. Inf. 
Cent., Energy Res. and Dev. Admin. TID-26113- 
P2. 

Deeming, J. E., J. W. Lancaster, M. A. Fosberg, R. 
W. Furman, and M. J. Schroeder. 1972. The Na- 
tional Fire-Danger Rating System. U.S. Dep. 
Agric. For. Serv., Res. Pap. RM-84, 165 p. U.S. 
Dep. Agric. For. Serv., Rocky Mt. For. and 
Range Exp. Stn., Fort Collins, CO. 

Ohmann, Lewis F., David F. Grigal, and Robert B. 
Brander. 1976. Biomass estimation for five 
shrubs from northeastern Minnesota. U.S. Dep. 
Agric. For. Serv., Res. Pap. NC-133, 11 p. U.S. 
Dep. Agric. For. Serv., North Cent. For. Exp. 
Stn., St. Paul, MN. 

Rothermel, R. C. 1972. A mathematical model for 
predicting fire spread in wildland fuels. U.S. 
Dep. Agric. For. Serv., Res. Pap. INT-115, 40 p. 
U.S. Dep. Agric. For. Serv., Intermountain For. 
and Range Exp. Stn., Ogden, UT. 

Rowe, J. S. 1959. Forest regions of Canada. Can. 
Dep. Natl. Aff. Nat. Res. For. Branch Bull. 123. 

Telfer, E. S. 1969. Weight-diameter relationships 
for 22 woody plant species. Can. J. Bot. 47:1851- 
1855. 



. 






S-U.S. GOVERNMENT PRINTING OFFICE: 1980-667-859/91 REG. #6 Pi 



- X ; Aic- - / /7 

SDA FOREST SERVICE 
ESEARCH PAPER NC-179 



# % 




GOVT. DOCUMENTS 
DEPOSITORY ITEM 



!AN 23 



CLEMSON 
LIBRARY 



Summer moisture contents 

of understory vegetation 

in northeastern Minnesota 



Robert M. Loomis, 

Peter J. Roussopoulos, 

and Richard W. Blank 



)rth Central Forest Experiment Station 
•rest Service, U.S. Department of Agriculture 









North Central Forest Experiment Station 

Robert A. Hann, Director 

Forest Service — U.S. Department of Agriculture 

1992 Folwell Avenue 

St. Paul, Minnesota 55108 

Manuscript approved for publication March 16, 1979 

1979 



SUMMER MOISTURE CONTENTS OF 
UNDERSTORY VEGETATION IN NORTHEASTERN 

MINNESOTA 



Robert M. Loomis, Fire Control Scientist, 

East Lansing, Michigan, 

Peter J. Roussopoulos, Forest Fuel Scientist, 

(now with the Rocky Mountain Forest and Range Experiment Station, Fort Collins, Colorado) 

and Richard W. Blank, Biological Technician, 

East Lansing, Michigan 



The behavior and effects of forest fires are often 
influenced by the character of the living under- 
story vegetation. Grasses, ferns, shrubs, mosses, 
tree reproduction, and herbaceous plants may 
either contribute actively to the energy of a fire, or 
they may serve as a heat sink and retard fire 
propagation and intensity. This can mean the dif- 
ference between a beneficial, easily managed fire, 
and a detrimental fire that is difficult to control. 
Whether living surface fuels play an active or pas- 
sive role in the combustion process depends 
largely on their abundance and moisture content. 
These factors are quantitatively addressed in 
state-of-the-art models of fire behavior (Rothermel 
1972, Albini 1976) and in the National Fire Dan- 
ger Rating System (Deeming et al. 1977). 

Although ongoing research holds promise for 
developing the capability to simulate live fuel 
moisture changes through knowledge of physical 
processes governing plant-water relations (How- 
ard 1978, Running 1978), currently available 
methods of predicting live fuel moisture content, 
without actual field measurement, are somewhat 
primitive. The algorithm proposed by Deeming et 
al. (1977) for fire danger rating represents the 
most comprehensive effort to date to provide a 
generally applicable model. It considers the pre- 
dominant type of vegetation (annual or perennial), 
the phenological stage of development, vegetative 
adaptation to moisture stress as reflected in the 
humidity provinces of Thornthwaite (1931), and 



antecedent precipitation and evaporative demand 
of the atmosphere. This method seems to work well 
in the northern Rocky Mountains and in other 
portions of the western United States, but it has 
not been thoroughly validated. 

Measured moisture contents of tree and shrub 
foliage have been reported in several studies rep- 
resenting diverse geographic areas (Olsen 1960, 
Philpot 1963, Reifsnyder 1961, Jameson 1966, 
Johnson 1966, Van Wagner 1967, Blackmarr and 
Flanner 1968, Gary 1971, Countryman 1974, Rus- 
sell and Turner 1975). Herbaceous moistures, on 
the other hand, are not as well documented. 
Although many ecologists, range and wildlife sci- 
entists, plant physiologists, and fire scientists 
measure herbaceous moisture contents regularly, 
little of this information is formally reported. The 
few published studies indicate that moisture con- 
tents of grasses and forbs growing under generally 
similar conditions are related (Turner 1972) and 
that many plants common to the West show "a 
decided decrease in moisture content" as the grow- 
ing season advances (Richards 1940). The degree 
to which these observations apply in other geo- 
graphic regions, however, has not been fully 
established. 

To assess the potential flammability of forests in 
the Boundary Waters Canoe Area (BWCA) in 
northeastern Minnesota (Roussopoulos 1978), 
moisture contents and fresh-to-dry weight conver- 
sion factors were required for common grasses, 



herbs, mosses, and small shrubs. 1 This paper pre- 
sents and examines summer moisture content 
measurements for 21 groups of understory plants 
that are common to the Great Lakes Forest Region 
(Rowe 1959). Results are useful for a variety of fire 
management activities including estimating liv- 
ing fuel biomass and predicting fire control diffi- 
culty and/or fire effects for actual or anticipated 
fires. They also offer a preliminary basis for vali- 
dating the National Fire Danger Rating System 
herbaceous moisture model (Deeming et al. 1977) 
in the upper Midwest. 

METHODS 

Field sampling of living herbs, mosses, and 
small shrubs for gravimetric moisture content de- 
termination was conducted in mature forest 
stands about 20 km south of Ely, Minnesota (47° 
50' N, 91° 45' W). Sampling began on June 24, 
after the period of primary plant growth, and 
ended on August 26, 1976. Throughout this period, 
plants representing 21 plant categories (individu- 
al species, genera, or groups of related species) 
found commonly in upland forest communities of 
the area were collected at intervals of one to sev- 
eral days. Forty-two subsamples were taken 
between 2:00 and 4:00 p.m. each sample day (3 
subsamples each from 14 plant groups). Sampling 
frequency for individual species or species groups 
varied approximately in proportion to the percent 
cover each represented. 

Each subsample consisted of at least 5 grams 
fresh weight of a single species or plant group. All 
above-ground plant parts were included. Sub- 
samples were sealed in metal cans and transported 
at once to the laboratory where they were weighed, 
dried at 105C for at least 16 hours, and reweighed 
to the nearest 0.1 gram. Dry weight conversion 
factors (CF) were computed for each subsample 



^Moisture content is weight loss expressed as a 
percentage of oven-dry weight: moisture content 
percent — 



( 



green weight — ovendry weight 
ovendry weight 



) 



x 100 



The conversion factor is the number that, when 
multiplied by green weight, yields oven dry weight: 
(conversion factor - oven dry weight ■*■ green 
weight). 



(dry weight/fresh weight) and a daily average was 
determined for each subsample triplet. These were 
converted to moisture contents (MC) by: 

100 

CF 



MC - 



100 



(1) 



At the end of the sample period, the time series 
of daily moisture content percentage values were 
examined and compared on the basis of magnitude 
and seasonal trend. Graphical analysis, t-tests, 
and regression analysis were the principal analyt- 
ical methods used. 



RESULTS AND DISCUSSION 

The number of sample days for each sampling 
category ranged from 3 for starflower (Trientalis 
borealis), false Solomon's seal (Smilacina race- 
mosa), and twinflower (Linnaea borealis) to 21 for 
large-leaved aster {Aster macrophyllus) and wild 
sarsaparilla (Aralia nudicaulis) (table 1). Sample 
size reflects the relative contribution to under- 
story plant cover in the sampling vicinity. 

Averages of seasonal moisture content percents 
ranged from 138 for Labrador tea {Ledum groen- 
landicum) to 1,027 for bluebead-lily {Clintonia 
borealis). 

Scatter diagrams showed downward seasonal 
trends in moisture content values for most plant 
groups, which reflects the influence of phenologi- 
cal development as well as the 1976 summer 
drought (Dickson 1976). Collections were made 
during the afternoon, the time of day when mois- 
ture stress with decreased moisture content would 
be most likely to occur due to an imbalance be- 
tween absorption and transpiration rates. Despite 
the drought, however, the observed trends were 
generally more subtle than has been reported for 
western plants (Richards 1940). Linear regression 
equations fit the time series data reasonably well 
and had significant correlation coefficients for 
Labrador tea, large-leaved aster, and for blue- 
bead-lily (fig. 1 ). Many species, such as wild lily-of- 
the-valley {Maianthemum canadense) on the other 
hand, showed no significant seasonal trend (fig. 1). 
To gain a clearer picture of the seasonal moisture 
responses by sampling group, average moisture 
content values and corresponding conversion fac- 
tors for all sampled plant groups were stratified by 
early (June 24 - July 24) and late (July 25 - August 
26) sampling periods (table 1). 



Table 1. — Moisture content of some grasses, forbs, mosses, and small shrubs in northeastern Minnesota 1 







All (June 24- Aug. 26 


) 




Early (June 24-July 24) 




Late (July 25 Aug 26) 










Standard 








Standard 








Standard 




Sample Moisture Conversion 


error ol Sample Moisture Conversion 


error ol 


Sample Moisture Conversion error ol 


Plant group 


day 


content 


(actor 


he mean 


day 


content 


(actor 


the mean 


day 


content 


factor 


the mean 




No 


Percent 






No 


Percent 






No 


Percent 






Labrador tea (Ledum groenlandicum) 


10 


138 


,42) 


7 


3 


160 


;.38) 


13 


7 


128 


,44) 


7 


Blueberry 2 {Vaccinium spp.) 


17 


151 


.40) 


6 


12 


157 


.39) 


7 


5 


136 


.42) 


7 


Clubmoss 3 (Lycopodium spp.) 


15 


153 


,40) 


8 


11 


156 


.39) 


10 


4 


144 


.41) 


11 


Grasses 4 


10 


170 


.37) 


13 


6 


189 


.35) 


17 


4 


142 


.41) 


10 


Rubus 5 (Rubus spp.) 


4 


208 


.32) 


22 


— 








4 


208 


.32) 


22 


Spreading dogbane (Apocynum androsaemifolium) 


6 


217 


.32) 


22 


4 


197 


.34) 


27 


2 


259 


.28) 


26 


Twinflower (Linnaea borealis) 


3 


239 


.29) 


41 


2 


270 


.27) 


47 


1 


178 


,36) 




Bracken fern {Pteridium aquilinum) 


13 


248 


.29) 


10 


10 


258 


.28) 


11 


3 


211 


[.32; 


18 


Wild sarsaparilla (Aralia nudicaulis) 


21 


253 


.28) 


6 


13 


258 


.28) 


7 


8 


244 


.29) 


11 


Strawberry 6 {Fragaria spp.) 


11 


258 


.28) 


14 


8 


271 


■27) 


14 


3 


224 


,31) 


34 


Bunchberry {Cornus canadensis) 


17 


261 


.28) 


7 


12 


270 


.27) 


6 


5 


237 


.30) 


12 


Bush honeysuckle {Diervilla lonicera) 


5 


290 


.26) 


31 


5 


290 


.26) 


31 


— 








False Solomon's seal {Smilacina racemosa) 


3 


304 


.25) 


28 


3 


304 


.25) 


28 


— 








Pearly everlasting (Anaphalis margaritacea) 


4 


319 


.24) 


49 


2 


390 


.20) 


36 


2 


247 I 


.29) 


53 


Other ferns 7 


5 


326 


.23) 


30 


2 


388 


.20) 


12 


3 


284 


.26) 


29 


Wood horsetail {Equisetum sylvaticum) 


4 


340 


.23) 


9 


1 


335 


.23) 




3 


342 


.23) 


13 


Starflower (Trientalis borealis) 


3 


369 


•21) 


7 


3 


369 


.21) 


7 


— 








Large-leaved aster (Aster macrophyllus) 


21 


380 


.21) 


15 


14 


418 I 


.19) 


12 


7 


305 I 


.25) 


17 


Dwarf solomons seal (Maianthemum canadense) 


17 


393 


.20) 


14 


11 


403 


.20) 


17 


6 


374 | 


.21) 


23 


Sweet coltsfoot (Petasites spp. ) 


4 


560 


.15) 


70 


2 


523 I 


.16) 


147 


2 


597 I 


.14) 


72 


Bluebead-lily (Clintonia borealis) 


15 


1,027 


.09) 


35 


10 


1,071 


.09) 


34 


5 


939 I 


.10) 


69 



'Moisture content percent equals (100 x moisture content - oven dry weight). Conversion factor equals (oven dry weight * green weight); dry weight = 

conversion factor x green weight. 

2 Vaccinium myritilloides and V. angustifolium are predominant. 

3 Lycopodium clavatum and obscurum are predominant. 

*Carex spp. and Oryzopsis asperifolia are predominant. 

5 Rubus strigous and R. pubescens are predominant. 

6 Fragaria vesca and F. virginiana are predominant. 

7 Athyrium spp., Onoclea sensibilis. and Osmunda cinnamomea are predominant. 



For all sampling groups represented by at least 
three sample days in each half of the season, mois- 
ture content was lower during the late sampling 
period. The difference between the early and late 
mean moisture contents (expressed as a percent- 
age of the seasonal average) for the groups with at 
least 10 sample collections ranged from 7 percent 
for wild lily-of-the-valley to 30 percent for large- 
leaved aster — with a mean of 16 percent. Groups 
showing less than 10 percent difference between 
early and late means were club-mosses (Lycopo- 
dium spp.), wild sarsaparilla, and wild lily-of-the- 
valley. Groups with a difference between early and 
late means of from 10 to 20 percent were blueberry 



(Vaccinium spp.), bracken fern (Pteridium aquili- 
num), strawberry (Fragaria spp.), bunchberry 
(Cornus canadensis), and bluebead lily. And those 
groups with a difference of more than 20 pecent 
between early and late means were Labrador tea, 
grasses, and large-leaved aster. These substantial 
variations among sampling groups may be due in 
part to species-related differences in morphology, 
phenological development, rooting characteris- 
tics, stomatal activity, etc., as well as micro-envi- 
ronmental preference. 

Admittedly, our sample was not adequate to 
conclusively identify and characterize seasonal 
moisture responses. It is also possible that the late 



200 



180 



160 



140 



K 120 

IU 

K 

2 100 

O 

o 

hi 

5 1400 



g 1300 



1200 

1100 

1000 

900 

800 

700 



LABRADOR-TEA 



R^.66 



500 



425 



350 



275 - 







BLUEBEAD-LILY 




- 




+ 




- 






R'=.31 


■ 




+ + + 




' 


f + 


+ ++ +^"^^ 




- 




+ 


^ N **"s» l ^^ + 








+ 



200 



600 



525 



450 



375 



300 



LARGE-LEAVED ASTER 



+ + 



R 2 =.64 



- 




WILD LILY-OF-THE-VALLEY 








+ 




- 




R 2 = 


.05 


- 


+ 


+ + 

+ + 




+ 








+ 


+ 


+ + ' 


+ 


, 


+ 










+ 


+ 






+ + 

' ' ' ' ' i i .. i i i . i .. i i ..... i i ..... i ... i i 





-> 



o 

3 
< 



3 

"5 



o 

3 
< 



Figure 1. — Plant moisture content percentages for some species found under 
mature upland stands in northeastern Minnesota. 



summer drought of 1976 produced atypical plant 
moisture trends in the sample area. Nonetheless, 
the average moisture content values obtained here 
compare favorably with reported values for simi- 
lar environments of the U.S.S.R. (Svanidze and 
Khidasheli 1973), and they may be useful in devel- 
oping crude estimates of dry- weight biomass from 
field-fresh weights or in predicting forest fire be- 
havior and effects. 

One question frequently asked by land manag- 
ers concerns the suitability of the National Fire 
Danger Rating System (NFDRS) herbaceous fuel 
moisture algorithm (Deeming et al. 1977) for pre- 
dicting actual fire behavior or evaluating fuel 



flammability. Available methods of predicting fire 
spread rates, intensities, and flame lengths (Roth- 
ermel 1972, Albini 1976) require that quantitative 
estimates of living fuel moisture be developed 
where these fuels are significant. Use of a simple 
plant moisture model such as that employed for 
fire danger rating would be desirable if acceptable 
accuracy could be achieved. Because fire danger 
rating is concerned mainly with identifying rela- 
tive levels and trends in fire potential for the most 
serious situations anticipated on broad rating ar- 
eas, it is uncertain whether the NFDRS herbace- 
ous moisture algorithm would be appropriate for 
other, possibly more demanding, applications. To 
gain insight on this question, we computed the 



USDA FOREST SERVICE 



NFDRS herbaceous moisture at weekly intervals 
(Burgan et al. 1977) using 1976 observations from 
the nearest fire weather station. Herbaceous mois- 
ture values were computed for all four "climate 
classes" and plotted for the entire sampling period 
(fig. 2). All four of these curves show only slightly 
reduced moisture contents in the late season, 
which is what we observed for most of the plant 
groups sampled. 

To facilitate comparison with the field results, 
average calculated moisture contents were deter- 
mined for the season and for both early and late 
sampling periods (table 1). In addition, corre- 
sponding observed plant moisture contents were 
calculated as a weighted mean of the moisture 
content values for all plant groups (table 2). The 
relative contribution of each plant group to total 
understory biomass (dry weight) from a broad- 
scale fuel inventory conducted roughly 50 km from 
the sample area (Roussopoulos 1978) provided the 
weighting coefficients for this computation. The 
mean NFDRS moisture contents for each climate 
class were also computed for this period and are 
substantially lower than the observed values. The 
approximate prediction errors are -63 percent, -66 
percent, -70 percent, and -75 percent, respectively, 
for NFDRS climate classes 1, 2, 3, and 4. The fact 
that climate class 3 yielded the second largest un- 
derprediction is especially significant because it is 



the one recommended for rating fire danger in the 
eastern United States (Deeming et al. 1977). 

Examination of the seasonal moisture trends, on 
the other hand, reveals slightly better agreement. 
The difference between the early and late 
weighted measurement means was roughly 18 
percent of the seasonal average, while the NFDRS 
computations yielded corresponding values of 6, 9, 
10, and 15 percent, respectively, for climate 
classes 1 through 4. The NFDRS herbaceous mois- 
ture algorithm is much better at representing 
relative changes — the principal intent of fire dan- 
ger rating — than it is at predicting actual values. 
The errors identified above should be closely ex- 
amined to determine their significance in applica- 
tions requiring absolute plant moisture content 
predictions. 

For the NFDRS fuel model G (dense conifer for- 
est with heavy detritus accumulations), we pre- 
dicted selected fire behavior properties using both 
measured and calculated mean herbaceous mois- 
ture contents (table 2). Fire behavior estimates 
were computed for climate class 1 and 4 to present 
the range produced by NFDRS. Fire behavior pre- 
diction errors resulting from the use of calculated 
moisture contents were expressed as percentages 
of the predictions based on measured moisture: 



150 



2 
UJ 

h- 

O 
O 

g 100 

3 



O 

o 

UJ 

o 

CD 
CC 
UJ 

I 

to 

2 



50 



• / N XX •-. .••'••.. — 

' / \ +* \ 

.•.«—* NFDRS Climate ^•++ 1 * ^ 




_l_ 



j_ 



j_ 



_L 



_J_ 



29 



6 13 20 27 4 11 18 25 1 8 15 22 

JUNE JULY AUGUST 

Figure 2- Manually calculated National Fire Danger Rating System (NFDRS) 
herbaceous moisture contents for all four climate classes at Ely, 
Minnesota, 1976. 



Table 2. — Average measured moisture contents and National Fire Danger Rating System (NFDRS) herbace- 
ous moisture predictions by sampling period 

(In percent) 



Sampling period 


Weighted mean 

of measured 
moisture contents 




NFDRS herbaceous moisture content 






Climate 
Class 1 


Climate 
Class 2 


Climate 
Class 3 


Climate 
Class 4 


All (June 24-August 26) 
Early (June 24-July 24) 
Late (July 25-August 26) 


299 

319 

264 


112 
115 
108 


103 
107 

98 


91 
95 
86 


74 
78 
67 



E(%) = 



Vc - Vr 
V m 



• 100 



(2) 



Where: E=the percentage error in a predicted 
fire behavior characteristic (e.g. 
spread rate, flame length, etc.) us- 
ing the NFDRS herbaceous fuel 
moisture 

V c =the predicted value obtained using 
the calculated NFDRS herbaceous 
fuel moisture 

V m =the predicted value obtained using 
the corresponding weighted mean of 
measured moisture contents (table 

2). 

Fuel moistures for the 10, 100, and 1,000 hour 
timelag fuels were assumed constant at 10, 15, and 
15 percent, respectively, for all computations. For 
1 hour timelag fuel moisture contents ranging 
from 5 to 15 percent and effective windspeeds 
ranging from 4 to 9 m/sec (10 to 20 m.p.h.), forward 
spread rate (m/min) was overestimated by roughly 
55 to 100 percent using the NFDRS climate class 1 
herbaceous moisture content, and by 70 to 150 
percent using the climate class 4 value. Estimated 
spread rate for a 1 hour timelag fuel moisture of 5 
percent and a windspeed of about 9 m/sec (20 
m.p.h.) was 2.4, 4.9, and 6.1 m/min (7.9, 16.1, and 
20.0 ft/min) for measured, estimated climate class 
1, and estimated climate class 4 fuel moisture val- 
ues, respectively. Energy release component was 
overestimated by about 5 to 15 percent and 15 to 25 
percent, respectively, for climate class 1 and cli- 
mate class 4, while corresponding error ranges for 
flame length predictions were overestimated by 25 
to 45 percent and 35 to 75 percent. Estimated 
flame length predictions for a 1 hour timelag fuel 



moisture of 5 percent and a windspeed of about 9 
m/sec (20 m.p.h.) were 1.2, 1.7, and 2.0 meters (3.8, 
5.6, and 6.6 feet) for measured, estimated climate 
class 1, and estimated climate class 4 fuel moisture 
values, respectively. Indiscriminant use of the cal- 
culated live fuel moistures to predict fire behavior 
for some planning and operational activities, 
therefore, would likely result in overly conserva- 
tive decisions — possibly leading to inefficient use 
of fire management resources. 

The moisture content prediction errors noted 
above do not significantly compromise the useful- 
ness of the NFDRS herbaceous moisture algor- 
ithm for fire danger rating in the Lake States in 
terms of its intended purpose. Predicted and mea- 
sured relative herbaceous moisture trends agree. 
Thus, the NFDRS algorithm may be used to esti- 
mate relative fire behavior trends for a season. 
Uses requiring greater precision, such as compari- 
sons of Lake States fuel moistures with western 
species, or estimation of absolute values for rate of 
spread and intensity are questionable unless 
actual field fuel moisture measurements are 
obtained. The average moisture content values re- 
ported here are appropriate when general esti- 
mates are needed for planning over broad areas in 
Great Lakes conifer forests. 

LITERATURE CITED 

Albini, Frank A. 1976. Computer-based models of 
wildland fire behavior: a user's manual. 68 p. 
U.S. Dep. Agric. For. Serv., Intermt. For. and 
Range Exp. Stn., Ogden, UT. 

Blackmarr, W. H., and William B. Flanner. 1968. 
Seasonal and diurnal variation in moisture con- 
tent of six species of pocosin shrubs. U.S. Dep. 



IDA FOREST SERVICE 



Agric. For. Serv., Res. Pap. SE-33, 11 p. U.S. 
Dep. Agric. For. Serv., Southeast For. Exp. Stn., 
Asheville, NC. 

Burgan, Robert E., Jack D. Cohen, and John E. 
Deeming. 1977. Manually calculating fire-dan- 
ger ratings — 1978 National Fire-Danger Rating 
System. U.S. Dep. Agric. For. Serv., Gen. Tech. 
Rep. INT-40, 51 p. U.S. Dep. Agric. For. Serv., 
Intermt. For. and Range Exp. Stn., Ogden, UT. 

Countryman, Clive M. 1974. Moisture in living 
fuels affects fire behavior. Fire Manage. 35(2): 
10-13. 

Deeming, John E., Robert E. Burgan, and Jack D. 
Cohen. 1977. The National Fire Danger Rating 
System— 1978. U.S. Dep. Agric. For. Serv., Gen. 
Tech. Rep. INT-39, 63 p. U.S. Dep. Agric. For. 
Serv., Intermt. For. and Range Exp. Stn., Og- 
den, UT. 

Dickson, Robert R. 1976. Weather and circulation 
of August 1976: Extremes of wetness in the West 
and dryness in the Midwest. Mon. Weather Rev. 
104(11): 1455-1460. 

Gary, Howard L. 1971. Seasonal and diurnal 
changes in moisture contents and water deficits 
of Engelman spruce needles. Bot. Gaz. 132(4): 
327-332. 

Howard, E. A., III. 1978. A simple model for esti- 
mating the moisture content of living vegeta- 
tion as potential wildfire fuel. Fifth National 
Conference on Fire and Forest Meteorology, p. 
20-23. Am. Meteorol. Soc, Boston, MA. 

Jameson, Donald A. 1966. Diurnal and seasonal 
fluctuations in moisture content of pinyon and 
juniper. U.S. Dep. Agric. For. Serv., Res. Note 
RM-67, 7 p. U.S. Dep. Agric. For. Serv., Rocky 
Mountain For. and Range Exp. Stn., Fort Col- 
lins, CO. 

Johnson, Von J. 1966. Seasonal fluctuations in 
moisture content of pine foliage. U.S. Dep. 
Agric. For. Serv., Res. Note NC-11, 4 p. U.S. 
Dep. Agric. For. Serv., North Cent. For. Exp. 
Stn., St. Paul, MN. 

Olsen, James M. 1960. 1959 green-fuel moisture 
and soil moisture trends in southern California. 
U.S. Dep. Agric. For. Serv., Res. Note PSW-161, 
8 p. U.S. Dep. Agric. For. Serv., Pac. Southwest 
For. and Range Exp. Stn., Berkley, CA. 



Philpot, C. W. 1963. The moisture content of pon- 
derosa pine and white-leaf manzanita foliage in 
the central Sierra Nevada. U.S. Dep. Agric. For. 
Serv., Res. Note PSW-39, 7 p. Pac. Southwest 
For. and Range Exp. Stn., Berkeley, CA. 

Reifsnyder, William E. 1961. Seasonal variation 
in moisture content of the green leaves of moun- 
tain laurel. For. Sci. 7(1): 16-23. 

Richards, Leon W. 1940. Effect of certain chemical 
attributes of vegetation on forest inflammabil- 
ity. J. Agric. Res. 60(12)^33-838. 

Rothermel, Richard C. 1972. A mathematical 
model for predicting fire spread in wildland 
fuels. U.S. Dep. Agric. For. Serv., Res. Pap. INT- 
115, 40 p. Intermt. For. and Range Exp. Stn., 
Ogden, UT. 

Roussopoulos, Peter J. 1978. A decision aid for 
wilderness fire prescriptions in the Boundary 
Waters Canoe Area. Fifth National Conference 
on Fire and Forest Meteorology, p. 52-58. Am. 
Meteorol. Soc, Boston, MA. 

Rowe, J. S. 1959. Forest regions of Canada. Can. 
Dep. North Aff. Nat. Resour. For. Branch Bull. 
123,711 p. 

Running, Steven W. 1978. A process oriented mod- 
el for live fuel moisture. Fifth National Confer- 
ence on Fire and Forest Meteorology, p. 24-28. 
Am. Meteorol. Soc, Boston, MA. 

Russell, R. N., and J. A. Turner. 1975. Foliar mois- 
ture trends during bud swelling and needle 
flush in British Columbia. Bi-Monthly Res. 
Notes 31(4):24-25. Environ. Canada For. Serv. 

Svanidze, M. A., and Sh. A. Khidasheli. 1973. For- 
est fire danger in eastern Georgia. Unedited 
translation from the Russian Dep. of the Secre- 
tary of the State, Translation Bureau, Multilin- 
gual Serv. Div., Ottawa, Canada. 4 p. 

Thornthwaite, C. W. 1931. The climates of North 
America according to a new classification. 
Geogr. Rev. 4:633-655. 

Turner, George T. 1972. A new approach to esti- 
mating herbage moisture content. J. Range 
Manage. 25(3)229-231. 

Van Wagner, C. E. 1967. Seasonal variation in 
moisture content of eastern Canadian tree fo- 
liage and the possible effect on crown fires. Dep. 
Pub. 1204, 15 p., For. Branch, Canada Dep. For. 
and Rural Development. 



ftU.S. GOVERNMENT PRINTING OFFICE: 1979--668007/1 19 



JfiDA FOREST SERVICE 



Loomis, Robert M., Peter J. Roussopoulos, and Richard W. Blank. 

1979. Summer moisture contents of understory vegetation in northeast- 
ern Minnesota. U.S. Dep. Agric. For. Serv., Res. Pap. NC-179, 7 p. U.S. 
Dep. Agric. For. Serv., North Cent. For. Exp. Stn., St. Paul, MN. 

Summer moisture contents and factors for converting fresh plant 
weights to ovendry biomass estimates are presented for some herbs, 
mosses, and small shrubs found in the upland forest stands of north- 
eastern Minnesota. 



OXFORD: 431.2(776). KEY WORDS: forest fuels, fuel modeling, 
plant biomass, fire danger. 



Loomis, Robert M., Peter J. Roussopoulos, and Richard W. Blank. 

1979. Summer moisture contents of understory vegetation in northeast- 
ern Minnesota. U.S. Dep. Agric. For. Serv., Res. Pap. NC-179, 7 p. U.S. 
Dep. Agric. For. Serv., North Cent. For. Exp. Stn., St. Paul, MN. 

Summer moisture contents and factors for converting fresh plant 
weights to ovendry biomass estimates are presented for some herbs, 
mosses, and small shrubs found in the upland forest stands of north- 
eastern Minnesota. 



OXFORD: 431.2(776). KEY WORDS: forest fuels, fuel modeling, 
plant biomass, fire danger. 




G-tve a hoo£. . .don' £ polhi£n! 



GOVT. DOCUMENTS 
DEI 0S1TORY ITEM 



APR 14 



CL£MSON 

LIBRARY 



A provisional assessment of 
triclopyr herbicide 

for use in Lake States' forestry 



Donald A. Perala 






^pLRSTT^ 




North Central Forest Experiment Station 
Forest Service, U.S. Department of Agriculture 



ACKNOWLEDGMENT 

I wish to thank Chris Peterson and Bob Morrow 
of the Blandin Paper Company, Woodlands Divi- 
sion, for providing the study sites, helicopter ex- 
pense, and assistance in conducting the trials. Dr. 
Jack Ryder, Dow Chemical, U.S.A., was especially 
instrumental in technical guidance and provided 
the chemicals. 



North Central Forest Experiment Station 

Robert A. Hann, Director 

Forest Service — U.S. Department of Agriculture 

1992 Folwell Avenue 

St. Paul, Minnesota 55108 

Manuscript approved for publication August 15, 1978 

1980 



A PROVISIONAL ASSESSMENT OF TRICLOPYR 

HERBICIDE FOR USE IN LAKE 

STATES' FORESTRY 



Donald A. Perala, Silviculturist, 
Grand Rapids, Minnesota 



A new herbicide is needed that controls a broad 
spectrum of hardwoods, is safe for releasing coni- 
fers, and breaks down rapidly in the forest environ- 
ment without moving from the site. The commonly 
used2,4-D [(2,4-dichlorophenoxy) acetic acid] does 
not control most northern hardwoods; though 2,4,5- 
T [(2,4,5-trichlorophenoxy) acetic acid ] controls 
more hardwoods (especially oaks), it may not con- 
tinue to be available for forest management. Piclo- 
ram (4-amino-3,5,6-trichloropicolinic acid) con- 
trols most Lake States hardwoods (Perala 1971, 
1974, Perala and Williams 1970), but because it is 
such a broad spectrum herbicide its use will likely 
be limited to site preparation rather than for aerial 
release of conifers. 

A recently developed chemical, triclopyr, (3,5,6- 
trichloro-2-pyridinyloxyacetic acid), controls a 
broad spectrum of broadleaf weeds (Haagsma 
1975). It is taken up by both foliage and roots, is 
readily translocated in the growing plant, and in- 
duces auxin-like responses similar to the action of 
2,4-D and 2,4,5-T. It is available as a water soluble 
triethylamine salt in Garlon 3 A Herbicide 1 com- 
prising 3 lbs a.e. per gallon. Breakdown in the soil 
depends on microbial activity when temperature 
and moisture are favorable. In the laboratory, 50 
percent breakdown occurs in 10 to 46 days at 35C. 
Field trials elsewhere show the breakdown rate is 
intermediate between that of 2,4,5-T and picloram 
(Haagsma 1975). Toxicity to birds and fish is very 
low, and moderate to animals. 

The objectives of this study were threefold: (1) 
compare rate-response curves of some typical Lake 
States hardwoods to triclopyr, (2) estimate the per- 
sistence and mobility of triclopyr in forest soil, and 

Mention of trade names does not constitute 
endorsement of the products by the USD A Forest 
Service. 



(3) evaluate the safety of triclopyr to white spruce 
during aerial release and for planting after site 
preparation. Aerial release of conifers is not a cur- 
rently registered use of triclopyr. However, forest 
site preparation is a registered use as long as coni- 
fer planting is deferred until six months after 
treatment. 

STUDY AREA 

The study area is located near Grand Rapids, 
Itasca County, Minnesota. The topography is roll- 
ing and soils are loamy but variable, ranging from 
sand to silt to clay. Soil drainage ranges from 
poorly-to well-drained. Mean July temperature 
and April to September rainfall are 66° F and 18 
inches, respectively. The dominant vegetation is 
pole-size northern hardwood forest (table 1). Ac- 
cording to Kuchler (1964) the potential natural 
vegetation is maple-basswood forest, indicating 
relatively high soil fertility. Some of the forest has 
been converted to white spruce [Picea glauca 
(Moench.) Voss] plantations where quaking aspen 
suckers comprise 86 percent of competitors and no 
other species makes up more than 5 percent. 

TREATMENTS AND 
METHODS 

Triclopyr, alone at three rates and in combina- 
tion with 2,4-D and Tordon 101 Mixture 2 Herbi- 
cide, was compared with Tordon 101 Mixture as a 
standard in six 5-acre site preparation treatments 
(table 2). Triclopyr was also compared to a stan- 
dard conifer release formulation of 1:1 mixture of 



2 The Dow Chemical Company. Contains 2 lbs. 
2,4-D + .54 lbs. a.e. picloram per gallon as triiso- 
propanol-amine salts. 



Table 1. — Average basal area and DBH for sample trees for six site preparation treatments 



Species 



Common name 



Scientific name 



Stand 
basal area 1 



Sample tree 

quadratic 
mean d.b.h. 



White birch 
Red maple 
Quaking Aspen 
Northern red oak 
American basswood 
Sugar maple 
Others 2 

Total or Average 



Betula papyrifera Marsh. 

Acer rubrum L. 

Populus tremuloides Michx. 

Quercus rubra L 

Tilia americana L 

Acer saccharum Marsh. 



ft 2 lacre 


inches 


15.8 


7.2 


12.3 


5.5 


36.5 


9.5 


17.7 


8.1 


3.9 


7.0 


16.2 


3.8 


4.2 


6.7 



106.6 



7.7 



1 Also mean number of trees sampled per treatment. 

2 Balsam poplar (Populus balsamifera L), green ash (Fraxinus Pennsylvania Marsh), American elm (Ulmus americana L), eastern hophornbeam (Ostrya 
virginiana (Mill.) K. Koch.), yellow birch (Betula alleghaniensis Britton). 



Table 2. — Rates of application of various chemicals 
for different treatments 



Chemical 



Treatment 


Triclopyr 1 


2,4-D 


Picloram 2 


2,4,5-P Total 




. . . Rate, lbs 


a.e. per acre . . . 


Site Preparation 4 










1 


1.5 


— 


— 


— 1.5 


2 


3.0 


— 


— 


— 3.0 


3 


4.5 


— 


— 


— 4.5 


4 


1.5 


5 4.0 


— 


— 5.5 


5 


1.5 


2 2.0 


0.5 


— 4.0 


6 


— 


2 2.0 


0.5 


— 2.5 


Spruce Release 6 










7 


0.375 


— 


— 


— 0.375 


8 


0.75 


— 


— 


— 0.75 


9 


1.125 


— 


— 


— 1.125 


10 


— 


3 1.5 


— 


1.5 3.0 



'As Garlon 3A Herbicide. 

2 As Tordon 101 Mixture. 

3 As Esteron Brush Killer. 

4 ln 8 or 10 gallons water mix per acre; 5 acres per treatment. 

s As Esteron 99 Concentrate. 

6 ln 4 gallons water mix per acre; 10 acres per treatment. 



2,4-D and 2,4,5-T 3 (table 2) on four 10-acre areas. 
All areas were sprayed on the morning of July 28, 
1975, by a helicopter equipped with a conventional 
boom fitted with "raindrop" nozzles for drift con- 
trol. Temperatures that afternoon and for the next 
3 days reached the low 90's. Calculated soil mois- 
ture (Thornthwaite and Mather 1955) on the 28th 
was 8.93 inches for the upper 6.7 feet of soil profile, 
or 74 percent of field capacity. The GDD sum 
[growing degree days, °F = V2 (daily maximum + 
daily minimum) - 40; positive values summed 
over period of interest] for the 1975 season had 
reached 2073° F. Soil moisture and GDD were cal- 
culated using standard weather station data (U.S. 
Department of Commerce 1975) averaged for 
Pokegama Dam and Remer #2 located 020°, 12 
miles; and 250°, 11.5 miles, respectively, from the 
study site. Neither soil moisture nor GDD sum (an 
index of plant maturity) were considered limiting, 
but the high temperatures during and after spray- 
ing may have increased herbicide volatility or re- 
duced herbicide activity. Winds during spraying 
did not exceed 5 mph. 

On April 27, 1976, 100 white spruce 2-2 stock 
were planted in each of the six site preparation 
treatments. On May 11 and 12, 1976, forty 0.8- 
inch diameter soil cores were extracted for a bioas- 
say for residual herbicides on a systematic grid 
from each of site preparation treatments 1, 2, 3, 



3 Esteron Brush Killer, the Dow Chemical Com- 
pany. Contains 2 lbs. per gallon each of 2,4-D and 
2,4,5-T propylene glycol butyl ether esters. 



and 6 (table 2) and from an adjacent untreated 
area. The soil cores were separated into 0- to 4-inch 
and 4- to 12-inch strata and randomly combined 
into four composite samples of ten per stratum. 
This method is believed to provide estimates of 10 
percent, 95 percent confidence for most soil proper- 
ties (Alban 1974), including, presumably, those 
that would influence herbicide behavior in the 
soil. The samples were stored at 38° F until a 
greenhouse bioassay similar to Leasure's (1964) 
was begun in June. Soybeans were planted in the 
herbicide treatment samples and in control 
samples spiked with a standard series of 0, 10, 100, 
and 1,000 ppb of either triclopyr or picloram (as 
Tordon 101 Mixture). After 22 days, plants were 
compared visually, harvested, ovendried, and 
weighed. Residual herbicide in treatment samples 
was assayed by interpolating from dose-response 
curves (fig. 1) developed from the standard series 
plant dry weights. The assay was confirmed by the 
visual comparisons. 



100 - 



« 



One year after treatment 10 sample points were 
systematically located down the center of each of 
the six site preparation treatments. A 10-basal 
area factor prism was used to select an average of 
11 sample trees at each point for dbh measure- 
ments, crown kill estimates to the nearest 10 per- 
cent, and presence of sprouting. In each of the four 
conifer release treatments, 40 white spruce were 
systematically selected and rated for release effec- 
tiveness based on estimated number of years be- 
fore need for another release (0 = immediate need, 
20 = 1 year ... 100 = 5 or more years), and her- 
bicide injury to the spruce. 



RESULTS AND DISCUSSION 

Bioassay. — The dose-response curves (fig. 1) were 
constructed from standard series aerial weights of 
soybeans for triclopyr, and the entire plant for 
picloram. Soybeans weighed more when grown 



80 



-Q 
O 

E 

3 
| 

"5 

E 



». 60 - 



c 
« 

if 

« 

Q. 



O 

i 

Q 

Ui 

O 



40 - 



20 - 



- 






















V TRICLOPYR 










N. PICLORAM 




- 




N. TRICLOPYR 








- 


a 


\ PICLORAM 




b 




- 


4- TO 12-INCH 
STRATUM 






0- TO 4-INCH 
STRATUM 




- 


| 


I 




1 


' 1 



10 



100 



1,000 1 



10 



100 



1,000 



SOIL HERBICIDE CONCENTRATION (ppb) 



Figure 1. — Dose-response curves for soybeans grown in control soils spiked with 0, 10, 100, and 
1,000 ppb triclopyr or picloram (as acid equivalents). In the figure, ppb is assigned a value ofl 
ppb to accommodate the logarithmic scale. The peak response in (a) is estimated from treatment 
sample data (see text). 



with 10 ppb triclopyr than when grown without 
herbicide in both the 0- to 4-inch and 4- to 12-inch 
strata. Furthermore, plants grown in the 4-to 12- 
inch strata from both the triclopyr and picloram 
treatments exceeded the weight of 10 ppb plants. 
Apparently, soybean growth is stimulated by both 
herbicides at low concentrations. Accordingly, the 
dose-response curves were modified by extrapolat- 
ing to the peak response found in treatment soils. 
In the future, standard series soil dosages should 
extend into the 1-10 ppb range and perhaps even 
lower. 

The soybean bioassay recovered less than 10 
percent of the triclopyr applied for site preparation 
and about 13 percent of the picloram, and practi- 
cally all herbicide was found in the 0- to 4-inch 
stratum (table 3). Even with 16.4 inches of precipi- 
tation between herbicide application and soil sam- 
pling (U.S. Department of Commerce 1975, 1976), 
both triclopyr and picloram strongly resisted 
leaching. 

It cannot be interpreted that the unrecovered 
herbicide was broken down entirely by microbio- 
logical activity because the amount of herbicide 
actually reaching the forest floor is unknown. 
Other losses include volatilization, photo-degra- 
dation, drift, and vegetation absorption and up- 
take, none of which were measured here. However, 
triclopyr was nearly as persistent as picloram af- 
ter 2012 GDD between application and sampling. 



This is 54 percent of 1975 seasonal GDD at the 
study site. 

The percent recovery of triclopyr varied in- 
versely with rate applied (table 3). This anomaly 
has been noted before for picloram (Perala 1974). 
One possible explanation is that low application 
rates are insufficient for the rapid adaptation and 
multiplication of soil micro-organisms responsible 
for herbicide degradation. 

Survival and growth were good in the planted 
white spruce in the site preparation treatments 
and no detectable herbicide injury was found. In 
light of the soybean bioassay, the retention of her- 
bicide in the 0- to 4-inch soil stratum effectively 
prevented herbicide contact with seedling roots. 

Site preparation. — Top-kill increased with rate 
of triclopyr applied and varied considerably by 
species (fig. 2). White birch was easily top-killed as 
was red maple at the highest rate. Extrapolation 
to rates of 6 lbs a.e. per acre suggest good top-kill of 
red oak and basswood as well. Quaking aspen and 
especially sugar maple were much more resistant. 

Triclopyr top-killed white birch and red maple 
better than did Tordon 101 Mixture, but the latter 
was superior on quaking aspen, basswood, and 
sugar maple (table 4). When triclopyr was applied 
in combination with 2,4-D or Tordon 101 Mixture, 
top-kill was either only slightly changed, or in the 



Table 3. — Recovery of triclopyr and picloram from four site preparation treatments by soybean bioassay 



0- to 4-inch STRATUM 




Soil weight 


Maximum possible 


Concentration 




Treatment 


per acre 


herbicide concentration 


recovered 1 


Recovery 




lb x 10 6 


ppb 


ppb 


Percent 


1 (Triclopyr) 


0.711 


2 2,110 


200 


9.5 


2 (Triclopyr) 


.745 


4,030 


220 


5.5 


3 (Triclopyr) 


.768 


5,860 


140 


2.4 


6 (Picloram) 


.745 


670 


86 


12.8 


4- to 12-inch STRATUM 


1 


2.049 


3 660 


4 


0.6 


2 


2.212 


1,290 


4 


0.3 


3 


2.196 


2,000 


4 


0.2 


6 


2.275 


190 


3 


0.5 



'Concentration recovered is based on the ovendry weight of the total plant for picloram and on aerial parts only for triclopyr. Plants harvested 22 days after 

sowing. 

2 Figures in this column for the 0- to 4-inch stratum assume all herbicide reached the soil surface and remained in the 0- to 4-inch stratum. 
3 Figures in this column for the 4- to 12-inch stratum assume herbicide not recovered in the 0- to 4-inch stratum leached unaltered to, and accumulated in, the 

4- to1 2-inch stratum. 



100 



80 



60 



40 



20 



WHITE BIRCH # 



RED MAPLE 




12 3 4 5 6 

TRICLOPYR APPLIED (lbs. a.e./acre) 

Figure 2. — Hardwood top-kill response curves to 
triclopyr applied at 1.5, 3.0, and 4.5 lbs a.e. per 
acre. Dashed lines are extrapolated trends. 



Table 4. — Top-kill of species in site preparation 
tests 4, 5, and 6 



Topkill 



Species 



Triclopyr + 
Triclopyr Tordon 101 Tordon 101 
+ 2,4-D Mixture Mixture 







. Percent 




White birch 


100 (+4) 1 


69 (-27) 


61 (-31) 


Red Maple 


57 (-37) 


18 (-70) 


35 (-34) 


Quaking aspen 


50 (+2) 


53 (+17) 


61 (+38) 


Northern red oak 


84 (-4) 


50 (-16) 


53 (+9) 


American basswood 


i 2 


58 (+13) 


100 (+88) 


Sugar maple 


8 (-12) 


21 (+1) 


73 (+53) 



'Numbers in parentheses indicate departures from the triclopyr rate- 
response curves in figure 2 on an equivalent total herbicide rate basis. 
2 i= Insufficient observations. 



case of red maple and white birch (especially with 
Tordon 101 Mixture), greatly reduced. Thus, little 
or no synergistic action of triclopyr in combination 
with 2,4-D or picloram is evident. Rather there 
seems to be antagonistic or blocking action with 
strong species interaction. 



A plot of sprouting percent over top-kill for tri- 
clopyr (fig. 3) indicates the root-kill character of 
the herbicide. Triclopyr tended to root-kill bass- 
wood, sugar maple, and red oak when top-kill was 
high. White birch sprouting was controlled moder- 
ately well but sprouting of red maple was poorly 
controlled and suckering of aspen not at all. 




TRICLOPYR TOP KILL (percent) 

Figure 3. — Hardwood sprouting response curves 
over increasing top-kill by triclopyr. Dashed lines 
are extrapolated trends. 



Sprouting of these species in treatments 4, 5, 
and 6 was compared to the sprouting expected if 
sprouting followed the curves in fig. 3. Tordon 101 
Mixture (either alone or with triclopyr) controlled 
sprouting as well or better than triclopyr alone, 
especially aspen suckering (table 5). 

Spruce release. — Triclopyr effectively released 
white spruce at a rate of lVs lbs a.e. per acre (fig. 4). 
Effectiveness was equal to that of 2,4-D + 2,4,5-T 
at 3 lbs per acre. Extrapolation of the effective- 
ness-rate response curve suggests that rates of 1.5 
- 2.0 lbs a.e. per acre would exceed 90 percent 
effectiveness. It seems unlikely that this increased 
rate would significantly injure white spruce since 
absolutely no herbicide injury was found at all. 



Table 5. — Sprouting of species in site preparation 
tests 4,5, and 6 







Sprouting 








Triclopyr + 






Triclopyr 


Tordon 101 Tordon 101 


Species 


+ 2,4-D 


Mixture 


Mixture 






. . Percent 




White birch 


59 (+31) 1 


20 (+2) 


18 (+1) 


Red maple 


25 (-14) 


14 (-4) 


(-25) 


Quaking aspen 


91 (-3) 


64 (-31) 


47 (-47) 


Northern red oak 


13(0) 


6 (-12) 


(-16) 


American basswood i 2 


(-23) 


0(0) 


Sugar maple 


13 (+8) 


(-26) 


4 (-6) 



1 Numbers in parentheses indicate departures from the triclopyr sprouting 
curves in figure 3 on an equivalent top-kill basis. 
2 i= Insufficient observations. 



100 - 



RATE (lbs. a.e./acre) 

Figure 4. — Rate response curves showing the effec- 
tiveness of conifer release from competing vegeta- 
tion (primarily aspen suckers) in a white spruce 
plantation sprayed with triclopyr at rates of 
0.375, 0.75, and 1.125 lbs a.e. per acre. Circles = 
triclopyr; dashed line - extrapolated trend; 
square = 2,4-D + 2,4,5-T at 3 lbs a.e. per acre. 



CONCLUSIONS 

1. Triclopyr and picloram are unlikely to be 
leached out of intact forest floor and upper 
soil strata high in organic matter. 




2. Triclopyr and picloram degradation rates in 
forest floor do not differ greatly during at 
least the first half-season (measured in GDD) 
after application. 

3. White spruce bare-root stock may be safely 
planted where the forest floor is undisturbed 
the next spring after site preparation with 
either triclopyr or picloram. 

4. White birch and red maple are readily top- 
killed with 4.5 lbs a.e. per acre triclopyr. 
Rates of 6 lbs per acre would likely top-kill 
red oak and basswood as well. 

5. Sugar maple and quaking aspen are more 
readily top-killed with Tordon 101 Mixture 
than with triclopyr. 

6. Triclopyr mixtures with 2,4-D or Tordon 101 
Mixture are not as effective as triclopyr or 
Tordon 101 Mixture alone at the same total 
rate. 

7. Compared to triclopyr, Tordon 101 Mixture 
reduced sprouting. 

8. Rates of 1-2 lbs a.e. triclopyr are highly effec- 
tive and safe for release of white spruce when 
sprayed at about 2100 seasonal GDD (2-3 
weeks after shoot growth is completed). 

9. Further study should test forestry uses of 
triclopyr at higher rates than used in this 
study. 



LITERATURE CITED 

Alban, D. H. 1974. Soil variation and sampling 
intensity under red pine and aspen in Minneso- 
ta. U.S. Dep. Agric. For. Serv., Res. Pap. NC- 
106, 10 p. U.S. Dep. Agric. For. Serv., North 
Cent. For. Exp. Stn., St. Paul, MN. 

Haagsma, T. 1975. DOWCO 233 Herbicide— a pos- 
sible new tool in vegetation management. Ind. 
Veg. Manage. 7(2): 13-15. 

Kuchler, A. W. 1964. Potential natural vegetation 
of the conterminous United States. Am. Geo- 
graphical Soc. Spec. Pub. 36, 155 p. and map. 

Leasure, J. K. 1964. Bioassay method for 4-amino- 
3,5,6-trichloropicolinic acid. Weeds 12:232-233. 

Perala, D. A. 1971. Controlling hazel, aspen suck- 
ers, and mountain maple with picloram. U.S. 
Dep. Agric. For. Serv., Res. Note NC-129, 4 p. 



U.S. Dep. Agric. For. Serv., North Cent. For. 
Exp. Stn., St. Paul, MN. 

Perala, D. A. 1974. Some invert formulations of 
picloram to prepare brushy sites for conversion 
to conifers. Ind. Veg. Manage. 6(1): 18-20. 

Perala, D. A., and C. S. Williams. 1970. Site prepa- 
ration for conifers using herbicides and subse- 
quent burning in a northern Minnesota 
hardwood stand. Down to Earth 26(3):5-8. 



Thornthwaite, C. W., and J. R. Mather. 1955. The 
water balance. Drexel Inst. Lab. of Climatology, 
Pub. in Climatology 8, 104 p. 

U.S. Department of Commerce. 1975, 1976. Clima- 
tological Data, Minnesota. National Oceanic 
and Atmospheric Admin. Environmental Data 
Serv., National Climatic Center, Asheville, NC. 



PESTICIDE 

PRECAUTIONARY 

STATEMENT 

This publication reports research involving pes- 
ticides. It does not contain recommendations for 
their use, nor does it imply that the uses discussed 
here have been registered. All uses of pesticides 
must be registered by appropriate State and/or 
Federal agencies before they can be recommended. 

CAUTION: Pesticides can be injurious to humans, 
domestic animals, desirable plants, and fish or 
other wildlife — if they are not handled or applied 
properly. Use all pesticides selectively and careful- 
ly. Follow recommended practices for the disposal 
of surplus pesticides and pesticide containers. 




FOLLOW IHI LAIIL 



*U.S. GOVERNMENT PRINTING OFFICE: 1 980--668420/1 64 



Perala, Donald A. 

1979. A provisional assessment of triclopyr herbicide for use in Lake 
States' forestry. U.S. Dep. Agric. For. Serv., Res. Pap. NC-180, 7 p. 
U.S. Dep. Agric. For. Serv., North Cent. For. Exp. Stn., St. Paul, MN. 

Compares rate-response curves of some typical Lake States 
hardwoods to tricolpyr. Estimates the persistance and mobility of 
triclopyr in forest soil. Evaluates the safety of triclopyr to white 
spruce. 



OXFORD: 231.324. KEY WORDS: Northern hardwoods, herbicide 
degradation, picloram, 2,4-D,2,4,5-T. 



Perala, Donald A. 

1979. A provisional assessment of triclopyr herbicide for use in Lake 
States' forestry. U.S. Dep. Agric. For. Serv., Res. Pap. NC-180, 7 p. 
U.S. Dep. Agric. For. Serv., North Cent. For. Exp. Stn., St. Paul, MN. 

Compares rate-response curves of some typical Lake States 
hardwoods to tricolpyr. Estimates the persistance and mobility of 
triclopyr in forest soil. Evaluates the safety of triclopyr to white 
spruce. 



OXFORD: 231.324. KEY WORDS: Northern hardwoods, herbicide 
degradation, picloram, 2,4-D,2,4,5-T. 




Toot youn. kootoji on pollation no\n\ 



USDA FOREST SERVICE 
RESEARCH PAPER NC-181 




GOVT. DOCUMENTS 
DEPO Y ITEM 

APR 17 1S80 
CLEMSON 

library: 



Twenty-year 

results of 

the 

Lake States 

JACK PINE 

seed source 
study 



Richard M. Jeffers 
Raymond A. Jensen 



North Central Forest Experiment Station 
Forest Service, U.S. Department of Agriculture 



CONTENTS 

Page 

Literature review 1 

Methods 2 

Results and discussion 6 

Seed collection recommendations 18 

Seed production areas 19 

Literature cited 19 

Acknowledgments 20 



North Central Forest Experiment Station 

Robert A. Hann, Director 

Forest Service - U.S. Department of Agriculture 

1992 Folwell Avenue 

St. Paul, Minnesota 55108 

Manuscript approved for publication September 18, 1978 

1980 



TWENTY- YEAR RESULTS OF THE LAKE STATES 
JACK PINE SEED SOURCE STUDY 



Richard M. Jeffers, Plant Geneticist, 

Rhinelander, Wisconsin 

(currently with the Rocky Mountain Region, 

Lakewood, Colorado) 

and Raymond A. Jensen, Associate Scientist, 

University of Minnesota, Cloquet Forestry Center, 

Cloquet, Minnesota 



Jack pine (Pinus banksiana Lamb.) is an impor- 
tant pulpwood species in the Lake States. It grows 
under a wide variety of environmental conditions 
and exhibits considerable phenotypic variation 
both within and among stands. Knowledge of ge- 
netic variation in this species is basic to a jack pine 
tree improvement program for the Lake States. 

In 1951 the North Central (then the Lake 
States) Forest Experiment Station and the Uni- 
versity of Minnesota began a regional seed source 
study to determine the nature and extent of genet- 
ic variation in jack pine. Private, State, and 
Federal forestry agencies collected seed from 29 
natural stands in Minnesota, Wisconsin, and 
Michigan. Seedlings produced from those seed col- 
lections were planted at 17 locations in the three 
Lake States and at a single location in Ontario. 

This paper includes data on survival, total 
height, diameter, volume per tree, and volume per 
acre 20 years after planting at 14 locations in the 
Lake States. 



LITERATURE REVIEW 

Results of plantings at one or more locations 
have already been published by various coopera- 
tors. 

Stoeckeler and Rudolf (1956) reported on height 
growth at age 2 and change in needle color in fall 
at ages 1,2, and 3 at the Hugo Sauer Nursery at 
Rhinelander, Wisconsin. Changes in fall needle 
color were found to be associated with latitude of 
seed source (the more northerly the seed source the 
more purplish the needles). Height growth of 2-0 



stock was negatively correlated with latitude of 
seed source and positively correlated with normal 
annual sum of average daily temperatures of 50°F 
or above (growing degree days) at seed origin. 

Arend et al. ( 1961 ) reported on 5-year results in 
three Lower Michigan plantations. Tree height 
differences among seed sources were significant at 
all three locations. 

Survival and height growth 5 years after plant- 
ing were also reported at six locations in 
Minnesota by Jensen et al. (1960), but statistical 
analyses were not included. 

Aim et al. (1966) found highly signifcant differ- 
ences among seed sources in height and diameter 
growth after 9 years in a plantation at the Univer- 
sity of Minnesota Cloquet Forestry Center in Carl- 
ton County, Minnesota. 

King (1966) reported on 10-year height growth 
of trees from 26 sources common to 1 1 plantings in 
Minnesota, Wisconsin, and Michigan and from a 
local commercial seed source (different at each 
location). At 10 locations there were significant 
differences in total tree height among test seed 
sources. At almost all locations test seed sources 
from nearest the planting sites outgrew the com- 
mercial nursery stock. King concluded that selec- 
tion of "good" jack pine stands for seed collection 
appears worthwhile in the species, but selection of 
tested nonlocal stock may offer even greater 
improvement. 

Utilizing the data reported by King (1966) and 
data from the Ontario planting, Morgenstem and 
Teich (1969) reported increases in height were ob- 
tained by planting trees 2 to 3° north of their seed 
origin. 



Rudolph (1964) described variation among trees 
from different seed sources in lammas growth and 
prolepsis in four Minnesota and two Wisconsin 
plantings. He found significant differences in 
these traits among 30 seed sources in the Cloquet 
plantation and significant differences among 10 
widely distributed seed sources evaluated at all 
locations. 

Variation among seed sources in susceptibility 
to several insects and diseases has been found in 
several of the plantings. Batzer (1961, 1962) found 
significant differences among sources in incidence 
of white pine weevil (Pissodes strobi (Peck)) in two 
northern Minnesota plantings. 

Arend et al. (1961) evaluated three plantings in 
Lower Michigan after 5 years and found signifi- 
cant differences among seed sources in incidence of 
white pine weevil, bark beetles {Pityophthorus 
spp.), and redheaded pine sawfly (Neodiprion le- 
contei (Fitch)), but no significant differences 
among sources in eastern gall rust (Cronartium 
quercuum (Berk.) Miy. ex Shirai) incidence. 

King (1971) reported on variation among seed 
sources in incidence of white pine weevil, eastern 
pineshoot borer (Eucosma gloriola Heinrich), and 
eastern gall rust at 11 locations in Minnesota, 
Wisconsin, and Michigan. He found significant 
differences among sources in white pine weevil 
incidence, and concluded that Lower Michigan 
sources are the best to use as a starting point in a 
white pine weevil resistance breeding program. 

Eastern pineshoot borer was found in eight of 
the 11 plantations evaluated after 5 years, but no 
significant differences among sources in borer in- 
cidence could be shown. However, after 10 growing 
seasons in the field, there were significant differ- 
ences among seed sources in borer incidence in two 
Wisconsin and one Upper Michigan plantings. 

King found eastern gall rust after 10 years in 
every plantation he examined. In seven planta- 
tions in which more than 15 percent of the trees 
were infected, differences among seed sources in 
gall rust incidence were significant. 

Trees from the northernmost sources showed 
high rust incidence while those from southern 
sources had the lowest rust incidence. King sug- 
gested that trees from the sourthern portion of the 
range have been subjected to more intense gall 
rust infection, and have developed some resistance 



to it, while those from farther north, where the 
alternate hosts are not as abundant, have not been 
subjected to as severe a selection for resistance. 

King and Nienstaedt (1965) described variation 
among seed sources in needlecast (Dauisomycella 
(Hypodermella) ampla (Dav.) Dank) incidence in a 
western Upper Michigan and a southern Wiscon- 
sin plantation. They found significant differences 
among sources in susceptibility to the needlecast 
fungus at both locations, indicating that suscepti- 
bility to this disease is under direct genetic 
control. 

Information published to date has given consid- 
erable insight into the development of this study. 
Significant differences among seed sources have 
been found for several traits, including height and 
diameter growth, lammas growth and prolepsis, 
and pest incidence. Variation among seed sources 
in most traits appears to be clinal, and individual 
genotypes react strongly to different test environ- 
ments. Northern and southern sources contribute 
most to this interaction, while middle-latitude 
sources contribute little. 

Yeatman (1974) showed that tree height at 5 
years was ineffective in predicting tree height at 
19 years for 12 Ontario seed sources grown at the 
Petawawa Forest Experiment Station. Variation 
in tree height at 1 1 years, however, was very effec- 
tive in predicting height at 19 years and provided a 
sound basis for selection of the better seed sources. 
King's (1966) data and our data on Lake States 
jack pine substantiate Yeatman's results. 



METHODS 

Seed Sources and 
Plantation Establishment 

In 1951 and early 1952 seed was collected from 
29 jack pine stands in Minnesota, Wisconsin, and 
Michigan. Each collection was made from domi- 
nant and codominant trees in a stand considered 
good for its locality. 

Seed from all 29 stand collections was sown in 
the spring of 1952 in both the General Andrews 
State Nursery at Willow River, Minnesota, and in 
the Hugo Sauer State Nursery at Rhinelander, 
Wisconsin. 



Two-year-old seedlings were planted by nine 
cooperating agencies at 17 locations, one in the fall 
of 1953 and 16 in the spring of 1954. Seedlings 
produced at the Genral Andrews Nursery were 
used to establish six plantations in Minnesota and 
two plantations in western Wisconsin. Seedlings 
produced at the Hugo Sauer Nursery were used to 
establish four additional plantings in Wisconsin 
and five plantations in Michigan. 

At each location a randomized complete-block 
design with four replications was used. Each 
source was represented in each replication by a 
square 64-tree plot with trees planted at a spacing 
of 5 x 5 feet. In addition to the selected seed 
sources, each replication contained one seedlot of 
stock supplied by a commercial nursery in the 
same area as the test plantation. Very little is 
known about the origin of the commercial stock in 
many instances. 

Because of shortages of seedlings from a few 
sources, substitutions were made at several loca- 
tions. Data for 26 seed sources (table 1, fig. 1) 
common to 14 locations 1 (fig. 1, table 2) are pre- 
sented here. 



Measurements 

Survival, height, and diameter 

In the fall of 1973, 20 years after establishment, 
the trees were evaluated for survival, total height, 
and diameter. Survival data are based on the 64 
trees originally planted in each plot. 

To save time and expense in tree measurement, 
a statistical sample was selected from each planta- 
tion. The sample was based on variances obtained 
with the 10-year measurements of height and 
d.b.h. plus an assumption that means and their 
standard errors would increase about in the same 
proportion between 10 and 20 years, which proved 
valid. We first determined that measuring trees on 
all four replications was the best procedure. To 
determine number of trees to be measured per plot, 
an estimate of the standard error of a source mean 
was plotted against number of trees measured per 
plot for both height and d.b.h. The standard error 
increased slowly as the number of trees measured 
decreased from 64 to about 15, and increased more 



Table 1. — Jack pine seed source origins 1 







MINNESOTA 














Growing 


Average 


Seed 








degree 


January 


source 


County 


Latitude 


Longitude 


days 2 temperature 






°N 


°W 




°F 


1589 


Cass 


47.4 


94.4 


9,200 


5 


1590 


Cass 


47.0 


94.6 


9,400 


7 


1591 


Itasca 


47.5 


94.1 


9,100 


5 


1592 


Lake 


47.7 


91.2 


7,400 


10 


1593 


Cook 


48.0 


90.3 


6,700 


14 


1594 


St. Louis 


48.1 


92.4 


8,500 


5 


1595 


Pine 


46.0 


92.6 


9,500 


10 


1596 


Pine 


46.4 


92.8 


9,000 


9 


1597 


Becker 


47.1 


95.4 


8,900 


4 


1600 


Cass 


46.8 


94.4 


9,400 


6 


1601 


Beltrami 


47.5 


95.0 


8,600 


5 


1602 


Itasca 


47.8 


93.3 


8,800 


6 


WISCONSIN 


1605 


Bayfield 


46.7 


91.0 


9,000 


13 


1606 


Forest 


46.0 


88.9 


8,500 


12 


1608 


Burnett 3 


45.9 


92.1 


10,000 


10 


1609 


Marinette 


45.2 


88.3 


9,600 


14 


1610 


Oneida 


45.8 


89.8 


9,000 


10 


1611 


Wood 


444 


89.7 


10,000 


13 


MICHIGAN (Upper Peninsula) 


1612 


Gogebic 


46.2 


89.2 


8,500 


12 


1613 


Ontonagon 


46.6 


89.0 


8,800 


15 


1614 


Alger 


46.3 


86.7 


8,100 


15 


1615 


Chippewa 


46.3 


84.8 


8,000 


15 


1621 


Luce 


46.6 


85.4 


7,900 


16 


MICHIGAN (Lower Peninsula) 


1616 


Manistee 


44.2 


86.2 


10,100 


22 


1617 


Ogemaw 


44.2 


84.1 


9,600 


19 


1618 


Alpena 


45.0 


83.5 


9,000 


19 



l One source planted at 13 locations only. 



'Data from Stoeckeler and Rudolf (1956). 

2 Normal annual sum of average daily temperatures of 50°F or above. 

3 Not planted at University of Minnesota CFC (Plantation 6). 



rapidly as numbers were reduced further. The pro- 
portional increase in standard error between mea- 
surement of 15 and eight trees per plot was 17 
percent for height and 24 percent for d.b.h., and 
the decrease in number of trees measured was 47 
percent. The standard error increased more rapid- 
ly as sample size decreased further. Taking costs 
and estimated standard errors into account, we 




Figure 1. — Location of seed sources and plantations included in 20-year measurements. Shaded 
area shows natural range of jack pine (Rudolf and Schoenike 1963). 



decided to measure eight trees per plot. Total 
height was measured to the nearest 0.5 foot and 
diameter to the nearest 0.1 inch. 



Volume 

An equation for estimating volume of a tree 
from its height and d.b.h. was developed for a sub- 
sample of trees, made up of 10 of the 26 sources at 
each of three plantations: Burnett County Forest 
(CF), Argonne Experimental Forest (EF), and Ot- 
tawa National Forest (NF). A single tree with 
height and diameter approximating the average 
height and diameter of the eight measured trees 
was selected from each of the 10 seed sources in 
each of the four replications. Sample trees were cut 
as close to the ground as possible and all limbs 



removed. The stems were cut into eight equal- 
length sections and the volume of wood in each 
section was determined using the volume formula 
for the frustum of a cone. 

By using transformations of the Behre equation 
of tree form (Bruce 1972), we found that although 
there appeared to be some slight differences in 
form among the three plantations, there were no 
apparent differences in form among sources. Con- 
sequently, this prediction equation was used for 
estimating tree volume for all seed sources at all 
locations: 
Vol (cuft) = .1781 + .3361 (d.b.h.) 2 ht - (.01/d.b.h.) 

Total volume per acre was estimated for each 
plot by taking (average volume per tree) x (num- 
ber of trees per acre initially = 1,740) x (propor- 
tion survival). 



Table 2. — Location of jack pine plantations 



MINNESOTA 













Growing 


Average 






Plantation 






degree 


January 


Number Forest 


County 


establishment 


Latitude 


Longitude 


days 


temperature 








°N 


°W 




f 


1 Superior NF 1 


Lake 


USDA For. Serv. 


47.6 


91.1 


7,400 


10 


2 Chippewa NF 


Beltrami 


USDA For. Serv. 


47.4 


94.5 


8,600 


5 


3 PillsburySF 


Cass 


Minn. Cons. Dep. 


46.4 


94.5 


9,400 


6 


5 General Andrews EF 


Pine 


Minn. Cons. Dep. 


46.4 


92.8 


8,700 


9 


6 Univ. Minn. CFC 


Carlton 


Univ. Minn. 


46.7 


92.5 


8,500 


8 


WISCONSIN 


7 Burnett CF 


Burnett 


Burnett Co. 


45.6 


92.8 


10,000 


10 


8 Mosinee IF 


Washburn 


Mosinee Pap. Co. 


46.2 


92.0 


10,000 


10 


9 Chequamegon NF 


Bayfield 


USDA For. Serv. 


46.3 


91.4 


9,000 


13 


10 Nepco IF 


Wood 


Nekossa-Edwards 
Pap. Co. 


44.2 


89.8 


10,000 


13 


11 Argonne EF 


Forest 


USDA For. Serv. 


45.8 


89.0 


8,500 


12 


12 Marinette CF 


Marinette 


Marinette Co. 


45.7 


88.0 


9,600 


14 


MICHIGAN (Upper Peninsula) 


13 Ottawa NF 


Ontonagon 


USDA For. Serv. 


46.3 


89.2 


8,500 


12 


MICHIGAN (Lower Peninsula) 


15 Univ. Mich. BS 


Emmet 


Univ. Mich. 


45.5 


84.7 


8,700 


17 


17 Fife Lake SF 


Grand 
Traverse 


Mich. Cons. Dep. 


44.5 


85.4 


9,700 


18 



1 NF: National Forest; SF: State Forest; EF: Experimental Forest; CFC: Cloquet 

Statistical Procedures 

Analyses of variance 

Two-way analyses of variance by source and rep- 
lication were performed for survival, height, diam- 
eter, volume per tree, and volume per acre for each 
plantation. Combined analyses of data from all 
locations were also run for these variables. Be- 
cause variances were quite different among plan- 
tations, log transformations of the data were used 
in the combined analyses to bring about homogen- 
eity. Data on volume per acre from plantations at 
Pillsbury State Forest (SF) in Minnesota and 
Nepco Industrial Forest (IF) in Wisconsin were not 
included in the combined analysis because the 
transformation gave logs of negative values. 

From the two-way analysis of variance for each 
plantation, an estimate of the "true" variation 
among sources was obtained as a "component of 
variance" after allowance for random variation 
(Snedecor and Cochran 1967). The corresponding 



Forestry Center; CF: County Forest; IF: Industrial Forest; BS: Biological Station. 
standard deviations, denoted by <r s , represent the 
variability among sources. In tables 3-7, estimated 
ranges among sources (plantation mean ± 2o- s ), 
are shown for each plantation. These ranges would 
include 95 percent of a normal distribution, and 
approximately 95 percent of a distribution with 
moderate deviation from normal. It does not ap- 
pear that the distributions of "true" values for 
sources differ markedly from normal, so that a 
value equal to the plantation mean + 2a s would 
represent a source near the top of the distribution, 
while a value equal to the plantation mean - 2o- s 
would represent a source near the bottom. These 
points are more stable than the maxima and min- 
ima of the observed means and were used to assess 
the potential gain from proper seed source selec- 
tion for a given location. A few sources will have 
values outside this range, so this treatment is con- 
servative. In the case of tree survival, values were 
calculated using the arcsin transformation and 
then transformed back. The range of variation 
among sources is much greater for some planta- 
tions than for others. 



Grouping of sources 

Because mean values for a single source at an 
individual plantation had fairly large standard 
errors, we tried grouping plantations and also 
sources. However, grouping of plantations was not 
very successful, primarily because of large differ- 
ences among plantations. 

We found that sources could be arranged into 
three broad groups based on simple correlation 
coefficients among individual sources over the 14 
plantations for measurements of height, d.b.h., 
and volume per tree. The separation was clearest 
for height, where 83 percent of the correlation 
coefficients between pairs of sources in the same 
group were 0.95 or above, as compared with 11 
percent for sources not in the same group. For 
d.b.h., correlations were lower and less consistent. 
Fifty-two percent of the correlations between pairs 
of sources in the same group were greater than 
0.90, compared with 25 percent for sources in dif- 
ferent groups. Results for volume per tree were 
intermediate, with 82 percent of the correlation 
coefficients in the same group greater than 0.90 
compared with 49 percent for sources in different 
groups. 

A few sources correlated well with both group 
two and group three sources, and could have been 
assigned to either group. A few sources that did not 
fit closely into any group were put into the group 
having the best agreement. 

Correlations 

For each plantation, simple correlations were 
run between (a) average source survival, height, 
diameter, volume per tree, and volume per acre 
and (b) growing degree days and latitude of seed 
origin. Simple correlations among sources were 
also run between heights at 5 and 10 years, 5 and 
20 years, and 10 and 20 years, after planting for 
each of 13 locations where measurements from the 
three periods were available. 



RESULTS AND DISCUSSION 
Survival 

Average tree survival 20 years after plantation 
establishment varied from 53 percent at Pillsbury 
SF in central Minnesota to 92 percent at Fife Lake 



SF in Lower Michigan (table 3). Survival at six 
locations was less than 80 percent. Much of the 
initial mortality at these locations was attributed 
to inadequate cultural treatment after planting, 
and insect and animal damage. Much continuing 
mortality is attributable to reduced tree vigor re- 
sulting from insect and disease incidence (King 
1971, King and Nienstaedt 1965). Continuing 
mortality is most prevalent at locations with high 
incidence of gall rust. Trees with galls on main 
stems are particularly susceptible to breakage 
from high winds, ice, and heavy, wet snows. 

Seed sources showed considerable variation in 
survival at all locations; differences were signifi- 
cant at all locations except the University of 
Michigan Biological Station (BS). In the combined 
analysis, differences among plantations and seed 
source x location interactions were significant. 

Five of eight plantations exhibiting high varia- 
bility among sources in survival had high gall rust 
incidence. The southernmost plantation, Nepco 
IF, showed much greater variation than the other 
plantations. This plantation not only had a high 
incidence of gall rust (King 1971), but also a high 
incidence of needlecast (King and Nienstaedt 
1965) and root tip weevil (Hylobius rhizophagus 
M.B. & W.). At this location, northern Minnesota 
sources had very poor survival and high gall rust 
and needlecast incidence, while sources from the 
southern portion of the species range had good 
survival and relatively low rust and needlecast 
incidence. 

At 10 of the 14 locations, the source from nearest 
the planting site (local) was among the top eight of 
the 26 sources in survival. At 12 locations survival 
of the local source exceeded that of the commercial 
source by 4 to 33 percent. At the remaining two 
locations, University of Minnesota Clouquet For- 
estry Center (CFC) and the University of Michi- 
gan BS, survival of both local and commercial 
sources was between 85 and 90 percent and dif- 
fered by less than 1 percent. Survival of commer- 
cial sources was relatively low, compared with 
other sources, at most locations. 

Survival was related to similarity between cli- 
mate and length of growing season at seed origin 
and at the plantation location. In the seven planta- 
tions having cooler climates and shorter growing 
seasons (growing degree days of 8,700 or less), best 
survival was attained by trees from Minnesota 



Table 3. — Survival percent of jack pine seed sources 
MINNESOTA 





1 


2 


3 


5 
General 


6 
Univ. 


7 


8 


9 


10 


11 


12 


13 


15 
Univ. 


17 
Fife 


Seed 


Superior 


Chippewa Pillsbury Andrews 


Minn. 


Burnett 


Mosinee Chequamegon 


Nepco Argonne Marinette Ottawa Mich. 


Lake 


source 


NF**i 


NF 


SF** 


EF " 


CFC * 


CF** 


IF ** 


NF ** 


IF '* 


EF ** 


CF** 
86 


NF** 

77 
89 


BS 

90 
95 


SF** 


Mean 


82 


81 


53 


85 


85 


85 


79 


63 


65 


68 


92 


Range 2 


93 


87 


68 


92 


90 


94 


95 


77 


89 


79 


93 


96 




67 


76 


38 


78 


81 


76 


61 


50 


37 


55 


79 


65 


86 


88 














MINNESOTA 
















1589 


84 


83 


58 


91 


87 


85 


82 


67 


52 


71 


91 


83 


90 


95 


1590 


73 


88 


60 


78 


81 


38 


85 


57 . 


59 


71 


91 


78 


94 


92 


1591 


88 


81 


60 


88 


91 


86 


82 


73 


43 


65 


89 


79 


85 


91 


1592 


3 91 


79 


47 


88 


89 


71 


64 


46 


45 


69 


79 


86 


91 


93 


1593 


92 


88 


52 


90 


88 


70 


74 


57 


34 


72 


90 


88 


93 


96 


1594 


89 


82 


48 


83 


84 


78 


6? 


50 


41 


70 


73 


77 


89 


89 


1595 


78 


80 


55 


88 


90 


88 


88 


71 


70 


67 


86 


82 


90 


91 


1596 


82 


86 


68 


3 90 


3 86 


89 


88 


71 


74 


73 


86 


74 


91 


94 


1597 


77 


85 


56 


85 


90 


84 


81 


59 


63 


67 


84 


74 


89 


95 


1600 


85 


87 


3 62 


84 


86 


88 


91 


68 


54 


69 


88 


77 


91 


93 


1601 


79 


3 85 


61 


89 


86 


89 


80 


68 


60 


68 


87 


70 


•93 


95 


1602 


81 


83 


43 


88 


84 


79 


59 


52 


46 


77 


77 


81 


85 


89 


WISCONSIN 


1605 


81 


81 


64 


84 


84 


84 


89 


3 72 


66 


71 


85 


73 


86 


89 


1606 


76 


74 


27 


78 


79 


81 


71 


55 


59 


3 69 


78 


75 


80 


83 


1608 


82 


84 


58 


84 


— 


3 89 


3 90 


82 


70 


66 


91 


82 


93 


95 


1609 


68 


82 


53 


86 


75 


93 


86 


63 


80 


60 


3 86 


71 


87 


91 


1610 


88 


79 


39 


89 


83 


88 


74 


62 


86 


66 


91 


67 


96 


91 


1611 


76 


82 


64 


73 


86 


91 


94 


61 


3 83 


48 


88 


57 


92 


95 


MICHIGAN (Upper Peninsula) 


1612 


85 


78 


32 


84 


90 


84 


76 


68 


67 


68 


87 


3 85 


88 


94 


1613 


84 


77 


54 


86 


89 


88 


76 


57 


60 


68 


86 


84 


88 


93 


1614 


91 


85 


54 


85 


84 


84 


77 


71 


71 


73 


80 


78 


87 


90 


1615 


90 


80 


55 


89 


89 


88 


74 


59 


75 


81 


90 


82 


91 


90 


1621 


87 


72 


52 


91 


86 


87 


70 


65 


76 


75 


89 


89 


96 


89 


MICHIGAN (Lower Peninsula) 


1616 


73 


71 


45 


84 


87 


89 


86 


71 


89 


60 


86 


70 


93 


3 94 


1617 


65 


83 


51 


73 


75 


91 


85 


56 


81 


53 


91 


66 


86 


95 


1618 


78 


75 


59 


88 


84 


84 


82 


67 


74 


59 


86 


79 


3 88 


91 


COMMERCIAL SEED SOURCE 




87 


73 


52 


76 


87 


65 


71 


50 


50 


64 


79 


78 


89 


89 



'Significant differences among seed sources: * = 5 percent; 
Estimated range (plantation mean ± 2& s ). 
3 Seed source nearest planting site (local). 



1 percent. 



and Upper Michigan. The majority of these trees 
was from areas having less than 9,000 growing 
degree days. 

Conversely, at the other seven locations (9,000 
growing degree days or more) best survival was 
attained by trees from areas having warmer cli- 
mates and longer growing seasons than at the 
planting site. Included in this group were sources 
1596 and 1600 from Minnesota, sources 1608, 
1609, 1611 from Wisconsin, and sources 1616 and 
1617 from Lower Michigan. Predominating 
among the sources with the poorest survival at 
these locations were the four northernmost 
sources— 1592, 1593, 1594, and 1602 from Minne- 
sota — and source 1606 from northern Wisconsin. 
Source 1606, from the coldest site in Wisconsin, 
was among the sources showing lowest survival at 
10 of the 14 locations. 

High incidence of gall rust at several of the 
warmer locations, including the Pillsbury SF, 
Burnett CF, Mosinee IF, Chequamegon NF, and 
Nepco IF (King 1971), probably resulted in in- 
creased mortality. Furthermore, since seed 
sources vary in their susceptibility to gall rust, the 
high incidence at these locations could have 
caused the increased variation among seed sources 
in survival. In general, seed sources with the high- 
est rust incidence had the poorest survival. 



Height 

Average tree height in the plantations ranged 
from 18.9 feet at the University of Michigan BS in 
northern Lower Michigan to 33.3 feet at the Pills- 
bury SF in central Minnesota (table 4). There were 
significant differences in tree height among 
sources at all locations except the University of 
Minnesota CFC; the combined analysis showed 
that differences among plantations and seed 
source x location interaction were significant. 

The estimated superiority of a source at the top 
of the distribution (plantation mean + 2& s ) com- 
pared with one at the bottom (plantation mean - 
2& s ), as a percent of plantation mean, varied from 
10 percent at the University of Minnesota CFC to 
49 percent at Mosinee IF. In general, variation 
among sources within plantations increased with 
the number of growing degree days at the planting 
site, but variation was not significantly correlated 



with latitude of plantation or site quality (based on 
average plantation height). The greatest variation 
among seed sources in tree height occurred at loca- 
tions with high incidence of gall rust; sources from 
northern Minnesota had the slowest growth and 
highest gall rust incidence, while sources from the 
southern portion of the species range in Minne- 
sota, Wisconsin, and Michigan had the greatest 
height growth and lowest rust incidence. 

In general, the "best" seed sources were local 
ones and those within 100 miles south of the plant- 
ing site. The number of growing degree days for 
these sources was nearly equal to or greater than 
the number at the planting site. Trees from the 
local source exceeded the average plantation tree 
height by 2 to 19 percent at all locations; the local 
source was among the seven tallest sources at 13 
locations. At 11 locations, the local source ex- 
ceeded the commercial source by 4 to 13 percent. 

Except at the Superior NF, where northeastern 
Minnesota sources were the best, one or more of 
the three Lower Michigan sources were among the 
five tallest sources at all locations. At eight plan- 
tations, one of the Lower Michigan sources was the 
tallest. Other sources producing the tallest trees at 
several locations were 1595, 1596, 1600, and 
1601 from Minnesota, and 1610 from northern 
Wisconsin. 

Six sources produced the shortest trees at sev- 
eral locations: 1592, 1593, 1594, and 1602 from 
northeast Minnesota (the four northernmost 
sources), source 1606 from northern Wisconsin, 
and source 1621 from Upper Michigan. All six of 
these sources have less then 8,900 growing degree 
days. 



Diameter 

Average plantation tree diameter varied from 
3.15 inches at the Mosinee IF to 4.75 inches at the 
Pillsbury SF (table 5). Average tree diameter for a 
source was significantly and positively correlated 
with average tree height for the source at each 
location. 

A combined analysis with data from all loca- 
tions showed significant tree diameter differences 
among sources and plantations, and a significant 
seed source x plantation interaction. 



Table 4.— Average height of trees from jack pine seed sources as a percent of the plantation mean 





1 


2 


3 


5 
General 


6 
Univ. 


7 8 


9 


10 


it 


12 


13 


15 17 
Univ. Fife 


Seed 


Superior 


Chippewa Pillsbury Andrews 


Minn. 


Burnett Mosinee Chequamegon 


Nepco Argonne Marinette Ottawa Mich. Lake 


source 


NF**i 


NF** 


SF** 


EF** 


CFC 


CF** IF** 


NF** 


IF** 


EF** 


CF** 


NF** 


BS** SF** 


Mean (ft) 


23.3 


23.9 


33.3 


27.5 


29.4 


25.9 23.5 


27.2 


23.8 


30.1 


29.2 


29.3 
31.8 


18.9 19.4 


Range 2 (ft) 


24.8 


25.6 


37.3 


30.0 


30.8 


27.8 29.3 


30.3 


28.7 


32.0 


31.8 


21.0 21.1 




21.8 


22.2 


27.3 


25.0 


280 


21.6 17.7 


24.1 


18.9 


28.2 


26.7 


26.8 


16.7 16.7 














MINNESOTA 














1589 


103 


99 


100 


109 


99 


99 106 


104 


96 


99 


106 


102 


103 104 


1590 


100 


102 


102 


100 


99 


100 106 


99 


95 


101 


102 


102 


102 102 


1591 


100 


100 


103 


101 


102 


98 101 


99 


98 


90 


94 


98 


97 95 


1592 


3 104 


102 


99 


97 


100 


87 71 


87 


91 


100 


92 


88 


98 97 


1593 


103 


95 


89 


84 


95 


86 79 


95 


83 


98 


93 


93 


94 93 


1594 


101 


91 


89 


92 


94 


79 71 


86 


81 


99 


91 


92 


90 87 


1595 


92 


104 


104 


109 


99 


109 116 


104 


107 


94 


101 


99 


100 105 


1596 


101 


105 


102 


3 104 


3 105 


105 108 


103 


103 


102 


98 


101 


101 102 


1597 


105 


105 


104 


103 


100 


103 107 


100 


98 


104 


100 


100 


101 99 


1600 


99 


105 


3 109 


103 


101 


103 113 


106 


94 


102 


101 


101 


99 99 


1601 


106 


3 107 


105 


98 


102 


105 104 


103 


102 


104 


101 


100 


106 102 


1602 


106 


95 


90 


97 


100 


89 83 


92 


89 


101 


92 


100 


94 95 


WISCONSIN 


1605 


94 


102 


103 


95 


94 


99 101 


3 102 


95 


101 


102 


104 


98 96 


1606 


101 


93 


82 


94 


88 


90 92 


96 


95 


3 102 


100 


100 


90 96 


1608 


101 


100 


107 


102 


— 


3 1 10 3 109 


106 


109 


94 


101 


101 


103 103 


1609 


96 


101 


104 


101 


97 


107 111 


101 


116 


100 


3 103 


101 


101 99 


1610 


104 


105 


101 


105 


106 


104 104 


104 


111 


103 


104 


105 


106 105 


1611 


94 


95 


104 


106 


97 


107 115 


97 


3 113 


89 


99 


91 


104 98 


MICHIGAN (Upper Peninsula) 


1612 


99 


104 


100 


99 


106 


105 101 


104 


99 


101 


100 


3 105 


98 100 


1613 


104 


101 


102 


104 


102 


96 95 


99 


90 


106 


105 


103 


96 98 


1614 


97 


98 


98 


97 


103 


100 98 


94 


98 


98 


99 


97 


92 89 


1615 


99 


97 


96 


98 


100 


97 91 


98 


92 


102 


98 


102 


96 97 


1621 


96 


91 


94 


96 


101 


93 92 


96 


95 


97 


92 


97 


96 89 


MICHIGAN (Lower Peninsula) 


1616 


98 


99 


106 


105 


102 


116 117 


109 


122 


102 


107 


105 


115 3 1 19 


1617 


98 


101 


100 


102 


101 


114 106 


105 


118 


102 


112 


104 


114 111 


1618 


99 


105 


108 


99 


103 


104 106 


112 


111 


106 


105 


109 


3 108 115 


COMMERCIAL SEED SOURCE 




107 


96 


104 


94 


105 


97 97 


97 


102 


98 


98 


101 


111 113 



'Significant differences among seed sources: 
Estimated range (plantation mean ± 2tf s ). 
3 Seed source nearest planting site (local) 



1 percent level. 



Table 5. — Average diameter of trees from jack pine seed sources as a percent of the plantation mean 





1 


2 


3 


5 
General 


6 
Univ. 


7 


8 


9 


10 


11 


12 


13 


15 17 
Univ. Fife 


Seed 


Superior 


Chippewa Pillsbury 


Andrews 


Minn. 


Burnett 


Mosinee Chequamegon 


Nepco Argonne 


Marinette 


Ottawa 


Mich. Lake 


source 


NF 


NF 


SF •' 


EF 


CFC 


CF** 


IF *' 


NF ' 


IF** 


EF ** 


CF* 


NF** 


BS* SF** 


Mean (in) 


3.77 


3.59 


4.75 


3.65 


3.88 


3.47 


3.15 


4.16 


3.48 


4.46 


3.99 


4.01 


3.27 3.31 


Range 2 (in) 


3.95 


3.83 


5.37 


3.88 


4.03 


3.75 


3.65 


4.56 


3.82 


4.90 


4.30 


4.43 


3.50 3.62 




3.59 


2.98 


4.13 


3.42 


3.73 


3.19 


2.65 


3.76 


3.14 


4.02 


3.68 


3.59 


3.04 3.00 


MINNESOTA 


1589 


106 


94 


92 


105 


93 


98 


103 


105 


100 


102 


100 


103 


.107 104 


1590 


104 


97 


93 


101 


100 


96 


100 


105 


91 


93 


103 


97 


100 100 


1591 


100 


98 


103 


101 


98 


95 


96 


92 


102 


90 


95 


94 


95 91 


1592 


3 100 


105 


108 


97 


99 


93 


77 


91 


96 


95 


96 


82 


96 92 


1593 


99 


90 


80 


85 


96 


95 


88 


94 


88 


97 


91 


86 


102 102 


1594 


97 


92 


89 


94 


92 


90 


78 


83 


95 


91 


96 


92 


90 88 


1595 


93 


109 


109 


116 


99 


106 


113 


104 


103 


102 


104 


100 


104 110 


1596 


99 


105 


95 


3 99 


3 105 


99 


102 


97 


98 


93 


103 


103 


105 101 


1597 


99 


101 


104 


100 


95 


99 


101 


103 


97 


105 


100 


99 


96 103 


1600 


100 


101 


3 110 


103 


101 


97 


108 


108 


92 


97 


107 


102 


98 95 


1601 


105 


3 105 


97 


101 


99 


100 


104 


101 


103 


104 


101 


104 


106 97 


1602 


106 


92 


105 


98 


106 


92 


87 


96 


97 


93 


98 


101 


99 99 


WISCONSIN 


1605 


95 


103 


96 


96 


99 


101 


96 


3 102 


99 


96 


105 


106 


104 101 


1606 


101 


106 


95 


97 


90 


98 


104 


102 


108 


3 105 


103 


104 


91 105 


1608 


101 


98 


107 


97 


— 


3 108 


3 104 


95 


107 


95 


96 


91 


104 102 


1609 


99 


100 


98 


100 


112 


105 


109 


104 


103 


107 


3 104 


99 


97 97 


1610 


100 


105 


109 


99 


109 


97 


99 


99 


101 


104 


98 


103 


102 104 


1611 


105 


104 


107 


113 


101 


105 


105 


100 


3 102 


98 


98 


104 


98 98 










MICHIGAN (Upper Peninsula) 












1612 


100 


109 


109 


99 


106 


106 


105 


112 


103 


105 


94 


3 107 


102 99 


1613 


101 


96 


102 


103 


99 


99 


95 


94 


94 


103 


105 


102 


97 99 


1614 


101 


95 


90 


100 


99 


101 


102 


91 


101 


91 


99 


92 


95 94 


1615 


93 


103 


92 


103 


94 


98 


99 


100 


96 


102 


97 


103 


98 98 


1621 


94 


95 


88 


98 


103 


95 


102 


99 


95 


99 


86 


96 


98 96 


MICHIGAN (Lower Peninsula) 


1616 


98 


99 


113 


100 


97 


109 


110 


97 


111 


113 


112 


109 


109 3 109 


1617 


109 


96 


101 


105 


109 


112 


103 


116 


113 


113 


111 


114 


110 105 


1618 


93 


105 


107 


95 


100 


101 


104 


107 


105 


112 


101 


106 


3 100 115 










COMMERCIAL SEED SOURCE 














103 


94 


106 


102 


111 


103 


97 


107 


112 


102 


102 


103 


100 107 



'Significant differences among seed sources: * = 5 percent; 
Estimated range (plantation mean ± 2<r s ). 
3 Seed source nearest planting site (local). 



= 1 percent. 



10 



There was a significant negative correlation be- 
tween average plantation diameter and survival 
(r = -0.696). Thus, considering all 14 plantings as 
a whole, high mortality resulted in reduced compe- 
tition and, hence, greater growth. The relation 
among sources within plantations was much more 
complex. D.b.h. -survival correlations were signifi- 
cant at six locations; they were negative at the 
University of Minnesota CFC, Argonne EF, and 
Ottawa NF, but positive at the Burnett CF, Mosi- 
nee IF, and Nepco IF. 

Seed sources varied significantly in tree diame- 
ter at all locations in Michigan and Wisconsin, but 
at only one location in Minnesota, the Pillsbury SF 
(table 5). Variation among sources in diameter 
was significantly and negatively correlated with 
average plantation survival, but was not corre- 
lated with growing degree days, latitude, or site 
quality of plantation. Variation among sources 
was greatest at the Pillsbury SF and least at the 
Superior NF. 

The average of the coefficients of variation 
among sources over all plantations was signifi- 
cantly less for diameter than for height. This sub- 
stantiates Funk's (1975) conclusion for white pine 
that height is a better indication of genetic differ- 
ences than diameter. 

Superiority of trees from local sources was not as 
evident for diameter as for height. Local sources 
ranked among the top eight of the 26 sources at 
only eight locations. They did, however, equal or 
exceed the average tree diameter of all sources at 
all locations. 

In general, trees from the Lower Michigan 
sources had the greatest average diameter; at 
least one of these sources was among the best 
sources at 10 of the locations, and all three Lower 
Michigan sources were among the top five sources 
at the Nepco IF and Argonne EF in Wisconsin, and 
at the Ottawa NF and Fife Lake SF in Michigan. 
Source 1595 from Pine County, Minnesota, was 
among the top five sources at three Minnesota and 
two northwest Wisconsin locations. And source 
1612 from western Upper Michigan was among 
the five best sources at three locations in Minne- 
sota, three in Wisconsin, and at the Upper Michi- 
gan location. 

Sources 1595 and 1618. which were among the 
best in diameter growth at several locations, were 
among the poorest at the Superior NF. And except 



at the Superior NF, sources 1592, 1593, and 1594 
from northeastern Minnesota were among the five 
poorest sources at 8, 9, and 11 locations, respec- 
tively. Also among the five poorest sources at 
six locations were 1614 and 1621 from Upper 
Michigan. 



Volume 



Volume per tree 



Average volume per tree ranged from 0.67 cubic 
feet at the University of Michigan BS to 2.09 cubic 
feet at the Pillsbury SF (table 6). Sources varied 
significantly in volume per tree at all locations 
except the University of Minnesota CFC, and the 
combined anlysis showed significant differences 
among plantations and a significant seed source x 
location interaction. 

Frequently, volume per tree may be affected by 
stand survival. However, in this study correla- 
tions between average seed source volume per tree 
and survival were significant at only four loca- 
tions. The correlations between these traits were 
positive at the Burnett CF, Mosinee IF, and Nepco 
IF, but negative at the Ottawa NF. Because vol- 
ume per tree is a function of diameter squared, the 
relations between survival and diameter at these 
locations also explain the relations between sur- 
vival and volume per tree. 

The estimated superiority of a source at the top 
of the distribution (plantation mean + 2<x s ) com- 
pared with one at the bottom (plantation mean - 
2& s ), as a percent of the plantation mean, varied 
from 13 percent at the University of Minnesota 
CFC to 94 percent at the Burnett CF. 

Variation among sources within plantations in 
volume per tree was greater than the variation 
among sources in survival, height, and diameter. 
Variation among sources in this trait was not sig- 
nificantly correlated with growing degree days or 
latitude at the planting site. Unusually large vari- 
ation among sources in volume per tree occurred at 
the Pillsbury SF in Minnesota. Not surprisingly, 
variation among sources in diameter was also 
greatest at this location. Variation among sources 
was relatively high at all locations in Wisconsin 
and in western Upper Michigan, while relatively 
low at the remaining four locations in Minnesota 
and at both locations in Lower Michigan. In gen- 
eral, variation among sources in volume per 



11 



Table 6. — Average volume per tree of trees from jack pine seed sources as a percent of plantation mean 





1 


2 


3 


5 
General 


6 
Univ. 


7 


8 


9 


10 


11 


12 


13 


15 
Univ. 


17 
Fife 


Seed 


Superior 


Chippewa Pillsbury 


Andrews 


Minn. 


Burnett 


Mosinee Chequamegon 


Nepco Argonne 


Marinette 


Ottawa 


Mich. 


Lake 


source 


NF 


NF* 


SF** 


EF * 


CFC 


CF** 


IF** 


NF** 


IF** 


EF** 


CF** 


NF** 


BS 


SF** 


Mean (cu. 


ft) .99 
j. ft) 1.10 


.94 


2.09 


1.09 


1.28 


.95 


.75 


1.36 


.89 


1.69 


1.34 


1.36 


.67 


.69 


Range 2 (ci 


1.08 


2.75 


1.29 


1.36 


1.57 


1.05 


1.70 


1.18 


2.04 


1.62 


1.67 


.80 


.87 




.88 


.80 


1.43 


.89 


1.20 


.68 


.45 


1.02 


.60 


1.34 


1.06 


1.05 


.54 


.51 


MINNESOTA 


1589 


112 


89 


85 


116 


88 


95 


108 


112 


97 


101 


104 


105 


113 


108 


1590 


107 


96 


89 


101 


99 


93 


102 


109 


81 


90 


106 


96 


101 


99 


1591 


99 


98 


106 


101 


100 


91 


94 


85 


100 


77 


87 


87 


90 


83 


1592 


3 103 


111 


113 


90 


98 


79 


52 


74 


86 


90 


86 


65 


91 


85 


1593 


101 


78 


59 


65 


88 


81 


68 


85 


70 


92 


80 


72 


97 


97 


1594 


95 


79 


73 


83 


81 


69 


53 


64 


78 


84 


85 


82 


78 


73 


1595 


82 


120 


120 


137 


96 


117 


136 


109 


110 


97 


108 


100 


105 


120 


1596 


101 


112 


91 


3 103 


3 114 


101 


108 


95 


99 


88 


102 


104 


107 


103 


1597 


101 


104 


111 


101 


92 


99 


106 


107 


92 


111 


99 


99 


94 


102 


1600 


100 


105 


3 127 


109 


103 


98 


122 


120 


82 


95 


113 


104 


97 


91 


1601 


115 


3 115 


97 


100 


100 


103 


108 


104 


105 


111 


102 


106 


115 


95 


1602 


116 


82 


101 


94 


110 


79 


68 


87 


87 


86 


90 


101 


93 


93 


WISCONSIN 


1605 


87 


108 


95 


89 


93 


98 


91 


3 104 


93 


94 


109 


113 


104 


97 


1606 


103 


103 


75 


38 


77 


91 


99 


101 


107 


3 110 


105 


105 


79 


104 


1608 


103 


96 


119 


95 


— 


3-122 


3 112 


95 


122 


86 


93 


84 


108 


105 


1609 


94 


101 


97 


99 


121 


115 


122 


107 


116 


111 


3 108 


99 


95 


95 


1610 


104 


113 


117 


103 


121 


97 


99 


99 


110 


108 


99 


109 


106 


110 


1611 


104 


101 


114 


128 


98 


114 


120 


95 


3 113 


86 


95 


98 


98 


95 


MICHIGAN (Upper Peninsula) 


1612 


99 


118 


116 


96 


115 


114 


108 


126 


103 


111 


91 


3 116 


101 


97 


1613 


106 


94 


104 


108 


101 


94 


87 


89 


82 


110 


114 


105 


92 


96 


1614 


98 


89 


81 


99 


99 


101 


102 


80 


99 


83 


97 


83 


86 


82 


1615 


87 


101 


82 


101 


89 


93 


90 


97 


86 


104 


92 


108 


93 


94 


1621 


87 


83 


74 


92 


104 


87 


95 


93 


86 


94 


71 


89 


94 


86 


MICHIGAN (Lower Peninsula) 


1616 


94 


97 


132 


103 


96 


130 


129 


101 


141 


125 


128 


120 


128 


3 131 


1617 


113 


92 


101 


110 


116 


137 


108 


136 


139 


127 


131 


132 


128 


117 


1618 


89 


114 


121 


90 


101 


104 


111 


123 


120 


130 


106 


118 


3 105 


143 










COMMERCIAL SEED SOURCE 
















112 


38 


113 


101 


125 


101 


91 


108 


124 


101 


100 


105 


108 


124 



'Significant differences among seed sources: * = 5 percent; ** = 1 percent. 
Estimated range (plantation mean ± 2& s ). 
3 Seed source nearest planting site (local). 



12 



tree was greater at locations with high gall rust 
incidence. 

The local source exceeded the plantation aver- 
age by 3 to 31 percent at all locations and the 
commercial source by 2 to 27 percent at nine loca- 
tions. Because volume per tree was determined 
from the diameter squared and height, ranking of 
sources for volume per tree was similar to that for 
diameter. 

Volume per acre 

In terms of productivity, volume per acre is the 
most important trait we studied. Average volume 
per acre ranged from 1,041 cubic feet at the Uni- 
versity of Michigan BS in Lower Michigan to 2,012 
cubic feet at Marinette CF in northeastern Wis- 
consin (table 7). The best plantations, Pillsbury SF 
and the University of Minnesota CFC in Minne- 
sota, and Argonne EF and Marinette CF in Wis- 
consin, produced 1,899 to 2,012 cubic feet of wood 
per acre. At each of these locations, average height 
exceeded 29 feet and average diameter was 
greater than 3.8 inches. The poorest plantations, 
Mosinee IF and Nepco IF in Wisconsin, and the 
University of Michigan BS and Fife Lake SF in 
Michigan, however, produced only 1,041 to 1,112 
cubic feet of wood per acre. Average height at these 
locations was less than 24 feet while average di- 
ameter was less than 3.5 inches. In addition to poor 
site quality, growth at the Mosinee IF and Nepco 
IF was probably also affected by high incidence of 
gall rust and/or other pests. 

A combined analysis with data from all loca- 
tions showed significant differences in this trait 
among sources and plantations, and a significant 
seed source x location interation. 

Except at the University of Michigan BS, corre- 
lations between average seed source volume per 
acre and height were greater than those between 
volume per acre and diameter. 

Differences in volume per acre among seed 
sources were considerable at all locations, and sta- 
tistically significant at 11 locations; variation 
among sources was not significant at the Chip- 
pewa NF, University of Minnesota CFC, and Ar- 
gonne EF (table 7). The least variation among seed 
sources in volume per acre occurred at the Univer- 
sity of Minnesota CFC, while the greatest varia- 
tion among sources occurred at the Nepco IF. High 
variability among sources also occurred at the 



other plantations having high gall rust incidence, 
including the Pillsbury SF, Burnett CF, Mosinee 
IF, and Chequamegon NF. The estimated superi- 
ority of a source at the top of the distribution (plan- 
tation mean + 2& s ) compared with one at the 
bottom (plantation mean - 2d s ), as a percent of the 
plantation mean, was 136 percent at the Nepco IF. 
Differences of this magnitude indicate that tre- 
mendous losses in volume production may occur if 
the wrong seed sources are used in a planting 
program. 

Local sources yielded the greatest volume per 
acre at the Superior NF, Chippewa NF, and Pills- 
bury SF in Minnesota, and at the Ottawa NF in 
Upper Michigan. Local sources ranked among the 
10 best sources at all locations. Local sources ex- 
ceeded the commercial sources at 11 locations, 
ranging from 15 percent better at the Fife Lake SF 
to 51 percent better at the Burnett CF. However, 
the commercial source produced more wood per 
acre than the local source at the Superior NF, at 
the University of Minnesota CFC, and at the Uni- 
versity of Michigan BS. 

The best seed sources included 1595 and 1600 
from Minnesota and 1616, 1617, and 1618 from 
Lower Michigan. The four northeastern Minne- 
sota sources— 1592, 1593, 1594, and 1602— and 
source 1606 from a cold location in Forest County, 
Wiscosnin, were the poorest sources overall. How- 
ever, at the northernmost plantation, Superior 
NF, sources 1592, 1593 and 1602 were among the 
best, while sources 1595, 1616 and 1618 were 
among the worst. 

Grouping of Seed Sources 

As indicated previously, the sources could be 
arranged into three broad groups based on simple 
correlation coefficients among individual sources 
over the 14 plantations for height, diameter, and 
volume per tree. Correlations among sources 
within each group were high whereas correlations 
among sources in different groups were lower. It 
turned out that this division of sources coincided 
with a geographical distribution into a northern 
group (Group 1 sources), central group (Group 2 
sources), and southern group (Group 3 sources) 
(table 8, fig. 1). They follow in a general way the 
more detailed seed collection zones developed by 
Rudolf (1956). The results from this study support 
the concept of geographic seed zones in jack pine; 



13 



Table 7. — Average volume per acre of trees from jack pine seed sources as a percent of plantation mean 





1 


2 


3 


5 
General 


6 
Univ. 


7 


8 


9 


10 


11 


12 


13 


15 17 
Univ. Fife 


Seed 


Superior 


Chippewa Pillsbury Andrews 


Minn. 


Burnett 


Mosinee Chequamegon 


Nepco Argonne 


Marinette 


Ottawa 


Mich. Lake 


source 


NF** 1 


NF 


SF** 


EF* 


CFC 


CF** 


IF " 


NF" 


IF** 


EF** 


CF** 


NF** 


BS SF** 


Mean (cu 


ft) 1408 


1330 


1899 


1616 


1900 


1420 


1061 


1493 


1045 


1968 


2012 


1814 


1041 1112 


Range 2 (cu. ft) 1660 


1520 


2777 


1912 


3 1900 


1955 


1654 


2086 


1755 


2287 


2509 


2142 


1277 1427 




1156 


1140 


1021 


1320 


1900 


885 


468 


900 


335 


1649 


1515 


1486 


805 797 


MINNESOTA 


1589 


114 


91 


97 


125 


91 


94 


1.11 


119 


79 


106 


110 


114 


113 111 


1590 


96 


102 


100 


92 


96 


94 


106 


92 


73 


94 


112 


97 


105 100 


1591 


106 


98 


123 


105 


106 


91 


95 


99 


66 


72 


89 


90 


86 82 


1592 


4 115 


106 


105 


94 


102 


65 


41 


55 


56 


92 


79 


73 


92 86 


1593 


114 


85 


59 


68 


90 


68 


61 


75 


38 


99 


84 


83 


101 101 


1594 


104 


80 


66 


80 


80 


62 


41 


51 


50 


87 


72 


83 


78 71 


1595 


79 


119 


128 


142 


102 


120 


149 


120 


118 


98 


108 


108 


105 119 


1596 


101 


118 


111 


4 107 


4 115 


105 


117 


108 


113 


96 


101 


101 


109 105 


1597 


95 


109 


117 


100 


97 


97 


106 


95 


88 


111 


97 


97 


94 106 


1600 


104 


113 


4 148 


107 


100 


101 


138 


129 


66 


97 


115 


103 


99 92 


1601 


111 


4 120 


113 


105 


101 


106 


106 


113 


93 


111 


103 


97 


120 98 


1602 


115 


84 


79 


97 


109 


73 


50 


71 


63 


100 


83 


109 


88 90 


WISCONSIN 


1605 


86 


108 


111 


88 


92 


96 


100 


4 118 


95 


102 


108 


108 


100 93 


1606 


96 


94 


38 


80 


73 


86 


88 


81 


93 


4 112 


95 


103 


70 93 


1608 


103 


100 


132 


93 


— 


4 126 


4 125 


123 


130 


86 


98 


90 


112 107 


1609 


78 


102 


98 


100 


105 


125 


129 


106 


140 


98 


4 108 


89 


91 93 


1610 


112 


110 


87 


108 


117 


100 


90 


98 


139 


105 


105 


93 


114 108 


1611 


97 


102 


139 


109 


99 


121 


139 


92 


4 140 


63 


98 


71 


101 98 


MICHIGAN (Upper Peninsula) 


1612 


103 


115 


67 


94 


122 


111 


102 


135 


102 


112 


91 


4 129 


99 98 


1613 


110 


90 


105 


110 


105 


96 


82 


80 


75 


110 


115 


114 


90 97 


1614 


111 


94 


81 


99 


98 


99 


95 


87 


107 


91 


91 


86 


83 80 


1615 


96 


99 


84 


106 


93 


94 


82 


90 


95 


126 


95 


115 


95 92 


1621 


93 


74 


72 


99 


106 


87 


83 


96 


97 


105 


74 


104 


101 82 


MICHIGAN (Lower Peninsula) 


1616 


84 


86 


111 


103 


98 


135 


137 


113 


187 


113 


127 


110 


132 4 134 


1617 


90 


93 


94 


95 


104 


145 


113 


121 


168 


100 


137 


113 


122 121 


1618 


85 


106 


135 


94 


100 


101 


112 


132 


128 


114 


106 


121 


4 103 141 


COMMERCIAL SEED SOURCE 




119 


79 


110 


91 


129 


75 


79 


87 


94 


94 


92 


107 


107 119 



'Significant differences among seed sources: 
Estimated range (plantation mean ± 2& s ). 
Estimate of & s is in this case. 
4 Seed source nearest planting site (local). 



5 percent; 



1 percent. 



14 



Table 8. — Growing degree days and January temperature of seed sources within groups 





Group 1 






Group 2 






Group 3 






Growing 






Growing 






Growing 




Seed 


degree 


January 


Seed 


degree 


January 


Seed 


degree 


January 


source 


days 


temp. 


source 


days 


temp. 


source 


days 


temp. 






°F 






°F 






°F 


1592 


7,400 


10 


1589 


9,200 


5 


1595 


9,500 


10 


1593 


6,700 


14 


1590 


9,400 


7 


1608 


10,000 


10 


1594 


8,500 


5 


1591 


9,100 


5 


1609 


9,600 


14 


1602 


8,800 


6 


1596 


9,000 


9 


1611 


10,000 


13 


1606 


8,500 


12 


1597 


8,900 


4 


1616 


10,100 


22 


Average 


7,980 


9.4 


1600 


9,400 


6 


1617 


9,600 


19 








1601 


8,600 


5 


Average 


9,800 


14.7 








1605 


9,000 


13 














1610 


9,000 


10 














1612 


8,500 


12 














1613 


8,800 


15 














1614 


8,100 


15 














1615 


8,000 


15 














1618 


9,000 


19 














1621 


7,900 


16 














Average 


8,790 


10.4 









based on our data, however, the large number of 
subdivisions proposed by Rudolf appears to be un- 
warranted for jack pine. 

Group 1 corresponds to Rudolfs collection zones 
5 and 6. This group includes the four northeastern 
Minnesota sources, 1592, 1593, 1594, and 1602, 
and source 1606 from Forest County, Wisconsin. 
Although source 1606 does not belong to this group 
geographically, it does belong to it climatically. 
Forest County, Wisconsin, is poorly covered by 
weather stations and many sites in the area are 
colder than the weather records indicate. There- 
fore, this seed source is probably characterized by 
a climate with fewer growing degree days than 
shown in table 1. Group 1 sources represent the 
most severe climate, with an average of 7,980 
growing degree days and an average January tem- 
perature of 9.4 C F. 

Group 3 corresponds to Rudolfs milder zones 2 
and 3. Sources in this group include 1595, 1608, 
1609, 1611, 1616, and 1617. These are the sources 
from the southern portion (mildest climate) of the 
species range in Minnesota, Wisconsin, and Michi- 
gan. Sources 1600, 1610, and 1618 and possibly 



others appear to be borderline between Groups 2 
and 3. Group 3 sources have an average of 9,800 
growing degree days and an average January tem- 
perature of 14.7°F. 

Group 2, which includes the remaining 15 
sources, corresponds to Rudolfs broad seed collec- 
tion zone 4. Group 2 sources have an average of 
8,790 growing degree days and an average Janu- 
ary temperature of 10.4°F. Variation among 
Group 2 sources in these climatic variables is 
considerable 

It is obvious that genetic variation in Lake 
States jack pine is continuous, expressing adapta- 
tion to climatic and other environmental factors. 
In some cases, the species shows adaptation to 
local conditions — source 1606, for example. 
Grouping of sources was done to bring out broad 
patterns of genetic variation. 

Table 9 enables comparison of seed source group 
means for height, diameter, volume per tree, and 
volume per acre. In each section, column 1 gives 
the mean value for Group 1; columns 2 and 3 show 
the ratios of the other two group means to the 



15 



Table 9. — Comparison of seed source groups for height, diameter, and volume 







Height (Group) 


Diameter (Group) 


Volume/tree (Group ) 


Volume/acre (Group) 




Plantation 1 


1 


2 


3 


1 


2 


3 


1 


2 


3 


1 


2 


3 








Percent of 




Percent of 




Percent of 




Percent of 






Ft. 


group 1 


In. 


grouf. 


) 1 


Ft. 3 


group 


/ 


Ft. 3 


group 1 


1 


Superior NF 


2 24.0 


98 


94 


3.80 


99 


100 


1.029 


96 


95 


2 1532 


93 


81 


11 


Argonne EF 


30.1 


101 


97 


4.29 


104 


109 


1.565 


109 


114 


1931 


105 


95 


2 


Chippewa NF 


2 22.7 


107 


105 


3.48 


104 


104 


.854 


113 


112 


1193 


115 


112 


6 


Univ. Minn. CFC 


2 28.0 


107 


105 


3.75 


103 


106 


1.166 


112 


115 


1723 


113 


111 


13 


Ottawa NF 


2 27.7 


108 


106 


2 3.74 


108 


110 


2 1.160 


121 


124 


1635 


116 


107 


17 


Fife Lake SF 


2 18.2 


106 


113 


3.22 


102 


107 


.629 


109 


122 


2 982 


112 


127 


12 


Marinette CF 


2 27.4 


107 


111 


3.86 


103 


108 


2 1.197 


112 


124 


2 1662 


122 


136 


15 


Univ. Mich. BS 


2 17.6 


107 


114 


2 3.12 


105 


108 


2 584 


114 


126 


2 891 


118 


129 


5 


General Andrews EF 


2 25.5 


109 


113 


2 3.43 


106 


112 


2 .919 


120 


133 


2 1355 


122 


127 


9 


Chequamegon NF 


2 24.8 


111 


114 


2 3.88 


108 


110 


2 1.122 


125 


131 


2 995 


159 


169 


3 


Pillsbury SF 


2 29.9 


113 


116 


4.53 


104 


111 


2 1.757 


119 


135 


2 1320 


149 


168 


7 


Burnett CF 


2 22.3 


117 


128 


2 3.25 


106 


115 


2 .758 


122 


153 


2 1006 


139 


182 


10 


Nepco IF 


2 20.8 


113 


131 


2 3.38 


101 


109 


2 .760 


112 


145 


2 628 


157 


245 


8 


Mosinee IF 
Average 


2 18.6 


129 


142 


2 2.74 


116 


123 


2 .512 


150 


179 


2595 


182 


236 




24.1 


26.3 


27.1 


3.60 


3.78 


3.94 


1.001 


1.156 


1.266 


1246 


1538 


1653 



'Plantations arranged according to magnitude of differences between group means for height and volume per acre. 
2 At least one significant difference exists among groups at 5 percent level. 



mean for Group 1 x 100 percent. The least varia- 
tion among groups occurred at planting sites with 
fewer than 8,700 growing degree days, and the 
greatest variation occurred at sites with 9,000 or 
more growing degree days. Plantings with the 
greatest variation also had high incidences of gall 
rust. The sources in Group 1 had the highest rust 
incidence, while sources 1595, 1608, and 1611 
from Group 3 had the lowest rust incidence (King 
1971). Genetic variation in susceptibility to rust 
probably accentuated the differences among 
groups at locations with high rust incidence, such 
as the Burnett CF, Mosinee IF, Chequamegon NF, 
and Nepco IF. 

Growing Season and Daylength 
at Seed Origin 

To determine the relation between growing sea- 
son or daylength at seed origin and performance 
variables, simple correlations were run for each 
plantation between (a) growing degree days and 
daylength (latitude) at seed origin and (b) surviv- 
al, height, diameter, volume per tree, and volume 
per acre (table 10). Survival at the Superior NF, 
General Andrews EF, Argonne EF, and Ottawa 
NF — four of the coldest locations — was signifi- 
cantly and negatively correlated with growing de- 



gree days, and positively correlated with latitude 
at seed origin. At the Burnett CF, Mosinee EF, and 
Nepco IF — three of the warmest locations — survi- 
val was significantly and positively correlated 
with growing degree days and negatively corre- 
lated with latitude at seed origin. 

Height at the Superior NF, Chippewa NF, Uni- 
versity of Minnesota CFC, and Argonne EF was 
not significantly correlated with growing degree 
days at seed origin. Correlations between these 
variables at all other locations, however, were sig- 
nificant, with coefficients ranging from 0.44 at the 
Ottawa NF to 0.80 at the Mosinee IF. 

Correlations between height and latitude at 
seed origin shifted from a significant positive cor- 
relation (r = 0.54) at the northernmost plantation 
(Superior NF), through nonsignificant correla- 
tions, to increasing negative correlations with de- 
creasing latitude of plantation. These results show 
that the effect of latitude at seed origin on height 
varies with latitude at the planting site (fig. 2). 
Graphs showing equally dramatic shifts could also 
have been drawn for survival and volume per acre. 

At the northernmost plantations, no significant 
correlations were found between seed source diam- 
eter and growing degree days or latitude at seed 



16 



Table 10. — Simple correlation coefficients between variables and (a) degree days or (b) latitude at seed origin 







Growing 




Survival 


Height 


Diameter 


Vol. /tree 


Vol. /acre 






degree 




Degree 




Degree 




Degree 




Degree 




Degree 






Plantation 1 


days 


Lat. 


days 


Lat. 


days 


Lat. 


days 


Lat. 


days 


Lat. 


days 


Lat. 


1 


Superior NF 


7,400 


°N 
47.6 


2-0.69 


0.63 




0.54 










-0.47 


0.64 


2 


Chippewa NF 


8,600 


47.4 




.40 


















6 


Univ. Minn. CFC 


8,500 


46.7 




.41 


















3 


Pillsbury SF 


9,400 


46.4 






.58 




.58 


-.43 


.62 


-.45 


.60 




5 


General Andrews EF 


8,700 


46.4 


-.45 


.50 


.75 


-.45 


.59 


-.44 


.64 


-.42 


.44 




9 


Chequamegon NF 


9,000 


46.3 


.44 




.60 


-.58 




-.47 


.45 


-.52 


.54 


-.49 


13 


Ottawa NF 


8,500 


46.3 


-.60 


.57 


.44 


-.44 


.54 


-.61 


.52 


-.61 






8 


Mosinee IF 


10,000 


46.2 


.69 


-.54 


.80 


-.65 


.60 


-.66 


.71 


-.68 


.76 


-.66 


11 


Argonne EF 


8,500 


45.8 


-.60 


.71 








-.66 




-.56 






12 


Marinette CF 


9,600 


45.7 






.60 


-.67 


.62 


-.50 


.64 


-.59 


.66 


-.62 


7 


Burnett CF 


10,000 


45.6 


.76 


-.63 


.73 


-.77 


,61 


-.82 


.68 


-.83 


.75 


-.84 


15 


Univ. Mich. BS 


8,700 


45.5 






.60 


-.63 






.49 


-.50 


.41 


-.42 


17 


Fife Lake SF 


9,700 


44.5 






.58 


-.64 




-.56 


.44 


-.63 


.45 


-.60 


10 


Nepco IF 


10.000 


44.2 


.52 


-.87 


.71 


-.88 


.55 


-.71 


.66 


-.84 


.64 


-.93 



1 Plantations are arranged from north to south. 

2 0nly correlations significant at the 5 and 1 percent levels are given (Significance levels r 



0.39 — 5 percent, r = 0.50 — 1 percent). 




LATITUDE AT SEED ORIGIN (°N) 

Figure 2. — Linear regression: average height on 
latitude at seed source origin. 



origin. At seven of the remaining locations, the 
correlations between diameter and growing de- 
gree days at seed origin varied from 0.54 to 0.62, 
while the correlations between diameter and lati- 
tude tended to increase negatively with decreas- 
ing latitude of plantation. 

As one might expect, the correlations between 
volume per tree and growing degree days or lati- 
tude were very similar to those between diameter 
and the latter variables. The correlations between 
volume per acre and growing degree days or lati- 
tude at seed origin were similar to those between 
height and growing degree days or latitude at seed 
origin. Only at the Superior NF was the correla- 
tion coefficient significant and negative (r = 
-0.47) for growing degree days at seed origin; at 
nine of the remaining locations the correlations 
between volume per acre and growing degree days 
at seed origin were significant and positive. 

Latitude of seed source appeared to influence 
growth more than growing degree days at seed 
origin in the northernmost and southernmost 
plantations, while growing degree days at seed 
origin appeared to influence growth more than 



17 



latitude of seed source in middle latitude plant- 
ings. The data for height, diameter growth, and 
volume production of trees from 26 seed sources 20 
years after planting add further support to the 
findings of Morgenstern and Teich (1969) for phe- 
notypic stability of height growth (based on height 
growth of 16 of the same sources 10 years after 
planting). Their results and ours show that 
sources from northeastern Minnesota (Group 1 
sources) and from the southern portion of the jack 
pine range in Minnesota, Wisconsin, and Michi- 
gan (Group 3 sources) contribute most to the geno- 
type x environment interactions while those from 
middle latitudes (Group 2 sources) contribute lit- 
tle to the interactions. Morgenstern and Teich sug- 
gested that an apparent reason for these differ- 
ences is the distance from origin of seed to the 
planting site (this distance being the least for mid- 
dle latitude seed sources). 



Comparison of Height Growth at 
5, 10, and 20 Years 

Coefficents of determination (r 2 x 100) between 
5-, 10-, and 20-year mean heights among sources 
at 13 locations are included in table 11. At nine 
locations, 67 to 92 percent of the variation in 
height at 20 years was accounted for by height at 
10 years. The highest coefficients of determination 
between 10 and 20 years were found for locations 
with milder climates. At the coldest locations — 
Superior NF, Chippewa NF, University of Minne- 
sota CFC, and Argonne EF — less than 60 percent 
of the variation in height at 20 years could be 
accounted for by height at 10 years. 

Our results indicate that height at age 5 is not a 
reliable estimate of height at age 20. Less than 67 
percent of the variation in height at age 20 could 
be accounted for by variation in height at age 5. 

The coefficients of determination for height at 
10 and 20 years for nine locations in the Lake 
States are similar to or exceed the coefficients of 
determination for Ontario jack pine heights mea- 
sured at 11 and 19 years at the Petawawa Forest 
Experiment Station in Ontario (Yeatman 1974). 
Therefore, by the time regional jack pine tests are 
10 to 15 years old, they can in most cases be used to 
develop reliable seed source recommendations. 
Exceptions are tests on the coldest sites, where 
final differentiation of response may be delayed. 



Table 11. — Coefficients of determination (r 2 x 100) 
between 5-, 10-, and 20-year mean 
heights among sources 
(In percent) 







5 and 10 


5 and 20 


10 and 20 




Plantation 


years 


years 


years 


1 


Superior NF 


58 


14 


41 


2 


Chippewa NF 


35 


5 


38 


3 


Pillsbury SF 1 


86 


46 


74 


5 


General Andrews EF 2 


56 


48 


72 


6 


Univ. Minn. CFC 


64 


22 


58 


7 


Burnett CF 


76 


52 


81 


8 


Mosinee IF 


49 


40 


92 


9 


Chequamegon NF 


42 


38 


86 


10 


Nepco IF 


66 


56 


76 


11 


Argonne EF 


72 


25 


49 


12 


Marinette CF 


79 


66 


83 


13 


Ottawa NF 


55 


28 


67 


17 


Fife Lake SF 


77 


61 


90 



'Heights measured at 5. 11, and 20 years. 
2 Heights measured at 5, 13, and 20 years. 



SEED COLLECTION 
RECOMMENDATIONS 

In general, the study results 20 years after 
planting support the seed collection recommenda- 
tions made by King (1966) based on results 10 
years after planting. If jack pine is to be planted in 
the Lake States for relatively short rotation 
pulpwood production (30 to 40 years), the results of 
this test can be used with confidence. 

In any planting program, environmental condi- 
tions such as climate, photoperiod, soils, nursery 
treatment, planting techniques, and damage from 
insects and diseases interact with the genetic 
makeup of the plant material and may drastically 
alter results. To realize the greatest yields in jack 
pine we must minimize these interactions and we 
must use the best genetic material for the locale 
and the best techniques available for raising 
planting stock and establishing and managing 
plantations. The first step is careful seed source 
selection based on the best available information. 

Where local seed sources appear to be superior, 
as in the Minnesota plantations, we recommend 
using seed from selected stands near the planting 
site. However, where superiority of nonlocal 



18 



sources is indicated, as in the Wisconsin plantings, 
we recommend a cautious approach. In these in- 
stances the forest manager might consider mixing 
seed from selected local stands with seed from the 
recommended nonlocal seed sources, to insure 
against possible failure of nonlocal material at 
later ages. 

It is obvious from this study that using the 
"wrong" seed source will result in considerable 
volume losses. Using the "right" seed source, how- 
ever, can result in modest to substantial gains in 
volume production. 

The following recommendations should be fol- 
lowed for planting jack pine in the Lake States: 

1. Collect seed from young to middle-aged stands 
having uniform, normal stocking on good sites. 

2. Collect seed from individual trees with good 
growth and form and little or no evidence of 
serious pest incidence. 

3. In Minnesota use seed collected from selected 
stands near the planting site. 

4. In Wisconsin, at locations having less than 
9,100 growing degree days, use seed from se- 
lected stands near the planting site and mix 
with seed from selected proven stands in Upper 
and Lower Michigan. At warmer locations use 
seed collected from the southern portion of the 
species range in Lower Michigan. 

5. In Upper Michigan use seed from selected 
stands near the planting site. In Lower Michi- 
gan use seed from the southern portion of the 
species range in Lower Michigan. 



SEED PRODUCTION AREAS 

The results of this study can be used to develop 
jack pine breeding populations for the Lake States. 
Seed source groups 1, 2, and 3 define effective 
breeding zones within which we can identify the 
best stands and best trees on which to base pro- 
grams for genetic improvement. 

The best stands in each breeding zone should be 
relocated and converted into seed production areas 
(SPA's). If the original stands are no longer in 
existence, it may be necessary to use other good 
jack pine stands in the immediate vicinity. King 
(1973) recommended that the best stands should 
be thinned to about 60 trees per acre on the basis of 



spacing, growth rate, and form. Commercial quan- 
tities of seed of better quality than those presently 
available should be available from these stands 
within 5 years. When seedlings from these seed 
collections are available, trees in the SPA's can be 
harvested and the sites replanted with seedlings 
originated from the same SPA's. By following this 
procedure, the genetic integrity of the selected 
stands will be maintained. 

The following stands are recommended for con- 
version to SPA's: 

Zone 1 (Rudolfs (1954) collection zones 5 and 6) 
1592 Lake County, Minnesota 
1602 Itasca County, Minnesota 
Zone 2 (Rudolfs zone 4) 

1600 Cass County, Minnesota 

1610 Oneida County, Wisconsin 
1612 Gogebic County, Michigan 
1618 Alpena County, Michigan 

Zone 3 (Rudolfs zones 2 and 3) 

1595 Pine County, Minnesota 

1596 Pine County, Minnesota 

1608 Burnett County, Wisconsin 

1609 Marinette County, Wisconsin 

1611 Wood County, Wisconsin 

1616 Manistee County, Michigan 

1617 Ogemaw County, Michigan 



LITERATURE CITED 

Aim, Alvin A., Bruce A. Brown, and Raymond A. 
Jensen. 1966. Height and diameter variation in 
a Minnesota jack pine seed source plantation. 
Minnesota For. Note 178, 2 p. 

Arend, John L., Norman F. Smith, Stephen H. 
Spurr, and Jonathan W. Wright. 1961. Jack pine 
geographic variation — five-year results from 
Lower Michigan tests. Michigan Acad. Sci., 
Arts, and Letters Proc. 46:219-237. 

Batzer, H. O. 1961. Jack pine from Lake States 
seed sources differ in susceptibility to attack by 
white-pine weevil. U.S. Dep. Agric. For. Serv., 
Tech. Note 595, 2 p. U.S. Dep. Agric. For. Serv., 
Lake States For. Exp. Stn., St. Paul, MN. 

Batzer, H. 0. 1962. White-pine weevil damage dif- 
fers significantly by seed source on two northern 
Minnesota jack pine plantations. U.S. Dep. 
Agric. For. Serv., Tech. Note 618, 2 p. U.S. Dep. 
Agric. For. Serv., Lake States For. Exp. Stn., St. 
Paul, MN. 



19 



Bruce, Donald. 1972. Some transformations of the 
Behre equation of tree form. For. Sci. 18(2): 164- 
166. 

Funk, David T. 1975. Sixteenth-year remeasure- 
ment — eastern white pine provenance study 
plantation at Yellow River Forest, McGregor, 
Iowa. Progress Rep. FS-NC-1402, CG-361, 8 p., 
U.S. Dep. Agric. For. Serv., North Cent. For. 
Exp. Stn., For. Sci. Lab., Carbondale, IL. 

Jensen, Raymond A., T. Schantz-Hansen, and 
Paul O. Rudolf. 1960. A study of jack pine seed 
source in the Lake States. Minnesota For. Note 
88, 2 p. 

King, James P. 1966. Ten-year height growth var- 
iation in Lake States jack pine. In Joint Proc. 
Second Genetics Workshop of the SAF and the 
Seventh Lake States For. Tree Improv. Conf. 
1965:84-88. U.S. Dep. Agric. For. Serv., Res. 
Pap. NC-6, 110 p. U.S. Dep. Agric. For. Serv., 
North Cent. For. Exp. Stn., St. Paul, MN. 

King, James P. 1971. Pest susceptibility variation 
in Lake States jack pine seed sources. U.S. Dep. 
Agric. For. Serv., Res. Pap. NC-53, 10 p. U.S. 
Dep. Agric. For. Serv., North Cent. For. Exp. 
Stn., St. Paul, MN. 

King, James P. 1973. Practical breeding programs 
for jack pine in the Lake States. In Tenth Lake 
States For. Tree Improv. Conf. and Seventh 
Central States For. Tree Improv. Conf. Joint 
Proc, 1971: 45-49. U.S. Dep. Agric. For. Serv., 
Gen. Tech. Rep. NC-3, 61 p. U.S. Dep. Agric. For. 
Serv., North Cent. For. Exp. Stn., St. Paul, MN. 

King, James P., and Hans Nienstaedt. 1965. Vari- 
ation in needlecast susceptibility among 29 jack 
pine seed sources. Silvae Genet. 14: 194-198. 

Morgenstern, E. K., and A. H. Teich. 1969. Pheno- 
typic stability of height growth of jack pine prov- 
enances. Can. J. Genet. Cytol. 11: 110-117. 

Rudolf, Paul O. 1956. A basis for forest tree seed 
collection zones in the Lake States. Minnesota 
Acad. Sci. Proc. 24:21-28. 

Rudolf, Paul O., and R. L. Schoenike. 1963. Botani- 
cal and commercial range of jack pine in the 
Lake States. U.S. Dep. Agric. For. Serv., Res. 
Note LS-15, 4 p. U.S. Dep. Agric. For. Serv., 
Lake States For. Exp. Stn., St. Paul, MN. 

Rudolph, Thomas D. 1964. Lammas growth and 
prolepsis in jack pine in the Lake States. For. 
Sci. Monogr. 6, 70 p. 



Snedecor, George W., and W. G. Cochran. 1967. 
Statistical methods. Sixth edition. 593 p. Iowa 
State Univ. Press, Ames. 

Stoeckeler, J. H., and Paul O. Rudolf. 1956. Winter 
coloration and growth of jack pine in the nursery 
as affected by seed sources. Z. Forstgent. un 
Forstpflanzenzuchtung 5:161-165. 

Yeatman, C. W. 1974. The jack pine genetics pro- 
gram at Petawawa Forest Experiment Station 
1950-1970. Dep. Environ., Can. For. Serv. Pub. 
1331, 30 p., Ottawa. 

ACKNOWLEDGMENTS 

This paper presents 20-year results of the Lake 
States Jack Pine Seed Source Study initiated by 
the University of Minnesota and the North 
Central (then Lake States) Forest Experiment 
Station. The study was planned, organized, and 
established largely through the efforts of the late 
T. Schantz-Hansen of the University of Minnesota 
and Paul O. Rudolf (retired, USDA Forest Service) 
of the North Central Forest Experiment Station. 
All involved with this study sincerely appreciate 
the foresight and efforts put forth by these 
individuals. 

This study has been made possible through the 
cooperative efforts of several Federal, State, and 
private forestry agencies that collected seed, pro- 
duced nursery stock, established, maintained, and 
periodically evaluated plantations included in the 
study. The generous assistance and cooperation of 
the following agencies and numerous individuals 
within these agencies is gratefully acknowledged: 

Industrial Corporations 

Mosinee Pulp and Paper Mills Company 

Nekossa-Edwards Paper Company 
Universities 

University of Minnesota 

University of Michigan 
Counties 

Burnett County, Wisconsin 

Marinette County, Wisconsin 
State Agencies 

Michigan Department of Natural Resources 

Minnesota Department of Natural Resources 

Wisconsin Department of Natural Resources 
Federal Agencies 

Eastern Region, USDA Forest Service 

North Central Forest Experiment Station, 
USDA Forest Service 

ftll.S. GOVERNMENT PRINTING OFFICE: 1980- -66841 3/165 



20 



Jeffers, Richard M., and Raymond A. Jensen. 

1980. Twenty-year results of the Lake States jack pine seed source 
study. U.S. Dep. Agric. For. Serv., Res. Pap. NC-181, 20 p. U.S. Dep. 
Agric. For. Serv., North Cent. For. Exp. Stn., St. Paul, MN. 

Variation among Lake States jack pine seed sources in survival, 
total height, diameter, volume per tree, and volume per acre grown for 
20 years at 14 locations in the Lake States are presented and dis- 
cussed. Seed collection and seed production area recommendations are 
included. 



OXFORD: 232.12:174.7 Pinus banksiana (77). KEY WORDS: seed 
source recommendations, genotype-environment interaction, juve- 
nile-mature correlations, breeding zones. 



Jeffers, Richard M., and Raymond A. Jensen. 

1980. Twenty-year results of the Lake States jack pine seed source 
study. U.S. Dep. Agric. For. Serv., Res. Pap. NC-181, 20 p. U.S. Dep. 
Agric. For. Serv., North Cent. For. Exp. Stn., St. Paul, MN. 

Variation among Lake States jack pine seed sources in survival, 
total height, diameter, volume per tree, and volume per acre grown for 
20 years at 14 locations in the Lake States are presented and dis- 
cussed. Seed collection and seed production area recommendations are 
included. 



OXFORD: 232.12:174.7 Pinus banksiana (77). KEY WORDS: seed 
source recommendations, genotype-environment interaction, juve- 
nile-mature correlations, breeding zones. 



Soil is for -plants.. 




not for tire tracks. 



\/~ ' 



USDA FOREST SERVICE 
RESEARCH PAPER NC-182 




orth Central Forest Experiment Station 
)rest Service, U.S. Department of Agriculture 



North Central Forest Experiment Station 

Forest Service — U.S. Department of Agriculture 

1992 Folwell Avenue 

St. Paul, Minnesota 55108 

1980 



CONTENTS 

Page 

METHODS 3 

Managed Harvest Procedure 3 

Harvest Costs 6 

Transport Costs 9 

Supply Curve Construction 10 

Estimating Stumpage Costs 12 

RESULTS 

Mead Corporation , Escanaba, Michigan 12 

Ten-mill Study Area — Northern Wisconsin 

and Upper Michigan 15 

SUMMARY 18 

SELECTED BIBLIOGRAPHY 19 



THE SUPPLY AND ENERGY POTENTIAL OF FOREST 

RESOURCES 
IN NORTHERN WISCONSIN AND MICHIGAN'S UPPER 

PENINSULA 



Dennis P. Bradley, Economist, 

Eugene M, Carpenter, Market Analyst, Duluth, Minnesota 

James A. Mattson, Mechanical Engineer, Houghton, Michigan 

Jerold T. Hahn, Resource Analyst, St. Paul, Minnesota 

and Sharon A. Winsauer, Computer Specialist, Houghton, Michigan 



Studies have shown that wood fuel has only a small 
potential for alleviating the national energy short- 
age; however, the opportunities for the pulp and pa- 
per industry are excellent. Because it is close to the 
wood resource, skilled in wood harvest and transport, 
familiar with wood properties, and already providing 
close to half of its energy needs from wood residues, 
the pulp and paper industry is ideally suited to gain 
complete energy independence through the use of 
wood-based fuels. 

This report contains two related assessments of 
forest energy potential. First, a detailed study (fig. 1) 
was made of Mead Corporation's pulp and paper mill 
in Escanaba, Michigan. This mill has a procurement 
area covering most of Michigan's Upper Peninsula 
and a small part of northeastern Wisconsin. Second, a 
broader evaluation was made for 10 of the 21 pulp 
and paper mills in the four Forest Survey Units of 
northern Wisconsin and Upper Michigan (fig. 2). 

For both the case study and the regional analysis, 
four questions were asked: 

1. How much wood is annually harvestable for 
solid wood products, for wood fiber, and/or wood 
fuel? 

2. How much will it cost to harvest and deliver? 

3. What are the energy requirements of the mill or 
mills, their current sources of energy, and their 
individual opportunities for converting to wood 
fuel? 



4. At what prices can wood fuel compete with fossil 
fuels and purchased electricity? 

The first two questions address the supply of wood 
and the last two the demand. 

This study integrates existing forest survey meth- 
ods, which describe physical characteristics, with 
methods that add essential economic perspectives. 
Traditionally, Forest Survey identifies current forest 
conditions and suggests appropriate strategies for 
national and regional forest management. Although 
these analyses have important economic implica- 
tions, primarily in regard to regional timber bal- 
ances, they are not designed to answer economic 
questions of greatest interest to indiviual firms: 
where is the harvestable timber, and how much will 
it cost to harvest and transport? While the specific 
objective of this report is to assess forest energy po- 
tentials in northern Wisconsin and Michigan, our 
approach, which combines silvicultural projections 
with harvest and transport cost estimates, should 
have wide application in forest resource appraisals. 

The study assumes that field chipping will be a 
part of all the harvest systems considered and that 
chips will be produced from the pulpwood portions of 
trees as well as from branches, tops, rough and rotten 
trees, and small sound trees. When the quantity of 
saw logs warrants it, they will be recovered before 
chips, but round pulpwood will not be considered. In 
other words, only two products will result: saw logs 
and chips. 



UPPER PENINSULA 




><£> 



LOCATION OF 
TIMBERSHED# CONCENTRATION YARD 



NORTHERN WISCONSIN 



1 


ESCANABA 


2 


CHAMPION 


3 


GULLIVER 


4 


L'ANSE 


5 


NEWBERRY 


6 


SHINGLETON 


7 


TROUT LAKE 



Figure 1. — Seven timbersheds for Mead Corporation's Escanaba, Michigan, mill. 



UPPER PENINSULA 




NORTHERN WISCONSIN 



A SURVEYED PULPMILLS 



Figure 2. — Forest residues energy program study area and the locations of surveyed pulpmills. 



This is a considerable departure from current prac- 
tices. But the productivity of chippers, the rapidly 
improving technology for cleaning and sorting chips, 
and the difficulties of recovering branches, tops and 
similar residues in any other way all point to the need 



for this assumption. The ultimate use of the chipped 
portion will depend on many factors, but it seems 
likely that the industry will find the way to "pulp the 
best and burn the rest". 



METHODS 

Five major tasks were carried out for both the 
regional analysis and Mead case study. 

1. Using field plot data, a "managed harvest" pro- 
cedure was used to calculate the amount of timber 
that should be cut each year from each of 164 "har- 
vest opportunities". Each harvest opportunity repre- 
sents the area and timber volume having the same 
general characteristics such as type, age, size, and 
stocking. These characteristics in turn determine 
harvest method as well as harvest costs. 

2. Given tree size and stocking characteristics for 
each of these 164 harvest opportunities, harvest costs 
were estimated using a computer simulation of the 
selected harvest systems. 

3. Transport cost distributions were determined 
by measuring the distance from plot to delivery point 
for all plots in each harvest opportunity and assum- 
ing similar distributions of transport costs for all the 
volume each harvest opportunity represents. 

4. A mill energy study based on detailed inter- 
views and a brief engineering analysis was used to 
estimate total wood requirements and the price that 
wood residues would have to meet to be competitve 
for each mill. 

5. Finally, mill price and supply curves were com- 
pared to see if both fuel and fiber requirements could 
be met by the forest resource. 

Managed Harvest Procedure 

Central to this analysis was a "managed harvest" 
procedure developed by the Renewable Resource 
Evaluation Project (RREP) of the North Central For- 
est Experiment Station (NCFES). 

It must be emphasized that Managed Harvest is 
based on existing forest conditions and does not rep- 
resent a fully regulated or intensively managed har- 
vest level. However, managed harvest is the 
amount of timber that should be cut each year 
for the next 10 years 1 to move the forest 
toward a more fully regulated condition. This 
approach extends previous methods used to calculate 



1 Ten years is an arbitrary projection period based 
on the current Renewable Resource Evaluation Proj- 
ect schedule to remeasure each State. In 10 years, a 
new survey will identify different forest conditions 
and calculate a new managed harvest. 



"allowable" or "desirable" cut. In simplest terms, a 
regulated forest has an even distribution of trees 
in each age class, and a constant proportion is har- 
vested each year and immediately regenerated to 
begin the cycle anew. An unregulated forest, in 
contrast, has an uneven distribution of age classes 
which, if harvested under strict age or tree size rules, 
would result in widely fluctuating annual harvests. 
By this definition, existing forests, whether public or 
private, are generally unregulated. 

Previous measures of timber supply conducted by 
the NCFES in Michigan and Wisconsin provided al- 
most 4,700 sample plots for this study. In addition, 
4,300 sample plots were provided from surveys car- 
ried out by the State of Michigan and Mead Corpora- 
tion, for a total of 9,000 plots. 

Deciding how to achieve a more regulated forest 
requires a set of rules for determining how each tim- 
ber stand should be treated in the next 10-year plan- 
ning period. The set of rules we used was adopted 
from guides used on National Forests in the Lake 
States and follows recommended silvicultural prac- 
tices (table 1). From this regional consensus, 20 
type/site index combinations were identified. Each 
combination elicits a recommendation from the 
guide — whether to cut it, thin it, or leave it alone. An 
example of the guide recommendations for the paper 
birch type is shown in a "decision tree" format ( fig. 3). 

Based on the timber management guides and the 
two basic silvicultural systems practiced in the Lake 
States (even-aged and uneven-aged management), 
three harvest methods were assumed for this study: 

1. Clear cutting of: Even-aged types at or past 
rotation age, any stand so poorly stocked that 
maintaining the few trees to rotation age is not 
economical, and stands judged unsuitable for 
the site (should be converted to another type). 

2. Thinning of: Selected types (both even-aged 
and uneven-aged) of less than rotation age or 
mature size, but for which stocking is too dense 
for best growth. 

3. Selection improvement cutting of: Saw log 
stands of northern hardwood, oak-hickory, and 
swamp hardwood types. In this category, har- 
vesting will be assumed to take place in two 
stages. First, because full-tree skidding of 
large, mature trees damages residual crop 
trees, saw log trees will be cut manually, 
bucked in the woods, and forwarded as logs to 
the landing. Second, a mechanized relogging 
operation will recover saw log tops, cull trees, 
and undersized trees scheduled for cutting. 



Table 1. — Timber management guide summary 1 



Forest 


Site 
index 


Rotation 
age 


Management 
objective 2 


Schedule of intermediate cuts 




type 


Stand age 


Minimum BA Residual BA 






Years 
50 




Years 
( 3 ) 


en ft /ztrrp -- 




Jack pine 


<60 


PW 


ol/. (i./auic 


— 




60 + 


60 


PW 


( 3 ) 


— 


— 


Red pine 


all sites 


100 


ST 


0-20 
20-90 


( 4 ) 

100 


90 


White pine 


all sites 


120 


ST 


20-110 


130 


110 


Balsam fir 


all sites 


50 


PW 


( 3 ) 


— 


— 


White spruce 


all sites 


90 


ST 


35-80 


130 


100 


Black spruce 


<40 


100 


PW 


( 5 ) 


— 


— 




40 + 


70 


PW 


( 5 ) 


— 


— 


Tamarack 


all sites 


100 


PW 


( 5 ) 


— 


— 


Northern white-cedar 


all sites 


100 


ST 


( 5 ) 


— 


— 


Aspen 


<60 


40 


PW 


( 3 ) 


— 


— 




60 + 


50 


PW/ST 


( 3 ) 


— 


— 


Paper birch 


<60 


50 


PW 


( 3 ) 


— 


— 




60 + 


80 


ST 


30-50 


95 


70 


Swamp hardwoods 


all sites 


120 


ST 


30-100 


100 


75 


Oak 


<60 


— 


convert 


( 6 ) 


— 


— 




60 + 


100 


ST 


30-60 

70-80 

90 


115 
90 
75 


90 
65 

45 


Northern Hardwoods 


<45 


— 


convert 


( 6 ) 








45 + 


( 7 ) 


ST 


all ages 


110 


90 


Nonstocked 8 


all sites 













1 AII stands at or above rotation age are clearcut. 
2 ST = Sawtimber PW = Pulpwood. 
3 No intermediate cuts, clearcut at rotation age. 
4 Remove overstory if one remains. 
5 No intermediate cuts, strip cut at rotation age. 
6 Clearcut and convert to a more desirable type. 

7 The objective is to achieve and maintain 90 sq. ft. of basal area distributed 60 sq. ft. in sawtimber, 20 sq. ft. in poletimber, and 10 sq. ft. in samplings. All cull and 
short log trees are removed first. 
8 These stands have less than the minimum basal area required for type classification. 




CLEAR CUT 
ALL STOCKING 
LEVELS AND 
REGENERATE 



STOCKING IS BELOW 
MINIMUM LEVEL TO 
MAINTAIN STAND 



FIGURE 4 
COLS 1-6 



95 SQ. FT. 
' MINIMUM 



THIN TO 
70 SQ FT 



LEAVE 
ALONE 



CLEAR CUT/ 
REGENERATE 




LEAVE 
ALONE 



CONVERT/ 
REGENERATE 



per cubic foot was determined for each species in the 
study area; these factors were used to convert volume 
to weight: 



Species 


Pounds cu. ft 


White pine 


37 


Red pine 


49 


Jack pine 


40 


White spruce 


36 


Black spruce 


42 


Balsam fir 


46 


Hemlock 


50 


Tamarack 


48 


Northern white-cedar 


32 


Oak 


60 


Yellow birch 


56 


Sugar maple 


56 


Red maple 


51 


Beech 


58 


White ash 


48 


Black ash 


52 


Balsam poplar 


50 


Cottonwood 


48 


Paper birch 


52 


Bigtooth aspen 


49 


Quaking aspen 


47 


Basswood 


44 


Black cherry 


48 


Elm 


55 



Figure 3. — Management guide for paper birch type. 



Ordinarily, a calculation of harvested volume in- 
cludes only sound, straight portions of trees — mate- 
rial larger than 4 inches for pulpwood, 7 inches for 
softwood saw logs, and 9 inches for hardwood saw 
logs. It ignores rough, rotten, and short-log trees, cull 
sections, branches, tops, and trees less than 5 inches 
d.b.h. In addition, areas with less than 3 cords of 
"usable" material per acre are considered uneconom- 
ical to log and are left out. However, because fuel is 
an ideal use for much of this material, our analysis of 
plot data includes these very significant quantities. 

In this report, managed harvest tonnages are pre- 
sented in terms of (1) saw log portions of trees, (2) 
pulpwood portions of trees, (3) rough and rotten trees, 
(4) branches, tops, and saplings (trees less than 5 
inches d.b.h.). 

Forest Survey has traditionally measured volume 
in cubic feet. However, in this study the green weight 



Harold Young at the University of Maine was the 
principal source of information for developing crown 
and sapling weight estimates. A universal equation 
(Young 1977) relating d.b.h. to complete tree weight 
in pounds was modified to determine total tree 
weight by d.b.h. for all species. Correction factors 
were then used to adjust these weights for individual 
species. Young estimates total tree weight to be dis- 
tributed 20 percent stump and roots, 55 percent bole, 
and 25 percent tops and limbs. When only above- 
stump components are considered, the distribution is 
approximately 69 percent bole and 31 percent tops 
and limbs. Based on this distribution, the top and 
limb weight was estimated from the bole weight. 
These crown weights were then adjusted for poletim- 
ber and sawtimber trees, by species, using factors 
developed from various biomass studies. 

Based on the timber management guides, the man- 
aged harvest program, and the inventory plots, we 



developed a set of 20 computer-generated type/site 
index tables. The general characteristics of these ta- 
bles are shown for paper birch/site index s= 60 (fig. 4). 
Each table has 13 columns summarizing area, total 
tonnage, tons per acre, and other average stand char- 
acteristics. Columns one through six report areas and 
tonnages from clear cuttings. Columns seven 
through 12 report areas and tonnages from thinnings 
of stands that are to continue in the same type and for 
which management will proceed normally to rota- 
tion. Column 13 reports tonnages and areas too 
sparse to justify maintaining them to merchantable 
sizes. 

Altogether, there are 260 possible columns sum- 
marizing harvest opportunities. However, because 
some types never require a thinning and other exist- 
ing types will not be maintained, the number of har- 
vest opportunities can be reduced to 164. Average 
d.b.h. and basal area for each of these 164 harvest 
opportunities are used in the subsequent estimation 
of harvest costs. 



Harvest Costs 



Using a computer-based harvest system simulator, 
which models the harvesting equipment and produc- 
tion in various stands, harvest costs were estimated 
for each of the 164 harvest opportunities. The simula- 
tors used in this study were written in GPSS (IBM 
1974) and represent the activities of the machines 
and operating rules for harvesting various stands of 
trees (Bradley et al. 1976, Bradley and Winsauer 
1976a). These models duplicate the essential features 
of typical whole-tree chipping systems and the re- 
logging harvest system. 

Thirty-three hypothetical stands representing 
most combinations of six average stand diameters (3 
to 18 inches d.b.h.) and six basal areas (30 to 180 
square feet) were developed. Selected characteristics 
of these stands are shown in table 2. 



AREA AND VOLUMES OF HARVEST AND THINNINGS BY STAN0 VOLUME CLASS AND FUEL, PRODUCT CLASS 
N0RTHEASTEPN WISCONSIN tl97? 



0-t 
399: 



33« 



400-t 
1199: 



PAPER BIRCH 



SI 61* 



HARVEST t THINNINGS 

PERCENT OF GROWING STOCK VOLUME IN SAWTIHBER TREES 
t <33S : 331 I 

CROWING STOCK CUT VOLUME PER ACRE 
t o-t '•oo-: : 0-: 400-: I 0-1 
1200*: 399: 1199: 1200»: 399: 119S1 1200*< 399: 



<33* 



400-J 
1199: 



I 

<33S » 

UNOERt 



1200«:STOCKEO< TOTAL 



COLUMN 



AREA 



.17 



.85 



7 8 

.(THOUSAND ACRES), 



VOLUMES BY 
FuEL/PRODUCT CLASSES 

SAWLCGS 

PULPkOOD 

RO'jGh.ROTTEN 

BRA^CHES'TOPS 

TOTAL 

VOLUMES PER ACRE BY 
FUEL/PRODUCT CLASSES 

SAWL0GS/ACRE 

PULPtOOO/ACRE 

RGUGM»ROT./ACRE 

BRANCHES»TOPS/A 

TOTAL 



.(TOTAL GREEN T0Ns>, 



6694. 

738. 

556. 

4632. 

12620. 



29012. 
12386. 
457. 
19511. 
61366. 



1*023. 

53569. 

4263. 

47096. 

118951. 



.(GREEN TONS PER ACRE). 



39. 

4. 

3. 
27. 
74. 



76. 

33. 

1. 

51. 

161. 



16. 

63. 

5. 

55. 

140. 



9 10 

'J* W24* 



11 

".Va 



12 



2040. 
10537. 

3130. 
12202. 
27909. 



2754. 
15285. 

2221. 
19106. 
39366. 



2. 

8. 

3. 
10. 
23. 



3. 

16. 

2. 

19. 

40. 



13 



3.86 



1790. 
2025. 
3198. 
2927. 
9940. 



56313. 

94540. 

13825. 
10S474. 
270152. 



7. 

8. 
13. 
12. 

41. 



IS. 

24. 

4. 

27. 
70. 



NUMEER OF PLOTS 



1. 



5. 



S. 



4. 



Figure 4. — Managed harvest output table for paper birch/site index 60. 



Table 2. — Selected characteristics of the 33 test stands 
used in the harvest simulation 

TREES PER ACRE (number) 



Average Average 


Basal area (square feet per acre) 


d.b.h. 


height 


30 


60 90 120 150 180 


Inches 


Feet 






3 


39 


486 


967 1,452 _ _ 


6 


43 


129 


265 395 526 660 790 


9 


51 


61 


123 186 246 308 370 


12 


57 


36 


72 108 143 179 215 


15 


62 


24 


47 71 94 118 141 


18 


66 


17 


34 50 67 84 101 


GROSS VOLUME (cubic feet per acre) 


3 


39 


613 


1,221 1,833 _ _ 


6 


43 


756 1,511 2,267 3,023 3,778 4,534 


9 


51 


896 


1,793 2,689 3,585 4,482 5,378 


12 


57 


1,002 2,005 3,007 4,009 5,012 6,014 


15 


62 


1,086 2,171 3,257 4,342 5,428 6,513 


18 


66 


1,182 2,363 3,545 8,727 5,908 7,090 


TREE SPACING (feet) 


3 


39 


9.5 


6.7 5.5 __ _ 


6 


43 


18.4 


12.8 10.5 9.1 8.1 7.4 


9 


51 


26.7 


18.8 15.3 13.3 11.9 10.7 


12 


57 


34.8 


24.6 20.1 17.5 15.6 14.2 


15 


62 


42.6 


30.4 24.8 21.5 19.2 17.6 


18 


66 


50.6 


35.8 29.5 25.5 22.8 20.8 



Detailed machine speeds and capacities were col- 
lected from a number of sources and supplied to the 
simulator. Machine operating characteristics are as 
follows: 

Two feller bunchers were considered in this study. 
The first is a rubber-tired, frame-steered machine 
with an accumulating shear mounted in front. This 
machine must drive up to each tree. The second feller 
buncher has a rotating boom and accumulating shear 
mounted on crawler tracks. It does not have to ap- 
proach each tree, but can swing and extend the shear 
while maintaining a fairly straight path. 

Bunches are picked up with an hydraulic grapple 
mounted on a rubber-tired, frame-steered skidder. 
Only one kind and size of skidder was considered in 
this analysis. At the landing, bunches are disassem- 
bled by the chipper's loader, chipped and blown im- 
mediately into a waiting van. When warranted, saw 
logs are first bucked, set aside and only the tops are 
chipped. 



Two chipper sizes were used, based on maximum 
acceptable butt diameter: small (12 inches and less) 
and large (22 inches and less). 

Full-tree skidding and saw log sorting at the land- 
ing in thinned or selectively cut sawtimber stands of 
oak-hickory, northern hardwoods, and swamp 
hardwoods result in too much damage to remaining 
crop trees. For these types, and in stands with aver- 
age d.b.h. greater than 9 inches and basal areas in 
excess of 90 square feet, saw logs are assumed to be 
removed by a short log operation, with the remaining 
material harvested by a topwood harvester. 

The topwood harvester which is undergoing devel- 
opment at the Forest Engineering Laboratory in 
Houghton, Michigan, has a rotatable and highly ma- 
neuverable shear mounted on a rubber-tired, frame- 
steered chassis. In addition to felling and bunching 
non-saw-log trees, its flexible shear is used to sever 
large limbs and compact the tops from the large 
sawtimber trees cut earlier. The compacted tops and 
other smaller trees can then be removed with little 
damage to the remaining stand. 

Given the equipment specifications and test stands, 
the simulator was used to select the combination of 
equipment that produced maximum net revenue per 
hour for each of the three harvest methods. Tables 3, 4, 
and 5 show the configuration, productivity, and cost 
of each system finally chosen. These decisions were 
reached by the following procedures: 

1. Costs per hour were calculated for each ma- 
chine, based on estimated purchase price, machine 
life, scheduled and actual hours to be worked each 
year, and cost of repairs, wages, insurance, taxes, 
fuel, and lubricants. 

2. Overhead costs and a margin for profit and risk 
were added to machine costs. 

3. Various combinations of machines, their 
speeds, capacities, costs, and the stand description 
were tested by the simulator. Each computer run 
resulted in an estimate of production per hour in 
cubic feet and tons. Because the chipper was the 
limiting factor, the number of skidders was increased 
until skidder output closely matched the chipper and 
net revenue per hour was maximized. 

4. The simulated production rates were adjusted 
to realistic levels using industry-developed estimates 
of achievable operator efficiency and machine utili- 
zation. 



Table 3. — Harvest systems selections — equipment, 
output, and cost — for clearcuts 

SYSTEM OUTPUT (green tons per hour) 



Table 5. — Harvest system selections — equipment' 
output, and cost for relogging 

SYSTEM OUTPUT (green tons per hour) 



Basal area (square feet per acre) 



Basal area (square feet per acre) 



D.b.h. 



30 



60 



90 



120 150 



180 



D.b.h. 



90 



120 



150 



180 



3 1 

6 

9 
12 
15 


6.00R 2 

22.34R 

35.93R 

46.40R 

61.43 

85.93 


6.03R 
23.19R 
38.13R 
43.39 
65.84 
86.19 


5.95R 
19.89 
33.85 
48.80 
68.24 
91.42 


20.46 
34.44 
51.56 
71.83 
94.91 


22.15 
36.10 
53.49 
73.49 
00.67 


22.66 
36.78 
55.11 
74.42 
02.54 


9 

12 
15 
18 


15.28 15.28 15.28 
13.96 13.96 13.96 
16.05 16.05 16.05 
22.05 22.05 22.50 
SYSTEM COST (dollars per green ton) 


15.28 
13.96 
16.05 
22.05 


18 


9 


13.85 


12.77 


12.41 


12.16 




SYSTEM COST (dollars per green ton) 




12 
15 
18 


13.35 

11.21 

8.11 


12.99 

11.00 

8.04 


12.87 

10.76 

8.00 


12.71 


3 1 
6 


20.78R 
8.03R 


20.40R 
7.56R 


20.43 
8.60 


8.38 


7.82 


6.72 


10.73 
7.94 


9 


4.74R 


4.39R 


4.81 


4.70 


4.49 


4.40 


'Equipment used: 1 topwood harvester, 3 grapple skidders, 1 large chipper. 


12 


3.56R 


3.69 


3.28 


3.08 


2.96 


2.87 


for all conditions. 








15 


2.78 


2.43 


2.33 


2.19 


2.13 


2.10 












18 


2.05 


1.91 


1.74 


1.65 


1.56 


1.50 













1 Two skidders and small chipper; for all other diameters, 3 skidders and large 
chipper. 
2 R = rubber-tired feller-buncher; No R = track-type feller buncher. 



Because the sawmill industry is concerned about 
the loss of saw logs due to field chipping, we identified 
stand conditions that would justify sorting saw logs 
before chipping. First, all species were ranked by 
their relative "woods run" saw log prices. Second, 
2,000 board feet or roughly 10 tons per acre was 
arbitrarily chosen as the minimum tonnage recover- 
able during chipping for the most valuable saw logs. 
The minimum recoverable tonnages for the other 
types were then ranked accordingly: 



Table 4. — Harvest system selections -equipment, out- 
put, and cost for thinning 

SYSTEM OUTPUT (green tons per hour) 





Basal area I 


square feet per acre 


) 


D.b.h. 


60 


90 


120 


150 


180 


6 1 


22.40R 2 


22.80R 


23.20R 


23.20R 


24.40R 


9 


34.50R 


37.60R 


38.70R 


38.90R 


39.06R 


12 


47.07R 


54.29R 


54.89R 


46.44 


48.77 


15 


62.45 


61.61 


61.52 


68.81 


66.49 


18 


82.85 


86.13 


84.21 


90.37 


91.43 


SYSTEM COST (dollars per green ton) 


6 1 


7.95R 


7.71R 


7.46R 


7.34R 


7.05R 


9 


4.95R 


4.47R 


4.31R 


4.22R 


4.17R 


12 


3.54R 


3.03R 


2.96R 


3.44 


3.27 


15 


2.77 


2.68 


2.61 


2.33 


2.38 


18 


2.11 


1.91 


1.94 


1.79 


1.74 



Type 

Jack pine 

Red pine 

White pine 

Balsam fir 

White spruce 

Black spruce 

Tamarack 

Northern white-cedar 

Aspen 

Paper birch 

Elm-ash-cottonwood 

Oak-hickory 

Northern hardwoods 



Minimum 
saw log tonnage per acre 

no saw logs recovered 

12 

12 
no saw logs recovered 

15 
no saw logs recovered 
no saw logs recovered 

10 

20 

10 

15 

10 

10 



For all diameters, 3 skidders and large chipper. 
2 R = rubber-tired feller-buncher; No R = track-type feller buncher. 



Finally, after all 164 harvest opportunities were 
matched to the proper cell in the three harvest-cost 
matrixes, all stands not relogged and having more 
than the minimum saw log tonnage for the type were 
assumed to be harvested by a field chipping and saw 



log sorting system. The cost of chips from this com- 
bined operation was increased as follows: 

Cost of chips = "chips-only" cost/ton > 
+ sawtimber tons/acre 



c- 



total tons/acre 



) 



Thus, if an aspen stand had a "chips-only" cost 
of $6.00/ton, a sawtimber volume of 26 tons/acre, 
and a total volume of 100 tons/acre, the cost of chips 
from the saw log and chip operation would be $6.00 
x / 1 + 26\ = $7.56/ton. 



/I + 26 \ 
V 100,! 



Transport Costs 

Recall that each harvest opportunity summarizes 
the acreage and volume of an entire type/site index, 
age, harvest method, cutting level, etc. Some of the 
stands in a harvest opportunity are close to the road 
and to delivery points, while others are far away. By 
measuring the transport distance from each plot to 
delivery point, we were able to estimate the distribu- 
tion of transport distances and costs for each harvest 
opportunity. The process required one assumption: 
total harvestable areas and tonnages in the entire 



harvest opportunity are distributed by distance and 
cost to delivery point in the same proportion as sam- 
ple plot areas and tonnages. 

In the Mead case study, distances from the plots in 
each of seven timbersheds were measured to the mill 
or concentration yard. For the regional analysis, 
because eventual delivery points were unknown, dis- 
tances were measured from each plot to the county 
center. However, when managed harvest was sum- 
marized for each survey unit, all counties were 
combined into one transport cost distribution. The 
analysis assumed that all chipped material would be 
hauled in 35-ton vans using per-ton charges provided 
by Mead Corporation. 

An example of managed harvest output is shown in 
figure 5 for Mead timbershed 1 (aspen-type/site index 
< 60). The boxed-in portion (column 3) shows one 
harvest opportunity. Using a "harvest and transport 
cost program" based on a method called ACCESS 
(Bradley 1972), the 55 plots from column 3 in the 
managed harvest table were distributed in a "harvest 
and transport cost table" (fig. 6). Each plot represents 
1/55 of the area in column 3, but the plots are now 
distributed across the table by distance to delivery 
point. 



AREA AND VOLUMES OF HARVEST AND ThIkNINGS BY STAND VOLUME CLASS AND FUEL PRODUCT CLASS 



399: 



33* 



400-1 
1199: 



TIMBER SHED 1 



ASPEN 



SI 0-60 



HARVEST t THINNINGS t : 
PERCENT OF GROWING STOCK VOLUME IN SAWTIMBER TREES 

: <33« » ? 33% : <33* « <33% : 

GROWING STOCK CUT VOLUME PER ACRE 

: 0-« 400-: : 0-: 400-: : 0-: 4oo-: : UNCER: 

1200*: 399: U99: 1200*: 399: H99: 1200»: 399: 1199: i200*:SToCKED: 



TOTAL 



AREA 



.39 



3.66 



VOLUMES BY 
FUEL/PRODUCT CLASSES 



SAULOGS 
PULPWOOD 
ROuGr. ROTTEN 
BRANCxES^TOPS 
TOTAL 



4916. 80679. 

2355. 50197. 

460. 71S3. 

3241. 66113. 

10974. 204172. 



volumes per acre by 
fuel/proouct classes .. 

sawlogs/acpe 13. 

PULPiOOO/ACRE 6. 

ROUGh.ROT./ACRE 1. 

BSANCH£S»TCPS/A 8. 

TOTAL 28. 



NUMBER OF PLOTS 



22. 

1*. 

2. 

18. 

56. 



46. 



1.11 



63568. 

30380. 

3539. 

42580. 

140067. 



57. 

27. 

3. 

38. 

126. 



55. 



3.51 



6.43 



1.26 



(THOUSAND ACRES) , 




.(TOTAL GREEN TONS), 



S286. 31389. l8o54. 

50683. 216604. 59345. 

4337. 7393. 1907. 

34233. 149472. 43792. 

94539. 404858. 123098. 



.(GREEN TONS PER ACRE). 



2. 
14. 

1. 
10. 
27. 



39. 



5. 
34. 

1. 
23. 
63. 



140. 



14. 

47. 

2. 

35. 

98. 



76. 



12.72 29.03 



9956. 213350. 
24483. 434052. 

3629. 28448. 
33880. 373311. 
71953. 1049661. 



1. 
2. 
0. 
3* 
6. 



25. 



7. 
15. 

1. 
13. 
34. 



Figure 5. — Example of managed harvest output table. 



HARVEST AND TRANSPORT COSTS 
OF ALLOCABLE CUTS. 





STATEl MICHIGAN 




T, 


SHEOl 


1 TYPE 


1 ASPEN 




•■lit o- 


,0 TOTAL ACRESi 


1110 


Table « 


, 11 




X SAXTJM3ER BY VOLUMEl ZllX 




CUT 


VOLUME PER 


ACRE CCU, 


: T,): 1200* 




TOTAL PLOTS 


55 


COLUMN 


•I 3 




KIND Of HARVESTl CLE< 


R CUT 




DBH 


CLASS! 9 


BASAL AREA CLASS 


1 120 






















DISTANCE 


TO MILL 


» MILES 


















0-20 


20-"0 


90-60 


60-80 


80-100 


100-120 


120-140 


140-160 


160-180 


180 * 


PLOTS 
TOTAL 




NUMBER OF PLOTS 




1 


2 


27 


7 


17 

















54« 




X TOTAL PLOTS 




1.85 


3.70 


50.00 


12.96 


31,48 


,00 


,00 


.00 


,00 


,00 






ACRES 




20 


41 


55S 


143 


319 

















1110 




AVE OIST IN HILES 




16.50 


25.50 


52.67 


73.32 


91.07 


,00 


.00 


,00 


,00 


.00 


65,76 




AVE OIST BY ROAD QUALITY 


























PAVED 




11.75 


21.12 


ai,86 


70.64 


84,82 


,00 


.00 


,00 


,00 


.00 


59,40 




GRAVELED 




2.50 


.00 


4.55 


1.82 


3,51 


.00 


.00 


,00 


,00 


.00 


3,66 




ALL HEATHER 




.00 


,00 


1.6" 


.11 


1.18 


,00 


,00 


,00 


,00 


.00 


1.23 




DRY LEATHER 




2,00 


,00 


.** 


.32 


1.22 


,00 


.00 


,00 


.00 


,00 


,94 




SEASONAL 




.25 


1.00 


.33 


.2' 


.07 


.00 


,00 


,00 


,00 


.00 


.27 




CROSS-CCUNTRY 




.00 


.3« 


.28 


.14 


.26 


,00 


.00 


,00 


,00 


,00 


.25 


Row A - 


HARVEST AND TRANSPORT 
PER TON 


COST 
S 11.44 


* 11.70 


S 12." 


S 13.09 


$ 13,60 


S ,00 s 


,00 


S .00 s 


,00 


S ,00 






SAkTT*3ER VOL (TONS) 
ACCUMULATIVE VOL 




1177 

1177 


2351 
3S31 


31781 
35315 


8210 
43555 


20012 
63567 



63567 




63567 



63567 



63567 



63567 


63567 




PULP-<000 VOL (TONS) 
ACCUMULATIVE VOL 




562 
562 


1125 

1687 


l!!?? 


3938 

20815 


9564 
30379 


30379 


3037° 


30379 



30379 


30379 


30379 




ROUC-M t ROTTEN (TONS) 
iCCUhULATIVE VOL 




65 
65 


l 2 l 

l«6 


1769 
1*65 


458 
2423 


1114 
3537 


353? 



3537 



3537 


353? 


353? 


35J7 




TOPS AND LI U BS (TONS) 
ACCUMULATIVE VOL 




788 


1577 
$365 


mi 


5519 

29174 


13104 
42578 


42578 


4257? 


42578 



42578 


42578 


42578 


Row B-l — 
B-2— 


TOTAL TCNS-3 PRODUCTS 
TOTAL ACC TONS 




1315 
1415 


2833 

1218 


38249 
02197 


9915 
52412 


24082 
76494 




76494 




76494 




76494 




76494 




76194 




C-1 
C-2 
C-3~ 


TOTAL COST. 3 PRODUCTS 
TOTAL ACC COST 
ACC COST/TON 


s 

s 


ISIS? 

11,11 


S 33116 

i 1«333 

% 11. si 


S 1777J0 

S 527063 

i 12,10 


i 129737 

S 656850 

J 12.53 


5 3.27515 

J 95C365 

i 12.87 


i OS 
S 961365 J 
S 12.87 1 


98436? 
12.87 


J 98131,5 S 
J-12.87 S 


98136? 
12,87 


S 98036? 
112. 87 





SAWLOG VOLUMES NOT INCLUDED - THE SAWLOGS HAVE HEEN SORTED OUT AT ThE CHIPPER 

Figure 6. — Example of harvest and transport cost output. 



One plot representing 20 acres and 1,415 tons is 
16.50 miles from the mill, 27 plots representing 555 
acres and 38,249 tons are 52.67 miles from the mill. 
These 1,415 tons and 38,249 tons are harvestable and 
transportable to the mill at an average cost of $11.44 
and $12.49 per ton, respectively. 



Supply Curve Construction 

Marginal and average delivered cost curves can be 
constructed from the harvest and transport cost ta- 
bles for practically any combination of (a) type/site 
index, (b) proportion of sawtimber in the stand, (c) 
growing stock volume cut per acre, and (d) kind of 
harvest (clear cut, mechanized thinning, or selection 
cut/relogging). 

Each harvest and transport cost table (fig. 6), al- 
ready includes marginal and average cost calcula- 
tions for the "three-product total" of pulpwood, rough 
and rotten, and tops and limbs. Row A shows the 
harvest and transport cost for the volumes found in 
each distance class. Row B-l shows the individual 



volumes of the "three-product total" in each distance 
class that are deliverable at the costs shown in row A. 
Row B-2 shows the cumulative volumes of the "three- 
product total" that can be delivered at or below the 
cost shown in row A. 

A marginal delivered cost curve can now be con- 
structed by plotting row B-2 against row A (fig. 7). 
This curve shows marginal cost or the cost of deliv- 
ering the last ton of any volume desired. For example, 
the mill would have to pay $13.60/ton (row A, 80-100 
miles) to recover the last of 76,494 tons (row B-2, 80- 
100 miles). That is, this cost would be paid for each of 
the last 24,082 tons (row B-l, 80-100 miles). If only 
52,412 tons are needed (row B-2, 60-80 miles), the 
cost of the last ton would be $13.09/ton (row A, 60-80 
miles). Of course, at either level of demand, some 
wood would cost substantially less; the delivered cost 
of the first 1,415 tons would be $11.44/ton. 

A more useful measure of supply is the average 
cost. That is, what price must be paid on the average 
to recover any desired amount? This is more useful 
because it is the average cost that will be compared to 
average revenue earned by the firm. Average cost is 



10 



14 



2 
O 

h- 

§ 13 
lU 
QC 
O 

69 



12 



<0 

o 
o 

Q 
Lu 
0C 
LU 



JU 11 

Q 



10 



ASPEN 

SITE INDEX ^ 60 
MATURE, CLEAR CUT 
SAWTIMBER TREES s 33% 
GROWING STOCK CUT VOL/AC 
AVE DBH ; 9 INCHES 
AVE BA = 120 SQ. FT. 



1200 FT 




A VERAGECOST_ 
^UPPL^) 



10 20 30 40 50 60 

CUMULATIVE VOLUMES (THOUSAND TONS) 

Figure 7. — Example of marginal and average delivered cost curves. 



70 



80 



shown in row C-3. For any distance class, it is the 
cumulative cost of the "three-product total" (row C-2) 
divided by the cumulative tonnage (row B-2). This 
curve is plotted in figure 7 along with marginal cost. 
Row C-3 in every harvest and transport cost table 
shows the average cost of the cumulative volumes up 
to each specific distance class. 

In the results section to follow, the average cost 
curves are called supply curves because they repre- 
sent the average cost of meeting a specific demand for 
wood. Figure 7 shows the supply of timber from all 
the aspen stands in Mead's timbershed 1 with a site 
index =£ 60 and with the other following characteris- 
tics: 

a. mature stand clearcut, 

b. sawtimber trees as a percent of growing stock 
trees 5= 33 percent, 

c. growing stock cut volumes/acre 3= 1,200 cubic 
feet, 

d. average d.b.h. = 9, 

e. average d.b.h. = 120 ft. 2 . 



The harvest cost portion of delivered cost for this 
harvest opportunity is $5.36 per ton, and was derived 
in the following way: Because a clear cut was recom- 
mended, table 3 (Harvest system selections-clear 
cuts) was examined for the stand averaging 9 inches 
d.b.h. and 120 square feet basal area. A harvest cost 
of $4.70/ton was indicated. However, this cost as- 
sumed that all the trees weighed 55 pounds per cubic 
feet and it was multiplied by the ratio of assumed tree 
weights to actual tree weights for the aspen type. 

$4.70/ton x (55 lbs./ft. 3 - 48.20 lbs./ft. 3 ) = $5.36/ton 

The harvest cost portion of each table is the same in 
each distance class, but of course, transport costs 
increase from left to right. Similar harvest and trans- 
port cost tables were prepared for each of the 164 
harvest opportunities. 

Generally, individual havest and transport cost 
tables are not based on enough data to be significant. 
Therefore, most of the supply curves used in the anal- 
ysis are aggregates of several tables. 



11 



Estimating Stumpage Costs 

The supply curves do not include stumpage cost 
because of the wide variation among owners and the 
complexity of the calculations (which often include 
volume per acre, accessibility, wood quality, road 
construction costs, and size of area). Therefore, any- 
one who wishes to use the delivered cost curves 
should add his own estimate for stumpage. 

However, for the approximate comparisons of sup- 
ply and demand to follow, we estimated an average 
stumpage cost for the combined volumes of 
roundwood and residues by species groupings. Resi- 
dues, including rough and rotten trees and tops, 
branches, and saplings, were valued at less than the 
minimum price now paid for mixed hardwood 
roundwood, as estimated in the Wisconsin Forest 
Products Price Review (1977). 

Estimated stumpage price, 



This resulted in the following tabulation for each 
species group: 



Species 




all products 




$/cord 


$/ton 


Pine 


9.14 


3.97 at 2.3 tons/cord 


Other softwoods 


4.82 


2.01 at 2.4 tons/cord 


Aspen 


4.00 


1.67 at 2.4 tons/cord 


Mixed hardwoods 


2.15 


0.77 at 2.8 tons/cord 



Since mixed hardwood stumpage was valued at about 
$0.77/ton, we assumed that all categories now classed 
as residues would be worth no more than $0.50/ton. 

Next, we used the following ratios of residues to 
roundwood for the species groups found in Michigan's 
Upper Peninsula: 



Species groups 

Pine 

Other softwood 

Aspen 

Mixed hardwoods 



Ratio: 

Residue volume/ 

Roundwood volume 

0.64 
.67 
.66 

.78 



That is, for pine, if a stand has 1 ton of roundwood per 
acre, it has 0.64 tons of rough, rotten, branches, tops, 
and saplings. 

Finally, using (1) the average roundwood stump- 
age price for each species group, (2) the assumed 
residue price of $0.50/ton, and (3) the roundwood/re- 
sidue ratios above, we determined a combined 
roundwood and residue weighted stumpage cost/ton. 
For example: 

1 ton of pine roundwood at $3.97/ton = $3.97 
+ .64 tons of pine residue at $ .50/ton = .32 



Species group 

Pine 

Other softwoods 

Aspen 

Mixed hardwoods 

Overall average 



Weighted stumpage cost 

(roundwood and residue $/ton) 
$2.62 

1.41 

1.20 

0.60 
$1.04 



$4.29 



$4.29/1.64 tons = $2.62/ton 



The last step was the calculation of an overall aver- 
age, weighted by the proportion of similar species 
groups found in the managed harvest for Michigan's 
Upper Peninsula. 

Mill Demand for Fiber and Fuel 

This study assumed that the entire demand for 
wood fiber and wood fuel of 10 of the 21 mills in the 
region would be met from the regions forest resource 
(fig. 2). Four steps were required. 

First, interviews were conducted at each of the 10 
mills to determine mill process and technology; wood 
procurement program; energy requirements, 
sources, and current levels of energy independence; 
internal steam, steam-electric, and hydro-electric fa- 
cilities; and current unused residues. 

Second, a brief engineering analysis for each mill 
estimated the opportunities (and costs) to achieve 
complete energy independence by either converting 
existing boilers or constructing new ones. 

Third, current and projected prices were deter- 
mined for the major fossil fuels as well as for pur- 
chased electricity. 

Fourth, using current and projected fuel prices and 
the costs of each conversion or new construction op- 
portunity, the price advantage of wood over existing 
fuels that just balanced capital costs was established. 
This advantage was then subtracted from each mill's 
existing fuel costs to establish the highest price that 
each mill could afford to pay for wood fuel. 



RESULTS 

Mead Corporation, 
Escanaba, Michigan 

Supply of wood fiber and fuel 

The combined area of Mead's seven timbersheds in 
Michigan's Upper Peninsula and northeastern Wis- 
consin is 8.9 million acres, of which 7.7 million acres 



12 



are classed as commercial forest land. This is distrib- 
uted as follows: 





Total 


Commercial 


Timbershed 


land area 

( 


forest 

acres) 


1 Escanaba mill 


5,023,600 


4,338,900 


2 Champion 


494,600 


458,200 


3 Gulliver 


778,100 


665,400 


4 L'Anse 


760,100 


671,700 


5 Newberry 


614,100 


550,900 


6 Shingleton 


275,400 


241,600 


7 Trout Lake 


997,600 


790,500 




8,943,400 


7,717,200 



Almost 3.5 million of the above acres are in public 
ownership, divided equally between State and Na- 
tional Forests. These acres have not been reserved for 
single use and presumably can be harvested as the 
timber matures. Two million acres are owned by for- 
est industry. Roughly 33 percent of this area is in 
softwood types, 22 percent is in aspen and paper 
birch, and 37 percent is in northern hardwoods. Be- 
cause over 50 percent of the area and inventory is in 
timbershed 1, the analysis will focus on this tim- 
bershed. 

Managed harvest recommendations for Mead's 
timbershed 1 call for the annual harvest of 146,000 
acres of all types, yielding over 7.4 million tons of all 
products (table 6). This total in timbershed 1 includes 
more than 1.4 million tons of saw logs assumed to be 
sorted during chipping or removed before relogging. 
The remaining saw log volumes are considered insuf- 
ficient per acre to warrant separation, and are to be 
chipped with the remaining stand. Thus, for tim- 
bershed 1, 6 million tons of wood, not including most 
saw logs, are deliverable at an average cost of 
$14.37/ton not including stumpage cost (fig. 8). Over 
one-half, or 3 million tons are available from rough 
and rotten trees and tops and branches. These ton- 
nages are shown by type groupings below: 



Maximum average 
delivered cost 



Type 



Pine 

Other softwood 

Aspen 

Northern hardwoods 

Other hardwoods 

All Types 



Volume 

(tons, less most 

sawtimber) 

287,000 

1,086,000 

1,662,000 

1,979,000 

952,000 

5,966,000 



($/ton less 
stumpage 

cost) 

13.57 

13.12 

13.71 

16.18 

13.44 

14.37 



If all material included in the managed harvest 
estimates is harvested, cost per ton will average 
$14.37. If less material is desired, the average price 
will decrease assuming that a mill can avoid stands 
with high harvest and transport costs. For example, 
for all types in timbershed 1, if only half the material 
is needed, average price will drop to $11 .33. 

Although the preponderance of the northern 
hardwood type is clear, its average delivered costs are 
higher than the overall average cost by almost 
$2/ton. However, if only half this type were har- 
vested, 1 million tons could be delivered at an aver- 
age cost of $12.30/ton, a considerable decrease from 
$16.18/ton. 

No timber removal data are available for Mead's 
timbersheds. However, the latest survey of removals 
for Michigan shows that in 1972, 101 million cubic 
feet of growing stock was harvested in the Upper 
Peninsula. This volume converts roughly to 3 million 
tons of all species. But recall that growing stock does 
not include rough and rotten trees, branches, tops, 
nor saplings. From ratios of merchantable 
roundwood to residues determined in this study, we 
know that roughly 0.7 tons of residue are generated 
for every ton of roundwood. Applying this ratio to 
growing-stock removals results in total removals of 
growing stock plus residues of nearly 5 million 
tons/year of all species in the Upper Peninsula. 

Without a breakdown by type we can make no 
precise comparison with managed harvest data, but 
in timbershed 1 alone our calculations show that 7.4 
million tons of wood and wood residues should be 
harvested each year. And this timbershed contains 
only 60 percent of Upper Michigan's commercial for- 
est land. 



25 



(fl 
O 

o 

Q 
Uj 

ec _^ 
£2 
~ O 

r 1 »- 
Ui -v 

oil 

Uj 
(3 

"5 
a 
Uj 



20 



15 - 



10 













WZS^' 












1980^^^-^ 


=^-- 








1 


1977__^_-- 





1 


2 


3 


4 


5 6 



CUMULATIVE VOLUME 
(MILLION TONS) 

Figure 8.— Supply curve for Mead Timbershed 1, all 
forest types. 



13 



Table 6. — Managed harvest by fuel/product classes, Mead timbershed 1 

(In green tons) 









Rough and 


Tops, branches, 






Type/Site Index 


Saw logs 


Pulpwood 


rotten 


saplings 


Total 


Acres 


Jack pine <60 


59,573 


50,038 


4,353 


54,231 


168,195 


7,640 


Jack pine s=60 


6,453 


4,747 


578 


5,746 


17,524 


390 


Red pine 


50,104 


12,691 


8,961 


44,939 


116,695 


5,120 


White pine 


37,983 


11,182 


1,710 


21,907 


72,782 


1,200 


Balsam fir 


191,914 


192,076 


18,098 


230,881 


632,929 


8,170 


White spruce 


31,919 


22,482 


2,636 


30,927 


87,964 


4,570 


Black spruce <40 


9,485 


25,393 


133 


23,269 


58,280 


3,940 


Black spruce &40 


19,035 


12,865 


1,024 


18,496 


51,420 


860 


Tamarack 


4,743 


11,832 


44 


12,033 


28,652 


1,760 


Northern white-cedar 


137,236 


84,473 


23,354 


139,035 


384,098 


5,430 


Aspen <60 


213,850 


434,052 


28,448 


373,311 


1,049,661 


29,080 


Aspen &60 


224,991 


288,434 


35,393 


337,396 


886,214 


14,840 


Paper birch <60 


24,230 


92,096 


10,627 


74,656 


201,609 


4,520 


Paper birch ^60 


6,509 


29,593 


10,639 


29,627 


76,368 


1,910 


Elm-ash-cottonwood 


145,209 


114,455 


40,420 


200,572 


500,656 


10,680 


Oak-hickory <60 


58,103 


144,719 


5,032 


142,930 


350,784 


4,780 


Oak-hickory s=60 


5,066 


6,091 


738 


7,456 


19,351 


140 


Maple-birch-beech <45 


371,874 


364,014 


79,909 


501,439 


1,317,236 


18,670 


Maple-birch-beech =M5 


451,064 


315,633 


94,633 


528,326 


1,389,656 


14,460 


Nonstocked 





289 


4,515 


2,263 


7,067 


8,070 


TOTAL 


2,049,341 


2,217,155 


371,245 


2,779,440 


7,417,141 


146,230 



Demand for wood fiber and fuel 

The mill energy study showed that Mead now uses 
887,000 tons of wood annually, about 100,000 tons of 
which goes to meet approximately 7 percent of its 
own energy needs (table 7). An additional 36 percent 
of Mead's energy requirements are met by black li- 
quor burning and hydroelectricity. Total energy in- 
dependence would require the additional purchase of 
about 1 million tons of wood fuel, bringing total wood 
use to about 1.9 million tons. 

Energy independence for Mead, or for any other 
existing plant, could rarely be achieved in one step. A 
preliminary engineering analysis made for the Mead 
mill shows the opportunities for reaching self-suffi- 
ciency (table 8). 

One question remains: Is there enough wood and 
wood fuel to satisfy these needs at a price Mead can 
afford? 



Supply and demand 

Although Mead does not get all its wood from tim- 
bershed 1, we have used this timbershed to illustrate 
the point that Mead has an abundance of wood at 
affordable prices. The following exercise will bear 
this out. On the horizontal axis of figure 8, locate 
Mead's demand in 1980, the time at which total en- 
ergy independence could be achieved: 1.9 million tons 
for all purposes (fuel and pulp). Draw a vertical line 
from this point on the horizontal axis to the 1980 
supply curve. Then run horizontally to the average 
delivered price less stumpage on the vertical axis: 
$12.50/ton. To this add the average stumpage value 
for all species, calculated earlier: $1.04 + $12.50 = 
$13.54/ton. This indicates that timbershed 1 can 
provide 1.9 million tons each year at an average 
delivered cost per ton including stumpage of $13.54, 
given the assumptions made in the analysis. 



14 



Table 7. — Wood fuel requirements for the ten study mills 









BILLION BTU's REQUIRED PER YEAR 












1 


II 


III IV V 


VI 


VII 


VIII 


IX 


X 


Total 

Current wood-fired 

New wood-fired 


9,688 

837 

8,851 


1,108 

20 

1,088 


5,445 631 3,355 

262 257 

5,445 369 3,098 


2,402 

209 

2,193 


1,290 



1,290 


6,575 

79 

6,496 


1,362 

157 

1,205 


4,885 
1,000 
3,885 


WOOD FUEL REQUIRED 
(Thousand tons per year) 


Total 

Current wood use 

New wood 


1,140 

99 

1,041 


130 

2 

128 


641 74 395 

31 30 

641 43 365 


283 

25 

258 


152 



152 


774 

9 

765 


160 

18 

142 


575 
118 
457 


HIGHEST AFFORDABLE WOOD FUEL PRICE 
(Dollars per million BTU's) 


1977 
1980 
1985 


.77 
1.66 

2.78 


1.51 

2.03 
2.84 


.81 1.70 1.79 
1.02 2.52 2.19 
1.46 3.91 3.07 


1.03 
1.49 
2.34 


1.03 
1.47 
2.41 


1.39 
1.89 
2.93 


1.06 
1.76 

2.91 


1.34 
1.77 
2.63 


(Dollars per green ton) 


1977 
1980 
1985 


6.12 
14.11 
26.63 


12.84 
17.26 
24.14 


6.88 14.45 15.22 

8.67 21.42 18.62 

12.41 33.24 26.10 


8.76 
12.66 
19.89 


8.76 
12.50 
20.48 


11.82 
16.06 
24.90 


9.01 
14.96 
24.74 


11.39 
15.04 
22.36 



Table 8. — Opportunities for reaching self-sufficiency 
at the Mead Mill 

Increase Energy Additional 

Cap. Annual in indepen- fuel 

Step cost saving energy dence wood 

($-10 6 ) (S-10 6 ) (Percent) required (ton/yr.) 



Present wood 












energy use 






— 


43 




Modify an 












existing boiler 


.3 


4.2 


7 


50 


75,000 


Convert a 












boiler 


1.4 


6 


10 


60 


113,000 


Replace a 












boiler 


15.3 


6.0 


34 


94 


769.000 


Replace direct 












heating 


2.0 


6 


6 


100 
TOTAL 


77,000 




1,034,000 



From the mill energy study, the maximum average 
price Mead would be able to pay in 1980 was esti- 
mated at $14.11/ton based on projected price 
increases in fossil fuels (table 7). Thus, by 1980, tim- 
bershed 1 could meet Mead's entire demand well 
within the projected cost limit. It must be emphasized 
that this price will meet not only Mead's fuel require- 
ments, but its fiber needs as well — and with no loss of 
raw material to the sawmill industry. 



Other fuel/product opportunities 

The fiber available from some forest types, espe- 
cially northern hardwoods, would cost significantly 
more than that from others. We have prepared three 
other supply curves to explain why, and to demon- 
strate some other facets of the analysis. 

The first shows the supply of wood and wood resi- 
dues (minus most sawtimber) that resulted from 
mechanized thinning (fig. 9). This harvest technique 
was applied to overstocked, uneven-aged types for 
which average stand diameter was less than 9 inches; 
it was also applied to all overstocked, even-aged 
types. Because of small tree size, full-tree skidding is 
not too damaging and a strip cut can remove the 
excess basal area (Biltonen et al. 1976). 

This curve shows that despite the fairly large ton- 
nages (over 420,000 tons/year), mechanized thinning 
is expensive. To harvest all thinnings would require 
a 1977 price, not including stumpage, of close to 
$13.80/ton, and a 1980 price of $15.81. The large 
volumes and high recovery costs suggest further 
study of ways to reduce the cost of mechanized thin- 
ning. Perhaps the development of specialized equip- 
ment could be justified if the market for fuel and fiber 
were larger. 



15 




500 



CUMULATIVE VOLUME 
(THOUSAND TONS) 

Figure 9. — Supply curve for Mead Timbershed 1. 
thinnings from all forest types. 



The next curve shows the supply of chips when saw 
logs are bucked and sorted before chipping (fig. 10). 
This material resulted from both thinnings and 
clearcuttings, and overlaps material contained in the 
thinning supply curve. Figure 10 illustrates a signifi- 
cant point: 1.6 million tons left after sorting can be 
recovered, at an average cost of $12.33/ton. Thus, a 
huge volume of material is available for fuel or fiber 
at a low price, and over 1 million tons of saw logs are 
saved from the chipper. We need to develop efficient 
ways to carry out saw log sorting and field chipping 
operations. 



25 



20- 



(/> 

O 
O 

Q 

85 s> 15 

0£ 10 
Ui 

o 

£ 5 





1985 




1980 


_ ' ' 


________ — — ' 1977 


1 


i i 



0.50 



1.00 



1.50 



CUMULATIVE VOLUME 
(MILLION TONS) 



Another supply curve represents the cost of relog- 
ging large sawtimber stands in which full-tree skid- 
ding is not recommended (fig. 11). The trees with 
large crowns in these stands would cause unaccepta- 
ble damage to future crop trees if skidded whole to the 
landing. We have assumed that all saw logs will be 
removed in separate, short-log operations. The large 
tops and the other trees to be removed could then be 
harvested with the topwood harvester, grapple skid- 
ders, and conventional chipping at the landing. 

This opportunity is large, but also expensive. The 
supply curve shows that 712,000 tons are deliverable 
at an average price of $20.30/ton in 1977 and 
$23.24/ton in 1980, not including stumpage. For the 
time being at least, much of this wood is out of Mead's 
price range. But here, as with thinning, the need for 
research and development of cheaper ways to recover 
this material is obvious. 




CUMULATIVE VOLUME 
(THOUSAND TONS) 



Figure 10. — Supply curve for Mead Timbershed 1, 
remaining volume after saw log sorting. 



Figure 11. — Supply curve for Mead Timbershed 1, 
relogging opportunities. 



Ten-mill Study Area — Northern 
Wisconsin and Upper Michigan 

Managed harvest area and volumes 

Earlier survey reports for northern Wisconsin and 
Upper Michigan show more than 18 million acres of 
commercial forest land equally distributed between 
the two States. Total inventory of growing stock ap- 
proaches 15 billion cubic feet, and this volume con- 
verts roughly to 700 million green tons, including 
residue. Of the total area, 10.5 million acres, or 58 
percent, is in two types: northern hardwoods and 
aspen. Volumes are similarly distributed — close to 
60 percent of the volume is found in the northern 
hardwood and aspen types. 



16 



Estimates of managed harvest levels for the two 
States show that about 300,000 acres containing 
close to 19.5 million tons including residue could be 
harvested from each (tables 9 and 10). However, 
there are some differences in the distribution of 
managed harvest by type: northern Wisconsin has a 
recommended harvest for aspen of 5.4 million tons 
including residues, compared with 3.2 million for 
Upper Michigan. But northern Wisconsin has a rec- 
ommended harvest of only 5.5 million tons of north- 
ern hardwoods, compared with 10.3 million tons for 
Upper Michigan. Northern Wisconsin also has more 
oak-hickory and elm-ash-cottonwood than Upper 
Michigan, 4.4 million tons to 1.9 million tons. 

Three more supply curves were constructed for the 
regional study area: (1) northern Wisconsin, all 
types, (2) Upper Michigan, all types, and (3) both 
regions, all types (figs. 12, 13, and 14). 

In northern Wisconsin, 15.9 million tons of fiber 
and fuel wood (most sawtimber not included) can be 
delivered at an average cost of $14.42/ton in 1977 and 
$16.51/ton in 1980, not including stumpage cost. For 
Michigan, a similar comparison shows 14.5 million 
tons at an average delivered cost of $13.62/ton in 
1977 and $15.59/ton in 1980. For both regions com- 
bined, 30.5 million tons are deliverable at an average 



cost of $14.05/ton in 1977 and $16.07/ton in 1980. 
Again, stumpage costs have not been included in any 
of these comparisons, and the tonnages do not include 
most of the saw logs, which were assumed to be recov- 
ered during saw log sorting or prior to relogging. 

Regional demand 

The 10 pulp and paper mills in the study area now 
use 2.44 million tons of wood, 13.6 percent or 332,000 
tons of which goes to meet their own energy needs. If 
all study mills were able to achieve energy indepen- 
dence, a total of 6.43 million tons would be required, 
4.32 million tons of which would be used as wood fuel. 

The opportunities for achieving full energy inde- 
pendence vary widely among mills. The brief engi- 
neering cost analysis for each mill suggested that 5 of 
the 10 could now justify some increased use of wood 
for fuel at a cost of $10/green ton or less (table 7). 
Further incentives may be provided by government 
policies affecting fuel prices, fuel availability, and 
investment tax credits. 

By 1980 all mills except one will probably find 
wood fuel attractive. These nine mills presently use 
2.39 million tons of wood annually. For energy inde- 
pendence, they would need an additional 3.35 million 
tons of wood annually for a total of 5.74 million tons of 



Table 9. — Managed harvest by fuel/product classes, northern Wisconsin 

(In green tons) 









Rough and 


Tops, branches, 






Type/Site Index 


Saw logs 


Pulpwood 


rotten 


saplings 


Total 


Acres 


Jack pine <60 


128,842 


187,696 


19,400 


217,789 


553,727 


8,440 


Jack pine ^60 


53,847 


58,921 


6,608 


71,856 


191,232 


2,900 


Red pine 


294,403 


91,440 


19,777 


237,435 


643,055 


14,430 


White pine 


166,537 


19,258 


15,665 


86,113 


287,573 


4,450 


Balsam fir 


293,656 


321,940 


38,325 


493,393 


1,148,314 


14,200 


White spruce 


5,327 


10,037 


1,270 


12,666 


29,300 


1,020 


Black spruce <40 


8,450 


14,894 


1,349 


58,235 


82,928 


2,930 


Black spruce 3=40 


11,998 


7,440 


3,763 


21,098 


44,299 


820 


Tamarack 


10,006 


46,584 


5,889 


50,367 


112,846 


4,910 


Northern white-cedar 


60,016 


51 ,835 


9,235 


82,179 


203,265 


3,950 


Aspen <60 


200,339 


567,511 


91,876 


688,586 


1,548,312 


34,730 


Aspen 3=60 


609,310 


1,440,085 


206,939 


1,598,304 


3,854,638 


57,370 


Paper birch <60 


61,885 


151,273 


14,553 


170,870 


398,581 


6,010 


Paper birch 3=60 


81,029 


186,198 


41,626 


210,435 


519,288 


9,650 


Elm-ash-cottonwood 


327,214 


285,334 


183,169 


614,211 


1,409,928 


37,140 


Oak-hickory <60 


368,895 


635,377 


311,681 


1,106,666 


2,422,619 


42,190 


Oak-hickory 3=60 


171,688 


119,823 


83,655 


234,606 


609,772 


9,670 


Maple-birch-beech <45 


150,045 


139,834 


73,566 


328,954 


692,399 


12,980 


Maple-birch-beech 3=45 


1,515,844 


1,059,042 


349,566 


1,860,487 


4,784,939 


48,480 


Nonstocked 


4,110 


3,695 


27,028 


39,811 


74,644 


18,380 


TOTAL 


4,523,441 


5,398,217 


1,504,940 


8,184,061 


19,610,659 


334,650 



17 



Table 10. — Managed harvest by fuel/product classes, Upper Michig 


an 








(In 


green tons) 














Rough and 


Tops, branches, 






Type/Site Index 


Saw logs 


Pulpwood 


rotten 


saplings 


Total 


Acres 


Jack pine <60 


142,186 


197,598 


12,879 


179,296 


531,959 


17,180 


Jack pine 2=60 


8,990 


25,759 


851 


20,838 


56,438 


800 


Red pine 


162,189 


79,467 


25,195 


167,994 


434,845 


15,990 


White pine 


88,304 


28,430 


3,164 


51,403 


171,299 


2,750 


Balsam fir 


369,657 


339,073 


34,733 


441,517 


1,184,980 


19,120 


White spruce 


55,195 


42,527 


3,416 


56,763 


157,901 


5,350 


Black spruce <40 


63,451 


83,309 


2,779 


93,884 


243,423 


10,290 


Black spruce 2=40 


51,703 


27,600 


3,121 


46,942 


129,366 


2,550 


Tamarack 


25,881 


22,641 


1,817 


28,100 


78,439 


4,150 


Northern white-cedar 


357,545 


191,204 


46,177 


333,768 


928,694 


15,250 


Aspen <60 


408,320 


677,383 


51,848 


625,087 


1,762,638 


47,250 


Aspen &60 


344,926 


486,300 


41,091 


544,361 


1,416,678 


21,510 


Paper birch <60 


56,960 


139,094 


13,986 


118,478 


328,428 


7,080 


Paper birch 2=60 


42,200 


55,424 


10,908 


62,949 


171,481 


2,160 


Elm-ash-cottonwood 


409,768 


311,565 


75,899 


536,308 


1,333,540 


26,350 


Oak-hickory <60 


89,464 


217,625 


14,597 


236,978 


558,664 


7,970 


Oak-hickory 2=60 


5,066 


14,181 


738 


16,056 


36,041 


590 


Maple-birch-beech <45 


1,806,138 


1,086,079 


299,053 


1,936,209 


5,127,479 


62,790 


Maple-birch-beech 2=45 


1,936,779 


977,994 


302,873 


1,921,446 


5,139,092 


47,960 


Nonstocked 





1,951 


6,341 


3,905 


12,197 


9,980 


TOTAL 


6,424,722 


5,005,204 


951,376 


7,422,282 


19,803,582 


327,070 




5 10 

CUMULATIVE VOLUME 
(MILLION TONS) 



15 



20 



Figure 12. — Supply curve for northern Wisconsin, all 
forest types. 




CUMULATIVE VOLUME 
(MILLION TONS) 

Figure 13. — Supply curve for Upper Michigan, all 
forest types. 



18 



CO 

O 
O 

Q 
lu 
QC 
lu 

UJ 

Q 
UJ 
CD 

QC 
UJ 



o 

*9 



25 



20 - 



15 - 



10 - 



5 - 





1985^^^ 




1980 ^ 




_________^— ' 1977 ^ 


I I 


I l l l 



10 



15 



20 



25 



30 



CUMULATIVE VOLUME 
(MILLION TONS) 

Figure 14. — Supply curve for all regions, all forest types. 



fiber and fuel. Thus, in 1980, the increased demand of 
nine mills plus the existing demand of the tenth mill 
could equal 5.79 million tons for all needs. Given 
these assumptions, the following questions remain: 
Can the existing forest resource sustain these de- 
mands, and most important, at a price the mills can 
afford? 

Table 7 shows maximum affordable prices that 
each mill can pay in 1977, in 1980, and in 1985. It 
seems that even in 1977 at least five mills could 
afford to pay more than $10/ton to achieve indepen- 
dence. In 1980, the demand of the 10 mills, if nine 
achieve independence, is estimated to be 5.8 million 
tons. Using the 1980 curve in figure 14, it can be seen 
that this 5.8 million tons of fuel and fiber can be 
delivered at a cost of $10.50/ton + $1.04/ton = 
$11.54/ton, including stumpage. This price is well 
below the maximum that nine of the mills can afford. 
And this volume is far below the 30.5 million tons of 
material suitable for fiber or fuel that our calcula- 
tions suggest should be harvested each year. 



SUMMARY 

The Mead case study and the regional analysis 
indicate that both fuel and fiber needs can be sup- 
plied from the forest resource with significant cost 
savings. Mead now uses almost 900,000 green tons 



each year; about 100,000 tons of this provide 7 per- 
cent of its energy needs. Achieving energy indepen- 
dence would require an additional 1 million tons, 
bringing total wood use to 1.9 million tons. From a 
volume point of view, this projected consumption can 
be favorably compared to the recommended harvest 
of 6 million green tons from only one of Mead's seven 
timber sheds. 

The analysis of costs in this same timbershed sug- 
gests that the 1.9 million tons could be harvested and 
delivered for $13.54 per green ton in 1980. This is 
$0.57 per ton less than the wood's equivalent fossil 
fuel value based on projected fossil fuel prices in 
1980. Thus, achieving energy independence would 
result in direct energy cost savings to Mead at these 
same price and volume levels of at least $570,000 per 
year. 

For the broader analysis of 10 pulp and paper mills, 
current wood consumption is 2.4 million green tons of 
which 330,000 tons go for energy. Energy self suffi- 
ciency by all 10 mills would require the added har- 
vest of 4.0 million tons, bringing total wood use to 6.4 
million tons. 

Our comparison of regional supply and the specific 
opportunities for using more wood fuel suggests that 
only 9 of these 10 mills could economically achieve 
independence by 1980. Total wood use for this situa- 
tion would be 5.8 million green tons. Again, this level 



19 



must be compared to the total recommended harvest 
for the study region of more than 30 million green 
tons of fiber and fuel material. 

The analysis also describes in detail several key 
residue opportunities that are not now being exploit- 
ed: ( 1 ) thinnings from pole timber stands, and (2) tops 
and nonsaw log trees from selectively harvested 
sawtimber stands. In Mead's largest timbershed for 
example, these two opportunities could yield over 1 
million green tons each year, over 17 percent of the 
total recommended harvest. However, existing har- 
vest systems are not designed for these tasks and 
these harvest cost estimates are significantly higher 
than the costs of clearcutting. However, research and 
development of new equipment and methods for 
these improvement harvests should be able to reduce 
costs and permit these over-stocked stands to achieve 
even higher productivity and tree quality in the 
future. 

A key factor in this analysis was the managed 
harvest procedure which estimated annual potential 
harvests based on existing forest conditions; it found 
some 30 million green tons suitable for either fiber or 
fuel or 39 million green tons if sawtimber is included. 
However, it must be emphasized that these estimates 
are in no way an upper limit on the region's forest 
potential. It is easily conceivable that in a regulated 
condition, northern Wisconsin and Upper Michigan 
could produce two to three times as much wood as 
reported here. 

Closely related to the issue of increasing productiv- 
ity, is a need to change residue definitions. Classify- 
ing trees as saw logs or pulpwood when they will fall 
down before they are needed obscures more realistic 
opportunities. Indeed, their use as a fuel now if no 
other market exists is especially appropriate. Cut- 
ting more to produce more, is not a contradiction, but 
is the principal means of achieving truly productive 
forests in the future. 



SELECTED BIBLIOGRAPHY 

Adams, T. C, and R. C. Smith. 1976. Review of the 
logging residue problem and its reduction through 
marketing practices. U.S. Department of Agricul- 
ture Forest Service, General Technical Report 
PNW-48, 22 p. U.S. Department of Agriculture 
Forest Service, Pacific Northwest Forest and 
Range Experiment Station, Portland, Oregon. 



Biltonen, F. E., W. A. Hillstrom, H. Steinhilb, and R. 
M. Godman. 1976. Mechanized thinning of north- 
ern hardwood pole stands: methods and economics. 
U.S. Department of Agriculture Forest Service, 
Research Paper NC-137, 17 p. U.S. Department of 
Agriculture Forest Service, North Central Forest 
Experiment Station, St. Paul, Minnesota. 

Blyth, James E., Allan H. Boelter, and Carl W. Dan- 
ielson. 1975. Primary forest products industry and 
timber use, Michigan, 1972. U.S. Department of 
Agriculture Forest Service, Resource Bulletin NC- 
24, 24 p. U.S. Department of Agriculture Forest 
Service, North Central Forest Experiment Station, 
St. Paul, Minnesota. 

Blyth, James E., E. F. Landt, J. W. Whipple, and J. T. 
Hahn. 1976. Primary forest products industry and 
timber use, Wisconsin, 1973. U.S. Department of 
Agriculture Forest Service, Resource Bulletin NC- 
31, 61 p. U.S. Department of Agriculture Forest 
Service, North Central Forest Experiment Station, 
St. Paul, Minnesota. 

Boyd, J. H., and W. P. Novak. 1977. A method of 
comparing logging system and machine costs. Spe- 
cial report No. 2, 11 p. Forest Engineering Re- 
search Institute of Canada, Pointe Claire, P.Q. 

Bradley, Dennis P. 1972. Improve forest inventory 
with access data — measure transport distance and 
cost to market. U.S. Department of Agriculture 
Forest Service, Research Paper NC-82, 21 p. U.S. 
Department of Agriculture Forest Service, North 
Central Forest Experiment Station, St. Paul, Min- 
nesota. 

Bradley, Dennis P., and Sharon Winsauer. 1976a. 
Solving wood chip transport problems with com- 
puter simulation. U.S. Department of Agriculture 
Forest Service, Research Paper NC-138, 8 p. U.S. 
Department of Agriculture Forest Service, North 
Central Forest Experiment Station, St. Paul, Min- 
nesota. 

Bradley, Dennis P., and Sharon Winsauer. 1976b. A 
full tree field chipping and trucking system simu- 
lator using GPSS. In Ninth Ann. Simulation Sym- 
posium Proceedings, p. 197-204. Tampa, Florida, 
March 17-19, 1976. 

Bradley, Dennis P., Frank E. Biltonen, and Sharon 
Winsauer. 1976. A computer simulation of full-tree 
field chipping and trucking. U.S. Department of 
Agriculture Forest Service, Research Paper NC- 
129, 14 p. U.S. Department of Agriculture Forest 
Service, North Central Forest Experiment Station, 
St. Paul, Minnesota. 



20 



International Business Machine Corporation. 1973. 
General purpose simulation system V users' man- 
ual, SH20-0851-1, 422 p. White Plains, New York. 

Keays, J. L. 1975. Biomass of forest residuals. 
AIChE, Symposium Series 71(146):10-21. 

Schlaegel, B. E. 1974. Estimating aspen weight and 
volume for individual trees, diameter classes, or 
entire stands. U.S. Department of Agriculture For- 
est Service, General Technical Report NC-20, 16 p. 
U.S. Department of Agriculture Forest Service, 
North Central Forest Experiment Station, St. 
Paul, Minnesota. 

Spencer, John S., Jr., and Ray E. Pfeifer. (n.d., circa 
1970). The growing timber resource of Michigan, 
1966, western Upper Peninsula. Michigan Depart- 
ment of Natural Resources, 92 p. 

Spencer, John S., Jr., and Harry W. Thorne. 1972. 
Wisconsin's 1968 timber resource — a perspective. 
U.S. Department of Agriculture Forest Service, 
Resource Bulletin NC-15, 80 p. U.S. Department of 
Agriculture Forest Service, North Central Forest 
Experiment Station, St. Paul, Minnesota. 

Stone, Robert N. 1976. A comparison of woodland 
owners' intent with woodland practice in Michi- 
gan's Upper Peninsula. 113 p. Ph.D. Thesis, School 
of Forestry, University of Minnesota, St. Paul, 
Minnesota. 

U.S. Department of Agriculture, Forest Service. 
1973. Weights of various woods grown in the 
United States. U.S. Department of Agriculture 
Forest Products Laboratory, Technical Note FPL- 
218, Madison, Wisconsin. 



U.S. Department of Agriculture, Forest Service. 
1973. The outlook for timber in the United States. 
U.S. Department of Agriculture Forest Service, 
Research Report 20, 367 p. 

U.S. Department of Agriculture, Forest Service. 
1976. Final report on feasibility of utilizing forest 
residues for energy and chemicals. A report to the 
National Science Foundation and the Federal En- 
ergy Administration, Washington, D.C. NSF-RA- 
760013. 

Wahlgren, H. Gus, and Thomas H. Ellis. 1978. Poten- 
tial resource availability with whole tree utiliza- 
tion. TAPPI 61(ll):37-39. 

Wisconsin Forest Products Price Review. 1977. 
Boltwood edition, Wisconsin Agricultural Experi- 
ment Station, University of Wisconsin, Madison, 
Wisconsin. 

Young, Harold E. 1977. Forest biomass inventory: 
The basis for complete tree utilization. 12 p. Com- 
plete Tree Institute. School of Forest Resources, 
University of Maine, Orono, Maine. 

Young, H. E., Leigh Hoar, and Marshall Ashley. 
1965. Weight of wood substance for components of 
seven tree species. TAPPI 48(81:466-469. 

Young, H. E., and A. J. Chase. 1965. Fiber weight and 
pulping characteristics of the logging residue of 
seven tree species in Maine. Technical Bulletin 17, 
Maine Agricultural Experiment Station, Univer- 
sity of Maine, Orono, Maine. 



■& U.S. Government Printing Office: 1980—669-122/37 Region Nov 6 



21 



Bradley, Dennis P., Eugene M. Carpenter, James A. Mattson, Jerold T. 
Hahn, and Sharon A. Winsauer. 

1980. The supply and energy potential of forest resources in northern 
Wisconsin and Michigan's Upper Peninsula. U.S. Department of Agri- 
culture Forest Service, Research Paper NC-182, 21 p. U.S. Depart- 
ment of Agriculture Forest Service, North Central Forest Experiment 
Station, St. Paul, Minnesota. 

Analyzes the economic potential of achieving energy independence 
by 10 pulp and paper mills in northern Wisconsin and Upper Michi- 
gan. Independence would require the annual harvest of 5.79 million 
green tons for both fiber and fuel wood needs, compared to a recom- 
mended harvest level of 31 million green tons. Delivered wood cost 
projections seem well within affordable levels. 

KEY WORDS: energy, wood fuel, forest residues, energy indepen- 
dence. 



Bradley, Dennis P., Eugene M. Carpenter, James A. Mattson, Jerold T. 

Hahn, and Sharon A. Winsauer. 

1980. The supply and energy potential of forest resources in northern 
Wisconsin and Michigan's Upper Peninsula. U.S. Department of Agri- 
culture Forest Service, Research Paper NC-182, 21 p. U.S. Depart- 
ment of Agriculture Forest Service, North Central Forest Experiment 
Station, St. Paul, Minnesota. 

Analyzes the economic potential of achieving energy independence 
by 10 pulp and paper mills in northern Wisconsin and Upper Michi- 
gan. Independence would require the annual harvest of 5.79 million 
green tons for both fiber and fuel wood needs, compared to a recom- 
mended harvest level of 31 million green tons. Delivered wood cost 
projections seem well within affordable levels. 

KEY WORDS: energy, wood fuel, forest residues, energy indepen- 
dence. 



Soil is for plants.. 




not for tire tracks, 



;iSDA FOREST SERVICE 
RESEARCH PAPER NC-183 



I 





GOVT. DOCUMENTS 

DEPOS5TORY STEM 

JUl 28 138Q 

CtEMSON 

U8RARX 



Yield and quality of seed from 

YELLOW BIRCH 
PROGENIES 

Knud E. Clausen 



Mh Central Forest Experiment Station 
F'est Service, U.S. Department of Agriculture 



Clausen, Knud E . 

1980. Yield and quality of seed from yellow birch 
progenies. U.S. Department of Agriculture Forest 
Service, Research Paper NC-183, 5 p. U.S. Department 
of Agriculture Forest Service, North Central Forest 
Experiment Station, St. Paul, Minnesota. 



Seed yield in 8- and 9-year-old yellow birch 



varied among families and 
than 1,500 seeds per tree, 
more seed than short ones, 
to insufficient pollination 
trees in flowering phenology. 



years but averaged more 

Long catkins contained 

Seed quality was poor due 

and to differences among 



KEY WORDS: Provenances, climate catkin dimensions, 
correlations, pollination, phenology, germination. 



Clausen, Knud E. 

1980. Yield and quality of seed from yellow birch 
progenies. U.S. Department of Agriculture Forest 
Service, Research Paper NC-183, 5 p. U.S. Department 
of Agriculture Forest Service, North Central Forest 
Experiment Station, St. Paul, Minnesota. 



Seed yield in 8- and 
varied among families and 
than 1,500 seeds per tree, 
more seed than short ones, 
to insufficient pollination 
trees in flowering phenology. 



9-year-old yellow birch 

years but averaged more 

Long catkins contained 

Seed quality was poor due 

and to differences among 



KEY WORDS: Provenances, climate catkin dimensions, 
correlations, pollination, phenology, germination. 



North Central Forest Experiment Station 

Robert A. Hann, Director 

Forest Service - U.S. Department of Agriculture 

1992 Folwell Avenue 

St. Paul, Minnesota 55108 



Manuscript approved for publication July 25, 1979 

1980 



YIELD AND QUALITY OF SEED FROM YELLOW 

BIRCH PROGENIES 



Knud E. Clausen, Principal Plant Geneticist, 
Carbondale, Illinois 



Seed production in open-grown yellow birch (Be- 
tula alleghaniensis Britt. ) was found to begin at age 7 
and to increase with age and size of the trees (Clausen 
19771. Although earliness of fruiting and proportion 
of fruiting trees varies among and within families, 
the presence and abundance of catkins is mostly de- 
termined by the size of the tree crowns (Clausen 
1979). However, early and abundant production of 
fruiting catkins will not necessarily result in a good 
crop of viable seeds. The purpose of this study was to 
determine the yield and quality of the seed produced 
by these young trees and possible relations of catkin 
size and shape to these characteristics. 

METHODS 

Yellow birch progenies were planted as 4-year-old 
seedlings in the spring of 1972 near Lake Tomahawk 
in Oneida County, north-central Wisconsin. The site 
is on Padus sandy loam and was clean-cultivated 
through the summer of 1975 and then mowed (Clau- 
sen 1977). Trees were planted ataspacingof8x8feet 
(2.4 m) and the plantation contained 155 open- 
pollinated families in 4-tree square plots and 4 
replications (table 1). Fruiting began in the fall of 
1974 but the number of fruiting trees and families 
was not large enough for the study until the following 
year. Therefore, the data presented here are for 
catkins and seeds produced in 1975 and 1976. 

In each year, fruiting catkins were collected and 
allowed to dry for several weeks before length and 



diameter of from 1 to 20 catkins per tree were mea- 
sured. Each catkin was then broken apart and the 
number of seeds counted and recorded. In 1975, all 
seeds from a tree were mixed and a sample of 100 
seeds drawn for use in a germination test. Four 100- 
seed samples per tree were counted out from the 1976 
seed crop, placed on type M film, and exposed to soft 
X-rays (15 kV, 1mA, 80 sec). The radiographs were 
examined and the number of seeds with normal-ap- 
pearing ovules, small or faintly opaque ovules (prob- 
ably inviable), and empty seeds recorded. Both the 
1975 and 1976 seedlots were tested for germination 
during the following winter by placing the seed on 
moist perlite in Petri dishes, which were kept under 
20-hr photoperiod in a greenhouse at 24C for 30 days. 

The catkin data were subjected to analysis of 
variance using tree means and a one-way nested 
classification with unequal sample sizes. Catkin 
diameter/length and catkin volume were calculated 
because the number of seeds per catkin may be re- 
lated to catkin shape and volume as well as to catkin 
dimensions. Catkin volume was determined as that 
of an ellipsoid. Simple correlations based on 
individual catkins were calculated in order to estab- 
lish possible relations between catkin size, catkin 
shape, number of seeds per catkin, percent filled 
seeds, and percent germination. Exceptions were 
those involving 1975 germination data where only 
tree averages were available. All percentage values 
used in the calculations were first converted by 
arcsin transformation. 



Table 1. — Origin of 21 yellow birch stands and number of open -pollinated families included in 

plantation 



Stand 


State or Province 


Origin 
code 




Location 


Number of 


No. 


Lat. 


Long. 


families 










Donrop .............. 






Uuyfuu ----- — 




3241 


Nova Scotia 


NS 


44.6 


60.5 


8 


3063 


Nova Scotia 


NS 


44.1 


65.8 


7 


3066 


New Brunswick 


NB 


47.4 


65.2 


6 


2998 


Quebec 


PQ 


48.2 


70.2 


8 


3000 


Quebec 


PQ 


47.5 


75.0 


9 


3004 


Ontario 


ON 


46.7 


79.6 


8 


3309 


Ontario 


ON 


47.5 


84.8 


10 


2977 


Maine 


ME 


44.8 


68.6 


8 


2986 


New Hampshire 


NH 


43.5 


71.4 


4 


2971 


Massachusetts 


MA 


42.7 


73.2 


5 


3312 


Pennsylvania 


PA 


41.6 


78.7 


6 


3299 


Virginia 


VA 


37.8 


79.1 


5 


2959 


North Carolina 


NC 


35.7 


82.3 


4 


2973 


Georgia 


GA 


34.8 


83.8 


5 


2983 


Illinois 


IL 


41.9 


89.4 


9 


2962 


Wisconsin 


Wl 


44.9 


87.2 


8 


4340 


Wisconsin 


Wl 


45.6 


88.6 


8 


2968 


Wisconsin 


Wl 


46.5 


92.1 


9 


2987 


Michigan 


Ml 


47.0 


88.7 


10 


2967 


Minnesota 


MN 


47.8 


90.2 


9 


2964 


Minnesota 
Total 


MN 


44.2 


94.1 


9 

155 



RESULTS 



Seed Yield and Quality 

Total seed yield was found to be determined by: 
(1) geographic origin, (2) number of seed-bearing 
trees per family, (3) number of fruiting catkins per 
tree, (4) number of seeds per catkin, and (5) climate. 
It may, therefore, vary greatly among and within 
families and from year to year. 

Fruiting was poorest in yellow birch from the 
Maritime provinces. Trees of New Brunswick origin 
set no seed in either 1975 or 1976 and those from 
Nova Scotia stand 3063 did not fruit in 1976 (table 2). 
The best seed producers were generaly Wisconsin 
and Minnesota trees that originated within 250 miles 
of the plantation. Trees of northern origin (latitude 
44.5° N or above) began to bear seed at an earlier age 
than those of southern origin (Clausen 1977). Seed 
yield followed a similar pattern (table 3). In both 



years, trees originating east of longitude 80° W aver- 
aged more catkins than western trees but had no 
more seeds per catkin. Geographic origin may thus 
have an effect on early seed yield in this species. 

Fruiting trees accounted for 4 percent of the sur- 
viving trees in both 1975 and 1976. Among the seed- 
bearing families in 1975, the number of fruiting trees 
per family varied from 1 to 12 or from 2 to 20 percent 
of the trees (table 2). Families with few survivors had 
higher but atypical percentages of fruiting trees. In 
1976, from 1 to 4 trees per family or from 3 to 12 
percent of the trees fruited. Similarly, the number of 
catkins per tree ranged from 1 to 196 in 1975 and 
from 1 to 188 in 1976. The total number of fruiting 
families, trees, and catkins was greater in 1975 than 
in 1976, possibly due to a severe drought in 1976. 

In 1975 the number of seeds per catkin averaged 
139 and ranged from 30 to 297 in the 50 families. The 
44 families fruiting in 1976 ranged from 9 to 306 
seeds per catkin and averaged 151 (table 4). Families 
from different stands also showed much variation 



Table 2. — Incidence of fruiting in yellow birch trees 
from 21 origins in 2 years 



Table 4.— Total seed yield in 1975 and 1976 seed 
crops of yellow birch families 





Fruiting in 1975 


Fruiting in 1976 




Stand 


1975 families 
Seeds per catkin 


1976 
Seeds 
Mean 

iber 

250 


families 


Stand 


Families' 

Percent 
13 


Trees per Catkins 
family per tree 

Number 
1 2 


Families' 

Percent 
22 


Trees per 
family 

Nu 
1 


Catkins 

per free 

Tiber 
1 


per catkin 


No. Origin 


No. 

3241 


Origin 

NS 


Mean 


Range 


Range 






3241 NS 


112 


112 


250 


3063 NS 
3066 NB 


43 



1-2 


1-14 










3063 


NS 


121 


87-165 






2998 PQ 


25 


1-2 


2-60 


12 


1 


4 


2998 


PQ 


111 


52-170 


222 


222 


3000 PQ 


33 


1-3 


1-12 


33 


1 


2-6 


3000 


PQ 


85 


30-128 


196 


174-219 


3004 ON 


12 


1 


27 


1? 


1 


4 


3004 


ON 


190 


190 


180 


180 


3309 ON 


30 


1 


1-7 


20 


1 


5 


3309 


ON 


136 


106-152 


84 


9-160 


2977 ME 


25 


1-3 


5-9 


50 


1-2 


1-23 


2977 


ME 


159 


108-209 


118 


45-211 


2986 NH 


50 


1-3 


1-23 


50 


2 


1-40 


2986 


NH 


155 


129-181 


120 


64-175 


2971 MA 


20 


1 


30 


20 


1 


1 


2971 


MA 


244 


244 


59 


59 


3312 PA 


33 


1 


1-4 


17 


2 


1-3 


3312 


PA 


85 


59-112 


110 


110 


3299 VA 
2959 NC 


40 

25 


1 ? 
2 


11-36 
1-4 


40 
25 


1 
1 


3-12 
10 


3299 


VA 


138 


113-163 


164 


158-170 


2973 GA 


40 


1 


6-14 


40 


1 


1-3 


2959 


NC 


148 


148 


171 


171 


2983 IL 


33 


1-2 


2-17 


44 


1-3 


1-65 


2973 


GA 


61 


45-77 


125 


125 


2962 Wl 


50 


1-4 


1-7 


62 


1-3 


1-7 


2983 


IL 


105 


69-131 


137 


84-189 


4340 Wl 


50 


1-2 


2-27 


38 


1-2 


1-32 


2962 


Wl 


173 


108-297 


134 


108-166 


2962 Wl 


33 


1-3 


2-196 


44 


1-2 


1-188 


4340 


Wl 


126 


63-228 


163 


131-222 


2987 Ml 


20 


1-12 


1-56 


10 


4 


1-10 


2968 


Wl 


152 


73-220 


187 


119-283 


2967 MN 


44 


1-2 


1-19 


33 


1-3 


1-16 


2987 


Ml 


157 


135-179 


98 


98 


2964 MN 


56 


1-6 


1-16 


33 


1-2 
65 


1-7 
672 


2967 
2964 


MN 
MN 


182 

156 


142-238 
94-295 


177 

168 


22-306 


Total 


85 


1,031 


124-192 


'Fifty families fruited in 1975, 44 in 1976. 








Range of trees 




18-297 




9-314 



Table 3.- 



- Seed yield in yellow birch of northern and 
southern origin in 2 years 



Fruiting 


Average number 
of catkins 


Average number of 
seeds/catkin 


stands 


1975 1976 


1975 1976 


11 northern 
9 southern 


66 39 

35 28 


143 164 
134 132 



with a range from 28 to 131 percent of the stand 
means in 1975 and from 8 to 178 percent in 1976. As 
expected, between- and within-stand variation 
among the individual trees was even greater (table 
4), but neither stand nor family differences were sta- 
tistically significant in either year. The number of 
seeds in individual catkins ranged from 4 to 305 in 
1975 and from 3 to 399 in 1976. Variation among 
catkins within a tree was usually less than that 
among trees. 

In contrast to the fair to good seed yield in both 
years, seed quality was generally poor. Average 
germination percentage of the 1975 seed ranged from 
in 5 families and 10 percent or less in 27 families up 
to 38 in one family (table 5). Among the individual 
trees it varied from in 8 trees and 10 percent or less 



Table 5. — Germination of 1975 and 1976 yellow birch 
seed crops and filled seed percentage of 1976 seed 



Stand 



1975 families 
Germination 



1976 families 



Filled seeds 



Germination 



No. 


Origin 


Mean 


Range 


Mean 


Range 


Mean 


Range 
















3241 


NS 


10 


10 














3063 


NS 


8 


4-12 


— 


— 


— 


— 


2998 


PQ 


4 


0-8 


12 


12 


2 


0.2 


3000 


PQ 


3 


1-6 


14 


0-24 


9 


0-17 


3004 


ON 


19 


19 


11 


11 


8 


8 


3309 


ON 


4 


2-7 


17 


8-26 


16 


6-26 


2977 


ME 


14 


8-20 


21 


9-27 


11 


4-17 


2986 


NH 


6 


0-13 


10 


8-12 


3 


2-4 


2971 


MA 


12 


12 


8 


8 


8 


8 


3312 


PA 


4 


3-4 


4 


4 


4 


4 


3299 


VA 


5 


0-9 


8 


6-11 


5 


3-8 


2959 


NC 


2 


2 














2973 


GA 








03 


03 








2983 


IL 


24 


8-38 


24 


11-44 


1 


4-13 


2962 


Wl 


13 


6-23 


12 


22 


3 


7 


4340 


Wl 


7 


4-9 


9 


8-18 


8 


0-16 


2968 


Wl 


10 


4-16 


11 


0-28 


9 


0-23 


2987 


Ml 


11 


7-14 


10 


10 


5 


5 


2967 


MN 


11 


0-18 


15 


2-34 


5 


0-14 


2964 


MN 


12 


1-19 


40 


24-61 


18 


4-34 


Range 


of trees 




0-62 




0-67 




0-52 



in 44 up to a maximum of 62 percent. As a result, the 
average germination was only 9 percent compared 
with 90 percent or better for mature trees in good 
seed years. The amount of filled seed in the 1976 seed, 



as determined by X-radiography, averaged only 10 
percent for the 44 families and was extremely vari- 
able (table 5). Six families had only empty seed while 
34 families had 30 percent or less filled seed and 4 
families had more than 30 percent. Twelve trees had 
no filled seed, 18 had up to 10 percent, 25 had 11-30, 6 
had 31-50, and 3 had more than 50 percent filled seed. 
In contrast, filled seed percentages exceeding 90 per- 
cent are common in natural stands in good seed 
years. Germination percentages were generally 
lower than filled seed percentages and ranged from 
in 8 families and less than 10 percent in 26 families 
up to 34 percent in 1 family (table 5). The best tree 
had 52 percent germination, but the average for the 
1976 seed crop was only 6 percent. 



Catkin Characteristics and 
Relations to Seed 
Yield and Quality 

On the average, catkin length and diameter were 
greater in 1975 than in 1976 (table 6). The smaller 
catkin dimensions in 1976 may be due to the drought 
because moisture availability affects the size of plant 
parts (Kozlowski 1971). Much variation among 
stands, families, and trees was recorded in both years 
but the analysis of variance showed no significant 
differences among seed origins or families in either 
year. The 1975 analysis included 74 trees of 43 fami- 
lies from 20 stands while the 1976 analysis included 
65 trees of 43 families from 19 stands. As shown 
previously, the analyses of number of seeds per cat- 
kin gave similar results. The lack of significance can 
probably be ascribed to the fact that unequal num- 
bers of trees per family and of families per stands 
were used in the analyses with the attendant reduc- 
tion in degrees of freedom. Because catkin dimen- 
sions were greater in 1975 than in 1976, both the 
diameter/length ratio and catkin volume followed 
the same pattern (table 6). 



Table 6. — Characteristics of catkins collected from 8- 
and 9 -year -old yellow birch in 1975 and 1976 



Catkin 


1975 families 


1976 families 


characteristic 


Mean 


Range 


Mean 


Range 


Length (mm) 
Diameter (mm) 
Diam /length 
Volume (cm 3 ) 


24.3 
15.2 
.629 
323 


13 3-36.3 

10.0-24.3 
.401-902 
0.75-6.85 


21.1 
10.0 
.484 
1.27 


12.3-34.2 

5.5-14 

.269-707 

0.32-2.72 



Correlations between catkin characteristics and 
seed yield were higher in 1975 than in 1976. In both 
years, number of seeds per catkin showed the closest 
relation to catkin length and catkin volume (table 7). 
One might expect a good correlation between catkin 
diameter and number of seeds because catkin diame- 
ter and volume were highly correlated in both years. 
This was true in 1975 but not in 1976. The apparent 
reason is that the catkins were proportionately thin- 
ner in 1976 as evidenced by the smaller diameter/ 
length ratio in that year (table 6) and also by the poor 
correlation between catkin length and diameter, 
which contrasts to the close relation between them in 
1976 (table 7). Catkin shape, as measured by the 
diameter/length ratio, was unrelated to catkin vol- 
ume and was a fair to poor indicator of seed yield 
(table 7). 

Of the characteristics measured, catkin length was 
the best indicator of potential seed yield but not of 
seed quality. Catkin length was poorly correlated 
with percent filled seed in 1976 and with seed germi- 
nation in both 1975 and 1976 (table 7). The correla- 
tion between catkin length and percent germination 
was closer in 1975 than in 1976 as a result of the 
slightly better germination in 1976. The relation be- 
tween seed yield and seed quality was poor. Number 
of seeds was uncorrelated with both percent filled 
seeds and percent germination in 1976 and only 
poorly correlated with germination percentage in 



Table 7. — Correlations between catkin characteris- 
tics, seed yield, and seed quality for 2 years 



Variables correlated 

Catkin length-catkin diameter 
Catkin length-catkin volume 
Catkin diameter-catkin volume 
Catkin diameter/length-catkin volume 
Catkin length-No. seeds/catkin 
Catkin diameter-No. seeds/catkin 
Catkin volume-No. seeds/catkin 
Catkin diameter/length-No. seeds/catkin 
Catkin length-percent filled seeds 
Catkin length-percent germination 
No. seeds/catkin-percent filled seeds 
No. seeds/catkin-percent germination 
Percent filled seeds-percent germination 

1 ** = significant at 0.01 level; * = significant at 0.05 level. 

2 First 8 correlations based on 237 individual catkins; last 2 on 74-tree 

averages 

3 First 8 correlations based on 307 individual catkins; last 5 on 155 catkins. 



2 1975 


3 1976 


0.668** 


0.117* 


.820** 


.522** 


.944** 


.868** 


-.253 


.131 


.621** 


.533** 


.529** 


.254** 


.574** 


.422** 


-.316** 


-.273** 


— 


.280** 


.329** 


.294** 


— 


-.042 


.271* 


.041 


— 


.550** 



1975 (table 7). The correlation between percent filled 
seed and percent germination (0.55) was only fair, 
indicating that a portion of the filled seeds was not 
viable. Although most of the correlations are statisti- 
cally significant, few of them show biologically 
significant relations. 



DISCUSSION 

These yellow birch trees were beginning to produce 
substantial amounts of seed by the time they were 8 
and 9 years old. Total yield of an average tree was 
more than 1,500 seeds in both 1975 and 1976. Trees 
with high numbers of catkins produced correspond- 
ingly greater numbers of seeds. Seed yield will, of 
course, increase with time as the number of catkins 
per tree goes up and with further increases in the 
average number of seeds per catkin. In 1975, 18 per- 
cent of the trees had more than 200 seeds per catkin 
compared with 22 percent in 1976. Of the latter, 3 
percent had more than 300 seeds per catkin. 

Variation between and within families in seed 
yield was substantial but not statistically signifi- 
cant. The lack of significance appears to be due to the 
plant material and the nature of the data. Unequal 
sample sizes were used in the analyses because the 
number of fruiting trees per family and the number of 
fruiting families per stand varied so much in both 
years. Therefore, the degrees of freedom were re- 
duced accordingly and even large differences were 
not significant. 

The correlation analyses indicate that catkin 
length, and, to a lesser degree, catkin volume have a 
fairly good positive relation with number of seeds per 
catkin. Catkin diameter, however, appears to vary 
more from year to year than other catkin variables 
and thus is a less reliable estimator of potential seed 
yield. 

Seed quality so far has not been as good as seed 
yield. Amount of filled seed was generally low, and 
seed germination in both years was less than 10 
percent compared with percentages above 90 percent 
for mature yellow birch in good seed years (Clausen 
1973). As a result, effective seed yield is still disap- 
pointingly low. 

The poor seed quality noted in these young trees 
may be due to a lack of pollen. No male catkins were 
produced in the plantation in the spring of 1975 and 
of the 16 trees with male catkins in the spring of 1976 
more than one-half had less than 10 catkins per tree 
(Clausen 1977). Thus, in both years most of the pollen 



had to come from the surrounding area where yellow 
birch is not common. Filled seed and germination 
percentages should improve as more trees begin to 
bear male catkins. 

In addition, the flowering phenology of some of the 
trees may be a problem. Trees of southern origin 
flowered later and may, therefore, be out of phase 
with the other trees in the plantation. The lack of 
seed germination in the trees from Georgia and the 
poorly filled and poorly germinating seed of trees 
from North Carolina stand 2959 and Virginia stand 
3299 suggest that these trees flower too late to be 
pollinated by trees of northern origin. On the other 
hand, the Wisconsin and Minnesota trees flower in 
synchrony with local pollen shed, and therefore, pro- 
duced more and better seed. 



LITERATURE CITED 

Clausen, K. E. 1973. Genetics of yellow birch. U.S. 
Department of Agriculture Forest Service, Re- 
search Paper WO-18, 28 p. U.S. Department of 
Agriculture Forest Service., Washington, D.C. 

Clausen, K. E. 1977. Early flowering and seed pro- 
duction in a yellow birch progeny test. Central 
States Forest Tree Improvement Conference Pro- 
ceedings 10:9-19. 

Clausen, K. E. 1979. The relation between tree size 
and flowering in yellow birch saplings. In Flower- 
ing and seed development in trees. IUFRO Sympo- 
sium Proceedings, p. 77-82. May 15-18, 1978. 
Starkville, Mississippi. 

Kozlowski, T. T. 1971. Growth and development of 
trees. Vol. I. Seed germination, ontogeny and shoot 
growth. 443 p. Academic Press, New York. 



ACKNOWLEDGMENT 

The assistance of David J. Polak in performing the 
statistical analyses is greatly appreciated. 

GU.S. GOVERNMENT PRINTING OFFICE: 1980--668906/5 




Put trash in the proper place. 



5DA FOREST SERVICE 
ESEARCH PAPER NC-184 






■V |T£8$ 



m\t 7 K 



CLEMSON 
LIBRARY 



Dimension yields from 
short logs of low-quality 

hardwood trees 



Howard N. Rosen 

Harold A. Stewart 

David J. Polak 




rth Central Forest Experiment Station 

rest Service, U.S. Department of Agriculture 






ACKNOWLEDGMENT 

The authors wish to thank David R. Schumann, 
Northeastern Area State and Private Forestry, 
USDA Forest Service, Portsmouth, New Hampshire, 
for his assistance in developing the yield charts. 



North Central Forest Experiment Station 

Forest Service - U.S. Department of Agriculture 

1992 Folwell Avenue 

St. Paul, Minnesota 55108 

Manuscript approved for publication October 26, 1979 

1980 



CONTENTS 

Page 

Method of Data Collection 1 

Method of Cutting Flitches 2 

Dimension Cuttings 2 

Data Summary 2 

Using the Charts 3 

Length 3 

Width adjustments 3 

Conversion of percent yield to number 

of cuttings per thousand board feet 20 

Comparison of Short Logs and Standard 

Grade Lumber — Yields and Costs 20 

Literature Cited 21 



DIMENSION YIELDS FROM SHORT LOGS 
OF LOW-QUALITY HARDWOOD TREES 



Howard N. Rosen, Research Chemical Engineer, 

Harold A. Stewart, Forest Products Technologist, 

and David J. Polak, Computer Programmer, 

Carbondale, Illinois 



As it becomes increasingly more difficult to obtain 
large sawtimber size logs for high-valued hardwoods, 
sawmills must look for alternative raw material 
sources. In the last decade researchers have sug- 
gested utilizing smaller diameter trees and shorter 
logs that are usually used only for pulp chips or fuel 
wood to make dimension cuttings for furniture parts 
(Bingham and Schroeder 1976a, 1976b, 1976c, 1977, 
Dunmire et al. 1972, Reynolds and Schroeder 1977). 
Before a manufacturer can choose among chips, fuel- 
wood, or solid wood as a final product for small diame- 
ter short logs, the expected yield of the product must 
be determined. The amount of chips or fuelwood can 
be evaluated by volume or weight, but more sophisti- 
cated methods are required to determine yield of solid 
wood dimension cuttings. 

A large quantity of material for short logs or bolts 
exists in the forest. About 87 percent of the commer- 
cial hardwood timber in the Eastern United States is 
in trees less than 21 inches in diameter (USDA For- 
est Service 1978). Additionally, thinning or pulp- 
wood cuts are often intermediate parts of the forest 
management plan. Smaller diameter logs and log- 
ging residue represent a considerable quantity of 
material unsuitable for standard lumber. This mate- 
rial can be cut into short lengths for dimension cut- 
tings. Although character marks are prevalent in the 
smaller logging residue, researchers have demon- 
strated that sound, character-marked cuttings can 
be manufactured from short logs or bolts (Cooper and 
Schlesinger 1974, Dunmire et al. 1972). In addition, 
the demand for character-marked hardwoods in 
furniture is increasing (Anonymous 1976). Thus, 
short logs contain a variety of material suitable for 
furniture. 



A method of estimating dimension stock yields 
from standard graded 4/4 lumber has been described 
(Englerth 1969, Schumann and Englerth 1967a, 
1967b, Schumann 1971). Nomographs were used to 
determine hard maple and black walnut yields for 
particular cutting bills. Comparisons were made be- 
tween grades so that the most economical grade or 
mix of grades could be found. And Landt (1974a, 
1974b, 1974c, 1974d) developed volume tables for 
predicting clear one-side (CIS) cuttings from small 
diameter short bolts of several species. A similar 
study on yellow-poplar by Cooper and Schlesinger 
(1974) predicted clear two-sides (C2S) and character- 
marked (CM) as well as CIS cuttings. However, the 
data were not in a form that could be easily converted 
to number of cuttings of a particular size per unit 
board feet. 

The purpose of this paper is to put in a nomograph 
form the dimension cutting yields of aspen, soft ma- 
ple, black cherry, yellow-poplar, and black walnut 
from logs 1.9 to 6.7 feet long and from 5 to 18 inches in 
diameter. This information may provide the furni- 
ture manufacturer or dimension plant manager a 
useful tool in selecting the most economical grade 
mix for a particular cutting bill. 



METHOD OF DATA COLLECTION 

Bolt material was removed from selected trees in 
stand improvement cuts or from residue remaining 
after logging on several sites from Wisconsin to 
Pennsylvania (table 1). The bolts were bucked from 
short tree sections and ranged in size from 1.9 to 6.7 
feet in length and from 5 to 18 inches in diameter. 



Table 1 


. — Basic 


information 


on bolt material 




Location 


Bolts 


Total 
volume 


Bolt 




Diameter Length 


Species 


Mean Range Mean Range 



No. Board feet Inches Feet 

Aspen Wisconsin 1 199 1,796 7.7 5.5-14.5 4.0 2.0-6.3 

Soft Southern 

maple Illinois 302 3,721 8.0 5.5-14.0 4.6 2.0-6.4 

Black Pennsylvania 

cherry 2 West Virginia 775 6,998 7.5 5.0-18.0 3.9 1.9-6.4 

Yellow- Southern 

poplar Illinois 351 4,241 7.7 5.5-16.5 4.7 2.0-6.7 

Black Southern 

walnut Illinois 200 1,621 6.8 5.5-12.0 4.5 2.1-6.4 

1 Each location represents one stand of trees. The results of this study only 
apply to the site from which the logs were cut. 

2 Black cherry material came from two stands each in a different location. 
This is not meant to be taken as a sample from the population of black cherry 
sites but just a combining of two sets of data. 



Bolt length was limited only by external defects or 
sweep exceeding 1 inch in 2 feet of length. The bolts 
were scaled by the International '/.-inch Rule — 
diameters were measured to the nearest /< inch and 
lengths to the nearest 1 inch. Bolt grades were re- 
corded using a grading system similar to the method 
described by Redman and Willard (1957). The bolt 
distributions, representative of residue material, are 
skewed toward the large end for diameter, length, 
and grade. 



defects on both faces. To determine maximum yields 
the boards were diagramed three separate times — 
first for the longest cuttings of grade C2S, then for the 
longest cuttings of grade CIS or better, and finally for 
the longest cuttings of grade CM or better. Because of 
the higher value of a long cutting, the longest and 
widest cuttings were diagramed first. Priority was 
established by the formula L 2 x W, where L was 
length and W was width of the cutting. Cuttings 
ranged from 1 to 6 inches wide and from 12 to 72 
inches long. 



DATA SUMMARY 

The cuttings for each species and cutting grade for 
bolts 2 feet and longer (2-foot minimum) were sum- 
marized on the computer to yield the number of cut- 
tings and total surface area by cutting length and 
width classes. A similar summary was made disre- 
garding bolts shorter than 4 feet (4-foot minimum). 
The classes ran from 1 to 6 inches, in increments of/2 
inch, for width, and from 12 to 72 inches, in incre- 
ments of 6 inches, for length. The surface area recov- 
ered in cuttings is reported as the total surface area of 
all cuttings represented as a percentage of the total 
surface area possible, as predicted by the Interna- 
tional '/-inch Rule (table 2). The small differences in 
surface area recovered among the cutting grades for 
each species is in part due to the priority established 



METHOD OF CUTTING 
FLITCHES 

The bolts were "live-sawn" into 1 /s-inch-thick 
rough unedged flitches on a portable bolter saw. Live- 
sawing consisted of slabbing one side of the bolt — the 
poorest side or the concave side of bolts with sweep — 
turning the bolt flat side down, and sawing the rest of 
the bolt without additional turning. The flitches were 
then air- and kiln-dried to a moisture content of 8 
percent. 



DIMENSION CUTTINGS 

The dried flitches, which had been skip-dressed to 
'/ie-inch thick, were diagramed to determine the 
dimension cuttings. The dimension cuttings were of 
three grades: C2S, which had two faces clear of knots 
and defects; CIS, which had one clear face and sound 
defects on the second face; and CM, which had sound 



Table 2. — Surface area recovered in cuttings by spe- 
cies, cutting grade, and minimum bolt length 





Cutting 


Two-foot 




Four-foot 


Species 


grade 


(minimum length) (minimum length) 








-Percent- 




Aspen 


C2S 


68 




69 




C1S 


69 




69 




CM 


69 




70 


Soft maple 


C2S 


64 




65 




C1S 


65 




66 




CM 


67 




67 


Black cherry 


C2S 


56 




58 




C1S 


57 




58 




CM 


57 




59 


Yellow-poplar 


C2S 


58 




58 




C1S 


58 




59 




CM 


59 




59 


Black walnut 


C2S 


62 




62 




C1S 


62 




63 




CM 


63 




63 



for longer cuttings (L 2 x W). The expected increase 
in surface area yield by allowing character marks on 
one of two faces was minimal because of the method of 
prioritizing cuttings (i.e., an increase in length was 
often accompanied by a decrease in width and an 
increase in waste). Also, the estimated yields are 
conservative because of the L 2 x W prioritizing. 

The summarized number of cuttings for each 
length and width class was then entered into a com- 
puter program designed to determine the dimension 
cutting yield charts for 1-inch-wide material and 
yield adjustment values for material greater than 1 
inch wide (figs. 1-14). Several yield charts were found 
to be nearly identical. In these cases one yield chart 
and width adjustment chart is reported and can be 
used to determine yield for several cutting grades 
and/or minimum bolt lengths without substantial 
loss in accuracy. 



new length is found by subtracting the unadjusted 
percent yield of the previous cutting length from the 
percent yield of the new cutting length. 

For example, if the cutting bill called for black 
walnut, C2S, 4-foot minimum bolt length, 1-inch- 
wide cuttings of lengths 48, 24, and 12 inches, the 
yield of 48-inch cuttings would be 14 percent (fig. 15). 
For 24-inch cuttings, the yield would be 22 percent 
(36 percent-14 percent), and the yield of 12-inch cut- 
tings would be 15 percent (51 percent-36 percent). 
The total yield of all cuttings would be 51 percent 
(which is also the accumulated percentage of the 
shortest length) or 510 square feet of surface area per 
1,000 board feet of bolts, scaled according to the In- 
ternational '/4-inch Rule. 



USING THE CHARTS 
Length 

The predicted percent yield of the longest desired 
cutting and subsequent shorter cuttings can be 
obtained from the charts given the length of the 
longest cutting required. To use the chart, first lo- 
cate the maximum cutting length required on the 
right-hand side of the chart. The percent yield, in 
surface area of a 1-inch-wide cutting, is found by 
moving horizontally to the left until the point of 
intersection with the percent yield scale on the far 
left. This is the percent yield of the longest 1-inch- 
wide cutting. To find the percent yield of the second 
longest cutting, begin at the point on the right corre- 
sponding to the maximum cutting length and pro- 
ceed vertically to the intersection with the line 
corresponding to the length of the second cutting. 
Now move horizontally to the percent yield scale. 
The percent yield of the second longest cutting, 
given the removal of the longest cutting, is obtained 
by subtracting the percent yield for the longest cut- 
ting from the percent yield for the second longest 
cutting. Other yields are obtained in a similar fash- 
ion always beginning at the point on the right 
corresponding to the maximum cutting length, pro- 
ceeding vertically to the line corresponding to the 
next desired length, and then moving horizontally to 
the percent yield scale. The percent yield for each 



Width Adjustments 

To determine the percent yield of cuttings other 
than 1-inch-wide, a correction multiplier chart is 
given. The percent yield for cuttings greater than 1 
inch are found by locating the length of the desired 
cutting at the base of the correction multiplier chart. 
Then move vertically to the intersection with the line 
for the needed width and then horizontally to the 
correction multiplier scale. Multiply the correction 
multiplier by the percent yield for a 1-inch-wide cut- 
ting of the required length to obtain the percent yield 
for the needed width. 

For example, given a cutting bill as before, but 
with a width requirement of 2.5 inches for 48-inch 
cuttings, 3 inches for 24-inch cuttings, and 2 inches 
for 12-inch cuttings, take the percent yield before 
subtraction for previous lengths and multiply it by 
the width correction multiplier of the length needed 
(fig. 16, table 3 ). Now the yield of 48-inch cuttings is 9 
percent — the original yield of 14 percent times the 
correction multiplier for a 2.5-inch-wide 48-inch-long 
cutting of 0.62. For the 3-inch-wide 24-inch cutting, 
the yield is now 7 percent — the original 36 percent 
times the correction of 0.43 and finally minus 9 per- 
cent for the 48-inch cuttings already removed. Simi- 
larly, the yield for the 2-inch-wide 12-inch cuttings is 
26 percent — 51 percent times 0.82 minus 16 for the 
48- and 24-inch cuttings already removed. The total 
yield of all desired cuttings is 42 percent or 42 square 
feet per 1,000 board feet, as scaled by the Interna- 
tional %-inch Rule. 




1.00 




LENGTH OF CUTTING (INCHES) 



Figure 1. — Yield prediction of 1 -inch-wide cuttings with width corrections for 
black walnut C2S — 2 -foot minimum. 



55 



50 



45 



40 



S 35 

O 

a: 

UJ 

a 

*- 30 
Q 

51 

Q 25 
Ul 



Q 

Ul 20 
QC 

a 



15 



10 



to 

UJ 

- i 
o 

z 

o 

- z 

1- 
»- 

D 

o 

- H 

Z 
UJ 

o 

- UJ 

(A 

CD 
D 
(/) 
u. 
" O 
X 
H 
O 

. z 

UJ 

-J 


12 

18 
24 

30 

36 

42 
48 

54 
60 

66 




















/ - 














"7 18 

24 


- 












"T^O 

A/ 




- 












~^36 
^^42 ^ 


- 






48 ^ 


> 


~~s r 66 

72 


^ — —*~54 
60 

i . j 


' 




• 



12 





1.00 




90 


cc 

UJ 


.80 






-J 




a. 


70 






k 




-j 




=) 


bO 


S 




2 


.50 


O 








l- 


40 


o 




UJ 




oc 


.30 


tt 




O 
O 


.20 




.10 








72 66 60 54 48 42 36 30 

LENGTH OF CUTTING (INCHES) 



24 



18 



1.5 
2.0 




•6.0 



12 



Figure 2.— Yield prediction of 1 -inch-wide cuttings with width corrections for 
black walnut CIS— 2 -foot minimum. 




1.00 
.90 
























- 

^~- 1.5 
2.0 " 

2.5 - 

' 3.0 " 
3.5 

4.0 

4.5 " 

. 5.0 


2 80 

•J 




















9: .70 
^ .60 














2 .50 












O 

P .40 












O 
Uj 
QC .30 














8 .20 
.10 

o 




























— ~~— - — 






5.5 " 

6.0 






72 


66 


i 

60 


i 

54 


48 


i 
42 


i 
36 


30 


24 


18 


12 



LENGTH OF CUTTING (INCHES) 



Figure 3. — Yield prediction of 1 -inch-wide cuttings with width corrections for 
black walnut CM — 2 -foot minimum. 




1.00 
.90 


- 






















1 5 


£ 80 

0. .70 
K 






















2.0 . 


^ 60 

s 

2 .50 




















2.5 

3.0 . 


O 

C .40 

o 






















Ul 

f£ .30 

QC 

§ 20 






















3.5 ' 






















4.0 . 

4.5 


10 






















5.0 - 
























5.5 
^r 60 







! 

72 


1 

66 


" T 

60 


1 

54 


i 
48 


42 


i 

36 


i 

30 


1 

24 


1 

18 


12 



LENGTH OF CUTTING (INCHES) 

Figure 4.— Yield prediction of 1 -inch-wide cuttings with width corrections for 
black walnut C2S^i-foot minimum. 



55 



50 



45 



12 



18 







V) 






40 


UJ 

X 

o 

z 


24 


— ^ 




^^ 






35 


. o 




Uj 
O 




z 




cc 




K 




Uj 




D 


^0 


a 


30 


. (J 




Q 

-J 




z 




Ul 




UJ 




Q 


25 


. O 

UJ 


36 


UJ 




</> 




t- 




CD 




o 




D 




n 




(0 


42 


UJ 


20 


- u. 




CC 




O 




a 




z 

O 


48 




1b 


" z 

UJ 

_l 


54 




10 




60 



66 





T 

54 48 42 36 30 

LENGTH OF CUTTING (INCHES) 



18 



12 



Figure 5. — Yield prediction of 1 -inch-wide cuttings with width corrections for 
black walnut CIS — 4 -foot minimum. 



55 



50 



45 



10 - 



12 



18 





40 


(A 

UJ 

" I 

O 

7 


24 










K 




•*— * 




2 

Uj 

o 


35 


. o 

Z 


30 


0- 




H 




Ul 




t- 




0. 




3 




Q 


30 


_ O 




-J 
Uj 




Z 
UJ 


36 


>» 




D 




Q 
Uj 


25 


. O 

UJ 




K 




v> 


4? 


O 




CD 








D 




U 




(/) 




Uj 
Q. 


20- 


U. 
O 

X 

»- 
o 


48 




15 


■ z 








UJ 


54 



60 



66 




1.00 




LENGTH OF CUTTING (INCHES) 

Figure 6.— Yield prediction of 1 -inch-wide cuttings with width corrections for 
black walnut CM — 4 -foot minimum. 



55 



50 



45 



10 



12 







</> 






40 


. UJ 

X 

o 




_^ 




z 




K 




s> 




UJ 
O 


35 


- o 

z 


18 


cc 

UJ 




1- 


24 


a 




3 




■— 


?n 


- o 




Ul 




1- 
z 








UJ 




V 




u 




Q 

UJ 


25 


L o 

UJ 


30 


K 




(A 




O 




CD 




Q 

UJ 
CC 

a 


20 


3 

0) 

■ u. 






O 


ib 






X 








1- 








o 






1b 


" z 


Al 



42 



48 



54 
60 
66 




1.00 




LENGTH OF CUTTING (INCHES) 

Figure 7. — Yield prediction of 1 -inch-wide cuttings with width corrections for 
black cherry C2S—2- and 4 -foot minimum. 



10 





72 



66 



60 



54 48 42 36 30 

LENGTH OF CUTTING (INCHES) 



24 



18 



Figure 8.— Yield prediction of 1 -inch-wide cuttings with width corrections for 
black cherry CIS— 2- and 4 -foot minimum. 



11 





72 



66 



T 
60 



-T 

54 



48 42 36 30 

LENGTH OF CUTTING (INCHES) 

c * „nh ,i,idP cuttings with width corrections for 



12 




1.00 




72 



66 



T" 

60 



4 48 42 36 

LENGTH OF CUTTING (INCHES) 



Figure 10.— Yield prediction of 1 -inch-wide cuttings with width corrections for 
yellow -poplar C2S—2- and 4 -foot minimum. 



13 





72 



66 



54 48 42 36 30 

LENGTH OF CUTTING (INCHES) 



24 



18 



Figure 11. — Yield prediction of 1 -inch-wide cuttings with width corrections for 
yellow -poplar CIS — 2- and 4 -foot minimum. 



14 




1.00 






3 
S 

O 

O 

lu 

ct 
oc 
O 
O 




72 



66 



54 48 42 36 30 

LENGTH OF CUTTING (INCHES) 



24 



12 



Figure 12 — Yield prediction of 1 -inch-wide cuttings with width corrections for 
yellow-poplar CM— 2- and 4 -foot minimum. 



15 



55 



50 



45 - 



15 



10 



12 



18 







ST 24 






UJ 




40 


. I 
O 

z 


h. 




o 


? 


35 


- z 


Uj 

o 




C 30 


UJ 




o 


■^ 


30 


\- H- 36 


Q 




Z 


-j 




UJ 


UJ 




D 


>- 




O 42 


Q 


25 


. UJ 

£ 48 

CD 






(A 


Q 




u. 


UJ 


20 


- O 54 


IT 






a 




H 60 



o 



UJ 



66 




1.00 



.90 ■ 



a- 






UJ 


80 










~j 






a 


.70 




K 






~J 






3 


.60 




S 






2 


.50 




O 












l- 

o 


.40 




UJ 






cc 


.30 




ec 






O 






o 


.20 


" 



.10 - 



— r~ 

72 



1.5 
2.0 




54 48 42 36 30 

LENGTH OF CUTTING (INCHES) 



Figure 13. — Yield prediction of 1 -inch-wide cuttings with width corrections for 
soft maple CIS, C2S, and CM— 2- and 4 -foot minimum. 



16 




72 



66 



60 



54 48 42 36 30 

LENGTH OF CUTTING (INCHES) 

Figure 14. — Yield prediction of 1 -inch-wide cuttings with width corrections for 
aspen CIS, C2S, and CM— 2- and 4 -foot minimum. 



17 



60 r- 



,51- 



50 



15%< 



40 



36 



22% < 



,14- 



10 - 



14% < 



.12- 





I 




- O 




* z 




(3 




Z 






h- 


1- 


s 


t- 


Uj 


3 


£ 30 


_ O 


a: 


1- 


Ul 


z 


Q. 


UJ 


Q 

-j 


3 
O 


Uj 


hi 




CO 


v 


CD 


Q 


3 


Uj 


(A 


K 


U. 


O 


O 






Q nn 


Z 


Uj 20 


— 1- 


Cc 


o 


Q. 


z 



1. 




72 66 60 54 48 42 36 30 

LENGTH OF LONGEST CUTTING (INCHES) 



24 



18 



12 



Figure 15. — Example for using the percent yield charts: black walnut C2S — 4- 

foot minimum. 



18 



s»u 








■■ -=- 


1-5 


- 


80 




2.0 


70 


- 




- 


60 






2 5 


- 


50 








30 


- 








40 








30 








35 


- 


20 








4 


- 


10 


_ 






45 
5.0 


- 


















5.5 
6.0 


- 





I I I I 


i i i 


i 



72 



54 48 42 36 30 

CUTTING LENGTH (INCHES) 



Figure 16. — Example for using the width adjustment chart: black walnut C2S- 

4 -foot minimum. 



Table 3. — Example for obtaining percent yields for 
cuttings of various widths from black walnut, C2S 
lumber, and 4 -foot minimum logs 





Yield 


Width 






Yield 


Cutting size 


ot 1 -inch-wide 


correction 


Adjusted 


given 


removal of 


Length Width 


boards 


multiplier 


yield 


previous cuttings 


Inches 

48 2.5 












14 


0.62 


9 




9 


24 3.0 


36 


0.43 


16 




7 


12 2.0 


51 


0.82 


42 
Total yield 




26 
42 



19 



Conversion of Percent Yield 

to Number of Cuttings 

per Thousand Board Feet 

Yields are easily converted from percents to num- 
ber of cuttings per 1,000 board feet by dividing the 
percent yield by 100, multiplying by 1,000, and divid- 
ing by the area of the cutting surface in square feet 
(table 4). 



Table 4. — Number of cuttings per 1,000 board feet of 
three cutting sizes from black walnut, C2S lumber, 
and 4 -foot minimum logs 









Cuttings 


Cutting size 


Surface 


Yield 


per 1,000 


Length Width 


area 




board feet 


Inches 


Ft 2 


Percent 


No. 


48 2.5 


0.83 


9 


108 


24 3.0 


0.50 


7 


140 


12 2.0 


0.17 


26 


1,529 



COMPARISON OF SHORT LOGS 

AND STANDARD GRADE 

LUMBER^ 

YIELDS AND COSTS 

Two general observations can be made comparing 
cutting yields from short logs to standard grade lum- 
ber: (1) overall yields are less for short logs and (2) 
distributions of cuttings are more heavily weighted 
to short and narrow pieces for short logs. For exam- 
ple, yields of short logs (C2S — 4-foot minimum) com- 
pared to No. 2 Common lumber for black walnut show 
that recoveries for 1 -inch-wide cuttings are similar, 
but recoveries from 3-inch lumber are better for No. 2 
Common (table 5). For black walnut, yields from 
short logs are less than 15 percent for all lengths of 
cuttings 4 inches wide and above, whereas FAS grade 
lumber can yield up to 60 percent for the same prod- 
uct (Schumann 1971). The same is true for the other 
wood species of this study. Integrated sawmill- 
dimension plants can best use short logs for dimen- 
sion lumber when the cutting bill requires narrow 
cuttings and a large percentage of the cuttings are 
short. 



Table 5. — Comparison of black walnut yields for No. 2 
Common and short logs(C2S — 4 -foot minimum) for 
different width cuttings 

(In percent) 



Cutting length 


1-inch width 3-inch width 


('Inches) 


No. 2 Common 1 Short logs No. 2 Common 1 Short logs 


48 
24 
12 


14 14 7 7 
28 22 24 9 
11 5 13 5 


Total 


53 51 44 21 



'From charts of Schumann (1971). 



Because of the large differences in prices of lumber 
that exist depending on area of the country and time 
of year, estimating a price on standard grade lumber, 
much less on short log material, is difficult. Because 
black walnut is the most valuable wood type in this 
study, we estimated the cost of this wood for different 
grades to fill a specific cutting bill. We assumed the 
following quantity and sizes: 400 48- by 2/2-inch, 
400 24- by 3-inch, and 2,000 12- by 2-inch. We also 
assumed a cost per 1,000 board feet for FAS of $1,300, 
No. 1 Common $800, No. 2 Common $400, and the 
lumber derived from short logs $250. 

After calculating lumber costs for several short log 
and grade combinations, we found that using entirely 
short log material was the most economical (table 6). 
Although three times as much short log lumber was 
required as for FAS lumber, the cost of cuttings from 
FAS lumber was almost twice as much. Combina- 
tions of No. 1 Common for the longer cuttings and 
short logs for the remainder reduced the lumber re- 
quired, compared to only short logs, by more than 
half; but the lumber cost was still about $150 per 
1,000 board feet higher. 

The cost comparison example given is dependent 
upon particular prices. Also processing, handling, 
and drying costs may be larger due to the greater 
quantities of lumber needed to fill the cutting bill. 
Nevertheless the analysis demonstrates that short 
log material has potential economic value and that 
hardwood dimension producers and furniture manu- 
facturers that produce their own raw material should 
consider this material when choosing the best grade 
mix to meet a specific cutting bill. 



20 



Table 6. — Comparison of costs to cut 400 48- by 2 1 h- 
inch, 400 24- by 3-inch, and 2, 000 12- by 2 -inch 4/4 
black walnut cuttings from various grades and 
short logs 1 



Lumber mix 



Lumber Cost per 

required 1,000 board feel Total 



Short logs for all 


Board feet 


Dollars - 




lengths 


3,774 


250 


944 


FAS for all lengths 


1,247 


1.300 


1,621 


No 1 Common for all 








lengths 


1,544 


800 


1,235 


No. 2 Common for all 








lengths 


3,008 


400 


1,203 


FAS for 48-inch lengths 2 


678 


1,300 




Short log for remainder 


913 


250 


1.110 


No. 1 Common for 48- and 








24-inch lengths 3 


1.278 


800 




Short log for remainder 


282 


250 


1,093 



1 Yields of standard grade from Schumann (1971). 
includes 230 of 24-inch and 199 of 12-inch, 
includes 1,348 of 12-inch. 



LITERATURE CITED 

Anonymous. 1976. Character marked hardwoods 
seminar. National Hardwood Magazine 50(8):40- 
41. 

Bingham, Samuel A., and James G. Schroeder. 
1976a. Short lumber in furniture manufacture, 
part I. National Hardwood Magazine 50( 11 ):34-35, 
48-50. 

Bingham, Samuel A., and James G. Schroeder. 
1976b. Short lumber in furniture manufacture, 
part II. National Hardwood Magazine 50(13):90- 
91, 112-113. 

Bingham, Samuel A., and James G. Schroeder. 
1976c. Short lumber in furniture manufacture, 
part HI. National Hardwood Magazine 50(13):38- 
39, 49. 

Bingham, Samuel A., and James G. Schroeder. 1977. 
Short lumber in furniture manufacture, part IV. 
National Hardwood Magazine 51(l):28-29, 32. 

Cooper, Glenn A., and Richard C. Schlesinger. 1974. 
Yield of furniture dimension from yellow-poplar 
thinnings. Southern Lumberman 229(2848): 83-85. 

Dunmire, D. E., E. F. Landt, and R. E. Bodkin. 1972. 
Logging residue is a source of valuable dimension 
stock. Forest Products Journal 22(1):14-17. 



Englerth, George H. 1969. Charts for calculating di- 
mension yields from hard maple lumber. U.S. De- 
partment of Agriculture Forest Service, Research 
Paper FPL-118, 12 p. U.S. Department of Agricul- 
ture Forest Service, Forest Products Laboratory, 
Madison, Wisconsin. 

Landt, Eugene F. 1974a. Aspen volume tables for 
furniture-type, flat, 4/4-inch dimension. U.S. De- 
partment of Agriculture Forest Service, Research 
Note NC-166, 2 p. U.S. Department of Agriculture 
Forest Service, North Central Forest Experiment 
Station, St. Paul, Minnesota. 

Landt, Eugene F. 1974b. Black cherry volume tables 
for furniture- type, flat, 4/4-inch dimension from 
small low-quality trees. U.S. Department of Agri- 
culture Forest Service, Research Note NC-167, 2 p. 
U.S. Department of Agriculture Forest Service, 
North Central Forest Experiment Station, St. 
Paul, Minnesota. 

Landt, Eugene F. 1974c. Black walnut volume tables 
for furniture-type, flat, 4/4-inch dimension from 
small low-quality trees. U.S. Department of Agri- 
culture Forest Service, Research Note NC-168, 2 p. 
U.S. Department of Agriculture Forest Service, 
North Central Forest Experiment Station, St. 
Paul, Minnesota. 

Landt, Eugene F. 1974d. Soft maple volume tables 
for furniture-type, flat, 4/4-inch dimension from 
small low-quality trees. U.S. Department of Agri- 
culture Forest Service, Research Note NC-169, 2 p. 
U.S. Department of Agriculture Forest Service, 
North Central Forest Experiment Station, St. 
Paul, Minnesota. 

Redman, G. P., and R. Willard. 1957. Short log 
bolter for furniture stock. 87 p. Southern Furniture 
Manufacturers Association, Highpoint, North 
Carolina. 

Reynolds, James W., and James G. Schroeder. 1977. 
System 6 — a way to use small logs to make grade 
lumber. Southern Lumberman 234(28971:9-10. 

Schumann, David R. 1971. Dimension yields from 
black walnut lumber. U.S. Department of Agricul- 
ture Forest Service, Research Paper FPL-162, 16 p. 
U.S. Department of Agriculture Forest Service, 
Forest Products Laboratory, Madison, Wisconsin. 

Schumann, David R., and George H. Englerth. 
1967a. Dimension stock-yields of specific width 
cuttings from 4/4 hard maple lumber. U.S. Depart- 
ment of Agriculture Forest Service, Research Pa- 
per FPL-85, 48 p. U.S. Department of Agriculture 
Forest Service, Forest Products Laboratory, Madi- 
son, Wisconsin. 



21 



Schumann, David R., and George H. Englerth. U.S. Department of Agriculture, Forest Service. 
1967b. Yields of random- width dimension from 4/4 1978. Forest statistics of the United States, 1977. 
hard maple lumber. U.S. Department of Agricul- (Review draft) 133 p. U.S. Department of Agricul- 
ture Forest Service, Research Paper FPL-81, 12 p. ture Forest Service, Government Printing Office, 
U.S. Department of Agriculture Forest Service, Washington, D.C. 
Forest Products Laboratory, Madison, Wisconsin. 

ftU.S. GOVERNMENT PRINTING OFFICE: 1980—669289/64 



22 



Rosen, Howard N., Harold A. Stewart, and David J. Polak. 

1980. Dimension yields from short logs of low-quality hardwood trees. 
U.S. Department of Agriculture Forest Service, Research Paper NC- 
184, 22 p. U.S. Department of Agriculture Forest Service, North 
Central Forest Experiment Station, St. Paul, Minnesota. 

Charts are presented for determining yields of 4/4 dimension cut- 
tings from short hardwood logs of aspen, soft maple, black cherry, 
yellow-poplar, and black walnut for several cutting grades and bolt 
sizes. Cost comparisons of short log and standard grade mixes show 
the estimated least expensive choice for a specific cutting bill. 

KEY WORDS: utilization, bolts, grades, dimension stock, cutting bill, 
costs, Liriodendron tulipifera, Juglans nigra, Prunus sertina, Acer 
rubrum, Populus grandidentata. 



Rosen, Howard N., Harold A. Stewart, and David J. Polak. 

1980. Dimension yields from short logs of low-quality hardwood trees. 
U.S. Department of Agriculture Forest Service, Research Paper NC- 
184, 22 p, U.S. Department of Agriculture Forest Service, North 
Central Forest Experiment Station, St. Paul, Minnesota. 

Charts are presented for determining yields of 4/4 dimension cut- 
tings from short hardwood logs of aspen, soft maple, black cherry, 
yellow-poplar, and black walnut for several cutting grades and bolt 
sizes. Cost comparisons of short log and standard grade mixes show 
the estimated least expensive choice for a specific cutting bill. 

KEY WORDS: utilization, bolts, grades, dimension stock, cutting bill, 
costs, Liriodendron tulipifera, Juglans nigra, Prunus sertina, Acer 
rubrum, Populus grandidentata. 




Gtvz a hoot. . .don' t pollute! 



/ ; v • ' 






SDA FOREST SERVICE 
ESEARCH PAPER NC-185 



: i rs 



OCT 27 



: 






The transmission of 

OAK WILT 






J.N. Gibbs and D.W. French 







North Central Forest Experiment Station 
Forest Service, U.S. Department of Agriculture 



North Central Forest Experiment Station 

Forest Service — U.S. Department of Agriculture 

1992 Folwell Avenue 

St. Paul, Minnesota 55108 

Manuscript approved for publication April 3, 1979 

1980 



CONTENTS 

Page 

History and severity of the disease 1 

Saprophytic existence of C. fagacearum in the 

dying tree 2 

Seasonal susceptibility of trees to infection. ... 4 

Underground spread 5 

Overland spread 6 

Information from research on control 12 

Information from the distribution of the disease ... 13 

Conclusions 13 

Literature cited 14 



THE TRANSMISSION OF OAK WILT 



J. N. Gibbs, Pathologist, Forestry Commission Research 

Station, Alice Holt Lodge, Farnham, Surrey, U.K. 

and D. W. French, Professor, Department of Plant Pathology, 

University of Minnesota, St. Paul, Minnesota 



Ceratocystis fagacearum (Bretzl Hunt, the cause of 
oak wilt, is a fungus with the potential to be one of the 
most destructive of all tree pathogens. Red oaks 
(subgenus Erythrobalanus) usually die within a few 
weeks of infection, and although white oaks 
(subgenus Lepidobalanus) are more resistant they 
are not immune. Trees that are diseased 1 year may 
recover the next but alternatively symptoms may 
recur. The host populations are enormous within the 
known range of the disease. In the eastern half of the 
United States the growing stock of oak amounts to 
60,000 million cu. ft., or about 35 percent of the total 
hardwood volume. In the 20 States in which oak wilt 
is now known to exist, the red oak growing stock 
amounts to 22,000 million cu. ft., and the white oak 
growing stock is almost as large (U.S. Department of 
Agriculture, Forest Service 1965). 

C. fagacearum has not caused the devastation once 
so greatly feared because its spread from diseased to 
healthy trees is slow and erratic. Three chief methods 
of transmission have been recognized: (1) through 
root grafts; ( 2 ) via sap-feeding insects such as Nitidu- 
lid beetles; and (3) via tree-wounding insects such as 
oak bark beetles. There is general agreement that 
the importance of these means of transmission varies 
in different parts of the oak wilt range, but in recent 
years little attempt has been made to evaluate all the 
data that have accumulated. The objective of this 
paper is to provide such an evaluation. 



HISTORY AND SEVERITY OF 
THE DISEASE 

Oak wilt was first described in Wisconsin in the 
early 1940's (Henry et al. 1944) but disease survey 
records suggest that it was present at least as early as 
1912 (French and Stienstra 1975). By 1947 it was 



known to be the major disease of oak in the Upper 
Mississippi Valley. In 1950 it was recorded in Penn- 
sylvania and by 1951 was reported from 18 States. 
However, in many places it was undoubtedly present 
for a number of years before it was recognized and 
identified. Thus, in Cumberland County, Pennsylva- 
nia, one disease center was probably in existence by 
1935 (Craighead and Nelson 1960), and in West Vir- 
ginia, True et al. (1951) established that it had been 
present for at least 5 to 10 years before it was discov- 
ered. Although the disease was subsequently rec- 
orded in Oklahoma (1958) and South Carolina 
(1969), its known distribution has changed little 
since 1951, despite the presence of apparently suit- 
able oak popu-lations in adjacent areas (fig. 1). In 
some places, for example in Minnesota, it is not now 
known in several counties where it was present ear- 
lier (French and Bergdahl 1973). 

Disease severity varies greatly and is worst within 
the northwest part of its range. During a 10-year 
period prior to 1953, the disease destroyed 4.4 percent 
of the oak woodland in four counties of central Wis- 
consin. But in four adjacent counties the loss 
amounted to only 0.2 percent (Anderson and Skilling 
1955). Even in the worst affected parts of States such 
as Missouri and West Virginia, less than one tree dies 
of oak wilt per square mile of forest each year (Lautz 
and Saufley 1970, Rexrode 1977). 

It should be noted that the dramatic and continuing 
development of the disease in parts of Minnesota and 
Wisconsin during the last 40 years can at least partly 
be explained by the presence of a relatively new, 
highly susceptible, host population. The dense stands 
of northern pin oak (Quereus ellipsoidalis E. J. Hill) 
through which infection is spreading have resulted 
from the vigorous coppicing habit of this species, 
which has enabled it to establish itself as the domi- 
nant tree after logging operations and fire destroyed 
the original, more diverse woodland communities. 




Figure 1. — The known distribution of oak wilt as of 1978. 



All oaks and related species in the family Fagaceae 
that have been subjected to artificial inoculation 
with C. fagacearum have shown some susceptibility 
(Bretz 1955). However, as indicated earlier, the white 
oaks are much less susceptible than the red oaks. In 
one study area in northern Iowa, Young ( 1949) found 
that 55 percent of Q. ellipsoidalis (northern pin oak) 
and 53 percent of Q. rubra L. (red oak) had died but 
only 28 percent of Q. alba L. (white oak) and 20 
percent of Q. macrocarpa Michx. (bur oak) had died. 
As might be expected, this difference in susceptibility 
has resulted in a change in the proportion of the two 
species in some areas. French and Bergdahl (1973) 
described an area in Sherburne County, Minnesota, 
in which 83 percent of the red oaks but only 11 per- 
cent of the bur oaks died from oak wilt between 1960 
and 1971. This increased the percentage of bur oaks 
in the stand from 48 to 83. 



SAPROPHYTIC EXISTENCE OF 

C. FAGACEARUM IN THE DYING 

TREE 

Before discussing the various means of disease 
transmission it is necessary to consider the status of 
the pathogen within the diseased tree — the infection 
source. Working with naturally infected red oaks, 
Henry and Riker (1947) in Wisconsin and Young 
(1949) in northern Iowa readily isolated C. faga- 
cearum from the xylem of roots, trunk, branches, 
twigs, and even leaf petioles. In the trunk Young 
( 1949) found that the fungus was present only in the 
outermost sapwood layer. This pattern of distribu- 
tion is not surprising. It is typical of vascular wilts of 
trees for the pathogen to be confined within the xy- 
lem vessels of the current annual ring until the host 



becomes moribund. Following the death of the tree 
the mycelium of C. fagacearum grows extensively in 
the xylem vessels, and the fungus penetrates the ray 
cells and begins to grow out toward the cambium 
(Struckmeyere^a/. 1958) and also inward. In one tree 
studied in Missouri about 8 months after its death, 
Jones and Bretz ( 1955) recovered the fungus from the 
tenth annual ring. Similarly Englerth et al. (1956) 
recorded inward colonization of the sapwood in logs 
from diseased trees. At the cambium the fungus may 
form a sporulating mat characterized by a layer of 
mycelium and conidiophores surrounding a raised 
pressure "pad" or "cushion". These mats are a famous 
feature of oak wilt, but it is not always realized that 
their formation requires the pathogen to be present 
simultaneously in the inner bark and adjacent xy- 
lem. Inner bark colonization was nicely demon- 
strated by Curl (1955) who found that when bark 
pieces from wilted trees were placed on the forest 
floor, 25 percent of them produced some kind of mat. 
He also found several mats that had developed en- 
tirely within the bark on standing trees. Such mats 
were also reported by Barnett et al. ( 1952 ) and Fergus 
and Stambaugh (1957). 

Because the formation of mats requires that C. 
fagacearum be present in the inner bark as well as 
the xylem, it follows that data on the distribution of 
mats on diseased trees give some information on bark 
colonization by the pathogen. Curl (1955) reported 
that in northern Illinois, mats were produced only on 
the trunk and large branches, and this is the common 
situation in other places also. However, Morris and 
Fergus (1952) in Pennsylvania reported that mats 
sometimes occurred on branches as small as 3 cm in 
diameter, and Engelhard (1955) in northern Iowa 
recorded them on 2.5 cm diameter branches. During 
the summer of 1977 in Minnesota, mats developed on 
2-to 4-cm diameter branches on young Q. ellipsoi- 
dalis that died in June, following artificial inocula- 
tion with C. fagacearum in May. No published 
accounts are available from more southerly States on 
the presence of mats on small branches, although C. 
0. Rexrode 1 has observed them in West Virginia. 

Data on mat formation can also pi'ovide informa- 
tion on the length of time during which C. faga- 
cearum can survive in the dead host. The formation of 
a sporulating mat represents the climax of the sapro- 
phytic growth of the pathogen in that part of the tree 
and its replacement by other microorganisms soon 
follows. Shigo ( 1958) in West Virginia was only able 
to isolate C. fagacearum from the xylem below 2 out 



of 58 sporulating mats, while other fungi such as 
Ceratocystis piceae (syn. Graphium rigidum), Glioc- 
ladium roseum, and Trichoderma lignorum (syn. T. 
viride) were very common. In the northern part of the 
oak wilt range where mats are formed frequently, 
trees that wilt in the first half of the summer (up to 
mid-July) produce mats in late August or September, 
while those that wilt later in the summer produce 
mats the following spring (Curl (1955) in Illinois; 
Morris and Fergus (1952) in Pennsylvania; and 
Campbell and French ( 1955a) in Minnesota). In Min- 
nesota mat formation begins on the upper part of the 
bole on the south side of the tree. The inner bark on 
the lower south side and on the north side is normally 
still fresh at this time although the fungus can read- 
ily be isolated from the xylem. 

In the southwest of the oak wilt range, the sapro- 
phytic survival of C. fagacearum is probably even 
more limited. Mats are rarely produced on diseased 
trees in Missouri, Arkansas, and Ohio. This is not due 
to genetic differences in the C. fagacearum isolates 
involved, because isolates from these States were just 
as capable of producing mats on inoculated trees as 
isolates from Wisconsin, Illinois, and Pennsylvania 
(Cobb and Fergus 1964). Some light is cast on the 
problem by experiments in which logs cut in the 
summer from diseased trees in Pennsylvania and 
Missouri and piled in Pennsylvania produced some 
mats, while similar logs piled in Missouri produced 
none (U.S. Department of Agriculture, Forest Serv- 
ice 1967). It was noted that the higher air tempera- 
tures and lower relative humidities in Missouri led to 
a lower wood moisture content, and also that coloni- 
zation by Hypoxylon punctulatum occurred earlier. 
Similarly in Arkansas, Tainter and Gubler (1973) 
have noticed that Hypoxylon atropunctatum rapidly 
establishes itself in the sapwood of artificially inocu- 
lated oaks, and have suggested that this fungus, to- 
gether with the drying out of the branches and the 
high summer temperatures, greatly reduces sapro- 
phytic survival of the pathogen. 

Some data on the effect of high temperatures on 
survival are provided by Houston et al. (1965). They 
recorded temperatures approaching 50C in the cam- 
bial region of infected northern pin oak logs exposed 
to the July sun for 3 days. The pathogen could not be 
isolated from the exposed side of the log but remained 
viable on the shaded side. Working with the same 
tree species in Minnesota, we 2 found that between 



Rexrode, C. O. Personal communication. 



2 Unpublished report on file at Stakman Hall, Uni- 
versity of Minnesota, Department of Plant Pathology, 
St. Paul, Minnesota. 



June 21 and July 1, 1977, the percentage of xylem 
chips yielding C. fagacearum dropped from 33 to 8 on 
the exposed side of 2-to 5-cm diameter branches, 
while it remained unchanged on the shaded side. 

Data on the survival of C. fagacearum in the xylem 
of standing trees come from the work of Turk (1955) 
who took samples at breast height from 52 northern 
pin oaks in Minnesota at various times after they had 
wilted. Within 6 months, recovery of the pathogen 
had dropped to 40 percent, by 9 months it was 20 
percent, and by 10 months it was 0. Competition from 
other fungi including the sapwood colonizer Nummu- 
laria bulliardi and antagonism from Trichothecium 
roseum were considered important in limiting the 
survival of C. fagacearum. Merek and Fergus (1954) 
in Pennsylvania found that they could isolate the 
fungus from the trunk of diseased trees that were 
felled and left lying in the forest for up to 44 weeks, 
while in twigs it could only be isolated for 3 weeks. 
Recently Gibbs (1979a) examined the survival of C. 
fagacearum in small branches (1 to 10 cm in diame- 
ter) of northern pin oak in Minnesota. C. fagacearum 
survived in the xylem for only 1 to 2 months in trees 
that died in May or June but survived longer in trees 
that wilted later in the summer. Dothiorella quercina 
and Coryneum kunzei were the two chief antagonists 
of C. fagacearum. 

Survival of the pathogen in sawn lumber has re- 
ceived some attention, principally because of the fear 
that the disease might thereby be introduced to other 
parts of North America or to other continents such as 
Europe (Gibbs 1978). Englerth et al. (1956) found 
that the frequency with which C. fagacearum was 
isolated from the wood dropped rapidly after sawing, 
but that the fungus could be recovered for up to 24 
weeks (6 weeks as roundwood, 18 weeks as bulk-piled 
boards). Steaming or kiln-drying killed the patho- 
gen. Little evidence exists linking the spread of infec- 
tion with the movement of wood from diseased trees, 
although French and Bergdahl (1973) found an iso- 
lated outbreak of the disease in central Minnesota 
that might have begun from transported firewood. 

From the above it would seem that within a few 
months in trees that die in early summer and well 
within a year in trees that die later, the oak wilt 
fungus has disappeared from the above ground parts 
of the tree. It is likely to survive longest on the lowest 
part of the north side of the trunk. As indicated above, 
antogonism by other fungi is one of the chief factors 
influencing survival of the pathogen. Additional in- 
formation on this comes from the work of Shigo ( 1958) 
on trees girdled as part of the oak wilt control program 



in West Virginia. This treatment reduces mat forma- 
tion due at least in part, to rapid colonization of the 
sapwood by Hy poxy Ion punctulatum. 

In the roots of a diseased tree, survival of C. faga- 
cearum may be more prolonged, although this seems 
to depend on whether the roots are grafted to those of 
a neighboring tree. In central Wisconsin, where root 
grafting is common, survival of up to 3 years has been 
recorded 3 and in Pennsylvania, with roots that were 
probably grafted, the fungus was isolated from one 
out of three trees that had been dead for 3 years and 
five out of eight trees that had been dead for 2 years 
(Yount 1955). Skelly (1967), also in Pennsylvania, 
obtained similar results. He recovered the fungus 
from the root systems of 52 percent of red oaks that 
had been dead for 1 year, 18 percent from trees dead 
for 2 years, and 3 percent from trees dead for 4 years. 
By contrast, Amos and True (1967) working with 
girdled trees not grafted to living trees in West Vir- 
ginia, found that although the fungus could readily 
be isolated from roots after 1 year, it was rarely found 
after 2. Armillaria mellea rhizomorphs were com- 
monly present on the roots after 1 year. Umbelopsis 
versiformis, Trichoderma viride, and Pencillium spp. 
were the fungi most commonly obtained from the 
bark, although Gliocladium roseum and Dothiorella 
spp. were also present. T. viride and Pencillium spp. 
were also commonly isolated from the wood. 

Survival of the pathogen has hitherto been dis- 
cussed only in relation to red oaks. In diseased white 
oaks the distribution of the fungus in the xylem of the 
current annual ring is much more restricted. And if 
the tree recovers, the infected ring will be buried 
under new wood (Parmeter et al. 1956). Isolation 
studies indicate that the pathogen survives in these 
buried rings for only a few years and in any event it is 
unlikely to constitute a significant source of inocu- 
lum. This is borne out by studies on infected but 
asymptomatic chestnut oak (Cobb et al. 1965a). If a 
white oak becomes infected and dies within a single 
season, the same events as described for red oaks 
could probably occur and an inoculum source could 
thus be created. 



SEASONAL SUSCEPTIBILITY 
OF TREES TO INFECTION 

One factor that may have some significance for all 
the possible means of disease transmission is the 
effect of season on host susceptibility. Engelhard 



3 Kuntz, J. E. Personal communication. 



(1956) in Iowa inoculated fresh stem wounds on red 
oaks at 4- to 5-week intervals throughout the year. 
Disease only occurred on trees inoculated between 
April 25 and August 28. Drake et al. (1956) in Wis- 
consin inoculated northern pin oak from April to 
September. The trees were most susceptible in early 
June. Many of the trees inoculated in April and May 
also became diseased, but symptoms did not appear 
as quickly. Few of the trees inoculated after summer- 
wood had begun to form became diseased. Nair and 
Kuntz (1963b) found that some northern pin oak 
trees wilted following branch inoculations in every 
month from February to late November, with a peak 
between May and August. All wilting trees died. 
Trees became diseased following stem inoculations in 
all months of the year. With both methods trees inoc- 
ulated in the dormant season showed symptoms the 
following June. Skelly and Merrill (1968) in Pennsyl- 
vania also found that some red oaks became diseased 
following stem inoculations during each month from 
November to March. They suggested that tempera- 
tures above freezing at the time of inoculation might 
be important. Nair and Kuntz (1963b) also inocu- 
lated bur oaks. The highest disease incidence follow- 
ing branch inoculations was from mid-May to early 
August. Some wilt occurred following stem inocula- 
tions in every month except December, January, and 
February. The highest incidence (40 to 80 percent) 
was from early May to early August. Cobb et al. 
(1965a) in Pennsylvania found that Chestnut oak 
was also most susceptible to stem inoculations in 
May and June. 



UNDERGROUND SPREAD 

Oak wilt spreads underground by the passive 
movement of spores of the fungus from a diseased tree 
to an adjacent healthy tree via the continuous xylem 
system that exists between root-grafted trees (fig. 2). 
Kuntz, Riker, and their associates in Wisconsin 
(Kuntz and Riker 1950, Beckman and Kuntz 1951) 
showed that almost all the northern pin oak within 
15 m of each other in the central part of the State 
were grafted together. Under these circumstances 
root graft transmission is, not surprisingly, the main 
means of disease spread. In Wisconsin and parts of 
southern Minnesota the disease characteristically 
develops as a number of clearly defined disease cen- 
ters which expand at the rate of about 7.5 m per year 
in each direction ( fig. 3 ) ( French and Bergdahl 1973 ). 

From similar studies conducted in other States in 
the oak wilt range it was concluded that root grafting 




Figure 2. — Grafted roots of northern pin oak. 




Figure 3. — Aerial view of an infection center in north- 
ern pin oak in Minnesota. The disease is spreading 
through root grafts. 

was less common and consequently root transmission 
of the disease less important. In Pennsylvania it was 
suggested that 20 percent of the oaks were grafted 
together (Craighead and Nelson 1960), in North Car- 
olina 15 percent (Boyce 1957a), and in West Virginia 
10 percent (True et al. 1960). In Missouri 14 percent of 
175 oaks were root grafted and it was considered from 
inoculation studies that many of these grafts were 
not functional as far as transmission of the pathogen 
was concerned (Jones and Partridge 1961). In these 
four States the maximum distance between grafted 
trees was 10 m. 



A number of factors influence grafting frequency. 
Site factors seem to play a role because grafting fre- 
quency in northern pin oak is high in parts of southern 
Minnesota but is apparently low in north-central 
Minnesota. In Crow Wing County no root transmis- 
sion occurred from inoculated to uninoculated trees 
although many were as close as 1.5 m (French and 
Schroeder 1969). Genetically controlled differences 
between species are also important. In that part of 
central Wisconsin in which virtually all the northern 
pin oak are grafted together, Parmeter et al. (1956) 
found that only 6 percent of the bur oaks were grafted. 
Jones and Partridge ( 1961 ) found that out of 29 grafts 
detected in Missouri, only one was between two 
different species (Q. velutina and Q. marilandica) so 
grafting between species is generally not important. 
However, in a situation where grafting is uncommon, 
interspecific grafting may comprise a significant por- 
tion of the total. Boyce (1957a) in North Carolina 
found that out of a total of seven grafts, three were 
between black oak and white oak. Similarly, Par- 
meter etal. ( 1956 ) in Wisconsin found that two of seven 
bur oak grafts were with northern pin oak. 

Recently, work in West Virginia has placed re- 
newed emphasis on the role of root transmission in 
that State. First, injections of wilting red oak trees 
with cacodylic acid showed that 31 percent were con- 
nected to healthy trees (Rexrode and Frame 1977), 
substantially more than the 10 percent suggested 
from earlier work in the State. Interestingly, 3 of the 
12 grafts were between red and white oaks. Second, 
and more important, Rexrode (1978) showed that 
transmission via the root system might take several 
years to occur — 4 out of 10 trees that wilted through 
root transmission of the fungus did not develop symp- 
toms until 3 years after the adjacent inoculated trees 
had died. As Rexrode points out, such a method of 
disease transmission would agree with the observed 
distribution of the disease in northeast West Virginia 
where 50 percent of the wilting trees are within 15 m 
of previously infected trees, but where several years 
may elapse between the death of one tree and the 
appearance of symptoms in another. Such a distribu- 
tion pattern is hard to explain in terms of transmis- 
sion by insect vectors. Delayed root transmission 
may also be important in other States. In the Sinis- 
sippi Forest in Illinois, Himelick and Fox (1961) 
noted that 91 percent of infected trees were within 9 
m of trees killed by the disease, but that the infection 
centers might remain dormant for several years be- 
fore breaking out again. In 25 percent of the infection 
centers 4 or more years elapsed between the appear- 
ance of the disease in one tree and its appearance in 



another. Similar situations have been described in 
Pennsylvania (Craighead and Nelson 1960) and 
southern Michigan 4 . In Pennsylvania, where it was 
the practice to fell all healthy oaks within 15 m of a 
diseased tree, Yount (1955) suggested that slow 
movement of the fungus through grafted roots of. 
these felled trees might explain the sporadic appear- 
ance of disease at the periphery of the cleared area 
several years later. 



OVERLAND SPREAD 

The Detection of C. fagacearam 
on Insects and Other Material 

C. fagacearum is well adapted to spread by ani- 
mals, and it has long been accepted that overland 
spread involves a vector. Yount etal. (1955) detected 
C. fagacearum on the bodies of insects collected from 
sporulating mats by plating out washings from these 
insects onto agar. However, C. fagacearum is nor- 
mally overrun by other microorganisms when iso- 
lated on conventional agar media and no adequate 
selective medium has been devised. Therefore, most 
workers use the spermatization technique as de- 
scribed by Jewell ( 1956), which relies on the fact that 
C. fagacearum possesses two mating types — A and B. 
In this technique insects are macerated in water and 
then aliquots of the macerate are spread over cul- 
tures of the A and B mating types. The production of 
perithecia in the cultures then demonstrates the 
presence of viable propagules of the opposite mating 
type. Both ascospores and conidia can be detected in 
this way (Stambaugh et al. 1955). 

Fergus et al. ( 1961 ) found that as few as 20 conidia 
could be detected by this technique. However, a num- 
ber of factors must be optimum for this degree of 
sensitivity to be realized: (1) the composition of the 
medium (Bell and Fergus 1967), (2) the receptivity of 
the female isolates (Fergus et al. 1961), (3) the age of 
the cultures (Rexrode and Jones 1971), and (4) the 
temperature of incubation (Cobb et al. 1962). Care 
must be taken that sterile perithecia, which will form 
even in unspermatized cultures (Bell and Fergus 
1967), are not regarded as evidence of fertilization. 
Recently Peplinski and Merrill ( 1974 ) suggested that 
pycnidia of Pyrenochaeta sp. and even synnemata of 
Graphium rigidum (syn. Ceratocystis piceae) devel- 
oping on C. fagacearum cultures might have been 
confused with fertile perithecia. Despite these com- 
plications the technique has provided valuable data. 



4 Hart, J. Personal communication. 



Transmission by sap-feeding insects 

The discovery of sporulating mats of C. faga- 
cearum excited an immediate interest in their func- 
tion. It was quickly realized that the fruity odor 
attracted insects and that these gained access to the 
mats at points where the central pressure pad had 
cracked the bark. Particularly abundant were Niti- 
dulid beetles (fig. 4). Leach etal. (1952) showed that 
these insects could act as agents of fertilization for C. 
fagacearum , transporting conidia of type A to mats of 
type B, and vice versa. Perithecia then developed on 
the mat. In rare cases, if a mixed thallus of A and B 
mating types was present in a single tree ( Hepting et 
al. 1952), perithecia might be formed before the bark 
cracked open to expose the mat. Perithecia are not 
produced on all mats — Curl (1955) found perithecia 
on 90 out of 393 mats in Illinois, Morris and Fergus 
(1952) found perithecia on only 1 tree out of many 
examined in Pennsylvania, and Shigo (1958) found 
them on only 19 out of 164 mats collected in West 
Virginia. 

Clearly the mats could serve as abundant reser- 
voirs of inoculum, initially in the form of conidia and 




Figure 4. — Nitidulids feeding on a mat o/"Ceratocys- 
tis fagacearum on a dead tree. 



later as ascospores. Knowledge that the Nitidulids 
(sap-feeding insects) were attracted to the mats indi- 
cated that they might act as vectors if they migrated 
from mats to wounds on healthy trees. In May 1953 
Dorsey et al. in West Virginia placed Nitidulids that 
had been artificially contaminated with C. faga- 
cearum in wounds on healthy oaks. Five out of six of 
the trees developed symptoms. In the same year in 
Iowa, Norris (1953) caged Nitidulids, freshly col- 
lected from mats, on wounds on red oaks and obtained 
infection in every case. 

In 1954 infection of wounds via Nitidulids was also 
reported from Illinois, Pennsylvania, and Wisconsin, 
but success rates were much lower. In Illinois Himel- 
ick et al. (1954) used naturally contaminated insects 
and obtained infection in 1 out of 36 trees wounded 
between April 22 and May 22. In Pennsylvania 
Thompson et al. (1955) working with field-collected 
insects subsequently placed on mats producing asco- 
spores, only obtained infection in 1 out of 175 trees 
wounded between April 19 and June 15. In the work 
of McMullen et al. (1955a) in Wisconsin, field-col- 
lected beetles were exposed overnight to mats or lab- 
oratory cultures. Infection was obtained in 15 out of 
52 trees wounded between May 1 and June 19. 

Many of these experiments were highly artificial. 
Indeed it has been suggested that glass marbles 
rolled on a sporulating mat and then placed in fresh 
wounds might also act as "vectors". Consequently, 
subsequent research was devoted to a more detailed 
examination of the inoculum source, behavior of the 
Nitidulids, and the infection court. 

Inoculum source. — Following the discovery of 
mats in Illinois, reports of their occurrence on wilt- 
killed red oaks quickly came in from many States. By 
contrast, there have been few reports for white oaks. 
Parmeter et al. ( 1956) did not find any mats on hun- 
dreds of inoculated bur-oaks in Wisconsin. However, 
in May 1954 Engelhard (1955) found 42 matlike 
structures on an inoculated bur oak in northern Iowa. 
Some consisted of both mycelium and pad; others of 
pad only. They were difficult to locate because of the 
furrowed or flaky nature of the bark. Nair and Kuntz 
( 1963a) also described minute mats on the trunks and 
branches of inoculated bur oaks in Wisconsin. The 
mats sporulated profusely and sap-feeding beetles 
were found on them. In November 1965 numerous 
oak wilt mats were observed on a recently killed 12- 
inch d.b.h. white oak in West Virginia. Thirty of 
these had apparently formed pads that had cracked 
the bark (Cones 1967). 



Data on the proportion of wilted red oaks producing 
mats are available from several workers. In north- 
eastern West Virginia about 25 percent of the trees 
produced mats, whereas in the southern part of the 
State the percentage was around 80, (Rexrode and 
Frame 1973 1 , Rexrode 1977). By contrast, mats are 
rarely formed in Missouri and Ohio (Donley 1959). T. 
W. Jones 2 indicated a figure of only 5percent for 
Missouri. Particularly important are data on mat 
production during the crucial period of peak host 
susceptibility between April and July or late June. In 
northern Illinois, Curl (1955) found that 22 out of 28 
trees wilting between June and August produced 
mats the following spring. In North Carolina, Boyce 
(1954) reported that 12 out of 22 trees infected in the 
summer of 1952 produced mats the following spring. 
In Minnesota, Campbell and French (1955a) estab- 
lished that spring mat production occurred princi- 
pally on trees wilting in August. Jeresek (1976) 
found that in 1972, 19 percent of oaks wilting in 
August and early September produced mats the fol- 
lowing spring and in 1973 the figure was 26 percent. 
In Pennsylvania, Cobb et al. (1965a) found that 
spring mat production occurred principally on trees 
that had been inoculated in July and that had devel- 
oped symptoms a few weeks later. 

The seasonal pattern of mat formation, with peaks 
in spring and autumn, has been related to tempera- 
ture — both high and low temperatures are unfavor- 
able. In Minnesota mean weekly temperatures of 
50°F ( IOC) are optimum for mat formation although 
they survive longer at cooler temperatures (Camp- 
bell and French 1955a). A high sapwood moisture 
content is also considered important (Campbell and 
French 1955b). Boyce (1957b) thought that high 
rainfall during autumn might influence the produc- 
tion of spring mats on felled trees. 

There is no evidence that the nature of the inocu- 
lum is critical. Himelick et al. ( 1954) wondered if the 
low level of infection in their experiments might be 
because they were using an endoconidia inoculum 
while Dorsey et al. ( 1953 ) and Norris ( 1953 ) had used 
beetles that could also have been carrying ascos- 
pores. However, in other experiments a low degree of 
infection has been obtained with ascospores (Thomp- 
son et al. 1955) and a higher level with endoconidia 
(McMullen et al. 1955a). 

It is possible that the Nitidulids are contaminated 
with inoculum from sources other than the sporulat- 
ing mats. In Pennsylvania, Craighead and Nelson 
( 1960) suggested that sporulation might occur in in- 
fected but symptomless chestnut oak but Cobb et al. 



( 1965b) found no evidence for this. However, they did 
show that sporulation could occur in wounds on dis- 
eased red oak that were close to the point of mat 
production. 

Vector behavior. — The biology of the Nitidulids 
was studied first in West Virginia (Dorsey and Leach 
1956) and in Iowa (Norris 1956). Subsequent studies 
included those of McMullen et al. ( 1960 ) in Wisconsin 
and Skalbeck (1976) in Minnesota. The time of peak 
abundance varies with each species but is usually in 
the spring. Dorsey and Leach ( 1956) considered mean 
weekly temperatures of about 50°F (IOC) to be opti- 
mum for Nitidulid activity. The circumstances under 
which Nitidulids leave the sporulating mats and fly 
to wounds on healthy trees have long been the subject 
of discussion. In Pennsylvania Morris et al. (1955) 
found that Nitidulids remained on the mats until the 
latter deteriorated or dried up. When migration did 
occur it was more often from mat to mat than from 
mat to wound. They concluded that "unless mats at 
the right stage of decline are available at the time the 
wounds are made or shortly thereafter, infection 
through these wounds is improbable". McMullen et 
al. (1960) concurred with this. They discovered that 
the largest numbers of insects were found on wounds 
made in mid-June and that this coincided with cessa- 
tion of mat production and the deterioration of the 
existing mats in the area. The highest percentage of 
wilting (about 20 percent) also occurred in trees 
wounded at this time. 

Little information is available on other sap-feeding 
animals that might be vectors but a number of possi- 
ble candidates are discussed by Craighead and Nel- 
son (1960). 

The infection court. — Evidence that wounding 
leads to infection is a vital part of the case for Nitidu- 
lids as vectors. In Pennsylvania, Guyton ( 1952 ) noted 
an association between wilt and wounds blazed to 
mark a future logging road, and Jeffrey (1953) 
reported many instances of wilt associated with 
pruning, climbing-iron damage, and logging wounds. 
Infections only occurred when the wounds were made 
between late April and early June, the period of 
spring wood formation in Pennsylvania. In northern 
Iowa, Norris (1955) observed that when a tree some 
distance from other wilted trees became diseased, a 
conspicuous wound was almost always present on the 
trunk or major limbs. Frequently the wilt symptoms 
were restricted to that part of the tree distal to the 
wound. (This last may not be relevant because symp- 
tom expression usually begins in the distal part of a 



branch.) In Wisconsin, McMullen et al. (1955a) re- 
ported that 26 trees wounded in June became dis- 
eased in July. Further evidence was provided by 
Kuntz and Drake ( 1957 ) who reported that 19 percent 
of 109 oaks pruned in mid-June became diseased, and 
that the removal of stump sprouts in May and June 
led to 27 percent of the remaining stems becoming 
infected. In Minnesota, D. W. French 2 has shown 
from evidence collected during a 20-year-period that 
overland infection is closely associated with tree 
wounding in late May or early June. He noted that 
trunk wounds such as those made by climbing irons, 
appeared to be more important infection courts than 
pruning wounds. 

Experiments involving deliberate wounding have 
confirmed that wounds made in May and June are 
most likely to lead to infection. In Pennsylvania, 
Craighead et al. (1953) reported that eight trees 
wounded in May were diseased in July and that lar- 
vae of Nitidulids and Diptera were abundant in the 
wounds of these trees. In Iowa, Norris (1955) found 
that disease occurred in 31 of 122 trees wounded 
between April 24 and June 22. Only 2 of 217 control 
trees died and these had been wounded by rodents. 
Similar results were obtained in Wisconsin with 11 
out of 60 trees wounded between May 15 and June 18 
becoming diseased (McMullen et al. 1955a, 1960). 
Recently Jeresek (1976) provided similar data for 
Minnesota. In 1974, 19 out of 82 trees wounded be- 
tween May 19 and June 9 became diseased and in 
1975, 6 out of 30 trees wounded between May 26 and 
June 9 became diseased. Trees on which the wounds 
were painted immediately did not become diseased, 
nor did any unwounded control trees. 

The rate of wound infection seems to vary from 
year to year. Nair and Kuntz ( 1963 ) wounded 15 bur 
oaks and 20 Northern pin oaks at weekly intervals 
from May to August each year from 1960 to 1962. In 
the first 2 years no trees were infected but in 1962 10 
bur oaks and 28 pin oaks became diseased. These 
trees were all wounded between May 21 and June 11. 

There is only one report of wound infection that 
does not fit the seasonal pattern. In Pennsylvania, 21 
of 101 red oaks pruned during the winter of 1957, 
wilted the following summer (Craighead and Nelson 
1960). 

Only one experiment has been described on the 
effect of wounding on infection from outside the north- 
ern part of the oak wilt range. In Missouri Buchanan 
(1960) found that oak wilt developed only in trees 
wounded between April 19 and April 26 — 7 out of 100 
trees wounded during this period became diseased. 



The possible importance of the condition of the 
wound and its location have received some attention. 
Clean cut or "bruise" wounds seem to be equally 
attractive to Nitidulids (Morris et al. 1955) and, as 
long as they reach the xylem, equally suitable for 
infection. Cobb et al. ( 1965a ) showed that wound loca- 
tion could be important for red and Chestnut oak; less 
infection occurred following inoculation of branch 
wounds than trunk wounds. Wound age also has an 
important effect. Kuntz and Drake (1957) in Wiscon- 
sin inoculated 10- to 30-cm diameter northern pin 
oaks at various intervals after wounding. They ob- 
tained 100 percent infection the first 4 days but no 
infection after 8 days. Natural infection occurred 
only within 24 hours of wounding. Morrises al. (1955) 
and Cobb et al. (1965a) found that wounds in red oak 
in Pennsylvania were not suitable as infection courts 
after 3 days. On bur oak, Nair and Kuntz (1963a) 
reported that they could only obtain infection for 24 
hours after wounding, and on Chestnut oak, Cobb et 
al. (1965a) found a progressive reduction in suscepti- 
bility, with no infection after the fifth day. The reduc- 
tion in wound susceptibility with time may be due in 
part to the formation of wound tyloses or the accumu- 
lation of phenolic compounds, as suggested by Cobb et 
al. (1965a). However, the colonization of the wound 
surface by other microorganisms is also important. 
Gibbs ( 1980b) showed that if Ceratocystis piceae (syn. 
Graphium rigidum), one of the fungi most commonly 
associated with wounds on healthy oak (Shigo 1958), 
was introduced to a fresh wound 24 hours prior to its 
inoculation with C. fagacearum, no infection re- 
sulted. However such a wound remained fully sus- 
ceptible in the absence of C. piceae. In nature C. 
piceae is probably brought to the wounds by insects 
(Jewell 1956). 

It has long been thought that moisture in the 
wound has an important influence on natural infec- 
tion (Himelick et al. 1954). It could affect both the 
transfer of spores from the insect to the xylem surface 
and also the germination and development of the 
fungus. Such moisture is probably most readily 
available in early summer when the vascular cam- 
bium is at its most active stage and many cells are in 
the process of differentiation. Rain water may also be 
important. In the experiments of McMullen et al. 
(1955a, 1960) in which Nitidulids were caged over 
drill holes, there was some evidence for a correlation 
between rainfall during the first 3 days after wound- 
ing and infection. The "fermenting sap" sometimes 
present in older wounds and regarded as highly at- 
tractive to Nitidulids (Morris et al. 1955) is almost 
certainly not an aid to infection. Its formation is 



9 



associated with the activity of other microorganisms, 
the presence of which is likely to prevent the estab- 
lishment of C. fagacearum. 

As with other wilt pathogens the number of spores 
introduced to the wound is not of great importance. 
Cobb et al. (1965a) found no great difference in the 
incidence of infection in trees inoculated with spore 
doses of between 10 3 and 10 6 . 



Transmission by Tree-wounding 
Animals 

Oak bark beetles 

The oak bark beetles Pseudopityophthorus spp. 
were among the first insects suspected as vectors of 
the disease. These minute beetles breed in wilt-killed 
trees and feed on the twigs of healthy oak (fig. 5). 
Using artificially contaminated beetles Griswold and 
Bart (1954) obtained infection in 2 out of 6 oak seed- 
lings and Donley (1959) obtained infection in 14 out 
of 135 seedlings. Attempts to achieve infection by 
caging field-collected beetles on young oaks were 
largely unsuccessful, however. In Pennsylvania 
Craighead and others 2 caged beetles from oak wilt 
trees on healthy saplings each year from 1954 to 
1957. As many as 300 beetles were placed in some 
cages but no disease resulted. Buchanan (1958) in 




Figure 5. — Pseudopityophthorus pruinosus feeding 
on scarlet oak (photograph courtesy of C. O. 
Rexrode). 



Missouri was a little more successful. He obtained 
oak wilt in 2 out of 204 red oak seedlings fed on by 
beetles that had emerged in spring from trees that 
had become diseased the previous year. (About 
140,000 beetles were involved in these tests.) With 
these results it is not surprising that in their review 
of oak wilt, True et al. ( 1960) did not consider the bark 
beetles to be of great importance in the transmission 
of the disease. More recently, however, the oak bark 
beetles have received much more attention. The 
main reason for this renewed interest has been the 
opinion that mat production in the south and west of 
the oak wilt range, is too rare an event to explain the 
observed incidence of the disease. In view of this 
change of emphasis, the relevant data on the biology 
of the oak bark beetles, and of the evidence for their 
role as vectors, are reviewed in some detail. 

The beetles and their life cycles. — Pseudopity- 
ophthorus spp. are present throughout and far be- 
yond the range of oak wilt. The two key species are P. 
minutissimus and P. pruinosus. The former appears 
to have a more northerly distribution and is the only 
species recorded in Wisconsin. The two species are, 
however, similar in behavior and research workers 
often have not differentiated between them. There 
are at least two generations per year as far north as 
the Lake States (McMullen et al. 1955b). In Ohio, all 
the stages successfully overwinter except the pupae 
(Rexrode 1969). In Wisconsin, however, the larger 
larvae are the only winter-resistant stage, and 
emerge as adults in May (McMullen et al. 1955b). 

Breeding and feeding habits. — Pseudopityo- 
phthorus spp. most commonly breed in oak, although 
other hosts have been recorded. Interestingly, 
McMullen et al. (1955b) in Wisconsin found that P. 
minutissimus was common in trees of the red oak 
group but was not found in white or bur oak. Breed- 
ing normally takes place in stems or branches from 1- 
to 10-cm diameter, although it has been recorded in a 
42-cm diameter tree in West Virginia (Rexrode et al. 
1965). The proportion of oak-wilted trees that are 
successfully colonized ranges from near to 50 per- 
cent. It seems to be at its highest in Missouri where 
most of the diseased black and scarlet oaks (Q. uelu- 
tina and Q. coccinea) were attacked (Buchanan 1956) 
and where successful breeding occurred in 45 out of 
87 wilted trees (Rexrode and Jones 1972). In West 
Virginia, attack in the main stem occurred in 8 of 27 
trees in 1961 and 4 of 30 trees in 1962 (Rexrode et al. 
1965). Studies of trees that had been dead for several 
years suggested that these figures were above aver- 
age. Attack on small branches occurred in half the 



10 



trees in both years. In 1967 in Ohio Rexrode (1967) 
found only 3 out of 27 trees to be attacked, and coloni- 
zation was light and restricted to the branches. In 
Minnesota, although small low branches on healthy 
trees almost invariably become colonized when they 
die from "shading", attack on the crowns of oak- 
wilted trees is usually light and concentrated on 
branches from 2 to 5 cm in diameter (Gibbs 1980a). 
From the observations in Minnesota, and earlier in 
Missouri (Buchanan 1956) and Wisconsin (McMullen 
et al. 1955b) it seems that Pseudopityophthorus spp. 
are best adapted for the colonization of slowly dying 
branches. Attacks on diseased trees before defolia- 
tion has occurred are normally abortive ( Rexrode and 
Jones 1970) presumably because host resistance is 
too high. After defoliation, branches quickly become 
unsuitable for breeding, probably because the tissues 
dry out rapidly and the bark is colonized by fungi 
such as Dothiorella quercina and Coryneum kunzei 
(Gibbs 1980a). 



from trees that had wilted earlier that year were 
contaminated with C. fagacearum. In Missouri Berry 
and Bretz (1966), also working on a Pseudopityo- 
phthorus generation that emerged during midsum- 
mer, found that beetles from 9 out of 12 trees were 
carrying C. fagacearum. As many as 30 percent of the 
beetles were contaminated from some of the trees. In 
general the mating type of the fungus on the beetles 
was the same as that from the xylem of the trees from 
which they emerged. A few beetles were carrying the 
other mating type, however, and both mating types 
were found on beetles from five of the nine trees. 
Several studies have shown that both mating types 
are rarely found together in the xylem of an infected 
tree (Boyce and Garren 1953, Barnett and Staley 
1953). Therefore, some of the C. fagacearum present 
on the emerging beetles was probably introduced to 
the branches by the parent beetles when they entered 
to breed. This is known to occur in elms infected with 
Ceratocystis ulmi (Lea 1977). 



Although it is clear that the bark beetles do not 
take full advantage of the death of trees through oak 
wilt, many can emerge from a single tree. In Ohio, 
Donley (1959) found that an average of 11,400 Scoly- 
tids ( virtually all Pseudopityophthorus spp. ) emerged 
in spring from a 15-cm diameter diseased black oak. 

The feeding habits of the oak bark beetles on 
healthy trees were first investigated by Griswold and 
Neiswander (1953) and Griswold and Bart (1954). 
They concluded that Pseudopityophthorus spp. com- 
monly made deep feeding wounds in the crotches, leaf 
axils, bud axils, and immature acorn axils of both red 
and white oaks. Rexrode and Jones (1970) reported 
that fresh feeding wounds could readily be found 
from mid-April onwards in Missouri, Ohio, and West 
Virginia. Feeding was primarily in the top branches 
of dominant trees and occurred mainly at the node 
between the previous year's and the current year's 
growth. The beetles bored through the bark, cam- 
bium, and xylem to the center of the twig. Rexrode 
and Jones found that such wounds acted as infection 
courts when they were inoculated with a spore sus- 
pension of C. fagacearum. 

Bark beetles and C. fagacearum. — The first evi- 
dence that bark beetles could carry C. fagacearum 
was provided by Buchanan (1956) who obtained some 
infection when large numbers of bark beetles from 
oak wilt trees were macerated in water and placed on 
wounds of healthy trees. Stambaugh et al. (1955) 
reported that between 0.7 and 7 percent of the Pseu- 
dopityophthorus beetles emerging in June and July 



Beetles emerging during the latter half of the sum- 
mer are not likely to act as vectors. Host susceptibil- 
ity is low and late summer feeding is concentrated on 
the petioles of the leaves of the current shoots which 
provide a less favorable avenue for infection than 
spring feeding wounds 1 . With these considerations 
in mind Rexrode and Jones (1971) examined beetles 
that emerged in early spring from the small branches 
of trees that had wilted the previous July. They rec- 
orded C. fagacearum on beetles from 8 of 17 trees 
from Missouri and 7 of 15 trees from West Virginia. 
The percentage of contaminated beetles from all the 
trees together was between 0.4 and 2.5. They also 
reported that the fungus was present on three imma- 
ture stages of the beetle — larvae, pupae, and tenerals 
— and also in the frass. No sporulating mats were 
found on any of the branches and attempts to isolate 
the fungus from the branches during the time of 
beetle emergence were unsuccessful. 

Parent beetles might also be transmitters of the 
disease in the spring. They may make a gallery sys- 
tem in a diseased tree, emerge to feed on twigs of 
healthy trees, and then breed again in a healthy tree. 
Rexrode et al. (1965) found that between 0.5 and 5 
percent of 440 parent adults re-emerging from the 
stems of wilted trees in West Virginia were carrying 
C. fagacearum. They suggested that the beetles in- 
gested the fungus from the xylem vessels while mak- 
ing the galleries. It is also possible, however, that 
they were already contaminated when they entered 
the trees and retained the fungus on or in their bodies 
while there. 



11 



Extensive microscopic examination of beetle gal- 
leries by J. G. Leach and others (see Rexrode et al. 
1965) have failed to reveal mycelium or spores of C. 
fagacearum, and the mechanism whereby the beetles 
become contaminated with the pathogen is not clear. 
Both inner bark and outer xylem are possible sources 
because early larval stages develop principally 
within the bark and later larval stages scar the xy- 
lem. If the inoculum comes principally from the 
strain of the fungus that kills the tree, the xylem is 
the more likely source because C. fagacearum rarely 
invades the inner bark of small branches (Gibbs 
1980a). Nothing is known about situations in which 
the fungus might be introduced to gallery systems by 
beetles entering to breed. By analogy with Dutch elm 
disease, development of the pathogen in either bark 
or xylem seems possible. 



Other tree-wounding insects 

Although the oak bark beetles have received the 
most attention, they are not the only possible vectors 
among the tree-wounding insects. The other chief 
candidates include the flat-headed borers (Bupresti- 
dae) such as the two-lined chestnut borer (Agrilus 
bilineatus) (Rexrode 1968). Young adults emerge 
from the trunks of wilted trees and then feed on twigs 
and leaves of healthy trees. Although the necessary 
period of at least a year for larval development would 
be expected to reduce greatly the number of adults 
carrying C. fagacearum, Stambaugh et al. (1955) 
found that between 4 and 20 percent of a sample of 
128 insects were contaminated with the fungus. 
However, Craighead et al. 2 did not achieve infection 
when 200 field-collected beetles were caged on small 
trees. Himelick and Curl (1958) also obtained nega- 
tive results, even though beetles they used had been 
exposed to cultures of C. fagacearum. The flat-headed 
apple tree borer, Chrysobothris femorata, has similar 
habits and Himelick and Curl (1958) and Donley 
( 1959) achieved some infection of seedlings with arti- 
ficially contaminated insects of this species. 

Donley (1959) found that artificially contaminated 
adults of the round-headed borer Urographis fascia- 
tus infected 16 of 135 oak seedlings. It is interesting 
to note that these results are almost identical to those 
obtained by him for P. minutisimus, although the 
number of U. fasciatus adults used per cage was only 
one-third of the number of oak bark beetles. Whether 
U. fasciatus ever carries the pathogen in nature, 
however, has not been determined. 

With a life cycle that may be as short as 6 weeks, 
the Ambrosia beetles might be expected to carry the 



pathogen frequently, and consequently they have 
been studied in some detail. Xyleborus spp. and Xy- 
loterinus politus were found emerging through oak 
wilt mats, and, not surprisingly, many of them were 
contaminated with C. fagacearum (Stambaugh et al. 
1955). Stambaugh et al. (1955) also found that be- 
tween 2 and 10 percent of 100 adults of X. politus from 
the sapwood of wilted trees were carrying the patho- 
gen, and Skelly ( 1966) found the pathogen on various 
Ambrosia beetles from the roots of wilted trees. How- 
ever, it has recently been concluded that these insects 
do not act as vectors, some species because they do not 
attack healthy trees and others, in particular X. poli- 
tus, because they are no longer carrying the fungus 
by the time their tunnels reach the xylem (Wertz et 
al. 1971). 

The only quantitative data about the relative 
abundance of these insects are those obtained by 
Donley ( 1959) in Ohio from artificially inoculated 15- 
cm diameter black oak. He found that about 230 
Buprestids and 150 Cerambycids (round-headed bor- 
ers) emerged from each tree, but that these were 
out-numbered by the Scolytids (99 percent Pseudopi- 
tyophthorus spp.) 50 to 1 and 75 to 1, respectively. 
Together with the other data presented here, this is 
enough to indicate that the oak bark beetles have 
good claim to be regarded as the most likely of the 
tree-wounding insects to spread infection. 

Squirrels and other animals 

In the north-central States, squirrels commonly 
feed on sporulating mats and consequently, have 
been suspected of transmitting the disease (Himelick 
et al. 1953). Transmission by squirrels has been re- 
ported under artificial conditions ( Himelick and Curl 
1955) but there is little reason to think that these 
animals are natural vectors (True et al. 1960). It has 
been postulated that birds might act as vectors, be- 
coming contaminated while feeding on insects that 
inhabit the sporulating mats. There is no evidence for 
this (Tiffany et al. 1954, True et al. 1960), however, 
the downy and hairy woodpeckers (Dendrocopus spp. ) 
might merit further investigation because they have 
been reported to make peck marks on healthy trees. 



INFORMATION FROM 
RESEARCH ON CONTROL 

Experiments on disease control are potentially a 
good source of information on the mechanisms of 



12 



disease spread. Early control work in central Wiscon- 
sin in which roots were severed by mechanical or 
chemical means provided strong supporting evidence 
for the importance of root graft transmission. Also, 
the prompt application of tree paints to wounds pro- 
vided evidence for the importance of those wounds as 
infection courts (Kuntz and Drake 1957). Other work 
has given more equivocal results, particularly be- 
cause any one treatment may influence several 
possible mechanisms of spread. In the large-scale 
experiments carried out in West Virginia between 
1970 and 1972, Rexrode and Frame (1973) found that 
felling wilted trees reduced bark beetle breeding by 
half and the production of fall sporulating mats al- 
most to zero. Deep girdling to the heartwood, as car- 
ried out in the West Virginia control program, had no 
effect on beetle breeding but did reduce fall mat pro- 
duction by about 75 percent. However, 2 years of 
these treatments had no effect on the incidence of 
new infection centers. 

A later series of experiments involved the injection 
of cacodylic acid, which was more effective than fell- 
ing both in reducing beetle breeding and mat produc- 
tion (Rexrode 1977). During the 4 years of the study, 
the number of new infection centers was 13 percent 
less in the treated than the untreated plots. In the 
last 2 years of the study, the reduction in new centers 
in the treated plots was 26 percent. There was a 
similar reduction in the number of 'breakover cen- 
ters' (newly infected trees within 15 m of a previously 
diseased tree) but part of the effect here might have 
been due to a lower frequency of root transmission. 



INFORMATION FROM THE 

DISTRIBUTION OF THE 

DISEASE 

Until now, little consideration has been given to 
the geographical distribution of oak wilt in the 
United States and, in particular, to its static nature 
and clearly defined boundaries. Some of these bound- 
aries can be readily explained. The western limits of 
the disease coincide with the original prairie/forest 
boundary beyond which few large populations of red 
oak exist. A reasonable explanation of the sharply 
defined southern boundary in Arkansas can be found 
in terms of the effects of high temperature and com- 
peting fungi on the saprophytic survival of the patho- 
gen in diseased trees (Tainter and Gubler 1973). 
Other boundaries cannot be so readily explained. In 
Minnesota and Wisconsin the disease has a distinct 
northern limit, which is not marked by any obvious 



difference in host population, climate, or soil. The 
same is true of the distribution of the disease in the 
East. True et al. ( 1960 ) produced a map showing those 
parts of the Appalachians in which oak wilt was 
concentrated. The same woodland types, physio- 
graphic characteristics, and shallow soils are found 
to the northeast and the southwest of this affected 
area and True et al. expected the disease to spread 
into these areas. No such spread has occurred. This is 
demonstrated in a striking way in Pennsylvania 
where the disease has a particularly abrupt eastern 
boundary (Craighead and Nelson 1960). Theoreti- 
cally studies conducted on either side of these 
transition zones should provide useful evidence on 
mechanisms of disease transmission. 



CONCLUSIONS 

Through much of the oak wilt range, transmission 
of the disease via root grafts is responsible for most of 
the mortality. This is particularly evident in parts of 
Minnesota and Wisconsin where infection centers 
enlarge steadily through the summer. It may also be 
true in Illinois, Pennsylvania, and West Virginia. 
The frequency of root grafts is lower in these States 
than it is further west but it seems that root trans- 
mission best explains the facts that (Da high propor- 
tion of wilting trees are within a few meters of ones 
previously killed by the fungus and (2) several years 
may elapse between the death of one tree and the 
appearance of symtoms in another. 

In these five States most overland spread can be 
explained in terms of transmission by Nitidulids. In 
Minnesota and Wisconsin, sporulating mats are com- 
monly found on diseased trees in spring, and for both 
these States and for Pennsylvania there is a wealth of 
evidence to show that disease commonly appears in 
trees that were wounded in the spring (May-June). 
The fact that many wounded trees escape infection is 
explicable in terms of factors influencing the move- 
ment of Nitidulids, the presence of spores on Nitidu- 
lids and the condition of the wound surface. 

It does not seem to be a coincidence that the part of 
the oak wilt range within which infection by Nitidu- 
lids is most important, is also the area in which root 
transmission is most conspicuous. The abundant root 
transmission leads to the wilting of many trees in 
July and August. These trees produce mats the fol- 
lowing spring at the time when trees are most suscep- 
tible to infection. 



13 



In the southern part of the oak wilt range, in par- 
ticular in southern Ohio and Missouri, workers are 
less persuaded of the importance of root infection 
than their colleagues further north; although here, 
as elsewhere, many of the trees that die are within 15 
m of infected ones. Natural infection of wounds has 
been shown to occur, but in general there is little 
enthusiasm for Nitidulids as vectors, chiefly because 
sporulating mats are so rarely produced. The oak 
bark beetles appear to be better equipped to act as 
vectors in these States than further north, princi- 
pally because all development stages can survive the 
winter. In addition it appears that the beetles may 
breed in larger diameter branch and stem material. 
Consequently, they have a greater chance of linking 
up with the pathogen. The relation between the di- 
ameter of the branch used for breeding and the occur- 
rence of C. fagacearum on beetles emerging from that 
branch is a matter meriting further investigation. 

It has been nearly 40 years since oak wilt was first 
described in Wisconsin, and, in many parts of the 
United States, the disease now arouses only limited 
interest. It must not be forgotten, however, that the 
present situation could change rapidly if transmis- 
sion of the disease were to become more efficient. The 
European oak bark beetle Scolytus intricatus Ratze- 
burg would seem equipped to be a formidable vector 
of C. fagacearum (Gibbs 1978), and care should be 
taken that this and other similar exotic insects do not 
become established in North America. 



LITERATURE CITED 

Amos, R. E., and R. P. True. 1967. Longevity of Cera- 
tocystis fagacearum in roots of deep-girdled oak- 
wilt trees in West Virginia. Phytopathology 
57:1012-1015. 

Anderson, R. L., and D. D. Stalling. 1955. Oak wilt 
damage, a survey in Central Wisconsin. U.S. De- 
partment of Agriculture Forest Service, Station 
Paper 33, lip. U.S. Department of Agriculture 
Forest Service, Lake States Forest Experiment 
Station, St. Paul, Minnesota. 

Barnett, H. L., and J. M. Staley. 1953. Isolation of 
compatible types of Endoconidwphora fagacearum 
from oak wilt trees naturally and artificially inoc- 
ulated. Phytopathology 43:341-343. 

Barnett, H. L., J. M. Staley, and R. P. True. 1952. 
Mycelial mats of Chalara quercina on killed oak 
trees as a potential source of perithecia in nature. 
Phytopathology 42:531-532. 



Beckman, C. H., and J. E. Kuntz. 1951. Translocation 
of poisons, dyes and radio iodine and its relation to 
oak wilt. Phytopathology 41:2-3. (Abstract) 

Bell, W. R., and C. L. Fergus. 1967. Cultural studies 
of sexual reproduction oi Ceratocystis fagacearum. 
Canadian Journal of Botany 45:1235-1242. 

Berry, F. H., and T. W. Bretz. 1966. Small oak bark 
beetle a potential vector of oak wilt. Plant Disease 
Reporter 50:45-49. 

Boyce, J. S., Jr. 1954. Mat formation by the oak wilt 
fungus on felled versus standing trees. Plant Dis- 
ease Reporter 38:676-677. 

Boyce, J. S., Jr. 1957a. Oak wilt spread and damage 
in the southern Appalachians. Journal of Forestry 
55:499-505. 

Boyce, J. S., Jr. 1957b. Relation of precipitation to 
mat formation by the oak wilt fungus in North 
Carolina. Plant Disease Reporter 41:948. 

Boyce, J. S., Jr., and K. H. Garren. 1953. Compatibil- 
ity types of the oak wilt fungus in 23 Appalachian 
trees. Phytopathology 43:644-645. 

Bretz, T. W. 1955. Some additional native and exotic 
species of Fagaceae susceptible to oak wilt. Plant 
Disease Reporter 39:495-497. 

Buchanan, W. D. 1956. Preliminary tests indicate 
that the small oak bark-beetle may be a vector of 
the oak wilt fungus. Plant Disease Reporter 
40:654. 

Buchanan, W. D. 1958. The small oak bark beetle 
transmits the oak wilt disease under caged condi- 
tions. Plant Disease Reporter 42:546-547. 

Buchanan, W. D. 1960. Insects associated with 
wounds on trees that develop oak wilt. Journal of 
Economical Entomology 53:578-581. 

Campbell, R. N., and D. W. French. 1955a. A 
study of mycelial mats of oak wilt. Phytopathology 
45:485-489. 

Campbell, R. N., and D. W. French. 1955b. Moisture 
content of oaks and mat formation by the oak wilt 
fungus. Forest Science 1:265-270. 

Cobb, F. W., and C. L. Fergus. 1964. Pathogenicity, 
host specificity and mat production of seven iso- 
lates of the oak wilt fungus. Phytopathology 
54:865-866. 

Cobb, F. W., C. L. Fergus, and W. S. Stambaugh. 
1962. The effect of temperature on ascogonial and 
perithecial development in Ceratocystis faga- 
cearum. Mycologia 53:91-97. 

Cobb, F. W., C. L. Fergus, and W. J. Stambaugh. 
1965a. Factors affecting infection of red and chest- 
nut oaks by Ceratocystis fagacearum. Phytopatho- 
logy 55:1194-1199. 



14 



Cobb, F. W., F. A. Wood, and R. A. Schmidt. 
1965b. Occurrence of Ceratocystis fagacearum in 
wounds on red and chestnut oaks. Phytopathology 
55:179-182. 

Cones, W. L. 1967. Oak wilt mats on white oak in 
West Virginia. Plant Disease Reporter 51:430-431. 

Craighead, F. C, C. L. Morris, and J. C. Nelson. 1953. 
A preliminary note on the susceptibility of 
wounded oaks to natural infection by the oak wilt 
fungus. Plant Disease Reporter 37:483-484. 

Craighead, F. C, and J. C. Nelson. 1960. Oak wilt in 
Pennsylvania. Journal of Forestry 58:872-881. 

Curl, E. A. 1955. Natural availability of oak wilt 
inocula. Illinois Natural History Survey Bulletin 
26:277-323. 

Donley, D. E. 1959. Studies of wood-boring insects as 
vectors of the oak wilt fungus. Ph.D. Thesis, Ohio 
State University. 

Drake, C. R., J. E. Kuntz, and A. J. Riker. 1956. 
Sapwood formation and infection by the oak wilt 
fungus. Phytopathology 46:11 (Abstract). 

Dorsey, C. K., and J. G. Leach. 1956. The Bionomics 
of certain insects associated with the oak wilt with 
particular reference to the Nitidulidae. Journal of 
Economical Entomology 49:219-230. 

Dorsey. C. K., F. F. Jewell, J. G. Leach, and R. P. 
True. 1953. Experimental transmission of oak wilt 
by four species of Nitidulidae. Plant Disease Re- 
porter 37:419-420. 

Engelhard, A. W. 1955. Occurrence of oak wilt 
fungus mats and pads on members of the red and 
white oak groups in Iowa. Plant Disease Reporter 
39:254-255. 

Engelhard, A. W. 1956. Influence of time of year and 
type of inoculum on infection of oak trees inocu- 
lated with the oak wilt fungus. Plant Disease Re- 
porter 40:1010-1014. 

Englerth, G. H., J. S. Boyce, and E. R. Roth. 1956. 
Longevity of the oak wilt fungus in red oak lumber. 
Forest Science 2:2-6. 

Fergus, C. L., and W. J. Stambaugh. 1957. An irregu- 
lar and unusual formation of mycelial mats in the 
oak wilt fungus. Mycologia 49:761-766. 

Fergus, C. L., W. J. Stambaugh, F. W. Cobb, and R. A. 
Schmidt. 1961. The effect of conidial concentration 
on perithecial formation by the oak wilt fungus. 
Plant Disease Reporter 45:736-738. 

French, D. W., and D. Bergdahl. 1973. Oak wilt in 
Minnesota 1972. University of Minnesota. Agri- 
culture Experiment Station Miscellaneous Report 
118, 8 p. St. Paul, Minnesota. 



French, D. W., and D. B. Schroeder. 1969. Oak wilt 
fungus, Ceratocystis fagacearum, as a selective sil- 
vicide. Forest Science 15:198-203. 

French, D. W., and W. C. Stienstra. 1978. Oak wilt. 
University of Minnesota. Agriculture Extension 
Service, Extenstion Folder 310, 6 p. St. Paul, 
Minnesota. 

Gibbs, J. N. 1978. Oak wilt. Arboricultural Journal 
3:351-356. 

Gibbs, J. N. 1980a. Survival of Ceratocystis faga- 
cearum in branches of trees killed by oak wilt in 
Minnesota. European Journal of Forest Pathology 
10.[In Press.] 

Gibbs, J. N. 1980b. The role of Ceratocystis piceae in 
preventing infection by Ceratocystis fagacearum in 
Minnesota. Transactions Brit. Mycological Society 
74:171-174. 

Griswold, C. L., and G. J. Bart. 1954. Transmission of 
Endoconidiophora fagacearum by Pseudopityo- 
phthorus pruinosus. Plant Disease Reporter 
38:591. 

Griswold, C. L., and R. B. Neiswander. 1953. Insect 
vectors of oak wilt fungus. Journal of Economical 
Entomology 46:708. 

Guyton, T. L. 1952. An unusual occurrence of oak 
wilt in Pennsylvania. Plant Disease Reporter 
36:386. 

Henry, B. W., and A. J. Riker. 1947. Wound infection 
of oak trees with Chalara quercina and its distribu- 
tion within the host. Phytopathology 37:735-743. 

Henry, B. W., C. S. Moses, C. Audrey Richards, and 
A. J. Riker. 1944. Oak wilt, its significance, symp- 
toms and cause. Phytopathology 34:636-647. 

Hepting, G. H., E. R. Toole, and J. S. Boyce, Jr. 1952. 
Sexuality in the oak wilt fungus. Phytopathology 
42:438-442. 

Himelick, E. B., and E. A. Curl. 1955. Experimental 
transmission of the oak wilt fungus by caged squir- 
rels. Phytopathology 45:581-584. 

Himelick, E. B., and E. A. Curl. 1958. Transmission 
of Ceratocystis fagacearum by insects and mites. 
Plant Disease Reporter 42:538-545. 

Himelick, E. B., and H. W. Fox. 1961. Experimental 
studies on control of oak wilt disease. Bulletin 680, 
48 p. Illinois Agriculture Experiment Station. 

Himelick, E. B., E. A. Curl, and B. M. Zuckerman. 
1954. Tests on insect transmission of oak wilt in 
Illinois. Plant Disease Reporter 38:588-590. 

Himelick, E. B., R. D. Schiem, and E. A. Curl. 1953. 
Rodent feeding on mycelial pads of the oak wilt 
fungus. Plant Disease Reporter 37:101-103. 



15 



Houston, D. R., C. R. Drake, and J. E. Kuntz. 1965. 
Effects of environment on oak wilt development. 
Phytopathology 55:1114-1121. 

Jeffrey, A. R. 1953. The relation of oak wounds made 
during spring wood formation to transmission of 
oak wilt. Plant Disease Reporter 37:568. 

Jeresek, J. D. 1976. Dissemination of spores and pre- 
vention of sporulation of the oak wilt fungus, Cera- 
tocystis fagacearum. MS Thesis, University of 
Minnesota. 71 p. 

Jewell, F. F. 1956. Insect transmission of oak wilt. 
Phytopathology 46:244-257. 

Jones, T. W., and T. W. Bretz. 1955. Radial penetra- 
tion of the oak wilt fungus into the boles of diseased 
trees. Plant Disease Reporter 39:872. 

Jones, T. W., and A. D. Partridge. 1961. The impor- 
tance of root grafts in oak wilt spread in Missouri. 
Plant Disease Reporter 45:506-507. 

Kuntz, J. E., and C. R. Drake. 1957. Tree wounds and 
long distance spread of oak wilt. Phytopathology 
47:22. 

Kuntz, J. E., and A. J. Riker. 1950. Root grafts as a 
possible means for local transmission of oak wilt. 
Phytopathology 40:16-17. 

Lautz, W., and G. C. Saufley. 1970. Evaluation of 
oak wilt incidence in southeastern Missouri. U.S. 
Department of Agriculture Forest Service, North- 
eastern Area State and Private Forestry 5230 
D-5-70. 5 p. 

Lea, J. 1977. A comparison of the saprophytic and 
parasitic stages of Ceratocystis ulmi. Ph.D. Thesis. 
University of London. 

Leach, J. G., R. P. True, and C. K. Dorsey. 1952. A 
mechanism for liberation of sproes from beneath 
the bark and for diploidization in Chalara quer- 
cina. Phytopathology 42:537-539. 

McMullen, L. H., C. R. Drake, R. D. Shenefelt, and J. 
E. Kuntz. 1955a. Long distance transmission of 
oak wilt in Wisconsin. Plant Disease Reporter 
39:51-53. 

McMullen, L. H., E. W. King, and R. D. Shenefelt. 
1955b. The oak bark beetle, Pseudopityophthorus 
minutissimus and its biology in Wisconsin. Cana- 
dian Entomology 87:491-495. 

McMullen, L. H., R. D. Shenefelt, and J. E. Kuntz. 
1960. A study of insect transmission of oak wilt in 
Wisconsin. Transactions Wisconsin Academy of 
Science, Arts, and Letters 49:73-84. 

Merek, E. L., and C. L. Fergus. 1954. Longevity of the 
oak wilt fungus in diseased trees. Phytopathology 
44:328. 



Morris, C. L., and C. L. Fergus. 1952. Observations 
on the production of mycelial mats of the oak 
wilt fungus in Pennsylvania. Phytopathology 
42:681-682. 

Morris, C. L., H. E. Thompson, B. L. Hadley, and J. M. 
Davis. 1955. Use of radioactive tracers for investi- 
gation of the activity pattern of suspected insect 
vectors of the oak wilt fungus. Plant Disease Re- 
porter 39:61-63. 

Nair, V. M. G., and J. E. Kuntz. 1963a. Mat formation 
in bur oaks infected with Ceratocystis fagacearum. 
Forest Research Note 95, 3 p. Wisconsin College of 
Agriculture. 

Nair, V. M. G., and J. E. Kuntz. 1963b. Seasonal 
susceptibility of bur oaks to artifical inoculation 
with the oak wilt fungus. Forest Research Note 97, 
4 p. Wisconsin College of Agriculture. 

Nair, V. M. G., and J. E. Kuntz. 1963c. Wound sus- 
ceptibility in bur oaks to natural infection by Cera- 
tocystis fagacearum. Forest Research Note 94, 3 p. 
Wisconsin College of Agriculture. 

Norris, D. M. 1953. Insect transmission of oak wilt in 
Iowa. Plant Disease Reporter 37:417-418. 

Norris, D. M. 1955. Natural spread of Endoconidio- 
phora fagacearum to wounded red oaks in Iowa. 
Plant Disease Reporter 39:249-253. 

Norris, D. M. 1956. The role of sap and fungus feeding 
insects in the dissemination of the fungus Endo- 
conidiophora fagacearum, the cause of oak wilt. 
Entomological Society of America, North Central 
Branch Proceedings 11:51-52. 

Parmeter, J. R., J. E. Kuntz, and A. J. Riker. 1956. 
Oak wilt development in bur oaks. Phytopathology 
46:423-436. 

Peplinski, J. D., and W. Merrill. 1974. Nonsurvival of 
Ceratocystis fagacearum in frassofoak bark beetles, 
ambrosia beetles. Phytopathology 64:1528-1530. 

Rexrode, C. O. 1967. Preliminary study on the time 
and frequency of oak bark beetle attacks on oak 
wilt trees. Plant Disease Reporter 51:755-757. 

Rexrode, C. O. 1968. Tree-wounding insects as 
vectors of the oak wilt fungus. Forest Science 
14:181-189. 

Rexrode, C. O. 1969. Seasonal development and hab- 
its of Pseudopityophthorus spp. in Southern Ohio. 
Canadian Entomologist 101:306-313. 

Rexrode, C. 0. 1977. Cacodylic acid reduces the spread 
of oak wilt. Plant Disease Reporter 61:972-975. 

Rexrode, C. O. 1978. Movement of oak wilt fungus in 
a tracer solution under pressure through root 
grafts. Plant Disease Reporter 62:982-984. 



16 



Rexrode, C. O., and R. E. Frame. 1973. The effect of 
two oak wilt control methods on oak bark beetle 
populations, mat production and disease incidence. 
Plant Disease Reporter 57:1055-1058. 

Rexrode, C. O., and R. E. Frame. 1977. Root graft 
incidence at oak wilt sites in West Virginia. Plant 
Disease Reporter 61:970-971. 

Rexrode, C. O., and T. W. Jones. 1970. Oak bark 
beetles, important vectors of oak wilt. Journal of 
Forestry 68:294-297. 

Rexrode, C. O., and T. W. Jones. 1971. Oak bark 
beetles carry the oak wilt fungus in early spring. 
Plant Disease Reporter 55:108-111. 

Rexrode, C. O., and T. W. Jones. 1972. Oak bark 
beetle attacks on oak wilt trees in Missouri. Envi- 
ronmental Entomology 1:57-58. 

Rexrode, C. O., H. M. Kulman, andC. K. Dorsey. 1965. 
Bionomics of the bark beetle Pseudopityophthorus 
pruinosus with special reference to its role as a 
vector of oak wilt, Ceratocystis fagacearum. Journal 
Economical Entomology 58:913-916. 

Shigo, A. L. 1958. Fungi isolated from oak-wilt trees 
and their effects on Ceratocystis fagacearum. My- 
cologia 50:757-769. 

Skalbeck, Thomas C. 1976. The distribution of Niti- 
dulidae in deciduous forests of Minnesota. Ph.D. 
Thesis, University of Minnesota. 260 p. 

Skelly, J. M. 1966. Detection of Ceratocystis faga- 
cearum on insects emerging from roots of wilt- 
killed red oaks. Phytopathology 56:902 (Abstract). 

Skelly, J. M. 1967. Survival of Ceratocystis faga- 
cearum in the roots of wilted red oaks to which the 
Pennsylvania control measures were applied. Phy- 
topathology 57:345 (Abstract). 

Skelly, J. M., and W. Merrill. 1968. Susceptibility of 
red oaks to infection by Ceratocystis fagacearum 
during the dormant season in Pennsylvania. Phy- 
topatholoy 58:1425-1426. 

Stambaugh, W. J., C. L. Fergus, F. C. Craighead, and 
H. E. Thompson. 1955. Viable spores oiEndoconi- 
diophora fagacearum from bark and wood-boring 
beetles. Plant Disease Reporter 39:867-871. 



Struckmeyer, B. E., J. E. Kuntz, and A. J. Riker. 
1958. Histology of certain oaks infected with the 
oak wilt fungus. Phytopathology 48:556-561. 

Tainter, F. H., and W. D. Gubler. 1973. Natural bio- 
logical control of oak wilt in Arkansas. Phytopa- 
thology 63:1027-1034. 

Thompson, H. E., B. L. Hadley, and A. R. Jeffery. 
1955. Transmission of Endoconidiophora faga- 
cearum by spore-infested nitidulids caged on 
wounded healthy oaks in Pennsylvania. Plant Dis- 
ease Reporter 39:58-60. 

Tiffany, L. H., J. C. Oilman, and D. R. Murphy. 1954. 
Fungi from birds associated with wilted oak in 
Iowa. Iowa State College Journal of Science 
29:659-706. 

True, R. P., H. L. Barnett, C. K. Dorsey, and J. G. 
Leach. 1960. Oak wilt in West Virginia. West Vir- 
ginia University of Agriculture Experiment Sta- 
tion Bulletin 448t. 

True, R. P., F. W. Craig, and H. L. Barnett. 1951. Oak 
wilt found in West Virginia. Plant Disease Re- 
porter 35:382. 

Turk, F. M. 1955. The biological relationship be- 
tween the oak wilt pathogen and the fungi found 
in wilted oak trees. MS Thesis. University of 
Minnesota. 

U.S. Department of Agriculture, Forest Service. 
1965. Timber trends in the United States. Forest 
Resource Report 17. 

U.S. Department of Agriculture, Forest Service. 
1967. Vascular and Virus diseases. Northeastern 
Forest Experiment Station Report 1966:48. 

Wertz, H. W., J. M. Skelly, and W. Merrill. 1971. 
Ceratocystis fagacearum not transmitted by am- 
brosia beetles. Phytopathology 61:1185-1187. 

Young, R. A. 1949. Studies on oak wilt, caused by 
Chalara quercina. Phytopathology 39:425-441. 

Yount, W. L. 1955. Longevity of the oak wilt fungus 
in oak roots as related to spread through root 
grafts. Plant Disease Reporter 39:256-275. 

Yount, W. L., A. R. Jeffery, and H. E. Thompson. 
1955. Spores of Endoconidiophora fagacearum on 
the external surface of the body of Nitidulids. Plant 
Disease Reporter 39:54-57. 



■ft U.S. Government Printing Office: 1980—669-244/54 Region No. 6 



17 



Gibbs, J. N., and D. W. French. 

1980. The transmission of oak wilt. U.S. Department of Agriculture 
Forest Service, Research Paper NC-185, 17 p. U.S. Department of 
Agriculture Forest Service, North Central Forest Experiment Sta- 
tion, St. Paul, Minnesota. 

Provides an up-to-date review of factors affecting the transmission 
of oak wilt, Ceratocystis fagacearum. Discusses the history and sever- 
ity of the disease, the saprophytic existence of the fungus in the dying 
tree, seasonal susceptibility of trees to infection, overland and under- 
ground spread, the role of animals and insects as vectors or tree 
wounders, and the distribution of the disease. 

KEY WORDS: Ceratocystis fagacearum, Quercus, vectors, fungus, 
forest management. 



Gibbs, J. N., and D. W. French. 

1980. The transmission of oak wilt. U.S. Department of Agriculture 
Forest Service, Research Paper NC-185, 17 p. U.S. Department of 
Agriculture Forest Service, North Central Forest Experiment Sta- 
tion, St. Paul, Minnesota. 

Provides an up-to-date review of factors affecting the transmission 
of oak wilt, Ceratocystis fagacearum. Discusses the history and sever- 
ity of the disease, the saprophytic existence of the fungus in the dying 
tree, seasonal susceptibility of trees to infection, overland and under- 
ground spread, the role of animals and insects as vectors or tree 
wounders, and the distribution of the disease. 

KEY WORDS: Ceratocystis fagacearum, Quercus, vectors, fungus, 
forest management. 




Leave parks and forests clean ... or cleaner. 






w3 / • ■' ' 



UDA FOREST SERVICE 
FISEARCH PAPER NC-186 



GOVT. NT9 



WOV 7 




>fr/p clearcutting to regenerate northern hardwoods 



Frederick T. Metzger 




Mh Central Forest Experiment Station 
Frest Service, U.S. Department of Agriculture 



ACKNOWLEDGMENT 

The author would like to thank the many staff 
members of the Forestry Sciences Laboratory in Mar- 
quette for their contributions in conceiving, imple- 
menting, and interpreting the study and guiding the 
author. Specific thanks are due Thomas Church, Jr., 
John Cooley, Richard Godman, Waino Salminen, 
Toivo Alanen, Robert Oberg, Ralph Peterson, Jr., and 
Gilbert Mattson. 






North Central Forest Experiment Station 

Forest Service - U.S. Department of Agriculture 

1992 Folwell Avenue 

St. Paul, Minnesota 55108 

Manuscript approved for publication May 9, 1979 

1980 



PESTICIDE PRECAUTIONARY 
STATEMENT 

This publication reports research involving pesticides. It does 
not contain recommendations for their use, nor does it imply 
that the uses discussed here have been registered. All uses of 
pesticides must be registered by appropriate State and/or Fed- 
eral agencies before they can be recommended. 

CAUTION: Pesticides can be injurious to humans, domestic 
animals, desirable plants, and fish or other wildlife — if they are 
not handled or applied properly. Use all pesticides selectively 
and carefully. Follow recommended practices for the disposal of 
surplus pesticides and pesticide containers. 




M 

PROTECT 

4^ 




FOLLOW THI LAIIL 



U S llfUDIII Of AGIICUUUK 



STRIP CLEARCUTTING TO 
REGENERATE NORTHERN HARDWOODS 



Frederick T. Metzger, Associate Siluiculturist, 
Marquette, Michigan 



The Great Lakes' northern hardwood forests are a 
collection of various combinations of species occur- 
ring on a broad spectrum of sites, but inevitably 
dominated by sugar maple and/or beech. These for- 
ests have natural resilience to disturbance, and nor- 
mally maintain themselves whether a single tree or 
an entire canopy is lost. These attributes permit use 
of a variety of regeneration techniques, both all-aged 
or even-aged. 

Clearcutting has been the least successful method 
of regenerating northern hardwoods to date (Metzger 
and Tubbs 1971). Yet, a workable method of clear- 
cutting could alleviate the problem of sugar maple 
and beech dominating the reproduction by improving 
conditions for germination and establishment of 
other northern hardwood species. 

It is intuitively appealing to clearcut in strips 
rather than larger blocks, because microclimates of 
strips can be better manipulated to meet regenera- 
tion requirements of individual species. Northern 
hardwood species have better germination and seed- 
ling height growth for up to 5 years under partial 
shade than in the open (Logan 1965, 1966, 1973; 
Godman and Krefting 1960; Marquis 1966; Tubbs 
1969). However, moderately shaded to open condi- 
tions usually provide the greatest increases in root 
and shoot weights (Logan 1965, 1966, 1973). 

Varying strip width and orientation changes the 
timing, duration, and amount of solar radiation 
reaching the ground and consequently the microcli- 
mate. Berry (1964), Marquis (1965), and Brown and 
Merritt (1971) have shown how manipulating strips 
affects shadow patterns. Other factors that can be 
influenced are exposure to prevailing winds, cold air 
drainage, diurnal and seasonal periods of subfreezing 
temperatures, and downward penetration of winds. 
Manipulating strip layout can influence snow accu- 
mulation and melting on level terrain (Clausen and 



Mace 1972). Thus the strip microclimate could be 
modified to provide a more suitable germination 
and growth environment than that of larger block 
clearcuts. 

A number of objectives must be met when regener- 
ating northern hardwoods by strip clearcutting. Ade- 
quate stocking and growth of a desirable species mix 
are basic requirements. Also important is the poten- 
tial for development of quality boles because north- 
ern hardwoods are normally managed for saw- and 
veneer-log production. Future bole quality is influ- 
enced early by both stand environment and repro- 
duction characteristics. Strip clearcutting also has 
advantages in wildlife habitat and watershed man- 
agement, requiring other criteria. 

In 1959 a series of strip cutting trials was begun to 
compare early establishment and development of re- 
production under two orientations, two widths, and 
with or without herbicide treatment of advance re- 
production. 

METHODS 
Stand Conditions 

The two types in the northern hardwood forest 
included in the trials are SAF type 25 (sugar maple- 
beech-yellow birch, referred to hereafter as northern 
hardwood) and SAF 24 (hemlock-yellow birch, re- 
ferred to hereafter as hemlock-hardwood). In all, four 
separate stands were studied: two northern 
hardwood stands and one hemlock-hardwood stand 
on the Upper Peninsula Experimental Forest 
(UPEF), 15 miles southeast of Marquette, Michigan; 
and one hemlock-hardwood stand on the Argonne 
Experimental Forest (AEF), 22 miles northeast of 
Rhinelander, Wisconsin. All four were old-growth 
stands containing large, overmature trees (table 1). 



Table 1. — Density, basal area stocking and volumes of original stands in clearcutting trials on the Upper 
Peninsula and Argonne Experimental Forests (based on all trees 5 inches d.b.h. and larger) 



Type and 




Basal area 








stand 


Density 


stocking 




Volumes 






Trees/acre 


Sq.ft. /acre 


Net MBF/acre 




Cords/acre 


Northern Hardwood: 












UPEF 1 


120 


120 


8.8 




15.6 


UPEF 2 


133 


118 


NA 1 




NA 1 


Hemlock Hardwood: 












UPEF 


177 


179 


11.4 




11.7 


AEF 


208 


161 


6.0 




24.4 



'NA = not available. 



Only light salvage cuttings had been made in the 
several decades preceding the study. 

All of the stands are on relatively level till plains. 
Soils on the UPEF site are usually sandy loams, free 
of rock; they vary from well drained to somewhat 
poorly drained in the northern hardwood stands, and 
from somewhat poorly to poorly drained in the hem- 
lock-hardwood stand. Soils at AEF are silt loams, 
with numerous large stones at or near the surface; 
they range from moderately well drained to some- 
what poorly drained. The better drained UPEF soils 
frequently have fragipans or in some cases bedrock 
within 24 inches of the surface. 

Seedlings were present in all stands prior to cut- 
ting. Their development and stocking were best in 
the northern hardwood stands, where 75 percent of 
the quadrats were stocked and sugar maple domi- 
nated. The AEF hemlock-hardwood stand had the 
most poorly developed seedling layer, dominated by 
sugar maple. The advance reproduction in the UPEF 
hemlock-hardwood stand was a more diverse mixture 
of species with red maple most common, but had the 
poorest stocking (51 percent). 



Silvicultural Treatment 






l: 



j 



UNCUT 



CUT-ONLY 



CUT-SPRAY 



Figure 1. — Generalized layout of clearcut strips in a 
UPEF stand. 



At UPEF, strips were clearcut by commercial oper- 
ators and were 1 or 2 chains wide by 8 chains long 
with the long axis oriented either east-west or north- 
south (fig. 1). Uncut strips of equal width alternated 
with the cut strips. All trees over 4 inches d.b.h. were 
cut and the merchantable material removed. Strips 
were cut in the northern hardwood stands in the 



winter of 1959-60 (stand 1 ) and in the winter of 1960- 
61 (stand 2). Strips in the hemlock-hardwood stand 
were cut the winter of 1964-65. Each winter two 
strips of each orientation of 2-chain width and four 
strips of each orientation of 1-chain width were cut. 






In the AEF hemlock-hardwood stand, all strips 
were oriented east-west to parallel an 8-percent slope 
of west aspect. Because most strips exceeded 8 chains 
in length, the study was confined to the central 8- 
chain portion. Saplings were cut in addition to the 
larger trees. During each of the winters of 1964-65 
and 1965-66 two 2-chain and four 1-chain-wide strips 
were cut. 

Herbicides were used to eliminate advance repro- 
duction on half of the strips in each combination of 
widths and orientations. They were sprayed the sum- 
mer following logging. Northern hardwood stands 
were treated with 2,4,5-T in a water-oil carrier at 8 
and 4 lbs. of herbicide in 100 gals, of carrier in stands 
1 and 2, respectively. Tordon 101 (Picloram and 2,4-D) 
was used in the hemlock-hardwood stands at a 
strength of 2.5 lbs. per 100 gals, of water. Herbicides 
were applied until they dripped from the foliage. 

Stumps in the UPEF sprayed strips were also 
treated separately with a stronger solution of herbi- 
cide in an oil carrier. Northern hardwood stand 1 was 
treated with 2,4,5-T at 20 lbs. per 100 gals., stand 2 
with a mixture of 2,4,5-T and 2,4-D at 30 lbs. per 100 
gals., and the hemlock-hardwoods with 2,4,5-T at 16 
lbs. per 100 gals. 



Survey Procedures 

Reproduction on uncut and cut-only strips was first 
measured one or two seasons after cutting; a second 
measurement was made six or seven seasons after 
cutting. Reproduction on cut-sprayed strips was mea- 
sured the first or second season after spraying and 
again in the fifth or sixth season. Seedlings were 
recorded by species, height class, density class, and 
origin (seedling or sprout); the species and height 
class of the tallest or dominant stem was also deter- 
mined. Quarter-milacre quadrats spaced at 10-link 
intervals on transects crossing the strip perpendicu- 
lar to the long axis were used. Between five and nine 
transects per strip were run. The second survey of 
hemlock-hardwood strips used actual counts of repro- 
duction instead of density classes and assessed the 
dominant stem on milacre and ninetieth-acre qua- 
drats. Each milacre quadrat was centered over a 
quarter-milacre quadrat. Three ninetieth-acre qua- 
drats were installed per chain of transect. The large 
quadrats sampled reproduction that had the best op- 
portunity of becoming one of approximately 90 final 
crop trees per acre. 



RESULTS 
Stand Establishment 

Cut-only strips were generally well stocked with 
reproduction 6 or 7 years after cutting. In the three 
UPEF stands, each cut-only strip had more than 
17,000 stems per acre (table 2), and 73 percent or 
more of the quarter-milacre quadrats contained at 
least one tree (table 3). The cut-only AEF hemlock- 
hardwood strips were not as well stocked. They 
averaged 9,000 seedlings per acre and 60-percent 
stocking of the quadrats. 

Reproduction on cut-sprayed strips was much 
more variable. Best results were obtained in UPEF 
northern hardwood stand 1 and in the UPEF hem- 
lock-hardwood stand. Each had in excess of 25,000 
seedlings per acre and 66 percent stocking of repro- 
duction on quadrats 6 years after spraying (tables 2 
and 3). Northern hardwood stand 2 at UPEF aver- 
aged only 9,000 stems per acre and 63 percent stock- 
ing. The herbicide-treated AEF strips had the poorest 
reproduction, averaging only 4,000 stems per acre 
and 30 percent stocking. 

Neither strip width nor orientation consistently 
led to increased reproduction (tables 2 and 3). Much 
of the variation within stands was due to factors 
unrelated to strip layout, such as soil moisture, origi- 
nal overstory, and advance reproduction. 

Desirable hardwoods — sugar maple, red maple, 
yellow birch, beech, white ash, and basswood — of 
seedling or seedling-sprout origin occurred on an av- 
erage of 95 percent of the quarter-milacre quadrats 
that were stocked with any species or form of tree 
reproduction at UPEF. Again, the cut-sprayed strips 
of northern hardwood stand 2 were the exception; the 
proportion of stocked quadrats with a desirable 
hardwood declined to an average of 72 percent on the 
2-chain-wide strips. Somewhat poorer results also 
occurred at AEF, where 87 and 57 percent of the 
stocked quadrats had desirable hardwoods on the cut- 
only and cut-sprayed strips, respectively. 

Effects of advance reproduction 

Regeneration success on cut-only strips depended 
on the presence of large advance reproduction. Stock- 
ing percentages 6 to 7 years after cutting were 
positively correlated (r 2 = .70) with the percent of 
quadrats stocked with advance reproduction over 4 
feet tall at the time of cutting. All successfully regen- 
erated cut-only strips originally had more than 9 
percent of their quadrats stocked with reproduction 
over 4 feet tall. 



Table 2. — Number of seedlings per acre, by cutting treatment and strip width and orientation 1 , 5 to 7 years after 
strip clearcutting on the Upper Peninsula and Argonne Experimental Forests 

(In thousands of seedlings per acre) 



Type and 
stand 



Cut-only 



Cut-sprayed 



Uncut 



1-NS 1-EW 2-NS 2-EW Mean 1-NS 1-EW 2-NS 2-EW Mean 1-NS 1-EW 2-NS 2-EW Mean 



Northern Hardwood: 2 

UPEF 1 

UPEF 2 
Hemlock-Hardwood: 3 

UPEF 

AEF 



26.3 28.0 28.3 27.9 27.6 40.8 29.1 35.5 25.3 32.7 29.1 24.2 21.4 17.2 23.0 

25.9 17.5 30.1 35.5 27.3 8.7 11.2 7.8 9.8 9.4 21.7 20.7 28.6 22.9 23.5 

26.1 27.7 34.5 22.3 27.7 31.5 30.9 50.9 32.0 36.3 16.5 19.4 13.3 19.6 17.2 

9.8 — 8.4 9.1 — 4.7 — 3.4 4.0 — 26.2 — 19.6 21.5 



'1-NS = 1 chain wide, oriented north-south, etc. 

Estimates are "minimum populations" because counts tor each species, size-class terminated at five on each quadrat. 

Estimates are based on counts of all seedlings on quadrats. 



Table 3. — Quarter -milacre quadrats stocked with one or more seedlings by cutting treatment and strip width and 
orientation 1 , 5 to 7 years after strip clearcutting on the Upper Peninsula and Argonne Experimental Forests 

(In percent) 



Type and 






Cut-only 








Cut-sprayed 








Uncut 




stand 


1-NS 


1-EW 


2-NS 


2-EW 


Mean 


1-NS 


1-EW 


2-NS 


2-EW 


Mean 


1-NS 


1-EW 


2-NS 


2-EW Mean 


Northern Hardwood: 






























UPEF 1 


89 


91 


95 


99 


93 


99 


91 


96 


89 


94 


96 


95 


91 


85 92 


UPEF 2 


79 


73 


89 


91 


83 


62 


59 


63 


66 


62 


86 


78 


90 


79 83 


Hemlock-Hardwood: 






























UPEF 


81 


81 


91 


77 


82 


66 


87 


83 


88 


81 


71 


62 


56 


76 66 


AEF 


— 


59 


— 


60 


60 


— 


30 


— 


30 


30 


— 


88 


— 


78 81 



'1-NS = 1 chain wide oriented north-south, etc. 



Other related parameters, such as desirable 
hardwood stocking 6 to 7 years after cutting, stocking 
of advance reproduction over 2 feet tall, and numbers 
of advance reproduction instead of stocking percent- 
ages, yielded significant correlations but reduced co- 
efficients of determination. The initial stocking of all 
sizes of advance reproduction was poorly correlated 
with stocking 6 to 7 years later (r 2 = .10). 



Stump sprouting 

Stump sprouts were limited to the cut-only strips 
because spraying effectively minimized them. Most 
species sprouted to some degree, but only red maple 
did so prolifically. Red maple sprouts were not well 
distributed, yet due to their rapid growth and large 
number per stump they occupied much growing space 



and readily overtopped other reproduction. They 
were most abundant on UPEF hemlock-hardwood 
cut-only strips, where they made up 6 percent of the 
dominants on milacre quadrats and 21 percent of the 
dominants on ninetieth-acre quadrats. 

Time of establishment 

Seedling age data showed most yellow birch and 
red maple seedlings originated after spraying in the 
UPEF hemlock-hardwood stand (93 percent of the 
birch and 85 percent of the maple). Establishment 
continued over 4 years following spraying. In the first 
season a third of both species became established. 
Birch establishment peaked the second year with 
half the birch seedlings becoming established then. 
Twenty percent of the maple seedlings became estab- 
lished in each of the next 2 years. 






Effects on quality 

The degree of crown competition within northern 
hardwood stands affects development of stem qual- 
ity. Thus, two factors affecting crown competition 
were considered: variation in height and distance to 
nearest competitor (tree at least two-thirds as tall) 
for the dominant tree on the ninetieth-acre quadrat. 
Cutting only trees larger than 4 inches d.b.h. in the 
hemlock-hardwoods at UPEF resulted in more varia- 
tion in height among dominants than at AEF, where 
trees over 2 inches d.b.h. were cut (table 4). Spraying 
at UPEF did not reduce this variability because a 
number of conifer and larger hardwood saplings were 
unaffected by the herbicide. Spraying at AEF, how- 
ever, created more variation in height than occurred 
on the cut-only strips. The majority of dominants on 
cut-only strips had a competitor within a distance not 
exceeding the dominant's crown diameter. On cut- 
sprayed strips, half or fewer of the dominants had a 
competitor within this distance. 



Stand Development 

Reproduction developed best in the cut-only strips 
in both stands and in the cut-sprayed strips in stand 1 
in the northern hardwoods at UPEF. There were 
substantial increases in quadrats stocked with repro- 
duction over 4 feet tall through the first 6 to 7 years 
after cutting ( table 5 ). The hemlock-hardwood stands 
did not develop as well, particularly at AEF. All the 
UPEF hemlock-hardwood strips had the potential to 
improve, however, because many quadrats had 
gained seedlings between 2 and 4 feet tall. The AEF 
strips had the poorest prospects for improvement. 



Poor development of reproduction as a stand at 
AEF and on cut-sprayed strips in UPEF stand 2 was 
not due to poor growth rates of the surviving trees, 
but due to loss of existing stocking and poor estab- 
lishment of new reproduction. Growth rates on these 
units (0.71 feet/year) were comparable to the overall 
average (0.73 feet/year). Height growth was calcu- 
lated as the difference between the two surveys in the 
weighted averages of class midpoints for the tallest 
seedling per quadrat. 

Neither strip width nor orientation affected repro- 
duction growth rates (table 6). Differences were usu- 
ally caused by other factors, often the proportion of 
fast-growing pioneer species present. 

Additional support for the conclusion that strip 
width did not affect height growth came from analyz- 
ing reproduction height across the strips for individ- 
ual species. Heights did not improve with distance 
from the timbered edge, although the opposite im- 
pression was apparent in the field, since the taller 
pioneer species — quaking aspen, paper birch, and 
pin cherry — were more abundant in the central por- 
tion of the strip. 



Species Composition 

Cut-only strips 

Strip clearcutting in the northern hardwood type 
had little effect on species composition in the regen- 
erated stand. Composition on cut strips was very 
similar to that on adjacent uncut strips (table 7). 
Sugar maple comprised most of this reproduction and 
its preponderance reduced species diversity of the 
reproduction below that of the original overstory. 



Table 4. — Characteristics of dominant trees on ninetieth -acre quadrats in hemlock -hardwood clearcut strips at 
the Upper Peninsula and Argonne Experimental Forests, 5 to 6 years after treatment 



Stand 




Heiqht 






Dominants 
with 


Mean 
distance to 


Mean 


and 






Coefficient 


crown diameter 


treatment 


Mean 


Range 


of variation 




competitor 1 


competitor 


of dominant 






Feet 




Percent 






- Feet 


UPEF: 
















Cut-only 


18.6 


4-67 


59 




63 


6.7 


7.2 


Cut-sprayed 


11.4 


4-46 


58 




50 


6.1 


4.3 


AEF: 
















Cut-only 


10.2 


4-15 


24 




75 


3.7 


4.3 


Cut-sprayed 


7.2 


3-17 


36 




42 


5.7 


2.5 



Percent of quadrats where dominant tree has a competitor (tree at least % height of dominant) not more than dominant crown diameter away 



Table 5. — Height of tallest reproduction stem on quarter milacre quadrats by cutting method and year after 
clearcutting or spraying on the Upper Peninsula and Argonne Experimental Forests 

(In percent of quadrats stocked) 









Cut-only 






Cut-sprayed 






Nonstocked 






Nonstocked 






Type and 




& 






& 








stand 


Year 


0-2' 


2-4' 


4' + 


0-2 




2-4' 


4' + 


Northern hardwood: 


















UPEF 1 


1-2 


64 


22 


14 


96 




2 


3 




6-7 


15 


16 


69 


15 




29 


55 


UPEF 2 


1-2 


66 


25 


9 


96 




2 


2 




5-6 


25 


17 


58 


65 




21 


15 


Hemlock-hardwood: 


















UPEF 


2 


77 


14 


10 


98 




2 


<1 




6 


32 


30 


38 


30 




45 


25 


AEF 


1-2 


91 


8 


1 


98 




1 


1 




5-6 


50 


16 


34 


76 




12 


12 



Table 6. — Average annual height growth 1 of dominant reproduction in the 4- to 5 -year period following treatment 
by strip width and orientation 2 on the Upper Peninsula and Argonne Experimental Forests 

(In feet per year) 



Type 


Cut-only 






Cut-sprayed 






and 


Means 




All 


Means 






stand 


1NS 1EW 2NS 2EW 1 2 NS 


EW 


1NS 1EW 2NS 2EW 1 2 NS 


EW 


All 



Northern 
hardwoods: 
UPEF 1 0.67 0.79 0.66 0.69 0.73 0.68 0.66 0.74 0.70 0.88 0.73 0.93 0.60 0.81 0.77 0.90 0.67 0.79 
.75 .84 .94 .86 .79 .90 .85 .85 .85 .44 .53 .74 .74 .48 .74 .59 .64 .61 



UPEF 2 
Hemlock- 
hardwoods: 
UPEF 
AEF 



.42 .66 .55 .62 .54 .58 .49 .64 .56 .65 .79 .80 .86 .73 .84 .74 .83 .78 
— .74 — .74 — — — — .74 — 1.09 — .47 — — — — .79 



'See text for method of derivation 

2 Strip width: 1 = 1 chain wide, 2 = 2 chains wide; orientation: NS = north-south, EW = east-west. 



Table 7. — Comparison between species composition of the original stand and of the reproduction 5 to 7 years after 
strip clearcutting on the Upper Peninsula and Argonne Experimental Forests 

NORTHERN HARDWOOD-UPEF Stand 1 





Original stand: 




Regenerated stand: s 


6 inches tall 






3= 5 inches d.b.h. 


Percent of numbe 


r 


Percent of 




Percent of 
basal area 


Percent of 


of seedlings 




dominant stems 


Species 


numbers Leave 


Cut-only 1 


Jut-spray 


Cut-only 


Cut-spray 


Sugar maple 


50 


43 74 


66 


60 


51 


44 


Red maple 


3 


5 7 


10 


3 


13 


4 


Yellow birch 


14 


10 6 


16 


33 


17 


47 


Other 














hardwoods 1 


25 


35 13 


8 


3 


18 


5 


Conifers 


8 


7 1 


1 


1 


1 





NORTHERN HARDWOOD-UPEF Stand 2 


Sugar maple 


63 


61 86 


88 


64 


76 


57 


Red maple 


1 


( 2 ) 1 


1 


2 


1 


2 


Yellow birch 


20 


13 5 


6 


12 


7 


9 


Other 














hardwoods 1 


13 


19 8 


6 


22 


15 


32 


Conifers 


3 


7 1 


1 


1 


1 


1 


HEMLOCK HARDWOOD-UPEF 


Sugar maple 


3 


4 10 


10 


1 


9 


1 


Red maple 


22 


28 37 


52 


6 


51 


15 


Yellow birch 


22 


24 45 


31 


84 


21 


39 


Other desirable 














hardwoods 


1 


2 1 


1 


1 


2 


1 


Pioneer 














hardwoods 





1 


1 


9 


4 


41 


Conifers 


53 


42 6 


5 


1 


14 


3 


HEMLOCK HARDWOOD-AEF 


Sugar maple 


16 


14 78 


56 


13 


37 


18 


Red maple 


5 


5 9 


11 


5 


19 


9 


Yellow birch 


21 


14 9 


12 


28 


8 


13 


Other desirable 














hardwoods 


6 


7 1 


3 


1 


3 


1 


Pioneer 














hardwoods 


1 


( 2 ) 1 


13 


53 


32 


57 


Conifers 


52 


60 1 


4 


1 


1 


2 



'Includes pioneer hardwood species, other desirable hardwood species, and ironwood. 
2 Less than 1 percent. 



Sugar maple's importance in the new stand may de- 
cline, allowing diversity to increase over time if the 
composition of the new stand shifts toward that of 
seedlings dominating the quadrats. Even so, the new 
stand will probably not reach the diversity of the 
parent stand. Offsetting the decline of sugar maple 
would be one or more species of the group of other 
hardwoods (beech, ironwood, quaking aspen, black 
ash, basswood, and American elm), which as a group 
make up a larger proportion of dominants than of all 
reproduction. Conifers made up a minor proportion of 
the new stand and probably will continue to do so. 
Red maple and yellow birch reproduction was incon- 
sistent; these species were more abundant than in the 
original overstory and the uncut strip reproduction 
in stand 1, but were about the same or declining in 
relation to these populations in stand 2. The net effect 
was that no important changes occurred in these two 
species due to strip cutting. 

The hemlock-hardwood stands at both UPEF and 
AEF underwent considerable change in character as 
hemlock-dominated overstories were replaced by 
hardwood-dominated reproduction. The change was 
not solely due to cutting, since the uncut and cut-only 
strips were more comparable to each other than to the 
original overstory (table 7). These stands would have 
probably reverted to hardwoods naturally, through 
attrition of overmature conifers over time and poor 
regeneration of hemlock. 

Sugar maple became the most common reproduc- 
tion species on both uncut and cut-only strips in the 
AEF hemlock-hardwood stands. However, cutting 
reduced the proportion of sugar maple compared to 
the uncut strips, while increasing the proportion of 
pioneer species (quaking aspen, paper birch, and pin 
cherry). This trend may continue, since the amounts 
of sugar maple and pioneer hardwood dominants 
were almost equal. 

On the UPEF hemlock-hardwood cut strips, red 
maple became the most common species and should 
remain so. It benefitted from cutting, as did the pio- 
neer species, while most other species remained un- 
changed or declined in relative abundance. These 
strips had the only significant amount of hemlock 
advance reproduction. Because it was well developed, 
it was often dominant. 



Cut-sprayed strips 

Spraying herbicides in combination with cutting 
was only partially effective in creating desired 
changes in species composition of the reproduction. 



Sugar maple, although relatively less abundant on 
cut-sprayed than on cut or uncut strips in all four 
stands, still comprised about the same proportion 
of the new stand as it did in the original overstory 
(table 7). 

The proportion of yellow birch increased after 
spraying in all four stands, but real increases in 
numbers and stocking were attained in only two. In 
northern hardwood stand 1, yellow birch became the 
leading dominant, with over 10,000 seedlings per 
acre. It was also the most abundant species in the 
UPEF hemlock-hardwood stand, where its 30,000 
seedlings per acre made up 84 percent of the repro- 
duction, but only 39 percent of the dominants. Higher 
proportions of yellow birch in the other two stands 
were caused by fewer stems of other species rather 
than by a gain in birch numbers. In these two stands, 
yellow birch numbers and stocking percentages were 
similar on cut-sprayed, cut-only, and uncut strips. 

Establishment of pioneer hardwood species 
(mostly quaking aspen, but including pin cherry and 
paper birch) was highly variable following spraying. 
They were a minor component of the reproduction in 
only the UPEF northern hardwood stand 1. In the 
remaining stands they ranked either first or second 
in abundance and proportion of dominants (table 7). 
In both hemlock-hardwood stands they were the most 
common dominants. 

Red maple was the only other species present in 
significant amounts in the cut-sprayed strips. It was 
relatively less abundant than in the other strips, but 
it remained at levels comparable with its position in 
the original overstories. 

Effect of strip design 

Changing strip width and orientation had little 
effect on the performance of desired species. Sugar 
maple was the only desired hardwood species to re- 
spond to changes in strip layout, but this occurred 
only in the cut-only strips of one stand. In this case 
sugar maple stocking was 12 percent greater on 1- 
chain than 2-chain strips and 9 percent greater on 
north-south than east-west strips. The pioneer spe- 
cies generally exhibited the opposite behavior, being 
better stocked on the more exposed cut-sprayed strips 
and AEF cut-only strips. Overall, pioneer hardwoods 
in the 2-chain strips averaged 12 percent and the 
east-west strips 5 percent better stocking than in the 
1-chain and north-south strips, respectively. These 
trends would probably have been better expressed if 
other factors affecting seedling establishment had 
been more uniform throughout each stand. 



8 



Effects of parent stand and site 

Species composition of reproduction was related to 
the composition of the original overstory and site 
factors, especially on the cut-only strips. The abun- 
dance of either sugar or red maple in the reproduction 
was positively correlated to its abundance in the 
overstory, but often negatively correlated to the 
abundance of the opposite species (table 8). Abun- 
dance of beech or hemlock in the overstory and red 
maple in the reproduction were positively correlated, 
while the opposite was often true for sugar maple in 
the reproduction. Yellow birch reproduction re- 
sponded similarly to red maple reproduction, being 



positively associated with overstory red maple but 
negatively with sugar maple. Abundance is the same 
as the ecologists' importance value for a species; it 
represents the mean of relative density and relative 
stocking for a reproduction species and the mean of 
relative density and relative basal area for each over- 
story species. 

There were fewer significant relations between 
reproduction and original overstory species in the 
cut-sprayed than in the cut-only strips (table 8). In 
northern hardwood stands, yellow birch reproduction 
was associated with overstory red maple, beech, and 
hemlock, but along with quaking aspen and pin 



Table 8. — Correlation coefficients for occurrence of species in reproduction and in overstory by strips, or segments 

of strips (see text for method of computing coefficients) 

NORTHERN HARDWOOD-UPEF STANDS 1 & 2 









Cut-only strips 




Cut-sprayed strips 




Reproduction 










Overstory species 










species 


Sugar 


Red 


Yellow 




Sugar 


Red 


Yellow 








maple 


maple 


birch 


Hemlock 


Beech maple 


maple 


birch 


Hemlock 


Beech 


Sugar 




















maple 


71 1 


-89 2 


38 


-91 2 


-77 1 32 


-61 


33 


-53 


-40 


Red 




















maple 


-77 1 


96 2 


-24 


91 2 


74 1 -46 


24 


-16 


16 


77 1 


Yellow 




















birch 


-53 


76 1 


-28 


70 


65 -79 1 


87 2 


-29 


88 2 


86 2 


HEMLOCK HARDWOOD-UPEF 


Sugar 




















maple 


87 1 


-37 


72 


-64 


87 1 


-14 


39 


-29 




Red 




















maple 


-31 


90 2 


18 


-29 


85 1 


-31 


32 


-17 




Yellow 




















birch 


-41 


-43 


-46 


57 


47 


37 


52 


-51 




Quaking 




















aspen 


-52 


-32 


-38 


60 


-85 2 


-6 


^16 


38 




HEMLOCK HARDWOOD-AEF 


Sugar 




















maple 


78 2 


-4V 


-4 


-58 2 


27 


5 


-12 


-21 




Red 




















maple 


-41 1 


52 2 


19 


20 


-1 


8 


27 


14 




Yellow 




















birch 


-55 2 


34 





40 


13 


-4 


-5 


-2 




Quaking 




















aspen 


-25 


-18 


-10 


33 


-6 


-17 


-6 


-2 




Pin 




















cherry 


-55 2 


13 


-2 


44 1 


-51 1 


11 


8 


51 1 





'Correlation coefficient of 95 percent acceptance. 
Correlation coefficient of 99 percent acceptance. 



cherry it was negatively related to overstory sugar 
maple. 

Soil moisture gradients within the stips also af- 
fected species composition. Sugar maple was most 
abundant on mesic sites, and red maple on somewhat 
poorly drained sites; yellow birch was slightly more 
common on somewhat poorly drained than on better 
drained sites. 

Lesser vegetation 

A dense layer of shrubs, herbs, and grasses fol- 
lowed cutting and cutting-spraying, initially domi- 
nating reproduction on many hemlock-hardwood 
strips (fig. 2). A similar vegetation survey was not 
made in northern hardwood stands. By the fifth to the 
seventh year, reproduction dominated the lesser veg- 
etation in most stands, yet this lesser vegetation 
remained an important component of the total plant 
cover. The cut-sprayed strips at AEF and UPEF 
stand 2 continued to be dominated by lesser vegeta- 
tion on the majority of quadrats even after 5 years. 

Raspberry was the most common shrub on mesic 
sites. Wetter sites had a mixture of shrubs, usually 
with mountain maple an important component. 



100 



90 



80 



111 
O 

QC 
LU 

Ul 

^ 70 

:d 
5 

O 

Q 

£ 40 

O 

o 



60 



50 



C/5 

K 

QC 
Q 
<* 

O 



30 

20 

10 


YEAR 



n 



_ 






: 



CUT 



6 
CUT 
SPRAY 



6 
CUT 



STAND 1 



5 
CUT 
SPRAY 

STAND 2 



NORTHERN HARDWOODS 



Other species of shrubs and small trees occurring in 
the strips were pin cherry, red-berried elder, black- 
berry, beaked hazel, and willow. Sedge-grass com- 
munities proliferated after spraying on wetter sites 
and on many of the AEF strips. 

DISCUSSION 

These trials demonstrated the great potential of 
strip clearcutting for re-establishing well-stocked 
stands of high-value hardwoods. Simultaneously, the 
poor results obtained served as a reminder that mis- 
application of the method can result in failure 
(Metzger and Tubbs 1971). Obviously, strip clear- 
cutting must be very carefully prescribed and used. 
The inconsistent results of these trials provide some 
leads toward improving our understanding of where 
and under what conditions the method is likely to 
succeed. 

Most strips in these trials had the potential to 
produce a fiber crop, in that the numbers and distri- 
bution of stems within the strips appear adequate to 
assure utilization of the site. Stocking on several of 
the cut-sprayed strips, however, was low, with much 






r 



I 









^ 



I 



- 



§ 



ra 



~ 






2 6 
CUT 



2 6 
CUT 
SPRAY 

UPEF 



1-2 6 
CUT 



1 5 
CUT 
SPRAY 



KEY 



AEF 



HEMLOCK HARDWOODS 



Figure 2. — Form of vegetation dominating quadrats on clearcut strips on the 
Upper Peninsula and Argonne Experimental Forest. 



10 



reproduction subordinate to lesser vegetation. In 
these cases, the well-established shrub and grass- 
sedge communities appeared likely to continue domi- 
nation of sizeable areas for some time (Levy 1970, 
Metzger and Tubbs 1971). 

The good distribution of desirable hardwoods in 
competitve positions on all strips at UPEF except the 
cut-sprayed strips in stand 2 create the potential for 
producing saw and veneer logs as well as fiber. The 
outlook for developing high-quality boles in these 
stands is not as readily apparent, but depends on 
maintaining high levels of crown competition to re- 
duce branch and fork caused bole defects (Godman 
and Books 1971). Achieving both vertical and hori- 
zontal competition among crowns requires well- 
spaced reproduction of uniform height in recently 
reproduced stands. Cutting to a 5-inch d.b.h. limit at 
UPEF left many potential wolf trees and considera- 
ble variation in height, even after spraying. Had the 
stands been weeded, residual density and distribu- 
tion should have resulted in crown closure, thus pro- 
moting higher stem quality. However, until research 
provides minimum stocking standards in relation to 
potential stem quality, the impact of stocking on 
quality can only be speculated upon. 

Poorer stocking and competitive positions of desir- 
able hardwoods in the AEF cut-only strips and UPEF 
cut-sprayed strips of stand 2 make the prospects of 
obtaining high-quality yields very marginal. Fur- 
thermore, these stands' crown structure, needed to 
promote stem quality, is not anticipated to develop 
for some time. Cut-sprayed strips at AEF are so se- 
verely understocked that there is no question of their 
failure. 

One frequently cited advantage of clearcutting is 
its potential to increase species diversity of regener- 
ated stands. On the contrary, these clearcutting 
trials without supplemental treatment reduced di- 
versity. Clearcutting released reproduction estab- 
lished before cutting, whose superior competitive 
position prevented establishment of any significant 
amounts of new reproduction that could have in- 
creased diversity. Sugar maple was usually the most 
abundant species of advance reproduction, but beech, 
red maple, and hemlock were important in a few 
situations. The use of herbicides also failed to in- 
crease species diversity, although the advance regen- 
eration was effectively eliminated or set back. Single 
species tended to dominate the regenerated stand, 
although the species varied from stand to stand de- 
pending on seed availability and site conditions. 



This study did not identify an optimum width or 
orientation for strips, since little of the variation in 
results could be associated with strip layout. Ringger 
and Stearns' ( 1972) microclimatic data for AEF open- 
ings suggest why stocking differences for many 
species are slight between 1- and 2-chain strips. A 
number of microclimatic parameters remain rela- 
tively constant until opening diameters exceed twice 
the height of the border trees, then change abruptly. 
In these trials, the ratio of strip width to border tree 
height was less than 2. Larger openings increase the 
risk of lower temperatures for longer periods and 
increase the drought stress brought about by a com- 
bination of longer durations of high temperatures, 
higher wind speeds, and increased direct solar radia- 
tion. There is considerable uncertainty about the suc- 
cessful establishment of key species in wider strips, 
and added trials are warranted. 

When advance reproduction was released by strip 
clearcutting, its success in the new stand depended 
on its ability to withstand major changes in light, 
heat, moisture, and competition. If the reproduction 
can withstand release, the size of area released ap- 
parently has little effect. Wider cut-only strips might 
be successful, because composition in these trials re- 
sembled that in larger clearcuts in Canada (Winget 
1968, Boivin 1971) and in northeastern United 
States (Nyland and Irish 1971, Richards and Farns- 
worth 1971). 

Reproduction height proved to be a good measure of 
its capability of withstanding overstory removal in 
the strips as well as in shelterwoods (Jacobs 1974). 
Site and climatic conditions also affect the outcome, 
so no single minimum stocking standard for wide- 
spread use is advisable. This study's results would 
apply to sites with low plant moisture stress similar 
to UPEF; a minimum of 15-percent stocking (a 50- 
percent safety factor has been added) of advance re- 
production over 4 feet tall is recommended in such 
cases. In situations with relatively high potential 
moisture stress, Jacobs' (1974) recommendations de- 
veloped for shelterwood release at AEF would be 
more appropriate (5,000 well distributed 2- to 4-foot- 
tall seedlings per acre). When these levels are not 
attainable, an alternative cutting method should be 
used; if an even-aged stand is desired, a shelterwood 
to promote further development of advance reproduc- 
tion before final release is necessary (Godman and 
Tubbs 1973). 

Species composition of the overstory and soil drain- 
age had a greater influence on composition of repro- 
duction on cut-only strips than did cutting or strip 



11 



design. Sugar maple reproduction was most aggres- 
sive under its own canopy and on moderately well- 
drained to well-drained sites. On sites with less than 
well-drained soils, where red maple, yellow birch or 
hemlock were abundant in the overstory, reproduc- 
tion of red maple and yellow birch increased in rela- 
tive abundance or became the dominant species. 
These results are generally contrary to the hypothe- 
ses of Fox (1977) and Forcier (1975) that different 
species tend to replace overstory species in climax 
forests. 

Successful regeneration from seed on the herbi- 
cide-treated cut strips depended on an adequate seed 
supply, elimination of advance reproduction, and a 
favorable environment. The importance of seed sup- 
ply was indicated by the successful establishment of 
reproduction following bumper crops of sugar maple 
and yellow birch seed. Conversely, reproduction es- 
tablishment declined drastically after crop failure or 
poor crops. It is also important that the seed is availa- 
ble the first season after spraying. Bumper crops of 
yellow birch occurred 2 and 3 years after spraying, 
but did not contribute measurably to the new stand. 
Rapid develpment of competing vegetation probably 
limited the chances of later seedling establishment. 

Herbicide spraying proved to be especially benefi- 
cial to the establishment of yellow birch in two stands 
by successfully eliminating the advance reproduc- 
tion. Sugar maple rebounded after spraying in the 
northern hardwood stands and became an important 
component of the reproduction again. Also benefitt- 
ing from spraying were the pioneer species — quaking 
aspen, paper birch, and pin cherry. The pioneer spe- 
cies' abundance on the strips was probably less than 
what would be more commonly expected. This was 
partially due to the lack of pioneer species in the 
original stand (except for one paper birch per acre on 
the AEF), which meant there was no vegetative re- 
production or seed available from adjacent uncut 
strips. There also would have been little seed availa- 
ble in the litter layer, because these stands had been 
undisturbed for 200 or more years (Graber and 
Thompson 1978). The source of seed was, therefore, in 
areas away from the strips, and in 2 of the 3 years of 
establishment at AEF, seed crops of quaking aspen 
and paper birch were failures (Godman and Mattson 
1976). Pioneer species' increased abundance on the 
wider or more open strips suggests that these species 
would be favored by even larger strips. 

Establishment of many other desirable species was 
hampered by too few trees, poor seed crops, or low 
seed mobility. For some, additional silvicultural 



practices may be needed. Basswood at AEF is an 
excellent example. Its reproduction was lacking de- 
spite good stocking in the overstory and plentiful 
seed. Supplemental treatments may be needed to 
overcome seed dormancy and to provide protection 
from decay and rodent predation (Godman and Matt- 
son 1976). 

Adequate dispersal of seed from the uncut border 
was no problem on the 1- and 2-chain-wide strips 
used. Benzie ( 1959) found high numbers of seeds from 
sugar maple and yellow birch up to 5 chains from 
their sources. 

The generally poorer regeneration at AEF sug- 
gests the environmental differences between AEF 
and UPEF are important and should be considered in 
any future applications of strip clearcutting. A com- 
bination of factors leads to increased moisture stress 
at AEF. The climate at AEF is more continental than 
at UPEF, which is in close proximity to Lake Supe- 
rior. Also contributing to greater stress at AEF are a 
boulder layer near the soil surface that reduces mois- 
ture-holding capacity, and the westerly aspect of the 
strips which may have increased evapotranspiration 
at the site. The other extreme, excessive soil mois- 
ture, caused reproduction failures of certain small 
areas at UPEF. Areas of high water tables through- 
out the growing season regenerated poorly. 

Successful strip clearcutting trials at UPEF were 
comparable in many respects to shelterwood (Tubbs 
and Metzger 1969) and seed tree (Godman and Kreft- 
ing 1960) trials also conducted there. These trials all 
became fully stocked after cutting and sugar maple 
stocking did not vary greatly among the different 
cutting methods. Yellow birch establishment on the 
successfully regenerated sprayed strips exceeded the 
results obtained in shelterwood trials with either 
seedbed scarification or seedbed scarification plus 
herbicide treatment (Tubbs and Metzger 1969). 

Results of strip clearcutting in the hemlock- 
hardwood stand at AEF were similar to those from an 
earlier study of strip and block clearcutting in imma- 
ture northern hardwood stands there (Metzger and 
Tubbs 1971). Reproduction in both trials was mar- 
ginal to unacceptable. Stocking often declined when 
the stand was heavily disturbed — that is, when 
larger areas were cut, when cutting diameter limits 
were lowered, or when strips were sprayed. Desired 
shifts in species composition were not obtained. 

Browse yields from shrubs and reproduction or 
clearcut strips at AEF reached 500 pounds per acre 
years after cutting (Stearns 1969). Other benefits tc 
wildlife from strip cutting are the interspersion 01 



i 



' 



12 



various stand structures, creation of edges and the 
increase in important food and cover species absent 
in mature northern hardwood stands. 



SUMMARY 

Strip clearcutting with or without supplemental 
treatments is a workable silvicultural option for 
even-aged management of northern hardwoods. It is 
not a technique that can be applied indiscriminately, 
because an error in application can result in long- 
term loss or reduction of tree cover and development 
of a community of lesser vegetation. Each situation 
requires careful evaluation, and strip clearcutting 
should be used only when it is the logical means to 
achieve management objectives. 

Strip clearcutting to release advance reproduction 
should only be done when the stocking of reproduc- 
tion capable of withstanding exposure is adequate. 
Guidelines for stocking should be more conservative 
on sites subject to greater moisture stress. This 
method does not allow significant manipulation of 
species composition in the regenerated stand. 

Strip clearcutting combined with herbicide treat- 
ment should be even more judiciously prescribed and 
is subject to more constraints. Our experience re- 
vealed that treatments must effectively eliminate 
the advance reproduction and seed must be available. 
Experience shows that at least good or better seed 
crops of yellow birch, sugar maple, and red maple are 
needed to restock clearcut areas. Good crops of other 
species, except the pioneer hardwoods, did not suc- 
cessfully establish them. Soil moisture availability 
at the site influences regeneration results but our 
limited data do not warrant specific recommenda- 
tions regarding soil moisture and stocking levels. 



LITERATURE CITED 

Benzie, John W. 1959. Sugar maple and yellow birch 
seed dispersal from a fully stocked stand of mature 
northern hardwoods in the Upper Peninsula of 
Michigan. U.S. Department of Agriculture Forest 
Service, Technical Note 561, 1 p. U.S. Department 
of Agriculture Forest Service, Lake States Forest 
Experiment Station, St. Paul, Minnesota. 

Berry, A. B. 1964. Effect of strip width on proportion 
of daily light reaching the ground. The Forestry 
Chronicles 40:130-131. 



Boivin, Jean-Louis. 1971. Etude de la regeneration 
apres coupe rase dans des peuplements feuillus et 
melanges de l'ouest quebecois. The Forestry 
Chronicles 47:82-85. 

Brown, K. M., and C. Merritt. 1971. Simulated sun- 
light duration maps of forest openings. Indiana 
Academy of Science Proceedings 1970:80:220-224. 

Clausen, John C, and Arnett C. Mace, Jr. 1972. Ac- 
cumulation and snow melt on north-south versus 
east-west oriented clearcut strips. Forestry Re- 
search Note 234, 4 p. College of Forestry, Univer- 
sity of Minnesota, St. Paul, Minnesota. 

Forcier, L. K. 1975. Reproductive strategies and the 
co-occurrence of climax tree species. Science 
189:808-810. 

Fox, John F. 1977. Alternation and coexistence of 
tree species. The American Naturalist 111:69-89. 

Godman, Richard M., and David J. Books. 1971. In- 
fluence of stand density on stem quality in pole-size 
northern hardwoods. U.S. Department of Agricul- 
ture Forest Service, Research Paper NC-54, 7 p. 
U.S. Department of Agriculture Forest Service, 
North Central Forest Experiment Station, St. 
Paul, Minnesota. 

Godman, Richard M., and Laurits W. Krefting. 1960. 
Factors important to yellow birch establishment in 
Upper Michigan. Ecology 41:18-28. 

Godman, Richard M., and Gilbert A. Mattson. 1976. 
Seed crops and regeneration problems of 19 species 
in northeastern Wisconsin. U.S. Department of 
Agriculture Forest Service, Research Paper NC- 
123, 5 p. U.S. Department of Agriculture Forest 
Service, North Central Forest Experiment Station, 
St. Paul, Minnesota. 

Godman, Richard M., and Carl H. Tubbs. 1973. Es- 
tablishing even-age northern hardwood regenera- 
tion by the shelterwood method — a preliminary 
guide. U.S. Department of Agriculture Forest 
Service, Research Paper NC-99, 9 p. U.S. Depart- 
ment of Agriculture Forest Service, North Central 
Forest Experiment Station, St. Paul, Minnesota. 

Graber, Raymond E., and Donald F. Thompson. 1978. 
Seed in the organic layers and soil of four beech- 
birch-maple stands. U.S. Department of Agricul- 
ture Forest Service, Research Paper NE-401, 8 p. 
U.S. Department of Agriculture Forest Service, 
Northeastern Forest Experiment Station, Broom- 
all, Pennsylvania. 

Jacobs, Rodney D. 1974. Damage to northern 
hardwood reproduction during removal of shelter- 
wood overstory. Journal of Forestry 72:654-656. 



13 



Levy, Gerald F. 1970. The phytosociology of northern 
Wisconsin upland openings. The American Mid- 
land Naturalist 83:213-237. 

Logan, K. T. 1965. Growth of tree seedlings as af- 
fected by light intensity. I. White birch, yellow 
birch, sugar maple and silver maple. Canada De- 
partment of Forestry Publication 1121, 16 p. 

Logan, K. T. 1966. Growth of tree seedlings as af- 
fected by light intensity. III. Basswood and white 
elm. Canada Department of Forestry, Rural Devel- 
opment, Forestry Branch Departmental Publica- 
tion 1176, 15 p. 

Logan, K. T. 1973. Growth of tree seedlings as af- 
fected by light intensity. V. White ash, beech, east- 
ern hemlock and general conclusions. Department 
of Environment Publication 1323, 12 p. Canadian 
Forestry Service, Ottawa. 

Marquis, David A. 1965. Controlling light in small 
clearcuttings. U.S. Department of Agriculture 
Forest Service, Research Paper NE-39, 16 p. U.S. 
Department of Agriculture Forest Service, North- 
eastern Forest Experiment Station, Upper Darby, 
Pennsylvania. 

Marquis, David A. 1966. Germination and growth of 
paper birch and yellow birch in simulated strip 
cuttings. U.S. Department of Agriculture Forest 
Service, Research Paper NE-54, 19 p. U.S. Depart- 
ment of Agriculture Forest Service, Northeastern 
Forest Experiment Station, Upper Darby, Penn- 
sylvania. 

Metzger, Frederick T., and Carl H. Tubbs. 1971. The 
influence of cutting method on regeneration of sec- 



ond-growth northern hardwoods. Journal of For- 
estry 69:559-564. 

Nyland, Ralph D., and H. Jack Irish. 1971. Early 
response to clearcutting in northern hardwoods. 
New York State University, College of Forestry, 
Applied Forestry Research Institute, Research 
Note 2, 1 p. Syracuse, New York. 

Richards, N. A., and C. E. Farnsworth. 1971. Effects 
of cutting level on regeneration of northern 
hardwoods protected from deer. Journal of For- 
estry 69:230-233. 

Ringger, Diane L., and Forest Stearns. 1972. Influ- 
ence of forest openings on climate. University of 
Wisconsin, Field Stations Bulletin 5:8-12. Milwau- 
kee, Wisconsin. 

Stearns, Forest W. 1969. Wildlife pressures. In Sugar 
Maple Conference, p. 51-59. Aug. 20-22, 1968. 
Michigan Technological University, Houghton, 
Michigan. 

Tubbs, Carl H. 1969. The influence of light, moisture, 
and seedbed on yellow birch regeneration. U.S. 
Department of Agriculture Forest Service, Re- 
search Paper NC-27, 12 p. U.S. Department of Ag- 
riculture Forest Service, North Central Forest 
Experiment Station, St. Paul, Minnesota. 

Tubbs, Carl H., and Frederick T. Metzger. 1969. Re- 
generation of northern hardwoods under shelter- 
wood cutting. The Forestry Chronicles 45:333-337. 

Winget, Carl H. 1968. Species composition and devel- 
opment of second-growth hardwood stands in Que- 
bec. The Forestry Chronicles 44(6):31-35. 



COMMON AND SCIENTIFIC NAMES OF TREES AND 

SHRUBS MENTIONED 

Ash, black Fraxinus nigra Marsh. 

white Fraxinus americana L. 

Aspen, quaking Populus tremuloides Michx. 

Basswood Tilia americana L. 

Beech Fagus grandifolia Ehrh. 

Birch, paper Betula papyrifera Marsh. 

yellow Betula alleghaniensis Britton 

Blackberry Rubus fsubgen. eubatus Focke) 

Cherry, pin Prunus pensylvanica L. f. 

Elder, red-berried Sambucus pubens Michx. 

Elm, American Ulmus americana L. 

Hazel, beaked Corylus cornuta Marsh. 

Hemlock, eastern Tsuga canadensis (L.) Carr 

Ironwood Ostrya virginiana (Mill) K. Koch 

Maple, mountain Acer spicatum Lam. 

red Acer rubrum L. 

sugar Acer saccharum Marsh. 

Raspberry Rubus idaeus L. 

Willow Salix spp. L. 



14 



;"r U.S. Government Printing Office: 1980—669-209/50 Region No. 6 



Metzger, Frederick T. 

1980. Strip clearcutting to regenerate northern hardwoods. U.S. Depart- 
ment of Agriculture Forest Service, Research Paper NC-186, 14 p. 
U.S. Department of Agriculture Forest Service, North Central Forest 
Experiment Station, St. Paul, Minnesota. 

Describes results of strip clearcutting trials in mature northern 
hardwood and hemlock-hardwood stands in the Lake States. Two strip 
widths and orientations were tested, with and without herbicide treat- 
ment of the advance regeneration. Establishment, growth, and species 
composition of the regeneration were assessed. 



KEY WORDS: Sugar maple, yellow birch, red maple, eastern hemlock, 
even-aged silvicultural systems, herbicides, Michigan, Wisconsin. 



Metzger, Frederick T. 

1980. Strip clearcutting to regenerate northern hardwoods. U.S. Depart- 
ment of Agriculture Forest Service, Research Paper NC-186, 14 p. 
U.S. Department of Agriculture Forest Service, North Central Forest 
Experiment Station, St. Paul, Minnesota. 

Describes results of strip clearcutting trials in mature northern 
hardwood and hemlock-hardwood stands in the Lake States. Two strip 
widths and orientations were tested, with and without herbicide treat- 
ment of the advance regeneration. Establishment, growth, and species 
composition of the regeneration were assessed. 



KEY WORDS: Sugar maple, yellow birch, red maple, eastern hemlock, 
even-aged silvicultural systems, herbicides, Michigan, Wisconsin. 




Nature is beautiful. . .leave only your footprints. 



ISDA FOREST SERVICE 
FESEARCH PAPER NC-187 



3€vt. cos m§ 



NOV 7 



CLEMSO& 







ra 


^ 






































9L 






























j 








































































_ 


- 










































































































































































































































1 


































































































































































I 




to 








































^'OSk.-M/^M -.; 










































|Pll >f! 










































































































































































































































































































































I 
















































R 














































































































































































































Evaluating stocking in 

upland hardwood forests using 

metric measurements 



Robert Rogers 



North Central Forest Experiment Station 
Forest Service, U.S. Department of Agriculture 



North Central Forest Experiment Station 

Forest Service - U.S. Department of Agriculture 

1992 Folwell Avenue 

St. Paul, Minnesota 55108 

Manuscript approved for publication February 6, 1980 

1980 



EVALUATING STOCKING IN UPLAND 

HARDWOOD FORESTS 

USING METRIC MEASUREMENTS 



Robert Rogers, Mensurationist, 
Columbia, Missouri 



Stocking standards have been established for 
upland hardwood stands. The standards define the 
range of stocking within which the amount of space 
available for tree growth is fully utilized. The meth- 
ods used to develop the stocking standards (Gingrich 
1964, 1967) and the use of these standards in making 
stand prescriptions have been published (Roach and 
Gingrich 1968, Sander 1977) and have been widely 
adopted for field use. 



Stocking is related to basal area per acre, number 
of trees per acre, and diameter of the tree of average 
basal area. These relations have been depicted in the 
form of a stocking chart (fig. 1). However, the data 
needed to interpret stocking percent and average tree 
diameter are based on measurements made in the 
English system. 



110 



140 



130 



P 120 

Hi 

III 

u. 

Uj 

q: 
t 

O 

CO 



100 



90 



80 



70 



60 



50 



40 









> 


s ' 






























k 


.;?<• 




























) 


^7 


• ' 


9% 




















t 


\[ 










"& 




**j 


> 
























o 


*» 


S 


//V c 


V 


\f 




















H 


•** 




^s 


























if 











^? 


■? 














-J 


1 


/ 






























r 


^ 


L 




/ 








7$ 
















S'<" 


t 


r v 


^ 


I 






















T 


?^ 


<** 


„/ 






/* 


p 
















~jr 


/ 




<. 


S 


/ 








A. 
















t 


/ 


l 




/ 






Z£ 




$ 








— 










s 


c 




/ 






& 


























~~y£ 


i 


/ 
















Oaj 
















.o C 














c 




£&*» 


r 






i^ 




h 




















•yi 






<0 






























^yk 
































/ 

























































UJ 
UJ 

u. 

UJ 
DC 

O 

CO 

Lu 

a 
u 

Uj 

a 



nn 








> 


































N? 


X. 






















A: 










***** 
*** 
































^ 


V. 






















o 










\ 


h 






















fc 


I'oi 








s 






















fc 






















/ 


X 


N, 


/ 
















70 










y? 


'(<> 




>^ 


N^ 










"V^i 
















$ 


%> 












A^ 






/ 














^i 


> 






s* 


? 








'JvV 






s 


>< 










~y& 


v 


















/ 










a 


/ 






























/ 












el 


f*c 


% 










y$ 


1 


ff* 


















n 


Sj 








•o 














30 










4> 


























yi 


i 




















on 

































50 



100 



400 



5 6 7 8 9 10 11 12 
HUNDRED TREES PER ACRE 



150 200 250 300 350 

TREES PER ACRE (NUMBER) 

Figure 1. — Relations of basal area, number of trees, and average tree diameter to stocking percent for upland 
hardwood forests of average uniformity. Tree diameter range 7-15 inches (left), 3-7 inches (right). The 
area between curves A andB indicates the range of stocking where trees can fully utilize the growing space. 
Curve C shows the lower limit of stocking necessary to reach the B level in 10 years on average sites. 
(Average tree diameter is the diameter of the tree of average basal area.) (From Gingrich 1967). 



13 14 15 16 



The purpose of this paper is to provide the informa- 
tion necessary to evaluate stocking when measure- 
ments have been made in the metric system and to 
provide some pertinent conversion factors. 



EVALUATING STOCKING 

The basic data needed to evaluate stand stocking 
are basal area and tree count. Stocking percent and 
average stand diameter are determined from figure 2 
using these two data items. Basal area and the num- 
ber of trees can be estimated by using the method 
suggested by Roach and Gingrich (1968) but modified 
to accomodate metric measurements. 



Table 1. — Basal area factor (BAF) conversions 



BAF 


BAF 


BAF 


BAF 


ft 2 /a 


m 2 /ha 


ft 2 /a 


m 2 /ha 


5 


1.15 


1 


4.36 


10 


2.30 


2 


8.71 


15 


3.44 


3 


13.07 


20 


4.59 


4 


17.42 


25 


5.75 


5 


21.78 


30 


6.89 


7 


30.49 



TREE COUNT 



BASAL AREA 

The use of an angle gage or wedge prism makes the 
determination of basal area in the field by the point 
sample technique easy and quick. The 10-factor gage 
has proved well suited for use with the central 
hardwoods and is widely used in this type. The ap- 
proximate metric equivalent of the 10-factor gage 
(ft 2 / A) is the 2-factor gage (m 2 /ha). This gage is cali- 
brated in square meters per hectare, so each tree 
tallied at a point contributes 2 square meters per 
hectare to the estimate of basal area per hectare. The 
metric equivalents of other basal area factors (BAFs) 
are given in table 1. 



A tree count can be made on a plot of fixed radius 
whose center is coincident with the point of the point 
sample. Normally a l/20th-acre plot is recommended 
for the tree count in the English system. In the metric 
system, a l/50th hectare plot whose radius is 7.98 
meters approximates l/20th acre. Trees lying within 
plot boundaries are counted and the total multiplied 
by 50 in order to put the number of trees counted on a 
per hectare basis. Thus, the number of trees in the 
metric system is expressed as the number of trees per 
hectare. 

An alternative method for obtaining the tree count 
is to use the conversion factors for the number of trees 
per hectare based upon the sizes of trees tallied using 



Hi 
OC 




28 


2 


CO 


26 


(1 


or 




Ul 


iu 


24 


T 


t- 






Uj 


22 


a 


Uj 


20 


2 

cr 


a 


18 


«S 


<J 


16 


-j 


00 




•<t 




14 


to 










12 

10 

8 

6 







yf» M .**BR A 


Gc>- 










-at4Aj 


/><i 


^o 2 ' 8 :r e °*M ETe 


R ten,/ 










24 


22 












4^*eb 




20 





































* 






tfr 1 


■7s? 


^ 






«^ 










/ - 


Y+ 


<? 






b/qv 








^_4 


4>* 






( 


:i^t 


l¥§r<- 


iXT"^ 


fe= ?£o 


*f* 










~/~ 7 cl 7 X^y kD y' v 





























































100 200 300 400 500 600 700 800 900 




500 1000 1500 2000 2500 3000 3500 4000 4500 5000 5500 6000 



TREES PER HECTARE (NUMBER) TREES PER HECTARE (NUMBER) 

Figure 2. — Relation of basal area, number of trees, and average tree diameter in metric units to stocking 
percentage for upland central hardwoods. Tree diameter range 20-40 cm (a), and 6-20 cm (b). The area 
between curves A and B on both charts indicates the range of stocking where trees can fully utilize the 
growing space. Curve C shows the lower limit of stocking necessary to reach the B level in 10 years on 
average sites. (Average tree diameter is the diameter of the tree of average basal area.) 



the angle gage. The per hectare conversion factors for 
BAF 2 (m 2 /ha) are given in the tabulation below. 1 



Diameter 

(cm) 

2 

6 

10 

14 
L8 

22 
26 
30 

34 
38 

42 
46 
50 
54 

58 
62 
66 
70 
74 
78 



Number of trees 

6366 

707 
255 
130 
79 
53 
38 
28 
22 
18 
14 
12 
10.2 

8.7 

7.6 

6.6 

5.8 

5.2 

4.7 

4.2 



If two 18 cm, two 30 cm, four 50 cm, and two 62 cm 
trees were tallied at a point, then 2 x 79 + 2 x 28 + 4 
x 10.2 + 2 x 6.6 = 268 trees/ha (the basal area 
would be 2 x 10 trees = 20 m 2 /ha). 



THE TALLY 

A tally such as proposed by Roach and Gingrich 
(1968; fig. 22, p. 24) can be modified for use with 
metric measurements. A 2-factor prism is used to 
estimate basal area in square meters per hectare 
(m 2 /ha) at 10 point samples and the number of trees 
per hectare is estimated by tree counts on l/50th- 
hectare plots. An example of a tally using metric 
measurements is presented in table 2. Stocking per- 
cent and average stand diameter are determined 
from figure 2 using the estimates of basal area and 
number of trees obtained from the data collected in 
table 2. 



l The tabulation is based on the following equation: 
Per-hectare conversion factor = BAFK7.85398 x 



ICT 6 !) 2 ) where BAF = 2. 



Table 2. — Sample tally sheet 



Sample 

point 

number 



Basal area tally 



Mature 
Timber 



Sawtimber 



Poletimber 



Small trees 



AGS 1 



UGS 2 



AGS 



UGS 



AGS 



UGS Cull 



Tree count 

No. trees per 
1/50-hectare 
plot 



1 




' 2 


1 


2 


2 






" 1 


10 


2 




~i 6 




" 3 


■ 2 




1 




13 


3 




1 


' ' 2 


~ 5 


1 






' ' 2 


10 


4 




3 


1 


' 2 


: 4 




1 




9 


5 




~l 6 




1 




1 




" ' 2 


11 


6 




" 2 


' 2 


' 2 


1 




' 2 




12 


7 




: 4 




1 


" 3 




1 




10 


8 




' 3 


' 2 




' 2 


' 2 




" ' 2 


9 


9 




~ 5 


1 


1 


3 


1 


' ■ 2 


1 


14 


10 




2 


1 


3 


" 2 




1 




12 


Total 





34 


10 


20 


20 


4 


8 


8 


110 


Per ha 3 




6.8 


2.0 


4.0 


4.0 


.8 


1.6 


1.6 


110 x 5 = 550 



Basal area: AGS 11.6 Total 20.8 
Number of trees/ha: 550 
Stocking percent: 83 
Average tree diameter: 22 cm 



1 Acceptable Growing Stock. 

2 Unacceptable Growing Stock. 



Obtained by multiplying 2 times total and dividing by 10 (number of points), e.g., 2 x 34 = 68 



68/10 = 6.8. General formula is BAF x tree coun t. 

No. of points 



STOCKING PERCENT 
PARAMETER CONVERSIONS 

Parameter estimates made from data collected in 
either the English or metric system can be converted 
to the other system using the conversion factors pres- 
ented on page xx. 

Example 1. Metric to English 
Given: 

Percent stocking = 85 
Number of trees/hectare = 530 
Basal area, square meters per hectare = 22 
Average tree diameter, cm = 23.0 
English equivalent: 

Percent stocking = 85 

Number of trees/acre = 0.4047 x 530 = 214 

Basal area, square feet per acre = 4.356 x 22 = 

96 

Average tree diameter, inches = 0.3937 x 23.0 = 

9.1 

Example 2. English to Metric 
Given: 

Percent stocking = 90 
Number of trees per acre = 275 
Basal area, square feet per acre = 96 
Average tree diameter, inches = 8.0 
Metric equivalent: 

Percent stocking = 90 

Number of trees per hectare = 2.471 x 275 = 

680 

Basal area, square meters per hectare = 0.22957 

x 96 = 22 

Average tree diameter, cm = 2.54 x 8.0 = 20.3 



English-Metric 
Conversion Factors 

Metric to English: 

Basal area per acre (sq. ft.) = 4.356 x basal area 
per hectare (m 2 ) 

Trees per acre = 0.4047 x trees per hectare 

Average tree diameter (inches) = 0.3937 x aver- 
age tree diameter (cm) 

Tree basal area (sq. ft.) = .0008454 x d.b.h. 2 (cm) 

Square feet = 10.764 x square meters 

Feet = 0.3048 x meters 

Acres = 2.471 x hectares 



English to Metric: 

Basal area per hectare (sq. meters) = 0.22957 x 

basal area per acre (ft. 2 ) 
Trees per hectare = 2.471 x trees per acre 
Average tree diameter (cm) = 2.54 x average tree 

diameter (in.) 
Tree basal area (sq. meters) = .0005067 x d.b.h. 2 

(in.) 
Square meters = 0.0929 x square feet 
Meters = 3.28 x feet 
Hectares = 0.4047 x acres 



TREE AREA EQUATION 

Gingrich's (1967) original tree area equation used 
to compute the stocking-density criteria expressed 
tree area in terms of mil-acres. This equation has the 
form: 



Tree area (mil-acres) = aN + b 2D + c 2D 2 , where 
N = number of trees per acre and D = tree diameter 
in inches. Solving this equation (given the appropri- 
ate constants a, b, and c) for N = 1 and D = any given 
diameter, defines the minimum tree area require- 
ments for that size tree. Thus a stand having 1,000 
mil-acres of tree area per acre was considered to be 
100 percent stocked (A-Level). 

In the metric system tree area is expressed in ares. 
One are equals 100 square meters and therefore 100 
ares equals one hectare (10,000 square meters). A 
stand having 100 ares of tree area per hectare is 
considered to be 100 percent stocked. N is the number 
of trees per hectare and D is measured in centimeters. 
The metric stocking equation is: 

Tree area (ares) = (-2.0518N + 2,7053 2D + 
0.19884 2D 2 )/1000. (Stocking percent) 

Minimum tree area requirements expressed in ares 
(stocking percent) for individual trees of varying di- 
ameters are given in the tabulation below. The met- 
ric equation expressing the maximum amount of 
area that trees can use (B-Level) is: 



Tree area (ares) 
2D 2 )/1000. 



(7.0820 + 3.2662 2D + 0.37636 



Xb.h. 


Tree area 


D.b.h. 


Tree area 


(cm) 


(ares) 


(cm) 


(ares) 


6 


0.0213 


40 


0.4243 


8 


0.0323 


42 


0.4623 


10 


0.0449 


44 


0.5019 


12 


0.0590 


46 


0.5431 


14 


0.0748 


48 


0.5859 


16 


0.0921 


50 


0.6303 


18 


0.1111 


52 


0.6763 


20 


0.1316 


54 


0.7239 


22 


0.1537 


56 


0.7730 


24 


0.1774 


58 


0.8238 


26 


0.2027 


60 


0.8761 


28 


0.2296 


62 


0.9300 


30 


0.2581 


64 


0.9855 


32 


0.2881 


66 


1.0427 


34 


0.3198 


68 


1.1013 


36 


0.3530 


70 


1.1616 


38 


0.3879 


72 


1.2235 




SUMMARY 





In an effort to carry out our Nation's commitment 
to adopt the metric standard, this publication makes 
available information necessary to evaluate stocking 



in upland hardwood forests when stocking variables 
are based on metric measurements. Metric-English 
conversions are presented so that comparisons can be 
made between stocking variables obtained using 
either measurement system. 



LITERATURE CITED 

Gingrich, Samuel F. 1964. Criteria for measuring 
stocking in forest stands. Society of American For- 
esters Proceedings 1964:198-201. 

Gingrich, Samuel F. 1967. Measuring and evaluat- 
ing stocking and stand density in upland hardwood 
forests in the Central States. Forest Science 13:38- 
53. 

Roach, Benjamin A., and Samuel F. Gingrich. 1968. 
Even-aged silviculture for upland central 
hardwoods. U.S. Department of Agriculture, Agri- 
culture Handbook 355, 39 p. 

Sander, Ivan L. 1977. Manager's handbook for oaks 
in the north-central States. U.S. Department of 
Agriculture Forest Service, General Technical Re- 
port NC-37, 35 p. U.S. Department of Agriculture 
Forest Service, North Central Forest Experiment 
Station, St. Paul, Minnesota. 



OU.S. GOVERNMENT PRINTING OFFICE: 1980--669290/63 



Rogers, Robert. 

1980. Evaluating stocking in upland hardwood forests using metric 
measurements. U.S. Department of Agriculture Forest Service, 
Research Paper NC-187, 5 p. U.S. Department of Agriculture For- 
est Service, North Central Forest Experiment Station, St. Paul, 
Minnesota. 

Allows stocking percent in upland forests as developed by Gingrich 
(1964, 1967) to be calculated when tree and stand variables are mea- 
sured in the metric system. 

KEY WORDS: Metric stocking percent, metric stocking conversion 
factors, metric tree area equation, tree area requirements, and oak- 
hickory forests. 




Help Aave. tkt bvuih, aywrnoJU and filoweAA. 



5DA FOREST SERVICE 
SEARCH PAPER NC-188 



GOVT. 

NOV 10 1980 

CLEMSON 
UBRARY 







SS/V 




■^ 



, ■ ,^- : 



^-y 



[*%?, *^pP" 



^VV\J ■^wv"-' 










Predicted yields from 
selected cutting prescriptions 
in northern Minnesota 



Pamela J. Jakes and W. Brad Smith 



>rth Central Forest Experiment Station 

rest Service, U.S. Department of Agriculture 



North Central Forest Experiment Station 

Forest Service — U.S. Department of Agriculture 

1992 Folwell Avenue 

St. Paul, Minnesota 55108 

Manuscript approved for publication February 22, 1980 

1980 



CONTENTS 

Page 

Methods 1 

Assumptions 3 

Findings 3 

Discussion 5 

Appendix 27 

Specifying cutting prescriptions 27 

Definition of terms 27 

Metric equivalents of units used 

in this report 28 

Principal tree species groups in 
northern Minnesota 29 



PREDICTED YIELDS FROM SELECTED 

CUTTING PRESCRIPTIONS 

IN NORTHERN MINNESOTA 



Pamela J. Jakes, Associate Resource Analyst, 
and W. Brad Smith, Associate Mensurationist 



As demand for forest products grows, it will become 
increasingly important to use forest resources effi- 
ciently. In Minnesota, the commercial forest area 
decreased 11 percent between 1962 and 1977, yet 
existing forest industries are expanding and new 
wood-using industries are entering the State. This 
increase in demand for timber coupled with a shrink- 
ing commercial forest land base, emphasizes the need 
for improved utilization of Minnesota's timber 
resources. 

By calculating predicted yields from selected cut- 
ting prescriptions, we identified deficiencies or sur- 
pluses in the forest resource. The two sets of cutting 
prescriptions used in this report should not be 
construed as management goals, but rather as side- 
boards defining a range of future yields, using cur- 
rent Minnesota forest conditions as a base. Predicted 
yields resulting from the prescriptions are only one 
facet of the total resource picture; they should be used 
in conjunction with other resource information in 
planning for the future. 



METHODS 

The first step involved in calculating predicted 
yields was specifying cutting prescriptions. We con- 
sidered two prescriptions, each defined by a set of 
rotation ages for 14 Minnesota forest types (table 1). 
The long rotation age option is a prescription based 
on criteria established jointly by representatives 
from Minnesota forest industries and the Minnesota 
Department of Natural Resources. The rotation ages 
used in this prescription approximate current har- 
vest practices, where the final product (sawtimber or 
pulpwood) depends on the forest type and site quality 
(specified by site index ranges). The short rotation 



age option uses current empirical yields to select 
rotation age at the age of maximum total volume 
yield. The final product from this prescription is pri- 
marily poletimber (see Appendix). 

For both prescriptions, an area control algorithm 
was used to calculate the number of acres cut in each 
forest type and site index range. The number of acres 
cut was set so that an even distribution of area among 
age classes would be achieved in each forest type by 
the end of one rotation. Harvest areas were calcu- 
lated for the decade 1977-1986; after 1986, the areas 
must be recalculated to account for changes in stand 
characteristics and the commercial forest land base. 

Plot data from the 1977 Minnesota Forest Inven- 
tory were then used in a modified version of the Tree 
Growth Projection System of the Forest Resources 
Evaluation Program (FREP). 1 The System "grew" 
each plot for 5 years and then scanned data from each 
commercial forest plot, selected plots that met the 
cutting prescription criteria, and calculated the area 
and volume represented by the plot. Since yields are 
estimated for a 10-year period, it was necessary to 
project each plot 5 years as an estimate of average 
growth on all harvest plots for the decade. 

The System assigns the highest cutting priority to 
overmature stands (fig. 1). In some forest types, there 
are large areas of stands too young to harvest. Some 
of these stands were "harvested" before rotation age 



1 U.S. Department of Agriculture, Forest Service. 
1979. A generalized forest growth projection system 
applied to the Lake States region. U.S. Department of 
Agriculture Forest Service, General Technical Report 
NC-49, 96 p. U.S. Department of Agriculture Forest 
Service, North Central Forest Experiment Station, St. 
Paul, Minnesota. 





1977-1986 
























40 




















20 






















ri 








„™~ 



1- 11- 21- 31- 41- 51- 61- 71- 81- 91- 
10 20 30 40 50 60 70 80 90 100 



40 - 



20 - 



2007-2016 














- 







































1- 11- 21- 31- 41- 51- 61- 71- 
10 20 30 40 50 60 70 80 



co 

o 

Q 

O 

l- 



40 - 



20 - 



1987-1996 












- 







































CO 
UJ 
QC 

o 
<* 

Q 
O 

a: 



40 


2017-2026 










20 






























n 





1- 11- 21- 31- 41- 51- 61- 
10 20 30 40 50 60 70 



1- 11- 21- 31- 41- 51- 61- 
10 20 30 40 50 60 70 





1997-2006 












40 












- 










20 






















n 





1- 11- 21- 31- 41- 51- 61- 
10 20 30 40 50 60 70 



40 



20 



2027-2036 



1- 11- 21- 31- 41- 
10 20 30 40 50 



STAND-AGE CLASSES (YEARS) 



STAND-AGE CLASSES (YEARS) 



Figure 1. — Example of the progression of the area control program to even area distribution 
among stand-age classes for a 50 -year rotation by decades. Shaded portions show area cut per 
decade. 



to bring about an even area distribution among age 
classes by the end of one rotation. The number of 
acres harvested in an age class cannot be directly 
compared to the number of acres in the age class 
as reported in the 1977 Forest Inventory of Minne- 
sota. Because plots were grown for 5 years before 
harvest, a shift in commercial land between age 
classes resulted. 



ASSUMPTIONS 

Calculations of yields from selected cutting pre- 
scriptions are based on these assumptions: (1) the 
area of commercial forest land will remain stable for 
the decade 1977-1986; (2) the current mix of area by 
forest type will remain stable — i.e., no stand conver- 
sion will occur; and (3) all timber is available for 
harvest and a ready market exists for all species and 
products. The method does not account for possible 
economic, social, political, or silvicultural con- 
straints on timber removals. 



FINDINGS 

Predicted yields from selected cutting prescrip- 
tions are presented for the Aspen-Birch and the 
Northern Pine Forest Survey Units, referred to here 
as the northern Minnesota region (fig. 2). Data from 
the 1977 Minnesota Forest Inventory show that these 
two Units contain 82 percent of the State's commer- 
cial forest land. Of the 11.2 million acres of commer- 
cial forest land in these Units, 62 percent is in public 
ownership. The State of Minnesota owns the largest 
portion of the commercial forest land base, 2.4 mil- 
lion acres. Growing-stock volume in the region totals 
9.5 billion cubic feet. The region has 98 percent of the 
State's softwood growing-stock volume. In 1976, 
growing-stock removals totaled 150.7 million cubic 
feet, 78 percent of the State total. 



Predicted Yields and 

Area Harvested — Long Rotation 

Age Option 

Between 1977 and 1986, 1.9 million acres of com- 
mercial forest land in northern Minnesota can be 
harvested, according to long rotation age criteria (ta- 
ble 2). The harvest acreage is equally divided be- 
tween the Aspen-Birch Unit and the Northern Pine 
Unit, with each having approximately 1.0 million 




Figure 2. — Northern Minnesota region. 



acres available for harvest over the decade (tables 3 
and 4). Of the 11.2 million acres of commercial forest 
land in the region, 2 percent would be cut annually 
under this harvest plan. 

The aspen forest type accounts for 49 percent of the 
harvest acreage, and the paper birch, balsam fir, and 
jack pine forest types cover an additional 25 percent. 
Although the distribution of harvest acreage among 
forest types differs slightly between the Survey 
Units, aspen makes up the largest portion of the 
harvest acreage in both — 45 percent in the Aspen- 
Birch Unit, 52 percent in the Northern Pine Unit. In 
the Aspen-Birch Unit, balsam fir accounts for 13 
percent of the harvest acreage, while in the Northern 
Pine Unit the type accounts for 5 percent. Black 
spruce occurs on 7 percent of the harvest acreage in 
the Aspen-Birch Unit, but on only 2 percent in the 
Northern Pine Unit. 

In most forest types, distribution of harvest acre- 
age among age classes is concentrated in overmature 
stands (tables 5-7). As the System works toward es- 
tablishing an even distribution among age classes for 
each forest type by the end of one rotation, it cuts an 
equal number of acres every year. During this proc- 
ess, overmature stands are assigned the highest cut- 
ting priority. For example, the rotation age of low site 
index jack pine is 50 years, yet during the decade 



1977-1986, 52 percent of the jack pine harvest acre- 
age would be over 70 years old. In low site index 
aspen, all the harvest acreage would be at least 10 
years past rotation age. 

A different problem arises in forest types where 
young stands dominate large areas. In the case of 
high site index red pine with a rotation age of 120 
years, all of the acreage harvested in the decade 
1977-1986 would be in stands less than 120 years 
of age. Again, this is necessary to achieve an equal 
area distribution among all classes by the end of one 
rotation. 

Harvest volume under long rotation age criteria 
would total 279.5 million cubic feet annually (table 
8). Of this total, 253.9 million cubic feet are growing 
stock and 25.6 million cubic feet are cull. Although 
harvest acreage is evenly divided between the Aspen- 
Birch Unit and the Northern Pine Unit, the greatest 
harvest volume would occur in the Aspen-Birch Unit 
(tables 9 and 10). This discrepancy is due to differ- 
ences in the volume per acre on commercial forest 
land between the two Units. The average growing- 
stock volume per acre on commercial forest land 
harvested in the Aspen-Birch Unit is 1,621 cubic 
feet, while in the Northern Pine Unit the volume per 
acre on commercial forest land harvested is 1,117 
cubic feet. 

Hardwood species comprise 65 percent of the har- 
vest volume. Aspen accounts for 36 percent of the 
growing-stock harvest, the largest percentage of any 
species. The softwood species accounting for the most 
harvest volume is balsam fir, with 8 percent of total 
harvest. 

A large portion of the harvest volume of many 
species is found in the aspen forest type (tables 11- 
13). Of the 5.4 million cubic feet of white pine harvest 
volume, 55 percent is in the aspen type; only 12 
percent is in the white pine type. Over one-third of 
the total softwood harvest volume is in aspen stands. 
The dominance of the aspen type in the hardwood 
harvest volume would be expected — nearly three- 
quarters of the hardwood harvest volume is found in 
the type. 

The hardwood annual harvest is fairly evenly di- 
vided between poletimber and sawtimber. The an- 
nual harvest of hardwood poletimber would total 83.8 
million cubic feet, while the annual harvest of 
hardwood sawtimber would total 82.1 million cubic 
feet. Softwood annual harvest is composed primarily 
of sawtimber, which accounts for 62 percent of the 
softwood growing-stock harvest. 



Predicted Yields and Area 

Harvested — Short Rotation 

Age Option 

In changing harvest criteria from a long rotation 
age option to a short rotation age option (or from 
sawtimber to poletimber harvest ages), the area of 
commercial forest land available for harvest over the 
decade increases from 1.9 million acres to 2.5 million 
acres (table 2). Under short rotation age criteria, 
timber would be harvested on 254,700 acres annu- 
ally. Harvest acreage remains fairly evenly divided 
between the Aspen-Birch Unit and the Northern 
Pine Unit (tables 3 and 4). 

As in the long rotation age option, the largest por- 
tion (50 percent) of the harvest acreage is in the aspen 
forest type. Paper birch, balsam fir, and black spruce 
types account for an additional 23 percent of the 
harvest acreage. The 1.3 million acres of aspen that 
would be harvested during the decade under the 
short rotation age option are 42 percent more than 
the acreage that would be harvested under the long 
rotation age option. Black spruce harvest acreage 
would increase 68 percent under short rotation age 
harvest criteria. Harvest acreages in the northern 
white-cedar and oak forest types would more than 
double. 

Although a large portion of the harvest acreage in 
the short rotation age option remains concentrated in 
overmature stands, additional younger aged stands 
would also be harvested (tables 5-7). For example, in 
the aspen forest type, 904,000 acres would be har- 
vested during the decade under long rotation age 
criteria, all in stands over 50 years old. Under short 
rotation age criteria, an additional 2,900 acres would 
be cut in the 61-70 age class, 357,800 acres in the 51- 
60 years age class and 18,000 acres in the 41-50 years 
age class, for a total of 1.3 million acres. 

Harvest volume under short rotation age criteria 
would total 372.7 million cubic feet annually — 337.3 
million cubic feet in growing-stock trees, and 35.4 
million cubic feet in cull (table 8). Harvest volumes 
under the short rotation age option are about 33 
percent higher than volumes under the high rotation 
age option. Differences between Survey Units in the 
growing-stock volume per acre on commercial forest 
land harvested result in more volume being har- 
vested in the Aspen-Birch Unit than in the Northern 
Pine Unit (tables 9 and 10). Harvest volume in the 
Aspen-Birch Unit would total 188.7 million cubic feet 
(1,526 cubic feet per acre) and in the Northern Pine 



Unit it would total 148.6 million cubic feet (1,134 
cubic feet per acre). 

Of the 337.3 million cubic feet of growing-stock 
harvest volume, 65 percent is hardwood (primarily 
aspen), and 35 percent is softwood. The 123.5 million 
cubic feet of aspen harvested is 37 percent of the total 
harvest and 56 percent of the hardwood harvest. Un- 
der short rotation age criteria, aspen harvest volume 
alone is greater than the total softwood harvest vol- 
ume. Balsam fir accounts for the largest portion of 
the softwood harvest volume (28.5 million cubic feet), 
and jack pine and black spruce harvest volumes are 
20.2 million and 20.9 million cubic feet, respectively. 

The aspen forest type, which is important in sup- 
plying softwood as well as hardwood harvest volume, 
contains 59 percent of total growing-stock harvest 
volume (tables 11-13). Fifty percent of the white pine, 
49 percent of the balsam fir, and 60 percent of the 
white spruce harvest volumes are found in the aspen 
forest type. 

Most of the increase in harvest volume resulting 
from the change in harvest criteria from long rota- 
tion age to short rotation age occurs in poletimber 
stands. Harvest volumes under short rotation age 
criteria are fairly evenly divided between sawtimber 
and poletimber stands; however the harvest volumes 
of species in the two size classes vary widely. Most of 
the softwood harvest volumes are concentrated in 
sawtimber trees — 95 percent of both the red pine and 
white pine harvest volume is in sawtimber stands. 
Three softwood species — balsam fir, black spruce, 
and tamarack — have the majority of their harvest 
volumes in pole-timber stands. Hardwood species are 
evenly divided between those with a majority of their 
volume in poletimber and those with a majority in 
sawtimber stands. 



DISCUSSION 

This study indicates that, under the assumptions 
specified above, the supply of timber from Minneso- 
ta's commercial forest land would be much higher 
than current timber removals (fig. 3). Under the long 
and short rotation cutting prescriptions considered, 
average annual growing-stock removals for the dec- 
ade would be from 69 percent to 124 percent higher 
than the 1976 growing-stock removals (tables 14- 
16). 2 In reality, however, these yields may never be 



2 Removals resulting from the transfer of commer- 
cial forest land to productive-reserved forest land are 
not included. 



realized for the following reasons: (1) the objectives of 
some landowners may be incompatible with timber 
production, (2) edaphic, topographic, or hydrologic 
characteristics of some sites may make it impossible 
to harvest timber, (3) some areas may be inaccessible, 
(4) markets may not exist for the products or species 
harvested. By using their knowledge of the local re- 
source and other inventory data, forest managers and 
planners can temper the findings of this report to fit 
the circumstances of their areas. 



Hardwood Removals Would 

More than Double 

Between 1977 and 1986 

Opportunities for increasing hardwood removals 
are greater than those for increasing softwood re- 
movals. Under the long rotation age option, average 
annual hardwood removals over the decade would be 
165.9 million cubic feet — nearly twice the 1976 re- 
movals of 83.4 million cubic feet. The annual remov- 
als under low rotation age criteria would be 220.9 
million cubic feet, 137.5 million cubic feet higher 
than the 1976 removals. For every hardwood species, 
empirical yields from harvests are higher than 1976 
removals. 

The distribution of hardwood removals volume 
among species is different under both cutting pre- 
scriptions from that in the 1976 harvest (fig. 4). 
Aspen will continue to provide the largest portion of 
hardwood removals volume; however, the relative 
importance of aspen in supplying harvest volume 
will decline. In 1976, aspen removals were 79 percent 
of total hardwood removals, while under the cutting 
prescriptions outlined here, aspen will contribute an 
average of 55 percent of harvest volume. The impor- 
tance of paper birch, balsam poplar and ash will 
increase under these cutting prescriptions. 



Softwood Removals Would 

Increase More than 

38 Percent 

Under long rotation age criteria, softwood annual 
removals would total 88.0 million cubic feet — 20.7 
million cubic feet higher than the 67.3 million cubic 
feet of softwood growing-stock removals in 1976. The 
difference between 1976 removals and predicted 



WHITE - 
RED PINE 



JACK 
PINE 



SPRUCE 
FIR 



OTHER 
SOFTWOODS 



PAPER 
BIRCH 



m 



ASPEN 



OTHER 
HARDWOODS 



] 1976 REMOVALS 
AVERAGE ANNUAL PREDICTED YIELDS 
LONG ROTATION AGE OPTION 



M 



10 



| } SHORT ROTATION AGE OPTION 



_L 



20 



30 



40 



50 



60 



70 



80 



90 



100 



110 



120 



GROWING STOCK REMOVALS (MILLION CUBIC FEET) 

Figure 3. — 1976 growing-stock removals and average annual predicted yields from selected 
cutting prescriptions, by species groups, northern Minnesota. 





ASPEN \ 




JACK PINE 






(79 PERCENT) \ 




(31 PERCENT) 


/ BALSAM FIR 
/ (20 PERCENT) 


OTHER HARDWOODS ~' S \^^\s/ / 
(6 PERCENT) X^S / 




WHITE PINE - JS \ 

(5 PERCENT) \^ 




/ \ BLACK SPRUCE 
/ \ (20 PERCENT) 


BALSAM POPLAR \S / 
(3 PERCENT) A/ 

OAK^ \. / 




NORTHERN WHITE - -^\ 
CEDAR (5 PERCENT) 






(3 PERCENT) 2^^^ 




WHITE SPRUCE 






PAPER ' 




(6 PERCENT) 






BIRCH 






RED PINE 


TAMARACK 


(9 PERCENT) 






(6 PERCENT) 


(7 PERCENT) 















BLACK SPRUCE\ 




ASPEN \ 






(18 PERCENT) \ 




(56 PERCENT) \ 


/ BALSAM FIR 
/ (24 PERCENT) 
















VsS^ JACK PINE 














•^fy\ (17 PERCENT) 










TAMARACK "^X^^^ 






OTHER HARDWOODS ^ \^^ 
(6 PERCENT) V^ 

OAK^ V/ 
(5 PERCENT) \y 




PAPER BIRCH yf 
(19 PERCENT) / 


(4 PERCENT) V 

WHITE JfC / 
PINE \^ 
(6 PERCENT) \/ 

WHITE /^ 






ASH / 
(6 PERCENT) 






SPRUCE 
(8 PERCENT) 




-+ ^^^ NORTHERN 

1 WHITE - CEDAR 




BALSAM 






RED (13 PERCENT) 




POPLAR 






PINE 






(8 PERCENT 


) 




(10 PERCENT) 



Figure 4. — (a) 1976 hardwood growing-stock remov- 
als by species and (b) average annual predicted 
yields under short-rotation-age criteria, northern 
Minnesota. 



Figure 5. — (a) 1976 softwood growing -stock remov- 
als by species and (b) average annual predicted 
yields under short-rotation-age criteria, northern 
Minnesota. 



harvest volume is even greater under short rotation 
age criteria, where removals would be 116.5 million 
cubic feet. 

Opportunities exist for increased removals from all 
softwood species, with the exception of tamarack and 
jack pine. Under both sets of harvest prescriptions, 
tamarack removals in 1976 exceed predicted harvest 
yields. The 1976 level of tamarack removals cannot 
be maintained if the objectives of the cutting pre- 
scriptions are to be achieved. Jack pine yields from 
the long and short rotation age criteria fall below the 
level of removals in 1976. 

Softwood removals are more evenly distributed 
among species under the harvest prescriptions than 



they were in the 1976 removals. Jack pine, balsam 
fir, and black spruce will continue to contribute the 
bulk of softwood removals, but their percentages will 
decline during the decade as removals of the other 
species increase (fig. 5). 

Aspen-Birch Unit has 
Greatest Potential 

Under the harvest prescriptions, the Aspen-Birch 
Unit will supply a greater proportion of total remov- 
als than it did in 1976. Growing-stock removals in 
northern Minnesota in 1976 were fairly evenly di- 
vided between the Aspen-Birch Unit and the North- 
ern Pine Unit. Under the cutting prescriptions, the 



Aspen-Birch Unit will contribute approximately 57 
percent of harvest volume. 

Predicted yields from cutting prescriptions range 
from 103.2 million cubic feet to 186.7 million cubic 
feet higher than 1976 removals. Most of this increase 
in removals volume would be in the Aspen-Birch 
Unit. The Unit would account for 69 percent of the 
increase in the long rotation age option, and 60 
percent of the increase in the short rotation age op- 



tion. Softwood yields are concentrated in the Aspen- 
Birch Unit. 

The Northern Pine Unit plays an increasingly im- 
portant role in supplying harvest volume of some 
species under the cutting prescriptions. For example, 
in 1976, jack pine removals were evenly divided be- 
tween the two northern Survey Units; under the pre- 
scriptions, the Northern Pine Unit would supply a 
majority of the jack pine harvest volume. 



Table l.--Site index and rotation age by forest type for long and short 
rotation age cutting prescriptions, northern Minnesota, 1977-1986 



Forest type 




Long rotation age option 
Site Rotation 
index age 


Short rotation age option 
Site Rotation 
index age 








years 




years 


Jack pine 




0-60 
61 + 


50 
60 


al" 


40 


Red pine 




0-55 
56+ 


100 
120 


al 


75 


White pine 




0-55 
56+ 


100 
120 


al 


75 


Balsam fir 




all 


50 


al 


45 


White spruce 




all 


80 


al 


65 


Black spruce 




0-40 
41 + 


120 
90 


al 


60 


Northern white- 


cedar 


all 


100 


al 


85 


Tamarack 




0-40 
41 + 


120 
90 


al 


45 


Oak 




0-55 
56+ 


100 
80 


al 


40 


Elm-ash-cottonwood 


0-55 
56+ 


70 
90 


al 


50 


Maple-basswood 




all 


»!/ 


al 


1 90i/ 


Aspen 




0-65 
66+ 


40 

60 


al 


! 35 


Paper birch 




0-65 
66+ 


40 

60 


al 


1 35 


Balsam poplar 




0-65 
66+ 


40 

60 


al 


1 40 



JVTubbs, Carl H. 1977. Manager's handbook for northern hardwoods in the North 
Central States. U.S. Department of Agriculture Forest Service, General Technical 
Report NC-29, 29 p. U.S Department of Agriculture Forest Service, North Central 
Forest Experiment Station, St. Paul, Minnesota (applied to stands over age 90). 



Table 2.-- Average annual predicted yields from selected cutting prescriptions 
by forest type and tree class, northern Minnesota, 1977-1986 

LONG ROTATION AGE OPTION 







Number 
















growing- 










1/ 






stock 




Pole- 


Saw- 




Other 


Forest 


Harvest 


trees 


Total 


timber 


timber 


Short-log 


cull 


type 


area 


harvested 


volume 


volume 


volume 


volume 


volume 




Thousand 


Thousand 














acres 


trees 










Jack pine 


9.2 


1,593 


14,933 


4,207 


9,870 


696 


160 


Red pine 


2.2 


377 


4,963 


814 


3,969 


^3 


87 


White pine 


0.5 


69 


1,262 


172 


955 


100 


25 


Balsam fir 


16.5 


2,634 


16,488 


8,294 


7,002 


923 


269 


White spruce 


1.0 


104 


1,240 


127 


1,049 


43 


21 


Black spruce 


8.4 


2,442 


8,960 


6,504 


2,230 


61 


165 


Northern white-cedar 


4.6 


1,044 


4,738 


1,208 


2,950 


324 


256 


Tamarack 


4.1 


760 


3,281 


2,050 


1,001 


72 


158 


Oak-hickory 


2.8 


361 


3,128 


950 


1,825 


284 


69 


Elm- ash- cottonwood 


7.3 


1,121 


7,825 


3,037 


3,697 


559 


532 


Mapl e-basswood 


7.4 


139 


1,903 


558 


11 


977 


357 


Aspen 


90.4 


16,473 


166,916 


71,455 


80,291 


9,126 


6,044 


Paper birch 


20.9 


3,607 


30,603 


13,562 


14,136 


1,882 


1,023 


Balsam poplar 


10.7 


1,366 


13,227 


4,747 


7,213 


760 


507 


All types 


186.0 


32,090 


279,467 


117,685 


136,209 


15,900 


9,673 






SHORT ROTATI 


OPTION 








Jack pine 


12.2 


2,570 


21,165 


7,614 


12,553 


765 


233 


Red pine 


3.2 


405 


5,896 


626 


5,113 


113 


44 


White pine 


0.7 


85 


1,703 


128 


1,424 


135 


16 


Balsam fir 


18.4 


3,066 


19,837 


9,601 


8,680 


1,194 


362 


White spruce 


1.2 


216 


1,628 


399 


1,165 


43 


21 


Black spruce 


14.0 


4,117 


13,570 


10,529 


2,703 


57 


281 


Northern white-cedar 


10.0 


2,163 


9,972 


2,589 


6,124 


790 


469 


Tamarack 


5.1 


1,046 


3,600 


2,601 


738 


60 


201 


Oak-hickory 


6.3 


855 


6,631 


2,195 


3,713 


559 


164 


Elm-ash-cottonwood 


10.9 


1,651 


11,801 


4,956 


5,423 


719 


703 


Mapl e-basswood 


7.4 


139 


1,903 


558 


11 


977 


357 


Aspen 


128.3 


22,362 


222,925 


97,175 


103,515 


13,115 


9,120 


Paper birch 


25.6 


4,854 


37,420 


18,626 


15,383 


2,127 


1,284 


Balsam poplar 


11.4 


1,442 


14,633 


5,236 


7,979 


815 


603 


All types 


254.7 


44,971 


372,684 


162,833 


174,524 


21,469 


13,858 



— Rough and rotten cull. 



Table 3.— Average annual predicted yields from selected cutting prescriptions 
by forest type and tree class, Aspen-Birch Unit, Minnesota, 1977-1986 



LONG ROTATION AGE OPTION 







Number 


















growing- 












1/ 






stock 




Pole- 


Saw- 






Other 


Forest 


Harvest 


trees 


Total 


timber 


timber 


Short- log 


cull 


type 


area 


harvested 


volume 


volume 


volume 


vo 


lume 


volume 




Thousand 


Thousand 
















acres 


trees 






nd cubic 


feet 






Jack pine 


3.1 


525 


5,539 


1,698 


3,307 




495 


39 


Red pine 


1.2 


274 


3,043 


630 


2,334 




32 


47 


White pine 


0.3 


29 


605 


78 


415 




97 


15 


Balsam fir 


12.2 


2,045 


13,797 


6,685 


6,099 




825 


188 


White spruce 


0.8 


68 


902 


78 


792 




32 





Black spruce 


6.0 


1,739 


7,046 


5,010 


1,865 




61 


110 


Northern white-cedar 


3.1 


620 


3,042 


687 


1,983 




238 


134 


Tamarack 


1.6 


235 


1,228 


632 


492 




46 


58 


Oak-hickory 


0.1 


12 


94 


50 


36 




8 





Elm- ash- cottonwood 


3.4 


504 


3,335 


1,485 


1,540 




188 


122 


Maple-basswood 


2.2 


27 


553 


133 


11 




294 


115 


Aspen 


40.9 


9,501 


95,114 


44,577 


45,520 


3 


,102 


1,915 


Paper birch 


12.2 


2,025 


18,490 


8,150 


8,535 


1 


,252 


553 


Balsam poplar 


4.5 


623 


6,370 


2,395 


3,257 




431 


287 


All types 


91.6 


18,227 


159,158 


72,288 


76,186 


7 


,101 


3,583 






SHORT ROTP 


OPTION 










Jack pine 


4.0 


1,057 


8,512 


3,787 


4,227 




458 


40 


Red pine 


1.6 


277 


2,981 


395 


2,580 




6 





White pine 


0.4 


39 


795 


90 


569 




125 


11 


Balsam fir 


13.6 


2,436 


16,908 


7,880 


7,665 


1 


,084 


279 


White spruce 


0.9 


163 


1,198 


329 


837 




32 





Black spruce 


9.9 


2,961 


10,239 


7,992 


2,033 




41 


173 


Northern white-cedar 


6.8 


1,411 


6,755 


1,697 


4,299 




498 


261 


Tamarack 


1.9 


347 


1,299 


835 


382 




29 


53 


Oak-hickory 


0.1 


27 


239 


120 


93 




26 





Elm- ash- cottonwood 


4.9 


816 


5,463 


2,541 


2,422 




279 


221 


Maple-basswood 


2.2 


27 


553 


133 


11 




294 


115 


Aspen 


57.5 


12,089 


117,826 


56,127 


54,537 


4 


,267 


2,895 


Paper birch 


14.8 


3,013 


22,779 


11,986 


8,960 


1 


,290 


543 


Balsam popl ar 


5.1 


721 


6,977 


2,757 


3,446 




463 


311 


All types 


123.7 


25,384 


202,524 


96,669 


92,061 


8 


,892 


4,902 



2/Rough and rotten cull 



10 



Table 4.-- Average annual predicted yields from selected cutting prescriptions 
by forest type and tree class, Northern Pine Unit, Minnesota, 1977-1986 



LONG ROTATION AGE OPTION 









Number 




















growing- 












1/ 








stock 




Pole- 


Saw- 






Other 


Forest 




Harvest 


trees 


Total 


timber 


timber 


Short- log 


cull 


type 




area 


harvested 


volume 


volume 


volume 


vo 


lume 


volume 






Thousand 


Thousand 


















acres 


trees 






nd cubic 


feet 






Jack pine 




6.1 


1,068 


9,394 


2,509 


6,563 




201 


121 


Red pine 




1.0 


103 


1,920 


184 


1,635 




61 


40 


White pine 




0.2 


40 


657 


94 


550 




3 


10 


Bal sam fir 




4.3 


589 


2,691 


1,609 


903 




93 


81 


White spruce 




0.2 


36 


338 


49 


257 




11 


21 


Black spruce 




2.4 


703 


1,914 


1,494 


365 







55 


Northern white- 


cedar 


1.5 


424 


1,696 


521 


967 




86 


122 


Tamarack 




2.5 


525 


2,053 


1,418 


509 




26 


100 


Oak-hickory 




2.7 


349 


3,034 


900 


1,789 




276 


69 


Elm-ash-cottonwood 


3.9 


617 


4,490 


1,552 


2,157 




371 


410 


Mapl e-basswood 




5.2 


112 


1,350 


425 







683 


242 


Aspen 




49.5 


6,972 


71,802 


26,878 


34,771 


6 


,024 


4,129 


Paper birch 




8.7 


1,582 


12,113 


5,412 


5,601 




630 


470 


Balsam poplar 




6.2 


743 


6,857 


2,352 


3,956 




329 


220 


All types 




94.4 


13,863 


120,309 


45,397 


60,023 


8 


,799 


6,090 








SHORT ROTP 


OPTION 










Jack pine 




8.2 


1,513 


12,653 


3,827 


8,326 




307 


193 


Red pine 




1.6 


128 


2,915 


231 


2,533 




107 


44 


White pine 




0.3 


46 


908 


38 


855 




10 


5 


Bal sam fir 




4.8 


630 


2,929 


1,721 


1,015 




110 


83 


White spruce 




0.3 


53 


430 


70 


328 




11 


21 


Black spruce 




4.2 


1,156 


3,331 


2,537 


670 




16 


108 


Northern white- 


cedar 


3.2 


752 


3,217 


892 


1,825 




292 


208 


Tamarack 




3.2 


699 


2,301 


1,766 


356 




31 


148 


Oak-hickory 




6.2 


828 


6,392 


2,075 


3,620 




533 


164 


Elm-ash-cottonwood 


6.0 


835 


6,338 


2,415 


3,001 




440 


482 


Mapl e-basswood 




5.2 


112 


1,350 


425 







683 


242 


Aspen 




70.8 


10,273 


105,099 


41,048 


48,978 


8 


,848 


6,225 


Paper birch 




10.8 


1,841 


14,641 


6,640 


6,423 




837 


741 


Balsam poplar 




6.3 


721 


7,656 


2,479 


4,533 




352 


292 


All types 




131.1 


19,587 


170,160 


66,164 


82,463 


12 


,577 


8,956 



—Rough and rotten cull. 



11 



Table 5. --Ten-year harvest area from selected cutting prescriptions by forest type, 
site index class, and stand-age class, northern Minnesota, 1977-1986 

(In thousand acres) 



LONG ROTATION AGE OPTION 





Site 
























Forest 


index 
class 


Total 








Sta 


nd-age cl 


ass (ye 


ars) 








type 


1-40 


41-50 


51-60 


61-70 


71-80 


81-90 


91-100 


101-120 


121-140 


141+ 


Jack pine 


0-60 


64.1 








7.8 


22.9 


7.4 


12.4 


11.8 


1.8 










61+ 


27.9 








20.1 


7.8 




















Red pine 


0-55 


8.9 

















4.4 


2.0 


2.5 










56+ 


13.4 








2.2 


2.1 





1.1 


8.0 











White pine 


0-55 


3.7 




















1.6 


2.1 










56+ 


1.6 




















1.6 











Balsam fir 


All 


165.1 








27.5 


80.3 


35.8 


9.7 


9.2 


2.6 








White spruce 


All 


10.0 











2.5 


1.2 


2.4 


2.8 


1.1 








Black spruce 


0-40 


54.4 

















8.7 


21.1 


12.5 


12.1 







41 + 


29.0 











1.8 


8.4 


9.5 


7.8 


1.5 








Northern 


























white-cedar 


All 


46.3 


























43.3 


3.0 


Tamarack 


0-40 


20.9 























2.9 


16.8 


1.2 




41+ 


20.0 











1.9 


3.1 





0.7 


9.8 


4.5 





Oak-hickory 


0-55 


17.2 





0.3 











1.4 


9.6 


5.9 










56+ 


10.4 


n 





0.4 


4.4 


2.8 


1.4 





1.4 








Elm-ash-cottonwood 


0-55 


55.9 





n 














17.9 


25.8 


12.2 







56+ 


16.9 

















1.7 


7.9 


3.4 


3.9 





Maple-basswood 


All 


73.7 




















24.5 


20.5 


28.7 





Aspen 


0-65 


504.9 





(J 


177.9 


209.5 


80.9 


26.9 


5.3 


3.0 


1.4 







66+ 


399.1 








229.5 


104.4 


32.9 


21.8 


9.1 


1.4 








Paper birch 


0-65 


179.3 











71.9 


61.7 


34.9 


5.5 


5.3 










66+ 


29.3 








1.1 


18.0 


6.3 


1.2 


2.7 











Balsam poplar 


0-65 


76.6 





1.9 


26.1 


24.4 


20.1 


2.8 


1.3 













66+ 


30.4 








10.4 


12.2 


5.1 


2.7 














All types 




1,859.0 





2.2 


503.0 


564.1 


265.7 


143.0 


150.4 


103.5 


122.9 


4.2 










SHORT 


ROTATION 


AGE OPTION 












Jack pine 


All 


122.2 








47.5 


41.3 


7.4 


12.4 


11.8 


1.8 








Red pine 


All 


32.4 

















14.3 


15.6 


2.5 








White pine 


All 


7.2 




















2.8 


4.4 








Balsam fir 


All 


183.8 








41.3 


85.2 


35.8 


9.7 


9.2 


2.6 








White spruce 


All 


12.3 








1.7 


3.0 


1.2 


2.4 


2.9 


1.1 








Black spruce 


All 


140.1 











4.5 


14.1 


45.2 


50.2 


14.0 


12.1 





Northern 


























white-cedar 


All 


99.7 




















0.5 


30.2 


66.0 


3.0 


Tamarack 


All 


51.2 

















1.0 


1.4 


23.8 


23.8 


1.2 


Oak-hickory 


All 


63.3 








1.3 


12.8 


25.3 


7.0 


9.6 


7.3 








Elm-ash-cottonwood 


All 


108.4 

















21.6 


41.6 


29.1 


16.1 





Mapl e-basswood 


All 


73.7 




















24.5 


20.5 


28.7 





Aspen 


All 


1,282.9 





18.0 


765.2 


316.8 


114.0 


48.8 


14.3 


4.4 


1.4 





Paper birch 


All 


256.1 








28.1 


110.3 


68.1 


36.1 


8.2 


5.3 








Balsam poplar 


All 


114.0 





1.9 


39.9 


37.5 


27.8 


5.5 


1.4 











All types 




2,547.3 





19.9 


925.0 


611.4 


293.7 


204.0 


194.0 


147.0 


148.1 


4.2 



12 



Table 6. --Ten-year harvest area from selected cutting prescriptions by forest type, 
site index class, and stand-age class, Aspen-Birch Unit, Minnesota, 1977-1986 

(In thousand acres) 



LONG ROTATION AGE OPTION 





Site 
























Forest 


index 

class 


Total 








Sta 


nd-age cl 


ass (years) 








type 


1-40 


41-50 


51-60 


61-70 


71-80 


81-90 


91-100 


101-120 


121-140 


141+ 


Jack pine 


0-60 


28.1 











2.2 


1.0 


12.4 


10.7 


1.8 










61 + 


3.1 








2.0 


1.1 




















Red pine 


0-55 


6.4 

















4.4 


0.9 


1.1 










56+ 


5.4 


n 





2.2 


2.1 





1.1 














White pine 


0-55 


2.7 




















1.6 


1.1 










56+ 


0.3 




















0.3 











Balsam fir 


ALL 


121.7 








27.5 


50.8 


24.9 


8.1 


9.2 


1.2 








White spruce 


ALL 


7.6 











1.3 


1.2 


1.2 


2.8 


1.1 








Black spruce 


0-40 


37.2 




















17.1 


10.9 


9.2 







41 + 


22.6 














6.9 


7.9 


7.8 











Northern 


























white-cedar 


ALL 


30.9 


























30.9 





Tamarack 


0-40 


7.8 























2.9 


3.7 


1.2 




41 + 


7.7 











1.9 


3.1 








1.4 


1.3 





Oak-hickory 


0-55 


0.3 





0.3 




























56+ 


0.4 








0.4 























Elm-ash-cottonwood 


0-55 


29.5 




















10.5 


12.3 


6.7 







56+ 


4.3 

















1.7 








2.6 





Maple-basswood 


ALL 


21.9 























12.9 


9.0 





Aspen 


0-65 


267.3 








88.7 


111.8 


46.0 


16.8 


2.7 


1.3 










66+ 


141.4 








72.4 


38.5 


15.8 


9.7 


5.0 











Paper birch 


0-65 


107.5 











34.6 


45.3 


23.9 


2.4 


1.3 










66+ 


13.9 








1.1 


6.6 


4.9 





1.3 











Bal sam poplar 


0-65 


30.9 








7.1 


11.1 


8.6 


2.8 


1.3 













66+ 


13.6 








4.3 


4.2 


3.8 


1.3 














All types 




912.5 





0.3 


205.7 


266.2 


161.5 


91.3 


73.6 


49.3 


63.4 


1.2 










SHORT 


ROTATION 


AGE OPTION 












Jack pine 


All 


39.8 











13.9 


1.0 


12.4 


10.7 


1.8 








Red pine 


All 


16.3 

















14.3 


0.9 


1.1 








White pine 


All 


3.9 




















2.8 


1.1 








Balsam fir 


All 


135.6 








41.3 


50.9 


24.9 


8.1 


9.2 


1.2 








White spruce 


All 


9.4 








1.7 


1.3 


1.2 


1.2 


2.9 


1.1 








Black spruce 


All 


98.6 














5.9 


26.4 


46.2 


10.9 


9.2 





Northern 


























white-cedar 


All 


67.7 




















0.5 


25.8 


41.4 





Tamarack 


All 


19.1 

















1.0 


1.4 


10.5 


5.0 


1.2 


Oak-hickory 


All 


1.3 








1.3 























Elm-ash-cottonwood 


All 


48.9 

















8.1 


19.3 


12.2 


9.3 





Mapl e-basswood 


All 


21.9 























12.9 


9.0 





Aspen 


All 


574.5 








324.1 


153.1 


61.8 


26.5 


7.7 


1.3 








Paper birch 


All 


147.8 








10.6 


58.0 


50.3 


23.9 


3.7 


1.3 








Balsam poplar 


All 


51.4 








18.3 


15.2 


12.4 


4.1 


1.4 











All types 




1,236.2 








397.3 


292.4 


157.5 


126.0 


106.7 


81.2 


73.9 


1.2 



13 



Table 7. --Ten-year harvest area from selected cutting prescriptions by forest type, 
site index class, and stand-age class, Northern Pine Unit, Minnesota, 1977-1986 

(In thousand acres) 



LONG ROTATION AGE OPTION 





Site 


























Forest 


index 
class 


Total 










Star 


d-age cl 


ass (yea 


rs) 








type 


1-40 


41-50 


51- 


50 


61-70 


71-80 


81-90 


91-100 


101-120 


121-140 


141+ 


Jack pine 


0-60 


36.0 








7 


.8 


20.7 


6.4 





1.1 













61 + 


24.8 








18 


.1 


6.7 




















Red pine 


0-55 


2.5 






















1.1 


1.4 










56+ 


8.0 






















8.0 











White pine 


0-55 


1.0 

























1.0 










56+ 


1.3 






















1.3 











Balsam fir 


All 


43.4 













29.5 


10.9 


1.6 





1.4 








White spruce 


All 


2.4 













1.2 





1.2 














Black spruce 


0-40 


17.2 



















8.7 


4.0 


1.6 


2.9 







41 + 


6.4 













1.8 


1.5 


1.6 





1.5 








Northern 




























white-cedar 


All 


15.4 




























12.4 


3.0 


Tamarack 


0-40 


13.1 




























13.1 







41 + 


12.3 






















0.7 


8.4 


3.2 





Oak-hickory 


0-55 


16.9 



















1.4 


9.6 


5.9 










56+ 


10.0 













4.4 


2.8 


1.4 





1.4 








Elm- ash- cottonwood 


0-55 


26.4 






















7.4 


13.5 


5.5 







56+ 


12.6 






















7.9 


3.4 


1.3 





Maple-basswood 


All 


51.8 






















24.5 


7.6 


19.7 





Aspen 


0-65 


237.6 








89 


.2 


97.7 


34.9 


10.1 


2.6 


1.7 


1.4 







66+ 


257.7 








157 


.1 


65.9 


17.1 


12.1 


4.1 


1.4 








Paper birch 


0-65 


71.8 













37.3 


16.4 


11.0 


3.1 


4.0 










66+ 


15.4 













11.4 


1.4 


1.2 


1.4 











Balsam poplar 


0-65 


45.7 





1.9 


19 


.0 


13.3 


11.5 



















66+ 


16.8 








6 


.1 


8.0 


1.3 


1.4 














All types 




946.5 





1.9 


297 


.3 


297.9 


104.2 


51.7 


76.8 


54.2 


59.5 


3.0 










SHORT 


ROTATION 


AGE OPTION 












Jack pine 


All 


82.4 








47 


.5 


27.4 


6.4 





1.1 











Red pine 


All 


16.1 






















14.7 


1.4 








White pine 


All 


3.3 

























3.3 








Balsam fir 


All 


48.2 













34.3 


10.9 


1.6 





1.4 








White spruce 


All 


2.9 













1.7 





1.2 














Black spruce 


All 


41.5 













4.5 


8.2 


18.8 


4.0 


3.1 


2.9 





Northern 




























white-cedar 


All 


32.0 

























4.4 


24.6 


3.0 


Tamarack 


All 


32.1 

























13.3 


18.8 





Oak-hickory 


All 


62.0 













12.8 


25.3 


7.0 


9.6 


7.3 








Elm-ash-cottonwood 


All 


59.5 



















13.5 


22.3 


16.9 


6.8 





Maple-basswood 


All 


51.8 






















24.5 


7.6 


19.7 





Aspen 


All 


708.4 





18.0 


441 


.1 


163.7 


52.2 


22.3 


6.6 


3.1 


1.4 





Paper birch 


All 


108.3 








17 


.5 


52.3 


17.8 


12.2 


4.5 


4.0 








Balsam poplar 


All 


62.6 





1.9 


21 


.6 


22.3 


15.4 


1.4 














All types 




1,311.1 





19.9 


527 


.7 


319.0 


136.2 


78.0 


87.3 


65.8 


74.2 


3.0 



14 



Table 8. --Average annual predicted yields from selected cutting prescriptions 
by species group and tree class, northern Minnesota, 1977-1986 



LONG ROTATION AGE OPTION 





G 


rowing stock 






C 


ull 






Species 




Pole- 


Saw- 








1/ 


Saw- 


group 


Total 


timber 


timber 


Toti 


Shortlog 


Other 


timber 




















Thousand 




















board„. 
feet — 








Thousand 


cub 


ic feet 








SOFTWOODS: 










White pine 


5,444 


107 


5,337 




195 




172 


23 


36,267 


Red pine 


9,404 


595 


8,809 




58 




44 


14 


53,498 


Jack pine 


14,083 


2,842 


11,241 




480 




338 


142 


57,737 


White spruce 


8,099 


2,314 


5,785 




64 




48 


16 


27,239 


Black spruce 


15,924 


10,166 


5,758 




177 




20 


157 


26,569 


Balsam fir 


21,464 


13,022 


8,442 




612 




259 


353 


39,201 


Tamarack 


3,778 


2,255 


1,523 




302 




97 


205 


6,586 


Northern white-cedar 


9,787 


2,554 


7,233 


1 


,613 


1 


,237 


376 


39,587 


Other softwoods 



























Total 


87,983 


33,855 


54,128 


3 


,501 


2 


,215 


1,286 


286,684 


HARDWOODS: 




















White oak 


2,616 


1,206 


1,410 




235 




142 


93 


7,337 


Select red oak 


4,686 


1,730 


2,956 




583 




416 


16 7 


14,781 


Other red oak 



























Hickory 


16 


8 


8 















41 


Yel low birch 


140 


56 


84 




116 




87 


29 


392 


Hard maple 


1,327 


99] 


336 


1 


,102 




735 


36/ 


1,925 


Soft maple 


1,337 


942 


395 




661 




288 


3 73 


1,938 


Ash 


11,360 


7,354 


4,006 




710 




265 


445 


22,874 


Balsam poplar 


15,493 


7,141 


8,352 




890 




SOS 


385 


43,021 


Paper birch 


32,070 


23,591 


8,479 


2 


,963 


1 


,356 


1,607 


43,791 


Bigtooth aspen 


4,109 


1,866 


2,243 




616 




392 


224 


10,627 


Quaking aspen 


86,253 


36,593 


49,660 


13 


,330 


8 


,971 


4,359 


265,170 


Basswood 


2,428 


1,344 


1,084 




346 




168 


178 


5,410 


Elm 


4,027 


981 


3,046 




447 




355 


92 


17,172 


Select hardwoods 


28 


6 


22 




12 







12 


108 


Other hardwoods 


21 


21 







27 




5 


22 





Noncommercial species 













58 







58 





Total 


165,911 


83,830 


82,081 


22 


,096 


13 


,685 


8,411 


434,587 


All species 


253,894 


117,685 


136,209 


25 


,597 


15 


,900 


9,697 


721,271 


SHORT ROTATION AGE OPTION 


SOFTWOODS: 




















White pine 


6,978 


346 


6,632 




230 




202 


28 


44,928 


Red pine 


11,317 


542 


10,775 




70 




62 


8 


66,228 


Jack pine 


20,183 


5,522 


14,661 




664 




484 


180 


75,472 


White spruce 


9,343 


2,726 


6,617 




81 




48 


33 


31,348 


Black spruce 


20,887 


14,518 


6,369 




252 




21 


231 


29,537 


Balsam fir 


28,469 


17,323 


11,146 




808 




394 


414 


51,479 


Tamarack 


4,332 


2,863 


1,469 




406 




115 


291 


6,338 


Northern white-cedar 


14,964 


3,727 


11,237 


2 


,390 


1 


,817 


573 


61,891 


Other softwoods 


23 


23 







3 







3 





Total 


116,496 


47,590 


68,906 


4 


,904 


3 


,143 


1,761 


367,221 


HARDWOODS: 




















White oak 


3,944 


1,852 


2,092 




410 




254 


156 


10,872 


Select red oak 


7,415 


3,020 


4,395 




880 




645 


235 


22,032 


Other red oak 


78 


67 


11 




3 







3 


54 


Hickory 


16 


8 


8 















41 


Yellow birch 


206 


63 


143 




139 




102 


37 


701 


Hard maple 


1,783 


1,358 


425 


1 


,211 




7 70 


441 


2,430 


Soft maple 


1,838 


1,295 


543 




844 




338 


506 


2,676 


Ash 


14,003 


9,155 


4,848 




877 




323 


554 


27,792 


Balsam poplar 


17,607 


8,386 


9,221 


1 


,150 




619 


531 


47,548 


Paper birch 


41,822 


31,425 


10,397 


4 


,226 


1 


,909 


2,317 


53,937 


Bigtooth aspen 


5,921 


2,935 


2,986 




774 




466 


308 


14,075 


Quaking aspen 


117,574 


52,181 


65,393 


18 


,757 


12 


,197 


6,560 


349,142 


Basswood 


3,657 


2,225 


1,432 




505 




243 


262 


7,091 


Elm 


4,922 


1,226 


3,696 




565 




454 


111 


20,648 


Select hardwoods 


40 


12 


28 




18 







18 


140 


Other hardwoods 


35 


35 







28 




6 


22 





Noncommercial species 













69 







69 





Total 


220,861 


115,243 


105,618 


30 


,456 


18,3 


12,130 


559,179 


All species 


337,357 


162,833 


174,524 


35 


,360 


21 


,469 


13,891 


926,400 



-Rough and rotten cull. 

2/ 

-International 1/4-inch rule. 



15 



Table 9. --Average annual predicted yields from selected cutting prescriptions 
by species group and tree class, Aspen-Birch Unit, Minnesota, 1977-1986 



LONG ROTATION AGE OPTION 





Growing stock 








Ci 


ill 






Species 




Pole- 


Saw- 










1/ 


Saw- 


group 


Total 


timber 


timber 


Tote 


Shortlog 


Other 


timber 




















Thousand 




















board„. 
feet -' 








ibic 


feet 








SOFTWOODS: 




















White pine 


3,690 


54 


3,636 




101 




84 


17 


26,092 


Red pine 


4,383 


388 


3,995 




27 




13 


14 


26,805 


Jack pine 


4,671 


713 


3,958 




122 




87 


35 


21,366 


White spruce 


6,875 


2,124 


4,751 




53 




43 


10 


21,775 


Black spruce 


13,049 


7,948 


5,101 




108 




6 


102 


23,890 


Balsam fir 


14,876 


9,555 


5,321 




452 




219 


233 


26,097 


Tamarack 


1,562 


891 


671 




141 




42 


99 


2,441 


Northern white-cedar 


5,760 


1,405 


4,355 




887 




734 


153 


25,448 


Other softwoods 



























Total 


54,866 


23,078 


31,788 


1 


891 


1 


228 


663 


173,914 


HARDWOODS: 




















White oak 


101 


12 


89 




22 




10 


12 


369 


Select red oak 


123 


50 


73 




13 




11 


2 


353 


Other red oak 



























Hickory 



























Yellow birch 


102 


25 


77 




67 




57 


10 


355 


Hard maple 


343 


287 


56 




328 




212 


116 


355 


Soft maple 


579 


393 


186 




372 




163 


209 


907 


Ash 


7,120 


4,526 


2,594 




214 




89 


125 


15,740 


Balsam poplar 


10,133 


5,007 


5,126 




587 




386 


201 


26,814 


Paper birch 


19,887 


15,295 


4,592 


1 


504 




853 


651 


24,507 


Bigtooth aspen 


542 


226 


316 




67 




38 


29 


1,692 


Quaking aspen 


52,775 


22,987 


29,788 


5 


,403 


3 


880 


1,523 


161,787 


Basswood 


272 


120 


152 




49 




37 


12 


800 


Elm 


1,631 


282 


1,349 




166 




137 


29 


7,784 


Select hardwoods 



























Other hardwoods 













1 







1 





Noncommercial species 













24 







24 





Total 


93,608 


49,210 


44,398 


8 


,817 


5 


873 


2,944 


241,463 


All species 


148,474 


72,288 


76,186 


10 


,708 


7 


101 


3,607 


415,377 






SHORT ROTATION AGE 


OPTION 










SOFTWOODS: 




















White pine 


4,595 


275 


4,320 




83 




69 


14 


31,462 


Red pine 


4,444 


288 


4,156 




25 




23 


2 


29,221 


Jack pine 


7,205 


2,202 


5,003 




122 




80 


42 


27,396 


White spruce 


7,713 


2,477 


5,236 




65 




43 


22 


24,037 


Black spruce 


16,665 


11,235 


5,430 




168 




15 


153 


25,681 


Balsam fir 


19,335 


12,645 


6,690 




597 




303 


294 


32,822 


Tamarack 


1,748 


1,102 


646 




168 




52 


116 


2,319 


Northern white-cedar 


9,437 


2,162 


7,275 


1 


,334 


1 


,076 


258 


42,501 


Other softwoods 



























Total 


71,142 


32,386 


38,756 


2 


,562 


~"T 


,661 


901 


215,439 


HARDWOODS: 




















White oak 


98 


31 


67 




21 




9 


12 


319 


Select red oak 


270 


121 


149 




36 




29 


7 


783 


Other red oak 



























Hickory 



























Yellow birch 


168 


32 


136 




92 




72 


20 


664 


Hard maple 


461 


405 


56 




341 




224 


117 


355 


Soft maple 


773 


556 


217 




473 




196 


277 


1,081 


Ash 


8,799 


5,671 


3,128 




293 




102 


191 


19,113 


Balsam poplar 


11,269 


5,727 


5,542 




702 




454 


248 


28,986 


Paper birch 


25,711 


19,986 


5,725 


2 


,009 


1 


,179 


830 


30,723 


Bigtooth aspen 


995 


580 


415 




103 




54 


49 


2,225 


Quaking aspen 


66,790 


30,516 


36,274 


6 


,930 


4 


,722 


2,208 


196,831 


Basswood 


427 


282 


145 




49 




37 


12 


754 


Elm 


1,827 


376 


1,451 




181 




152 


29 


8,379 


Select hardwoods 



























Other hardwoods 













2 




1 


1 





Noncommercial species 













33 







33 





Total 


117,588 


64,283 


53,305 


11 


,265 


7 


,231 


4,034 


290,213 


All species 


188,730 


96,669 


92,061 


13 


,827 


8 


,892 


4,935 


505,652 



1/ 



Rough and rotten cull. 

2/ 

— International 1/4-inch rule. 



16 



Table 10. --Average annual predicted yields from selected cutting prescriptions 
by species group and tree class, Northern Pine Unit, Minnesota, 1977-1986 



LONG ROTATION AGE OPTION 





G 


rowing stock 






Cull 






Species 




Pole- 


Saw- 






1/ 


Saw- 


Group 


Total 


timber 


timber 


Total 


Shortlog 


Other 


timber 
















Thousand 
















board- , 
feet ' 






Th 


ousand ci 


ibic feet 






SOFTWOODS: 








White pine 


1,754 


53 


1,701 


94 


88 


6 


10,175 


Red pine 


5,021 


207 


4,814 


31 


31 





26,693 


Jack pine 


9,412 


2,129 


7,283 


358 


251 


107 


36,371 


White spruce 


1,224 


190 


1,034 


11 


5 


6 


5,464 


Black spruce 


2,875 


2,218 


657 


69 


14 


55 


2,679 


Balsam fir 


6,588 


3,467 


3,121 


L60 


4i) 


120 


13,104 


Tamarack 


2,216 


1,364 


852 


161 


55 


106 


4,145 


Northern white-cedar 


4,027 


1,149 


2,878 


726 


50 3 


22 3 


14,139 


Other softwoods 























Total 


33,117 


10,777 


22,340 


1,610 


987 


623 


112,770 


HARDWOODS: 
















White oak 


2,515 


1,194 


1,321 


213 


132 


81 


6,968 


Select red oak 


4,563 


1,680 


2,883 


570 


405 


If, 6 


14,428 


Other red oak 

















CI 





Hickory 


L6 


8 


8 


(J 








41 


Yellow birch 


38 


31 


7 


49 


30 


19 


37 


Hard maple 


QH4 


704 


280 


774 


523 


251 


1,570 


Soft maple 


758 


549 


209 


289 


125 


164 


1,031 


Ash 


4,240 


2,828 


1,412 


496 


1/6 


320 


7,134 


Balsam poplar 


5,360 


2,134 


3,226 


303 


119 


1 84 


16,207 


Paper birch 


12,183 


8,296 


3,887 


1,459 


503 


956 


19,284 


Bigtooth aspen 


3,567 


1,640 


1,927 


549 


354 


195 


8,935 


Quaking aspen 


33,478 


13,606 


19,872 


7,927 


5,091 


2,836 


103,383 


Basswood 


2,156 


1,224 


932 


297 


131 


166 


4,610 


Elm 


2,396 


699 


1,697 


281 


218 


6 3 


9,388 


Select hardwoods 


28 


6 


22 


12 





12 


108 


Other hardwoods 


21 


21 





26 


5 


21 





Noncommercial species 











34 





34 





Total 


72,303 


34,620 


37,683 


13,279 


7,812 


5,467 


193,124 


All species 


105,420 


45,397 


60,023 


14,889 


8,799 


6,090 


305,894 


SHORT ROTATION AGE OPTION 


SOFTWOODS: 
















White pine 


2,383 


71 


2,312 


14/ 


133 


14 


13,466 


Red pine 


6,873 


254 


6,619 


45 


39 


6 


37,007 


Jack pine 


12,978 


3,320 


9,658 


54? 


404 


138 


48,076 


White spruce 


1,630 


249 


1,381 


16 


5 


11 


7,311 


Black spruce 


4,222 


3,283 


939 


84 


6 


78 


3,856 


Balsam fir 


9,134 


4,678 


4,456 


21] 


91 


120 


18,657 


Tamarack 


2,584 


1,761 


823 


238 


63 


176 


4,019 


Northern white-cedar 


5,527 


1,565 


3,962 


1,056 


741 


315 


19,390 


Other softwoods 


23 


23 





3 





3 





Total 


45,354 


15,204 


30,150 


2,343 


1,482 


860 


151,782 


HARDWOODS: 
















White oak 


3,846 


1,821 


2,025 


389 


245 


144 


10,553 


Select red oak 


7,145 


2,899 


4,246 


844 


616 


228 


21,249 


Other red oak 


78 


67 


11 


3 





3 


54 


Hickory 


16 


8 


8 





(1 





41 


Yellow birch 


38 


31 


7 


4/ 


30 


17 


37 


Hard maple 


1,322 


953 


369 


870 


546 


324 


2,075 


Soft maple 


1,065 


739 


326 


371 


142 


229 


1,595 


Ash 


5,204 


3,484 


1,720 


584 


221 


363 


8,679 


Balsam poplar 


6,338 


2,659 


3,679 


448 


165 


283 


18,562 


Paper birch 


16,111 


11,439 


4,672 


2,217 


730 


1,487 


23,214 


Bigtooth aspen 


4,926 


2,355 


2,571 


671 


412 


259 


11,850 


Quaking aspen 


50,784 


21,665 


29,119 


11,827 


7,475 


4,352 


152,311 


Basswood 


3,230 


1,943 


1,287 


456 


206 


250 


6,337 


Elm 


3,095 


850 


2,245 


384 


302 


82 


12,269 


Select hardwoods 


40 


12 


28 


18 





18 


140 


Other hardwoods 


35 


3 b 





26 


5 


21 





Noncommercial species 











36 





36 





Total 


103,273 


50,960 


52,313 


19,191 


11,095 


8,096 


268,966 


All species 


148,627 


66,164 


82,463 


21,533 


12,577 


8,956 


420,748 



— Rough and rotten cull. 

2/ 

— International 1/4-inch rule. 



17 



Table 11. --Average annual predicted yields of growing-stock from selected cutting prescriptions 
by species group and forest type, northern Minnesota, 1977-1986 

(In thousand cubic feet) 



LONG ROTATION AGE OPTION 



Species 
group 



Total 



Northern 
Jack Red White Balsam White Black white- Tama- 
pine pine pine fir spruce spruce cedar rack 



Elm-ash- 
Oak- cotton- 
hickory wood 



Maple- 

bass- 

wood Aspen 



Paper 
birch 



Bal sam 
poplar 



SOFTWOODS: 
White pine 
Red pine 
Jack pine 
White spruce 
Black spruce 
Balsam fir 
Tamarack 
Northern 
white-cedar 
Other softwoods 
Total 



5,444 

9,404 
14,083 

8,099 
15,924 
21,464 

3,778 

9,787 



87,983 



48 

563 

10,161 

37 

1,229 

119 





616 

3,554 

10 

15 

71 

200 





672 

140 


111 


111 





427 

226 

137 

1,287 

2,606 

5,715 

282 



18 


12,157 i,m~ 



1,072 


l,W 



11,752 



63 

28 

2 

598 

1 

102 

1 

107 

0_ 

~90"T 



119 
63 
143 
107 
7,186 
203 
350 

289 



8,460 



40 
25 

23 
135 
68 
83 

3,368 


3,742 







16 



306 

31 

2,572 

85 

0_ 

3,010 



21 

143 



6 



11 










64 

11 

184 

7 



TBI 



1,427 









55 







2,990 
3,254 
3,407 
4,769 
4,129 
10,038 
171 

1,990 



30,748 



431 

1,408 

188 

884 

167 

2,999 



999 



7,076 



17 


19 

198 

83 

1,328 

312 

432 



2,389 



HARDWOODS: 
White oak 
Select red oak 
Other red oak 
Hickory 
Yel low birch 
Hard maple 
Soft maple 
Ash 

Balsam poplar 
Paper birch 
Bigtooth aspen 
Quaking aspen 
Basswood 
Elm 

Select hardwood 
Other hardwoods 



2,616 

4,686 



16 

140 

1,327 

1,337 



11,360 

15,493 

32,070 

4,109 

86,253 

2,428 

4,027 

28 

21 



5 
26 








34 
668 

69 

1,089 

4 

9 



16 













73 
44 
182 













2 


101 



















29 

206 

238 

771 

,303 



978 

8 

11 







3 











1 

206 


b4 
















25 

12 

144 

6 

76 


11 










24 



205 
11 

169 



7 












5 
14 
16 

6 







561 

1,252 







20 

46 

5 b 

6 

143 

64 

265 

133 

44 



5 



30 







8 



148 

,941 

87 

368 

57 

119 

121 

862 







39 

6 







107 

14 

170 

6 

65 



19 

28 

60 







1,781 

2,914 





20 

810 

737 

5,550 

8,545 

14,904 

3,731 

78,197 

1,626 

2,155 

28 





157 


40 


432 


56 








16 





65 


23 


310 


51 


184 





1,099 


1,072 


421 


5,585 


13,124 


786 


118 


20 


3,834 


1,424 


414 


94 


448 


420 














20,622 


9,571 



Total 


165,911 


1,920 


299 


103 


3,544 274 ?74 416 41 2,594 


4,741 


5 14 


120,998 


20,622 9,571 


All species 


253,894 


14,077 


4,783 


1,137 


15,296 1,176 8,734 4,158 3,051 2,775 


6,734 


569 


151,746 


27,698 11,960 


SHORT ROTATION AGE OPTION 



SOFTWOODS: 
White pine 
Red pine 
Jack pine 
White spruce 
Black spruce 
Balsam fir 
Tamarack 
Northern 
white-cedar 
Other softwoods 



,978 
,317 
,183 
,343 
,887 
,469 
,332 

,964 
23 



891 

14,776 

62 

1,599 

163 









959 

3,951 

229 

15 


1/4 







787 
461 


138 


96 





427 

226 

137 

1,288 

2,713 

7,074 

282 



76 

28 

26 
835 



38 

82 

221 

185 

20 10,815 

144 408 

10 642 



100 

25 



73 

463 

193 







16 



401 

35 



126 2,663 



1,647 




107 




395 
23 



6,834 




159 




128 
212 
180 
6 

9 
9 






63 



15 

78 

18 

861 

7 

,968 









55 







3,472 
3,887 
4,377 
5,620 
4,573 
14,040 
281 

2,281 




844 


36 


1,554 





187 


19 


894 


149 


210 


75 


3,512 


1,705 





312 



1,011 




562 




Total 


116,496 


17,539 


5,328 


1,482 


13 


,794 


1,246 


12 ,S09 


7,814 


3,274 


544 


3,010 


55 


38,531 


8,212 


2,858 


HARDWOODS: 


































White oak 


3,944 


8 













3 











1,258 


91 


39 


2,249 


246 


50 


Select red oak 


7,415 


43 

























2,516 





6 


4,146 


636 


68 


Other red oak 


78 





































78 








Hickory 


16 













(J 


























16 





Yellow birch 


206 






















24 








72 





20 


66 


24 


Hard maple 


1,783 













29 














19 





107 


1,228 


339 


61 


Soft maple 


1,838 













206 














71 


160 


14 


1,163 


219 


5 


Ash 


14,003 













255 





8 


272 


5 


89 


4,557 


170 


6,477 


1,149 


1,021 


Balsam poplar 


17,607 


40 










775 


12 


12 


36 


14 


6 


154 


6 


10,000 


459 


6,093 


Paper birch 


41,822 


723 


119 


70 


2 


,052 


213 


244 


483 


40 


389 


497 


65 


19,736 


16,355 


836 


Bigtooth aspen 


5,921 


103 


81 













6 








104 


57 





5,362 


188 


20 


Quaking aspen 


117,574 


1,682 


211 





1 


,135 


90 


142 


71 


6 


609 


337 


19 


106,605 


5,074 


1,593 


Basswood 


3,657 


4 










24 














222 


204 


28 


2,469 


591 


115 


Elm 


4,922 


9 


(J 







11 





11 


13 





64 


1,240 


60 


2,584 


459 


471 


Select hardwood 


40 




























12 








28 








Other hardwoods 


35 


16 

























5 








14 









1,861 2,628 



411 



70 4,487 



318 



423 



899 



65 5,364 7,369 514 162,159" 



All species 337,357 20,167 5,739 1,552 18,281 1,564 13,232 8,713 3,339 5,908 10,379 569 2007690" 



25,797 
34,009 



10,357 
131215 



18 



Table 12. --Average annual predicted yields of growing-stock from selected cutting prescriptions 
by species group and forest type, Aspen-Birch Unit, Minnesota, 1977-1986 

(In thousand cubic feet) 



LONG ROTATION AGE OPTION 



















Northern 




Elm-ash- 


Mapl e- 








Species 




Jack 


Red 


White 


Bal sam 


White 


Black 


white- 


Tama- 


Oak 


- cotton- 


bass- 




Paper 


Bal sam 


group 


Total 


pi ne 


pine 


pine 


fir 


sprues 


spruce 


cedar 


rack 


hicko 


ry wood 


wood 


Aspen 


birch 


poplar 


SOFTWOODS: 
































White pine 


3,690 


31 


411 


:i,<j 


427 


(1 


119 


39 











Li 


2,290 


1.04 





Red pine 


4,383 


15 


2,246 





308 


ii 


63 


83 


II 











1,074 


7 9,' 





Jack pine 


4,671 


2,316 


10 





137 


U 


143 








11 








2,036 


29 





White spruce 


6,875 


22 


IS 


in 


1,243 


31-: 


107 


14 








64 





3,972 


737 


72 


Black spruce 


13,049 


1,218 


71 





1,815 





5,602 


86 


98 











3,980 


134 


75 


Bal sam fir 


14,876 


29 


4,-: 


75 


5,023 


54 


143 


09 


31 


8 


32 3 


23 


6,526 


1,785 


752 


Tamarack 


1,562 











282 


LI 


3 74 


8 3 


4-;.) 





o 





12 





32 


Northern 
































white-cedar 


5,760 


n 


10 





584 


83 


313 


2,031 


71 





352 


ii 


1,775 


476 


158 


Other softwoods 















































Total 


54,866 


3,631 


2,819 


455 


9,719 


655 


6,663 


2,280 


1,109 


2 


739 


23 


21,665 


4,017 


1,089 


HARDWOODS: 
































White oak 


HI! 





n 




















8 


30 





51 


19 





Select red oak 


123 











li 











1) 


59 


o 





23 


o 


4 9 


Other red oak 





il 


11 


1) 


i) 


11 


II 





11 


o 

















Hickory 








n 


II 


il 


11 

















11 











Yel low birch 


102 














II 


11 


84 








8 





10 


37 


9 3 


Hard maple 


343 





1) 





'•) 


11 











II 





37 


72 


1/9 


26 


Soft maple 


579 








8 


191 


1) 














22 





250 


114 





Ash 


7,120 


II 








1 1 3 


11 


17 


1813 


9 


9 


1,451 


22 


4,301 


478 


541 


Bal sam popl ar 


10,133 





11 


I) 


699 





12 


11 





o 


54 





6,402 


252 


2,697 


Paper birch 


19,887 


'il'j 


11 


56 


1,168 


173 


101 


L60 


10 





153 


84 


8,933 


8,258 


354 


Bigtooth aspen 


542 





(1 











6 

















516 


14 


6 


Quaking aspen 


52,775 


859 


14 





858 


4n 


76 





o 


8 


110 


19 


46,844 


3,184 


632 


Basswood 


272 











8 





il 











44 





103 


71 


41 


Elm 


1,631 








II 


il 


1) 


il 


7 





2 


404 


14 


92 / 


69 


1 98 


Select hardwooc 


s 


II 


1) 






































Other hardwoods 















































Total 


93,608 


1,374 


145 


38 


3,065 


215 


212 


390 


15 


84 


2,286 


121 


68 L 432 


12,668 


4,563 


All species 


148,474 


5,005 


2,964 


493 


12,784 


870 


6,875 


2,670 


1,124 


86 


1,089 


144 


90,097 


16,685 


5,652 












SHORT ROTATION AGE 


OPTION 
















SOFTWOODS: 
































White pine 


4,595 


31 


627 


379 


427 


13 


J8 


99 


o 





63 





2,511 


39h 


11 


Red pine 


4,444 


175 


1,900 





208 





83 


25 














1,302 


798 





Jack pine 


7,205 


4,387 


229 





137 





189 











19 


o 


2,219 


29 





White spruce 


7,713 


22 


lb 


13:-: 


1,244 


7 3,' 


185 


63 





11 


78 





4,416 


74,9 


72 


Black spruce 


16,665 


1,534 








1,818 


9 


8,174 


406 


1 19 





7 





4,326 


177 


75 


Balsam fir 


19,335 


29 


,'4 


96 


6,305 


81 


344 


1 9 7 


35 





353 


23 


8,820 


2,297 


771 


Tamarack 


1,748 





11 


11 


282 


:l 


44" 


34 


934 











12 





33 


Northern 
































white-cedar 


9,437 





11 





1,153 


83 


305 


4,512 


/h 





9b9 





2,066 


488 


,'85 


Other softwoods 















































Total 


71,142 


6,178 


2,795 


613 


11,574 


918 


9,666 


5,301 


1,184 





1,085" 


23 


25,672 


4,887 


1,246 


HARDWOODS: 
































White oak 


98 



































63 


35 





Select red oak 


270 





l) 








n 











186 








33 


16 


46 


Other red oak 





1) 








11 








11 











11 











Hickory 

















il 

















11 











Yel low birch 


IhH 





II 





11 








84 








72 





10 


38 


84 


Hard maple 


461 








(1 


29 




















37 


189 


1/9 


27 


Soft maple 


773 











L91 

















34 


(1 


417 


126 


5 


Ash 


8,799 





IJ 





113 








801 


5 





2,662 


22 


4,705 


999 


562 


Balsam poplar 


11,269 











699 


11 


13 


32 








121 





7,138 


270 


2,997 


Paper birch 


25,711 


47'-: 





46 


1,917 


1M3 


194 


376 


88 





391 


84 


11,246 


10,559 


405 


Bigtooth aspen 


995 

















6 

















969 


14 


6 


Quaking aspen 


66,790 


1,358 


181) 





1,015 


66 


143 


53 





80 


208 


19 


58,950 


4,125 


146 


Basswood 


427 











8 


il 














28 





250 


100 


41 


Elm 


1,827 




















7 








909 


19 


1,032 


69 


198 


Select hardwood 


s 








(J 



































Other hardwoods 















































Total 


117,588 


1 ,836 


180 


46 


3, '171 


248 


359 


655 


38 


213 


3,878 


181 


84,992 


16,059 


4,957 


All species 


188,730 


8,014 


2,975 


659 


15,545 


1,166 


10,025 


5,996 


1,217 


213 


4,963 


144 


110,664 


20,946 


6,203 



19 



Table 13. --Average annual predicted yields of growing-stock from selected cutting prescriptions 
by species group and forest type, Northern Pine Unit, Minnesota, 1977-1986 

(In thousand cubic feet) 



LONG ROTATION AGE OPTION 





















Northern 


Elm-ash- 


Maple- 










Species 




Jack 


Red 


•White 


White 


Black 


Bal sam 


Tama- 


whi te- 


Oak- 


cotton- 


bass- 




D 


iper 


Bal sam 


group 


Total 


pine 


pine 


pine 


spruce 


spruce 


fir 


rack 


cedar 


hickory wood 


wood 


Aspen 


b 


rch 


poplar 


SOFTWOODS: 


































White pine 


1,754 


17 


205 


403 


63 











1 


21 








700 




327 


17 


Red pine 


5,021 


548 


1,308 


140 


28 





13 


□ 





143 








2,180 




656 





Jack pine 


9,412 


7,845 








2 


n 





16 














1,371 




159 


19 


White spruce 


1,224 


15 








80 





44 





9 


6 








797 




147 


126 


Black spruce 


2,875 


11 








1 


1,584 


■"J I 


238 


49 





11 





149 




33 


8 


Balsam fir 


6,588 


90 


152 


36 


43 


60 


692 





6 


9 


161 


32 


3,512 


1 


214 


576 


Tamarack 


2,216 











1 


76 





1,633 


60 





7 





159 







280 


Northern 


































white-cedar 


4,027 











24 


77 


488 


14 


1,337 





1,075 





215 




523 


274 


Other softwoods 

















































Total 


33,117 


8,526 


1,665 


579 


247 


1,797 


2,033 


1,901 


1,462 


179 


1,254 


32 


9,083 


3 


,059 


1,300 


HARDWOODS: 


































White oak 


2,515 


5 





n 


3 














553 





39 


1,730 




145 


40 


Select red oak 


4,563 


26 

















il 





1,197 





6 


2,891 




432 


11 


Other red oak 

















































Hickory 


16 

















n 






















16 





Yellow birch 


33 

















n 





13 











10 




28 





Hard maple 


984 

















ii 





il 


20 





70 


738 




131 


25 


Soft maple 


758 





(J 











15 








46 


126 


14 


487 




70 





Ash 


4,240 














8 


126 





17 


50 


1,490 


148 


1,249 




621 


531 


Balsam poplar 


5,360 


34 








1 


3 


72 


14 








33 


6 


2,143 




169 


2,888 


Paper birch 


12,183 


153 


73 


65 


31 


43 


135 


6 


9 


143 


215 


41 


5,971 


4 


,866 


432 


Bigtooth aspen 


3,567 


69 


44 


ii 


13 








ii 





64 


57 





3,215 




104 


14 


Quaking aspen 


33,478 


231) 


3/ 





34 





120 


6 





257 


9 





31,353 




650 


792 


Basswood 


2,156 


4 























133 


72 


28 


1,523 




343 


53 


Elm 


2,396 


9 








1) 


11 


11 


1) 





42 


453 


41 


1,228 




379 


222 


Select hardwood 


s 28 



































28 










Other hardwoods 


21 


16 























5 



















Total 


72,303 


546 


154 


65 


59 


62 


479 


26 


26 


2,510 


2,455 


393 


52,566 


7 


,954 


5,008 


All species 


105,420 


9,072 


1,819 


644 


306 


1,859 


2,512 


1,927 


1,488 


2,689 


3,709 


425 


61,649 


11 


,013 


6,308 












SHORT ROTATION AGE 


OPTION 


















SOFTWOODS: 


































White pine 


2,383 


17 


332 


408 


63 











1 


128 








961 




448 


25 


Red pine 


6,873 


716 


2,051 


4b 1 


33 





18 








212 








2,585 




802 





Jack pine 


12,978 


10,389 








26 


33 


(i 


16 





180 








2,158 




158 


19 


White spruce 


1,630 


40 








103 





44 





10 


6 








1,204 




146 


77 


Black spruce 


4,222 


65 








11 


2,641 


895 


262 


57 





11 





247 




33 





Balsam fir 


9,134 


134 


150 





63 


64 


7-3) 





36 


9 


508 


32 


5,220 


1 


,215 


934 


Tamarack 


2,584 











10 


193 





1,729 


87 


9 


7 





269 







280 


Northern 


































white-cedar 


5,527 











24 


190 


494 


Si 


2,322 





1,399 





215 




523 


277 


Other softwoods 


23 














23 































Total 


45,354 


11,361 


2,533 


869 


328 


3,143 


2,220 


2,090 


2,513 


544 


1,925 


32 


12,859 


3 


,325 


1,612 


HARDWOODS: 


































White oak 


3,846 


8 


n 





3 














1,258 


91 


39 


2,186 




211 


50 


Select red oak 


7,145 


43 























2,331 





6 


4,123 




620 


22 


Other red oak 


78 
































i) 


78 










Hickory 


16 























1) 
















16 





Yellow birch 


38 



































10 




28 





Hard maple 


1,322 


























19 





70 


1,039 




160 


34 


Soft maple 


1,065 

















15 








71 


126 


14 


746 




93 





Ash 


5,204 


ii 











3 


143 


1) 


69 


89 


1,895 


148 


1,772 




621 


459 


Bal sam poplar 


6,338 


40 








12 





76 


14 


4 


6 


33 


6 


2,862 




189 


3,096 


Paper birch 


16,111 


245 


119 


24 


31 


45 


135 


12 


107 


389 


246 


41 


8,490 


5 


,796 


431 


Bigtooth aspen 


4,926 


103 


31 




















104 


57 





4,393 




174 


14 


Quaking aspen 


50,784 


324 


31 





34 





133 


6 


18 


581 


129 





47,655 




949 


947 


Basswood 


3,230 


4 














16 








222 


176 


28 


2,219 




491 


74 


Elm 


3,095 


9 











11 


11 





6 


64 


738 


41 


1,552 




390 


273 


Select hardwood 


s 40 


























12 








28 










Other hardwoods 


35 


16 























5 








14 










Total 


103 , ? 7 3 


792 


231 


24 


70 


64 


516 


32 


. 204 


5,151 


3,491 


393 


77,167 


9 


,738 


5,400 


All species 


148,627 


12,153 


2,764 


893 


398 


3,207 


2,736 


2,123 


2,717 


5,695 


3,4 lb 


425 


90,026 


13 


,063 


7,012 



20 



Table 14. --Growing-stock removals for 1976—' and average annual 
predicted yields from selected cutting prescriptions 
for 1977-1986 by species, northern Minnesota 

(In thousand cubic feet) 





1976 


Predicted yie 


Ids from selected 




Growing- 
stock 


cutting p 


rescriptions 


Species 


Long rotation 


Short rotation 


group 


removals 


age option 


age option 


SOFTWOODS: 








White pine 


2,724 


5,444 


6,978 


Red pine 


4,025 


9,404 


11,317 


Jack pine 


20,977 


14,083 


20,183 


White spruce 


4,153 


8,099 


9,343 


Black spruce 


13,733 


15,924 


20,887 


Bal sam fir 


13,768 


21,464 


28,469 


Tamarack 


4,629 


3,778 


4,332 


Northern white-cedar 


3,296 


9,787 


14,964 


Other softwoods 








23 


Total 


67,305 


87,983 


116,496 


HARDWOODS: 








White oak 


605 


2,616 


3,944 


Select red oak 


2,101 


4,686 


7,415 


Other red oak 








78 


Hickory 


5 


16 


16 


Yellow birch 


11 


140 


206 


Hard maple 


347 


1,327 


1,783 


Soft maple 


346 


1,337 


1,838 


Ash 


1,539 


11,360 


14,003 


Balsam poplar 


2,499 


15,493 


17,607 


Paper birch 


7,793 


32,070 


41,822 


Bigtooth aspen 


2,867 


4,109 


5,921 


Quaking aspen 


63,126 


86,253 


117,574 


Basswood 


853 


2,428 


3,657 


Elm 


1,257 


4,027 


4,922 


Select hardwoods 





28 


40 


Other hardwoods 


22 


21 


35 


Noncommercial species 











Total 


83,371 


165,911 


220,861 


All species 


150,676 


253,894 


337,357 



-1976 Growing-stock removals are trend removals for the period 
1962 to 1976. 



21 



Table 15.--Grov/ing-stock removals for 1976—' and average annual 
predicted yields from selected cutting prescriptions 
for 1977-1986 by species, Aspen-Birch Unit, Minnesota 

(In thousand cubic feet) 





1976 


Predicted yie 


Ids from selected 




Growing- 
stock 


cutting 


prescriptions 


Species 


Long rotation 


Short rotation 


group 


removals 


age option 


age option 


SOFTWOODS: 








White pine 


1,711 


3,690 


4,595 


Red pine 


1,485 


4,383 


4,444 


Jack pine 


10,718 


4,671 


7,205 


White spruce 


2,888 


6,875 


7,713 


Black spruce 


10,275 


13,049 


16,665 


Balsam fir 


7,728 


14,876 


19,335 


Tamarack 


2,503 


1,562 


1,748 


Northern white-cedar 


1,495 


5,760 


9,437 


Other softwoods 











Total 


38,803 


54,866 


71,142 


HARDWOODS: 








White oak 


10 


101 


98 


Select red oak 


113 


123 


270 


Other red oak 











Hickory 











Yellow birch 


3 


102 


168 


Hard maple 


28 


343 


461 


Soft maple 


77 


579 


773 


Ash 


302 


7,120 


8,799 


Balsam poplar 


1,044 


10,133 


11,269 


Paper birch 


3,253 


19,887 


25,711 


Bigtooth aspen 


665 


542 


995 


Quaking aspen 


32,686 


52,775 


66,790 


Basswood 


48 


272 


427 


Elm 


207 


1,631 


1,827 


Select hardwoods 











Other hardwoods 


9 








Noncommercial species 











Total 


38,445 


93,608 


117,588 


All species 


77,248 


148,474 


188,730 



—1976 Growing-stock removals are trend removals for the period 
1962 to 1976. 



22 



Table 16. --Growing-stock removals for 1976—' and average annual 
predicted yields for selected cutting prescriptions 
for 1977-1986 by species, Northern Pine Unit, Minnesota 

(In thousand cubic feet) 





1976 


Predicted yi 


elds from selected 




Growing- 
stock 


cutting 


prescriptions 


Species 


Long rotation 


Short rotation 


group 


removals 


age option 


age option 


SOFTWOODS: 








White pine 


1,013 


1,754 


2,383 


Red pine 


2,540 


5,021 


6,873 


Jack pine 


10,259 


9,412 


12,978 


White spruce 


1,265 


1,224 


1,630 


Black spruce 


3,458 


2,875 


4,222 


Balsam fir 


6,040 


6,588 


9,134 


Tamarack 


2,126 


2,216 


2,584 


Northern white-cedar 


1,801 


4,027 


5,527 


Other softwoods 








23 


Total 


28,502 


33,117 


45,354 


HARDWOODS: 








White oak 


595 


2,515 


3,846 


Select red oak 


1,988 


4,563 


7,145 


Other red oak 








78 


Hickory 


5 


16 


16 


Yellow birch 


8 


38 


38 


Hard maple 


319 


984 


1,322 


Soft maple 


269 


758 


1,065 


Ash 


1,237 


4,240 


5,204 


Balsam poplar 


1,455 


5,360 


6,338 


Paper birch 


4,540 


12,183 


16,111 


Bigtooth aspen 


2,202 


3,567 


4,926 


Quaking aspen 


30,440 


33,478 


50,784 


Basswood 


805 


2,156 


3,230 


Elm 


1,050 


2,396 


3,095 


Select hardwoods 





28 


40 


Other hardwoods 


13 


21 


35 


Noncommercial species 











Total 


44,926 


72,303 


103,273 


All species 


73,428 


105,420 


148,627 



—1976 Growing-stock removals are trend removals for the period 
1962 to 1976. 



23 



01 

a. 
>> 



to to 
+-> CO 

O en 



OJ 




E c 




o c 




S- -r- 




M- 2! 




o> c 


_^ 


E S- 


^J 


3 OJ 


OJ 


— -C 


OJ 


O 4-> 


H- 


> s_ 







tJ 


^ c 


■r- 



(_> 


.£3 


C3 


O » 


3 


< 


-l-> CO 


O 




to </) 




IT 


1 <o 


■O 


C 


oil— 


C 


> — 1 


C (J 


ro 


r— 


f— 


1/1 


< 


3 oj 


3 


I— 


en 


O 


O 


s- <n 


-C 


cc 


00 1 


■(-> 




T3 




3Z 


+-> C 


c 


CD 


00 ID 


M 


j— 1 


OJ +J 


1 ' 


X 



■a 




c;j 


[/) 


01 


t/1 


rTD 


10 


c 


, — 


m 


u 


E 






X 


1 — 





10 


T3 


3 


c 


C 


•r— 


EZ 




IS 


01 




4-> 


OJ 


•1— 


01 


f/l 


10 




(_ 




V 




> 




< 





o 
oj «^- 

S- i-H 

o 

E 

o 



+-> 1— 1 
■=3- 



x 10 

oj oj to 
+-> -a <o 

•1- C r— 
00 -r- O 



to 

ai a> 

s- o. 

o >, 

Lu. 4-> 



oooooooooor--o 



OOOOOOOtOOOi-HO 
CO tO to 

O CO >* 



r-.Or-.Oinococoi-HCOOO 
CO CO O CO CO «3" CM r--. 
cm ro .-h r-. cm cm en 



oONHrsinNWHi^o* 

"d"-— i in Cn CM CO ■—< CO o 

>* >* co ro co r- 1 en cn •— i 



iOroiooo<a-oina30o 

CTi CM Cn r~ O CO 

CO CO CM r-» to cn 



oooooooco^i-oin 

r*- CM un CO tO 

0~> ^h .— ( CM •— < 



Lono^oor-^otDCMO^a- 

"d" <3" r~» CM O «* CM 

r-. r-» ro co cm r-» cm 



cocMOoooooocnoo 

CO O in *3" 

Cn ^J" O f— t 



OOOOOOOOOOOO 



000000000000 



CMincM.— icMuntoO"3-tocoro 
ro^3-cMtO'd-cnr--councncMCM 

CTlt— HDr- lNfOHNCfl!\Jr- < C3~> 



I^OCDOOOOOOOOO^- 
ur"> o 



.— iOO^i—Ci-hCOOOOOO 
O OhNPI 

0> CO tO CM >— I 



owcornDN^tiriNOOo 

fHrHlOMNinOlO 

ro ro ^h in ■— 1 .— 1 r-- 



OOOCOtOt— ISHOIHHO 

CTi CTiCOCM>^-<^-LnC0r-~- 

r^~ incMCMtountotoi-H 



O^irjOiHOr^OicoMOioi 

co un co cn cn co to to >3- 

.-h co r-- id ro Cvj «t in 



co -=r un 



OOhOOOOihOMOOih 

co to ro en cn to un 

**• co ^- 00 r-~. to co 



oo«3-ooor~-~unintouno 
cri cr> cm un <z? to i-h 

tn co un in ro i-h 



cn o r-~ cm cm cm 

CM CM 



oocooootor-^Ocoun«3- 



un cn 

CM O 



r-H CM 

co «d- 



un to cn 

CM CM ro 



OCOOOOOOOOOr--0 



OOOOOOOOOOOO 



coroMO^aicocO'^q-m'a- 
uni— <tocootouncoun*3-uno 

H(\nr>ONlONOiHlOOOl 



co in .-h ro 



H<*IOl/lHH 



o + in + un + 0+ + 

tO ■— I tn tO IO tO 1 — <tHr-fl-H 
IO ILD I Uli — l<3-r— I rt 

o o o <c o <c o 



^O ^- CJ in r% ^t 
r-~ r-~ CM 



10+LD+ un + un + un + 
r— un to tn to 1 — totototototo 

1 — ILO ILOr— ItO ItO ItO 

<c o o < o o o 



o 
o 
3 
rz 

o 



T3 

0) 
OS 



a> ai 






<o 









■0 


3 3 


1- 




ai 


S_ i- 


■r~ 







O- Q. 


4- 


^ 


C 1 


l/> (/> 




U 


s- qj 




H 


"3 


a> +j 


c j*: 


fO 


S. 


.c •■- 


+-> 


to 


T3 


+-> -C 


1- T3 


1 — 


F 


S- 3 


JZ. r- 


03 


to 


O 


3 CQ 


CD 


I— 


z 



■a 

o 
o 

3 



1 



a> c 

Q. CL 

<a 00 



o o o o O O C~- 
«3- 



O O O O to O CM 
CO 00 

O *3" 



NNCOcoinajo 
r~- 00 o o to ^r i-h 
cm o ■— 1 00 cm r- 



O co «3- r«» to tn to 

CM CTl «d" CM CTi CO to 

iHNinro hoi 



r-. O O O O O O 
un 

CM 



to O to i-h CO o o 
■-h cm r~ ro 
CTi O-i CM ^h 



■— 1 •—< un •— 1 



CM CM 



>* O <J r-- CO "* 
to un oi co co ro 

Ol to CM «3" CTi 



.— I O O O 00 «3- CD 

>=3- cm un ro 

<T> i-H 1 — l CM 



to O O CTi O to O 

■— I CTi to 

01 <3- «3" CTi 



CM O O to O O CD 
tO CTi ■— 1 

d CM CTi 



OOOOOOO 



OOOOOOO 



r~~ CTi CM >^- CM 1— I CTi 

to ro in to co co co 
1— c 1--. un un cm cm ro 



ai r»- rH rv a. r-. o 
to co to r~^ *j- 
«J- to to ro 1 — 



■— 1 o co t— 1 cm r-» i-h 

(^ cr> in cm 00 ro r-^ 

r-^ r~~ cm •— 1 ro .— 1 



o cri en cd ro ■— 1 ro 
o ro o i-h cm 
in o un to o 



O CO o o *s- cm r^. 
to co o CM 

co 00 f~- in 



o i-h o o cr> to i-h 
en to co CO 

ro r-~ 1— 1 ^j- 



C31 cm «d- 

"3" -h 



O co O O <n to un 

•-H CO to CM 

CM rH^fCO 



en ^- ^j- 



O O O CD to O CO 
CM CO 



OOOOOOO 



co 00 en cr> en in 
i-h o r~~ to en o i-h 
r- en ro in to o cm 



O *3- ro 
OtOH 
CM 



<<<<<<< 55<5555 



cu o> 

O (J 



<0 

■a 
01 
u 



■a 
o 
o 

3 

c -a 
o o 

4-> o 
4J 3 

o 00 



to 



01 3 3 s_ 

01 C i- S_ -i- 
C0)-i-O_Cm_j»;c I 
■1— C O- to (/) O4-0J 
Q.-I- E <0 O) +-> 
Q-0) O).^ <D ! C-i- 

otj-i-'i- <Oi — E i- 3-*: E Q.O- o_i — 
fOOJ-C-Ci — (onso fDi — 15 « m id 

oaS3£flttlHZ OLUZ<Q.IM 



JO -Q 



H3 Ol C S- <0 

r— OJ 0J U0 



24 



Table 18. --Average annual managed harvest growing-stock volume from harvested plots by forest type, 
site index class, and stand-age class, Aspen-Birch Unit, Minnesota, 1977-1986 

(In thousand cubic feet) 



HIGH ROTATION AGE OPTION 





Site 
index 














Stand- 


age c 


lass (years 


) 








Forest 


less than 




























more than 


type 


class 


Total 


41 41 


-50 


51-f 


61-; 


71-i 


81-90 


91 


-100 


101-120 


121-140 


140 


Jack pine 


0-60 


4,438 















168 




145 


1 


,961 


1 


,987 


177 










61 + 


567 


(i 







409 




158 


























Red pine 


0-55 


1,214 

























893 




205 


116 










56+ 


1,750 








1 


,050 




374 









326 
















White pine 


0-55 


472 















(i 




H 









357 


115 










56+ 


21 










I) 



















21 











White spruce 


All 


870 


n 


n 









174 




120 




141 




32/ 


108 








Black spruce 


0-40 


3,509 





I) 






















1 


,921 


690 


898 







41 + 


3,366 


















1 


,105 


1 


,350 




911 











Balsam fir 


All 


12,784 








4 


,149 


4 


,918 


1 


,805 




800 




935 


177 








Tamarack 


0-40 


488 

































220 


221 


47 




41 + 


636 


i) 












224 




165 












128 


1 1 9 


n 


Northern 




































white-cedar 


All 


2,670 















(I 




















2,670 





Oak 


0-55 


23 


11 


23 






































56+ 


63 


(1 







63 































Elm-ash- cottonwood 


0-55 


2,288 





il 
























623 


919 


746 







56+ 


737 


u 


n 



















331 










406 





Maple-basswood 


All 


144 

































73 


71 





Aspen 


0-65 


53,220 








22 


,096 


19 


,840 


7 


,800 


2 


,952 




442 


90 










66+ 


36,877 








19 


,379 


9 


,947 


3 


,979 


2 


,865 




707 











Paper birch 


0-65 


13,707 













3 


,331 


5 


,950 


3 


,907 




348 


1/1 










66+ 


2,978 










253 




779 


1 


,594 









352 











Balsam poplar 


0-65 


3,088 










585 


1 


,081 




782 




469 




1/1 













66+ 


2,564 










555 


1 


,037 




741 




231 
















All types 




148,474 





23 


48 


,639 


42 


,031 


24 


,186 


16 


,226 


9 


,307 


2,984 


5,131 


47 








LOW 


ROTATION AGE OPTION 




















Jack pine 


All 


8,014 













3 


,744 




145 


1 


961 


1 


,987 


177 








Red pine 


All 


2,975 


n 




















2 


,654 




205 


116 








White pine 


All 


659 


1! 


o 
























544 


115 








White spruce 


All 


1,166 


o 


o 




296 




174 




120 




141 




32/ 


108 








Black spruce 


All 


10,025 




















625 


2 


877 


4 


,935 


690 


898 





Balsam fir 


All 


15,545 








6 


,910 


4 


,918 


1 


,805 




800 




935 


1// 








Tamarack 


All 


1,217 





n 



















34 




b6 


731 


339 


17 


Northern 




































white-cedar 


All 


5,996 






























/I 


2,212 


3,713 


o 


Oak 


All 


213 


n 







213 




11 


























Elm- ash- cottonwood 


All 


4,963 


n 


11 

















1 


071 


1 


,812 


919 


1,161 





Maple-basswood 


All 


144 

































/3 


71 





Aspen 


All 


110,664 


(i 





61 


701 


30 


134 


11 


779 


5 


817 


1 


,143 


90 





(J 


Paper birch 


All 


20,946 








2 


953 


5 


666 


7 


549 


3 


907 




700 


1/1 





(J 


Balsam poplar 


All 


6,203 








1 


,674 


2 


,124 


1 


,529 




705 




171 











All types 




188,730 








73 


,747 


46 


,760 


23 


,552 


19 


967 


12 


,896 


5,579 


6,182 


47 



25 



Table 19. --Average annual managed harvest growing-stock volume from harvested plots by forest type, 
site index class, and stand-age class, Northern Pine Unit, Minnesota, 1977-1986 

(In thousand cubic feet) 
HIGH ROTATION AGE OPTION 





Site 
index 














Stand-age cl 


ass (years) 








Forest 


less tha 


1 






















more than 


type 


class 


Total 


41 


41-50 


5] 


-60 


6] 


-70 


71-80 


81-< 


91-100 


101-120 


121-140 


140 


Jack pine 


0-60 


4,454 










988 


2 


577 


796 







133 













61 + 


4,578 








2 


993 


1 


585 






















Red pine 


0-55 


408 


























237 


171 










56+ 


1,411 


























1,411 











White pine 


0-55 


270 





























270 










56+ 


374 


























374 











White spruce 


All 


306 















153 







153 














Black spruce 


0-40 


1,271 























779 


261 


93 


138 







41 + 


588 















206 


48 




253 





81 








Balsam fir 


All 


2,512 















824 


1,429 




188 





71 








Tamarack 


0-40 


640 
































640 







41 + 


1,287 


























104 


842 


341 





Northern 
































white-cedar 


All 


1,488 
































1,231 


257 


Oak 


0-55 


1,190 










il 












84 


790 


316 










56+ 


1,499 















594 


431 




156 





318 








Elm-ash-cottonwood 


0-55 


2,742 


























970 


1,214 


558 







56+ 


967 





n 




















236 


526 


205 





Maple-basswood 


ALL 


425 


























221 


4 


200 





Aspen 


0-65 


23,538 


(1 





9 


160 


9 


557 


3,569 




845 


205 


64 


138 







66+ 


38,111 








22 


718 


10 


,078 


2,452 


1 


,834 


834 


195 








Paper birch 


0-65 


8,447 













4 


,224 


1,949 


1 


,431 


307 


536 










66+ 


2,566 













1 


767 


202 




268 


329 











Balsam poplar 


0-65 


3,968 





117 


1 


680 


1 


284 


887 





















66+ 


2,340 










839 


1 


,073 


110 




318 














All types 




105,420 





117 


38 


,378 


33 


,922 


11,873 


6 


,309 


6,412 


4,701 


3,451 


257 


LOW ROTATION AGE OPTION 


Jack pine 


All 


12,153 








7 


062 


4 


162 


796 







133 











Red pine 


All 


2,764 


(.) 























2,593 


171 








White pine 


All 


893 





























893 








White spruce 


All 


398 















245 







153 














Black spruce 


All 


3,207 















490 


533 


1 


,610 


261 


175 


138 





Balsam fir 


All 


2,736 













1 


048 


1,429 




188 





71 








Tamarack 


All 


2,122 





























979 


1,143 





Northern 
































white-cedar 


All 


2,717 





























257 


2,203 


257 


Oak 


All 


5,695 













1 


391 


2,368 




509 


790 


637 








Elm-ash-cottonwood 


All 


5,416 























968 


1,941 


1,742 


765 





Maple-basswood 


All 


425 


























221 


4 


200 





Aspen 


All 


90,026 





2,726 


57 


488 


19 


635 


6,055 


2 


,686 


1,039 


259 


138 





Paper birch 


All 


13,063 








1 


513 


6 


520 


2,153 


1 


,704 


637 


536 








Balsam poplar 


All 


7,012 





188 


3 


151 


2 


,357 


998 




318 














All types 




148,627 





2,914 


69,214 


35,848 


14,332 


8 


,136 


7,615 


5,724 


4,587 


257 



26 



APPENDIX 



Specifying Cutting Prescriptions 

The level of yield from cutting prescriptions varies 
according to the criteria specified. Two prescriptions 
are considered here. 3 The long rotation age option 
uses harvest criteria established jointly by repre- 
sentatives from Minnesota forest industries and the 
Minnesota Department of Natural Resources (table 
1). Harvest criteria for this option are based on forest 
type, age, and site index. 

The short rotation age option is based on current 
stand conditions, and sets the rotation age for a spe- 
cies at the age of maximum total volume yield. Forest 
type and age are the stand characteristics considered 
in this option. Age of maximum total volume yield is 
calculated for each forest type by ( 1) determining the 
number of rotation acres in each age class per 
year — this is equal to the area in the forest type 
divided by the midpoint of the age class; (2) calculat- 
ing the growing-stock volume per acre of commercial 
forest land for each age class — this is equal to the 
growing-stock volume in the age class divided by the 
commercial forest area in the age class; (3) 
determining the yield of an age class — this is equal to 
the rotation acres in the age class multiplied by the 
growing-stock volume per acre of commercial forest 
in the age class. The rotation age is selected from the 
age class where total volume yield is highest. 

Definition of Terms 

Land-use classes 

Forest land. — Land at least 16.7 percent stocked 
by forest trees of any size, or formerly having such 
tree cover, and not currently developed for nonforest 
use. Includes afforested areas. The minimum forest 
area classified was 1 acre. Roadside, streamside, and 
shelterbelt strips of timber must have a crown width 
of at least 120 feet to qualify as forest land. Unim- 
proved roads and trails, streams, and clearings in 
forest areas were classed as forest if less than 120 feet 
wide. 



3 The Minnesota Department of Natural Resources 
has made an estimate of yield under a third set of 
cutting prescriptions. 



Commercial forest land.— Forest land that is 
producing or is capable of producing crops of indus- 
trial wood and that is not withdrawn from timber 
utilization by statute or adiministrative regulation. 
This includes areas suitable for management to grow 
crops of industrial wood generally of a site quality 
capable of producing in excess of 20 cubic feet per acre 
of annual growth. This includes both inaccessible and 
inoperable areas. 

Tree class 

Growing-stock trees. — All live trees of commer- 
cial species except rough and rotten trees. 

Sawtimber trees. — Growing-stock trees of com- 
mercial species containing at least a 12-foot saw log 
or two noncontiguous saw logs, each 8 feet or longer. 
At least 33 percent of the gross volume of the tree 
must be sound wood. Softwoods must be at least 9.0 
inches d.b.h. and hardwoods at least 11.0 inches. 

Poletimber trees. — Growing-stock trees of com- 
mercial species at least 5.0 inches d.b.h. but smaller 
than sawtimber size, and of good form and vigor. 

Rotten trees. — Live trees (any size) of commer- 
cial species that do not contain a merchantable 12- 
foot saw log or two noncontiguous 8-foot or longer saw 
logs, now or prospectively, because of rot (that is, 
when more than 50 percent of the cull volume of the 
tree is rotten). 

Rough trees. — Live trees that do not contain at 
least one merchantable 12-foot saw log or two non- 
contiguous 8-foot or longer saw logs, now or prospec- 
tively, because of roughness and poor form, as well as 
all live noncommercial species. 

Short-log (rough trees). — Sawtimber sized trees 
of commercial species that contain at least one mer- 
chantable 8- to 11-foot saw log but not a 12-foot saw 
log. 

Other classifications 

Site index. — An expression of forest site quality 
based on the height of a free-growing dominant or 
codominant tree of a representative species in the 
forest type at age 50. 



27 



Stand-age. — Age of the main stand. Main stand 
refers to trees of the dominant forest type and stand- 
size class. 

Forest types 

A classification of forest land based upon the spe- 
cies forming a plurality of live-tree stocking. Major 
forest types in Minnesota are: 

Jack pine. — Forests in which jack pine comprises 
a plurality of the stocking. (Common associates in- 
clude eastern white pine, red pine, aspen, birch, and 
maple.) 

Red pine. — Forests in which red pine comprises a 
plurality of the stocking. (Common associates in- 
clude eastern white pine, jack pine, aspen, birch, and 
maple.) 

White pine. — Forests in which eastern white pine 
comprises a plurality of the stocking. (Common asso- 
ciates include red pine, jack pine, aspen, birch, and 
maple.) 

White spruce. — Forests in which white spruce 
and balsam fir comprise a plurality of the stocking, 
with white spruce more common. (Common associ- 
ates include aspen, maple, birch, northern white- 
cedar, tamarack.) 

Black spruce. — Forests in which swamp conifers 
comprise a plurality of the stocking with black spruce 
most common. (Common associates include tama- 
rack and northern white-cedar.) 

Balsam fir. — Forests in which balsam fir and 
white spruce comprise a plurality of stocking, with 
balsam fir more common. (Common associates in- 
clude aspen, maple, birch, northern white-cedar, and 
tamarack.) 

Tamarack. — Forests in which swamp conifers 
comprise a plurality of the stocking, with tamarack 
most common. (Common associates include black 
spruce and northern white-cedar.) 

Northern white-cedar. — Forests in which 
swamp conifers comprise a plurality of the stocking, 
with northern white-cedar most common. (Common 
associates include tamarack and black spruce.) 

Oak. — Forests in which northern red oak, white 
oak, or bur oak, singly or in combination, comprise a 
plurality of the stocking. (Common associates in- 
clude elm, maple, and aspen.) 



Elm-ash-cottonwood. — Forests in which elm, 
ash, cottonwood and red maple, singly or in combina- 
tion, comprise a plurality of the stocking. (Common 
associates include basswood and balsam poplar.) 

Maple-basswood. — Forests in which sugar ma- 
ple, basswood, yellow birch, American elm, and red 
maple, singly or in combination, comprise a plurality 
of the stocking. (Common associates include white 
pine and elm.) 

Aspen. — Forests in which quaking aspen or big- 
tooth aspen, singly or in combination, comprise a 
plurality of the stocking. (Common associates in- 
clude balsam poplar, balsam fir, and paper birch.) 

Paper birch. — Forests in which paper birch com- 
prises a plurality of the stocking. (Common associ- 
ates include maple, aspen, and balsam fir.) 

Balsam poplar. — Forests in which balsam poplar 
comprises a plurality of the stocking. (Common asso- 
ciates include aspen, elm, and ash.) 

Timber removals 

Timber removals from growing stock. — The 

volume of sound wood in growing-stock trees re- 
moved annually for forest products (including 
roundwood products and logging residues) and for 
other removals. Roundwood products are logs, bolts, 
or other round sections cut and used from trees. Log- 
ging residues are the unused portions of cut trees plus 
unused trees killed by logging. Other removals are 
growing-stock trees removed but not utilized for 
products, or trees left standing but "removed" from 
the commercial forest land classification by land use 
change — examples are removals from cultural opera- 
tions such as timber stand improvement work, land 
clearing, and changes in land use. 



Metric Equivalents of Units 
Used in this Report 

1 acre = 4,046.86 square meters or 0.405 hectare. 

1,000 acres = 405 hectares. 

1,000 board feet (International Winch rule) = 3.48 

cubic meters. 

1 cubic foot = 0.0283 cubic meter. 



28 



Principal Tree Species Groups 
in Northern Minnesota 4 

Eastern white pine Pinus strobus 

Red pine Pinus resinosa 

Jack pine Pinus banksiana 

White spruce Picea glauca 

Black spruce Picea mariana 

Balsam fir Abies balsamea 

Tamarack Larix laricina 

Northern white-cedar Thuja occidentalis 

Other softwoods: 

Eastern redcedar Juniperus virginiana 

Scotch pine Pinus syluestris 

White oaks: 

White oak Quercus alba 

Bur oak Quercus macrocarpa 

Select red oak: 

Northern red oak Quercus rubra 

Other red oaks: 
Northern pin oak Quercus ellipsoidalis 

Hickories: 

Butternut hickory Carya cordiformis 

Yellow birch Betula alleghaniensis 

Hard maples: 

Sugar maple Acer saccharum 



Soft maples: 

Red maple Acer rubrum 

Silver maple Acer saccharinum 

Ashes: 

White ash Fraxinus americana 

Black ash Fraxinus nigra 

Green ash Fraxinus pennsyluanica 

Balsam poplar Populus balsamifera 

Paper birch Betula papyrifera 

Aspens: 

Bigtooth aspen Populus grandidentata 

Quaking aspen Populus tremuloides 

Basswood Tilia americana 

Elms: 

Amercian elm Ulmus americana 

Slippery elm Ulmus rubra 

Rock elm Ulmus thomasii 

Select hardwoods: 

Butternut Juglans cinerea 

Black cherry Prunus serotina 

Other hardwoods: 

Boxelder Acer negundo 

Eastern cottonwood Populus deltoides 

Black willow Salix nigra 



*The common and scientific names are based on: 
Little, Elbert L., Jr. 1979. Check List of United States 
Trees (Native and Naturalized). U.S. Department of 
Agriculture, Agricultural Handbook 541, 375 p. 



2>U.S. GOVERNMENT PRINTING OFFICE: 1980-669-277/62 REG. #6 



29 



Jakes, Pamela J., and W. Brad Smith. 

1980. Predicted yields from selected cutting prescriptions in northern 
Minnesota. U.S. Department of Agriculture Forest Service, Research 
Paper NC-188, 29 p. U.S. Department of Agriculture Forest Service, 
North Central Forest Experiment Station, St. Paul, Minnesota. 

Includes predicted yields based on two sets of cutting prescriptions 
in northern Minnesota. Indicates that given a specific set of assump- 
tions, average annual growing-stock removals for the decade 1977- 
1986 would be from 69 percent to 124 percent higher than 1976 
growing-stock removals. 



KEY WORDS: Minnesota, timber removals, timber harvest. 



Jakes, Pamela J., and W. Brad Smith. 

1980. Predicted yields from selected cutting prescriptions in northern 
Minnesota. U.S. Department of Agriculture Forest Service, Research 
Paper NC-188, 29 p. U.S. Department of Agriculture Forest Service, 
North Central Forest Experiment Station, St. Paul, Minnesota. 

Includes predicted yields based on two sets of cutting prescriptions 
in northern Minnesota. Indicates that given a specific set of assump- 
tions, average annual growing-stock removals for the decade 1977- 
1986 would be from 69 percent to 124 percent higher than 1976 
growing-stock removals. 



KEY WORDS: Minnesota, timber removals, timber harvest. 




Mon.2. bicycler and i>koo. l£a£hQA...leAA 6moq. 



)A FOREST SERVICE 
SEARCH PAPER NC-189 



i.hTS 



FEB % 



StEMSOtf 




Comparing Jack Pine Slash 

and Forest Floor 

Moisture Contents and 

National Fire Danger 

Rating System Predictions 



Robert M. Loomis and William A. Main 




1h Central Forest Experiment Station 
cest Service, U.S. Department of Agriculture 



Loomis, Robert M., and William A. Main. 

1980. Comparing jack pine slash and forest floor moisture contents and 
National Fire Danger Rating System predictions. U.S. Department of 
Agriculture Forest Service, Research Paper NC-189, 10 p. U.S. 
Department of Agriculture Forest Service, North Central Forest 
Experiment Station, St. Paul, Minnesota. 

Relations between certain slash and forest floor moisture contents 
and the applicable estimated timelag fuel moistures of the National 
Fire Danger Rating System were investigated for 1-year-old jack pine 
fuel types in notheastern Minnesota and central Lower Michigan. 
Only approximate estimates of actual fuel moisture are possible for 
the relations determined, thus emphasizing the need for on-site mea- 
surements when fuel moisture may be critical. 

KEY WORDS: Forest fuels, fire hazard. 



North Central Forest Experiment Station 

Forest Service — U.S. Department of Agriculture 

1992 Folwell Avenue 

St. Paul, Minnesota 55108 

Manuscript approved for publication February 15, 1980 

1980 



COMPARING JACK PINE SLASH AND 

FOREST FLOOR MOISTURE CONTENTS AND 

NATIONAL FIRE DANGER RATING 

SYSTEM PREDICTIONS 



Robert M. Loomis, 

Fire Management Scientist 
and William A. Main, 

Computer Programmer 
East Lansing, Michigan 



Thousands of acres of jack pine are clear-cut in the 
Lake States each year (fig. 1). Prescribed fire to re- 
duce fuel amounts and prepare the site for planting or 
seeding frequently follows this cutting. The land 
manager must select the time when fuel and weather 
conditions will produce a fire that will accomplish the 
desired results and assist in planning burning opera- 
tions (Deeming et al. 1977). The National Fire 
Danger Rating System (NFDRS) estimates and in- 
terprets fuel moisture, a critical variable in fire be- 
havior. Information is limited as to the actual fuel 
moisture content of natural fuels in relation to 
NFDRS moisture content estimates. Our study re- 
ports the moisture content of specific fuels found on 
1-year-old jack pine clear-cut areas in relation to 
NFDRS 1-hour, 10-hour, and 100-hour timelag fuel 
moisture content estimates. 1 



RESULTS 

We found only a general relation (significant, but 
with much variation) between the moisture content 
of certain naturally occurring fuels on 1 -year-old jack 
pine slash areas and NFDRS fuel moisture estimates. 
Statistically significant but low correlation exists 



between actual and NFDRS-estimated moisture con- 
tent for Fines and 1-hour timelag fuels (fig. 2); be- 
tween L-layer and 1-hour timelag fuels (fig. 3); and 
between F-layer and 100-hour timelag fuels (fig. 4). 2 
The actual moisture content of Fines and of L-layer 
almost always exceeded the estimated 1-hour time- 
lag fuel moisture. We examined each of these rela- 
tions using the paired t-test. In all cases, the null 
hypothesis was rejected — the mean of the population 
differences was not zero. 



THE STUDY 

Two study areas were selected on the Superior 
National Forest in Minnesota. The previous summer, 
55 cords per acre of jack pine pulpwood had been 
harvested by clear-cutting. The soils were well- 
drained loamy sands. We selected a third study area 
on State land near Roscommon, Michigan, where 
clear-cutting had removed about 15 cords per acre of 
jack pine pulpwood the winter preceding the summer 
study. This site had well-drained sandy soil. 



1 This study was designed by Rodney Sando (for- 
merly with the North Central Forest Experiment 
Station) who was responsible for data collected in 
Minnesota. 



2 Fines consist of jack pine slash needles and twigs 
less than 'A inch in diameter aerially exposed at one 
foot or more above ground; L-layer consists of forest 
floor loose surface needles, leaves, grass, twigs, etc.; 
F-layer consists of the forest floor layer beneath the 
L-layer— the compacted decomposing organic materi- 
als still identifiable as to source. 




Figure 1. — Thousands of acres of jack pine are clear- 
cut in the Lake States each year. 



Actual fuel moistures were measured for Fines, for 
forest floor L-layer and for forest floor F-layer from 
fuel sample collections taken: June 9 through Sep- 
tember 9, 1968, in Lake County, Minnesota; and 
June 13 through September 3, 1969, in St. Louis 
County, Minnesota; and June 12 through September 
17, 1975, in Roscommon County, Michigan. 

On-site weather instruments and records were 
maintained to provide data to compute NFDRS fuel 
moisture values. Samples were collected daily except 
on weekends or when fuels were wet from rainfall 
at collection time. We had no 10-hour timelag fuel 
moisture sticks so this value for the NFDRS was 
computed. 

In Minnesota a steel sampling cutter was used to 
remove 2-inch diameter circular samples of forest 
floor materials, and in Michigan a 5-inch diameter 
sampling cylinder was used. Five samples of slash 
and 10 samples each of the forest floor layers were 
collected daily, 15 samples of the F-layer were col- 
lected beginning the second summer because of the 
variability found the first year. We determined mois- 
ture content by oven drying the samples at 105°C for 
about 24 hours. The observations and measurements 
were made during early or mid-afternoon to relate 
closely with time of basic fire weather observations. 

Covariance analysis indicated 1968 and 1969 data 
should not be combined for regressions. Data for 
these two years (locations) came from apparently 



similar areas in Minnesota. Because of differences in 
moisture response — moisture estimate relations, a 
similar data set was obtained in Michigan in 1975 
and each year (location) was considered separately. 



DISCUSSION 

The NFDRS fuel classes relate directly to dead, 
round wood. They relate only "roughly" to portions of 
the forest floor. The aerially exposed slash studied is 
a 1-hour timelag fuel while forest floor fuel compo- 
nents are not clearly defined in terms of timelag 
(Deeming e£ a/. 1977). Sticks (Vis-inch pine dowels) are 
an excellent source of 10-hour timelag fuel moisture 
estimates and contributed to the 1-hour timelag fuel 
moisture estimate in the 1978 system. These fuel 
moisture sticks integrate effects of the meteorologi- 
cal environment including day length and cloudi- 
ness. Stick weight measurements were not available 
so 1-hour and 10-hour timelag fuel moistures were 
computed. Without stick moisture measurements, 
the system has an adjustment for 1-hour timelag 
moisture estimates based on state of weather. Tem- 
perature and relative humidity adjustments of shel- 
ter observations (4.5 feet above ground) are made to 
better estimate the temperature of fuels on the 
ground and the relative humidity at their level before 
computing an equilibrium moisture content. 

Because 1-hour and 10-hour timelag fuel mois- 
tures are "multiples" of each other when stick mois- 
ture contents are not available, we considered only 
the 1-hour estimates related to the actual moisture 
content of Fines and the L-layer actual. However, the 1 
10-hour timelag fuel moisture also could have been 
used. The L-layer probably has between a 1-hour and 
10-hour timelag response. Fosberg (1971) reported 
the NFDRS 100-hour timelag fuel moisture esti- 
mates are inapplicable for duff and litter. This study 
found significant, but low correlation between the 
100-hour estimates and the F-layer moisture 
contents. The F-layer response had slightly higher 
correlation with the 100-hour than with the 10-houi 
timelag fuel moisture estimates. 

This study did not determine a timelag response 
value for the fuel components, but does show theii 
relation to NFDRS timelag fuel moisture values. 

The moisture content of the Fines and L-layei 
was almost always equal to or greater than the esti 
mated NFDRS 1-hour timelag fuel moisture. Th< 
regression lines are distinctly above the 1-hour time 
lag estimates. 

Similar discrepancies have been reported b} 
others. Hough and Albini (1978) found NFDRS 



(Deeming et al. 1972) estimates of 1-hour and 10-hour 
timelag fuel moisture inadequate to estimate dead 
fuel moisture for their palmetto-gallberry fuel 
model. Actual moisture content of field plot L-layer 
ranged from 9 to 33 percent while 1-hour estimates 
ranged from 5 to 11 percent and 10-hour esti- 
mates ranged from 6 to 14 percent. 

The NFDRS 1-hour timelag fuel moisture predic- 
tions are unlikely to be exactly the same as those 
found for various naturally occurring 1-hour timelag 
fuels. NFDRS predictions are based on equilibrium 
moisture content for wood while the fuels we ob- 
served included needles, leaves, bark, and grasses in 
addition to wood. Investigators have shown equilib- 
rium moisture content for such other materials to be 
slightly higher than for wood (Dunlap 1932, Black- 
marr 1971, VanWagner 1972, Anderson et al. 1978). 

Some interacting variables influencing moisture 
content are not directly considered in the national 
system. Forest floor and surface soil temperature and 
moisture gradients; fuel amounts, composition and 
.potential equilibrium moisture contents; wind speed, 
aspect, slope, stability, and the radiation absorption 
and emission characteristics of the fuels and sur- 
rounding environment contribute to NFDRS fuel 
moisture value variation. 

A clear division between L-layer and F-layer, and 
between F-layer and H-layer usually does not exist. 
In addition, forest floor and upper soil moisture is 
influenced by vegetative roots contributing to a tran- 
spirational drain. Because the Fines were the most 
uniform and uniformly exposed materials, they had 
the smallest standard errors of estimates. 



A question arises as to whether the low correla- 
tions result from natural variability or a poor rela- 
tion between forest floor moisture content and 
NFDRS predictions. To check the data, we compared 
it to the Canadian Fire Weather Index (FWI) which 
VanWagner (1975) reports is related to forest floor 
moisture content. This independent comparison re- 
sulted in an r 2 (coefficient of determination) of 0.71 
when the FWI Fine Fuel Moisture Code was re- 
gressed against L-layer moisture content. 

These results indicate that natural variability 
of the data was not the underlying cause of low 
correlations. 



SUMMARY 

This study emphasized the range and variability in 
moisture response for various slash and forest floor 
fuels associated with 1-year-old jack pine slash areas 
in Minnesota and Michigan when the moisture val- 
ues were related to NFDRS fuel moisture estimates. 
Our results indicate the National Fire Danger Rat- 
ing System, which rates fire danger for the worst 
possible conditions — on south or southwesterly as- 
pects, based on aerially exposed woody material 
(Deeming et al. 1977) — does not work well for litter 
and duff on the forest floor. Only approximate esti- 
mates can be made for the moisture content of Fines, 
L-layer, and F-layer using NFDRS estimates of the 
moisture content of 1-hour and 100-hour timelag 
fuels. These results emphasize the value of on-site 
fuel moisture measurements when fuel moisture con- 
tent may be critical. 



□ 



MS .. 



Ld 40 
a: 

UJ 

Q- 35 - 



Ld 30 .. 



□ 




-^> 


25 


Ld 




tK 




El 

h- 


20 


m 




□ 




7": 


IE 


Ln 




LJ 





10 .. 



LI 



LJ 




□ 







EI 


□ 






EI 

n 


□ 


n 


H 










El 














□ El 


ri 




n 


El 






El 


tH 


El 

U 






n 






DJ 


El 




El 






23 EI 


H 


i i 


M 








^ El 




■ ! 










El 















El 



□ 



Y = .60 + 2.70X 
r 2 = .58 
Sy-x = 9. 



E/3 - 3/9 I9EB LRKE CD. Ml 



■+■ 



-+- 



■+■ 



3 E 9 12 15 IB 21 24 

NFDR5 DNE-HDUR TIMELRG FUEL MDI5TURE PERCENT 



27 



MS .. 



LxJ 40 

cr 

Ld 

a- 35 



lj 30 .. 



z 




[ J 




'.,' 


2E 


LiJ 




["r 




LJ 

h 


20 


1X1 




□ 




>. 


15 


1X1 




Ld 





10 .. 



□ 




EI □ , 
□ □ Q^EI □ □ 

"" El 

El El 

Bg B 



E/13 - 9/3 I9E9 ST. LULlIS CD. MINN. 



3 E 3 12 15 IB 21 2H 

NFC>R5 DNE-HnUR TIMELRE FUEL MDI5TURE PERCENT 



27 



H5 .. 



UJ 


H0 


l / 




DC 




UJ 




ri 


35 



U 30 .. 

I— 

□ 

^ 25 

UJ 

cr 

un 

□ 
?: 

ui 

E 10 .. 



15 .. 



LI 



El 



L.l 



EJ 



n 



n 







L'l 






L:i 






[-1 






El 






□ 












[.1 








L7J 

LI 

EL 


El 


LJ 
El 

Ul 


El 

DM 




""El 
El □ 


EI 
EI 

m Fl 

El 


El 

LJ 

n 

L.l 
L.l 


El 
El 


El 

L'l 

LI 


[ : 





LI 



El 



13.68 

,1 7 . 



. Hi) 



LI 



Sv- 



E/ 12 - 3/17 1375 RD5CDMMDM CD. MICH. 



3 E 3 12 15 IB 21 2H 

NFDRE DNE-HDUR TIMELRG FUEL MOISTURE PERCENT 



27 



Figure 2. — Aerially exposed (1 foot or more above ground) slash needles and 
twigs (equal to or less than V4 inch in diameter), (Fines) moisture content- 
National Fire Danger Rating System (NFDRS) 1-hour timelag fuel moisture 
relations from 1 -year -old jack pine slash area. 



Ld 

V 9E 
q: 

Ld 

□_ 

BH 



P 72 .. 



□ 
U 

nr 



E0 



Ln HB 



□ 

cr 
u 

>- 

LE 



35 .. 



2H .. 



12 .. 









* 






4* 






* 
«* 


* 








* 




* 

«* 


Y = -16.90 + 6.55X 


* * 


* 






>> 






«* 


R 2 = .48 
Sy-x = 26. 


$ 


* 


/4 


* 
«* 




* 


«3« 






/• 


1 

* 


* 


$ 












1 1_ 


« 


— 1— 






— t 


E/a - 
— i 1 


3/9 I3EB LRKE CD. MINN. 
1 1 1 



3 E 3 12 IS IB 21 2H 27 

NFDR5 QNE-HLUR TIMELR5 FUEL MOISTURE PERCENT 



I0B .. 



Ld 

V 9E 
cr 



BH .. 



£ 72 1 



□ 




KS 






RPI 


Ld 




q: 




m 




U1 


HB 


□ 




z: 






3R 


IL 




U 




>- 




CE 


-3U 




g g I * i 



Y = -.39 + 2.50X 

R 2 = .21 



& Sy-x = 17. 



E/13 - 9/3 I3E3 5T. LDUI5 CD. Ml 



■+■ 



-+- 



3 E 9 12 15 IB 21 2H 27 

NFDR5 DNE-HDUR T1MELRG FUEL MDI5TURE PERCENT 



IBS .. 



B0 + 



UJ 

K^> 3B 

\x. 

uj 

□_ 

BH 



£ 72 4- 
□ 

UJ 

rt. 
m 

En 

□ 

i 

12 .. 




Y = -11.28 + 4.65X 
R 2 = .37 
Svx = 22. 



B/12 - 3/17 1975 RD5CDMMQN CD. MICH. 



3 B B 12 IS IB 2! 2H 27 

NFDR5 ONE- HOUR TIMELRG FUEL MOISTURE PERCENT 



Figure 3. — Forest floor L-layer moisture content — National Fire Danger Rating 
System (NFDRS) 1 -hour timelag fuel moisture relations for 1 -year-old jack pine 
slash area. 



2: 

UJ 

m 

Ld 
Q. 



Ld 



□ 

Ld 

DC 
Zl 
h- 
LT1 

□ 

cr 

Ld 
>- 

rr 











+ 










215 H 














+ 




132 . 






+ 


+ 




+ + 
4- + 

+ / 






IBB . 






4- 

+ * 


+ 


* 

4= 


■/:■ 

+ 






IHH - 




+ 


+ 
+ + 


+ 


/ + 






120 . 




+ 


+ / 




* 


* 
+ 




Y = -47.17 + 11.93X 
r 2 = .29 


3E . 
72 . 




+ 


/ * 

/ + * 

/ + + 


+ 
+ 




+ 




Sy-x = 38. 


HB . 






/ + + 


+ 










2H . 




/ 


















( 1 


J 1 „ 


— H- 




E/3 - 
1 


3/3 
1 — 


I3EB LRKE CD. MINN. 
1 ,__ — 



3 E 3 12 15 IB 21 2H 27 

NFDR5 DNE HUNDRED-HDUR TIMELRE FUEL MDI5TURE PERCENT 



i 

Ld 

rr. 

Ld 

a. 



Ld 

h- 

rj 

Ld 

rr: 

I— 

m 

□ 
^: 

re 

Ld 
>- 

rr 
1 



2IE 1 

! 

132 j. 

iGB i 

IHH I 

123 .. 

35 .. 

72 .. 

HB .. 
2H 




Y = -161.11 + 15.94X 
r 2 = .11 
Sy-x = 33. 



5/13 - 3/3 1353 5T. LOUIS CD. Ml 



3 5 3 12 15 IB 21 2H 27 

NFDR5 DNE HUNDRED- HOUR TIMELRE FUEL MDI5TURE PERCENT 





2iE 4_ 


LJ 

K 
Ld 
CL 


132 | 
IBB 4. 


UJ 

i— 


IHH I 




120 | 






1X1 

r 3 


<3E | 


UJ 

>- 
en 


MB j 



2H 4 






3 E 3 



12 



m* 



Y = -165.93 * 

R 2 = .W 

Sy-x = 23. 



B/12 - 3/17 1975 RD5CGMMGN CG. MICH. 



IB 21 2H 27 

NFDR5 ONE HUNDRED- HtJUR TIMELRE FUEL MU15TURE PERCENT 



Figure 4. — Forest floor F -layer moisture content — National Fire Danger Rating 
System (NFDRS) 100-hour timelag fuel moisture relations for 1 -year-old jack 
pine slash area. 



;:- 



LITERATURE CITED 

Anderson, Hal E., Robert D. Schuette, and Robert W. 
Mutch. 1978. Timelag and equilibrium moisture 
content of ponderosa pine needles. U.S. Depart- 
ment of Agriculture Forest Service, Research Pa- 
per INT-202, 28 p. U.S. Department of Agriculture 
Forest Service, Intermountain Forest and Range 
Experiment Station, Ogden, Utah. 

Blackmarr, W. H. 1971. Equilibrium moisture con- 
tent of common fine fuels found in southeastern 
forests. U.S. Department of Agriculture Forest 
Service, Research Paper SE-74, 8 p. U.S. Depart- 
ment of Agriculture Forest Service, Southeastern 
Forest Experiment Station, Asheville, North 
Carolina. 

Deeming, John E., Robert E. Burgan, and Jack D. 
Cohen. 1977. The National Fire Danger Rating 
System — 1978. U.S. Department of Agriculture 
Forest Service, General Technical Report INT-39, 
63 p. U.S. Department of Agriculture Forest Serv- 
ice, Intermountain Forest and Range Experiment 
Station, Ogden, Utah. 

Deeming, John E., J. W. Lancaster, M. A. Fosberg, R. 
W. Furman, and M. J. Schroeder. 1972. The Na- 
tional Fire Danger Rating System. U.S. Depart- 
ment of Agriculture Forest Service, Research 



Paper RM-84, 165 p. U.S. Department of Agricul- 
ture Forest Service, Rocky Mountain Forest and 
Range Experiment Station, Fort Collins, Colorado. 

Dunlap, M. E. 1932. Drying rate of hardwood-forest 
leaves. Journal of Forestry 30:421-423. 

Fosberg, Michael A. 1971. Moisture content calcula- 
tions for the 100-hour timelag fuel in fire danger 
rating. U.S. Department of Agriculture Forest 
Service, Research Note RM-199, 7 p. U.S. Depart- 
ment of Agriculture Forest Service, Rocky Moun- 
tain Forest and Range Experiment Station, Fort 
Collins, Colorado. 

Hough, W. A., and F. A. Albini. 1978. Predicting fire 
behavior in palmetto-gallberry fuel complexes 
U.S. Department of Agriculture Forest Service, 
Research Paper SE-174, 44 p. U.S. Department of 
Agriculture Forest Service, Southeastern Forest 
Experiment Station, Asheville, North Carolina. 

Van Wagner, C. E. 1975. A comparison of the Cana- 
dian and American Forest Fire Danger Rating 
Systems. Petawawa Forest Experiment Station, 
Information Report PS-X-59, 19 p. Chalk River, 
Ontario. 

VanWagner, C. E. 1972. Equilibrium moisture con- 
tents of some fine forest fuels in Eastern Canada. 
Petawawa Forest Experiment Station, Informa- 
tion Report PS-X-36, lip. Chalk River, Ontario. 



10 



U.S. Government Printing Office: 1980 — 766-419/123 Region No. 6 



3DA FOREST SERVICE 



ESEARCH PAPER NC-190 OOVJT. tXKUHQfm 

fEB 171981 











£CH. & ^ 



Height and diameter 
of tamarack seed sources 
in northern Wisconsin 



Don E. Riemenschneider and R. M. Jeffers 



>rth Central Forest Experiment Station 

rest Service, U.S. Department of Agriculture 



Riemenschneider, Don E., and R. M. Jeffers. 

1980. Height and diameter growth of tamarack seed sources in northern 
Wisconsin. U.S. Department of Agriculture Forest Service, Research 
Paper NC-190, 6 p. U.S. Department of Agriculture Forest Service, 
North Central Forest Experiment Station, St. Paul, Minnesota. 

Reports height and diameter growth of tamarack seed sources 
planted in northern Wisconsin and makes recommendations for se- 
lecting the highest-yielding sources. 



KEY WORDS: Larix laricina, seed source test, planting recom- 
mendations. 



North Central Forest Experiment Station 

Forest Service — U.S. Department of Agriculture 

1992 Folwell Avenue 

St. Paul, Minnesota 



Manuscript approved for publication August 7, 1979 

1980 



HEIGHT AND DIAMETER OF TAMARACK 

SEED SOURCES 
IN NORTHERN WISCONSIN 

Don E. Riemenschneider, Plant Geneticist, 

Rhinelander, Wisconsin 

and R. M. Jeffers, Plant Geneticist, 

(currently with the Rocky Mountain Region, Lakewood, Colorado), 

Rhinelander, Wisconsin 



Tamarack (Larix laricina (DuRoi) K. Koch) is one 
of the most widely distributed North American coni- 
fer species. It will grow under a wide range of climatic 
conditions and site characteristics and is one of the 
fastest growing boreal conifers on well-drained sites 
(Roe 1957). Although it accounts for a small propor- 
tion of pulpwood production in the Lake States ( Blyth 
and Hahn 1974), harvesting is increasing (Johnston 
1975). Recent investigations have been designed to 
evaluate genetic variation (Pauley 1965, Jeffers 
1975, Sajdak 1970, Cech et al. 1977) and the potential 
for short rotation intensive culture (Zavitkovski and 
Dawson 1978). 

This paper presents survival, height, and diameter 
data for 13- and 14-year-old trees in a seed source test 
of tamarack. The investigation is part of a coopera- 
tive study of rangewide variation in tamarack begun 
by the University of Minnesota (Pauley 1965). Ear- 
lier height and survival data have been reported by 
Jeffers (1975). 



MATERIALS AND METHODS 

Two experimental plantings were established on 
old field sites in north-central Wisconsin in October 
1967. The planting sites and experimental design 
have been previously described (Jeffers 1975). 

One planting consists of 24 seed sources and is 
located on a well-drained, high quality tamarack site 
in Forest County (table 1) (Jeffers 1975). The other 
planting is in Oneida County and includes 17 of the 
same seed sources. In both cases, a few seed sources 



were planted as 2-2 transplants but the majority 
were planted as 3-0 seedlings. A randomized com- 
plete block design with 10 replicates and 4-tree row 
plots was used at both locations. 

All trees were measured in November 1977 for 
total height and diameter at breast height. Data were 
subjected to analysis of variance with plot means as 
entries. Variation among sources was then parti- 
tioned to determine if significant differences existed 
between age groups and between sources within each 
age group. The relation between seed source perfor- 
mance and the latitude and longitude of origin was 
examined using linear and curvilinear regression 
models. Relations among variables were tested by 
simple correlation analysis. 

RESULTS 
Survival 

Mean survival for 13- and 14-year-old trees at the 
Forest County site was 76 percent and 91 percent, 
respectively (table 1). Most of the losses occurred in 
the first 2 years after planting (Jeffers 1975). Analy- 
sis of variance showed significant differences in sur- 
vival between age groups and among 13-year-old 
sources but not among 14-year-old sources (table 2). 

Mean survival for 13- and 14-year-old trees at the 
Oneida County site was 60 percent, which is consid- 
erably less than the survival at the Forest County 
site (table 1). Mortality did not drop off after 2 years 
like it did at the Forest County site. 



Table 1. — Origin, survival, height, and diameter of tamarack seed sources 



Source 


State or 
Province 


County 


Latitude 


°W 
Longitude 


Forest County 




Oneida County 


Age no. 


Survival 


Height 


D.b.h. 


Survival 


Height 


D.b.h. 












Percent 


m 


cm 


Percent 


m 


cm 


14 3036 


ME 


Somerset 


45.7 


71.2 


88 


6.36 


8.28 








3019 


Wl 


EauClaire 


44.7 


91.0 


98 


5.51 


6.79 


68 


4.85 


6.75 


3014 


MN 


Anoka 


45.1 


93.0 


100 


5.43 


6.57 


72 


4.76 


6.05 


3007 


Wl 


Washburn 


46.0 


91.8 


90 


5.41 


6.55 








3038 


Ml 


VanBuren 


42.2 


86.1 


90 


5.18 


5.82 


60 


4.36 


5.32 


3011 


Wl 


Waukesha 


43.0 


88.2 


82 


5.16 


5.67 


45 


4.68 


5.50 










Average 


91 


5.51 


6.61 


61 


4.66 


5.90 


13 3333 


NS 


Annapolis 


44.8 


65.0 


75 


5.05 


5.61 


80 


4.86 


5.89 


3319 


MN 


Anoka 


45.2 


93.1 


52 


5.04 


5.66 


42 


3.63 


3.96 


3266 


Wl 


Oneida 


45.8 


89.2 


90 


4.96 


5.43 








3330 


ME 


Somerset 


45.6 


70.3 


85 


5.02 


5.80 


72 


4.40 


5.29 


3282 


Wl 


LaCrosse 


43.8 


91.1 


80 


5.02 


5.67 


45 


4.59 


5.61 


3265 


Wl 


Forest 


45.8 


88.9 


88 


4.94 


5.54 


58 


4.13 


4.50 


3323 


Wl 


Trempaleau 


44.2 


91.5 


80 


4.89 


5.18 








3332 


ONT 


Oxford 


43.2 


80.6 


62 


4.81 


5.69 


72 


4.58 


5.24 


3284 


MN 


Itasca 


47.4 


93.6 


85 


4.76 


5.14 


52 


4.19 


5.13 


3272 


Ml 


Alger 


46.5 


87.0 


82 


4.72 


5.16 








3283 


MN 


Itasca 


47.5 


94.1 


95 


4.61 


4.85 


65 


4.10 


4.57 


3320 


MN 


St. Louis 


47.0 


93.0 


80 


4.55 


4.91 


65 


3.94 


4.37 


3324 


MAN 




50.1 


95.4 


90 


4.55 


4.64 


60 


4.16 


5.30 


3327 


Ml 


Chippewa 


46.3 


84.2 


55 


4.55 


4.99 


58 


4.94 


6.14 


3331 


Ml 


Houghton 


47.0 


88.4 


78 


4.40 


4.69 


45 


3.83 


4.30 


3273 


Ml 


Alger 


46.5 


87.0 


68 


4.35 


4.81 


58 


3.86 


4.73 


3337 


Ml 


Ingham 


42.5 


84.8 


62 


3.66 


3.48 








3332 


ATA 




56.6 


111.2 
Average 


68 


3.04 


3.08 










76 


4.61 


5.02 


59 


4.25 


5.00 






Least significant difference 


18 


0.61 


0.93 


25 


0.55 


0.85 



Table 2. — Mean squares for survival, height, and diameter of tamarack seed source 







Forest County 


Oneida County 






Mean square 


df 


Survival 


Height 


Diameter 


df 


Survival 


Height 


Diameter 


Among sources 


23 


1 2.674 


1 3.988 


1 10.728 


16 


2 2.031 


1 1.624 


1 5.862 


Between age groups 


1 


1 32.289 


1 36.453 


1 113.128 


1 


2 0.163 


M.370 


1 26.888 


Among 14-year-old sources 


5 


2 0.670 


11.984 


1 8.783 


3 


2 2.300 


2 0.383 


1 4.280 


Among 13-year-old sources 


17 


1 1.404 


1 2.668 


1 5.276 


12 


2 1.119 


1 1.706 


'4.503 


Error 


205 


0.678 


0.482 


1.130 


129 


1.293 


0.392 


0.947 


'Significant at P < 0.01. 


















2 Not significant. 



















Height 

Mean height at Forest County was 5.51 m for 14- 
year-old trees and 4.61 m for 13-year-old trees. All 
components of variation between and within sources 
were significant (table 2). The mean of the shortest 
14-year-old trees (Source 3011, Waukesha County, 
WI) was greater than the mean of the tallest 13-year- 
old trees (Source 3333, Annapolis County, Nova Sco- 
tia). The tallest trees in both age groups were from 
Wisconsin, central Minnesota, Maine, and Nova Sco- 
tia. However, trees from Source 3319 from Anoka 
County, Minnesota, have shown susceptibility to 
snow damage (Sajdak 1970, Jeffers 1975). The short- 
est trees were from Michigan, northern Minnesota, 
and northwestern Canada. 

Mean height at Oneida County was 4.66 m for 14- 
year-old trees and 4.25 m for 13-year-old trees (10 
percent less than at Forest County). Differences were 
significant between age groups and among sources of 
13-year-old trees but not among sources of 14-year- 
old trees. The tallest 13-year-old trees were from 
Chippewa County, Michigan, and Annapolis County, 
Nova Scotia. Source 3319 from Anoka County, Min- 
nesota, was the poorest source at Oneida County and 
was one of the best sources at Forest County. 



Diameter 

Results for this variable were similar to those for 
total height. The most notable difference was that the 



range of seed source means was greater than for 
height (96 percent of the mean for d.b.h. vs. 69 per- 
cent for height at Forest County and 54 percent vs. 28 
percent at Oneida County). Mean d.b.h. at Forest 
County was 6.61 cm for 14-year-old trees and 5.02 cm 
for 13-year-old trees. Age groups overlapped some- 
what. Fourteen-year-old trees from Source 3011 had 
a mean d.b.h. of 5.67 cm while 13-year-old trees from 
Source 3330 had a mean d.b.h. of 5.80 cm. The tallest 
trees also had the largest diameters. 

Differences were also significant between age 
groups and among sources within age groups in One- 
ida County. Again, sources that ranked high in 
height also ranked high in diameter. The poorest 
source once more was 3319 from Anoka County, 
Minnesota. 

Correlation and Regression 
Analyses 

Simple correlation analysis was performed be- 
tween all variables measured in this study at ages 13 
and 14 and also the 8- and 9-year heights presented 
by Jeffers ( 1975) (table 3). The age-age correlation for 
total height was remarkably high— 0.92 at the Forest 
County site and 0.81 at the Oneida County site. The 
height-diameter correlations were 0.98 at Forest 
County and 0.93 at Oneida County. The correlation 
between seed source means at different sites was 0.55 
for height and 0.59 for d.b.h. 

Height and diameter at the Forest County planting 
were significantly and negatively correlated with 



Table 3. — Simple correlations 





1 


2 


3 


4 


5 


6 


7 


8 


Forest County 


















7,8-year height 


















13,14-year survival 


1 2 0.58 
















13,14-year height 


1 .92 


3 0.49 














13,14-year d.b.h. 


1 .93 


3 .45 


1 0.98 












Oneida County 


















7,8-year height 


\82 


4 .42 


1 .62 


1 0.85 










13,14-year survival 


4 .06 


4 .29 


4 .22 


4 .31 


4 0.39 








13,14-year height 


4 .47 


4 .17 


3 .55 


3 .56 


1 .81 


4 0.45 






13,14-year d.b.h. 


3 .57 


4 .28 


3 .58 


3 .59 


1 .86 


4 .15 


1 0.93 




Source latitude 






1 -.54 


3 -.46 






4 -.44 


4 -0.30 


Source longitude 






3 -.49 


3 -.47 






4 -.40 


4 .26 



Significant at P < 0.01. 

2 df = 22 for Forest County; 15 for Oneida County; 15 between sources common to both locations. 
Significant at P < 0.05. 
"Not significant. 



latitude and longitude of the source origin. The effect 
of latitude is easily interpreted as a photoperiod con- 
trolled adaptation to environmental gradients, but 
the effect of longitude is not as easily understood. It 
should be noted that latitude and longitude were 
significantly correlated (r = 0.56), i.e., the collections 
ranged from northwestern Canada to the eastern 
United States, suggesting that the relation between 
growth and longitude of seed source may be spurious. 
Partial correlation analysis did not provide an an- 
swer to this question. 

The effect of latitude of origin on tree height at the 
Forest County site was also investigated via linear 
regression of seed source mean height on latitude of 
origin for the 18 sources of 13-year-old trees. The 
regression (P < 0.025) accounted for 38 percent of the 
variation in height (fig. 1). There was a response 
gradient of 2 percent of the experiment mean per 
degree of latitude with trees of more northern lati- 
tudes performing poorly. An examination of raw data 
and standardized residuals indicated, however, that 
they departed from the linear model. Two sources 
from north and south of the planting site (3332 and 
3337) were found to lie more than two standard devi- 
ations below the regression line. Although a third 



degree polynominal model accounted for 69 percent 
of the variation, the cubic independent variable was 
on the borderline of significance (P < 0.10) and the 
relation was not investigated further. 



DISCUSSION 

Seed sources of tamarack differed significantly in 
survival, height, and diameter at two sites in north- 
central Wisconsin. The presence of significant differ- 
ences tended to depend on the age group of the trees 
and the site on which they were grown — more differ- 
ences existed among seed sources on the better site 
and among seed sources of the younger trees. How- 
ever, this may be a result of differences in precision 
due to the unequal numbers of sources in the two age 
groups. 

The site with the best survival and growth was a 
well-drained upland site adjacent to a stand of natu- 
ral tamarack (Jeffers 1975). The best trees at this 
location were from Wisconsin, central Minnesota, 
Maine, and Nova Scotia. In other cooperative tests, a 
source from Clare County, Michigan, was tallest 




42 



48 50 52 

LATITUDE OF ORIGIN (°N) 



54 



56 



Figure 1. — The regression of mean height on latitude 
of origin for 18 sources of 13 -year-old tamarack. 
The regression equation is height = 9.43 — .10 (°N 
Latitude). 



(Sajdak 1970, Cech et al. 1977). This source was not 
represented in the Wisconsin tests, but sources from 
similar latitudes (3282 and 3323) ranked high. 

Due to the limited nature of the experimental pop- 
ulation, no estimate of genotype x environment in- 
teraction was made. Some inference, however, can be 
made on the basis of seed source rank and correla- 
tions. The correlations between sites were 0.55 for 
height and 0.59 for d.b.h. (r 2 = 0.30 and 0.35). Al- 
though the correlations are significant (P < 0.01), 
they are not extremely high values and reflect a lack 
of consistent performance over the two sites. An ex- 
amination of seed source means shows that sources 
3319 (Anoka County, Minnesota) and 3327 (Chip- 
pewa County, Michigan) are extremely unstable in 
this experiment (fig. 2). The reason for this instabil- 
ity remains unknown however, because the two 
sources do behave erratically, they may not be suited 
to planting in north-central Wisconsin. 

Due to the strong age-age and height-diameter 
correlations, there appears to be no reason to modify 
the following seed source recommendations for 
north-central Wisconsin (Jeffers 1975): 



1. Maine, Somerset County (3036 and 3330) 

2. Wisconsin, Eau Claire County (3019) 

LaCrosse County (3282) 
Oneida County (3266) 

3. Nova Scotia, Annapolis County (3333) 
These sources are those that grew best on the highest 
yeilding site and also appeared to have reasonably 
stable performance. Although collections from the 
original stands represented in this study may be the 
best approach to seed procurement, the curvilinear 
regression of mean seed source height on latitude of 
origin indicates that there may be a range (43.5°N to 
46.0°N) where high performing genotypes can be 
found. However, because one of the sources that lead 
to the fitting of a polynominal model is present at 
only one location and a source with known instability 
(3319) is included in the latitudinal range, there may 
be some risk associated with such a generalization. 



LITERATURE CITED 



I 

o 

uj 

Uj 
O 

a 

=> 
O 

CO 

Q 

Uj 
Uj 
CO 

2 
Uj 



5.50 



5.00 



4.50 



4.00 



3.50 



3.00 



FOREST COUNTY 
SITE 



5.50 




3.00 

ONEIDA COUNTY 
SITE 



Figure 2. — Height of 13 -year-old trees from 13 
sources of tamarack at two locations in northern 
Wisconsin. 



Blyth, J. E., and J. T. Hahn. 1974. Pulpwood produc- 
tion in the north central region by county 1974. 
U.S. Department of Agriculture Forest Service, 
Resource Bulletin NC-29, 26 p. U.S. Department 
of Agriculture Forest Service, North Central For- 
est Experiment Station, St. Paul, Minnesota. 

Cech, F. C, R. N. Keys, and D. H. Weingartner. 1977. 
Seventh-year results of a tamarack provenance 
study. In Twenty-fourth Northeastern Tree Im- 
provement Conference Proceedings, 1976. p. 55- 
65. University of Maryland Center for Environ- 
mental and Estuarine Studies, College Park, 
Maryland. 

Jeffers, R. M. 1975. Survival and height growth of 
tamarack planted in northern Wisconsin. U.S. 
Department of Agriculture Forest Service, Re- 
search Note NC-190, 3 p. U.S. Department of Ag- 
riculture Forest Service, North Central Forest 
Experiment Station, St. Paul, Minnesota. 

Johnston, W. F. 1975. Reproducing lowland conifer 
forests. Journal of Forestry 37(l):17-20. 

Pauley, S. S. 1965. Seed sources of tamarack, Larix 
laricina (DuRoi) K. Koch. In Fourth Central 
States Forest Tree Improvement Conference Pro- 
ceedings, 1964. p. 31-34. Nebraska Agricultural 
Experiment Station, Lincoln, Nebraska. 

Roe, E. I. 1957. Silvical characteristics of tamarack. 
U.S. Department of Agriculture Forest Service, 



Station Paper 52, 22 p. U.S. Department of Agri- 
culture Forest Service, Lake States Forest Exper- 
iment Station, St. Paul, Minnesota. 
Sajdak, R. L. 1970. Provenance testing at Michigan 
Technological University. In Ninth Lake States 
Forest Tree Improvement Conference Proceed- 
ings, 1969. U. S. Department of Agriculture For- 
est Service, Research Paper NC-47, p. 4-5. U.S. 



Department of Agriculture Forest Service, North 
Central Forest Experiment Station, St. Paul, 
Minnesota. 

Zavitkovski, J., and D. H. Dawson. 1978. Structure 
and biomass production of 1- to 7-year-old tama- 
rack in Wisconsin. Journal of the Technical 
Association of the Pulp and Paper Industry 
61(6):68-70. 



•j A f U.S. Government Printing Office: 1981 — 766-505/135 Region No. 6 



USDA FOREST SERVICE 
RESEARCH PAPER NC-191 




GOVT. DOCUMENTS 

depository: itew 

MAR 30 1981 

CLfMSON 
LIBRARY 



CH. & ^ 



Summary of 
green weights 

and volumes for 

five tree species 

in Michigan 

Sharon A. Winsauer and Helmuth M. Steinhilb 



North Central Forest Experiment Station 
Forest Service, U.S. Department of Agriculture 



North Central Forest Experiment Station 

Forest Service — U.S. Department of Agriculture 

1992 Folwell Avenue 

St Paul, Minnesota 55108 

Manuscript approved for publication February 19, 1980 

1980 



SUMMARY OF GREEN WEIGHTS 
AND VOLUMES FOR FIVE TREE SPECIES 

IN MICHIGAN 



Sharon A. Winsauer, Computer Specialist, 

and Helmuth M. Steinhilb, Research Forester, 

Houghton, Michigan 



In the past few years more and more small, whole 
xees, and the residue remaining from conventional 
ogging operations have been converted to chips for 
Iber or fuel. As fossil fuels become scarcer, chipping 
rees and logging debris in the woods will probably 
)ecome a common way to harvest much of this material 
vhich was previously considered unmerchantable. So, 
oresters and loggers increasingly need to estimate the 
veight of trees, boles, and logging residue. 

Studies made in the western part of the Upper Penin- 
iula of Michigan in 1970 (Steinhilb and Erickson 
.970), 1972 (Steinhilb and Erickson 1972), and 1976 
Steinhilb and Winsauer 1976) produced estimating 
•quations, tables, and graphs for the weights and cubic 
bot volumes of trees, boles, and residue for aspen, 
pruce, balsam fir, red pine, and sugar maple. This 
nformation is summarized here in a condensed and 
nore usable form. 



Definitions: the "total tree" is the entire tree 
above the stump, while the "bole" is the delimbed 
stem cut at a point 3 inches in diameter outside 
the bark for all species except sugar maple, where 
the top diameter of the bole was 4 inches outside 
the bark. "Residue" is all remaining portions of 
the tree except the stump and roots, including 
tops, limbs, and foliage. 



In the original studies, field data were obtained from 
7 aspen, 71 balsam fir, 58 white spruce, 58 red pine 
nd 58 pulpwood-sized and 79 saw log-sized sugar ma- 
le trees. 



FIELD PROCEDURES 

Each tree was numbered and its d.b.h. recorded to 
ie nearest 0.1 inch and felled carefully to minimize 



breakage and loss of limbs. Each tree was weighed 
within 24 hours of felling. 

All limbs were severed and the main stem lopped at 
its 3-inch-diameter point outside the bark (d.o.b.) ex- 
cept for sugar maple which was severed at 4-inch d.o.b. 
Taper measurements were taken on each stem outside 
the bark at the stump cut, at 2 feet and 4 feet above the 
stump and every 8 feet thereafter to the point of lop- 
ping. Total length of the tree and the bole were also 
measured to the nearest foot. 



ANALYTIC PROCEDURE 

A standard form of estimating equation for all five 
species was felt desirable. The equations used in the 
sugar maple study (Steinhilb and Winsauer 1976) 
were chosen because of their simplicity and the gen- 
eral availability of the independent variables from 
timber inventory. 

Cubic foot volume was calculated for the whole tree 
and its residue from the tree weight and pound per 
cubic foot figures. Regression equations were devel- 
oped for each species. Several other regression mod- 
els were also tested to insure that the common form 
chosen was sufficiently accurate for all species. 



RESULTS AND DISCUSSIONS 

The variance in weight and volume measurements 
increased as the diameter of the trees increased, at a 
rate of approximately (d.b.h.) 4 . Therefore, the final 
equations were obtained from weighted regressions 
with a weighting factor of (1/d.b.h.) 4 to compensate 
for the nonhomogenity of weights and volumes of 
trees, boles, and residue. 



The saw log size (12 inches d.b.h. or larger) sugar 
maple trees were included in the analyses of tree 
weights and volumes but could not be used for the 
bole and residue equations because their bole length 
had been defined differently. 

The similarity of the tree weight curves (Young 
1976) would suggest the possibility of creating a set of 
general curves to predict green tree weight of any 
species. It also seems to imply that if d.b.h. is known, 
a practical estimate of green tree weight can be ob- 
tained regardless of species. Although a larger varia- 
tion appears in the residue weight curves, these also 
indicate the possibility of a set of general curves. 



GRAPHS AND 

Subject 

Tree weight vs. d.b.h. 

and tree height 
Tree volume vs. d.b.h. 

and tree height 
Bole weight vs. d.b.h. and 

bole length 
Bole volume vs. d.b.h. and 

bole length 
Residue weight vs. d.b.h. 
Residue volume vs. d.b.h. 



EQUATIONS 

Table Figure 

No. Page No. Page 

1-5 3-7 1-5 14-15 

i 

1-5 3-7 6-10 15-16 

6-10 8-12 11-15 16-17 

6-10 8-12 16-20 17-18 

11 13 21-25 19-20 

11 13 26-30 20-22 



Tables of tree weights and volumes with 95 
percent confidence limits on the mean for each spe- 
cies are presented in tables 1-5. Tables of bole 
weights and volumes with 95 percent confidence lim- 
its on the mean are presented in tables 6-10. Residue 
weights and volumes for all five species are presented 
in table 11. 



Two additional graphs are included — one for tre 
weight for all five species based on d.b.h. (fig. 31) an 
one containing all five residue weight curves (fig. 32 



LITERATURE CITED 

Steinhilb, H. M., and John R. Erickson. 197 
Weights and centers of gravity for quaking aspt 
trees and boles. U.S. Department of Agricultu 
Forest Service, Research Note NC-91, 4 p. U. 
Department of Agriculture Forest Service, Nor 
Central Forest Experiment Station, St. Paul, Mi 
nesota. 

Steinhilb, H. M., and John R. Erickson. 197 
Weights and centers of gravity locations for r 
pine, white spruce, and balsam fir trees and bolt 
U.S. Department of Agriculture Forest Servit 
Research Paper NC-75, 7 p. U.S. Department 
Agriculture Forest Service, North Central Fore 
Experiment Station, St. Paul, Minnesota. 

Steinhilb, H. M., and Sharon A. Winsauer. 19 1 ; 
Sugar maple: tree and bole weights, volumes, ce 
ters of gravity and logging residue. U.S. Depa 
ment of Agriculture Forest Service, Research I 
per NC-132, 7 p. U.S. Department of Agricultu 15 
Forest Service, North Central Forest Experime 
Station, St. Paul, Minnesota. 

Young, Harold E. 1976. A summary and analysis 
weight table studies, p. 75-99. Oslo Biomass Sti 
ies, College of Life Sciences and Agriculture, U 
versity of Maine, Orono. 



kii* 



i'able 1. — Aspen tree weight and tree volume (95 percent confidence limit on the mean) 















Tree length (feet) 












D.b.h. 


30 




40 




50 


60 




70 




80 




Weight 


Volume 


Weight 


Volume 


Weight 


Volume 


Weight 


Volume 


Weight 


Volume 


Weight 


Volume 


nches 


Pounds 


Feet 3 


Pounds 


Feet 3 


Pounds 


Feet 3 


' Pounds 

1 


Feet 3 


Pounds 


Feet 3 


Pounds 


Feet 3 


5 


142.5 


3.52 


182.0 


4.27 


221.6 


5.04 


261.1 


5.80 


300.7 


6.56 


340.2 


7.32 




(24. 2) 1 


(.57) 


(23.1) 


(-61) 


(22.0) 


(.67) 


I (21.1) 


(.74) 


(20.2) 


(.82) 


(19.5) 


(.91) 


6 


194.7 


4.52 


251.6 


5.62 


308.6 


6.71 


365.5 


7.81 


| 422.5 


8.91 


479.4 


10.00 


I 


(22.7) 


(.63) 


(21.3) 


(.73) 


(20.1) 


(-84) 


(19.1) 


(.97) 


1 (18.5) 


(1.10) 


(18.1) 


(1.24) 


e- 7 


256.4 


5.71 


333.9 


7.20 


411.4 


8.69 


488.9 


10.18 


I 566.4 


11.67 


643.9 


13.17 


IB 


(21.2) 


(.74) 


(19.6) 


(90) 


(18.6) 


(107) 


(18.1) 


(1.26) 


| (18.3) 


(1.46) 


(19.1) 


(1-66) 


8 


327.5 


7.08 


428.8 


9.03 


530.0 


10.98 


631.3 


12.92 


732.5 " 


14.87 


1 833.8 


16.82 


11 


(19.7) 


(.88) 


(18.4) 


(1.12) 


(18.1) 


(1.37) 


(19.0) 


(1.63) 


(20.7) 


(1.89) 


1 (23.3) 


(2.16) 


» 9 


408.2 


8.63 


536.4 


11.10 


664.5 


13.56 


792.7 


16.03 


920.8 


18.50 


|1 ,048.9 


20.96 




(18.6) 


(1.07) 


(18.2) 


(1.38) 


(19.4) 


(1.71) 


(22.2) 


(2.05) 


(25.8) 


(2.39) | (30.2) 


_(2_71L 


; 10 


498.4 


10.37 


I 656.6 


13.41 


814.8 


16.46 


973.0 


19.50 


1,131.2 


22.55 


1,289.4 


25.59 


n 
e| 11 


(18.1) 


(1.29) | (19.3) 


(1-69) 


(22.7) 


(2.11) 


(27.5) 


(2.53) 


(33.1) 


(2.96) 


(39.2) 


(3.39) 


598.1 


12.29 


| 789.5 


15.97 


980.9 


19.65 


1.172.3 


23.34 


1,363.8 


27.02 


1,555.2 


30.70 


c 
t 


(18.6) 


(1.54) 


LJ22.1) 


(2.04) 


(27.8) 


(2.55) 


(34.7) 


(3.07) 


(42.2) 


(3.59) 


(49.9) 


(4.11) 


12 


707.2 


14.39 


935.0 


18.77~ll, 162.8 


23.16 


1,390.6 


27.54 


1,618.5 


31.93 


1,846.3 


36.31 




(20.2) 


(1.83) 


(26.3) 


(2.43) 


(34.3) 


(3.05) 


(43.2) 


(3.66) 


(52.6) 


(4.28) 


(62.1) 


(4.91) 


„.13 


825.9 


16.67 


1,093.2 


21.82 


1,360.6 


26.96 


1,627.9 


32.11 


1,895.3 


37.25 


2,162.7 


42.40 


e 


(23.0) 


(2.14) 


(31.8) 


(2.86) 


(42.0) 


(3.58) 


(53.0) 


(4.31) 


(64.2) 


(5.04) 


(75.7) 


(5.77) 


■ 14 


954.0 


19.14 


1,264.1 


26.10 


1,574.2 


31.07 


1,884.2 


37.04 


2,194.3 


43.01 


2,504.4 


48.98 


1 


(26.9) 


(2.48) 


(38.2) 


(3.32) 


(50.7) 


(4.16) 


(63.7) 


(5.01) 


(77.0) 


(5.86) 


(90.4) 


(6.71) 


ul 15 


1,091.7 


21.79 


1,447.6 


28.64 


TT803.6 " 


"3519" 


2,159.5 


42.34 


2,515.5 


49.19 


2,871.4 


56.04 


w 


(31.7) 


(2.85) 


(45.5) 


(3.82) 


(60.3) 


(4.79) 


(75.5) 


(5.76) 


(90.9) 


(6.74) 


(106.4) 


(7.71) 


16 


1,238.8 


24.62 


1,643.8 


32.41 


2,048.8 


40.21 


2,453.8 


48.00 


2,858.7 


55.80 


3,263.7 


63.59 


is 


(37.2) 


(3.25) 


(53.6) 


(4.35) 


(70.8) 


(5.46) 


(88.2) 


(6.57) 


(105.8) 


(7.68) 


(123.6) 


(8.79) 


' 17 


1,395.4 


27.63 


1,852.6 


36.43 


2,309.8 


45.23 


2,767.0 


54.03 


3,224.2 


62.83 


3,681.4 


71.63 


Ui 


(43.4) 


(3.68) 


(62.4) 


(4.92) 


(82.0) 


(6.17) 


(101.8) 


(7.43) 


(121.8) 


(8.68) 


(141.9) 


(9.93) 


18 


1,561.5 


30.83 


2,074.1 


40.70 


2,586.6 


50.56 


3,099.2 


60.43 


3,611.8 


70.29 


4,124.3 


80.16 




(50.2) 


(4.13) 


(71.8) 


(5.53) 


(94.0) 


(6.93) 


(116.4) 


(8.34) 


(138.8) 


(9.74) 


(161.4) 


(11.15) 



le numbers in parentheses give a confidence interval on the mean. 



Table 2. 


— White spruce tree i 


veight and tree volume (95 percent confidence limit 


on the mean). 














Tree length (feet) 












30 




40 




50 




60 




70 




D.b.h. 


Weight 


Volume 


Weight 


Volume 


Weight 


Volume 


Weight 


Volume 


Weight 


Volumi 


Inches 


Pounds 


Feet 3 


Pounds 


Feet 3 


Pounds 


Feet 3 


Pounds 


Feet 3 


Pounds 


Fee 


5 


273.5 


4.00 


313.7 


4.86 


354.0 


5.72 


I 394.2 


6.58 


434.5 


7.4' 




(72. 7) 1 


(1.10) 


(69.2) 


(1.05) 


(65.8) 


(1.00) 


L_ ( 62 - 7 ) 


(.95) 


(59.8) 


(.9 


G 


326.6 


5.14 


384.6 


6.38 


442.5 


7.62 


500.5 


8.85 


558.4 


10.0<j 




(68.1) 


(1.03) 


(63.4) 


(.96) 


(59.2) 


(.90) 


(55.6) 


(.84) 


(52.6) 


(•8( 


7 


389.4 


6.48 


468.3 


8.17 


547.2 


9.85 


626.1 


11.54 


I 705.0 


13.21; 




(63.1) 


(.96) 


(57.5) 


(-87) 


(53.2) 


(.81) 


(50.3) 


(.76) 


I (49-1) 


(-7- 


8 


461.8 


8.03 


564.9 


10.23 


667.9 


12.43 


771.0 


14.64 


874.0 


16.3 




(57.9) 


(.88) 


(52.4) 


(-79) 


(49.4) 


(-75) 


(49.5) 


(■75) 


(52.7) 


(J 


9 


544.0 


9.78 


674.4 


12.57 


804.8 


15.36 


935.2 


18.15 


1,065.6 


20.9i 




(53.3) 


(-81) 


(49.3) 


(-75) 


(50.2) 


(.76) 


(55.8) 


(85) 


(64.9) 


(.9, 


10 


635.7 


11.75 


796.7 


15.19 


957.7 


18.63 


1,118.7 


22.07 


1,279.7 


25.5 




(50.0) 


(.76) 


(50.0) 


(.75) 


(57.2) 


(.87) 


(69.3) 


(1.05) 


(84.3) 


(1-1 


11 


737.2 


13.91 


932.0 


18.08 


1,126.8 


22.24 


1,321.6 


26.41 


1,516.4 


30.5 




(49.1) 


(.74) 


(55.7) 


(.84) 


(70.0) 


(1.06) 


(88.4) 


(1-34) 


(108.9) 


(1.1 


12 


848.2 


16.29 


1,080.1 


21.25 


1,311.9 


26.20 


1,543.8 


31.16 


1,775.6 


36.1 




(51.6) 


(-78) 


(66.1) 


(1.00) 


(87.5) 


(1.32) 


(111.9) 


(1.69) 


(137.7) 


(2.0 


13 


969.0 


18.87 


1 1,241.1 


24.69 


1,513.2 


30.50 


1,785.3 


36.32 


2,057.4 


42.1 




(57.9) 


(.88) 


| (80.5) 


(1.22) 


(108.6) 


(1.64) 


(138.8) 


(2.10) 


(170.1) 


(2.E , 


14 


1,099.4 


21.66 


| 1,415.0 


28.40 


1,730.5 


35.15 


2,046.1 


41.90 


2,361.6 


48.e 




(67.7) 
1,239.5 


(1.02) 
24.65 


| (98.1) 


(1.49) 


(132.6) 
| 1,964.0 


(2.01) 
40.14 


(168.8) 
2,326.2 


(2.56) 
47.89 


(205.8) 
2,688.5 


( 3 - 1 :iS 
55.1 


15 


1,601.7 


32.40 




(80.3) 


(1.22) 


(118.3) 


(1-79) 


(159.3) 


(2.41) 


(201.6) 


(3.05) 


(244.5) 


(3.- |£ 


16 


1,389.2 


27.85 


1,801.4 


36.67 


I 2,213.5 


45.48 


2,625.7 


54.29 


3,037.8 


63/ ' 




(95.4) 


(1.44) 


(140.7) 


(2.13) 


Q188.4) 


(2.85) 


(237.1) 


(3.59) 


(286.2) 


(4.| 


17 


1,548.6 


31.26 


2,013.9 


41.21 


2,479.2 


51.16 


2,944.5 


61.10 


3,409.8 


71.1 




(112.4) 


(1.70) 


(165.1) 


(2.50) 


(219.7) 


(3.33) 


(275.0) 


(4.16) 


(330.8) 


Hi! 


18 


1,717.6 


34.88 


2,239.3 


46.03 


2,760.9 


57.18 


3,282.6 


68.33 


3,804.2 


79. 




(131.2) 


(1.99) 


(191.4) 


(2.90) 


(253.1) 


(3.83) 


(315.5) 


(4.78) 


(378.2) 


(5.'p 


1 The numbers in parentheses 


give a confidence interval 


on the mean 





















ible 3. — Red pine tree weight and tree volume (95 percent confidence limit on the mean) 













Tree length (feet) 










.b.h. 


40 




50 




60 




70 




80 




Weight 


Volume 


Weight 


Volume 


Weight 


Volume 


Weight 


Volume 


Weight 


Volume 


iches 


Pounds 


Fee? 


Pounds 


Feet 3 


Pounds 


Feel 3 


I Pounds 


Feel 3 


Pounds 


Feel 3 


1 5 


187.5 


3.59 


232.5 


4.34 


277.6 


5.09 


| 322.6 


5.84 


367.7 


6.59 




(18.1) 1 


(.51) 


(17.0) 


(.53) 


(16.0) 


(.55) 


j (15-D 


(.58) 


(14.2) 


(.60) 


6 


266.8 


4.91 


331.6 


5.99 


396.5 


7.08 


461.4 


8.16 


I 526.3 


9.24 


i 


(16.3) 


(.55) 


(14.9) 


(.58) 


(13.8) 


(62) 


(12.8) 


(.65) 


|_(12.2) 


(.69) 


7 


360.5 


6.47 


448.8 


7.95 


537.1 


9.42 


625.4 


10.90 


713.7 


12.37 




(14.4) 


(-60) 


(13.0) 


(.65) 


(12.1) 


(.70) 


(11.8) 


(.74) 


(12.2) 


(.79) 


8 


468.6 


8.28 


583.9 


10.20 


699.3 


12.13 


814.6 


14.05 


929.9 


15.98 




(12.7) 


(.66) 


(11.9) 


(.72) 


(12.1) 


(-79) 


(13.5) 


(.85) 


(15.6) 


(-92) 


9 


591.1 


10.32 


737.1 


12.76 


883.1 


15.20 


1,029.0 


17.63 


1,175.0 


20.07 




(11.9) 


(.73) 


(12.5) 


(.81) 


(14.6) 


(-89) 


(17.8) 


(-97) 


(21.5) 


(1.05) 


10 


728.1 


12.61 


908.3 


15.62 


1,088.5 


18.62 


1,268.7 


21.63 


1,448.9 


24.64 




(12.4) 


(.80) 


(15.1) 


(-90) 


(19.2) 


(1-00) 


(24.1) 


(1.11) 


(29.2) 


(1.21) 


11 


879.5 


15.14 


1,097.5 


18.77 


1,315.5 


22.41 


1,533.6 


26.05 


1,751.6 


29.69 




(14.6) 


(89) 


(19.5) 


(1.01) 


(25.4) 


(1.13) 


(31.7) 


(1.26) 


(38.3) 


(1.38) 


12 


1,045.2 


17.90 


1,304.7 


22.23 


1,564.2 


26.56 


1,823.7 


30.89 


2,083.2 


35.22 




(18.2) 


(.98) 


(25.1) 


(1.13) 


(32.6) 


(128) 


(40.5) 


(1.42) 


(48.5) 


(1.57) 


13 


1,225.4 


20.91 


M ,530.0 


25.99 


1,834.5 


31.07 


2,139.1 


36.15 


2,443.6 


41.24 


i 


(22.9) 


(1.08) 


|_(31.6) 


(1.26) 


(40.8) 


(1.43) 


(50.2) 


(161) 


(59.8) 


(1.78) 


14 


1,420.1 


24.16 


1,773.2 


30.05 


12,126.4 


35.94 


2,479.6 


41.84 


2,832.8 


47.73 


15 


' (28.4) 
1,629.1 


(1.19) 
27.64 


(39.0) 
2,034.5 


(1.40) 
34.41 


I (49.9) 


(1.60) 


(60.9) 


(1-80) 


(72.0) 


(2.01) 


2,440.0 


41.18 


2,845.4 


47.94 


3,250.9 


54.71 




(34.6) 


(1-31) 


(47.0) 


(1-55) 


(59.7) 


(1.78) 


(72.4) 


(2.01) 


(85.3) 


(2.25) 


16 


1,852.5 


31.37 


2,313.8 


39.07 


2,775.2 


46.77 


3,236.5 


54.47 


3,697.8 


62.17 




(41.4) 


(1.44) 


(55.7) 


(1-71) 


(70.2) 


(1.97) 


(84.8) 


(2.24) 


(99.5) 


(2.51) 


17 


2,090.4 


35.34 


2,611.2 


44.03 


3,132.0 


52.72 


3,652.7 


61.41 


4,173.5 


70.10 




(48.7) 


(1.58) 


(65.0) 


(1.88) 


(81.5) 


(2.18) 


(98.1) 


(2.48) 


(114.6) 


(2.78) 


18 


2,342.7 


39.55 


2,926.5 


49.30 


3,510.4 


59.04 


4,094.2 


68.78 


4,678.1 


78.52 


! 


(56.6) 


(1.72) 


(75.0) 


(2.06) 


(93.5) 


(2.40) 


(112.1) 


(2.74) 


(130.7) 


(3.07) 



■ numbers in parentheses give a confidence interval on the mean. 



Table 4. — Balsam fir tree weight and tree volume (95 percent confidence limit on the mean) 













Tree length (feet) 












30 




40 




50 




60 




70 




D.b.h. 


Weight 


Volume 


Weight 


Volume 


Weight 


Volume 


Weight 


Volume 


Weight 


Volun 


Inches 


Pounds 


Feet 3 


Pounds 


Feet 3 


Pounds 


Feet 3 


i Pounds 


Feet 3 


Pounds 


Fe 


5 


202.7 


3.79 


245.7 


4.64 


288.7 


5.48 


331.7 


6.32 


374.7 


7. 




(47.4) 1 


(.40) 


(44.2) 


(.39) 


(41.2) 


(-38) 


(38.4) 


(-37) 


(35.9) 


(•: 


6 


259.5 


4.91 


321.4 


6.12 


383.3 


7.34 


445.2 


8.55 


507.1 


9. 




(43.2) 


(.39) 


(39.0) 


(.37) 


(35.4) 


(.36) 


LJ32.5) 


(.36) 


(30.6) 


(•: 


7 


326.5 


6.22 


410.8 


7.88 


495.1 


9.53 


579.4 


11.18 


663.7 


12.1 




(38.7) 


(-37) 


(34.0) 


(.36) 


(30.9) 


(-35) 


(29.7) 


(.35) 


(30.8) 


(•: 


8 


403.9 


7.74 


514.0 


9.90 


624.1 


12.06 


7342 


14.22 


844.3 


16.: 




(34.4) 


(-36) 


(30.4) 


(.35) 


(30.0) 


(35) 


(33.3) 


(.36) 


(39.3) 


(•: 


9 


491.7 


9.46 


631.0 


12.20 


770.3 


14.93 


909.6 


17.66 


1,048.9 


20, 




(31.0) 


(-35) 


(30.1) 


(-35) 


(35.0) 


(.37) 


(43.7) 


(39) 


(54.5) 


(• 


10 


589.7 


11.39 


761.7 


14.76 


933.7 


18.14 


1,105.7 


21.51 


1,277.7 


24. 




(29.7) 


(.35) 


(34.6) 


(.36) 


(45.5) 


(.40) 


(59.2) 


(.45) 


(74.3) 


(• 


11 


698.1 


13.51 


906.2 


17.60 


1,114.3 


21.68 


1,322.4 


25.76 


1,530.5 


29. 




(31.8) 


(36) 


(43.5) 


(39) 


(60.0) 


(.45) 


(78.3) 


(.52) 


(97.4) 


(• 


12 


816.7 


15.84 


J 1.064.4 


20.70 


1,312.1 


25.56 


1,559.8 


30.42 


1,807.5 


35. 




(37.6) 


(.37) 


(55.8) 


(.43) 


(77.3) 


(-52) 


(100.1) 


(.62) 


(123.4) 


(• 


13 


945.7 


18.37 


1 1,236.4 


24.08 


1,527.1 


29.78 


1,817.8 


35.48 . 


2,108.5 


41. j 




(46.4) 


(.40) 


I (70.6) 


(49) 


(97.1) 


(.61) 


(124.4) 


(.74) 


(152.1) 


(• 


14 


1,085.1 


21.11 


1,422.2 


27.72 1 1,759.3 


34.33 


2,096.4 


40.95 


2,433.5 


47.j 




(57.5) 


(.44) 


(87.4) 


(.56) 


(J118.9) 


(.71) 


(151.0) 


(.87) 


(183.4) 


(1 


15 


1,234.7 


24.04 


1,621.7 


31.63 


2,008.7 


39.23 


2,395.7 


46.82 


2,782.7 


54., 




(70.4) 


(.49) 


(105.9) 


(-65) 


(142.6) 


(.82) 


(179.7) 


(1.01) 


(217.1) 


(1 


16 


1,394.7 


27.18 


1,835.0 


35.82 


2,275.3 


44.46 


2,715.6 


53.10 


3,155.9 


61 J 




(84.8) 


(-55) 


(126.0) 


(.74) 


(168.1) 


(95) 


(210.6) 


(1.17) 


(253.2) 


(1- 


17 


1,564.9 


30.53 


2,062.0 


40.27 


2,559.1 


50.03 


3,056.2 


59.78 


3,553.3 


69 j 




(100.6) 


(62) 


(147.7) 


(.85) 


(195.5) 


(1.09) 


(243.6) 


(1.34) 


(291.8) 


(1 


18 


1,745.5 


34.06 


2,302.8 


45.00 


2,860.1 


55.93 


3,417.4 


66.87 


3,974.7 


77, ; 




(117.6) 


(.70) 


(170.8) 


(.97) 


(224.6) 


(1.24) 


(278.6) 


(1.52) 


(332.8) 


d 


'The numbers in parentheses give a confidence interval 


on the mean 















5. — Sugar maple tree weight and tree volume (95 percent confidence intervals on the mean) 















rree height (feet) 






<. 




b.h. 


40 




50 


I 


61 


) 70 


80 


Weight 


Volume 


Weight 


Volume 


Weight 


Volume Weight 


Volume 


Weight 


Volume Weight 


Volume 


ches 


Pounds 


Feel 3 


Pounds 


Feet 


Pounds 


Feet 3 ^Pounds 


Feet 3 


Pounds 


Feet 3 Pounds 


Feet 1 


5 


194.0 


2.82 


236.0 


3.59 


277.9 


4.37 I 320.0 


5.20 


362.0 


5.97 403.9 


6.75 




(21. 3) 1 


(.39) 


(20.8) 


(.37) 


(20.3) 


(.37) I (20.1) 


(.37) 


(19.8) 


(.36) (19.5) 


(.36) 


6 


267.8 


4.18 


328.2 


5.30 


388.7 


6.42 I 449.3 


7.58 


509.7 


8.70 570.2 


9.82 




(20.4) 


(-37) 


(19.8) 


(-36) 


(19.5) 


(-36) | (19.4) 


(.35) 


(19.3) 


(.35) (19.3) 


(-35) 


7 


355.1 


5.80 


437.3 


7.31 


519.5 


8.83 601.8 


10.35 


| 684.3 


11.92 766.6 


13.44 




(19.7) 


(.40) 


(19.2) 


(.34) 


(19.0) 


(.34) (19.3) 


(-35) 


L09.9) 


(.36) (20.6) 


(-38) 


8 


455.8 


7.66 


563.2 


9.64 


670.6 


11.63 778.0 


13.61 


885.3 


1575V 993.3 


17.62 




(19.2) 


(.35) 


(19.1) 


(-35) 


(19.6) 


(.36) (20.6) 


(.37) 


(21.8) 


(.39)1 (23.85) 


(-43) 


9 


569.9 


9.76 


705.8 


12.28 


841.7 


14.79 977.6 


17.30 


1,113.6 


19.81 11 ,250.2 


22.37 




(19.1) 


(-34) 


(19.8) 


(.36) 


(21.3) 


(.39) (23.3) 


(.42) 


(25.8) 


(■47)| (28.9) 


(53) 


10 


697.8 


12.17 


865.3 


15.22 


1,033.0 


18.32 1,200.8 


21.42 


1,368.6 


24.53 11,537.3 


27.67 




(20.0) 


(.36) 


(21.6) 


(-39) 


(24.3) 


(.44) (27.5) 


(.50) 


(31.2) 


(-57) | (35.7) 


(.65) 


11 


838.8 


14.78 


1,041.4 


18.48 


1,244.4 


22.23 1,447.5 


25.98 


1,650.5 


29.73 i1 ,854.6 


33.53 




(21.5) 


(-39) 


(24.4) 


(.44) 


(28.4) 


(.52) (33.1) 


(.60) 


(38.1) 


(.69)' (43.8) 


(-80) 


12 


993.8 


17.63 


1,234.4 


22.04 


1,496.0 


26.51 1,717.6 


30.98 


1,959.3 


35.44 12,202.1 


39.94 


13 


(23.8) 
1,161.2 


(.43) 
20.73 


(28.3) 
1,444.1 


(.51) 
25.92 


(33.8) 
1,727.7 


(.61) (39.8) 
31.16 2,011.3 


(-73) 
36.40 


(46.1) 
2,294.9 


(.83) | (53.1) 


(.97) 


41.64 2.578.4 


46.88 




(27.0) 


(.49) 


(33.0) 


(60) 


(40.0) 


(-73) (47.5) 


(.86) 


(55.1) 


(1.00) (62.9) 


(1.14) 


14 


1,342.5 


24.07 


1,670.6 


30.11 


1,999.5 


36.18 2,328.4 


42.26 


2,657.3 


48.34 2,986.2 


54.42 




(31.0) 


(.56) 


(38.5) 


(.70) 


(47.1) 


(.85) (56.0) 


(1-02) 


(65.1) 


(1.18) (74.3) 


(1.35) 


15 


1,537.3 


27.67 


1,914.0 


34.60 


2,291.5 


41.58 2,669.1 


48.56 


3,046.6 


55.53 3,424.2 


62.51 




(35.7) 


(.65) 


(44.9) 


(.81) 


(55.0) 


(1.00) (65.5) 


(1.19) 


(76.0) 


(1.37) (86.7) 


(1.57) 


16 


1,745.5 


31.51 


2,175.3 


" 39.45 


^2,603.6 


47.35 3,033.2 


55.29 


3,462.8 


63.22 3,892.3 


71.16 




(40.9) 


(.74) 


(52.4) 


(-95) 


(63.6) 


(1.16) (75.7) 


(1.38) 


(87.8) 


(1.59) (100.0) 


(1.81) 


17 


1,967.1 


35.60 


2,452.3 


44.56 


2,935.9 


53.49 3,420.8 


62.45 


3,905.7 


71.41 4,390.7 


80.37 




(46.8) 


(-85) 


(60.0) 


(1.09) 


(72.9) 


(1.32) (86.6) 


(1.57) 


(100.4) 


(1.82) (114.3) 


(2.07) 


18 


2,202.1 


39.94 


2,746.1 


49.99 


3,288.2 


60.00 3,831.9 


70.05 


4,375.6 


80.09 4,919.3 


90.14 




(53.1) 


(.97) 


(68.2) 


(1.24) 


(82.8) 


(1.51) (98.3) 


(1.79) 


(113.9) 


(2.06) (129.5) 


(2.35) 


19 


2,450.6 


44.53 


3,056.8 


55.72 


3,660.8 


66.88 4,266.5 


78.08 


4,872.3 


89.27 5,478,0 


100.47 




(60.0) 


(1.09) 


(77.0) 


(1.40) 


(93.5) 


(1.69) (110.7) 


(2.01) 


(128.2) 


(2.32) (145.6) 


(2.65) 


>0 


2,712.6 


49.37 


3,384.2 


61.77 


4,053.4 


74.14 4,724.6 


86.54 


5,395.8 


98.95 6,067.0 


111.35 




(67.3) 


(1-22) 


(86.3) 


(1.57) 


(104.6) 


(1.90) (123.9) 


(2.24) 


(143.2) 


(2.60) (162.6) 


(2.95) 


>1 


2,987.9 


54.45 


3,728.4 


68.12 


4,466.2 


81.77 5,206.2 


95.44 


5,946.2 


109.12 6,686.2 


122.80 




(75.0) 


(1.36) 


(96.2) 


(1-75) 


(116.5) 


(2.11) (137.8) 


(2.50) 


(159.1) 


(2.89) (180.5) 


(3.28) 


>2 


3,276.7 


59.78 


4,089.3 


74.78 


4,899.1 


89.77 5,711.3 


104.78 


6,523.4 


119.79 7,335.6 


134.80 




(83.2) 


(1.51) 


(106.6) 


(1.94) 


(128.9) 


(2.34) (152.5) 


(2.77) 


(175.8) 


(3.19) (199.3) 


(3.62) 


!3 


3,578.9 


65.36 


4,467.1 


81.76 


5,355.3 


98.15 6,239.8 


114.55 


7,127.5 


130.95 8,015.2 


147.35 




(91.9) 


(1.67) 


(117.5) 


(2.13) 


(143.2) 


(2.60) (167.5) 


(3.05) 


(193.3) 


(3.51) (219.0) 


(3.97) 


>4 


3,894.6 


71.19 


4,861.7 


89.04 


5,828.8 


106.90 6,791.9 


124.75 


7,758.4 


142.61 8,725.0 


160.47 




(101.0) 


(1.83) 


(128.9) 


(2.34) 


(157.0) 


(2.85) (183.5) 


(3.33) 


(211.5) 


(3.84) (239.6) 


(4.35) 



numbers in parentheses give a confidence interval on the mean. 



Table 6.- 


—Aspen bole weight and bole volume (95 percent confidence limit on the mean) 






Bole length (feet) 
20 3G 40 50 


60 


D.b.h. 


Weight Volume Weight Volume Weight Volume Weight Volume 


Weight Volum 



Inches 
5 



10 



11 



12 



13 



14 



15 



16 



17 



Pounds 
104.3 
(16.7) 1 
140.7 
(15.8) 
183.7 
(14.9) 
233.3 
(14.0) 
289.5 
(13.3) 
352.3 

421.7 

(12.8) 
497.7 

(13.4) 
580.4 

(14.8) 
669.6 

(16.8) 
765.5 

(19.4) 
868.0 

(22.5) 
977.1 

(26.0) 
1,092.8 

(30.0) 



Feel 3 
2.51 
(-47) 
3.21 
(-47) 
4.05 
(.46) 
5.00 
(-46) 
6.09 
(-45) 
7.31 

il 45 l 
8.65 

(.45) 
10.12 

(-45) 
11.72 

(-46) 
13.44 

(.47) 
15.29 

(.49) 
17.28 

(.51) 
19.38 

(.54) 
21.62 

(58) 



Pounds 
145.6 
(15.7 
200.2 
(14.6 
264.7 
(13.6 
339.0 
(12.9 
423.3 
(12.8 
517.6 
(13.7 
"I 621.7 
I (15.6 
| 735.8 

I (18.5 

859.7 

(22.2 

993.6 

(26.6 

1,137.4 

(31.6 

1,291.2 

(37.1 

1,454.8 

(43.0 

1,628.4 

(49.4 



Feel 3 
3.31 
(-47) 
4.37 
(.46) 
5.61 
(.45) 
7.05 
(.45) 
8.68 
(.45) 

10.50 
(.46) 

12.51 
(.47) 

14.72 
(.48) 



Pounds 
187.0 
(14.9) 
259.7 
(13.6) 
345.7 
(12.8) 
444.8 
(12.9) 
557.2 
(14.3) 
682.9 
(17.1) 
821.7 
(21.0) 
973.8 
(25.9) 



17.12 ~|1. 139.1 

(.51) , (31.6) 

19.70 1,317.6 

(.55) (38.0) 

22.48 ' 1,509.4 

(.59) I (45.0) 

25.46 ( 1 ,714.3 

(.65) [_(52- 6 ! 

28.62 1,932.5 

(.71) (60.8) 

31.98 2,164.0 

(.78) (69.5) 



Feel 3 
4.11 
(.46) 
5.52 
(.45) 
7.18 
(.45) 
9.09 
(.45) 

11.27 
(.46) 

13.70 
(.48) 

16.38 
(.50) 

19.32 
(.54) 

22.52 
(.59) 

25.97 
(.66) 

29.68 
(.73) 

33.64 
(.82) 



Pounds 
228.3 
(14.1) 
319.2 
(13.0) 



Feel 3 
4.91 
(.45) 
6.67 
(-45) 



37.86 
(.92) 
42.33 
(1.02) 



426.7 

(12.8) 

550.6 

(14.2) 

691.1 

» (17.3) 
848.2 
(21.9) 

1,021.7 
(27.6) 

1,211.8 
(34.20) 
1,418.5 

(41.7) 
1,641.6 

(49.9) 
1,881.3 

(58.9) 
2,137.5 

(68.5) 
2,410.3 

(78.9) 
2,699.5 

(89.8) 



8.74 

(-45) 
11.14 

(-46) 
13.86 

(.48) 
16.89 

(.51) 
20.25 

(.56) 
23.92 

(.62) 
27.92 

(.70) 
32.23 

(.79) 
36.87 

(.89) 
41.82 
(1.01) 
47.09 
(1.14) 
52.68 
(1.28) 



Pounds 

269.6 

(13.5) 

378.7 

(12.7) 
"' 507.7 
I (13.6) 
| 656.4 
L_(16_4)_ 

825.0 

(21.1) 
1,013.5 

(27.3) 
1,221.7 

(34.6) 
1,449.9 

(42.8) 
1,697.8 

(52.0) 
1,965.6 

(62.0) 
2,253.2 

(72.9) 
2,650.7 

(84.6) 
2,888.0 

(97.0) 
3,235.1 
(110.3) 



38.5 
(• 
44.C 
(1.1 
50.( 
(1.1 
56.v 
d 

63.( 
d 



'The numbers in parentheses give a confidence interval on the mean. 



ible 7. — White spruce bole weight and bole volume (95 percent confidence limit on the mean) 













Bole length (feet) 












20 




30 




40 




50 




60 




.b.h. 


Weight 


Volume 


Weight 


Volume 


Weight 


Volume 


Weight 


Volume 


Weight 


Volume 


iches 


Pounds 


Feet 3 


Pounds 


Feet 3 


Pounds 


Feet 3 


I Pounds 


Feel 3 


Pounds 


Fee? 


5 


196.3 


2.85 


228.8 


3.55 


261.3 


4.24 


293.8 


4.94 


326.4 


5.64 




(45. 4) 1 


(.29) 


(43.0) 


(.28) 


(40.7) 


(26) 


(38.5) 


(.25) 


(36.7) 


(.24) 


6 


224.9 


3.47 


271.7 


4.47 


318.5 


5.47 


365.4 


6.47 


412.2 


7.47 




(43.2) 


(-28) 


(40.0) 


(.26) 


(37.1) 


(.24) 


L_(34.7) 


(.22) 


(33.0) 


(.21) 


7 


258.7 


4.19 


322.4 


5.55 


386.2 


6.92 


449.9 


8.28 


" I 513.7 


9.64 




(40.8) 


(.26) 


(36.9) 


(.24) 


(33.9) 


(.22) 


(32.1) 


(•21j 


1 (31.9) 


(.21) 


8 


297.7 


5.02 


381.0 


6.81 


464.3 


8.59 


547.5 


10.37 


I 630.8 


12.15 




(38.3) 


(.25) 


(34.1) 


(.22) 


(32.0) 


(-21) 


(32.3) 


(.21) 


Q35.1) 


(.23) 


9 


342.0 


5.97 


447.3 


8.23 


552.7 


10.48 


658.1 


12.73 


763.5 


14.99 




(35.8) 


(.23) 


(32.2) 


(.21) 


(32.4) 


(.21) 


(36.5) 


(.24) 


(43.4) 


(-28) 


10 


391.4 


7.03 


521.5 


9.81 


651.6 


12.60 


781.7 


15.38 


911.8 


18.16 




(33.7) 


(.22) 


(31.9) 


(.21) 


(36.2) 


(.23) 


(44.7) 


(-29) 


(55.6) 


(36) 


I 11 


446.0 


8.20 


603.5 


11.57 


760.9 


14.93 


918.3 


18.30 


1,075.7 


21.67 




(32.2) 


(.21) 


(34.0) 


(.22) 


(43.2) 


(28) 


(56.2) 


(.36) 


(71.1) 


(.46) 


12 


505.9 


9.48 


693.2 


13.49 


880.6 


17.49 


1,067.9 


21.50 


1,255.3 


25.51 




(31.8) 


(.21) 


(38.6) 


(-25) 


(52.9) 


(-34) 


(70.3) 


(.45) 


(89.0) 


(.57) 


13 


570.9 


10.87 


790.8 


15.57 


1,010.7 


20.28 


1,230.5 


24.98 


1,450.4 


29.69 




(32.9) 


(.21) 


| (45.4) 


(.29) 


(64.8) 


(.42) 


(86.5) 


(.56) 


(109.2) 


(70) 


14 


641.2 


12.37 


896.2 


17.83 


1 1,151.2 


23.28 


1,406.2 


28.74 


1,661.2 


34.20 




(35.6) 


(.23) 


(54.2) 


(.35) 


I (78.5) 


(-51) 


(104.6) 


(.67) 


(131.4) 


(85) 


15 


716.7 


13.99 


1,009.4 


20.25 


| 1,302.1 


26.51 


1,594.8 


32.78 


1,887.6 


39.04 




(40.1) 


(.26) 


(64.7) 


(-42) 


(93.8) 


(.60) 


(124.3) 


(80) 


(155.4) 


(1.00) 


16 


797.3 


15.71 


1,130.4 


22.84 


1,463.4 


29.97 


1,796.5 


38.09 


2,129.5 


44.22 


i 


(45.9) 


(.30) 


(76.4) 


(.49) 


(110.5) 


(.71) 


(145.7) 


(.94) 


(181.4) 


(1.17) 


i 17 


883.2 


17.55 


1,259.2 


25.59 


1,635.2 


33.64 


2,011.1 


41.68 


2,387.1 


49.73 




(53.1) 


(.34) 


(89.4) 


(.58) 


(128.6) 


(.83) 


(168:7) 


(109) 


(209.1) 


(1.35) 


18 


974.2 


19.50 


1,395.8 


28.52 


1,817.3 


37.54 


2,238.8 


46.56 


2,660.3 


55.57 




(61.4) 


(.40) 


(103.5) 


(.67) 


(147.9) 


(95) 


(193.1) 


(1.24) 


(238.6) 


(1.54) 



numbers in parentheses give a confidence interval on the mean. 



Table 8. — Red pine bole weight and bole volumes (95 percent confidence limits on the mean) 



30 



40 



Bole length (feet) 
50 



60 



70 



D.b.h. Weight Volume Weight Volume Weight Volume Weight Volume Weight Volumi 



Inches 
5 



10 



11 



12 



13 



14 



15 



16 



17 



18 



Pounds 
177.4 
(12.4) 1 
234.3 
(11.3) 
301.5 
(10.2) 
379.0 

(9.2) 
466.9 

(8.4) 
565.1 
JM. 
673.7 

(9.1) 
792.6 
(10.8) 
921.8 
(13.2) 
1,061.4 
(16.2) 
1,211.3 
(19.7) 

1,371.5 
(23.6) 

1,542.1 
(27.8) 

1,723.0 
(32.4) 



Feet 3 
3.35 
(.20) 
4.30 
(.19) 
5.42 
(.17) 
6.72 
(.15) 
8.18 
(.14) 
9.82 
U_4) 

11.64 
(-15) 

13.62 
(.18) 

15.78 
(.22) 

18.11 
(.27) 

20.61 
(-33) 

23.28 
(.39) 

26.13 
(-46) 

29.15 
(.54) 



Pounds 
220.5 
(11.6) 
296.3 
(10.3) 
385.9 
(9.1) 
489.3 
(8.4) 
606.5 
(8.5) 
737.4 
(9.9) 

| 882.2 

\J\2A) 

1,040.7 
(15.8) 

1,213.0 
(19.8) 

1,399.1 
(24.3) 

1,598.9 
(29.2) 

1,812.6 
(34.6) 

2,040.0 
(40.5) 

2,281.2 
(46.7) 



Feet 3 
4.07 
(.19) 
5.34 
(.17) 
6.83 
(.15) 
8.55 
(.13) 

10.51 
(.14) 

12.70 

(.17) 
15.12 
J.21)_ 
17.76 

(.26) 
20.64 

(-33) 
23.74 

(.40) 
27.08 

(.49) 
30.65 

(.58) 
34.44 

(.67) 
38.47 

(.78) 



Pounds 
263.6 
(10.8) 
358.4 
(9.4) 
470.4 
(8.4) 
599.6 
(8.5) 
746.0 
(10.1) 
909.7 
(13.0) 

1,090.6 
(16.9) 

1,288.8 
(21.6) 

1,504.2 
(26_9)_ 

1,736.8 
(32.7) 

1,986.6 
(39.1) 

2,253.7 
(46.0) 

2,538.0 
(53.3) 

2,839.5 
(61.1) 



Feet 3 
4.79 
(.18) 
6.37 
(.16) 
8.24 
(.14) 

10.40 
(.14) 

12.84 
(.17) 

15.58 
(.22) 

18.60 
(.28) 

21.90 
(.36) 

25.50 
__(.45J_ 

29.38 



Pounds 

306.7 
_(10J) 

420.4 
(8.8) 

554.8 
(8.3) 

709.9 
(9.5) 

885.6 

(12.5) 
1,082.0 

(16.7) 
1,299.1 

(21.8) 
1,536.9 

(27.7) 
1,795.4 

(34.2) 
2,074.5 



(.55) |_(4JL3) _ ( 



Feet 3 
5.51 

il 7 L 

7.41 

(.15) 

9.65 

(.14) 
12.24 

(.16) 
15.17 

(.21) 
18.45 

(.28) 
22.08 

(.36) 
26.05 

(46) 
30.36 

(.57) 
35.02 
.69) 



Pounds 

349.8 
(9.5) 

482.4 
|_(8_4) 

639.2 
(8.8) 

820.1 

(11.3) 
1,025.2 

(15.4) 
1,254.3 

(20.7) 
1,507.6 

(27.0) 
1,785.0 

(33.9) 
2,086.5 

(41.7) 
2,412.2 

(50.0) 



35.S 



33.55 
(.65) 

38.01 
(.77) 

42.76 
(.89) 

47.79 

(1.02) 



2,374.3 

(49.1) 
2,694.8 

(57.4) 
3,035.9 

(66.2) 
3,397.7 

(75.6) 



40.02 
(-82) 
45.37 
(.96) 
51.07 
(1.10) 
57.11 
(1.26) 



2,762.0 

(59.1) 
3,135.8 

(68.8) 
3,533.9 

(79.2) 
3,956.0 

(90.2) 



1 The numbers in parentheses give a confidence interval on the mean. 



10 



fable 9. — Balsam fir bole weight and bole volume 



?ent confide, 



20 



m 



D.b.h. 


Weight 


Volume 


Weight 


Voiurne 














'Inches 


Pounds 


Feet 3 


Pounds 


Feet 3 


i Pounds 












5 


144.8 


2.65 


180.5 


3.35 








4.76 








(29.3) 1 


(.24) 


07 1 \ 






/ 9n\ 




(.19) 


(21.7) 




6 


176.2 


3.27 












6.30 








(27.3) 


(.22) 




(.20) 














7 


213.6 


4.00 










425.0 










(25.2) 


(.20) 


\c i .yj 


(.18) 




(.16) 




(.16) 


(22.0) 


(.18) 


? 


256.8 


4.84 


348.8 


6.64 






I 532.9 


10.23 




12.02 




(23.0) 


(.19) 


(20.2) 


(.16) 




(.16) 


| (23.6) 


(.19) 


(28.8) 


(.23) 


9 


305.6 


5.79 


| 422.1 


8.07 




1 n "%& 


655.1 


12.61 


m.<o 






(21.2) 


(.17) 


(20.1) 


(.16) 


{co.o/ 


('19) 


(30.8) 


(.25) 


j (39.2) 


(.32) 


10 


360.3 


6.86 


i 504.1 


9.67 


647.9 


12.47 


791.7 


15.28 


935.5 


18.08 




(20.0) 


(.16) 


1 (22.3) 

1 


(.18) 


(30.3) 


(.25) 


(40.8) 


(.33) 


(52.3) 


(.42) 


11 


420.7 


8.04 


1 594.7 


11.43 


768.7 


14.83 


942.7 


18.22 


1,116.7 


21.61 




(20.1) 


(.16) 


| (27.0) 


(.22) 


(39.0) 


(.32) 


(52.9) 


(.43) 


(67.4) 


(.54) 


12 


486.8 


9.33 


693. S 


13.37 


901.0 


17.41 


1,108.1 


21.45 


1,315.1 


25.49 




(21.7) 


(.18) 


1 (33.5) 


(.27) 


(49.5) 


(.40) 


(66.6) 




(84.3) 


(.68) 


13 


558.7 


10.73 


| 801.8 


15.47 


1 044 8 


20.21 


1,287.8 




1,530.8 


29.69 




(24.9) 


(-20) 


(41.6) 


(.34) 




(.50) 


(81.9) 




(102.9) 


(.83) 


14 


636.4 


12.25 


I 918.2 


17.74 


1,200.1 


23.24 


1,481.9 


28.74 


1,763.8 


34.24 




(29.6) 


(.24) 


|J_50.9) 


(.41) 


(74.4) 


(.60) 


(98.7) 


(.80) 


(123.2) 


(.99) 


15 


719.8 


13.87 


1,043.3 


20.18 


1,366.9 


26.50 


1,690.4 


32.81 


2,014.0 


39.12 




(35.4) 


(■29) 


(61.2) 


(.49) 


(88.7) 


(.72) 


(116.8) 


(.94) 


(145.0) 


(1.17) 


16 


808.9 


15.61 


1,177.1 


22.79 


1,545.2 


29.97 


1.913.3 


37.15 


2,281.5 


44.33 




(42.1) 


(.34) 


(72.5) 


(.59) 


(104.1) 


(.84) 


(136.2) 


(1.10) 


(168.5) 


(1.36) 


17 


903.9 


17.46 


1,319.4 


25.57 


1,735.0 


33.68 


2,150.6 


41.78 


2,566.2 


49.89 




(49.7) 


(-40) 


(84.6) 


(.68) 


(120.7) 


(.97) 


(157.0) 


(1.27) 


(193.5) 


(1.56) 


• 18 


1,004.5 


19.43 


1,470.4 


28.51 


1,936.3 


37.60 


2,402.3 


46.69 


2,868.2 


55.78 




(58.0) 


(.47) 


(97.7) 


(.79) 


(138.2) 


(1.12) 


(179.1) 


(1.45) 


(220.1) 


(1.78) 



he numbers in parentheses give a confidence interval on the mean. 



]J 



Table 10.— Sugar maple bole weight and bole volume (95 percent confidence intervals on the mean) 













Bole length (feet) 














10 




20 




30 




40 




50 




60 




D.b.h. 


Weight 


Volume 


Weight 


Volume 


Weight 


Volume 


Weight 


Volume 


Weight 


Volume 


Weight 


Volumi 


Inches 


Pounds 


Feel 3 


Pounds 


Feel 3 


Pounds 


Feel 3 


Pounds 


Feet 3 


Pounds 


Feef 


Pounds 


Feet 


5 


90.7 


1.36 


136.7 


2.18 


182.6 


2.99 


228.6 


3.81 


274.5 


4.62 


320.5 


5.44 




(7.8) 1 


(.13) 


(7.2) 


(.12) 


(6.7) 


(.11) 


(6.4) 


(.11) 


(6.5) 


(.11) 


(6.8) 


(-12 


6 


111.0 


1.72 


177.1 


2.89 


243.3 


4.07 


309.5 ~ 


5.24 


I 375.6 


6.41 


441.8 


7.59 




(7.5) 


(.13) 


(6.7) 


(.11) 


(6.4) 


(.11) 


(6.7) 


(.11) 


I (7.5) 


(.13) 


(8.6) 


(.15 


7 


134.9 


2.14 


224.9 


3.74 


315.0 


5.34 


405.0 


6.94 


I 495.1 


8.53 


585.2 


10.13 




(7.2) 


(.12) 


(6.5) 


(.11) 


(6.8) 


(.11) 


(7.9) 


(-14) L (97) 


(.16) 


(11.8) 


(.20 


8 


162.4 


2.63 


280.1 


4.72 


397.7 


6.81 


515.3 


8.90 


633.0 


10.98 


| 750.6 


13.06 




(6.9) 


(.11) 


(6.5) 


(.11) 


(7.8) 


(.13) 


(10.1) 


(.18) 


(13.0) 


(.22) 


I (16.0) 


(.27 


9 


193.7 


3.19 


342.6 


5.83 


491.4 


8.47 


640.3 


11.11 


789.2 


13.76 


' 938.1 


16.39 




(6.6) 


(.11) 


(7.0) 


(.12) 


(9.6) 


(-16) 


(13.1) 


(.22) 


(17.0) 


(.29) 


I (21.0) 


(.36 


10 


228.6 


3.81 


"| 412.4 


7.07 


596.2 


10.33 


780.0 


13.59 


963.8 


16.86 


1,147.6 


20.12 




(6.5) 


(.11) 


(8.1) 


(.14) 


(12.0) 


(-20) 


(16.8) 


(v28) 


(21.8) 


(-37) 


(26.9) 


(.46 


11 


267.2 


4.49 


| 489.6 


8.44 


712.0 


12.39 


934.4 


16.33 


1,156.8 


20.28 


1,379.2 


24.22 




(6.5) 


(.11) 


| (9-6) 


(.16) 


(15.0) 


(-26) 


(21.0) 


(.35) 


(27.2) 


(.46) 


(33.4) 


(.57 


12 


309.5 


5.24 


574.1 


9.94 


] 838.8 


14.63 


1,103.5 


19.32 


1,368.2 


24.02 


1,632.8 


28.71 




(6.7) 


(.11) 


(11.5) 


(-20) 


I (18.3) 


(.31) 


(25.6) 


(.43) 


(33.1) 


(.56) 


(40.6) 


(.65 


13 


355.4 


6.06 


666.0 


11.57 


I 976.7 


17.07 


1,287.3 


22.58 


1,597.9 


28.09 


1,908.5 


33.6C 




(7.2) 


(.12) 


(13.8) 


(.23) | (22.1) 


(.38) 


(30.8) 


(.52) 


(39.6) 


(.67) 


(48.5) 


(■« 


14 


405.0 


6.94 


765.3 


13.33 


1,125.5 


19.72 


1,485.8 


26.10 


1,846.0 


32.49 


2,206.3 


38.81 




(7.9) 


(-13) 


(16.4) 


(.28) 


(26.3) 


(.45) 


(36.4) 


(.62) 


(46.7) 


(.79) 


(57.0) 


(.9/ 


15 


458.3 


7.88 


871.9 


15.22 


1,285.4 


22.56 


1,699.0 


29.89 


2,112.5 


37.22 


2,526.1 


44. 51 




(8.9) 


(.15) 


(19-2) 


(.33) 


(30.8) 


(.52) 


LJ42.5) 


(.72) 


(54.3) 


(.92) 


(66.2) 


(1.1! 


16 


515.3 


8.89 


985.8 


17.24 


1,456.4 


25.58 


1,926.9 


33.93 


2,397.4 


42.27 


2,868.0 


50.6; 




(10.1) 


(.17) 


(22.4) 


(-38) 


(35.6) 


(-60) 


(49.0) 


(.83) 


(62.5) 


(106) 


(76.1) 


(1.2! 


17 


576.0 


9.97 


1,107.2 


19.39 


1,638.3 


28.81 


2,169.5 


38.23 


2,700.7 


47.65 


3,231.9 


57.0" 




(11.5) 


(.20) 


(25.7) 


(.44) 


(40.8) 


(.69) 


(56.0) 


(.95) 


(71.3) 


(1.21) 


(86.5) 


(1.4 


18 


640.3 


11.11 


1,235.8 


21.67 


1,831.3 


32.23 


2,426.8 


42.80 


3,022.4 


53.36 


3,617.9 


63.9! 




(13.1) 


(.22) 


(29.4) 


(.50) 


(46.3) 


(.78) 


(63.4) 


(107) 


(80.5) 


(1.36) 


(97.7) 


(1.6 



'The numbers in parentheses give a confidence interval on the mean. 



12 



-Residue weight and volume (95 percent confidence limit on the mean) 





















Sugar maple 




Aspen 


White 


spruce 


Red 


pine 


Balsam fir 


pulp wood 


D.b.h. 


Weight 


Volume 


Weight 


Volume 


Weight 


Volume 


Weight 


Volume 


Weight 


Volume 


Inches 


Pounds 


Feet 3 


Pounds 


Feel 3 


Pounds 


Feet 3 


Pounds 


Feet 3 


Pounds 


Feel 3 


5 


38.2 


0.99 


65.0 


0.93 


26.5 


0.51 


89.8 


1.77 


103.2 


1.77 




(17.3) 1 


(.37) 


(49.0) 


(1.01) 


(9.0) 


(.16) 


(32.9) 


(72) 


(12.9) 


(.27) 


6 


63.2 


1.47 


101.0 


1.68 


48.5 


0.87 


119.9 


2.37 


130.8 


2.27 




(14.9) 


(-32) 


(42.5) 


(-87) 


(7.4) 


(.13) 


(27.6) 


(60) 


(9.9) 


(22) 


7 


92.9 


2.04 


143.5 


2.58 


74.4 


1.31 


155.3 


3.07 


163.5 


2.85 




(12.9) 


(.28) 


(36.2) 


(.74) 


(6.2) 


(.11) 


(22.8) 


(50) 


(9.8) 


(23) 


8 


127.1 


2.69 


192.5 


3.61 


104.3 


1.81 


196.3 


3.88 


201.2 


3.53 




(12.4) 


(.27) 


(31.6) 


(.65) 


(6.4) 


(.11) 


(20.8) 


(45) 


(13.3) 


(.31) 


9 


165.8 


3.43 


248.1 


4.79 


138.1 


2.37 


242.7 


4.79 


243.9 


4.30 




(14.3) 


(.31) 


(31.4) 


(65) 


(8.4) 


(.15) 


(24.3) 


(.53) 


(19.5) 


(45) 


10 


209.1 


4.25 


310.3 


6.10 


176.0 


3.01 


294.6 


5.82 


291.6 


5.26 




(18.4) 


(.39) 


(37.4) 


(.77) 


(118) 


(-21) 


(32.8) 


(-72) 


(27.3) 


(.62) 


11 


257.0 


5.17 


378.9 


7.54 


217.9 


3.71 


351.9 


6.95 


344.4 


6.10 




(24.3) 


(.52) 


(48.8) 


(100) 


(16.0) 


(28) 


(44.8) 


(98) 


(36.2) 


(82) 


12 


309.4 


6.17 


454.1 


9.13 


263.7 


4.48 


414.7 


8.19 


402.2 


7.14 




(31.3) 


(.67) 


(64.0) 


(1.32) 


(20.8) 


(37) 


(59.0) 


(1.29) 


(46.3) 


(1.05) 


13 


366.4 


7.26 


535.9 


10.85 


313.5 


5.31 


482.9 


9.54 


465.0 


8.27 




(39.4) 


(.84) 


(81.9) 


(168) 


(26.2) 


(46) 


(75.1) 


(1.64) 


(57.3) 


(129) 


14 


427.9 


8.43 


624.2 


12.71 


367.3 


6.21 


I 556.7 


10.99 


532.9 


9.49 




(48.3) 


(1.03) 


(102.0) 


(2.10) 


(32.1) 


(-57) 


I (92.9) 


(2.02) 


(69.2) 


(1-56) 


15 


494.0 


9.69 


719.0 


14.71 


I 425.1 


7.18 


635.8 


12.55 


605.8 


10.80 




(58.0) 


(1.24) 


(124.1) 


(2.55) 


(38.4) 


(68) 


(112.1) 


(2.44) 


(82.1) 


(1.84) 


16 


564.6 


11.04 


820.4 


16.85 


I 486.9 


8.21 


720.5 


14.23 


683.7 


12.20 




(68.4) 


(1.46) 


(148.0) 


(3.04) _ 


I (45.3) 


(.80) 


(132.8) 


(2.90) 


(95.9) 


(215) 


17 


639.9 


12.48 


928.3 


19.12 


552.7 


9.31 


810.5 


16.00 


766.6 


13.69 




(79.6) 


(1.70)j 


(173.6) 


(3.57) 


(52.6) 


(.93) 


(155.0) 


(3.38) 


(1107) 


(2.47) 


18 


[719.6 


14.00 


1,042.7 


21.53 


622.4 


10.48 


906.1 


17.89 


854.6 


15.27 




| (91.5) 


(1.96) 


(200.9) 


(4.13) 


(60.4) 


(1.07) 


(178.5) 


(3.89) 


(126.3) 


(2.82) 



'The numbers in parentheses give a confidence interval on the mean. 



13 



10 000 r TREE WEIGHT = 23.80 + 0.16 TREE HEIGHT D.B.H.' 
' I STANDARD ERROR ESTIMATE = 1.29 D.B.H.' 

9,000 
8,000 ■ 




4 6 8 10 12 14 16 18 20 22 24 
D.B.H. (INCHES) 

Figure 1.— Tree weight— aspen. 



10 000 r TREE WEIGHT = 7.288 + 0.18 TREE HEIGHT D.B.H/ 
STANDARD ERROR ESTIMATE = 0.76 D.B.H. 1 

9,000 

8,000 
§ 7,000 
2 8,000 

1 5,000 

ts 

2 4,000 

Hi 

Uj 3,000 

E 

2,000 
1,000 

c 




4 6 8 10 12 14 16 18 20 22 24 
D.B.H. (INCHES) 

Figure 3.— Tree weight— red pine. 



10 000 r TREE WEIGHT = 152.70 + 0.16 TREE HEIGHT D.B.H.' 
STANDARD ERROR ESTIMATE = 2.34 D.B.H.' 

9,000 
^ 8,000 

v> 

§ 7,000 
§ 6,000 

1 5,000 

2 

2 4,000 

lu 

Ui 3,000 

s 

2,000 
1,000 






4 6 8 10 12 14 16 18 20 22 24 
D.B.H. (INCHES) 

Figure 2. — Tree weight — white spruce. 



10 000 r TREE WE| G HT = T3.69 + 0.17 TREE HEIGHT D.B.H.' 
STANDARD ERROR ESTIMATE = 1.81 D.B.H. 2 



9,000 
8,000 



Ui 



Ul 

lu 


-j 
2 
3 


7,000 


it 


O 

0. 


6,000 


K. 






ID 




5,000 


lu 


O 




I 


Ul 


4,000 


Ul 


i 




LU 


Ul 

Ui 


3,000 


K 




2,000 

1,000 






Ul 

s. 

I 
O 
Lu 

Ul 
Ul 

oc 



4 6 8 10 12 14 16 18 20 22 24 
D.B.H. (INCHES) 

Figure 4.— Tree weight— balsam fir. 



U 



10 000 r TREE WE| G HT = 26.16 + 0.17 TREE HEIGHT D.B.H.' 
STANDARD ERROR ESTIMATE = 1.57 D.B.H. 1 




160 



r TREE VOLUME = 1.42+0.0034 TREE HEIGHT D.B.H. 1 
STANDARD ERROR ESTIMATE = 0.035 D.B.H. 2 



BO 






70 


K 


Ul 

u. 




Ly 


o 




Ul 




60 


LC 


03 

3 




h> 


o 


50 


CD 


Uj 




Uj 


S 


40 


I 


3 




Ui 


O 




Ul 


^ 




E 


Uj 
Ui 
CC 



4 6 8 10 12 14 16 18 20 22 24 
O.B.H. (INCHES) 

Figure 5. — Tree weight — maple. 




70 



o0 



Uj 
50 It! 



40 



30 



6 8 



10 12 14 16 18 20 22 24 
D.B.H. (INCHES) 



a: 

CD 
Uj 

Uj 
Uj 

E 



Figure 7. — Tree volume — white spruce. 



Ul 

St 
o 

o 

Ul 

I 

g 



160 

140 

120 

100 

80 

SO 

10 



TREE VOLUME = 1.23+0.0030 TREE HEIGHT D.B.H. 2 
STANDARD ERROR ESTIMATE = 0.025 D.B.H. 1 



TREE VOLUME = 0.58+0.0030 TREE HEIGHT D.B.H.' 
STANDARD ERROR ESTIMATE = 0.014 D.B.H. 2 




80 



70 



60 Ul 

It! 



50 



40 r= 



3d 



4 6 8 



10 12 14 16 18 20 22 24 
D.B.H. (INCHES) 




10 12 14 16 18 20 22 24 
D.B.H. (INCHES) 



Figure 6— Tree volume — aspen. 



Figure 8. — Tree volume — red pine. 



15 



160 r TREE vol - UME = 1 26+0.0034 TREE HEIGHT D.B.H.' 
STANDARD ERROR ESTIMATE = 0.032 D.B.H. ' 



iu 

Uj 

U. 
O 

s 

o 

Ul 

£ 
iu 




6 8 



10 12 14 16 18 20 22 24 
D.B.H. (INCHES) 



Figure 9. — Tree volume — balsam fir. 



V) 

Q 

O 

CD 

UJ 

UJ 

-J 

o 

00 



3,000 



2,500 



2,000 



1,500 



1,000 



500 



BOLE WEIGHT = 21.67 + 0.165 BOLE LENGTH D.B.H.' 
STANDARD ERROR ESTIMATE = 0.91 D.B.H.' 



70 60 




50 



30 



20 



8 10 12 14 
D.B.H. (INCHES) 



16 18 



UJ 

a 

6 

CD 

S 

•j 
Ul 

•J 

o 

00 



Figure 11. — Bole weight — aspen. 



UJ 

8! 

o 
bo 

o 

Ul 

I 

-J 

UJ 
UJ 



TREE VOLUME = -0.23+0.0031 TREE HEIGHT D.B.H.' 
STANDARD ERROR ESTIMATE = 0.029 D.B.H.' / 90 



80 




70 



50 



40 ^ 



'4 6 8 10 12 14 16 18 20 22 24 
D.B.H. (INCHES) 

Figure 10.— Tree volume — maple. 



3,000 



2,500 



BOLE WEIGHT =131.20 + 0.13 BOLE LENGTH D.B.H.' 
STANDARD ERROR ESTIMATE = 1.52 D.B.H.' 70 



60 





</> 




K 


Q 




Uj 
Ul 


2 
3 


2,000 


u. 


o 




K 


a 




ar 


K 




CD 


I 


1,500 


Ul 


CD 




Ul 


^ 




Ul 

E 


UJ 
00 


1,000 



500 




8 10 12 14 
D.B.H. (INCHES) 



50 _ 
K 
UJ 

40 t 



30 



CD 
UJ 

•j 

Ul 

20 -J 

§ 



18 



Figure 12. — Bole weight — white spruce. 



16 



BOLE WEIGHT = 48.23 + 0.172 BOLE LENGTH D.B.H. 1 
STANDARD ERROR ESTIMATE = 0.53 O.B.H. 1 



BOLE WEIGHT = 44.79 + 0.184 BOLE LENGTH O.B.H. 1 
STANDARD ERROR ESTIMATE = 0.51 D.B.H. 1 



3,000 


70 60 


2,500 




— ^ 




V) 




Q 




§ 2,000 




O 




a 








I 1,500 




o 








Ul 




* 




lu 1,000 




•J 




o 




CD 




500 
n 


— i — i — i — j.. .. . . i . i i i ■ i , i 



50 



40 



30 



8 10 12 14 16 18 
D.B.H. (INCHES) 



3,000 r 



2,500 



_ 


C/> 


K 


Q 




§ 2,000 
O 


fC 


a 


O 

s 


!t 1,500 
(3 


-J 


Ul 


Uj 


^ 


5 

QQ 


114 1,000 

-J 

o 




QQ 



500 



60 50 




Uj 

E 

s 

•J 

iuu 

to 



6 8 10 12 14 16 18 
D.B.H. (INCHES) 



Figure 13. — Bole weight — red pine. 



Figure 15. — Bole weight— maple. 



BOLE WEIGHT = 72.69 + 0.144 BOLE LENGTH D.B.H.' 
3 000 r STANDARD ERROR ESTIMATE = 1.21 D.B.H. 1 






K 




Ul 


K 
UJ 


Uj 

u. 


^ 


o 


u. 






CD 


I 


S 


K 


O 




Ul 

S 


~J 


^ 


UJ 

-J 

o 




QQ 


Uj 




-J 




O 




08 



60 r 



BOLE VOLUME = 0.91 + 0.0032 BOLE LENGTH D.B.H. 1 
STANDARD ERROR ESTIMATE = 0.012 D.B.H. 1 

70 60 



8 10 12 14 
D.B.H. (INCHES) 



16 18 




tu 

z. 

Uj 
-J 

ua 

§ 



8 10 12 14 16 18 
D.B.H. (INCHES) 



Figure 14.— Bole weight— balsam fir. 



Figure 16.— Bole volume— aspen. 



17 



BOLE VOLUME = 1.46 + 0.0028 BOLE LENGTH D.B.H. : 
STANDARD ERROR ESTIMATE =0.010 D.B.H.' 



SI 



O 
UJ 

1 

s 

UJ 
03 




70 






50 






50 




Uj 




— ^ 


UJ 




1- 


u. 


40 


Uj 

8. 
E 


s 

O 


30 


O 


UJ 




2: 

UJ 


s 

-j 




Ul 


o 


20 


-J 


s 




o 

0Q 


UJ 

•J 

o 

CO 



60 T BOLEVOLUME = 1.25 + 0.0028BOLELENGTHD.B.H.' 
STANDARD ERROR ESTIMATE = 0.010 D.B.H.' 



8 10 12 14 16 18 
D.B.H. (INCHES) 




6 8 10 12 14 16 18 
D.B.H. (INCHES) 



Figure 1 7.— Bole volume — white spruce. 



Figure 19. — Bole volume — balsam fir. 



BOLE VOLUME =1.19 + 0.0029 BOLE LENGTH D.B.H.' 
STANDARD ERROR ESTIMATE = 0.009 D.B.H. 1 



UJ 

o 
o 

-j 

-j 
O 
oa 






h- 




Ul 


UJ 


B! 


Ul 


o 


u. 


3 


I 


s 


K 


o 


O 

yj 


S 


-j 


d 


UJ 

-j 

o 


•j 
§ 


00 


UJ 




-i 




O 




an 



60 r 



-v 50 



40 



30- 



20 



10 



BOLE VOLUME = 0.546 + 0.0033 BOLE LENGTH D.B.H.' 
STANDARD ERROR ESTIMATE = 0.009 D.B.H.' , 60 




UJ 

8 
E 

C3 
Ul 



20 Uj 



O 

no 



6 8 10 12 14 16 18 
D.B.H. (INCHES) 



8 10 12 14 16 18 
D.B.H. (INCHES) 



Figure 18. — Bole volume— red pine. 



Figure 20. — Bole volume — maple. 



18 



RESIDUE WEIGHT = -18.81 + 2.28 D.B.H.' 
STANDARD ERROR ESTIMATE = 0.88 D.B.H. 




1,400 



RESIDUE WEIGHT = -16.75 + 3.27 D.B.H.' 
STANDARD ERROR ESTIMATE = 1.47 D.B.H. 



> 8 10 12 14 

D.B.H. (INCHES) 

Aspen residue weight with 95 percent 




Figure 21 

confidence limits on the mean. 



8 10 12 14 
D.B.H. (INCHES) 

Figure 22. — White spruce residue weight with 95 
percent confidence limits on the mean. 



1,400 



RESIDUE WEIGHT = -23.29 + 1.99 D.B.H.' 
STANDARD ERROR ESTIMATE = 0.39 D.B.H. 




8 10 12 14 
D.B.H. (INCHES) 

Figure 23.— Red pine residue weight with 95 percent 
confidence limits on the mean. 



19 



1,400 r 



RESIDUE WEIGHT = 21.57 + 2.73 D.B.H. 2 
STANDARD ERROR ESTIMATE = 1.27 D.B.H/ 




1,400 r 



RESIDUE WEIGHT = 40.34 + 2.51 D.B.H. 2 
STANDARD ERROR ESTIMATE = 0.77 D.B.H. 1 



6 8 10 12 14 16 18 

D.B.H. (INCHES) 

Figure 24. — Balsam fir residue weight with 95 per- 
cent confidence limits on the mean. 




16 



18 



6 8 10 12 14 

D.B.H. (INCHES) 

Figure 25. — Maple residue weight with 95 percent 
confidence limits on the mean. 



28 r 



RESIDUE VOLUME = -0.10 + 0.044 D.B.H.* 
STANDARD ERROR ESTIMATE = 0.02 D.B.H. 2 



K 
Ui 
UJ 

u. 
o 

Si 

o 

Ui 

§ 

Q 
? 
QC 




16 



18 



8 10 12 14 
D.B.H. (INCHES) 

Figure 26. — Aspen residue volume with 95 percent 
confidence limits on the mean. 



20 



28 



Uj 
Uj 

u. 
o 

£ 

o 

UJ 

$ 

-J 

§ 

UJ 

Q 

CO 
UJ 



RESIDUE VOLUME = -0.80 + 0.069 D.B.H.' 
STANDARD ERROR ESTIMATE = 0.030 D.B.H.' 



RESIDUE VOLUME = -0.33 + 0.033 D.B.H.* 
STANDARD ERROR ESTIMATE = 0.007 D.B.H. 1 




lU 
UJ 
U. 



QQ 
O 
UJ 

§ 

Uj 

Q 

co 
Uj 



8 10 12 14 

D.B.H. (INCHES) 



16 18 




8 10 12 14 
D.B.H. (INCHES) 



16 18 



Figure 27. — White spruce residue volume with 95 
percent confidence limits on the mean. 



Figure 28. — Red pine residue volume with 95 percent 
confidence limits on the mean. 



28 



RESIDUE VOLUME = 0.43 + 0.054 D.B.H. 2 
STANDARD ERROR ESTIMATE = 0.028 D.B.H. 2 



I- 
LU 

UJ 

u. 
o 

CD 
O 
Uj 

§ 

Uj 

Q 
co 

Uj 

ec 




8 10 12 14 

D.B.H. (INCHES) 



16 



18 



Figure 29. — Balsam fir residue volume with 95 per- 
cent confidence limits on the mean. 



21 



RESIDUE VOLUME = 0.64 + 0.045 D.B.H. 2 
STANDARD ERROR ESTIMATE = 0.017 D.B.H. : 



m 

a 

o 
o 

UJ 

§ 

:> 




8 10 12 14 

D.B.H. (INCHES) 



16 18 



Figure 30. — Maple residue volume with 95 percent 
confidence limits on the mean. 



1,400 




RED PINE 

SUGAR 
MAPLE 
ASPEN 

WHITE 
SPRUCE 

BALSAM Fll 



4 6 8 10 12 14 16 18 20 22 24 
D.B.H. (INCHES) 

Figure 31 . — Tree weight based on d.b.h. — 5 species 







WHITE 
SPRUCE 

BALSAM FIR 
SUGAR 
MAPLE 

ASPEN 

RED PINE 



4 6 8 10 12 14 16 18 
D.B.H. (INCHES) 

Figure 32. — Residue weight based on d.b.h. — 5 spe- 
cies. 



22 



tV U.S.GPO: 1981-766-506/13. 



/ 



nA FORFST SERVICE 



Winsauer, Sharon A., and Helmuth M. Steinhilb. 

1980. Summary of green weights and volumes for five tree species in 
Michigan. U.S. Department of Agriculture Forest Service, Research 
Paper NC-191, 22 p., U.S. Department of Agriculture Forest Service, 
North Central Forest Experiment Station, St. Paul, Minnesota. 

Presents and summarizes the green weights and volumes of trees, 
boles and residue for sugar maple, white spruce, aspen, balsam fir and 
red pine in Northern Michigan. Equations, tables and graphs are 
included for each of the five species. 

KEY WORDS: Sugar maple, white spruce, aspen, balsam fir, red pine, 
tree, bole, residue. 



/ 



SDA FOREST SERVICE 
ESEARCH PAPER NC-192 



Um 241981 

CUM&OH 

mm. 




A program and documentation 

for simulation of a 

tracked feller/buncher 

Sharon A. Winsauer 








Morth Central Forest Experiment Station 
Forest Service, U.S. Department of Agriculture 



North Central Forest Experiment Station 

Forest Service — U.S. Department of Agriculture 

1992 Folwell Avenue 

St. Paul, Minnesota 55108 

Manuscript approved for publication January 28, 1980 

1980 



A PROGRAM AND DOCUMENTATION 

FOR SIMULATION OF A TRACKED 

FELLER/BUNCHER 



Sharon A. Winsauer, Computer Specialist, 
Hough ton , Mich iga n 



Computer modeling of forest harvesting systems 
has many applications, each requiring a different 
approach to simulation. Simulations range from 
technical descriptions of machine components for 
improving equipment design to overviews of major 
systems for estimating large-scale changes in pro- 
ductivity. Forestry Sciences Laboratory personnel at 
Houghton have been involved for the past several 
years in developing harvesting simulators for use in 
forest engineering research (Bradley et al. 1976, 
Bradley and Winsauer 1976, 1978, U.S. Department 
of Agriculture Forest Service 1978). 

The purpose of this paper is to present a computer 
model and documentation for a tracked feller/bunch- 
er (fig. 1). The model was developed to provide a 
detailed study of this machine's operation in the 
woods during harvesting operations. It allows the 
user to identify productive and unproductive condi- 
tions, to experiment with ways of increasing machine 
efficiency, to predict the effects of system changes, 
and to determine harvest costs for various types of 
stands and conditions. 



THE EQUIPMENT STUDIED 

The feller/buncher is a tracked vehicle with a 
shear mounted on the end of a rotatable hydraulic 
boom. The shear head can either sever and bunch 
single stems or accumulate several stems at once. 
The boom enables the feller/buncher to shear trees up 
to the boom's maximum reach while maintaining a 
fairly straight path through the woods; cutting a 
strip for thinning or in working the face of a clearcut 
stand. 



MODEL OBJECTIVES 

The model was designed to study the productivity 
and operation of the feller/buncher in the woods. In- 



put required includes data on the machine's operat- 
ing characteristics, the stand itself, and the type of 
harvest to be carried out. 

Model output includes productivity figures such as 
number of trees, weight and volume harvested per 
hour or load, number of total cycles (accumulator 
head loads), number of trees per cycle, total distance 
traveled, and skid bunches completed. 

Additional output includes details of the woods 
operation, distribution of cycle times, distances, 
number of trees per bunch, size of skidder bunches, 
and frequency and length of delays. 




5« 






Figure 1. — Tracked feller/buncher. 



SIMULATION LANGUAGE 
— GPSS 

The simulation is written in General Purpose Sim- 
ulation Language (GPSS), with output subroutines 
written in FORTRAN. GPSS is a discrete-event sim- 
ulation language developed by IBM (Schriber 1974). 
The user constructs a block diagram by arranging the 
discrete events of a system in their logical structure. 
The block diagram is made up of a group of specific 
GPSS block types, which then become the GPSS pro- 
gram. A basic understanding of the GPSS language 
allows the user to accurately interpret or modify the 
simulation. 

Most versions of GPSS have a standard output 
format that concisely presents the data. The FOR- 
TRAN subroutines are used to present important 
output values in an organized, labeled form. 



MODEL ASSUMPTIONS 

The basic time unit used in the model is a "centi- 
minute" — that is, 1/100 of a minute. The feller works 
one shift a day for the number of days chosen by the 
user. All volume measures are in 0.01 cu. ft. 

The feller/buncher is assumed to operate far 
enough in advance of the skidder to minimize inter- 
actions between machines. To simulate a "hot" log- 
ging operation, this model can be combined with a 
complete skidding-chipping or skidding-loader 
model. Another alternative would be to add a "black 
box" skidder segment, a simplified segment used to 
create the machine interactions without a detailed 
study of the skidder operation. 

Since an efficient skidder load usually contains 
more trees than the shear can hold, the feller/buncher 
may shear and drop several loads at one location to 
create the skidder bunch. 

The following assumptions are made about the 
feller/buncher's behavior: 

1. The machine has an accumulating shear. (The 
model can easily be changed to single-head shear). 

2. The feller/buncher travels in a relatively 
straight row, either along the edge of a clearcut or 
down a thinning strip, harvesting all trees within 
reach that are scheduled for cutting. At the end of the 
strip, it travels through the woods to the next strip, 
then returns along that strip toward the landing (fig. 
2). 

3. If there are no more trees within the feller's 
reach, the bunch it was working on is considered 




Figure 2. — Feller/buncher travel pattern. 



complete and the feller must move forward — it do 
not back up. 

4. If a bunch is not started at a given location (om 
one accumulator head load has been cut), the fellr 
will carry the trees in the shear head to the net 
location. (Exception: if the end of the row (strip) h 
been reached, the feller will drop the load there I 
fore traveling to the next strip. 

5. The feller/buncher can quit for the day or stop !(r 
a break only after completing a bunch for the skidd* 

6. Mechanical and nonmechanical delays are i 
corporated into the model. For ease of programmir , 
they are assumed to occur after the feller/buncr 
drops the current load of trees. 



MODEL DESCRIPTION 

The model consists of two GPSS segments: .. 
Timer segment, 2. Feller/buncher segment) ajp 
two FORTRAN subroutines for easily read output \ 
complete variable list, program listing, and fl v 
charts are contained in Appendixes A, B, and C 



The Timer segment (fig. 3) controls the daily 
schedule, keeps track of the days worked, and sends 
information to the subroutines. The Timer signals 
the start of the day, the rest breaks and the lunch 
break (two 15-minute breaks a day and V2 hour lunch 
break are assumed), and the end of the workday. The 
Timer then completes the 24-hour day and produces 
a report of the previous day's operation and produc- 
tion. Output is produced at the end of 24 hours — not 
at the end of the shift — to include the time needed for 
the feller to complete the bunch being worked on 
before quitting for the day. If the required number of 
days has been simulated, the model shuts off. Other- 
^ wise the start of another work day is signaled and the 
-1 process continues. 



TIMER SEGMENT - OVERVIEW 




END Run 



Figure 3. — Overview of the timer segment. 



The Feller/buncher segment (fig. 4) simulates a 
tracked feller/buncher with a boom-mounted shear. 
This boom enables the machine to remain at a har- 
vesting station while it shears and bunches trees 
within a radius of 20 feet, (the radius depends on the 
stand and the harvesting treatment as well as on the 
physical reach of the boom). 

The feller is assumed to move into the woods, cut 
one or more trees, then drop them to one side to form a 
bunch. After all the trees to be cut at that location are 
processed, the feller moves to the next set of trees and 
repeats the process. 



INPUT DATA 

GPSS accepts description data in several forms 
(see table 1). For additional information on data 
types and card formats, see Appendix D or a GPSS 
Programming Manual (Schriber 1974). 

Most of the data required by this program are 
needed in the form of VARIABLES. VARIABLES 
can be defined in terms of the other input types (Ap- 
pendix D), thus allowing the user flexibility of input 
data without card shuffling in the main body of the 
program. For example, a shear-time change from a 
constant (SAVE VALUE) to a distribution (FUNC- 
TION) can be accomplished by simply redefining the 
shear-time VARIABLE. 

The major exception to this convention is the use of 
SAVEVALUES (constants) for the summary values 
(average stand d.b.h., average tree height, etc.) that 
are used simply for output information. 



DATA REQUIRED BY MODEL 

Four types of data describing the equipment and 
harvesting operation must be supplied to run the 
simulation: (1) stand data (table 2); (2) machine data 
dependent on stand conditions (table 3); (3) machine 
data independent of stand conditions (table 4); and (4) 
simulation run control data (table 5). 

The sample output presented in this paper re- 
sulted from running the simulation with the input 
data in tables 2-5. These data were gathered from 
time studies carried out during actual logging opera- 
tions. Although the data in the tables may not repre- 
sent the operating characteristics of the feller/ 
buncher under all conditions, they do present a rea- 
sonably accurate picture of its performance in a stand 
of this type (table 2). 



FELLER BUNCHER SEGMENT - OVERVIEW 




TAKE BREAK OR 
QUIT UNTIL 
HORNING 



F.B. WILL DROP 

LOAD IF BUNCH 

STARTED OR IF 

END OF STRIP 




Figure 4. — Overview of the feller/buncher segment. 



SIMULATION OUTPUT 



The simulation run produces two pages of labeled 
output for each day of the run (figs. 5 and 6). These 
pages contain a summary of the stand data and a 
cumulative productivity report. Additional output, 
with complete run details, is produced at the end of 
the run by the GPSS processor. 



The first page of standard output (fig. 7) presents a 
simulation "map" listing all the blocks in the model 
and the number of transactions that have moved 
through each block. This is of primary value while 
debugging the model. The listing of the QUEUE sta- 
tistics provides detailed information on how the 
feller/buncher spends its time in the woods. The 
SAVEVALUE list contains all the totals necessary to 
calculate productivity if the FORTRAN subroutines 



Table l.—GPSS Data input forms 



Input data type 



Input form in GPSS 



Example 



Simple averages 

or constants 

Range of values 



Equations 
Frequency 

distributions 
Mean and variance for 

a standard distribution 

such as normal, Poisson 

Exponential, etc. 



Initialized SAVEVALUES 

Initialized SAVEVALUES 
for mean and one-half 
width of interval 

VARIABLES 

FUNCTIONS 

SAVEVALUES and 
FUNCTIONS for the 
distribution 



Ave. d.b.h. = 6 in. 

Shear time = 10 to 14 centi-min. 
= 12 ± 2 centi-min. 



Vol. 



.13 d.b.h. 2 - .28 



D.b.h. (in.) 

percent trees 

Wait time is normally 

distributed with 

mean = 10 

variance = 9 



23 



25 



30 



15 



Table 2. — Stand data 



Data required 



Case study 



Form required 
by program 



Name 



Ave. stand diameter 
Ave. stand volume 
Ave. tree height 
Ave. trees/acre 
Strip length 

Distance between strips 
Wood density 
Hours in daily work 

shift 
Actual tree diameters 



Tree volume 



5.5 in. d.b.h. 

3,450 cu. ft./acre 

43 ft. 

844 trees/acre 

725 ft. 

35 ft. 

55 Ib./cu. ft. 

8 hrs. 

Diameter distribution: 
D.b.h. (in.) 
percent trees 
Volume of trees (cu. ft.): 
Volume = -0.283 + .0031 (DBH 2 ) tree ht. 
= -0.283 + 0.133 (DBH 2 ) 
assuming tree ht. = 43 ft. 



3 


4 


5 


6 


7 


8 


9 


5 


23 


30 


24 


14 


3 


1 



SAVEVALUE 


XSAVDBH 


SAVEVALUE 


XSAVVOL 


SAVEVALUE 


XSAVHGT 


SAVEVALUE 


XSTRPAC 


VARIABLE 


VSSTPLN 


SAVEVALUE 


XSMVSTP 


SAVEVALUE 


XSLBCFT 


SAVEVALUE 


XSWKDAY 


VARIABLE 


VSTDIA 


VARIABLE 


VSTVOL 



i cannot be used. For example: 

FBHR = 4020 = feller hours x 100 
FBTRE = 6663 trees cut 
6663/40.2 = 165.75 trees/hr. 
Additional details of the woods operations are 
provided by the output tables (figs. 8-10). The cycle 
j time table (cycle time starts after the feller/buncher 
I drops one load and ends when it drops the next) shows 
there were 2,667 cycles with an average cycle time of 



79 centi-minutes. Of all cycles, 209 cycles or 7.8 per- 
cent took between 100 and 125 centi-minutes. 

It should be noted that all of the output is based on 
random processes in the model. Therefore, the output 
data are themselves random, only a sample of what 
could happen. If the model is run again with different 
random numbers, the results will be different. An 
average over several days or several runs is the best 
estimate of an actual system. 



Table 3. — Machine data dependent on stand conditions 



Data required 



Case study 



Form required 
by program 



Name 



Number of trees within 
reach of accumulator 
head from a given 
location 



Distance feller must 
move to be well 
positioned to cut 
next set of trees 



Distribution: trees in head 

Number of trees 
percent time 
Number of trees 
percent time 
Distribution: feller travel 



ft. 



VARIABLE 



VSREACH 



2-4 


4-6 


6-8 


10-12 


12-16 


16-18 


3 
18-20 


7 
20-22 


10 
22-24 


20 
24-26 


20 
26-30 


8 
30-38 


7 


6 


5 


4 


5 


5 



VARIABLE 



VSFBMOV 



percent time 

ft. 

percent time 



0-5 


5-10 


10-25 


25-35 


35-50 


10 
50-60 


15 
60-90 


30 
90-120 


20 


10 


5 


5 


5 







FULL TREE FIELD CHIPPING SIMULATOR 

NORTH CENTRAL FOREST EXPERIMENT STATION 
FOREST SERVICE, U, S, DEPT, OF AGRICULTU 



RE 



FOREST PRODUCTS MARKETING ANO UTILIZATION PROJECT, DULUTH, MINNESOTA 

and ThE 
FOREST ENGINEERING LABORATORY, HOUGHTON, MICHIGAN 

AUTHORS 

DENNIS P, BRADLEY, ECONOMIST 

SHARON A, HINSAUER, PROGRAMMER 

I, STAND CHARACTERISTICS, 

JOS0.00 CU FT, AVERAGE VOLUME PER ACRE, 

5,50 INCHES, AVERAGE DBH, 

«3,00 FEET, AVERAGE TREE HEIGHT, 

884,00 AVERAGE NUMBER OF TREES PER ACRE, 

7,02 FEET, AVERAGE TREE SPACING, 

II, SYSTEM CHARACTERISTICS AS SPECIFIED BY THE ANALYST, 

A, FELLER-BUNCHER, 1 MACHINE TRACKED TYPE-ROTATABLE BOOM WITH ACCUMULATOR SHEAR 

MACHINE LIMITATIONS 

THE ACCUHULATOR-SHEAR WILL TRY TO 08TAIN A LOAD OF UP TO 1,0 TREES 
BUNCH LIMITATIONS DICTATED BY SKIDDER C*P*CITY| 

THE FELLER-BUNCHER3 TRY TO ACHievE 8KIDOEP BUNCHES WITH A TOTAL SUM OF DIAMETERS EQUAL TO 120,0 INCHES 

Figure 5. — Simulation output: system and stand characteristics. 



KNOWN GPSS SYSTEMS 
DIFFERENCES 

The program listed in Appendix B is operational 
on a UNI VAC 1110 under an implementation of 
GPSS called GPSS-X8, which can be obtained from 
Use Program Library Interchange (UPLI). It should 



run under most GPSS Processors with only minoi 

changes. The following should be checked for youi 

system: 

INITIAL cards— In GPSS-X8 the INITIAL cards ap 
pear last in the deck. They should be moved to th< 
front of the deck for most other GPSS processors 

Job control cards — These are unique to each con^ j" 



Table 4. — Machine data 



Data required 



Case study 



Form required 
by program Name 



Travel time 
(based on distance) 

Shear time per tree 



Drop and bunch time 
per load 



Number of trees per 
accumulator load 



Maximum size of 
skidder bunch FB 
tries to create 
(total sum 
of d.b.h.) 
Gal./hr. fuel used 
Length of nonmechan- 
ical delays/cycle 
(including cycles 
with zero delay) 



Length of mechanical 

delays/cycle 
(including cycles 

with zero delay) 



0-10 


10-20 


20-30 


30-40 


29 
40-50 


39 
50-60 


21 
60-70 


8 
70-90 


1 


1 


.5 


.5 



Travel speed: 
mean 49 ft./min. 
s.d. 19 ft./min. 
Distribution: shear time 
time (centi-min. 
percent of cuts 
time (centi-min. 
percent of cuts 
Distribution: drop and bunch time 

time (centi-min.) 

percent of cuts 

time (centi-min.) 

percent of cuts 

Distribution: trees/accumulator head load 



number of trees 
percent loads 
120 in. 



1-10 


10-20 


20-30 


30-40 


3 
40-50 


67 
50-60 


15 
60-70 


3 


1 


.5 


.5 





1 


2 


3 


4 


5 


22 


23 


28 


17 


10 



5.5 gal./hr. 

Distribution of nonmechanical delays (min.) 

length of delay 

percent of cycles 

length of delay 

percent of cycles 

Distribution of mechanical delays (min.) 



length of delay 
percent of cycle 
length of delay 
percent of cycle 



VARIABLE V$FBTRAV 



VARIABLE VSFBSHR 



VARIABLE VSFBDRP 



VARIABLE VSACCLM 



VARIABLE VSBUNLM 



SAVE VALUE 
VARIABLE 



XSFBGPH 
VSNMDLY 






.15 


.25 


.35 


.40 


.50 


.75 


98.5 
.80 


.2 

1.0 


.2 
2.0 


.2 
3.0 


.1 
4.0 


.1 
5.0 


.3 


.3 


.3 


.1 


.1 


.1 


.1 





VARIABLE VSMCDLY 






.5 


1.0 


1.5 


2.0 


2.5 


5.0 


99 
10.0 


.1 
20.0 


.1 
30.0 


.1 
40.0 


.1 


.1 


.1 


.1 


.1 


.1 


.1 





puter lab, and must be obtained locally. 
HELP BLOCKS— In GPSS-X8, HELP blocks (lines 
65-80 in the program) are used to pass an array of 
up to five integer values to a FORTRAN subrou- 
tine. Only a one-way transfer is permitted; i.e., the 
subroutine cannot pass arguments back to GPSS. 
Some versions of GPSS do not allow HELP blocks. 
In such cases the program can be run without lines 48 
to 80; these data can instead be obtained from the 
standard output. If some form of HELP blocks are 



allowed, their format and the subroutines may have 
to be changed. See the GPSS manual for your com- 
puter installation. 



SENSITIVITY 

Table 6 presents the results of a preliminary study 
of the model's sensitivity. The model was run to fit 



Table 5. — Simulation run control 



Data required 



Case study 



Form required 
by program 



Random number generators 



Number of days 
Standard output 



Random number use 

1 — used by GPSS scheduler 

2 — Shear time and drop time distribution 

3 — trees in reach and distance to move 

distributions 
4 — d.b.h. distribution 
5 — not used 
6 — delays distribution 
5 days 
every 5th day 



RMULT CARD 



START CARD 



FELLER-BUNCHER 



PRODUCTIO 



COST 



S U M M A » Y 



I, A. <J0.20 "OURS WORKED BY FELLER-BUNCHFR ( S) INCLUDING 
2Q8.22 MINUTES Of PERSONAL BREAKS, 
213.36 MINUTES MECHANICAL DELAY, 
26.67 MINUTES NON. MECHANICAL DELAY 
B. 1 FELLEK-OUnChER. TRACKED TYRE-ROT A TABLE BOOM WITH ACCUMULATOR SHEAR 



II. 



ibt>y 



B, 



III. 



--. 


0663 


0. 


2435". 2 


L, 


66«.7« 


f . 


1JRJU 


i. 


66.3" 


B. 


11.97 


C. 


165.75 


C. 


605.81 


E, 


16.66 


F. 


3U6.62 



TOTAL ACCUMULATOR CYCLES COMPLETED - ALL MACHINES. 
2.50 AVERAGE NUMBER OF TREES PER CYCLE. 
R.li AVERAGE VOLUME PER CYCLE - CU.FT. 

BUNCHES COMPLETED 
13. e5 AVERAGE NUMBER OF TREES PER BUNCH. 

5.5<J AVERAGE NUMBER OF CYCLES PER BUNCH, 
50.61 AVERAGE VOLUME PEP BUNCH - CU.FT. 

TOTAL TREES FELLFD, 

TOTAL VOLUME FELLED - CU.FT, 

TOTAL TONS FELLED 

TOTAL OISTANCE TRAVELED - FEET 

AVERAGE NUMBER OF CYCLES PER MACHINE HOUR. 

AVERAGE NUMBER OF BUNCHES COMPLETED PER MACHINE HOUR. 

AVERAGE Numbfr OF TREES FELLED PER MACHINE HOUR, 

AVERAGE VOLUME FELLED PER MACHINE HOUR. 

AVERAGE NUMBER OP TONS FELLED PER MACHINE HOUR. 

AVERAGE DISTANCE TRAVELED PER MACHINE HOUR - FEET. 



iv, a. 22i.io total gallons of fuel used by the feller-buncher. 

5,50 GALLONS PER HOUR, 

,33 GALLONS PER TOM, 

,'1 GALLONS PER CUNIT, 



Figure 6. — Simulation output: productivity. 



»EL*TI vt CLOCK I 



720101 



TRAnS COUNTi 



BLOC" 
Tlv E p 



IMT 



DONE 
DROP 



TO NT 



••• GPSS/xg \0B"«l COMPLETION OF SIMULATION »UN 
ABSOLUTF CL^C"! 720001 



T04V.S 



C 



11 8 

21 



u 

"1 

46 

1 i 



EVENTS 

BPT 

72000 



66 

n 
n 
n 
\n 



C«AIN 

61K PR 
23 12 

108 10 



J .ol 



24 



T0T»L 



48 1 



j" r 
1 7 



BLOC 
NXDA^ 



SUIT 



7 
12 
17 
22 

27 

!i 

4 2 

J7 

07 

02 

112 

1 17 



T"ANS, 


TOT 


0, 




: , 




0, 




0, 




ol 


4 


Oi 


U 


0. 


66 


c, 


•i6 


J, 


- 


:. 


26 


:, 


26 


0, 


26 


0, 


?6 


■. I 


^b 


:, 


2b 


0. 


u 


0, 


a 



5' : 
1 



SOIP 



Oi 

0, 



k i 



s 

! 

si 

se 
hi 

67 

67 
67 
67 
67 

* T 
81 

! 

481 

126 

1 7 



PARAMETERS 


62"28S 




bloc* 



TRAPES, TOTAL 



II 

DUTPT 
28 

FB«oy 

58 

TREE 

IB 
S3 
5f 
63 

te 

73 

78 
8 i 

«e 

93 

98 
103 

hi 



6J7 



0, 

o, 







572 





455 

use 



o ( 

0, 
0, 

0, 666 

0, 666 

0, 666 

0, «U 

0. 2o6 

0, 266 

0, 266 



?bt 
cb 
SO 

48 
48 

1 

UP 

1 



J: a ?7 



1 



6663 





BLOC" 
Day 

o 

1' 

ACCFS 

29 

3u 
39 

M 

UC 

Su 
59 

6(1 

eO 
7<l 
70 
8a 
Bunch 

94 
09 

N09RK 
109 

1 iu 







3 120 





NS. TOTAL 

a! s 

Oi 5 

Oi 5 

i, 1 

o, usA 

o, use 

0i 6663 

0, 6663 

0, 6205 

0. 29 

0. 2667 

0. 2667 

0, 2o67 

0, 2667 

0, 2667 

Oi 481 

0, 481 

Oi 481 

0, 481 

0, 480 

0. 17 



BLOC" TRANS 
5 



10 
15 



35 

:i 

V 

STRIP 
65 

n 



1 1? 



TOTAL 

5 
5 

5 
1 

4SB 

458 

6663 

666} 

2209 

17 

Itll 
2667 
266? 
2667 

481 

us! 

»8J 

-2 



QUEUE STATISTICS 



QUEUE 


MAXIMUM 


AVERAGE 


TOTAL 

entrie! 


ZERO 


PERCENT 
ZEROS 


CONTENTS 


CON^ 


'ENTS 


ENTRIES 


BREAK 
M CBLV 


1 




,04 


481 
2667 


466 


96.0 


1 




•M 


2644 


99. 1 

98.8 


NMDlV 
FdTRv 


1 




2667 


2b36 




,04 


U58 


9 


2.0 


FBSHR 
DROP 




•M 


6663 


2 2? 


3! 5 


1 




2667 


, 7 



AVERAGE 


$ 


AVERAGE 
E/TR1NS 


TIME/TPANS 


TIC 


b l:ll 




'm-M 


1.20 

70.95 




103.55 




72.37 


16,43 




16,98 


16.27 




16.40 



lUMBER 



CURRENT 
CONTENTS 



SAVE VALUES 



SAVEx 







H«DA» 


266? 


m 


884 
6663 


LBCFT 

Sntre 


55 


100 


FBOUT 


ACCLM1 


4 



NR. . . . 




NR 


DATS 


24120J 


FBHR 


FBTI- 

NUMfB 


AWOL 


243541? 


ACLM1 


FBVOL 


BNCOM 
LASTS 

BNLOC 


FBGPM 


55 


BNDIF 


M 


DBH 


FTRvl 


37 


c 


38 



"Wo 

345000 

46 1 
5T2 

■ 



NR,., 
8RKT* 
AVDBH 
FBTYP 

r S\l 

BNVOL 
FTRv2 
39 



•m 



139J4 

725 

C 

19 





NR,., 

DtLT» 
avmgT 

KM 

BndTa 
strip 

MVSTP 
40 



43 

i! 





Figure 7. — Simulation output: standard GPSS form. 



actual data as closely as possible. Then one parame- 
ter at a time was varied and the effect on productivity 
noted. 

These sensitivity figures can be used (1) to give 
some indication of the accuracy required for the input 
data (the greater the sensitivity the more accurately 
that parameter should be specified ) , and ( 2 ) to give an 
indication of what factors should be further exam- 
ined for their impact on the machine's productivity. 

It should be noted that even parameters which did 
not show major changes in productivity at the levels 
of change examined could have larger effects when 
changed to a greater degree or when combined with 
other changes. In all cases, the better the input infor- 
mation, the more accurate the output data will be. 



CONCLUSION 

With the proper choice of input variables, the 
simulation model can be used to study productivity of 
feller/ bunchers under a variety of stand or harvest- 
ing conditions. It is possible to study the effects of a 
change in just one parameter or the effects of many 
changes at the same time. This model can also be 
combined with a model of skidding and chipping op- 
erations to provide productivity information for 
whole-tree systems. 

Perhaps the greatest value of a GPSS model is the 
ease by which it can be modified to suit the exact 
requirements of a harvesting system. Although this 
requires some knowledge of GPSS, it makes simula- 
tion an extremely valuable tool. 



9 



CYCLE TIME ( CENTI-MINUTES) 



FBTIM 



ENTRIES IN TABLE 
2667 




OBSERVED 
FREQUENCY 



PE ?0T^ 



OF 
8 
31 

^b 
is 

7 
3 
1 
1 



U7 
3b 

85 

ho 
64 

u 

16 
q 3 
75 
U8 

n 

7 

§* 
07 
07 
15 
03 
03 
33 



MEAN ARGUMENT 
79,19 

CUMULATIVE 
PERCENTAGE 
8,5 
39,9 
66,7 
82.3 
90,1 
93.9 
95.2 

96. a 

97.3 
98.1 
90,6 
98.9 
99,1 
99,2 
99.2 
99.3 
99. a 
99, U 
99,6 
99.6 
99,7 
100,0 



STANDARD DEVIATION 
167,76 



SUM OF ARGUMENTS 
211203. 



CUMULATIVE 

REMAINDER 

91.5 

60,1 

33.3 

■i:l 

:J 

3.6 

1:1 

i.« 

':J 
il 

il 

:i 



MULTIPLE 
OF MEAN 

6=3 

9a 

it 

89 

ji 

3^79 

alio 






DEVIATION 
FPOM MEAN 



VOLUME (CU. FT. X 100) PER ACCUMULATOR HEAD LOAD 



TABLE 
ACCVL 



ENTRIES IN TABLE 
2667 



UPPE* 

LIMIT 

500 
LOO 

500 

000 

500 
(000 



OBSERVE 
FREQUEN 

Z* 7 
861 

626 

hi 

15 



PERCENT 

OF TOTAL 

28.76 

32.28 

23.47 

11,66 

5.26 

.56 



MEAN ARGUMENT 
913.17 

CUMULATIVE 
PERCENTAGE 
28.8 

84,5 

96,2 

99,4 

100,0 



STANDARD DEVIATION 
538,03 

cumulative 



JEMAINQEI 
71.2 
39,0 

.0 



SUM OF ARGUMENTS 
24354l7t 



.54 

I'M 



fIom mean 
-,76 



Figure 8. — Simulation output: productivity tables (h 



10 



NUMBER OF TREES PER ACCUMULATOR HEAD LOAD 



TABLE 
ACCTB 


ENTRIES IN 
2667 


TABLE 


MEAN ARGUMENT 
2.50 


standard deviation 


SUM 


Of 


ARGUMENTS 
6663. 


UPPER 
LI M IT 


OBSERVED 


PERCENT 
OF TOTAL 


cumulative 
percentage 


CUMULATIVE 
REMAINDER 


^ULTTPLt 

0" MEAN 




DEVIATION 
FROM MEAN 


FREQUENCY 




\ 


706 
685 


26.47 
25,68 


26,5 


.40 




-1.22 


*>!■,?■ 


a? 1 8 


.80 




..(JO 


3 


714 


26.77 


78.9 


21,1 


1.20 




1.23 

2,05 


a 


365 


13.69 


92.6 


7!u 


1.60 
2,00 




5 


t<>7 


7.39 


100,0 


.0 





NUMBER OF TREES PER SKIDDER BUNCH 



TABLE 
BTREE 



UPPER 

LIMIT 

2 

3 
u 
5 
6 
7 
8 
9 
10 

w 

1 a 
5 
6 

I! 

Is. 



OVERFLOW 



ENTRIES In 


TABLE 


MEAN ARGUMENT 


STANDARD DEVIATION 


SUM OF ARGUM 


f-NTS 


461 




13.85 


6.37 




6603. 




OBSERVED 


PERCENT 
OF TOTAL 


CUMULATIVE 
PERCENTAGE 


CUMULATIVE MULT I » 
REMAINDER 0* Mf 


>LE 


DEVIATION 


FREQUENCY 


AN 


F ROM 


MEAN 


2 


2.08 


,4 


99.6 


1" 




-1.86 


10 


2.5 


97.5 


le 




-1 ,70 


lb 


3.33 


5,8 


94,2 




-1.55 


» 


3 12 
5.61 

4.99 


8,9 
14.6 


u-a 


n 




-1.39 
-1 .23 


?n 


19.5 


80,5 


50 




-1.08 


23 


4,78 


24.3 


75.7 


57 




-.91 


25 


5.20 


29.5 


70.5 


65 




-.76 


28 


5.20 


34,7 


65,3 


*6 




-,60 


4. 16 
5.82 


38,9 

44,7 


m 




-.44 

-.29 


11 


7.69 


5$. 4 


47.6 


93 




-.13 


5.20 


42.4 1 
38,7 1 


n 




.02 


18 


3.74 


61,3 




:5 


fl 


a, 78 


6fc, 1 
69.4 


33,9 1 


to 




3 74 


30.1 1 
26,6 t 




,49 


16 


3 33 


73,2 


50 




'$ b 


\t 


3.33 


76,5 


23,5 1 


57 




,80 


3.74 


80,2 


19,8 1 


«u 




.96 


1» 


3.33 


83.6 


16.4 f 


\\ 




1.12 

1 ,2b 


U 


3.33 


86,9 


li *:\ \ 




J 


4. 37 


91.3 

95.0 


hb 




1 ,4a 


3.74 


5.0 1 


73 




1.59 


2a 


4,99 


100,0 


.0 









Figure 9. — Simulation output: productivity tables (II). 



i I 



TABLE 

9V0L 



UPPER 
LIMIT 

1000 

1500 

2000 

2500 

3000 

3500 

aooo 

4500 

5000 

5500 

6000 

6500 

70 0*1 

7500 

6000 

6500 

9000 

9500 
10000 
10500 



DISTANCES (FT) BETWEEN CONSECUTIVE BUNCHES 



lift 1 


ENTRIES IN 

481 


TABLE 


MEAN 
25 


ARGUMENT 
.93 


STANDARD DEVIATION 
27.60 






Su M 


OF AR&UM 
12474, 


ENTS 


UPPER 
LI M IT 


OBSERVED 
FREQUENCY 


PERCENT 




CUMULATIVE 


CUMULATIVE 


MULTIPLE 


DEVIATION 


OF TOTAL 




PERCENTAGE 


REMAINDER 


OF 


MEAN 


FROM 


MEAN 





56 


12.06 




12.1 
23.3 


87,9 




0,00 




-,9a 


5 


54 


11.23 




76,7 






19 




::| 


ii 


5S 


11 .43 




34.7 


65,3 






36 




17 


7,69 




42,4 


57.6 






57 




-.39 


» 


10.60 




53,0 


47.0 
37.2 






77 




:: 


25 


9 77 




62,8 






96 




si 


44 
26 


9.15 

5.41 
1.87 

1:^ 




7 l. q 
77 3 


28,1 

22.7 




1 

! 


it 




:1 


tt 


9 




79,2 


20.6 




54 




:i 


14 




82 1 
85.7 


17,9 




1 


74 




5 


l 6 




14 3 




93 




!87 


55 

60 


i:» 




87.5 
89,8 


12.5 

10.2 




If 




!:« 


6S 


l)25 




21«i 

q 1.5 


8.9 




2 


:W 




1.42 


70 


2 


21 




8.5 




2 




1,60 


OVERFLOW 


41 


8,5? 




100,0 


.0 













VOLUME (CUFT X 100) PER SKIDDER BUNCH 



ENTRIES IN 
481 


TABLE 




OBSERVED 
FREQUENCY 


PEP 


OF 1 


6 


1 . 


19 


3, 


\l 


3, 
6. 


S3 


o. 


S8 


7, 


30 


6. 


n 


P. 
b. 


« 


7, 
6. 


31 


6. 


2a 


a . 


13 


3. 


I 


3. 


29 


6. 


13 


2. 


a 




1 





MFAN ARGUMENT 
5063,24 

CEMT CUMULATIVE 

OTAL PERCENTAGE 

66 1,7 

95 5.6 

74 9,4 

03 15.4 

86 22,2 

90 30.1 

24 36.4 

52 44.9 
03 50.9 

69 58,6 
03 64.7 
44 7" 
99 7 
7a 7 

53 83,4 
86 90,2 
03 96.3 

70 99.0 
83 °9,8 
20 100,0 



STANDARD DEVIATION 
2341,80 



SUM OF ARGUMENTS 
?43f 



CUMULATIVE 
REMAINDER 
96,3 
94.4 
90,6 
64,6 

77, e 

69,9 
63.6 
55.1 
49,1 
41,4 
35.3 
28.9 
23.9 
20.2 
16,6 
9.8 
3.7 

l.f 



•1 

.0 



MULTIPLE 
OF MtAN 
19 
29 
39 
49 
59 
69 
79 
66 
96 
09 

2l 

ue 
se 

t! 

88 
98 
07 



Figure 10. — Simulation output: productivity tables (III). 



15417, 

DEVIATION 
FROM MEAN 



12 



Table 6. — Sensitivity of the simulation 



Change in 
run parameters 




Productivity 


Tons/hr. 


Effect 


Standard run 


16.66 




Decrease shearing time 
by 10 percent 


17.48 


Increase 5 percent 


Decrease swing & drop 
time by 10 percent 


16.84 


( 1 ) 


Increase travel speed 
by 10 percent 


16.70 


( 1 ) 


Increase average number 
of trees in accumulator 
head load from 2.5 
to 3.5 


17.82 


Increase 7 percent 



Increase average number 16.72 ( 1 ) 
of trees within reach 
from 14 to 16 

Increase average d.b.h. 23.94 
by 1 inch 

Increase frequency of 14.64 

mechanical delay from 
1 percent to 2 percent 
of the cycles 

1 No major change in productivity observed for this level of change in 
parameter. 



Increase 43 percent 
Decrease 12 percent 



LITERATURE CITED 

Bradley, Dennis P., Frank Biltonen, and Sharon 
Winsauer. 1976. A computer simulation of full tree 
field chipping and trucking. U.S. Department of 
Agriculture Forest Service, Research Paper NC- 
129, 14 p. U.S. Department of Agriculture Forest 
Service, North Central Forest Experiment Station, 
St. Paul, Minnesota. 

Bradley, Dennis P., and Sharon Winsauer. 1976. 
Solving wood chip transport problems with com- 
puter simulation. U.S. Department of Agriculture 
Forest Service, Research Paper NC-138, 8 p. U.S. 
Department of Agriculture Forest Service, North 
Central Forest Experiment Station, St. Paul, Min- 
nesota. 

Bradley, Dennis P., and Sharon Winsauer. 1978. 
Simulated full-tree chipping: Model compares fa- 
vorably to the real world. Forest Products Journal 
28(10):85-88. 

Schriber, T. J. 1974. Simulation using GPSS. 533 p. 
John Wiley & Sons, New York. 

Mattson, James A., Dennis P. Bradley, Eugene Car- 
penter, Sharon Winsauer, and Jerold T. Hahn. 
1978. Forest residues energy program; — final re- 
port. U.S. Department of Agriculture Forest Ser- 
vice, North Central Forest Experiment Station. 
Contract to ERDA, Contract No. E-(49-26)-1045. 
(U.S. Department of Energy Report TID-28416.). 

U.S. Department of Agriculture, Forest Service. 
1978. Forest residues energy program. 297 p. Final 
report. U.S. Department of Agriculture Forest 
Service, North Central Forest Experiment Station, 
St. Paul, Minnesota. 



13 



APPENDIX A 

VARIABLE LISTS AND 
DEFINITIONS 



Definitions 

IMPLICIT TIME UNIT— Centi-minute 
MODEL SEGMENTS, TRANSACTIONS, AND PA- 
RAMETERS 

Segment 1 — timer 
Transactions — 1 time keeper 
Parameters — NONE 
Segement 2 — feller/buncher 
Transactions — feller/buncher 
Parameters 
PI — not currently used. 
P2 — clock time marking start of cycle. 
P3 — next location of feller/buncher (distance 
in feet from the beginning of this strip or 
row). 
P4 — current location of feller bunches (dis- 
tance down strip in feet). 
P5 — tree number being cut. 
P6 — limit for trees in accumulator head (num- 
ber of trees). 
P7 — maximum size of skidder bunch at- 
tempted (sum of diameters). 
P8 — diameter of current tree. 
P9 — distance feller/buncher must travel to 

next set of trees. 
P10 — number of trees currently within reach 

of boom. 
Pll — actual sum of diameter of accumulator 

load. 
P12 — total volume in accumulator head. 
P13 — number of trees in load. 



Functions 

ACCLM Distribution of the number of trees in ac- 
cumulator loads. 

DBH Distribution of diameters in the stand. 

FBDRP Distribution of drop times. 

FBMOV Distribution of the distances feller must 
move to reach more trees. 

FBSHR Distribution of shear times. 

MCDLY Distribution of mechanical delay times in 
each cycle (including cycles with zero 
delay). 



NMDLY Distribution of nonmechanical delay 
times in each cycle (including cycles with 
zero delay). 

REACH Distribution of trees within reach from a 
random spot. 

SNORM Distribution used to obtain a normal dis- 
tribution for sampling with a mean = 0, 
variance = 1. Lower end truncated. 



Savevalues 

ACLMl Mean accumulator head limit (number of 

trees). 
AVDBH Mean stand diameter (inches x 100). 
AVHGT Mean stand height (feet). 
AVVOL Mean vol./acre (cu.ft. x 100). 
BNCOM Number of skidder bunches complete and 

ready for skidding. 
BNDIA Sum of diameters of trees in the bunch 

(inches). 
BNDIF Distance between bunches (feet). 
BNLM1 Limit for skid bunch size — maximum sum 

of diameters. 
BNLOC Location of bunch (distance from end of 

strip). 
BNTRE Number of trees in a bunch. 
BNVOL Volume of bunch (cu.ft. x 100). 
BRKTM Total time spent on breaks (centi-min- 

utes). 
DAYS Number of days worked. 
DBH Diameter breast high (d.b.h.) of current 

tree (inches). 
DELNM Total time lost to nonmechanical delays 

(centi-minutes). 
DELYM Total time lost to mechanical delays 

(centi-minutes). 
FBACC Total number of feller/buncher accumula- 
tor loads. 
FBDIS Total distance feller/buncher has traveled 

(feet). 
FBGPH Feller/buncher fuel usage (gal./hr.). 
FBHR Total time feller has worked (hours x 

100). 
FBOUT Distance feller travels to get to the woods 

(feet). 



14 



FBTIM Total time feller has been on job (minutes 
x 100). 

FBTRE Total trees cut. 

FBTYP Feller/buncher type; tracked = 0, rubber- 
tired = 1. 

FBVOL Total volume cut (cu.ft. x 100). 

FTRV1 Mean travel speed (ft./min.). 

FTRV2 Standard deviation of travel speed. 

LASTB Location of last bunch. 

LBCFT Density of plot species (lb./cu. ft.). 

MVSTP Distance feller/buncher travels between 
strips. 

NUMFB Number of feller/bunchers on the job. 

STPLN Length of strip or row (feet). 

STRIP Number of strips harvested. 

TRPAC Mean stand density (trees/acre). 

WKDAY Length of work shift (hours). 



Switches 

COFFE Set by timer to indicate coffee break. 
DAY Set by timer to indicate beginning of work 

day. 
FBMOV Set by feller when he must move (no more 

trees in reach). 
LUNCH Set by timer to indicate lunch break. 
STRIP Set by feller/buncher when he has cut the 

last tree of current strip. 



Tables 

ACCTR Trees per accumulator load. 

ACCVL Volume per accumulator load (cu.ft. x 

100). 
BDIA Sum of diameter of trees in a skid bunch 

(inches). 
BDIF Distance between bunches (feet). 
BLOC Bunch location — distance down strip 

(feet). 
BTREE Number of trees in skidder bunch. 
BVOL Volume per skidder bunch (cu.ft. x 100). 
FBTIM Cycle time for feller/buncher, from drop to 

drop, including delays (centi-minutes). 



Queues 

BREAK Coffee and lunch break time. 

DROP Swing and drop time. 

FBSHR Position and shear time. 

FBTRV All travel time. 

MCDLY Mechanical delays. 

NMDLY Nonmechanical delays. 

Variables 

ACCLM Accumulator head size limit (number of 
trees). 

BNDIF Calculate distance between bunches 
(feet). 

BREAK Time between breaks (centi-minutes). 

BRKTM Time spent on breaks (centi-minutes). 

BUNLM Skidder bunch limit (sum of diameter). 

DELNM Nonmechanical delay time (centi-min- 
utes). 

DELYM Mechanical delay time (centi-minutes). 

FBDRP Swing and drop time (centi-minutes). 

FBHR Total hours worked. 

FBMOV Distance feller/buncher must move to put 
more trees in reach (feet). 

FBSHR Shear time (centi-minutes). 

FBTRV Travel time (centi-minutes). 

MCDLY Length of mechanical delay (centi-min- 
utes). 

NEXTY Next harvesting location down strip = 
current position + 95 percent distance 
traveled (allows for travel not in a straight 
line). 

NIGHT Remainder of 24 hours after work shift 
(centi-minutes). 

NMDLY Nonmechanical delay time (centi-min- 
utes). 

REACH Number of trees within reach of boom at 
current feller/buncher position. 

SPEED Travel speed (ft./min.). 

STEND Distance to the end of strip (feet). 

STPLN Length of strip or plot (feet). 

TDIAM Tree diameter (inches). 

TVOL Tree volume (0.01 cu. ft.). 

WKDAY Length of work shift (centi-minutes). 



15 



APPENDIX B 
PROGRAM LISTING 



/ - 






u 



19 

20 
21 



FELLER 8UNCHFR SIMULATOR 
UPQ4TE0 JULY 1979 



U,S, FOREST SERVICE. NORTH CEnTRai FOREST EXPERIMENT STATION 
MARKETING PROJECT, DULUTH, MINNESOTA 
ENGINEERING PROJECT, HOUGHTON, MICHIGAN 



AUTHORS; SHARON A, WINSAUER 
DENNIS P. BRAOLEY 



MASTER TIMER AND REPORT GENERATOR 



THIS SEGMENT CONTROLS THE STOP AND START OF THE FELLER BUNCHER, 
THE HOURS WORKED PE« DAY, XSWKDAY, AND THE NUMBER OF 
DAYS WORKEDCARGUMENT A ON STARTjCARD) 

IT ALSO CALCULATES OUTPUT VALUES, VALUES TO BE SENT TO HELP BLOCKS, 



WKDAY VARIABLE 
K FVARIABL! 
T VARIABLE 



BREAK FV 
NIG*" 



TIMER GENERATE 
NXDAY SAVEVALUE 

LOGIC 5 
DAY ADVANCE 
LOGIC S 
ADVANCE 
LOGIC S 
ADVANCE 
LOGIC S 
ADVANCE 
LOGIC R 
ADVANCE 



XSWKDAYofcOOO 

V$WKpAY/4 

(2«-X$WKDAY)*6000 

DAvUll 

DAY 

VSBREAK 
COFFE 
VSBREAK 

VSBREAK 

COFFE 

VSBREAK 

DAY 

VSNIGHT 



WORKDAY IN CENTIMINUTES 

TIME BETWEEN BREAKS 

REST OF 2« HOURS IN CENTl^iNUTES 

NUMBER OF CURRENT WORKING DAY 

TIME FOR COFFEE BREAK 

WORK UNTIL LUNCH TIME 

AFTERNOON COFFEE 

END OF WORK DAY 

WAIT UNTIL NEXT MORNING 



* • * HELP BLOCKS GENERATE REPORT AT END OF EACH DAY 



* * * REPORT VARIABLES 



SAVEVALUE 
SAVEVALUE 
SAVEVALUE 
SAVEVALUE 
FBHR FVARlAglE 
BRKTM FVAR; ABLE 
DELYM FVARllABLt 
DELNM FVARIABLE 



FBHR. VSFBHR 
BRKTM, VSBRKTM 
DELYM, VSDELYM 
DELNM, vSDELNM 
XfFBTIM/60 

ot$break*qc$break 
otsmcdly*qcsmcdly 
ot$nm5ly*jc$nmdly 



HOURS FB HAVE WORKED 
TIME FB SPENT ON PREAKS IN MIN*100 
MECHANICAL DELAY TIME OF FB 
NON-MECHANICAL DELAY TIME OF FB 



* * • STAND DATA AND MACHINE CHARACTERISTICS REPORT 



HELP 

HELP 



SYSTM,X$AVVOL,X$AVDBH,X$AVHGT,X$TRPAC 

SYSTM,X$NUMFB,X$ACLM1»X$FBTYP,XSBNLM1 

AVE VOL, AVE DBH.AVE HE IGHT, TREES/ ACRE » NUM OF FEL 

CUMULATOR HEAD LIMIT, FB TYPE,SKIDDER BUNCH LIMIT 



LCRS 



FB AC 

FELLER BUNCHER PRODUCTION REPORT 

HELP FELBN,XSNUMFB.XSFOTYP,X$BRKTM.XsDELYM,X$DELNM 
NUM OF FELLER BUNCHERS, FB TYPE, BREAKTIME, DELAY TIME 

HELP FELBN,X$FBTRE,XSFBVOL.X$BNCOM,XSFBHR.XSFBDlS 

TOT TREES, TOT VOL»SKlD BUNCHES COMPLETED#FB HOURS, TOT DIST TRAVELED 

HELP FELBN,X$FBACC.XSLBCFT,X$FBVOL t X$FBGPH 

NUM OF ACC LOADS, WOOD DENSITY LB/CUFT, TQT VUL, GAL PER HOUR USED 



!2 SPLIT 1,NXDAY 

!3 OUTPT TERMINATE 1 



16 



GENERAL FUNCTIONS NEEDED IN MODEL 



• • * 

* 

# 
* 

* 

SNORM 





* FN$SNORm is used to obtain A SAMPitng of an appro* 
NORMAL DISTRIBUTION OF mean AND STANDARD DEV J 

note! to avoid data loss from intergerazation, 
function vaiueS have Seen • 

YLTiplied By 10 

meanWfn$8norm 



Imately 



NOTE! LOWER END OF FUNCTION HAS BEEN TRUNCATED 
TO AVOID NEGATIVE ADVANCE BLOC* 
WHEN STANDARD DEVIATIONS ARE LARGE 



- S,DEV)/10 

ADVANCE I ICKS 



FUNCTION RN2,C20 



JCTI 



•5H!;^;f;^;^!ar»!!B!!<;t!!f;«n^^ 4 ''''^ {4 -^ 



TRACKED TYPE FELLER BUNCHER 



THIS SEGMENT IS A FELLER BUNCHER WITH A BOOM MOUNTED SHEAR 

SINGLE 



IT CAN BE USED AS I 
THE F 



r ELLER BUNTHER IS 
OUT ONE STRIP BA 



ASSUMED TO W 
CK THE NEXT, 



TEM OR ACCUMULATING HEAD FB 
ORK IN A ZIG-ZAG PATTERN 



THE s &3cS s?rao&i;feWT&rak^ he,d suNCHts to obt,,n * 



* 
* 
* 

* 
* 

* 

* 
« 
* 

* * * 

* * * 

* 



FELLER BUNCHER PARAMATER ASSIGNMENTS 
NOT USE 



JOT USED 

Current location of fb 

bunch time 

tree number 

current accumulator limi 

current bunch size limit 

TREE DIA 

DIST FB WILL TRAVEL 

* TREES IN REACH 
PI 1 SUM OF DIAMETER IN ACC. H 
PJ2 TOTAL VOLUME IN ACC, HEAD 
P1J NUMBER OF TREES IN LOAD 






HEAD 



GENERATE FELLER 8UNCHERS , INITILIZE ACC H£AD 
DEFINING VARIABLES FOR FELLER BUNCHER 



FVARIABLj 

FVARIABL 

FVARlAgt 



P9l9§7l06*PJ 
PJ-XSSTPLN 



8TEN0 
VSTIH TABI 'P2»0,2 



{STANCE BETWEEN BUNCHE 
CTUAL Dill CPWN STRIP 
ISTARCE TO END Of 3TRI 



OUTPUT TABLES FOR FB 
CYCLE DATA 



FT) 



ME 



TABL 
TAB. 
TAB. 

tabC 



X|BNDIA,0 
XSBNTftEiO 
X|BNDIF,0 
XSBNLOCO 
XSBNV0L»0 



''11 



ml 



TABLES OF SKIDDER 
JCHES CREATED 



17 



64, 

« 



tt 






J! 
| 



ill: 
If: 



ill: 
III: 



«?4 

25 

ft 

30 



33 
36 



39 

40 
4J 



ua 



43 

oa 

45 

2* 

48 

a9 

!i 

53 

5a 

55 

fi 



* * • GENERAT 
CCFB GENERATE 



SAVEVALUE 
INIT ASSIGN 
ASSIGN 
ASSIGN 
ASSIGN 
ASSIGN 
ASSIGN 
MARK 

f 

* • • FB MOVES 

' TRAVEL 

WITHIN 

FBMOV QUEUE 

ADVANCE 

DEPART 

SAVEVALUE 



E FELIE* BJNCHERS, INITIALIZE FIRST 01 ST, ACCUMULATOR HEAD 
1 1» I* 10,15 
STRIP, 1 



!k? 



.#V$ACCLM 
', v$bun[m 
>,x$fbout 



DONE. 



ASSIGN 
ASSIGN 

ASSIGN 

ASSIGN 
LOGIC R 



-.0 
13,0 
6,V$ACCLM 

1: 

TO NEXT LOCA 
DISTANCE AND 
REACH AT THI 

FBTRV 
VSFBTRV 
FBTRV 
FBDIS*,P9 

10. VSREACH 
4,P3 

9, VSFBMOV 

3, VSNEXTY 
FBMOV 



INITIALIZE STRIP COUNT 
INITIALIZE ACCUMULATOR HEAD 



ACC. LOAD LIMIT 

BUNCH SIZE LIMIT 

DIST FB TRAVELS To START WORK 

MARK START OF CYCLE 

TION. RESETS SWITCH FBMOV, RECORDS 

TIME AND DETERMINES « OF TREES 
S LOCATION, 



FB MOVES TO NEW LOCATION 
RECORD TRAVEL TIMF 
AND DISTANCE 

DETERMINE * OF TREES IN REACH 
ACTUAL DISTANCE down The STRIP OF 

THE FELLER BUNCHER 
DISTANCE FB WILL HAVE TO MOVE 

WHEN DONE HERE 
NEXT LOCATION OF FB 
RESET TO INDICATE FB NOT 

READY TO MOVE 



# * * CHECK IF 

IF SO G 

TEST E 

* * * TREE- 



•REE ASSIGN 
ASSIGN 

QUEUE 

ADVANCE 

DEPART 

ASSIGN 
ASSIGN 
">AyEvAL 



FB HAS 
TO D 

P13,0 

'ELLER 

TREES 

8*vjT 

FBSHR 
VSFBS 
FBSHR 

1 



CARRIED A TREE TO THIS LOCATION 
ROP 



SAVEVALUE 

ASSIGN 

ASSIGN 

TEST G 
TEST GE 

TRANSFER 

tflfV 



23< 



59 

60 S 

* 



TRANSFER 
TRIP LOGIC 5 



,DROP 

BUNCHE 
S 

HI 



IS FUL 
"THI 



♦ ,P 

12*, v 
io-ii 



j 



,DROP 

FBMOV 
P3.VS 

XSBNT 



DIAM 

HP 

8 

8 
STVOL 



,DONE 
6. TREE 



IF OF TREES IN HEAD > DROP 

R DEFINES AND SHEARS TREES UNTIL ACCUMULATOR 
L OR UNTIL HE RUNS OUT OF 
N REACH 

TREE NUMBER 
DBH 



STPLN, 
RE,0,D 



STRIP 
ROP 



l? 



BMOV 
RIP 



SHEAR T 
RECORD 

TOTAL S 
* OF TR 

TOTAL 
ONE LE 

IF NO M 
IF ACQ 

GO TO 
ACC HEA 



DONE 
REA 



CH 



SET SHI 
IF AT E 

IF BUNC 
OTHER 
ALONG 

SET SWITCH INDICATING END OF STRIP 



REE 

SHEAR TIME 

UM OF DBH OF LOAD 
EES IN ACC HEAD 

IQL OF ACC, LOAD 
>S TREE IN REACH 

ORE TREES HERE GO TO 
NOT FULL I A TREE IN 

TREE TO CUT IT 
D FULL GO TO DROP 

TCH THAT THE FB MUST MOVE 
NO OF STRIP (jo TO STRIP 
H IS ALREADY STARTEDiDROP LOAD 
WISE, CARRY TREES IN ACC. HEAD 
WITH (FBMQV) 



18 



61 

ti 

64 
65 
66 
67 



66 

W 



7a 

n 

n 

79 

ec 
e l 

8c 
63 

?! 

87 
86 



,07 

06 

oo 

10 

t 



* * * DROP' 



DPQP 



QUEUE 

ADVANCE 

DEPART 



Swing and crop the fb 
See if the bunch is 

DROP 
VSFBDRP 

DROP 



LOAD, 
LARGE 



ACCUMULATE LOAD STATISTICS 
ENOUGH FOR SKICDERS 



* • 



savevalue 

savevaluC 
savevalue 

• * DELAY- 
QUEUE 
ADVANCE 
DEPART 
QUEUE, 
ADVANCE 
DEPART 

SAVEVALUE 
SAVEVALUE 
SAVEVALUE 
SAyEVAi UE 

TABULATE 



8NL0C,P« 
BNVf ' 
B> " 
Bl 



in 

1KB 

ilGN 



TABULATE 

MARK 

A! 

A! 

ASSL. 

ASSIGN 

ASSIGN 

TEST L 

GATE LS 

* 

* 

• • * * BUNCH. 

* 
* 

6UNCH TABULATE 
[E 

•E 



sm::Hf 

NTftE*,Pl3 

IF A DELAY IS DUE 

HCDLY 
ViMCDLY 

HCDLY 

N M D L Y 

V$NHDLY 

NMDLY 

FBACC*,1 

*BTRE*,P13. 

F8V0U,P12 

FBTIM+,MP2 

F3TIH 

2 

11,0 

12,0 

13,0 

6, V$ACCL* 

7, VSPUNLH 

X$BNDIA,P7, BUNCH 

FBMQV,TREE 



SWING AND DROP BUNCH 
RECORD SWING & DROP TIME 

DISTANCE DOWN STRIP 



TOTAL VOLUHF 
TOTAL DIA OF LOA 
TOTAL NO OF TREE 



TO 



OCCUR, IT HAPPENS HERE 
MECHANICAL DELAY 
NON-MECHANICAL DELAY 



DS 



KEEP TRACK OF NUMBER OF ACC. LOA 
FB RECORDS PRODUCTION OF 
ROQUETS AND OF TIME 



REINITIALIZE ACC HEAD 



NEW ACCUMULATOR LIMIT 
NEW BUNCH LIMIT 



TO BUNCH 
IF MORE TREES IN REACH 
GO TO TREE FOR 
ANOTHER ACC, LOAD 



TABULATE BUNCH DATA TO USE FOR SK IDDERS, Rt IM T I AL I ZE 
PARBN MATRIX , CHECK IF FB IS AT END OF STRIP 



TABULAT! 
TABULATI 
TABULAT! 



SAVEVALUE 
E 

UE 



•ABULATE 
iAVEVAl ■■ 



'ALU 
iAVEVALU, 
SAVEVALUE 
SAVEVALUE 
SAVEVALUE 



NOBRK 



# * # 
*QUIT 



ADVANCE 
LOGIC R 
GATE LS 
ADVANCE 
LOGIC R 

* QUIT- 

SAVEVALUE 
DEPART 
GATE LS 
MARK 
GATE LS 
GATE LS 

ASS 



BDIA 
BTREE 
BVOL 
BLOC 

BNDIF, VSBNDIF 

BDIF 

LASTB,P4 

BNVOL»0 

BNDIA,0 

BNTftE,0 

BNCOM+, 1 

BREAK 
COFFE, NOBRK 

LUNCH, QUIT 
3000 

LUNCH 



RECORD BUNCH DATA FOR SKID SEG 

DISTANCE BETWEEN BUNCHES 
ZE BUNCH DATA 



n nviM 



: BUNCH 
IUNCH 



ON 



TABULATE TIME 

IT'S TIME TO QUIT FOR THE DAY 



IF TIME FOR COFFEE OR LUNCH, 
TAKE BREAK 

CHECK IF BREAK 

15 MINUTE COFFEE BREAK 



30 MINUTE LUNCH BREAK 
BREAKS AND CHECK IF 



FBTIM*,MP2 

BREAK 

DAY 

2 

FBMOV,TREE 

STRIP, ■ 



FBMOV 



kVEVALUE 

avevatue strip*, 

logic r strip 

Transfer ,fbmov 



3, VSSTEND 
ASTB,0 
1 



TABULATE TIME SPENT ON BREAK 

IF SHIFT IS OVER, STOP UNTIL TOMORROW 

MARK START OF CYCLE 

MORE TREES IN PEACH -GO TO TREE 

IF NOT END OF STRIP -GO To FBMOVE 

STRIP DISTANCE 

COUNT STRIPS 



L9 



HI 

111 
"« 



SI 



/ • * • INPUT DATA 

FBTRV FVARIABLE P9* 1 OO/VSSPEED 

* 



TRAVEL TIME (DIST/SPEEDJ IN 



P9 • 01 31 *N 

SPEED FVARIABLE X$FTRVl*XSFTRV2*FN$3N0RM/ig FELLER BUNCHER TRAVI 
FB3HR FVARIABLE PNJFBSHR POSITION & SHEAR TIME 

FBORP FVARIABLE FNSF8DRP SWING AND OROP TlMgS 



jENTIMIN, 



ACCLM FVARIABLE 
BUNLM FVARIABLE 



FNSACCLM 
XSBNLMI 



ACCUMULATOR HEAD LIMIT 

skiodeB bunch 8IZ? limit 



REA< 



EACH FVARIABLE FNSREACt 

BMOV FVAR ABLE FN$FBMO\ 

TDIAM FVAR ABLE FN$pBH 

MCDLY FVAR ABLE FNSMCDLY 

NMDLY FVAR A8LE FNSNM- 

8TF-LN VARIIBl! X$8TP 

Tvot fvaAiabV 



E FNlNMDLY 

X$8TPLR 
E U33*P6*P8-283)/10 



NUMBER OF TREES WITHIN REACH 

DIST FB MUST MOVE TO REACH MORE TREES 

SUPPLIED FROM STAND DBH DATA 

MECHANICAL DELAY 

NON-MECHANICAL DELAY 



TREE 
100 



VOL IN 
TH'S OF 



TERMS OF DIA(P8) 
CU FT, 



DEFINING FUNCTIONS • MACHINE DATA 



THIS INPUT DATA WAS OBTAINED FROM ACTUAL TIME STUDIES 

'm zD 8KLH 197e 



* POSITION AND S . 
oj85^9^0/ t fcsl , 20/j8^5?/,»7,a0/,98 # 50/,99,60/,«)95,70/l,0,90 



PER STEM 



* TREES IN ACC HEAD 
TIME* IN CENTI-MIN 



SWING AND DROP TIME IN CENTI-MJN 
JNCTION 



oJ8/,Il,i'o5,i81 l 20/.^lj5S/,98,aO/,99,30/,«»95,60/l,0,70 

• DEFINING FUNCTIONS ■ STAND DATA 

REE DIAMETER 



DBH FUNCTION RNU.D7 TREE DIAMETER INCHES 

-08, 3/, 28, J/, 58, 5/, 82; 6/, 96, 7/, 99, 8/ 1,0, 9 

REACH FUNCTION RN3,C12 TREES IN REACH AT A GIVEN SPOT 

0,27,OJ,«/,i,6/.2,8/,6,U/,6e,16/,75#20/,8t,22>,86 > 2tt/,9,26/,*5,30/t,36 

*FBMOV FUNCTION RN3.C9 PJ8TANCE FELLER MUST MOVE 

0,0/ t l,5/,2S,rO/,55,2S/,75,35/,e5,50/,9,5d/,55,9o/t,,r20 

*MCDLY FUNCTION RN6.D11 MECHANICAL DELAY 

NMDLY FUNCTION RN6.DU NON-MECHANICAL DELAY 

INITIAL CONDITIONS FOR STAND 

HIHJUJJ* '^^ "°XvE M : H D5H MC ;?Wr00 
aS AVE TREE HGT. apprc 

JaSOOO CU FT/ACREMOO 

II? 



« * • 



INITIAL 
N T A . 

n: It! a 

;:n:M: 
n :t.:a. 

: n!tiac 



X 
X,. 

X|MyS 

xjwkda> 



:ft 



lox ai< 



I5l"«fi« 



V 



M 



. STRIP LENGTH 

TANCE between stri 

PER CU FT • RED MA 
IN A WORK DAY 



PS 

PLE 



• * ■ « INITIAL CONDITIONS FOR EQUIPMENT -DATA TAKEN FROM USFS MPH TIME STUDY 



mm 

UN 1 - A 
,, N , ,,,. A 

; n ;;: a 

, N A^ 
NfTIAt 

END 



;i 



Ifbout 



51 

X 

X 
X 



ACLM1 
BNLMl 
FTRVl 
FTRvi 



(1 

i! 

loo 



MAX 
M 

M 



MAX ATTEMPTE 
AVEL 3 

v. Travel speed 



AN TR 
TO, D 

1ST F 



■M 



TREES 

~~MP~ 
VE 

/HR 

TRAVELS TO 



RUBBE 

" WOODS 



OAD 

UNCH SIZE 

T/MIN 

FT/MIN 



4"S\ 

5, ,5» l 



20 



1?: 



5. 
6< 

,9. 



2a, 
25- 
2b, 
27- 
28- 
29- 

i!! 

33. 

36. 

37- 
38- 

11: 

43- 

aa. 

45- 
46. 

U: 

??: 

!* ; 

53- 

5a. 

55- 
56- 
57- 
58- 
59- 
60- 

& 

63- 

64- 

Si: 

67- 
68- 

»?' 



7a, 

?i; 

77- 
78. 
79, 
80- 



SUBROUTI^E SY§TM(IX) 
DIMENSION IX(5) 
DATA 1/0/ 



bX 



a. 'NORTH CENTRA! • 
SERVICE. U, S, DEP 



FQRMAT(lHl.a3X, 'FULL TREE FIELD CHIPPING '. 
A 'SIMUI AT0ft'/lH0,5ffx f »DFVPi OPED BY THE'/lH .4 
B.'F0PE"t fxpe r I m ENT StaTIoni/ih .uax, 'FOREST 
CT. OF AGRICULTURE '//1H0,29X, tFO&EST PRODUCT 

TutlLlZATIQN PROJECT, DUL.UTH. mt NNESOT A • / 1 H ,61X.'AND THE'/lH . 
FaQX, 'FOREST ENGINEERING LABORATORY, HOUGHTON. MICH IGAN i /'i mq . 62X 
F, 'AUTHORS '/1H ,52X, 'DENNIS P, BRADLEY* ECONOMIST • /1H ,5lX,'SHAR0N 
G A. WINSAUER, PROGRAMMER'//) 



*CE. U. 
maPKETInC 



Ind 



2 FOP 
A ' 
B' P 
C A 

D'PE 

3 FOP 
A' AN 
B 5 
C , 



a FOP 

A'AN 
B5X, 



MATMH 
AVERAG 

AGE 
R ACRE 

MATtlH 
ALYST. 
X, iTRi 

//<>x, 

M £M H 



0,'I, STAND CHARACTERISTICS, '//1H0, Fi2,2,' CU FT, 

E VOLUMFt. 

e.'/Tho.f! 

tree heig 

, '/1H0,F12 



2.2,' inches, average dbh, '/1h0,f12.2, • 
ht, '/lh0 t f12,2,' average number of trees 
,2,' feet, average tree spacing,'/) 



FEET 



STEM CHARACTERISTICS AS SPECIFIED BY THE ', 
»A. FELLEP-BUNCHER, ', 10X ,110.' MACHINE', 
ROTATABLE BOOM WITH ACCUMULATOR SHEAR' 



•RUBBE* 
5 FORMATflH 



6 FOR 
A ' 

7 FOR 
AlHO 
B » 

C F6 



MAT( 1 
A LOAD 

MATU H 

SKIDDE 
.1* ' 

IX C 1 



0,'II, SY 
•//1H ,2x, 

CKED TYPE- 

'MACHINE LIMITATIONS '/) 

0,'II, SYSTEM CHARACTERISTICS AS SPECIFIED BY ThE ', 
'//1H .2X,»A FELLER-BUNCHER, S 10X ,110, • MACHINE", 
R-TIRED TYRE', //,9X, 'MACHINE LIMITATIONS, •/) 

, lax, 'THE SHEAR CAN HANDLE ONLY ONE TREE,') 

1 OX, 'THE ACCUMULATOR-SHEAR WILL TRY TO OBTAIN', 
,F6,1,' TREES') 



H , 10X, ' 
OF UP TO' 



0,8X,'BUNCH LIMITATIONS DICTATED RY SKIDDER CAPACITY|'/ 
'THE FELLJR-6UNCHER5 TRY TO. 

INCHES')" 



_ 1ERS TRY TO ACHIEVE 
WITH A TOTAL SUM OF DIAMETERS EQUAL TO', 



XI 

X2 s IX(2 

x3 = !xh 

xa = ixf 4 

X5_ = IX(5! 

!f"( I*,GT, 2 I I 
GO TO (10,20), I 



1 



10 CONTINUE 

AWOLs Xl/100. 
AVVOL = AVERAGE VOLUMN PER ACRE - CUFT, 

AvDBHs X2/100. 
ACDBH r AVERAGE D§H - INCHES 

AVHGTsXJ 
AVHGT s AVERAGE TREE HEIGTH - FEET 

TRPAC=Xa 
TRPAC = AVERAGE NUMBER TREES PER ACRE 

TRSPC=(a3560./TRPAC)**,5 
TRSPC e AVERAGE TREE SPACING - FEET 

WRITE (6.1) „ „ 

WRITE (6, 2) AWOL, AVDBH, AVHGT, TRPAC, TRSPC 

RETURN 

20 CONTINUE 
NUMFBsl 
NUMFB s NUMBER OF FELLER-BUNCHERS 

ACCL M = X2 
ACCLM ACCUMULATOR L I M I T SUM OF DBH 
FBTYPsX3 
FBTYP FELLER BUNCHER TYPE a DROTT 

BUNLM s xa 
BUNLM BUNCH SIZE LIMIT 



1 s RUBBER-TIRED TYPE 



LM BUNCH SIZE LIMIT 
IF(FBTYP,FQ,0)WRTTE(6,3)NUMFB 
lF(FBTYP,E0.1)WRlTE(6.a)NUMFB 
IF (ACCL^ .L.T, 1.) WRJTE(6,5) 
IF (ACCLM ICE! 1.5 WRITE(6,6) 
WRITE(6,7) BUNLM 



ACCLM 



RETURN 
END 



21 



s 

D 

¥ 

I F 
A • 
B/ 



UBROUTINE FELBN (IX) 
IHENSION IX(5) 



$hl m 



R FBACC,FBDIS,BNCQM 



\ 



HER PRODUCT», 
R Yi,//,1H 1 i32('*'),//// 

D BY FELLER-BUNCHER(S) ' 



a 5* 

2 F 
A' 
B/ 



c/ 

D 1 

3 F 

A' 
B' 

a F 
Al 

5 F 

6 F 
A7 

B> 

:f 



&' 



7 F 
A< 

8 F 
A' 
8' 

C ' 
D« 
E' 



A ' 



ORMAT (1H1,UX.'F ELLER-BUNC 
ION AND COST summa 
/#1H,'I. A. ■ »F8,2,5X# 'HOURS WORKE 

,' INCLUDING'//16X.F8.2. 
§X <minuTES OF PERSONAL BREAKS, • , //16X.F8.2. 

5X, 'HINUTES MECHANICAL DELAYi ',/ / 16ft, fA,2, 
5X, 'mInuTES NON-meCHANICAL DELAY 1 , 
' L<*L ' BVtlS, 5 X, 'FELLER-BUNCHER. ',5x. 
'TRACKE6 T?PE-ROTaTaBLE BOOM WITH ACCUMULATOR SHEAR'///) 

ORMAT (l'Hl, flX.'F ELLER-BUNCHER PRODUCT*, 

ION AND COST SUMMAR Y',//,1H . 1 32 ('*'), //// 
/#1H,'I, A. »#F8,2#5X, 'HOURS WORKED BY FELLER-BUNCHER ' 
,' INCLUDING'//16X,F8 1 2 1 
5X. 'MINUTES OF PERSONAL BREAKS, ' , // 1 6X, F8 . 2, 

5X, 'MTNUTFS MECHANICAL DEL*Y. './ / 16X, F8 , 2 , 
5X, 'MINUTES NON-MECHANICAL DELAY', 
/,7X, »B. ' . 18,5 X, ' FELLER-BUNCHER, ' ,5x, 
RUBBER-TIRED TYPE.'///) » 

ORMATC1M. »H, A, ' ,I8,5X, 'TOTAL ACCUMULATOR CYCLES COMPLETED', 
- ALL MACHINES. '!//,16X.F8. 2, 5X, 'AVERAGE NUMBER OF TREES PER 1 , 
CYCL*, ',//,16X,FB, 2, 5X, 'AVERAGE VOLUME PER CYCLE - CU.FT,',/) 

ORMAT (1H.6X, 'B. ',I8,5X, 'BUNCHES COMPLETED',//, 

6X, F8, 2, 5X, * AVERAGE NUMBER OF TREES PER BUNCH,',/) 

ORMATC1H, 15X,F8, 2, 5X, 'AVERAGE NUMBER OF CYCLES PER BUNCH,',/) 

ORMATdH- 15X,F8-2,5X, • AVERAGE VOLUME PER BUNCH • CU.FT,',//, 
X.'C, i, IB, 5X, 'TOTAL TREES FELLED, ', //,7X, ' D. ', F8, 1 ,5X, 
TOTAL VOLUME FELLED - CU . FT ,',//, 7ft, ' E .' , 
8.2.5X, 'TOTAi TONS FELLED', //#7X,'F,', 18, 5X, 
•TOTAL DISTANCE TRAVELED - FEET',///) 

ORMAT(l H # 'III. A, ' ,F8,2,5X, 'AVERAGE NUMBER OF CYCLES PER', 
MACHINE HOUR,',/) 

0"MAT(1H,6X, 'B, '.F8, 2, 5X. 'AVERAGE NUMBER OF BUNCHES COMPLETED', 
PER MACHINE HOUR. ' ,//,7x, 'C, ',F8. 2, 5X, 'AVERAGE NUMBER OF', 
TREES FELLED PER MACHINE HOUR. • , //, 7X, ' D, T , F8.2, 5X, T a vERAGE ' , 
FELLED PER MACHINE HOUR. ' ,//,7x, 'E ,', F8.2.5X, • AVERAGE ' , 
"ELLED PER MACHINE HOUR. ' , //, 7x, • F ' , F8,2,5X, 
TRAVELED PER MACHINE HOUR - FEET,', ///J 



VOLUME 

NUMBER OF TONS F 
AVERAGE DISTANCE 



BX 

X 
X 
X 
X 
X 

I 



ORMATflH, ' IV. A. ' ,F8,2,5X, 'TOTAL GALLONS OF FUEL USED B*», 

THE F^ LER-SUNCHER i,//,16X,Fe.2,5X.TgALL0NS PER HOUR.' t //»16 
,F8, 2, Sx, 'GALLONS PER TON. ',//, 1 6X, F8 ,2, 5X, ' GALLONS PER CUNIT.' 



10 



I 



c 

N 

NUMFB 

F 

FELL 



3 s IX 3 
a s I x fa 

5 » ixh: 
■ !♦ i 

\l \ lfi,2oj30)7l l 



ONTINUE 
U M FB » 1 
s NUMB 
BTYP a X 

er-bunchEr TYPE 



OF FELLER-BUNCHERS 
s DROTT 1 



s RUBBER-TIRED TYPE 



22 



C BREAK AND DELAY TIMES IN ^IMJTES 
BREAK=X3/100 
DELAYsXa/IOO 
DELNMsX5/100 
RETURN 
20 CONTINUE 
NTPEE = XI 



FBVOL « X2/100, 
FBVOL = VOL. FELLED ALL MACHINES 

FBCNT = F&VQL/100. 
FBCNT FELLER-BUNCHER VOL, IN CUNITS, 

BNCOM s X3 
BNCO* = BUNCHES COMPLETED 

FBHR s XU/100, 

FBDIS = X5 
FBDIS s DIST, TRAV, ALL MACHINES 

RETURN 
30 CONTINUE 

PHArr ~ xi 

FBACC NUMBER OF ACCUMULATOR COMPLETED CYCLES. 

XDEN s X2 
POUNDS OF WOOD PER CUBIC FOOT, 

FBTNS = X3*XDEN/200000 



FBGPH s Xa/10 

R-BUNCMEP TOTAL GAI 



:LLER-BUNCMEP TOTAL GALLONS USED 
LLER-BUNCHER VOL, IN TONS 

Ff FBTYP.FQ.O ) WR I TF (6, 1 ) FBHR, BREAK , DELAY , DELNM, NUMFB 
F( FBTYP.EQ.l 5 WR I TE (6, d) FBHR, BRE AK, DEL A Y, DELNM, NUMFb 



IF 

I 



FBTPC = NTREEM, /FBACC 
FBTPC TREES PER CYCLE 



FBVPC = FBVOL/FBACC 

PC VOLUME PER CYCLE 

WRITE (6, 3) FBACC, FBTPC, FBVPC 



FBTPB = NTREEM./BNCOM 
FBTPB = AVE. TREES P£R COMPLETED BUNCH, 
WRITE(6,a)BNC0M, FBTPB 

PBCP8 » FBACC*1 ,/BNCOM 
WRITE(6,5)FBCPB 

FBVPB = FBVOL/BNCOM 
FBVPB = AVE. VOL. PER BUNCH 

WRITE (6 ,6) FBVPB, NTREE, FBVOL, FBTNS, FBDIS 

CYCHR = FBACC/FBHR 
WPITE(6,7)CYCHR 

FBBMH = BNCOM/FBHR 
FBRHH = AVE, BUNCHES PER MACHINE HOUR 

FBTMH = NTREE/FBHR 
FBTMH = TREES FELLED PER MACHINE HOUR 

FBVMH = FBVOL/ FBHR 
FBVMH = VOL. FELLED e ER MACHINE HOUR 

FBATH = FBTNS/FBHR 
FBATH AVERAGE TONS PER HOUR 

FBDMH r FBDIS/FBHR 
FBDMH s DIST. TRAV. PER MACHINE HOUR 

WRITE (6 ,8) FBBMH, FBTMH, FBVMH, FBATH, FBDMH 

FELLER-BUNCHER GALLONS/HOUR 
FBGPT = FBGPH/FBATH 

FELLER-BUNCHER GALLONS PER TON 

FBGPC = FBGPH/CFBVMH/100) 
FELLER-BUNCHER GALLONS PER CUNIT 

FB T GU = FBGPH*FBHR 

WRITE (fe, 11 ) FBTGU, FBGPH, FBGPT, FBGPC 

RETURN 

END 



23 



APPENDIX C 
COMPLETE FLOW CHARTS 




FELLER BUNCHER SEGMENT FLOW I MAPI 
(Pa-ie 1 of 4 ) 



CREATE 

1 
FELLER BUNOIER 



r 7^ im ' 

l SAVEVALUE ) STRI 

V_ y cnuN 



INITIAL IZE 
Al r . HEAD LOAD 
LIMIT, BUNCH 
SIZE. DISTANCE 



NITIALIZE I ■ 



)-- MAPI 
l) CYCLE 



QUE 
FBTRV 



- 6 



ADVANCE 
VSFBTRV 



DEPART 
FBTRV 



TRAVEL 

TCI HE* I 
LOCATION 



R RECORD 
TRAVEL 
TIMES 



( SAVEVALUE ) 



RECORD 

TRAVEL DISTANCE 



(ICi.VSREACH "\ 



C9.VSFBMOV ^\ 
3.VSNEXTY J 

T" 



NUMBER OF 
TREES IN F.B.'s 
REACH. 
CURRENT LOCATION 



DISTANCE F.B. 

HILL HAVE TO 

MOVE. 

NEXT LOCATION 



l 0rIC __ F.B. HAS TREES 
PES ' [T IN REACH 

FBMOV — ' ( D0ES N0T HAVE 

TO MOVE) 



YES (DROP) 

F.B. DROPS TREE 

IT TARRIED WITH. 




(TREE) 



.B. READY TO 
SHEAR TREES 



FELLER BUNCHER SEGMENT FLOW CHART 
(Paqe 2 of 4 ) 



TREE) L_ 



B.VSTDIAM 1 



ASSIGN 
TREE » 
AND 
DIAMETER 



QUEUE 
FBSHR 



T 



3 



ADVANCE 
VSFBSHR 



SHEAR 
TREE 



DEPART 
FBSHP 



Q 



RECORD 

SHEAR 

IMES 



( SAVEVALUE ) ^S 

N r^ FUNCT 

DBH.PE 

T 



ADD TO 

TOTAL DIAMETER 

AND NUMBER 

OF TREES 

IN ACC. HEAD 



E IN 
ON 



/l2+,VSTVOLS 



ADD TREE 
VOLUME TO 
HEAD 

DECREASE BY 1 
NUMBER OF TREES 
STILL IN REACH 




24 



FELLER BUNCHE R SI GMI NT FLOW i MALI 
(Paqe i of 4) 



(DROP) 



QUEUE 

DROP 



ADVANi L 
VSDROP 



DEPART 
DROP 



b 



Q 



( SAVEVALUE J 



BUNCH 
DATA 



QUEUE 
MCDLY 



iDV INI I 
VSMCDLY 



DEPART 
MCDLV 



SUING AND 
DROP 

ACCUMULATOR 
LOAD 



DROP TIMES 



RECORD BUNCH 
LOCATION, 
ADD VOL, SUM 01 
DIA, AND NUMBER 
OF TREES TO 
ACC. HEAD 



13 



MECHANICAL 

DELAY - IF 
ANY 



r ^ MEC 
0( DEL 



HANICAL 
DELAY 




II i 1 

N'lUI Y 


6 


' 




ADVANCI 
VSNMDLY 




y 




NMDLY 


3 


' 


1 


-N 


) 


( SAVEVALUE 


F.B. 




1 




TABULATE 

II 


^ 


" 




MARK 


N 




" 






f INITIALIZ 


^ 





HON IIEl.HANIl AL 
DELAY - 
II ANY 



N N " : HANICAL 



TOTAL VOLUME 
TOTAL TREES 
TOTAL TIME 



INITIALIZE 



SET - 







. •« 




1 Nil (if 
) ' If IP, 
/ Plf" 


until 




1 











' N 



AVEVALI1E I DUMBER 
J 



H 



' 



HOP! UNTIL 
riMI FOP 

IDIIII IX! A! 



□ . 



/SBREAl 



. 



□ ' 



■ N 
ill 





i mill BPfA 






lllil'l UNTIL 

N 



I 1 FOR 



u 



UNTIL 



) 
J 




HON ri'iiiiui ". T I v J 
on 



ppniiui i 



25 



Winsauer, Sharon A. 

1980. A program and documentation for simulation of a tracked feller/ 
buncher. U.S. Department of Agriculture Forest Service, Research 
Paper NC-192, 26 p. U.S. Department of Agriculture Forest Service, 
North Central Forest Experiment Station, St. Paul, Minnesota. 

Presents a computer model written in GPSS (General Purpose 
Simulation System) designed to simulate and study the productivity 
and operation of a tracked feller/buncher. 

KEY WORDS: GPSS (General Purpose Simulation System), com- 
puter, harvesting, productivity, modeling. 



f o , J 



V V. / 



;da forest service 
;esearch paper nc-193 







. 



- 

- . - 



The effect of initiaTrftimber of trees per acre 
and thinning densities on timber yields from 
RED PINE PLANTATIONS in the Lake States 




North Central Forest Experiment Station 
Forest Service, U.S. Department of Agriculture 



Allen L. 



U.S. Department of Agriculture 
1992 Folwell Avenue 
St. Paul, Minnesota 55108 
roved for publication January 8, 1980 



Contents 

Page 

Summary 1 

Volume yields 1 

Tree and stand characteristics 4 

Plantation management regimes 5 

Growth and yield model 6 

Volume yield response to stand density 7 

Total cubic-foot volume 8 

Merchantable cubic-foot volume 9 

Board-foot volume 10 

Timber production in relation to site 12 

Effect of stand density on tree and stand characteristics 12 

Tree size and form 12 

Timber quality 14 

Crown width and site occupancy 16 

Stand characteristics 17 

Harvest factors 18 

Risks in density choices 20 

Conclusion 23 

Literature cited 24 



THE EFFECT OF INITIAL NUMBER OF TREES PER 
ACRE AND THINNING DENSITIES ON TIMBER 
YIELDS FROM RED PINE PLANTATIONS IN THE 

LAKE STATES 



Allen L. Lundgren, 

Principal Economist 



In red pine (Pinus resinosa Ait.) stands the initial 
density ( number of established trees per acre ), timing 
of thinnings, and stand density left after thinning 
(basal area per acre) greatly affect tree diameter 
growth, and thus the salable timber output. Planting 
fewer trees reduces establishment costs and acceler- 
ates diameter growth of individual trees, resulting in 
larger, but fewer, trees at any given age. Thinning to 
low densities also results in fewer but larger trees. 
But the effects of initial density and subsequent 
thinning on the quantity and quality of timber prod- 
uct yields are not immediately obvious. A careful 
evaluation of stand density alternatives is needed to 
determine the best course of action to meet manage- 
ment goals. 

A wide range of initial tree spacings have been 
recommended for red pine (Evert 1973). Generally, 
recommendations based primarily on physical yield 
considerations tend to favor a higher initial density 
than those based on economic considerations. Warn- 
bach (1967), in his analysis of initial spacing in red 
pine plantations, concluded that a wide spacing, with 
400 established trees per acre, is economically prefer- 
able to a higher number of trees per unit area, and 
suggested that under some conditions even fewer 
trees may be more desirable. Wambach's findings 
were not published, and his suggestion has not been 
followed by an evaluation of lower initial densities. It 
has become apparent that an analysis of such low 
initial densities is needed. 

This paper summarizes the effects of stand density 
on tree and stand characteristics and timber yields in 
red pine plantations in the Lake States region of the 
United States for a wide range of initial numbers of 
trees and subsequent thinning densities. Simula- 
tions of growth and yield in red pine plantations were 
made using REDPINE, an unpublished growth and 
yield computer program developed by the author 
from growth models by Buckman (1962) and Warn- 
bach (1967), along with other data. 



The initial densities analyzed ranged from 50 to 
1,600 established trees per acre; the thinning densi- 
ties ranged from 60 to 180 square feet of basal area 
per acre left after thinning. This information should 
help forest landowners and managers who must 
decide which initial stocking level and subsequent 
program of thinnings will best meet their objectives 
in growing red pine for timber. 



SUMMARY 
Volume Yields 

Several graphs have been prepared to summarize 
and better illustrate the effects of initial density, 
thinning density, and site index on volume yields 
from red pine plantations. These graphs display 
contour lines of equal annual volume production for a 
range of initial and residual thinning densities for 
each site index. They show major differences in 
production over the range of sites, initial densities, 
and thinning densities considered. 

Total cubic-foot volume production (mean annual 
increment) drops rapidly on all sites with fewer than 
200 established trees per acre (figs. 1-3). As initial 
density increases above 200 trees per acre, produc- 
tion rises gradually. As initial densities approach 
1,200 to 1,600 trees per acre, production levels off, 
rising very little with additional trees per acre. 
Production is highest with a thinning density of from 
120 square feet on site index 50 to 140 square feet on 
site index 70. Production falls off for thinning densi- 
ties lower or higher than these. 

Merchantable cubic-foot volume production dis- 
plays a different pattern (figs. 4-6). Again, a deep 
trough of low merchantable volume production is 
noticeable with fewer than 200 trees per acre, but a 
peak of maximum production is evident at about 800 
trees and 120 square feet of basal area on site index 



20 




cu. It. /acre/year 



400 600 800 1.000 1,200 

INITIAL DENSITY (trees/acre) 



Figure 1. — The effect of initial and thinning densities 
on maximum total cubic-foot mean annual incre- 
ment in red pine plantations on site index 50. 




100 110 120 130 140 150 



cu. II. /acre/year 



400 600 800 1,000 1.200 1.400 1,600 

INITIAL DENSITY (trees/acre) 



Figure 3. — The effect of initial and thinning densities 
on maximum total cubic-foot mean annual incre- 
ment in red pine plantations on site index 70. 




200 400 600 800 1,000 1,200 1,400 1.6C 

INITIAL DENSITY (trees acre) 

Figure 2. — The effect of initial and thinning densities 
on maximum total cubic-foot mean annual incre- 
ment in red pine plantations on site index 60. 




400 600 800 1,000 1,200 

INITIAL DENSITY (Irees/acre) 



Figure 4. — The effect of initial and thinning densitiet 
on maximum merchantable cubic-foot mean an 
nual increment in red pine plantations on site index 
50. 




400 600 800 1.000 1.200 

INITIAL DENSITY (Irees/acre) 



Figure 5. — The effect of initial and thinning densities 
on maximum merchantable cubic-foot mean an- 
nual increment in red pine plantations on site index 
60. 



50, and 1,000 trees and 140 square feet on site index 
70. Production drops off at thinning densities above 
and below these levels. 

Board-foot volume production displays an even 
more noticeable peak in production (figs. 7-9). Once 
again, there is a distinct depression in production 
below 200 trees per acre. A sharp peak in production 
occurs with 200 trees per acre on all sites in stands 
thinned to 120 square feet of basal area on site index 
50, and to 140 square feet on site index 70. Production 
falls off as numbers of trees per acre increase beyond 
200. The choice of thinning density becomes critical 
at higher numbers of trees per acre, with rapid 
decreases in production resulting from thinning 
either too much or too little. 

Users of these results should keep in mind that the 
term "initial density" refers to the number of trees 
per acre that became established following early 
mortality the first few years after planting, and 
assumes reasonably even spacing of the established 
trees. 




400 600 800 1.000 1.200 

INITIAL DENSITY (Irees/acre) 



1.400 1.600 



Figure 6. — The effect of initial and thinning densities 
on maximum merchantable cubic-foot mean an- 
nual increment in red pine plantations on site index 
70. 




400 600 800 1.000 1,200 

INITIAL DENSITY (Irees/acre) 



1.400 



Figure 7. — The effect of initial and thinning densities 
on maximum board-foot mean annual increment in 
red pine plantations on site index 50. 




400 600 800 1,000 1.200 

INITIAL DENSITY (trees/acre) 



Figure 8. — The effect of initial and thinning densities 
on maximum board-foot mean annual increment in 
red pine plantations on site index 60. 



Tree and Stand Characteristics 

Diameter growth is strongly affected by both the 
initial number of trees per acre and by the stand 
density left after thinning. Lower initial numbers of 
trees and lower densities result in fewer but larger 
trees at a given age, and the effect is great. One of the 
lowest initial and residual density combinations 
evaluated here (100 initial trees/acre thinned to 60 
square feet of basal area/acre ) produced trees 6 inches 
larger at age 40 and 14 inches larger at age 80 than 
the highest combination (1,600 initial trees/acre 
thinned to 180 square feet). 

Wood produced with a low initial number of trees/ 
acre has a slightly lower specific gravity than wood 
produced with a high initial number of trees, varying 
on site index 60 from 0.34 for 100 trees/acre to 0.35 for 
1,600 trees per acre. The average diameter of dead 
branches at a height of 17 feet was larger for low 
initial densities, but the difference in branch size 
between 100 and 1,600 initial trees per acre was less 
than % inch. 

The initial number of trees per acre has a major 
impact on the size of trees harvested over a given 
rotation. With fewer initial trees, fewer trees are 




INITIAL DENSITY (Irees/acre) 



Figure 9. — The effect of initial and thinning densities 
on maximum board-foot mean annual increment in 
red pine plantations on site index 70. 






harvested over a rotation, but the trees harvested are 
larger in diameter. On site index 60 stands thinned 
regularly to 140 square feet, with only 200 estab- 
lished trees per acre, almost all trees harvested for 
timber over an 80-year rotation were larger than 12 
inches d.b.h. and they averaged about 16% inches. In 
contrast, almost all trees harvested in the same 
stands over the same rotation with 1,200 established 
trees were smaller than 12 inches, and averaged only 
7 3 /4 inches. 

For a given established number of trees per acre, 
leaving lower basal areas after thinning decreases 
the average size of trees harvested. Although indi- 
vidual trees grow more rapidly in diameter in stands 
thinned to low densities, more trees must be cut at an 
earlier age to reduce the stand to this lower density, 
and this reduces the average diameter of trees cut. In 
stands thinned to low basal area densities, more trees 
have to be harvested to get a thousand cubic feet of 
total volume than in stands thinned to high basal 
areas. However, the reverse is true for board-foot 
volume. Fewer trees have to be cut to get a thousand 
board feet from stands thinned to low basal areas. 
These findings, and the analyses from which they are 
derived, are explained more fully in the sections that 
follow. 



PLANTATION MANAGEMENT 
REGIMES 

Through choices of the initial spacing, the timing, 
kind, and intensity of thinning, and the final rotation 
age, a manager can greatly influence the kind, quan- 
tity, quality, and timing of salable products har- 
vested from a red pine plantation. An almost endless 
array of management alternatives exists. Only a 
limited number are considered here, but these en- 
compass a wide range of choices, making it possible to 
define regimes that can accomplish a variety of 
silvicultural and economic objectives. 

Initial density is defined as the number of trees 
successfully established about 5 years after planting; 
it ranges from 50 to 1,600 trees per acre (T/A) in this 
analysis, a range in spacing from 29.5 to 5.2 feet 
between trees. 



Trees 


Area 


Square 


er acre 


per tree 


spacing 




(sq. ft.) 


(ft.) 


50 


871 


29.5 


100 


436 


20.9 


150 


290 


17.0 


200 


218 


14.8 


300 


145 


12.0 


400 


109 


10.4 


500 


87 


9.3 


600 


73 


8.5 


800 


54 


7.4 


1,000 


44 


6.6 


1,200 


36 


6.0 


1,600 


27 


5.2 



Trees are considered to be 3-year-old seedlings at the 
time of planting, and spacing is assumed to be 
approximately square. The young red pine trees will 
be released from competing vegetation at 3 and 6 
years after planting, and at 15 years if necessary 
because of low initial density. 

Thinnings are to be made at not less than 10-year 
intervals. The first thinning is to be made 20 years 
after planting, but only if: (1) height to a 3-inch top 
d.i.b. is at least 17 feet; (2) the average stand d.b.h. is 
at least 5 inches; and (3) the potential cut is at least 
25 percent of the stand basal area per acre. If the 
plantation does not meet the above criteria at age 20, 
the thinning is postponed 5 years, and the stand is 
checked again. After a thinning, the stand must wait 
10 years and have a potential cut of at least 25 
percent of the stand basal area per acre to qualify for 
another thinning. If it does not, thinning is postponed 
for 5 years, at which time the stand is rechecked to 



see if it qualifies for a thinning. If it does, it is 
thinned; if not, it is rechecked each 5 years until it 
qualifies. This process continues throughout the life 
of the stand until final harvest. 

In this analysis a fixed residual basal area level 
after thinning is assigned to each stand. This basal 
area density ranges from 60 to 180 square feet per 
acre (SF/A), and at each thinning the stand is cut 
back to the assigned density. The management op- 
tion of varying residual densities throughout the 
rotation is not considered. Since basal area growth 
generally declines with age after the first thinning, 
the basal area level in each stand follows a declining 
saw-tooth pattern throughout the rotation (fig. 10). It 
is assumed that in all thinnings the proportion of 
trees cut is the same in all diameter classes, so that 
the diameter of the tree of average basal area does not 
change. At the end of the assigned rotation the stand 
is clearcut, and an identical management cycle is 
repeated. Rotation ages range from 20 to 150 years. 

This analysis focuses on only one of many kinds of 
thinning options: that of choosing a fixed level of 
residual basal area to which the stand is thinned at a 
fixed interval of 10 years, and thinning from above 
and below so that the tree of average basal area is not 
changed by thinning. Many other thinning options 
could be selected by the manager, including varying 
stand densities throughout the rotation, varying 
intervals between thinnings, and thinning from be- 
low or above. Although the findings reported in this 
study cover only a selected group of management 
practices, they do cover a broad range of typical 
recommendations and should help managers in se- 
lecting regimes for red pine plantations. 




STAND AGE (years) 



Figure 10. — Basal area in a red pine plantation with 
an initial density of 400 established established 
trees per acre on site index 60, thinned at 10 -year 
intervals to 120 square feet of basal area per acre. 



GROWTH AND YIELD MODEL 

A growth and yield model for even-aged red pine 
plantations and natural stands in the Lake States 
was developed by modifying, combining, and enlarg- 
ing the growth and yield models reported by Buck- 
man (1962) and Wambach (1967). Only a brief 
description of the growth and yield model is given 
here. Details will be published elsewhere. 

Wambach's (1967) reported basal area equation is 
a relatively simple function that assumes a constant 
basal area growth regardless of age for a given site 
and initial number of trees per acre. The original 
equation (corrected for a misprint of one of the 
coefficients, and using different symbols for the vari- 
ables) is: 

B = 32.278 + 0.001518 (N) (BHA) + 

0.6915 (I) (BHA) 
R 2 = 0.654 SE = 29.9ft 2 

where B = basal area in square feet per acre, 
N = trees per acre, 
BHA = breast-height age of the trees, and 
I = site index expressed as a 5-year 

intercept above breast height, in feet. 

A more flexible nonlinear function fit to Wam- 
bach's original data provided a considerable im- 
provement in fit: 



B = 6.565 (S) [1-e 



-04018A1 1.1677 



Ll-e 



001885N] 



R 2 = .751 SE = 25.0ft 2 

where S = site index (total height of dominants 
and codominants at 50 years), and 
A - age of the trees in years. 

This equation was used to estimate basal area 
development in stands up to age 25. In practice, basal 
area growth was determined each year by using the 
first derivative of the above equation with respect to 
age. From age 25 on, Buckman's basal area growth 
equation was used to project stand development: 

AB = 1.689 + 0.04107B-0.000163B 2 -0.07696A + 
0.0002274A 2 + 0.06441S, 

where AB = annual basal area growth in square feet 
per acre. 

Wambach's data were from unthinned, fully 
stocked plantations. When applied to young thinned 
stands, or to stands with low numbers of trees per acre, 
the diameter growths implied by the equation derived 
from his data appeared under some conditions to be too 



high. To overcome this deficiency, a nonlinear maxi- 
mum annual diameter growth constraint was im- 
posed on the basal area growth equations: 

ADmax = .007 S e -° 1BHA . 

This closely approximates the linear constraint 
used by Wambach (1967) to overcome this problem. 
At best this is only an approximation of the complex 
relation that determines the maximum potential 
diameter growth of a tree in the absence of competi- 
tion, but it does impose a needed constraint when 
applying the model to extremes of spacing. In prac- j 
tice, basal area growth estimated from the equations 
was reduced when necessary so that this maximum 
diameter growth was not exceeded. 






Starting with an initial number of established 
trees at a given age below 25 years on a specified site, 
the revised equation fit to Wambach's data projects 
stand basal area at future ages. Knowing the site, ; 
age, and basal area, Buckman's equation is then used 
to project basal area growth beyond 25 years. The 
diameter of the tree of average basal area (D) was 
computed from stand basal area and number of trees 
per acre. To determine the proportion of the stand in 
poletimber and sawtimber size classes, stands were 
assumed to have a standard deviation (SD) of tree 
diameters expressed by the following function fit to 
data reported by Stiell and Berry (1973): 

SD = .37628 D e - 093346 ° , 
where Demean stand diameter. 

Heights of dominant and codominant trees (H) 
were estimated using the nonlinear function re- 
ported by Lundgren and Dolid (1970): 



H- 1.890 S(l-e 



.01979AU 3892 



Total cubic-foot volumes (CF) in poletimber trees 
(from 5 to 9 inches d.b.h.), in sawtimber trees (9 
inches d.b.h. and larger), and in all trees were esti- 
mated using the ratio volume equation developed by 
Buckman (1961): 

CF = 0.4085 BH. 

Merchantable cords were computed as a constant 
proportion for the total cubic-foot volumes in trees 

greater than 5 inches d.b.h.: 

i I 

Cords = .0097 (CF, + ). 



A special function to estimate the board feet (Inter- 
national Winch) per cubic-foot (total) ratio in trees of 
specified diameter and height was developed from 
data in Gevorkiantz and Olsen (1955): 

BdFt/CF = - 8.76 + 1.985D - 0.07253D 2 + 
0.0008421D 3 + 0.04951H -.00892DH + 
0.0003169D 2 H - 0.000002786D 3 H. 

This ratio was applied only to the total cubic-foot 
volume in trees larger than 9 inches d.b.h. 

Mortality was approximated by a nonlinear func- 
tion that limited basal area per acre to a maximum 
accumulation depending upon site: 

Basal Area Mortality = 
B e - 20SB . 

The number of trees that died to make up this 
calculated mortality was estimated by assuming that 
the dead trees had an average diameter equal to the 
average stand diameter less one standard deviation. 

The final equations were used in a computer simu- 
lation program to forecast yields from various sites, 
initial densities, and thinning schedules. Test runs 
closely simulated red pine growth and yield reported 
in stands not used in developing the equations. 
Subsequent use and comparisons of projected versus 
observed growth and yield have reaffirmed the re- 
sults of these earlier tests. 



VOLUME YIELD RESPONSE TO 
STAND DENSITY 

The impacts of site, initial density, density after 
thinning, and rotation age on volume yields in red 
pine plantations are great. Red pine is a versatile tree 
and is sold for many products in the Lake States, 
including posts, pulpwood, poles, piling, and saw 
logs. Volume outputs can be measured in many ways, 
but this analysis will consider only three contrasting 
measures: total cubic feet per acre of the entire stem 
under bark, including stump and tip for all trees in 
the stand; merchantable cubic-foot volume per acre 
in trees 5 inches d.b.h. and larger to a 3-inch top 
diameter inside bark; and board feet per acre (Inter- 
national Winch rule) in trees 9 inches and larger 
above a 1-foot stump to a variable top diameter inside 
bark of not less than 6 inches. 



The total-cubic-foot measure reflects total utiliza- 
tion of wood in the central stem. It was determined 
using Buckman's (1961) stand volume ratio equa- 
tion, V = 0.4085BH, 
where V = volume per acre in cubic feet, 

B = basal area per acre in square feet, and 

H = mean total height in feet of dominant 

and codominant trees. 

Thus, cubic-foot volume is determined solely by the 

basal area and height of the stand; the diameters of 

individual trees play no part. 

The merchantable cubic-foot volume equation was 
derived by multiplying Buckman's (1961) cordwood 
equation, V = 0.003958BH, by 79 cubic feet per cord 
to give: 

Vm = 0.3127BH, 
where Vm = merchantable volume per acre in cubic 
feet. 

In effect, this ratio equation assumes a constant 
103 cubic feet of total peeled volume of the central 
stem per merchantable cord, for all merchantable 
trees. Because only trees 5 inches d.b.h. and larger 
are considered merchantable, tree diameters do af- 
fect this volume estimate. 

In contrast, the board-foot volume of a stand is 
determined by multiplying the total cubic volume by 
a board-foot recovery factor that is a nonlinear func- 
tion of average tree diameter and total height. The 
larger the tree diameter, the higher the board-foot 
volume per cubic foot of total volume. Thus, the 
board-foot volume measure is affected by basal area 
and height, but also is greatly affected by the diame- 
ters of trees in a stand. 

Comparing volume outputs over time for different 
sites, initial densities, and thinning schedules can be 
complex, because the kind, quantity, and quality of 
outputs vary over time. With the large number of 
alternatives considered here, some simplification is 
necessary. This analysis will ignore the timing of 
outputs and use mean annual increment 1 (MAI) over 
a rotation as the measure of volume output, ex- 
pressed as cubic feet per acre per year (CF/A/Y) and 
board feet per acre per year (BF/A/Y). Most of the 
examples will be for site index (SI) 60, which is an 
average site for red pine. Although the details are not 
included herein, the general conclusions regarding 
initial density and thinning density appear to hold 
for a broad range of sites. 

l The MAI used here was determined by summing 
the volume of all prior thinnings (if any), adding the 
potential volume available for harvest at a given age, 
and dividing this sum by the total age of the stand 
since establishment. 



Total Cubic-foot Volume 

The pattern of total cubic-foot MAI over a rotation 
is similar for all stand conditions. A rapid increase for 
the first 40 to 50 years is followed by a more gradual 
increase to a maximum MAI at about 60-100 years 
for most sites, initial numbers of trees per acre, and 
basal area densities left after thinning. After reach- 
ing a maximum, the MAI declines gradually through 
150 years. Site index (SI) has a pronounced effect on 
the level of MAI (fig. 11). For an initial density of 400 
established trees per acre (T/A) thinned to 120 square 
feet per acre (SF/A), for example, maximum MAI 
ranges from 91 CF/A/Y for SI 50 to 166 CF/A/Y for SI 
70, culminating at about 80 to 90 years. Shortening 
or lengthening the rotation age by 10 years has only a 
small effect on MAI, usually reducing it by less than 1 
percent. 

Increasing the initial density also increases cubic- 
foot MAI, but the effect is less pronounced, varying 
from 116 CF/A/Y for 200 T/A to 141 CF/A/Y for 1,200 
T/A when stands are thinned every 10 years to 120 
SF/A of basal area (fig. 12). Quadrupling stocking 
from 200 to 800 T/A increases total cubic-foot volume 
production per acre by only 17 percent over the 
rotation. 

Cubic-foot MAI increases with increasing basal 
area left after thinning, up to a maximum attained 
with 120-140 SF/A (fig. 13). But with 400 T/A initial 
density, leaving 120 rather than 80 SF/A (a 50- 
percent increase in residual stocking) increases 
cubic-foot volume production by less than 20 percent. 
With higher initial numbers of trees the percent 
increase in MAI is less. 

In general, the higher the initial and residual 
stand densities, the higher the total cubic-foot vol- 
ume production (table 1). It is important to note, 
however, that although total cubic-foot volume pro- 
duction increases as density increases, it does so at a 
declining rate. For example, the first 200 red pine 
trees per acre, thinned to 120 SF/A of basal area, 
produces 116 CF/A/Y; the second 200 trees (bringing 
the total to 400 T/A) adds only 12 CF/A/Y to this total 
MAI production; the third 200 trees increases the 
total MAI by only 5 CF/A/Y; and the fourth 200 trees 
adds only 4 CF/A/Y. This decreasing marginal prod- 
uctivity becomes especially important in 
determining optimum initial densities. 

The effect of initial density and density after 
thinning on MAI is more easily seen when it is 
graphed (fig. 14). Here it is readily apparent that the 



160 - 



140 



i 120 



80 



2 60 

Z 

UJ 40 

S 



SITE INDEX 


70 


/ 


60 


/ / 


50 





i i i i . i i i 



20 



80 



100 



STAND AGE (years) 

Figure 11. — The effect of site index on total cubic-foot 
mean annual increment over a rotation in red pine 
plantations with an initial density of 400 estab- 
lished trees per acre, thinned to 120 square feet of 
basal area per acre. 

160 







f 


S^~~^ ■~-^>-^1.200 


// 


,< \(i0&^. 


/ 1 / 


y^^ ~"~"~---^^___^ 400^- 


// / / 


200^ 


If If 


Initial Density 


11 


(trees/acre) 


1 1 1 1 1 


i i i i i i i i i 



u. 100 

2 
Ui 

S 

Lu 80 
0C 

o 

2 

^ 60 

2 
2 

2 40 

<t 

lu 

5 

20 



STAND AGE (years) 

Figure 12. — The effect of initial number of established 
trees per acre on total cubic-foot mean annual 
increment over a rotation in red pine plantations on 
site index 60, thinned to 120 square feet of basal 
area per acre. 



^ 100 



80 



60 - 



1 

2 40 

2 

2 

| 20 



THINNED TO: 
120 




80^--^^ 


2 
IU 




S 




IU 




C£ 




O 


60 — 


2 






Basal Area 


~i 


(sq. It. /acre) 


3 



20 40 60 80 100 120 140 

STAND AGE (years) 

Figure 13. — The effect of residual basal area left after 
thinning on total cubic-foot mean annual increment 
over a rotation in red pine plantations on site index 
60, with an initial density of 400 established trees 
per acre. 

Table 1. — Maximum mean annual increment in thinned red pine 
stands on site index 60, for selected initial numbers of 
trees and basal areas left after thinning 



Initial stand 


Basa 


I area 


left after thinning (sq. 


ft. /acre) 




density 


60 


80 


100 


120 


140 


160 


(trees/acre) 


TOTAL STEM VOLUME UNDER BARK (cu. 


ft. /acre/year) 


100 


64 


76 


87 


96 


99 


97 


150 


72 


87 


99 


110 


112 


107 


200 


78 


93 


107 


H6 


117 


113 


300 


86 


102 


115 


124 


123 


118 


400 


93 


108 


122 


128 


127 


121 


500 


98 


113 


125 


131 


130 


124 


600 


103 


117 


128 


133 


132 


126 


800 


108 


123 


133 


137 


136 


132 


1,000 


115 


127 


136 


139 


139 


134 


1,200 


119 


131 


138 


141 


141 


136 


1,600 


124 


134 


140 


144 


143 


139 


MERCHANTABLE CUBIC-FOOT VOLUME (cu. ft. /acre/year) 


100 


49 


58 


67 


73 


76 


74 


150 


55 


66 


76 


84 


86 


82 


200 


59 


71 


82 


89 


90 


87 


300 


66 


78 


88 


95 


95 


90 


400 


70 


82 


93 


98 


97 


92 


500 


73 


85 


96 


100 


100 


95 


600 


75 


87 


97 


101 


101 


96 


800 


78 


90 


99 


102 


103 


100 


1,000 


79 


90 


49 


102 


103 


100 


1,200 


78 


90 


98 


101 


102 


98 


1,600 


77 


88 


96 


99 


98 


95 




BOARD-FOOT VOLUME 1 (bd. 


ft. /acre/year) 






100 


417 


508 


588 


656 


681 


649 


150 


439 


555 


647 


729 


736 


687 


200 


456 


569 


673 


743 


741 


694 


300 


446 


562 


657 


718 


704 


645 


400 


437 


551 


643 


687 


656 


583 


500 


427 


535 


615 


659 


608 


526 


600 


421 


523 


612 


634 


567 


485 


800 


404 


508 


573 


593 


524 


424 


1,000 


393 


493 


551 


557 


479 


380 


1,200 


386 


467 


527 


521 


436 


334 


1,600 


367 


445 


485 


471 


375 


228 



140 


THINNED TO: — *" 




120^ 




^ _- — — 


120 


y^/vM ^ — """ - — — " — 




II / /*Q ^__— -^"^ 


100 






I / /^60 




/// / / Basal Area 


80 


//// / (sq It. /acre) 


60 




40 




?0 


i i i i i i i i i 



400 600 800 1,000 1.200 1,400 1,600 

INITIAL DENSITY (trees/acre) 



International %-inch rule. 



Figure 14. — The effect of initial and thinning densi- 
ties on maximum total cubic-foot mean annual 
increment in red pine plantations on site index 60. 



first 200 T/A achieve a relatively high MAI, espe- 
cially if the stand is thinned to 120-140 SF/A. The 
increase in MAI from additional trees after the first 
200 T/A is relatively minor. 

The manager can considerably influence the total 
cubic-foot output from a red pine plantation by the 
choice of initial number of trees and the basal area 
left after thinning. A combination of low initial 
density, say 200 established T/A, and a schedule of 
thinnings to low basal areas, say 60 SF/A, will result 
in an uninspiring 78 CF/A/Y. At the other extreme, a 
choice of high initial and thinning densities (for 
example, 1,200 T/A and 140 SF/A) could increase 
production to 141 CF/A/Y, a sizable gain. However, 
production from a plantation with low initial density 
can be increased considerably by choosing a high 
density after thinning. With 200 T/A, thinning to 140 
rather than 60 SF/A will increase production from 78 
to 117 CF/A/Y, a substantial increase. 



Merchantable Cubic-foot Volume 

The relations of merchantable cubic-foot MAI to 
rotation age, initial density, and thinning density are 
similar to the total cubic-foot volume relations, with 
a few significant differences. Merchantable cubic- 
foot MAI begins at a later stand age, but culminates 
at about the same age as total cubic-foot MAI, or 



perhaps only 5 years or so later in some instances. 
Rotation ages are similar over the range of densities 
considered here. Merchantable cubic-foot MAI is, of 
course, less than total cubic-foot MAI at each stand 
age. 

The highest MAI is produced by an initial density 
of 800-1,000 T/A (fig. 15). Increasing the initial 
density beyond 1,000 T/A reduces total stand produc- 
tion of merchantable wood over the rotation, in 
contrast to an increase in total cubic-foot volume 
production. However, the effect of initial density on 
merchantable wood production is small over a wide 
range of initial densities. Production varies by only 
about 4 percent from 400 to 1,600 T/A in stands 
thinned to 120-140 SF/A of basal area. 

For a given initial number of trees per acre the 
manager can influence merchantable volume yields 
by his choice of thinning densities, but over a wide 
range of basal area densities the effect is not large 
(fig. 16). With 800 T/A, volume yields vary only 4 
percent for thinning densities ranging from 100 to 
160 SF/A of basal area. Within this range, the choice 
of thinning density has only a slight effect on mer- 
chantable yields from red pine plantations, but the 
highest yields are produced from stands thinned to 
120-140 SF/A of basal area. 

As with total cubic-foot volume, the manager can 
substantially influence merchantable cubic-foot vol- 
ume yields from red pine plantations by his choice of 
initial and thinning densities (table 1). A low initial 
density and low basal area left after thinning — for 
example, 200 T/A thinned to 60 SF/A— would pro- 
duce only 59 CF/A/Y. In contrast, 800 T/A thinned to 
140 SF/A would produce 103 CF/A/Y, a 75-percent 
increase over the low-density choice. 



120 



S 100 - 



THINNED TO: 
120, 140 




1.400 1.600 



INITIAL DENSITY (trees/acre) 

Figure 15. — The effect of initial density on maximum 
merchantable cubic-foot mean annual increment 
for selected basal area densities left after thinning 
in red pine plantations on site index 60. 



10 




100 



120 



140 



160 



BASAL AREA LEFT AFTER THINNING (sq. ft./acre 

Figure 16. — The effect of basal area density left after 
thinning on maximum merchantable cubic-foot 
mean annual increment for selected initial numbers 
of established trees per acre. 

Board-Foot Volume 

A different picture emerges if we look at board-foot 
production in relation to stand density. First, al- 
though MAI curves for board feet show the same 
general pattern as those for cubic feet (a rapid initial 
increase followed by several decades of only slight 
change ), they start at a later date and culminate later 
than the cubic-foot curves (fig. 17). In general, the 
better the site, the earlier the initiation of board-foot 
production. 

But the big difference between cubic-foot and 
board-foot production in relation to stand density 
shows up when we examine MAI for a range of initial 
densities (fig. 18): whereas cubic-foot MAI increases 
as initial density increases from 200 to 1,200 T/A, 
board-foot MAI shows a reverse pattern, decreasing 
from the highest MAI with 200 T/A to the lowest with 
1,200 T/A. In addition, although the age at which 
cubic-foot MAI culminates is only slightly affected by 
initial density (fig. 12), the age at which board-foot 
MAI culminates is strongly affected by initial den- 
sity, particularly if the stand is thinned to higher 
densities (100 SF/A and above). For SI 60 stands 
thinned to 120 SF/A, board-foot MAI culminates at- 
age 100 with 200 T/A, compared with age 140 with 
1,200 T/A (fig. 18). 



700 



600 



2 

UJ 500 

s 

UJ 

<t 
U 



2 

2 300 

2 
UJ 

* ?nn 



100 



- 


SITE INDEX 


70 y ^^^^ 


- 




My' -^ 


- 




50^ ~ 


1 


y i i i 


i i i i i i I i 



20 40 60 80 100 

STAND AGE (years) 



140 



Figure 17. — The effect of site index on board-foot 
mean annual increment over a rotation in red pine 
plantations with an initial density of 400 estab- 
lished trees per acre, thinned to 120 square feet of 
basal area per acre. 





Initial Density 


200 






(trees/acre) 






700 




' 400 




600 




800^-- 

/ 1,200^-' 




500 




400 








300 








200 
100 


i i 1 1\ / \S i i i 


i i i i 


i i i 



120 



140 



STAND AGE (years) 



Figure 18. — The effect of initial number of established 
trees per acre on board-foot mean annual increment 
over a rotation in red pine plantations on site index 
60, th inned to 120 square feet of basal area per acre. 



700 



Stand density after thinning strongly affects the 
pattern of board-foot MAI over stand age. In general, 
the higher the basal area left after thinning (up to 
about 120 SF/A), the higher the production and the 
later the culmination of MAI. With 400 T/A, thinning 
a stand to 120 SF/A rather than 60 SF/A increases 
production from 437 to 687 BF/A/Y, a 57-percent 
increase (fig. 19). However, at the same time, the age 
maximizing MAI increases from 85 to 110 years. 

The effect of stand density on board-foot volume 
production (as measured by maximum board-foot 
MAI) is dramatically evident if MAI is plotted over 
initial density for a range of basal area densities left 
after thinning (fig. 20). On SI 60, the highest board- 
foot MAI's are obtained with 120 SF/A left after 
thinning for initial densities below 1,200 T/A, and 
with 100 SF/A for initial densities above 1,200 T/A. 
The most striking feature about these curves is the 
rapid increase in MAI from the first 100 T/A, followed 
by a further increase to a peak MAI with 200 T/A, 
followed by a decline in board-foot volume production 
over a rotation as initial stocking is increased beyond 



soo 



2 400 
UJ 

UJ 

oc 

O 300 

2 
~i 

3 

2 200 

2 
•z 

2 

2 100 

s 



THINNED TO: / 




120 


// 




140 
100 


if 




80 


1/^^ 




60 




Basal Area 






(sq. tt./acre) 




/ 


i i i i 


i 



20 40 60 80 100 

STAND AGE (years) 



140 



Figure 19. — The effect of residual basal area density 
left after thinning on board-foot mean annual in- 
crement over a rotation in red pine plantations on 
site index 60, with an initial density of 400 estab- 
lished trees per acre. 



11 



THINNED TO: 




400 600 800 1.000 1.200 

INITIAL DENSITY (trees/acre) 



1,400 1.600 



Figure 20. — The effect of initial number of established 
trees per acre on maximum board- foot mean annual 
increment in red pine plantations on site index 60, 
for selected basal area densities left after thinning. 



200 T/A. It is evident from this figure that the 
maximum board-foot volume production occurs with 
an initial density of about 200 T/A thinned to 120 
SF/A of basal area every 10 years. In such a stand the 
first thinning would be made at age 45 and the final 
harvest would be at age 100. 

The combined effect of initial number of trees and 
density after thinning is large. A choice of 1,600 T/A 
thinned to 60 SF/A produces less than 370 BF/A/Y. In 
contrast, choosing 200 T/A thinned to 120 SF/A 
doubles production, to about 740 BF/A/Y. Thus, 
through the choice of initial density and densities 
after thinning, the forest manager can greatly alter 
the production of board-foot volume from his stand. 
Once a stand is established with a given number of 
trees, and has grown for 20-40 years with no thinn- 
ings, the manager is somewhat restricted in the 
effects he can achieve with thinning prescriptions. 

For initial densities of 200 to 400 T/A and thinning 
densities of 80 to 160 SF/A, board-foot volume pro- 
duction is maximized at rotation ages ranging from 
80 to 110 years. For initial densities of 500 to 1,000 
T/A and the same range of thinning densities, pro- 
duction is usually maximized at rotation ages rang- 
ing from 105 to 135 years. Generally, the more 
established trees per acre, the longer the rotation 
needed to maximize board-foot MAI. 



Timber Production in 
Relation to Site 

To avoid presenting too much detail, most of the 
previous information was given only for SI 60. How- 
ever, analyses of timber production for a wide range 
of Si's (40-80) indicate that the same general conclu- 
sions hold for all sites. Total cubic-foot volume pro- 
duction is highest with a high number of trees per 
acre, but the first 200 T/A accounts for about 80 
percent of the MAI obtained with 1,200 T/A. Gener- 
ally, the highest cubic-foot MAI for any initial num- 
ber of trees is obtained by thinning to basal areas of 
about 120 SF/A on poor sites to 140 or more SF/A on 
good sites. Site quality has a substantial effect on 
volume production. An initial density of 800 T/A 
thinned to 120 SF/A on SI 50 produces 98 CF/A/Y 
over an 80-year rotation. The same number of trees 
per acre thinned to 140 SF/A on SI 70 will produce 
181 CF/A/Y over the same rotation, an increase of 85 
percent. In general, maximum MAI in merchantable 
cubic feet is obtained with slightly higher densities 
on better sites, ranging from 73 CF/A/Y with 800 
trees thinned to 120 SF/A on SI 50, to 137 CF/A/Y 
with 1,000 trees thinned to 140 SF/A on SI 70. 

Board-foot volume production is highest on all sites 
with an initial density of about 200 T/A. With 200 
T/A optimal basal area densities after thinning are 
lowest (about 120 SF/A) on poor sites. Better sites 
appear able to support higher basal area densities 
throughout a rotation. 



EFFECT OF STAND DENSITY 

ON TREE AND STAND 

CHARACTERISTICS 

Volume yields are greatly affected by initial num- 
ber of trees and thinning density. But stand density 
also affects other tree and stand characteristics im- 
portant in choosing a management regime for red 
pine. In the following discussion, all examples are for 
SI 60 unless otherwise stated. 



Tree Size and Form 

In the growth and yield simulation model, height 
growth of trees in the stand was assumed to follow 
published site index curves (Gevorkiantz 1957, 
Lundgren and Dolid 1970) regardless of stand den- 
sity, and to apply to all trees in the stand on a given 
site at a given age regardless of diameter (fig. 21). 



12 




10 20 30 40 50 60 70 80 90 100 110 120 130 140 150 
STAND AGE (years) 

Figure 21. — Average height of dominant and codomi- 
nant red pine trees at different ages. 

Although Stiell and Berry (1977) reported a stand 
density effect on height for a SI 80 red pine planta- 
tion, it amounted to only a 2-foot difference at age 20 
between extremes of density. Thus, the effect of stand 
density on height will be ignored in this analysis. 
Because heights apply to dominant and codominant 
trees, heights of intermediate and suppressed trees 
will be overestimated. The exact effect of this is not 
known, but is expected to be small in all but the 
densest stands. 

In unthinned stands, diameter growth (change in 
diameter of the tree of average basal area) on a given 
site is strongly affected by the initial stand density 
(fig. 22). With 1,200 T/A, for example, diameter 
growth appears to slow down when the stand is about 
11-13 years old, and falls below that in less dense 
stands. With 800 T/A, diameter growth falls behind 
at about 12-15 years; with 400 trees, at about 17-20 
years; and with 200 trees, at about 25 years. These 
densities correspond to basal areas of about 40 to 60 
SF/A, and mark the onset of competition. This decline 
in diameter growth occurs earlier on better sites and 
later on poorer sites. 

Average tree diameters in thinned stands are 
affected by both the initial number of trees and the 
basal area density left after thinning (Lundgren and 
Wambach 1963). For example, with 400 T/A, thin- 
ning every 10 years to 80 SF/A rather than 120 SF/A 
results in larger trees throughout the rotation (trees 
average 3 inches larger in d.b.h. by the end of an 80- 
year rotation) (fig. 23). Initial density has a similar 



i 

00 
Q 

C5 

DC 
UJ 

-3C 



16 



14 



12 



10 



2 



Initial Density 
(trees/acre) 




50 



STAND AGE (years) 



Figure 22. — Average stand diameters for different 
numbers of initial trees per acre in unthinned red 
pine plantations on site index 60. 



effect on tree diameters. Establishing 400 T/A rather 
than 800 results in larger trees throughout the 
rotation, by about 3 inches towards the end of an 80- 
year rotation when thinning to 120 SF/A (fig. 24). 
Establishing only 200 T/A rather than 800 produces 
trees averaging more than 6 inches larger at the end 
of an 80 year-rotation. In summary, lower initial 
numbers of trees and lower basal areas left after 
thinning both lead to increased tree diameters at a 
given age, and the effect is substantial. 

The bole form of a tree can influence the product 
output. For example, less taper in a tree of a given 
d.b.h. and height indicates not only more total cubic 
volume, but also more board-foot volume output. 



13 



24 



22 



20 



18 



16 



14 



12 



10 



2 - 



THINNED TO: 




Basal Area 
(sq. ft. /acre) 



20 



40 



60 



80 



100 



STAND AGE (years) 



Figure 23. — Diameter of the tree of average basal area 
in red pine plantations thinned to different basal 
area densities on site index 60, with an initial 
density of 400 established trees per acre. 



In estimating the product output from red pine 
stands for this analysis, no allowance was made for 
possible changes in tree form due to changes in age, 
site, or stand density. Wambach (1967), after study- 
ing form quotients 2 in Lake States' red pine planta- 
tions, reported: "No significant relationship was 
found between form quotient and either stocking, 
site, or age." He concluded that "...the experience 
gained in the course of this study lends support to the 
conclusions drawn by Berry, Eyre and Zehngraff, and 
others... that spacing does affect tree form, with close 
spacing leading to slightly better form. But the effect 
is small and temporary. The influence on yield and 
quality over the course of a full rotation will not be 
very important." Apparently, ignoring tree form will 
have little effect on the outcome of silvicultural and 
economic analyses. 



2 Form quotient is defined as the ratio between the 
diameter of a tree at one -half its total height and at 
breast height. 



14 



i 
cri 
d 

UJ 

O 

a: 

UJ 



22 
20 

18 

16 

14 

12 

10 

8 

6 

4 

2 





- 


Initial Density 
(trees/acre) 


100/ 
/20o/ 






400/ 


- 




' 800/ 


- 




/■Goo/" 


> 




Xl,600 


/ I I 


i i i i i i 


1 



20 



40 



60 



80 



100 



STAND AGE (years) 

Figure 24. — Diameter of the tree of average basal area 
in red pine plantations thinned to 120 square feet of 
basal area per acre on site index 60, for selected 
initial numbers of established trees per acre. 



Timber Quality 






Earlier we saw that the highest board-foot volume 
production per acre is achieved with a low initial 
density (about 200 T/A thinned to 140 SF/A). But 
what about the quality of wood produced at such wide 
initial spacing? Larson ( 1962 ) has shown that the size 
and distribution of the living crown affect the amount 
and quality of wood produced. Crown development is 
in turn governed to a large extent by stand structure. 
Larson has pointed out that stem wood produced 
within crowns of conifers has a higher proportion of 
juvenile wood (which is of low quality) than stem 
wood produced below crowns. Open-grown trees have 
larger, deeper crowns than more closely spaced trees 
where crown competition has begun. As a result, the 
quality of wood from open-grown trees is expected to 
be lower than that from trees in denser stands. Baker 
(1969) has documented this for red pine plantations 
for a range of spacings. 



Unfortunately, data were not available to accu- 
rately model crown size development in red pine 
stands in relation to site, age, initial density, and 
thinning schedules and to relate this in turn to wood 
quality. Instead, three indirect indicators of the 
impact of spacing on wood quality were used: diame- 
ter growth rate (expressed as rings per inch ), number 
and size of branches, and specific gravity. 

Previous studies have reported that conifer trees 
with more than 6 to 8 rings per inch (a d.b.h. growth 
rate of from Vb- to Winch per year) are acceptable for 
saw logs (Lundgren 1965). In stands thinned to 120 
SF/A, only at the earliest ages does tree growth 
exceed Vb-inch in diameter per year and produce fewer 
than six rings per inch. Diameter growth rates of six 
or more rings per inch are produced after stand age 15 
even in stands with only 200 T/A. Even with this low 
initial density, stands thinned to 120 SF/A produced 
at least eight rings per inch from age 40 on (table 2). 

Diameter growth rates on better sites are higher, 
prolonging the production period of wood with fewer 
than six rings per inch. But even on SI 80, with 200 
T/A thinned to 120 SF/A, diameter growth has slowed 
by stand age 30 to produce more than six rings per 
inch. Thus, it appears that except for a juvenile core 
of wood, red pine plantations with low initial densi- 
ties (200 T/A) will produce wood of acceptable quality 
for saw logs throughout most of a rotation. 

Table 2. — Diameter growth throughout a 100-year rotation 
of red pine plantations on site index 60, for a 
range of initial trees per acre thinned to 120 sq. ft. 
of basal area per acre 

(In inches) 



Stand 




Initial density (trees/acre) 




age 


D 1 


200 400 


800 


(years) 


AD 2 RPI 3 D AD RPI D 


AD RPI 



10 2.3 

20 5.6 

30 8.6 

40 11.2 

50 13.3 

60 15.5 

70 17.8 

80 19.8 

90 21.7 



0.33 6.1 

.30 6.7 

.26 7.7 

.21 9.5 

.22 9.1 

.23 8.7 

.20 10.0 

.19 10.5 

.16 12.5 



2.3 

5.3 
7.4 
8.9 
10.6 
12.4 
14.3 
16.0 
17.7 



0.30 6.7 
.21 9.5 
.15 13.3 
.17 11.8 
.18 11.1 
.19 10.5 
.17 11.8 
.17 11.8 
.18 11.1 



2.3 

4.8 

6 1 

7.4 

8.8 

10.2 

11.8 

13.3 

14.7 



0.25 8.0 
.13 15.4 
.13 15.4 
.14 14.3 
.14 14.3 
.16 12.5 
.15 13.3 
.14 14.3 
.16 12.5 



100 23.3 — — 19.5 — — 16.3 

'D.b.h. (outside bark). 

2 Average annual growth in d.b.h. over the decade. 

3 Rings per inch growth in d.b.h. 



The number and size of branches on a tree affect 
the number and size of knots in the saw logs from that 
tree, which in turn affect the grade of lumber output 
from the tree. Wambach ( 1967 ) studied the number of 
branches per whorl from breast height to 20 feet 
above the ground. He found that the average number 
of branches per whorl was related to site, increasing 
from 4.3 to 5.9 branches per whorl as SI increased 
from 40 to 70. However, he reported that initial stand 
densities from 400 to over 2,000 T/A had no signifi- 
cant effect on number of branches per whorl. 

If we accept Wambach's conclusion, the manager's 
choice of initial density on a given site has no effect on 
the number of branches per whorl. Since height 
growth is essentially unaffected over the range of 
stand densities considered here, the number of 
whorls for a given lineal length of stem (say the butt 
16-foot log) will be controlled by the site. A rough 
estimate indicates the number of whorls in a 16-foot 
butt log ranges from 16 on SI 40 to 9 on SI 70 (table 3). 
Thus, even though the average number of branches 
per whorl increases on better sites, the number of 
branches per foot of stem decreases because there are 
fewer whorls due to greater height growth. 

Wambach ( 1967) found that the average diameter 
of dead branches (measured 1 inch from the stem) 
was affected by initial stand density. Branch diame- 
ter increased as initial density decreased, as site 
quality increased, and as distance from the ground 
increased. But the differences in average branch 
diameter were not large. On SI 40, with 1,200 trees 
per acre, dead branch diameter at breast height 
averaged 0.5 inch (table 4). In contrast, on SI 70 with 



Table 3. — Estimated number of branches per butt log in 
relation to site index in red pine plantations in 
the Lake States 



Site 



Years to 

grow to Whorls per Branches 



index 17 feet 1 butt log 2 per whorl 3 



Branches 
Branches per foot 
per butt log of butt log 



40 
50 
60 
70 



21 

17 
15 
13 



16 

13 

11 

9 



4.3 
4.7 
5.2 
5.9 



69 
61 
57 
53 



4.3 
3.8 
3.6 
3.3 



'Computed from Lundgren and Dolid (1970). Based on total years from 
seed. 

determined by subtracting an estimated 5 years from seed to reach a 1- 
foot stump height on SI 40 and 4 years for SI 50-70. from total years to 
reach 17 feet. 
Computed from Wambach (1967). 



15 



Table 4. — Average diameter of dead branches in red pine 
plantations at selected heights in relation to site 
index and initial stand density 1 

(In inches) 





Average dead branch 


Average dead branch 




diameter at 4.5 feet 


diameter at 17.0 feet 






whorl height on 






whorl height on 


Initial 




site index: 






site 


index: 


trees 
















per acre 


40 


50 


60 


70 


40 


50 


60 70 


200 


0.7 


0.7 


0.8 


0.8 


1.0 


1.0 


1.1 1.1 


400 


.7 


.7 


.7 


.8 


.9 


1.0 


1.0 1.1 


600 


,6 


.7 


.7 


8 


.9 


.9 


.9 1.0 


800 


.6 


.6 


.7 


.7 


.8 


.9 


.9 1.0 


1,000 


.6 


.6 


.6 


.7 


.8 


.8 


.8 .9 


1,200 


.5 


.6 


.6 


.7 


.7 


.8 


.8 .9 



1 From Wambach (1967). 



200 T/A, at 17 feet (the top of the butt log), dead 
branch diameter averaged 1.1 inches. Although on a 
given site at a given distance up the stem initial 
density did affect dead branch diameter, the effect 
was not large. An increase in initial density of 1,000 
T/A would reduce average branch diameter only by 
about Va inch. 

If we accept Wambach's findings we can conclude 
that the choice of initial stand density has no effect on 
the number of branches (and thus the number of 
potential knots) in the butt log, and has only a slight 
effect on the average size of dead branches over the 
range of initial densities considered in this study. 
Laidly and Barse 3 reported that dead branch diame- 
ters in the butt log of trees from a SI 70 red pine 
plantation ranged from 0.7 inches at 5-foot spacing 
(1,742 T/A) to 1.2 inches at 11-foot spacing (360 T/A). 
This closely parallels Wambach's findings. Persson 
(1977) reported that the largest branches in young 
Scotch pine plantations in Sweden ranged from 0.7 
inch at 4,000 T/A to 1.1 inches at 454 T/A, findings 
similar to the red pine results. 

Wambach ( 1967) estimated specific gravity of trees 
in the red pine plantations he studied from measure- 
ments of large-diameter increment cores. He found 
that the average specific gravity of cores at breast 
height decreased with increasing site quality and 
increased with increasing initial density, although 



site had considerably more effect than stocking. An 
increase in SI from 40 to 70 reduced specific gravity 
from about 0.36 to 0.33, approximately 10 percent, 
while a change in initial density from 200 to 1,200 
T/A increased specific gravity by only about 2 percent 
(table 5). Wambach found that adding tree age as a 
variable did not improve the estimates, an unex- 
pected result that he thought might be due to the 
small range of ages sampled (from 14 to 49 years). 
Baker and Shottafer (1970) reported similar results 
in plantation red pine, where an increase in stand 
density from 440 T/A (10-foot spacing) to 1,210 T/A 
(6-foot spacing) resulted in an increase in specific 
gravity of only 2 percent. 

Although the specific gravity of an increment core 
is not necessarily the average specific gravity of the 
tree, if we accept Wambach's results as indicative of 
the relative differences in specific gravity due to site 
and initial density, we must conclude that on a given 
site the choice of initial density has only a small 
impact on the specific gravity of the wood produced. It 
is likely that over an entire rotation the effect of 
initial spacing on the specific gravity of the trees will 
be slight. This is in line with Larson's (1972) conclu- 
sion that wide spacing has relatively little effect on 
wood specific gravity in young pines. 



Crown Width and Site Occupancy 

We saw earlier that the first 200 T/A account for a 
large amount of the potential stand productivity. 
This point is so important to this analysis that it 
bears further explanation. In the following exercise 
we assume SI 60 with a rotation age of 80 years (the 
age maximizing MAI of total cubic-foot volume under 
bark of the entire stem for stands thinned to about 
120 SF/A). 



Table 5. — Specific gravity of increment cores at breast 
height in red pine plantations, in relation to site 
index and initial stand density 1 



^Personal communication, 1978. 



Initial 










trees 




Site index 




per acre 


40 


50 


60 


70 


200 


0.355 


0.347 


0.337 


0.323 


400 


.357 


.349 


.339 


.325 


600 


.358 


.350 


.340 


.326 


800 


.359 


.351 


.341 


.327 


1,000 


.361 


.353 


.343 


.329 


1,200 


.362 


.354 


.344 


.330 


1 From Wambach (1967). 









16 



Ek (1971) developed a relation between tree diam- 
eter and crown width based on 95 red pine trees on 76 
different sites in Michigan, Minnesota, and Wiscon- 
sin. If we assume that his equation holds for square- 
spaced red pine plantations, the crown width (CW) in 
feet just before the crowns touch and competition sets 
in can be predicted from tree d.b.h. in inches (D) by 
his equation: 

CW = 4.2334 + 1.4616D. 

Using the average stand diameters predicted over 
time by the red pine growth simulation model for the 
various initial spacings, and assuming all trees have 
the diameter of the tree of average basal area, it is 
possible to predict the age at which the crown width 
just equals the tree spacing, and the crowns just 
touch. At this age, assuming a full, circular crown 
and perfectly square spacing of equal-diameter trees, 
the crowns in a stand would occupy about 79 percent 
of the available area. Any growth beyond this point 
would result in overlapping crowns and increasing 
crown competition. 

Knowing the average stand diameter at each age, 
one can compute the area per acre occupied by tree 
crowns and thus determine the percent of the total 
area covered by crowns at each age until the crowns 
touch (fig. 25). For example, 400 T/A will average 4.1 
inches d.b.h. at stand age 15. A tree this size has an 
estimated crown width of 10.2 feet, or a crown area of 
82 square feet. With 400 trees per acre, crowns would 
occupy 32,860 SF/A, 75 percent of the total area. In 
such a stand crowns should begin touching about 1 
year later, when the stand is 16 years old and crown 
width equals tree spacing (10.4 feet). With 200 T/A, 
crowns would touch about 26 years after planting; 
with 800 trees, about 10 years after planting; and 
with 1,200 trees, about 8 years after planting. 

As we have seen (fig. 22), projected diameters in 
simulated stands with 200 T/A begin to slow in 
potential growth at about 25 years; with 400 T/A, at 
17-20 years; with 800 T/A, at 12-15 years; and with 
1,200 T/A, at 10-13 years. If we assume that competi- 
tion among trees begins about the time the crowns 
touch, the ages at which this occurs agree closely 
with our estimates of the onset of competition based 
on reduction in diameter growth. This comparison 
substantiates the suitability of the simulation model 
underlying these analyses. Newnham's (1966) ear- 
lier model, CW - 3.417 + 1.609D, based on a 15- 
year-old natural stand of red pine in Ontario, predicts 
almost identical stand ages for crown closure, further 
substantiating our results. 



Crowns Touch 




STAND AGE (years) 



Figure 25. — Estimated percentages of land area cov- 
ered by tree crowns in unthinned red pine planta- 
tions on site index 60, for a range of initial numbers 
of established trees per acre. 



The important thing to note is that on SI 60 a 
decline in potential diameter growth has set in as 
early as 17-20 years with only 400 T/A, indicating the 
onset of competition at this age. With only half that 
many trees, the onset of competition is delayed from 5 
to 10 years at most, until about stand age 25. If we use 
the age at which trees begin to compete with one 
another as an indicator of full site occupancy (only 
approximately), we see that even with only 200 T/A, 
the site would be fully occupied at age 25 for the 
remaining two-thirds of a 75-year rotation. If we 
double this initial density to 400 T/A, we get full site 
occupancy at 17-20 years, but with a 75-year rotation 
this increases full occupancy only from about 67 
percent to perhaps 77 percent of the rotation, at most. 
The first 200 trees are able to capture a large portion 
of potential site productivity over the rotation 
lengths considered here. This leaves only a relatively 
small amount of unused potential to be captured by 
increasing the initial density. 



Stand Characteristics 

Mortality in regularly thinned stands of red pine in 
the Lake States is relatively small after establish- 
ment, except in high-density stands where diameter 
growth rates are very low (Buchman 1979). In this 



17 



analysis very few trees died from competition. Essen- 
tially all changes in numbers of trees per acre after 
initial establishment were due to timber harvesting. 
On a given site, number of trees is controlled by the 
choice of initial density, thinning schedules, and 
rotation age (table 6). In general, following a pre- 
scribed thinning schedule throughout a rotation on a 
poor site results in leaving more trees per acre after 
thinnings than following the same schedule on a good 
site. 

Basal area growth in young plantations is affected 
by site index, tree age, and initial number of trees per 
acre, and at later ages by site index, tree age, and 
basal area density. Establishing more trees per acre 
allows faster accumulation of basal area per acre; for 
example, on SI 60, plantations reach 120 SF/A at 38 
years with 200 T/A, at 31 years with 400 T/A, at 23 
years with 800 T/A, and at 20 years with 1,200 T/A 
(fig. 26). The effect of increased initial stand density 
is to shorten the time to full occupancy, when the first 
thinning can be made. 

Buckman's (1962) basal area growth equation 4 
specifies that on a given site at a given age, all stands 
with a specified basal area will have the same basal 
area growth, regardless of number of trees or past 
stand history. According to our model, an unthinned 
plantation with 400 T/A will have about 85 SF/A at 
age 25; an unthinned plantation with 800 trees will 
have about 130 SF/A at this age (fig. 26). If, at age 25, 
the denser stand is thinned from 130 to 85 SF/A 
( removing about 35 percent of the basal area and thus 
about 35 percent of the trees), the two stands would 
be projected to grow the same amounts of basal area 
during the ensuing years. 



Harvest Factors 

Initial density and the timing and intensity of 
thinning can greatly affect harvesting operations, 
and thus indirectly affect the value of volume yields 
to the timber purchaser. For example, the number of 
trees per acre affects tree spacing, which can in turn 
affect future timber harvest options. Access may be 
difficult in closely spaced plantations, particularly 



Table 6. — Number of trees per acre in red pine plantations 
before thinning, for selected initial stand densities and 
basal areas left after thinning on site index 60 



J A B = 1.6889 + 0.041066B - 0.00016303B 2 - 
0.076958A + 0.00022741 A 2 + 0.06441 S, 
where: Afi = periodic net annual basal area 
increment, 
B - basal area in square feet per acre, 
A = tree age in years, and 
S = site index (50 -year base). 



Initial density (trees per acre) 






200 


1,200 




Stand 


60 sq. ft./ 120 sq. ft./ 


60 sq. ft./ 120 sq. ft./ 


age 


acre 


acre 


acre 


acre 


Years 













200 


200 


1,200 


1,200 


10 


200 


200 


1,200 


1,200 


20 


200 


200 


1,200 


1,200 


30 


200 


200 


1.190 


1,190 


40 


148 


200 


386 


769 


50 


80 


149 


206 


526 


60 


52 


108 


116 


373 


70 


38 


79 


73 


273 


80 


26 


79 


52 


186 


90 


26 


56 


35 


185 


100 


19 


56 


35 


133 


110 


19 


40 


25 


132 


120 


19 


39 


25 


93 


130 


14 


39 


25 


92 


140 


14 


29 


18 


69 


150 


14 


29 


18 


69 



for mechanized harvest operations. Assuming square 
spacing and full survival, an initial planting of 1,200 
T/A results in rows about 6 feet apart, with trees 
within rows 6 feet apart. By the time of the first 
thinning the crowns would be touching. Removal of a 
complete row would leave an access strip at most 6 
feet wide between crowns, 12 feet between stems. 
With only 200 T/A the rows would be spaced about 15 
feet apart, leaving a 15-foot access strip with removal 
of a row. Generally, establishing fewer trees per acre 
provides more working space for tree harvesting. 

Of course, spacing need not be square. For a given 
number of trees per acre, spacing between rows can 
be increased by spacing trees within rows closer 
together. With 1,200 trees per acre, rows could be 8 
feet apart with trees spaced 4.5 feet within rows. 
Although this spacing increases the width of the 
access strip, it leaves less space for tree felling within 
rows. 



18 



200 


Initial Oensity 
(trees/acre) 


2,400/l,200 


/800 


/400 


/200 


150 


120sq.lt. / 










100 






50 




25 years 








n 


£*£--— r i 


i i 


i 


i 





40 



60 



STAND AGE (years) 



Figure 26. — Basal area growth in unthinned red pine 
plantations on site index 60, for a range of initial 
numbers of established trees per acre. 



In stands with fewer trees per acre and higher 
residual basal area levels, fewer thinnings would be 
scheduled than in stands with more trees per acre 
and lower residual basal area levels. In stands with 
low initial numbers of trees to be thinned to high 
basal area levels, first thinnings generally are de- 
layed until age 40 to 50. With higher initial densities 
and lower basal area levels, first thinnings may be 
scheduled as early as 20 to 30 years. The timing and 
thus the number of thinnings can be controlled to 
some extent by the manager, but generally lower 
initial densities result in one or two fewer thinnings 
over a rotation than higher initial densities. Higher 
numbers of trees per acre may produce more cubic- 
foot volume over a given rotation, but that also may 
require more harvest entries, thus increasing timber 
sale and harvesting costs. 

Studies of how tree size affects harvest costs gener- 
ally have shown a sharp decline in costs as diameters 
increase immediately above the minimum merchan- 
table limit, followed by a more gradual decline as 
diameters increase further. For example, Manthy et 
al. (1967) recorded 4.8, 3.5, 2.8, 2.5, and 2.4 man- 
hours per cord to fell-limb-buck red pine trees in 
plantations for trees 5, 6, 7, 8, and 9 inches in d.b.h., 
respectively. Hannula ( 1971 ) showed a sharp decline 
in logging costs in pine and spruce stands in Canada 
from 5 to 7 inches, with a more gradual decline 
thereafter. 

The initial number of trees per acre has a major 
impact on the size of trees harvested over a given 
rotation age. In red pine plantations thinned to 100 



SF/A every 10 years over an 80-year rotation, the 
average diameter of all trees harvested varied from 
16 inches for plantations with 200 initial T/A to only 
7 inches for plantations with 1,200 T/A (table 7). 

The diameter distributions of trees cut over an 80- 
year rotation in stands thinned to 140 SF/A every 10 
years were estimated for extremes of initial stand 
density (fig. 27). 5 Of the 1,137 trees harvested from 
the stand with 1,200 trees initial density, 12 percent 
were less than 5 inches d.b.h. and thus were consid- 
ered nonmerchantable. More than 40 percent of all 
the trees cut were less than 7 inches d.b.h. Only 30 
percent of those cut were greater than 9 inches d.b.h., 
the minimum diameter for sawtimber, and only eight 
trees per acre were larger than 13 inches d.b.h. 

Table 7. — Average diameter of all trees harvested over 
an 80-year rotation in red pine stands on site index 
60, for a range of initial and thinning densities 
(In inches) 



Initial 














stand 


Basal 


area 


left after thinning (sq. 


ft. /acre) 


density 


60 


80 


100 


120 


140 


160 


Trees/acre 














200 


13 


15 


16 


16 


16 


16 


400 


10 


11 


12 


12 


12 


12 


600 


9 


10 


10 


10 


10 


10 


800 


8 


9 


9 


9 


9 


9 


1,000 


7 


8 


8 


8 


8 


8 


1,200 


7 


7 


7 


8 


8 


8 


1,600 


6 


6 


7 


7 


7 


7 







1,200 ss 


Initial Density 
\ (trees/acre) 










150 












a. 

•3 
















100 












a 
















50 






200 tree 







Figure 27. — Number of cut trees by 1-inch diameter 
classes in red pine plantations on site index 60 over 
an 80-year rotation, for initial densities of 200 and 
1200 trees per acre, thinned to 140 square feet of 
basal area per acre. 

5 'Assumes a normal distribution of trees around the 
average diameter of trees cut, with a standard devia- 
tion calculated from the formula given earlier. 



19 



In contrast, all of the 195 trees harvested from the 
stand with 200 trees initial density were greater than 
10 inches d.b.h., and more than half were larger than 
16 inches d.b.h. 

Densities left after thinning did affect the size of 
trees harvested, but the effect was not as great as the 
effect of initial density. Furthermore, the effect was 
the opposite of what would be expected intuitively. 
Since trees in stands managed with a low basal area 
density grow faster, one would expect the average 
size of trees harvested from these stands to be larger. 
But the reverse is true. 

With a given initial density, leaving lower basal 
areas after thinning tends to decrease the average 
size of trees harvested and produce less volume in 
larger trees. Individual trees grow more rapidly in 
diameter if a stand is kept at low density by thinning, 
but more trees are cut in the first thinning to reduce 
the stand to this level than would be cut if a higher 
density were maintained (table 8). This heavy first 
cut, which consists of the smallest trees that will be 
cut during the rotation, may remove up to half the 
trees in the stand. The effect of such early heavy 
thinnings is to reduce the number of trees available 
for harvest at a later date when they are larger. 

The initial number of trees and the thinning den- 
sity level interact to affect tree size and volume 
production from the stand. This interaction is re- 
flected in a measure that can affect timber harvest 
costs — the number of trees that must be cut to 
produce a unit of volume. 

If we establish 200 T/A, and mortality is light, then 
over the rotation we can expect to harvest almost 200 
trees. If we establish 1,200 T/A we can expect to 
harvest nearly 1,200 trees, or almost six times as 
many. We have already seen (fig. 14) that 1,200 T/A 
does not produce six times the cubic-foot volume, but 
only about 20 percent more than is produced by 200 
T/A. So, we should expect that many more trees 
would have to be harvested to get a unit of volume 
output from stands with high initial densities, and 
such is the case. In stands thinned to 120 SF/A every 
10 years, we would have to harvest about 20 trees 
over the rotation for every thousand cubic feet of total 
tree volume if initial density is only 200 trees (fig. 
28). We would have to harvest more than 100 trees for 
every thousand cubic feet if initial density is 1,200 
trees. The effect is even more pronounced for board- 
foot volumes (fig. 29). 



20 



Table 8. — Diameters and numbers of trees cut per acre over an 80- 
year rotation in red pine plantations on site index 60, for a range 
of initial and thinning densities 



Basal area left after thinning (sq. ft. /acre) 



Stand 



60 



100 



140 



age 


Ave. 
d.b.h. 


No. of Ave. No. of Ave. No. of 
trees cut d.b.h. trees cut d.b.h. trees cut 


200 TREES PER ACRE INITIAL DENSITY 


30 


8.6 


52 — — — — 



35 
40 
45 
50 
55 
60 
65 
70 
75 
80 
Entire rotation 



11.7 


67 


11.2 


53 


— 


— 


14.5 


28 


13.7 


49 








— 


— 


— 


— 


13.7 


60 


17.1 


14 


162 


28 


— 


— 



20.5 

21.5 
13.4 



11 

26 

198 



19.6 
20.6 

15.7 



— 16.3 

22 

48 
200 



18.1 
16.4 



38 

97 
195 



1,200 TREES PER ACRE INITIAL DENSITY 



30 
35 
40 
45 
50 
55 
60 
65 
70 
75 
80 
Entire rotation 



5.7 



380 



5.3 804 5.3 546 

7.3 180 6.7 229 

67 202 
9.7 90 8.1 137 — — 



12.3 



43 



98 



85 



170 



14.6 



21 



11.6 



54 



16.7 52 

6.2 1,190 



13.4 137 
74 1,188 



9.5 104 

10.1 283 

7.7 1,139 



RISKS IN DENSITY CHOICES 

In the preceding analyses we have assumed that 
future yields can be predicted with certainty, that the 
unexpected will not happen, and that there are nc 
risks to consider in choosing initial number of trees oi 
thinning densities. This section decribes some of tht 
risks that should be kept in mind when planning foi 
red pine plantation establishment and management 









THINNED TO 






- 


60 , 


/ / 




80," 




/ / 


120 




/ / // 


160 






Basal Area 


/ //y 


(sq 


ft/acre) 


iii 




i i i 



400 600 800 1.000 1.200 

INITIAL DENSITY (trees/acre) 



Figure 28. — Average number of trees cut per thousand 
cubic feet of total tree volume harvested from redpine 
plantations over an 80 -year rotation on site index 60, 
for a range of initial and thinning densities. 



The analyses have assumed that the forest man- 
ager or planner can accurately predict the number of 
established trees per acre. But in truth, the manager 
may have only limited control over how many trees 
become established in a plantation. The manager can 
usually control the selection of areas to be planted, 
methods and intensity of site preparation, condition 
of the planting stock, timing of and technique used in 
planting, spacing and number of trees planted, and 
subsequent follow-up treatments to improve sur- 
vival. Yet, at best one can only estimate survival 
rates, and thus guess at the number and spacing of 
trees that will result from the planting operation. 

Past experience may lead the manager to expect an 
average tree loss of 25 percent— or a 75-percent 
survival rate. On this basis, if the objective is to 
establish 600 T/A, 800 T/A would be planted. How- 
ever, this 75-percent survival is an average rate, and 
variations are likely to occur from year to year and 
from site to site. If conditions are ideal, all planted 
trees might survive, so the plantation will have 800 
established T/A instead of 600. Or, if conditions are 
bad, half or more of the planted trees might be lost, 
resulting in a stand with only 400 T/A or fewer. This 
variation in survival must be considered in choosing 
the number of trees to plant. How much is the 
potential loss from a wrong guess? 



80 



THINNED TO 




Basal Area 
(sq. It. /acre) 



600 



800 



1,000 1.200 1,400 1,600 
INITIAL DENSITY (trees/acre) 

Figure 29. — Average number of trees cut per thousand 
board feet of timber harvested from red pine planta- 
tions over an 80-year rotation on site index 60, for a 
range of initial and thinning densites. 



One approach to answering this question is based 
on a simple variant of game theory (Luce and Raiffa 
1957, Thompson 1968, Halter and Dean 1971). The 
planting of trees can be viewed as a game against 
nature. The number of trees to be planted is to be 
chosen by the forest manager from a range of trees 
per acre that could be planted. The percentages of 
planted trees surviving are the states of nature. 

We start by asking how many trees per acre would 
survive and become established for different num- 
bers of trees per acre planted, if we experience 
different survival rates in a SI 60 red pine stand. We 
can consider as many planting actions and states of 
nature as desired, but to keep this example simple we 
will use only three initial density levels and five 
survival rates (table 9A). The number of trees surviv- 
ing (S) for each initial density (N) and survival rate 
(R) is easily determined by using the formula: 

S = N x R. 

For each number of surviving trees (table 9A) we 
estimate (in this example, from table 1) the maxi- 
mum MAI we can expect from a stand with that many 
surviving trees (assuming that we thin to the basal 



21 



Table 9. — Strategy analysis for deciding how many 
trees per acre to plant for red pine plantations on site 
index 60 






Percent of 


Trees planted per 


acre 


trees surviving 


200 


300 


400 


A. NUMBER OF TREES SURVIVING (PER ACRE) 


100 


200 


300 


400 


75 


150 


225 


300 


50 


100 


150 


200 


25 


50 


75 


100 














B. EXPECTED MAI WITH OPTIMAL THINNING (BD. FT./ACRE/YEAR) 


100 


743 1 


718 


687 


75 


736 


739 


718 


50 


681 


736 


743 1 


25 


485 


580 


681 














C. LOSS IN MAI FROM MAXIMUM (BC 


I. FT./ACRE/YEAR) 


100 





25 


56 


75 


7 


4 


25 


50 


62 


7 





25 


258 


163 


62 





743 


743 


743 



'Maximum possible MAI. 



area density that gives the highest MAI for this 
number of surviving trees). For this example we will 
use board foot MAI. These board foot MAI's are 
recorded in table 9B. From table 1 we estimate that 
the highest board-foot MAI we can get from a red pine 
stand on SI 60 comes from a plantation with 200 
established T/A. We assume that this maximum MAI 
occurs at one of the initial densities listed in the 
table, and not a density falling between those listed 
(for SI 60, this maximum MAI is 743 BF/A/Y). The 
expected loss (table 9C) is the difference between the 
maximum MAI, 743 BF/A/Y, and the MAI expected 
from a given combination of trees planted per acre 
and percent of trees surviving (from table 9B). For 
example, if we planted 300 T/A and only 50 percent 
survived, we would get a MAI of 736 BF/A/Y (table 
9B), a loss of 7 BF/A/Y from the site's potential (743- 
736 = 7). 

This table of expected losses can be used to explore 
the consequences of different planting actions if we 
can assign probabilities of occurrence to each "state 
of nature", the percentage of trees surviving. For 
example, suppose past experience indicates that 60 



22 



percent of the time we can expect a 75-percent 
survival rate; 10 percent of the time a 100-percent 
survival rate; 20 percent of the time a 50-percent 
survival rate; and 10 percent of the time a 25-percent 
survival rate; and no complete failures. With these 
estimates we can calculate the probability of loss in 
MAI for each number of trees planted. 



In each "number of trees planted" column, the loss 
expected to occur for a given survival rate (table 9C) 
is multiplied by the corresponding probability that 
that rate will occur. For 200 trees planted per acre, if 
100 percent of the trees survive, we expect a loss of 
BF/A/Y. Multiplying this by a probability of 0.1 still 
leaves BF/A/Y as the expected loss. 



If 75 percent of the trees survive, the loss is 7 
BF/A/Y. Since we expect this to occur 60 percent of 
the time, the expected loss is 7 BF/A/Y x 0.6 = 4.2 
BF/A/Y. If only 50 percent of the trees survive, the 
expected loss is 62 BF/A/Y x 0.2 = 12.4 BF/A/Y. 
With 25-percent survival, the expected loss is 258 
BF/A/Y x 0.1 = 25.8 BF/A/Y. Finally, if no trees perl 
acre survive, the loss is 743 BF/A/Y, but since there is, 
no chance of this occurring (in our example) the 
expected loss is 743 BF/A/Y x O = O BF/A/Y. These 
expected losses for each probability of survival can be 
summed to get the total expected loss for each plant- 
ing action (table 10A). For this example, planting 20C 
T/A results in a total expected loss of 42.4 BF/A/Y! 
The loss expected from planting 300 T/A is 24. C 
BF/A/Y, and from 400 T/A, 26.8 BF/A/Y. To minimize 
expected losses in board-feet MAI per acre with thest 
probabilities of tree survival, one would choose 30C 
T/A from among the planting options considered hert 
(200, 300, or 400 T/A). 

The effects of other tree survival probabilities or 
the choice of planting density can be explored (tablet 
10B and IOC). A slight change in probability of £ 
given survival rate can change the expected loss by £ 
relatively large amount (table 10B), although in this 
example the density choice is not affected. A fairlj 
large shift in probabilities of survival may have littl 
or no effect on the manager's choice of the number o 
trees per acre to plant (table IOC). 



The exploration of risk examples given here use 
board-foot MAI, but similar analyses could be don< 
for other product yields. The simple calculations 
above do not tell the manager what to do, but thej 
help him explore the consequences of an array o 
actions and so may help him choose a defensible one 



Table 10. — Expected loss in mean annual increment 

for varying probabilities of tree survival 

(In bd. ft./acre/yr.) 



Percent of Probability of 


Trees planted per acre 


trees surviving occurrence 


200 


300 400 


A. FIRST PROBABILITY OF OCCURRENCE EXAMPLE 


100 0.10 





2.5 5.6 


75 .60 


4.2 


2.4 15.0 


50 .20 


12.4 


2.8 


25 .10 


25.8 


16.3 6.2 











Total expected loss 


42.4 


24.0 26.8 


B. SECOND PROBABILITY OF OCCURRENCE EXAMPLE 


100 .10 





2.5 5.6 


75 .60 


4.2 


2.4 15.0 


50 .30 


18.6 


2.1 


25 

















Total expected loss 


22.8 


7.0 20.6 


C. THIRD PROBABILITY OF OCCURRENCE EXAMPLE 


100 .333 





6.2 18.7 


75 .333 


2.3 


1.3 8.3 


50 .333 


20.7 


2.3 


25 

















Total expected loss 


23.0 


9.8 27.0 



CONCLUSION 

This analysis of product yields expected from red 
pine stands with various combinations of initial and 
thinning densities has led to several interesting 
conclusions. 

The initial number of trees per acre has a major 
impact on the amount and quality of product yields 
from red pine stands. The more trees established per 
acre, the higher the total cubic-foot volume yield. 
However, this increased volume must be harvested 
from more and smaller trees. Furthermore, the added 
production drops off sharply after the first 200 T/A. 
Merchantable cubic-foot yields (pulpwood) are max- 
imized with about 800 to 1,000 established T/A, but 
increasing the initial density above 500 established 
T/A increases production by less than 5 percent. 
Board-foot yields, which are strongly related to indi- 
vidual tree size, are maximized with relatively low 
levels of initial density — about 200 established T/A. 



With low initial densities it is necessary to main- 
tain stands at higher densities throughout the rota- 
tion to obtain highest board-foot yields; this can be 
accomplished by thinning stands to 120-140 SF/A 
every 10 years. These are higher than the 90 SF/A 
residual densities recommended by Benzie (1977) for 
poletimber stands, but close to his recommended 
thinning density for sawtimber stands. The higher 
the site index, the higher the basal area density 
required to achieve maximum yields. 

The red pine growth equations used in this analy- 
sis do not explicitly include the effect of shrub and 
other understory competition on tree growth, other 
than the undocumented effects of whatever condi- 
tions existed on the study plots used in developing the 
original growth models. Fear has been expressed that 
the yields reported here for low initial densities may 
not be achieved in practice because other trees, 
shrubs, and grasses may invade some sites and 
compete strongly with the widely spaced red pine 
trees. The yields reported here were developed from 
Buckman's and Wambach's growth models, but the 
200 T/A that produces maximum board-foot growth is 
just below the lower end of their data. Wambach's 
study of red pine plantations did include plots with as 
few as 270 T/A, and he developed diameter-growth 
constraints for low initial densities. Diameter growth 
projections by REDPINE, over a wide range of initial 
and thinning densities, were compared with those 
made by an updated version of the FREP Tree 
Growth Projection System (USDA Forest Service 
1979). At low initial densities (400 T/A and fewer) 
both systems projected almost identical diameters for 
the tree of average basal area during the first 30-40 
years, so the diameter-growth constraint equation in 
REDPINE that governs maximum diameter growth 
in early years at low densities appears reasonable. 
Nevertheless, to be conservative, one should assume 
that to achieve these growth rates, periodic tree 
release treatments to remove competing vegetation 
in low-density red pine stands will be required until 
the first thinning. 

The low initial densities indicated by this analysis 
for maximum board-foot volume production fall be- 
low the densities commonly used in establishing red 
pine plantations in the Lake States. For this reason 
these results should be considered tentative until 
more experience has confirmed their applicability. 
Still, all the available growth and yield evidence 
points towards establishing plantations with fewer 
trees per acre than has been common in the past. This 
will not only reduce initial costs, but increase future 
board-foot yields. 



23 






Our immediate need is for growth and yield studies 
to test these lower densities, and to document more 
accurately the development of low-density planta- 
tions throughout the region. But until this informa- 
tion becomes available, this analysis can serve as a 
guide to those wishing to manage red pine planta- 
tions with fewer trees per acre. 



LITERATURE CITED 

Baker, Gregory. 1969. Influence of tree spacing in a 
red pine plantation on certain wood and tree quali- 
ties. Agricultural Experiment Station Bulletin 
668, 10 p. University of Maine, Orono, Maine. 

Baker, Gregory, and James E. Shottafer. 1970. The 
effect of tree spacing in a red pine plantation on 
tree growth and wood quality. Agricultural Exper- 
iment Station Bulletin 685, 38 p. University of 
Maine, Orono, Maine. 

Benzie, John W. 1977. Manager's handbook for red 
pine in the North Central States. U.S. Department 
of Agriculture Forest Service, General Technical 
Report NC-33, 22 p. U.S. Department of Agricul- 
ture Forest Service, North Central Forest Experi- 
ment Station, St. Paul, Minnesota. 

Buchman, Roland G. 1979. Mortality functions, p. 47- 
55. In A generalized forest growth projection 
system applied to the Lake States region. U.S. 
Department of Agriculture Forest Service, Gen- 
eral Technical Report NC-49, 96 p. U.S. Depart- 
ment of Agriculture Forest Service, North Central 
Forest Experiment Station, St. Paul, Minnesota. 

Buckman, Robert E. 1961. Development and use of 
three stand volume equations for Minnesota. Jour- 
nal of Forestry 59(8):573-575. 

Buckman, Robert E. 1962. Growth and yield of red 
pine in Minnesota. U.S. Department of Agriculture 
Technical Bulletin 1272, 50 p. Washington, D.C. 

Ek, Alan R. 1971. Size-age relationships for open 
grown red pine. University of Wisconsin Forestry 
Reseach Note 156, 5 p. Madison, Wisconsin. 

Evert, F. 1973. Annotated bibliography on initial 
tree spacing. Canadian Forestry Service, Depart- 
ment of Environment, Forest Management Insti- 
tute Information Report FMR-X-50, 149 p. Ottawa, 
Ontario. 

Gervorkiantz, S. R. 1957. Site index curves for red 
pine in the Lake States. U.S. Department of Agri- 
culture Forest Service, Technical Note 484, 2p. 
U.S. Department of Agriculture Forest Service, 
Lake States Forest Experiment Station, St. Paul, 
Minnesota. 



24 



Gevorkiantz, S. R., and L. P. Olsen. 1955. Composite 
volume tables for timber and their application in 
the Lake States. U.S. Department of Agriculture 
Technical Bulletin 1104, 51 p. Washington, D.C. 

Halter, Albert N., and Gerald W. Dean. 1971. Deci- 
sions under uncertainty. 266 p. Southwestern Pub- 
lishing Co., Cincinnati, Ohio. 

Hannula, 0. 1971. The effect of average stand diame- 
ter on tree-length logging costs. Pulp and Paper 
Magazine of Canada 72(2):96-100. 

Larson, Philip R. 1962. A biological approach to wood 
quality. TAPPI 45(6):443-448. 

Larson, Philip R. 1972. Evaluating the quality of 
fast-grown coniferous wood. p. 146-152. Perma- 
nent Association Committee Proceedings, 1972. 
Western Forestry and Conservation Association, 
Portland, Oregon. 

Luce, R. Duncan, and Howard Raiffa. 1957. Games 
and decisions: introduction and critical survey. 509 
p. John Wiley and Sons, Inc. New York. 

Lundgren, Allen L. 1965. Thinning red pine for high 
investment returns. U.S. Department of Agricul- 
ture Forest Service, Research Paper LS-18, 20 p. 
U.S. Department of Agriculture Forest Service, 
Lake States Forest Experiment Station, St. Paul, 
Minnesota. 

Lundgren, Allen L., and William A. Dolid. 1970. 
Biological growth functions describe published site 
index curves for Lake States timber species. U.S. 
Department of Agriculture Forest Service, Re- 
search Paper NC-36, 9 p. U.S. Department of 
Agriculture Forest Service, North Central Forest 
Experiment Station, St. Paul, Minnesota. 

Lundgren, Allen L., and Robert F. Wambach. 1963. 
Diameters and numbers of trees in red pine stands 
are greatly affected by density, age, and site. U.S. ! 
Department of Agriculture Forest Service, Re- 
search Note LS-20, 4 p. U.S. Department of 
Agriculture Forest Service, Lake States Forest 
Experiment Station, St. Paul, Minnesota. 

Manthy, R. S., W. A. Lemmien, V. J. Rudolph, and 
L. M. James. 1967. A time study for plantation 
logging. Michigan State University Agriculture 
Experiment Station, Quarterly Bulletin 49(3): 
328-341. 

Newnham, R. M. 1966. Stand structure and diameter 
growth of individual trees in a young red pine 
stand. Project S-25 Progress Report. Forest Man- 
agement Research and Services Institute Internal 
Report FMR-1, 19 p. Department of Forestry, Ot- 
tawa, Ontario. 






Persson, Anders. 1977. Quality development in 
young spacing trials with Scots pine. Royal College 
of Forestry, Department of Forest Yield Research, 
Research Note 45, 152 p. Stockholm, Sweden. 

Stiell, W. M., and A. B. Berry. 1973. Yield of un- 
thinned red pine plantations at the Petawawa 
Experiment Station. Canadian Forestry Service, 
Department of Environment, Publication 1320, 16 
p. Ottawa, Ontario. 

Stiell, W. M., and A. B. Berry. 1977. A 20-year trial of 
red pine planted at seven spacings. Canadian For- 
estry Service, Department of Environment, Forest 
Management Institute, Information Report FMR- 
X-97, 25 p. Ottawa, Ontario. 



Thompson, Emmett F. 1968. The theory of decision 
under uncertainty and possible applications in 
forest management. Forest Science 14(2): 156-163. 

U.S. Department of Agriculture, Forest Service. 
1979. A generalized forest growth projection 
system applied to the Lake States region. U.S. 
Department of Agriculture Forest Service, Gen- 
eral Technical Report NC-49, 96 p. U.S. Depart- 
ment of Agriculture Forest Service, North Central 
Forest Experiment Station, St. Paul, Minnesota. 

Wambach, Robert F. 1967. A silvicultural and eco- 
nomic appraisal of initial spacing in red pine. 282 
p. University of Minnesota Ph.D. Thesis, Microfilm 
#67-14,665, University Microfilms Inc., Ann 
Arbor, Michigan. 



25 



Lundgren, Allen L. 

1981 . The effect of initial number of trees per acre and thinning densities 
on timber yields from red pine plantations in the Lake States. U.S. 
Department of Agriculture Forest Service, Research Paper NC-193, 
25 p. U.S. Department of Agriculture Forest Service, North Central 
Forest Experiment Station, St. Paul, Minnesota. 

Describes an analysis of initial density and subsequent thinning 
options for red pine (Pinus resinosa Ait.) plantations in the Lake 
States. Results showed that the initial number of established trees per 
acre has a major impact on the amount and quality of timber product 
yields, with 200 trees per acre (500/ha) thinned to 120 square feet of 
basal area per acre (27.5 m 2 /ha) maximizing board-foot volume yields. 

KEY WORDS: Volume yields, timber management, stand density, 
Pinus resinosa, timber quality. 



U.S. Government Printing Otfice: 1981 — 766-852/185 Region No. 6 



A FOREST SERVICE 
EARCH PAPER NC-194 





GOVT, 
DEPOSITORY !TE^' 

JUN x 1981 
CLEMSON 






Estimating northern 

RED OAK CROWN 
COMPONENT WEIGHTS 

in the northeastern 
United States 



by Robert M. Loomis and Richard W. Blank 



h Central Forest Experiment Station 

st Service, U.S. Department of Agriculture 



North Central Forest Experiment Station, 

Forest Service — U.S. Department of Agriculture 

1992 Folwell Avenue 

St. Paul, Minnesota 55108 

Manuscript approved for publication March 13, 1980 

1981 



ESTIMATING NORTHERN RED OAK CROWN 

COMPONENT WEIGHTS IN THE 

NORTHEASTERN UNITED STATES 



Robert M. Loomis, Fire Management Scientist 

and Richard W. Blank, Biological Technician 

East Lansing, Michigan 



Tree crowns, particularly those remaining as de- 
bris after logging operations, wind or ice storms, or 
insect epidemics, may be a significant forest fire fuel. 
Northern red oak, (Quercus rubra L.) widely distrib- 
uted in the eastern United States, is found in most 
oak-hickory forests. To appraise potential fire behav- 
ior following disturbance in stands containing north- 
ern red oak, a method of estimating individual crown 
fuel weights was developed. These weight estimates 
are suitable for use in fuel models and also provide 
estimates of potentially usable fiber. While these 
estimates are for northeastern northern red oak, they 
are probably satisfactory for northern red oak and 
similar black oak species throughout the eastern 
United States. 

A number of methods have been developed for 
estimating eastern hardwood tree crown (foliage 
and/or branchwood) 1 weights in addition to, inde- 
pendently of, or together with weights of "merchan- 
table" or specified portions of the upper bole (Storey 
and Pong 1957, Yonngetal. 1964, Ralston and Prince 
1965, Young and Carpenter 1967, Oak Ridge Na- 
tional Laboratory 1971, Zavitkovski 1971, King and 
Schnell 1972, Sando and Wick 1972, Ribe 1973, 
Schlaegel 1975, MacLean and Wein 1976, Phillips 
1977, Wartluft 1977, Wiant etal. 1977, and Wartluft 
1978). Two studies established branchwood size 
classes as needed for fire behavior prediction using 
the Rothermel (1972) model— (Loomis 1975) for 
northern red oak, and (Loomis and Roussopoulos 
1978) for aspen. 

Most investigators developing crown weight pre- 
diction methods have used d.b.h. (diameter at breast 
height) as an independent variable, either alone or 
combined with tree height, crown length, or crown 
ratio. 2 While bole diameter at base of crown is an 



excellent single estimator of crown weights, (Storey 
and Pong 1957, Loomis et al. 1966) it can be difficult 
to obtain. Using d.b.h. and crown ratio to estimate 
shortleaf pine crown weights produced more accurate 
estimates than using d.b.h. alone ( Loomis et al. 1966). 
These results were almost as accurate as those ob- 
tainable by using the bole diameter at base of crown. 

METHODS 

Twenty-eight trees from Michigan and 28 from 
Pennsylvania were destructively sampled in 1973, 
1974, and 1978. The trees were from four locations in 
Wexford, Manistee, and Calhoun Counties in Michi- 
gan and from one location in Huntingdon County, 
Pennsylvania. 

The trees, from fully stocked stands on medium or 
better sites, ranged in size from 1.0 to 20.6 inches 
d.b.h., and in crown ratio from 27 to 82 percent. 3 A 
wide range of crown ratios was selected for each d.b.h. 
class. Vigorous dominant or codominant trees with 
relatively uniform crowns and branching were cho- 
sen. Preference was given to trees with well defined 
boles extending well into the crown. Sampling was 
done from mid-summer through early fall while trees 
were in full foliage, after seasonal growth was 
completed. 

There was noticeable insect defoliation on many 
sample trees in both States. Selecting trees that were 
least affected by this minimized influence on foliage 
weights and resulting estimates but did not com- 
pletely avoid it. 

Tree measurements made included d.b.h., total 
height, live crown length and width, and basal diam- 
eters of all branches at 2 inches from the bole. 4 One 
hundred fifty-four branches were randomly selected 



l "Branchwood" and also "bolewood" as used in this 
paper refer to both wood and bark. 

2 Crown ratio is the ratio of live crown length to total 
tree height expressed as a percent. 



^English-metric equivalents: 1 inch = 2.54 cm; 1 
pound = 0.4536 kg; 1 acre = 0.40469 hectare. 

4 All bole, branch, and branchwood diameters in this 
paper are outside bark (d.o.b.) measurements. 



and cut from sample trees for a branch sample 
representing all crown parts and all trees. Bole 
sections measuring 1 to 3 inches in diameter were 
weighed. 

Field weights of foliage, total live branchwood, and 
live branchwood in four fuel groups — with diameters 
of to l A inch, X A to 1 inch, 1 to 3 inches, and 3 inches 
or more — were determined for each sample branch. 
Sample branches ranged from 0.5 to 10.4 inches in 
basal diameter. 

Field weights for dead branchwood in the four size 
groups were obtained for each tree. The wood was 
weighed for approximately half the trees. As this was 
considered a suitable base, ocular weight estimates 
were included for the other trees, particularly for 
material falling within the smaller diameter groups. 

Factors for converting all field weights to ovendry 
weights for analysis were obtained by ovendrying 
foliage, branch, and bole sections at 105° C for 24 
hours or more. 

Logarithmic transformations were made to adapt 
to the general equation Lny = a + bLnx for predict- 
ing weights of foliage and total live wood per branch 
using basal diameter as the independent variable. 
These and all subsequent equations were adjusted for 
logarithmic transformation bias (Baskerville 1972). 

To aid subsequent mathematical representation, 
measured dry weights of wood in the three mutually 
exclusive size classes (0 to Vi inch, l A to 1 inch, 1 to 3 
inches) were arithmetically combined by branch into 
overlapping classes: to l A inch, to 1 inch, and to 3 
inches. The percent of total branchwood weight per 
branch in each overlapping size class was then plot- 
ted against branch diameter for all sample branches. 
Curves were drawn defining the relations. Next, the 
percentage values for size groups — to l A inch, l A to 1 
inch, 1 to 3 inches, and 3 + inches were computed 
using curve values. These basal branch diameter 



percentages were multiplied by total live branch- 
wood weight for each applicable basal diameter to 
obtain class weights per branch. Foliage and branch- 
wood weights (total and by size class) for each tree 
were then computed by summing the predicted 
weights for the tallied live branch basal diameters. 
Weight and dimension data for the 56 trees were then 
used to develop estimating equations. 



RESULTS 

Regression analysis yielded good relations for pre- 
dicting foliage weights and total weight of live wood 
per branch when using basal diameter as the inde- 
pendent variable ( table 1 ). Equations were developed 
to predict weights of foliage and branchwood per tree 
using various variables: bole diameter at base of 
crown, d.b.h., and the combination of d.b.h. and 
crown ratio (table 2). Resulting estimates for the pair 
of equations using d.b.h. and crown ratio are pre- 
sented on table 3 and 4. The prediction improvement 
from adding crown ratio as a second independent 
variable was significant at the 0.01 level. Equations 
to compute ratios of branchwood weights within 
diameter size class (0 to Vi inch, to 1 inch, to 3 
inches) to total branchwood weight per tree were 
developed (table 5). These predictions are con- 
strained at a 1.0 value as ratios cannot exceed unity. 
These ratios, together with available estimates of 
total live branchwood weight per tree, allowed com- 
putation of estimates of live branchwood weights per 
tree within each of the diameter size groups (table 6). 
Estimates of bolewood weight in the 1 to 3 inch 
diameter class were obtained from a curve drawn 
through data for the bole section weight plotted over 
tree d.b.h. (table 7). 



Table 1. — Equations for estimating northern red oak foliage and Hue branchwood 1 dry weights for individual 

branches 



Dependent variable 


Equations 


R 2 


Sy x 


Percent 

of 
mean 


Foliage 

Total branchwood 


Wf - 0.3925(Bd) 1 5648 
Wb = 1.1852(Bd) 26883 


0.86 
.98 


0.08 
.84 


3.9 
2.6 



'Branchwood includes the topmost section of the bole that is less than 1 inch in diameter. Branchwood refers to both wood and bark. 
The abbreviated terms are: Wf = foliage weight (pounds) 

Wb = live branchwood weight (pounds) 

Bd = branch basal diameter (inches) 

R 2 = coefficient of determination 

Syx = standard deviation about the regression 

Percent of mean = percent error of the mean 



Table 2. — Equations for estimating total dry weight of northern red oak foliage and live branchwood 1 using 

various crown and stem measurements 











Percent 








Percent 










of 








of 


Foliage 




R 2 


Sy x 


mean 


Live branchwood 


R 2 


Sy x 


mean 


Wf = 0.5953DC 1 


6428 


0.98 


0.65 


3.6 


Wb = 0.4928DC 2 8932 


0.99 


11.42 


3.7 


Wf = .4590Dbh 1 


.5018 


.96 


1.18 


6.5 


Wb = .3328Dbh 26360 


.97 


34.07 


10.9 


Wf = .0136Dbh 1 


5791 








Wb - .0005Dbh 2 7768 








Cr 8612 




.98 


.69 


3.8 


q.1 5685 


.99 


16.13 


5.2 



'Branchwood includes the topmost section of the bole that is less than one inch in diameter. Branchwood refers to both wood and bark. 
The abbreviated terms are: Wf = foliage weight (pounds) 

Wb = live branchwood weight (pounds) 

Dc = bole diameter at base of crown, outside bark (inches) 

Dbh = diameter at breast height, outside bark (inches) Cr = crown ratio: crown length in feet h- tree height in feet, expressed 

as percent 
Ft 2 = coefficient of determination 
Syx = standard deviation about regression 
Percent of mean = percent error of the mean 



Table 3. — Dry weight of northern red oak foliage 
(In pounds) 



Crown ratio 1 (percent) 



D.b.h. 
(inches) 



20 30 40 50 



60 



70 80 



1 


0.2 


0.3 


0.3* 


0.4 


0.5* 


0.5* 


0.6 


2 


0.5 


0.8 


1.0 


1.2* 


1.4* 


1.6* 


1.8 


3 


1.0 


1.4 


1.8 


2.2* 


2.6 


3.0* 


3.4 


4 


1.6 


2.3 


2.9* 


3.5 


4.1 


4.7 


5.3 


5 


2.3 


3.2 


4.1 


5.0 


5.9 


6.7* 


7.5 


6 


3.0 


4.3 


5.5 


6.7 


7.8 


9.0 


10.0 


7 


3.9 


5.5 


7.0* 


8.5* 


10.0 


11.0 


13.0 


8 


4.8 


6.8* 


8.7* 


11.0 


12.0 


14.0 


16.0 


9 


5.8 


8.2 


10.0* 


13.0 


15.0 


17.0 


19.0 


10 


6.8 


9.7 


12.0 


15.0* 


17.0 


20.0 


22.0 


11 


7.9 


11.0* 


14.0* 


17.0* 


20.0* 


2?.0 


26.0 


12 


9.1 


13.0 


16.0* 


20.0* 


23.0 


27.0 


30.0 


13 


10.0 


15.0 


19.0* 


23.0 


26.0 


30.0* 


34.0 


14 


12.0 


16.0* 


21.0* 


26.0 


30.0* 


34.0 


38.0 


15 


13.0 


18.0 


23.0* 


28.0* 


33.0 


38.0 


43.0 


16 


14.0 


20.0 


26.0* 


31.0 


37.0 


42.0 


47.0 


17 


16.0 


22.0 


29.0* 


35.0 


40.* 


46.0 


52.0* 


18 


17.0 


24.0 


31.0 


38.0 


44.0 


51.0 


57.0 


19 


19.0 


27.0 


34.0 


41.0 


48.0* 


55.0* 


62.0 


20 


20.0 


29.0 


37.0* 


45.0 


52.0* 


60.0 


67.0 


21 


22.0 


31.0 


40.0 


48.0 


57.0* 


65.0 


73.0 


22 


24.0 


34.0 


43.0 


52.0 


61.0 


70.0 


78.0 


23 


25.0 


36.0 


46.0 


56.0 


65.0 


75.0 


84.0 


24 


27.0 


38.0 


49.0 


60.0 


70.0 


80.0 


90.0 


25 


29.0 


41.0 


53.0 


64.0 


75.0 


85.0 


95.0 



Note: Asterisks identify observed data. 

1 Crown ratio is the ratio of live crown length to total tree height expressed 
as a percent. 



An equation to estimate total dead branchwood 
weight per tree was developed: 

Wb = 0.356 Dbh 1713 

where: 
Wb = dead branchwood weight per tree (pounds) 

and 
Dbh = diameter at breast height (inches). 

The coefficient of determination (r 2 ) was 0.71 and the 
standard deviation from the regression (Sy.x) was 
3.46 for this equation. 

Total dead branchwood weight was subdivided into 
three size groups by plotting each tree's percentage of 
the total within size groups (0 to V* inch, to 1 inch, 
and to 3 inches in diameter) over d.b.h. Curves were 
then drawn through the data. Dead branchwood 
weights within size groups — to V* inch, l A to 1 inch, 
1 to 3 inches, and 3+ inches in diameter — were 
computed using appropriate calculations on curve 
values and the equation for estimating total dead 
branchwood weight per tree (table 8). 



Table 4. — Dry total weight of northern red oak live branchwood 1 

(In pounds) 



1 Branchwood includes the topmost section of the bole that is less than 1 inch in diameter. Branchwood refers to both wood and bark. 
The abbreviated terms are: R,,R 2 ,R 3 = Ratios of branchwood within a group to total branchwood weight. 
Dbh = Diameter at breast height, outside bark (inches). 
Dc = Bole diameter at base of crown, outside bark (inches). 
Cr = Crown ratio: Crown length in feet tree height in feet, expressed as a percent. 
R 2 = Coefficient of determination. 
Syx = Standard deviation about regression. 
Percent of mean = Percent error of the mean. 



D.b.h. 






Crown ratio 2 (percent) 










(inches)D.b.h. 


20 


30 


40 


50 




60 




70 


80 


1 


0.1 


0.1 


0.2 


0.2 




0.3 




0.4 


0.5 


2 


0.4 


0.7 


1.1 


1.6 




2.1 




2.7 


3.3 


3 


1.2 


2.2 


3.4 


4.9 




6.5 




8.3 


10.0 


4 


2.6 


4.9 


7.7 


11.0 




14.0 




18.0 


23.0 


5 


4.8 


9.1 


14.0 


20.0 




27.0 




34.0 


42.0 


6 


8.0 


15.0 


24.0 


33.0 




45.0 




57.0 


70.0 


7 


12.0 


23.0 


36.0 


51.0 




68.0 




87.0 


107.0 


8 


18.0 


33.0 


52.0 


74.0 




99.0 




126.0 


155.0 


9 


25.0 


46.0 


73.0 


103.0 




137.0 




175.0 


216.0 


10 


33.0 


62.0 


97.0 


138.0 




184.0 




234.0 


289.0 


11 


43.0 


81.0 


127.0 


180.0 




240.0 




305.0 


376.0 


12 


54.0 


103.0 


162.0 


229.0 




305.0 




389.0 


479.0 


13 


68.0 


129.0 


202.0 


286.0 




381.0 




485.0 


599.0 i 


14 


84.0 


158.0 


248.0 


352.0 




468.0 




596.0 


735.0 


15 


101.0 


191.0 


300.0 


426.0 




567.0 




722.0 


891.0 


16 


121.0 


229.0 


359.0 


510.0 




679.0 




864.0 


1,065.0 


17 


143.0 


271.0 


425.0 


603.0 




803.0 




1,023.0 


1,261.0 


18 


168.0 


317.0 


498.0 


707.0 




941.0 




1,198.0 


1,478.0 


19 


195.0 


369.0 


579.0 


822.0 




1,094.0 




1,393.0 


1,717.0 I 


20 


225.0 


425.0 


668.0 


947.0 




1,261.0 




1,606.0 


1,980.0 


21 


258.0 


487.0 


765.0 


1,085.0 




1,444.0 




1,839.0 


2,267.0 


22 


293.0 


554.0 


870.0 


1,234.0 




1,643.0 




2,092.0 


2,580.0 


23 


332.0 


627.0 


984.0 


1,397.0 




1,859.0 




2,367.0 


2,919.0 


24 


373.0 


705.0 


1,108.0 


1,572.0 




2,092.0 




2,664.0 


3,285.0 


25 


418.0 


790.0 


1,241.0 


1,760.0 




2,343.0 




2,984.0 


3,679.0 


'Branchwood includes bolewood less than 1 


inch in diameter. Branchwood and bolewood include all woody parts (wood and bark). 




2 Crown ratio is the ratio of live 


crown length to total tree height expressed as a percent. 












Table 5. — Equations for 


estimating ratios of northern red oak live branchwood 1 with 


in a 


size group, to total 








branchwood 


per tree 












Live branchwood 












Percent 




diameter class 




Equation 


R 2 


Syx 


ol 


mean 


n 


0- 1 /4 inch 


R, = 0.6262DbfT 1 0795 


0.94 


0.007 




5.4 








R, = 0.5094DC" 1 1863 


.97 




005 




3.8 




56 




Rt = 6.4735Dbh" 1 1313 




















Cr " 5777 


.96 




007 




5.4 






0-1 inch 


R 2 = 3.1018Dblr 1 0114 


.70 




016 




6.4 








R 2 = 1.7710D<T 9479 


.82 




012 




4.8 




42 




R 2 = 36.8351 Dbh" 9345 




















Cr " 7014 


.79 




014 




5.6 






0-3 inches 


R 3 = 8.7376DblT 1 0113 


.64 




018 




3.0 








R 3 = 4.0290DC" 8443 


.78 




015 




2.5 




34 




R 3 = 28.2916Dbh" 8658 




















Q.-4084 


.72 




017 




2.8 













c/> 


c 


a> 


cu 


.c 


u 












a. 


— 




CO 




GO 




re 








o 








CU 








cu 




E 




re 








-o 



OOOOOOOOOOOOOOOOOOO 



■*— co co o en 






TOIOOOOOOOOOOOOOOOOOOOO 



CM CM 
CD CD 



OCOWtDCOTr^lDN^CMOOlDOlO^lOl^^T 



CM ^f CD OO O (M IT) 



(Di- 1 O) ■» 
CO "3- -*r •>3- LT) 



cd c\j 

CO CO 



WWCD^OOOOOOOOOOOOOOOOOOOOO 



CD N en N O) 

n i in cd N 



in co en oi 



CMOOLOLOLncO-T— CD 

oor-«n^cos 



CD CD O 

i— ■>3- en 

CM CM CM 

nqooooooooooooooo 

O) r- CDLOr-~CT>T— CO CD O t— ^TCDt— ^J" CD 00 

'-i-'-i-'-cMWNnconn'jf'r^ 



■- o o o 

CD CO CD en 
CM Tf N 



ooooooooooooooo 



n n n cm in in n 
*— cd cm cd r— r-~ co 
t— t— CM cm co -<3- en 



O OJ N t- CD CM O) 

in t- n co co i- in 
co cd i— co en co co 



y— l~- CO 
CD CNJ CO 



OOOOOOOOOOOOOOOOOOOOO 



■•— co en cr> co oo 

n en n on co 

T- r— t— CM CM CM 



n in co co 

on n i- 

CO CO CO ^T 



CD CD 

co en 
en cd 



CM CD CM 



OOOOOOOOOOOOOOOOOOOOO 



CM O N N O O O 

^r in en co n oi o 



T CD CD "3- 



t- CM CO If 



•^3- t^. t— CD 

en co co C7J 



ooooooooooooooo 
t- n in n a) t 



■2 

O 
O 

S 
u 

e 

a 

~o 

cu 

a —■ 
•** cn 

•— -a 

~S£ C 

S 3 
o o 

■X3 & 

^ c 
c w 

to 

o 
e 






c- 



EH 









to 




cu 


a> 
o 


cz 






cu 






CO 




to 


cz> 


re 


UD 


u 




w 




cu 








cu 




E 




re 








-o 








to 




cu 




.c 




o 




c= 






C 




CU 






CO 


cu 




Q. 


u 


O 


fc_ 


en 


cu 




cu 




E 




re 








•o 



soooooooooooooooooo 

ocMaicMoicDOeococnTfSinNineocOT-6 

T-cNinNT-ior-smnwcoinainocoN 

i-i-wc\in\rinioNcoot\inin 

(Doooooooooooooooooooooo 

T-CDeOSM'NT-COCOCOOlCOnNCTicMNOLnr-CO^Ol 

t- cvi^eONcoocMTioair- ->3~ co i— ^rr^f— ^rooco 

y— i— ■.— t— t— CMCMCMCOCOCOTT^r^Ten 



N CM CO 
CO CD oi 



oooooooooooooooooooo 

o oo r»~ 



CM CD t- CO 



N CD CO 
CM CO f 



CO CD O O CD O O 

co ai >- n in co co 



ooooooooo 



sooooooooooooooooo 

ninooiocDcociimcoinwooniDOT 

T-coinNO^-ai^ONint^eocoo)^ 

■■—■■— i— Nnn^incoMooir- 

SOlfOOOOOOOOOOOOOOOOOOOO 

^cococooiNinnT-aicom 
neoNOi'-^NOiiM^aio 

t— i— -r— i— t— t— CMCMCMCMCOCOCO^r 

'-t-weo'joooooooooooooooooooo 



•- in o co cnj oi co 

O O t- t- CM CM CO 



to o> t- ^r o o 

en CD CO CD i— CM 



OOOOOOOOOOO 

^rencDooocM^renr^cDCM 

t— t— t— t— CMCMCMCMCMCOCO 



o^ re 



■JOOOOOOOOOOOOOOOO 

fins'jinciiO'TCOTinTf'-iOi-ins 

i-w^iocowincDTOiincjcDNCoeo 

i— T~-r-CMCMCOTj-^3-encDr-- 

conoooooooooooooooooooo 

T-inCNO'-CNlT-^enNcD'^OlinoioiOCNJ^lDCOO 

T-CMCO^incONCOCD'-CMTeOSOCM^'COCOT- 

t— i— i— i— i— CM CM CJ CM CM CO 

i-NCOCOOlOOOOOOOOOOOOOOOOOOOO 



CO CM N ^r t- CO 
O t— i— CM CO CO 



in ir> ar> m 

iniOSCOOl'-CMCO 



OOOOOOOOOOO 

^- CO N Oi 



«- .= CO 









CO 




cu 








o 




c 






t= 




cu 


CO 
CO 


cu 




a. 


u 


<=> 


fc_ 


CO 


cu 








cu 




E 




re 








-o 



inNOOOOOOOOOOOOOO 

t— CMCOCOCOOCDCOCMGOO 

IDCOOCOlDOCOCOCOCOin 
i— t— r— CM PJ CM CO CO T 

coroirooooooooooooooooooo 
o cm en o r-- 



TO CU 

en ■£■ 



F <" 

TO i— 



cd cd en ao 
i— co in r— 



ooooooooooooooooooo 



CD CD CD CD CD CD 









CO 




CU 








u 




cz 






c 




cu 
o 


to 

CO 


cu 


re 


a. 


u 


C3 


fc- 


CM 






cu 




E 




re 








■a 



t-cocooicocococoincMONLncoooio 

OOOO'-'-CMCMtO^^incONCOCOO'-lMCO^incOlOCT) 



(OOOOOOOOOOOOOOO 



■^raocDcocDcDenen 

eOCOCOOCMTNO 



CO CM CO 
1— CO CD 



oooooooooooooo 

TTONTOOCO^ 



T en en cd (^ 



o o o 

CO CM y— 



5 — 



CMCOOOlinCMNOOO 
OOCMCOinNoicM 



OOOOOOOOOOOOOO 



T-cM^riooiMino^cocoNcooeO'-s 
ooboO'-i-CMCMwco'co^^inioio 



T-CMCO^incONCOO)Oi-CMCOTincOS 



CD OO CD CD CD CD CD 
0O OO CD O ■>— CM CO 






Table 7. — Dry weight of bole section from 1 to 3 inches 
in diameter for northern red oak per tree by d.b.h. 





Bole 


D.b.h. 


section 


(inches) 


(lbs.) 





Bole 


D.b.h. 


section 


(inches) 


(lbs.) 



Table 8. — Dry weight of northern red oak deai 
branchwood 1 by four diameter classes and total pe 
tree by d.b.h. 



D.b.h. 




Diameter 




Total dead 


(inches) 


0-.25 


.25-1 1-3 


3 + 


branchwoo 



1 


2 


16 


10 


1 


0.1 


0.3 






0.4 


2 


11 


17 


10 


2 


0.2 


1.0 






1.2 


3 


12 


18 


9 


3 


0.3 


2.0 






2.3 


4 


13 


19 


9 


4 


0.5 


3.3 






3.8 


5 


13 


20 


8 


5 


0.6 


5.0 






5.6 


6 


13 


21 


8 


6 


0.7 


6.9 


0.1 




7.7 


7 


13 


22 


8 


7 


0.8 


6.1 


3.1 




10.0 


8 


13 


23 


7 


8 


0.9 


5.6 


6.0 




13.0 


9 


13 


24 


7 


9 


0.9 


5.2 


9.1 


0.2 


15.0 


10 


12 


25 


7 


10 


0.9 


4.8 


12.0 


0.9 


18.0 


11 


12 






11 


1.0 


4.4 


15.0 


1.7 


22.0 


12 


12 






12 


1.0 


4.0 


17.0 


2.8 


25.0 


13 


11 






13 


1.0 


3.6 


21.0 


3.7 


29.0 


14 


11 






14 


1.0 


3.4 


23.0 


4.9 


33.0 


15 


10 






15 
16 


1.0 
1.0 


3.4 
3.1 


27.0 
30.0 


5.9 

7.4 


37.0 










41.0 










17 


1.0 


2.7 


33.0 


8.6 


46.0 










18 


1.0 


2.7 


37.0 


10.0 


50.0 










19 


1.1 


2.5 


40.0 


12.0 


55.0 










20 


1.2 


2.4 


43.0 


13.0 


60.0 










21 


1.3 


2.3 


48.0 


14.0 


66.0 










22 


1.4 


2.3 


51.0 


16.0 


71.0 










23 


1.5 


2.3 


55.0 


18.0 


77.0 










24 


1.7 


2.5 


59.0 


20.0 


82.0 










25 


1.8 


2.6 


63.0 


21.0 


88.0 



1 Branchwood refers to both wood and bark. 



DISCUSSION 

_ The northern red oak equations with two inde- 
ed pendent variables (d.b.h. and crown ratio) were used 
I0f|to estimate foliage and branchwood weight for two 
independent sets of data. The first set used six species 
of hardwood trees (silver maple, Acer saccharinum 
L.; sweet birch, Betula lenta L.; pignut hickory, 
Carya glabra Sweet; American beech, Fagus grandi- 
folia Ehrh.; yellow-poplar, Lirwdendron tulipifera L.; 
and scarlet oak, Quercus coccinea Muenchh.) from 
the Pisgah National Forest in North Carolina (Storey 
and Pong 1957). A second data set was for quaking 
aspen, (Populus tremuloides Michx.) from northeast- 
ern Minnesota (Loomis and Roussopoulos 1978). Re- 
sults of analysis with this independent data were 
inconclusive. 

The actual and estimated weights were compared 
by a paired t-test (table 9). No significant difference 



between actual and estimated foliage weights was 
indicated for sweet birch, pignut hickory, and aspen. 
Foliage weight differences for all other species tested 
were significant. No significant difference between 
actual and estimated branchwood weights was indi- 
cated for pignut hickory, American beech, scarlet 
oak, and aspen while differences for sweet birch, 
silver maple, and yellow-poplar were significant. 

Specific gravity for wood varies by species, and, to a 
lesser extent, by location (Phillips 1977). Thus, the 
effect of specific gravity adjustment on branchwood 
estimates was examined. Specific gravity values, 
based on volumes at 12 percent moisture content, 
were taken from Wood Handbook (USDA 1974). The 
ratio of specific gravity of each species to specific 
gravity of northern red oak was computed. This ratio 
was used as a multiplier to obtain adjusted branch- 
wood weight estimates. The specific gravity adjust- 
ment yielded a significant improvement in the gap 



Table 9. — Comparison of actual dry weights of foliage and branchwood 1 per tree with estimated weights using the 
equations developed for northern red oak 2 





Crown 
component 


Number 
of trees 


Specific 4 

gravity 

adjustment 


D.b.h. 

range 

(inches) 


Crown ratio 

range 
(percent) 


Means 


Standard 
error 
of the 

difference 




Species 3 


Actual 
(pounds) 


Estimated 
(pounds) 


Paired 5 
t 


Aspen, quaking 

(Populus tremuloides Michx.) 


Foliage 
Branchwood 


15 


0.60 


1.2-15.0 


31-79 


12 

88 
88 


11 
115 

53 


0.8 

14.9 
10.0 


1.00 NS 
1.80 NS 
3.53** 


Beech, American 
(Fagus g rand i folia Ehrh.) 


Foliage 
Branchwood 


14 


1.02 


4.1-15.8 


42-79 


24 
262 
262 


15 
199 
203 


3.4 
34.5 
33.3 


2 61* 
1.82 NS 
1.77 NS 


Birch, sweet 
(Betula lenta L.) 


Foliage 
Branchwood 


17 


1.03 


1.9-138 


31-68 


17 
107 
107 


14 
140 

144 


1.5 

12.5 
13.2 


2.14* 
2.61* 

2.78* 


Hickory, pignut 
(Carya glabra Sweet) 


Foliage 
Branchwood 


16 


1.19 


1.9-23.4 


25-65 


31 
260 
260 


18 
327 
390 


6.7 
55.7 
69.8 


2.02 NS 
1.21 NS 
1.86 NS 


Maple, silver 

(Acer saccharinum L.) 


Foliage 
Branchwood 


16 


.75 


2.2-14.0 


27-77 


21 
108 
108 


15 
163 
122 


2.2 

14.6 
8.8 


2.64* 
3.77** 
1.60 NS 


Oak, scarlet 

(Quercus coccinea Muenchh.) 


Foliage 
Branchwood 


14 


1.06 


4.1-20.2 


30-58 


28 
197 
197 


16 
199 

211 


4.7 
15.2 
14.1 


2.60* 
0.13 NS 
0.99 NS 


Yellow-poplar 
(Liriodendron tulipifera L.) 


Foliage 
Branchwood 


18 


.67 


2.1-21.6 


21-63 


11 
48 
48 


14 
180 
121 


.9 

51.2 
30.0 


4.09** 

2.58* 

2.42* 



1 Branchwood refers to both wood and bark. 

2 Dry weight estimates for foliage and live branchwood per tree obtained using (northern red oak) equations: Wf = 0.0136 Dbh 1 5791 and Wb = 0.005 Dbh 27 

Cr 1 5685 where Wf and Wb = weight of foliage and branchwood respectively in pounds; Dbh = diameter breast height, outside bark in inches; Cr = crown ratio 

(live crown length -total tree height expressed as a percent). 
3 AII species groups except aspen are from Storey and Pong (1 957); aspen is the sample tree group used for Loomis and Roussopoulos (1 978) publication— oata on 

file at North Central Forest Experiment Station, East Lansing field office. 
"Specific gravity adjustment is a multiplier for the computed branchwood weight estimates; it is the ratio of the specific gravity of the concerned species to the 

specific gravity of northern red oak. Specific gravity values obtained from Wood Handbook (USDA 1974). 
5 Levels of significance: 0.01 (**), 0.05 (*), and not significant (NS). 



between estimated and actual weights for silver 
maple, and an insignificant improvement for Ameri- 
can beech and yellow-poplar. In contrast, sweet birch, 
pignut hickory, and scarlet oak had greater differ- 
ences between estimated and actual values. Quaking 
aspen results were significantly poorer. In general, 
using specific gravity adjustment to increase the 
accuracy of applying the red oak equations to other 
species has variable results. 

The results support use of the northern red oak 
equations for foliage and for total live branchwood 
weight for those species where tests indicated no 
significant difference between actual and estimated 
values. Under most circumstances, however, better 
results would probably be obtained by using estimat- 
ing equations based on data for each species. 

Scarlet oak estimated foliage weights averaged 
only 73 percent of actual weights. Although species 
and/or site related differences are possible, estimates 
for northern red oak may be lower due to insect 
defoliation on sample trees. Foliage, not as constant 
as branchwood, represents an annual crop, and its 
quantity may be altered by many things — not only 
insects, but also unusually strong wind and drought. 

The equation, using two independent variables, 
was also tested on an additional data base for esti- 
mating branchwood weight. These data concerned 71 
northern red oak trees ranging from 6 to 24 inches 
d.b.h. from uneven-aged stands on better than aver- 
age sites on the Pisgah National Forest in North 
Carolina. The estimates from the equations devel- 
oped here compared favorably with those using equa- 
tions that had been developed from the independent 
data base. 5 This further supports the application of 
this method throughout the Eastern deciduous 
forest. 

The use of the combination of d.b.h. and crown ratio 
as independent variables is believed to minimize 
effects of differences in stand density and site. 

The equations presented here for estimating 
northern red oak crown component weights are ap- 
plicable to the range of northern red oak in the 
eastern United States. The branchwood estimates for 
northern red oak are considered usable for other 
black oak species, and for approximations for other 
hard hardwoods with similar crown form such as 
hickory. (For practical application, see Loomis and 



5 Personal communication with Alexander Clark HI, 
Forestry Sciences Laboratory, Athens, Georgia, Feb- 
ruary 11, 1980. 



Blank 1981.) However, it is suggested foliage esti 
mates be used for other black oak species and other 
hard hardwoods only when no other estimating pro 
cedure is available. 



LITERATURE CITED 

Baskerville, G. L. 1972. Use of logarithmic regres- 
sions in the estimation of plant biomass. Canadian 
Journal of Forest Research 2:49-53. 

King, W. W., and Robert L. Schnell. 1972. Biomass 
estimates of black oak tree components. Tennessee 
Valley Authority Technical Note Bl, 24 p. Division 
of Forestry, Fisheries, and Wildlife Development, 
Tennessee Valley Authority, Norris, Tennessee. 

Loomis, Robert M., Robert E. Phares, and John S. 
Crosby. 1966. Estimating foliage and branchwood 
quantities in shortleaf pine. Forest Science 12:30- 
39. 

Loomis, Robert M. 1975. Diagnostics of northern red 
oak (Quercus rubra) biomass for predicting fire 
behavior. Michigan Academician 7(4):515-520. 

Loomis, Robert M., and Richard Blank. 1981. How to 
estimate weights of northern red oak crowns in a 
stand. U.S. Department of Agriculture Forest 
Service, General Technical Report (In prepara- 
tion). U.S. Department of Agriculture Forest Ser- 
vice, North Central Forest Experiment Station, St. 
Paul, Minnesota. 

Loomis, Robert M., and Peter J. Roussopoulos. 1978. 
Estimating aspen crown fuels in northeastern 
Minnesota. U.S. Department of Agriculture Forest 
Service Research Paper NC-156, 6 p. U.S. Depart- 
ment of Agriculture Forest Service, North Central 
Forest Experiment Station, St. Paul, Minnesota. 

MacLean, David A., and Ross W. Wein. 1976. Bio- 
mass of jack pine and mixed hardwood stands in 
northeastern New Brunswick. Canadian Journal 
of Forest Research 6:441-447. 

Oak Ridge National Laboratory. 1971. Ecological 
sciences division, annual progress report. ORNL- 
4759, 188 p. Oak Ridge National Laboratory, Oak 
Ridge, Tennessee. 

Phillips, Douglas R. 1977. Total-tree weights and 
volumes for understory hardwoods. TAPPI 
60(6):68-71. 

Ralston, Charles W., and Allan B. Prince. 1965. 
Accumulation of dry matter and nutrients by pine 
and hardwood forests in the lower Piedmont of 
North Carolina. In Forest-Soil Relationships in 
North America: papers presented at the second 



8 



North American Forest Soils Conference, Oregon 
State University, 1963. p. 77-94. Oregon State 
University Press, Corvallis, Oregon. 

Ribe, John H. 1973. Puckerbrush weight tables. 
Miscellaneous Report 152, 92 p. Life Sciences and 
Agriculture Experiment Station, University of 
Maine, Orono, Maine. 

Rothermel, Richard C. 1972. A mathematical model 
for predicting fire spread in wildland fuels. U.S. 
Department of Agriculture Forest Service, Re- 
search Paper INT-115, 40 p. U.S. Department of 
Agriculture Forest Service, Intermountain Forest 
and Range Experiment Station, Ogden, Utah. 

Sando, Rodney W., and Charles H. Wick. 1972. A 
method of evaluating crown fuels in forest stands. 
U.S. Department of Agriculture Forest Service, 
Research Paper NC-84, 10 p. U.S. Department of 
Agriculture Forest Service, North Central Forest 
Experiment Station, St. Paul, Minnesota. 

Schlaegel, Bryce E. 1975. Estimating aspen volume 
and weight for individual trees, diameter classes, 
or entire stands. U.S. Department of Agriculture 
Forest Service, General Technical Report NC-20, 
16 p. U.S. Department of Agriculture Forest Ser- 
vice, North Central Forest Experiment Station, St. 
Paul, Minnesota. 

Storey, T. G., and W. Y. Pong. 1957. Crown charac- 
teristics of several hardwood tree species. U.S. 
Department of Agriculture Forest Service, Interim 
Technical Report AFSWP-968, 36 p. 

U.S. Forest Products Laboratory. 1974. Wood hand- 
book; Wood as an engineering material. U.S. De- 
partment of Agriculture, Agriculture Handbook 
72, revised, 398 p. U.S. Government Printing Of- 
fice, Washington, D.C. 



Wartluft, Jeffrey L. 1977. Weights of small Appala- 
chian hardwood trees and components. U.S. De- 
partment of Agriculture Forest Service, Research 
Paper NE-366, 4 p. U.S. Department of Agriculture 
Forest Service, Northeastern Forest Experiment 
Station, Upper Darby, Pennsylvania. 

Wartluft, Jeffrey L. 1978. Estimating top weights of 
hardwood sawtimber. U.S. Department of Agricul- 
ture Forest Service, Research Paper NE-427, 7 p. 
U.S. Department of Agriculture Forest Service, 
Northeastern Forest Experiment Station, Upper 
Darby, Pennsylvania. 

Wiant, Harry V., Carter E. Sheetz, Andrew Colan- 
inno, James C. DeMoss, and Froylan Castaneda. 
1977. Tables and procedures for estimating 
weights of some Appalachian hardwoods. Bulletin 
659T, 36 p. Agriculture and Forest Experiment 
Station, West Virginia University, Morgantown, 
West Virginia. 

Young, Harold E., Lars Strand, and Russell Alten- 
berger. 1964. Preliminary fresh and dry weight 
tables for seven tree species in Maine. Technical 
Bulletin 12, 76 p. Maine Agriculture Experiment 
Station, University of Maine, Orono, Maine. 

Young, Harold E., and Paul M. Carpenter. 1967. 
Weight, nutrient element and productivity studies 
of seedlings and saplings of eight tree species in 
natural ecosystems. Technical Bulletin 28, 39 p. 
Maine Agriculture Experiment Station, Univer- 
sity of Maine, Orono, Maine. 

Zavitkovski, J. 1971. Dry weight and leaf area of 
aspen trees in northern Wisconsin. In Forest Bio- 
mass Studies, p. 193-205. H. E. Young, editor. Life 
Science and Agriculture Experiment Station, Uni- 
versity of Maine, Orono, Maine. 



-", U.S. Government Printing Office: 1981--766-968/200 



Loomis, Robert M., and Richard W. Blank. 

1981. Estimating northern red oak crown component weights in the 
Northeastern United States. U.S. Department of Agriculture Forest 
Service, Research Paper NC-194, 9 p. U.S. Department of Agriculture 
Forest Service, North Central Forest Experiment Station, St. Paul, 
Minnesota. 

Equations are described for estimating crown weights for northern 
red oak trees. These estimates are for foliage and branchwood 
weights. Branchwood (wood plus bark) amounts are subdivided by 
living and dead material into four size groups. Applicability of the 
equations to other species is examined. 

KEY WORDS: forest fuels, fuel modeling, biomass. 



/9<s~ 



/ U ' / I 

5DA FOREST SERVICE 
SEARCH PAPER NC-195 




Evaluation of a vest-pocket park 




Rachel Kaplan 



North Central Forest Experiment Station 
Forest Service, U.S. Department of Agriculture 



CONTENTS 

Page 

PRE-DESIGN SURVEY 1 

POST-DESIGN SURVEY 2 

The Sample 2 

The Questionnaire 3 

Background Characteristics 3 

Uses and Importance 5 

Satisfaction 6 

Problems 7 

Participation 8 

Preferred Places 8 

CONCLUSIONS AND IMPLICATIONS 10 

LITERATURE CITED 10 

APPENDIX— SAMPLE SIZE FOR ANALYSES 11 



North Central Forest Experiment Station 

Forest Service — U.S. Department of Agriculture 

1992 Folwell Avenue 

St. Paul, Minnesota 55108 

Manuscript approved for publication July 12, 1979 

1981 



EVALUATION OF AN URBAN VEST-POCKET PARK 



Rachel Kaplan 

Professor, School of Natural Resources and Urban and Regional Planning Program, and Associate Professor, 
Department of Psychology, University of Michigan, Ann Arbor, Michigan 




Liberty Plaza Park 



Liberty Plaza is a vest-pocket park in downtown 
Ann Arbor, Michigan. The park's design was based, 
in part, on local people's reactions to photographs of 
simulated park scenes (Kaplan 1978a). These reac- 
tions indicated that people desired a "green" place, 
complete with flowers, grass, and large trees. People 
living in the vicinity of the proposed park were 
concerned about safety and security as well. The park 
designers proceeded with these objectives in mind, 
and Liberty Plaza was opened to the public in the fall 
of 1977. 

A year later, at the request of the Parks and 
Recreation Department, I conducted a follow-up sur- 
vey to find out if people were satisfied with the park. 



Based on the park's popularity as a place to eat lunch, 
I expected the frequent users to express a high level of 
satisfaction. But to determine the public's overall 
reaction to the park, I felt it was important to reach 
beyond the group that clearly enjoys it as a place to sit 
and rest. Therefore, I also sampled the opinions of 
people who live or work near the park, but who do not 
necessarily spend time there. The results are detailed 
here. 

PRE-DESIGN SURVEY 

Two landscape architects, Terry Brown and 
Charles Cares, were hired in 1975 to design the park. 



The Parks and Recreation Department, as well as the 
designers, wanted to obtain citizen input during this 
process. People were asked to respond to photographs 
of simulated park settings depicting various 
arrangements of green space, trees, benches, etc. 
Although the 24 scenes lacked detail, the 180 partici- 
pants had no difficulty indicating their preferences. 
Among the most obvious results was that people 
wanted a "green" place — a park with color, flowers, 
grass, and big trees. In addition, the survey showed 
strong differences in preferences between those who 
live in the central business district and those who 
work there. Clearly, the workers saw such a park as 
an amenity, while for the nearby residents it posed 
some threats. Because these concerns played a major 
role in the design of the park, we examined them 
again in the post-design survey. 

POST-DESIGN SURVEY 

The post-design evaluation took place in the fall of 
1978, about a year after the park dedication. The 
evaluation was not triggered by problems such as 
vandalism or citizen discontent, but by the commend- 
able desire of the Parks and Recreation Department 
to find out if the park was achieving its objectives. 

Ann Arbor, with a population of about 106,000, has 
extensive parks and open space. In addition to the 
2,500 acres managed by the Parks and Recreation 
Department, the school system owns playfields and 
natural areas, and the University of Michigan has 
over 600 acres that are available for public use. In the 
central business district, however, Liberty Plaza — a 
mere one-fourth of an acre in size — is the only park. 

Bordering Liberty Plaza on the south is the Kempf 
House, a classic example of Greek Revival Architec- 
ture, built around 1850. The City's Historical Com- 
mission purchased the home in 1969, and it is now 
open to the public on weekends. Immediately to the 
west of the park is a large office building that was 
constructed at about the same time as the park. The 
building has a restaurant overlooking the park as 
well as several shops a half-story below street level 
that open to the park side of the building. 

To keep the park compatible with this building, the 
park was built on several levels, the lowest of which 
connects to the shops. Seating, in the form of wooden 
benches with backs, is available at an intermediate 
level as well as at street level. The benches can 
accommodate about 100 people, and low ledges 
around numerous planters also provide seating, es- 
pecially during planned events. At peak times, dur- 
ing concerts or other performances, there may be as 
many as 500-600 people in the park. 



The Sample 

Liberty Plaza is at the intersection of two major 
thoroughfares. One of these, Division Street, is a one- 
way arterial carrying heavy traffic volumes much of 
the time. The other, Liberty Street, is a commercial 
corridor linking the downtown area with the State 
Street-campus area. Within a radius of about two 
blocks, there are apartment units and multi-family 
houses, shops, offices, relatively large commercial 
establishments, and some public facilities. 

With this diversity of uses in the vicinity, one 
would expect to find people at Liberty Plaza who livei 
in the area, work nearby, or come to the park while 
they are in the area shopping, running errands, or 
whatever. But in addition to asking park users how 
they feel about the park, we felt it important toj 
sample the opinions of people who have the park in 
easy reach and may or may not use it. Is the park 
serving them in any sense? The evaluation thus 
included two separate groups: park users (on-site 
sample), and people who live or work within a speci- 
fied radius of the park (off-site sample). 

To obtain the on-site sample, interviewers ap- 
proached 163 people who were in the park at different 
times of day and on different days of the week over 
several weeks. (Only one interviewer was present at 
a time.) A particularly cold October made the park a 
less popular place than it had been only a short time 
earlier. Thus, the interviewing period was extended 
in the hope that milder days would return. 

To obtain the off-site sample, questionnaires were 
delivered to residences and business establishments 
within a prescribed radius (fig. 1). For the residences, 
339 questionnaires were dropped off and 82 were 
returned (24 percent). For the businesses, approxi- 
mately 380 questionnaires were left at 90 places, and 
about 40 percent were returned. At each business, a 
responsible person was asked to see that employees 
had an opportunity to respond. 

The business establishments included several doc- 
tors' offices and small law firms, many small shops, 
and a few large employers such as the telephone 
company, the newspaper, the public library, and 
some banks. We do not know how the questionnaires 
were distributed to employees, nor which establish- 
ments returned the most questionnaires. 

While 61 percent of the off-site sample (n = 233) 
was comprised of people who work in the a. ea, the on- 
site sample consisted mostly of people who neither 
live nor work in the central business district (63 
percent). Of these, almost half (45 percent) indicated 









J 



jUi 





g o 

> I 

= o 



200 400 600 

Questionnaire 

Distribution Area __ — _. 
Downtown Area " I ■!■!■! 




I 1 ( 1 i 

Figure 1. — Distribution of questionnaires 



The Questionnaire 

All participants were administered the same ques- 
tionnaire (Appendix). It included a cover letter, 
signed by the Superintendent of the Parks and Recre- 
ation Department, urging people to reply. It also 
mentioned the fact that citizens had participated in 
designing the park, and pointed out the importance of 
continued input. The bottom of the page included a 
small sketch showing the location of the park. 

The three-page questionnaire included both open- 
ended items and scaled items. The major sections of 
the survey covered the kinds and frequencies of use, 
satisfactions, problems, and particular places within 
the park that were favored. The last of these was 
based on a map of the park (fig. 2) which identified six 
regions, designated by the letters "A" to "F." The map 
took most of the middle page of the questionnaire; at 
the top of the page the question inquired how much 
the participant liked to be in each area "to sit in this 
area, or walk through it" using a 5-point scale. 
Participants were also asked to indicate on the map 
their favorite places within the park. 

The major independent variables included: age, 
sex, student or not, full-time employment or not, 
frequency of park use, length of work or residence in 
the downtown area, whether the park is passed on 
foot, by bike, or by car, and — of greatest interest — 
whether the participant lives or works in the area. 



they are in the area "quite often," while 33 percent 
indicated "very often" and 22 percent responded "not 
very often." 

The two samples were remarkably similar with 
respect to other variables, however. About half of 
each sample was female, about 40 percent was male, 
and the rest did not indicate sex on the questionnaire. 
About 55 percent of the people in each sample were in 
their twenties; the on-site sample included a few 
more people in their teens (12 percent as opposed to 3 
percent for the off-site group), and a few less people in 
their forties or fifties (9 percent as opposed to 20 
percent). The off-site sample included 29 percent who 
indicated they were students; for the on-site group 37 
percent were students. 

The on-site sample was about evenly divided be- 
tween people who frequent the park at least weekly 
as opposed to those who are there less often. About 
one-fifth of the off-site participants indicated they do 
not spend time in the park, while 42 percent said they 
visit it at least once a week and 36 percent said they 
frequent it less often than that. 



Background Characteristics 

As mentioned earlier, one of the most important 
differences revealed by the pre-design survey was in 
the attitudes of people who live downtown versus 
those who work there. The park was designed with 
their separate concerns in mind, and the post-design 
evaluation was structured to obtain responses from 
both groups. 

A little less than half the overall sample (47 
percent) consisted of people who work downtown 
(table 1). Of these, 82 percent had full-time jobs, and 
only 10 percent were students. Women outnumbered 
men almost two to one, and about half of the working 
group was 30 or older. 

About a quarter of the sample were residents of the 
downtown area. Of these, only 37 percent held full- 
time jobs, and 60 percent were students. Of the latter, 
83 percent attended the University of Michigan; the 
others attended a variety of schools and colleges. The 
"student" and "full-time job" categories are not mu- 
tually exclusive (about 12 percent of the students also 



NORTH V 



LIBERTY SQUARE(OFFICE BLDG.) 






KJJJU^ 



DOWN 





a g o„„ .1 



DIVISION STREET 

Figure 2. — Sketch map of the park. 



LU 

IJJ 

cc 

I- 

> 
H 
QC 

OJ 

CO 



held full-time jobs). Not surprisingly, the "live down- 
town" participants were younger, with 80 percent 
under 30. Men and women were equally represented. 
The remaining quarter of the sample (28 percent 
actually) neither lived nor worked downtown. About 
43 percent were students and 42 percent were em- 
ployed full-time. Women accounted for 52 percent of 
this group, and about one-third were over 30. This 
group also included more people in their teens than 
did either of the other two. 



It should be mentioned that 21 participants who 
lived and worked in the downtown area were in-'i 
eluded only in the "live downtown" group. The defini- 
tion of "downtown" was somewhat more inclusive 
than the area included in the survey itself. People 
living or working within the area bound >d by Main 
Street, Huron, State, Madison, and Packard were 
included (fig. 1). 

Of those who lived downtown, 46 percent had been 
there less than a year, and 40 percent for a year or 



Table 1. — Background characteristics of respondents 
(In percent) 







Group 




All 




Background 


Live downtown 


Work downtown 


Other 


groups 


characteristic 


(n = 98) 


(n = 186) 


(n == 112) 


(n 


3%) 


Sex: 












Male 


44 


32 


38 




37 


Female 


46 


61 


52 




55 


No data 


10 


7 


10 




8 


Age: 












Under 20 


6 


4 


1? 




7 


20-29 


74 


45 


56 




56 


30-49 


12 


38 


23 




27 


50 + 


3 


11 


4 




6 


No data 


5 


2 


5 




4 


Full-time job 


37 


82 


42 




59 


Student 


60 


10 


43 




32 



longer. Among the downtown workers, a majority (59 
percent) had worked there for a year or longer and 25 
percent for less than a year. Information was not 
available for the others. 

Participants were asked how often they pass the 
park on foot, by car, and on bike. For those who lived 
downtown, passing on foot was a very frequent occur- 
rence (70 percent), while passing by car was "not 
often" for the majority (53 percent). About half of the 
"work downtown" participants indicated that they 
pass the park "very often" on foot, and a similar 
number pass it that frequently by car. 

Participants varied considerably in frequency of 
park use. Just under half of the "live downtown" and 
"work downtown" samples (45 and 48 percent, re- 
spectively) indicated at least weekly use of the park 
(keep in mind the survey was conducted in the fall). 
For the rest of the sample, 38 percent indicated at 
least weekly use. 

Participants who worked downtown used the park 
most heavily during lunchtime, and not much at 
other times. Those who lived downtown used the park 
less during lunchtime (only 22 percent), but much 
more extensively in the evenings, during events, and 
on weekends. 



Uses and Importance 

What do people do in the park and in what ways is 
it important to them? Not surprisingly, the answers 
to these questions are strongly interrelated. So, for 
example, the park is important as "a place to rest," 



"to sit down where there are trees," "to get away from 
things for a little while" and to "rest, sun, and think." 
These four items, describing the "sit and rest" func- 
tions of the park, define one of its salient features. For 
the sample as a whole this cluster rated an average of 
3.8 on a 5-point scale (table 2). The "sit and rest" 
cluster is least important to people who live in the 
downtown area, and particularly important to those 
who neither live nor work there (mean 4.0). Younger 
participants (under 30) rated these items signifi- 
cantly higher, especially those who work downtown. 1 

The park is also "a place to have lunch," to "eat and 
drink." As noted earlier, the downtown residents are 
not as likely to be lunchtime users and these items 
rate 2.6 for them. The other two groups, however, 
average 3.6 on this item. The "eating" function is far 
more important for women than for men, and higher 
for the on-site sample than for the off-site. 

Some people arrange to meet a friend in the park, 
or go there to be with other people they know. This is 
particularly true for the women in the sample (mean 
3.3) and for the younger members of the "work 
downtown" group (mean 3.4). By contrast, "people- 
watching" seems to be everybody's sport, indulged in 
by people of all backgrounds and represented by a 
mean of 3.6 for the sample as a whole. 



l The results discussed in the report are based on t tests 
and analyses of variance. Only results significant at the 
95 percent level are included in this discussion. (Ap- 
pendix shows sample sizes for the various analyses.) 



Table 2. — Analysis of uses and importance 1 












Respondents' 






Access 






Park iirp Rating (1-5) Location Age 

Mean SD on-site/off-site <30/30 + 


Sex 
F/M 


Bike 
Yes/no 


Foot 
Yes/no 


Car 
Yes/no 


Relation to park 
Live/work/other 



Link, shops 
Know it's there 
See it 
Sports 

Read, sketch 
Meet friends 
Eat, have lunch 
People-watch 
Sit and rest 



2.8 
3.4 
4.2 
1.3 
3.1 
3.2 
3.3 
3.6 
3.8 



1.2 

1.4 
1.1 
.8 
1.5 
1.2 
1.4 
1.2 
1.0 



+■ 
+ + + 

+ rf + 



+ + + 

+ 



+ 



+ + 
+ + + 
+ + + 



+ 

+ + 
+ + + 



+ 

+ 



+ 



+ + + 
+ + + 
+ + + 



+ 



1 + indicates first sub-group in comparison rated higher 
- indicates second sub-group in comparison rated higher 
x indicates the middle group is different from the other two 
+ or - = p<0.05 
+ + = p<0.01 
+ + + = p<0.005. 



The "read, sketch, write" function was more impor- 
tant to younger (under 30) than older participants 
(means 3.3 and 2.7), and more important to women 
than men among the downtown workers (means 3.3 
and 2.5). This function was especially important to 
frequent park users who neither live nor work down- 
town (mean 3.9). 

The participants were asked how likely they are to 
engage in sports (jog, skateboard, etc.) while in the 
park. The results show they are extremely unlikely 
to do such things (mean 1.3). Those who bike rated 
the item higher than did the others, but not much 
higher. While there are occasional skateboarders in 
the park, they are not represented in the sample. 
Generally, the park is not conducive to such activites. 

The park serves some other functions as well, 
which are not related to specific activities. The park 
has shops near it and serves as a midway point 
between two shopping areas. But in general, the 
"shopping" item did not receive very high ratings. 
The most important function of the park is that it is 
"a nice place to see as you go by." This item received a 
mean rating of 4.2 for the sample as a whole and 
seemed equally important to men and women, resi- 
dents, shoppers, and employees, young and old. The 
very existence of the park also plays an important 
role. To "know it is there" is important to people, 
quite independently of their pattern of use. 



Satisfaction 



Do people find Liberty Plaza a pleasing place, does 
it meet their needs? Eight of the items dealing with 
satisfaction were strongly interrelated and were 
combined to form an "overall satisfaction" scale. 
These included such things as the placement of 
plantings, the "variety of kinds of places," the seating 
facilities, and the overall appearance of the park. The 
park fared extremely well in terms of these consider- 
ations. The mean rating for the entire sample was 4.1 
on a 5-point scale (table 3). Women rated overall 
satisfaction somewhat higher than men (4.2 and 3.9), 
as did people who more often pass the park on foot 
(compared with those who are less likely to walk by 
it). Among the people who live in the downtown area, 
those who frequent the park more often (at least 
weekly) show a great overall satisfaction (means 4.3 
and 3.7). While these differences are interesting, the 
most striking result is that by and large the overall 
satisfaction level was high for everyone. 

Because seating was a source of concern in the pre- 
design study, the two items included in the overall 
satisfaction scale that dealt with "kinds uf benches" 
and "seating arrangement" were examined sepa- 
rately as well. The mean rating for these was 4.1, and 
all the differences mentioned above were true for the 



Table 3. — Summary of satisfaction analyses (see table 2 for instructions on interpreting table) 





Rating (1-5) 


Respondents' 




Access Relation to park 
Foot Live/work/ 


Time 


in park 


Satisfaction item 


Location 


Age 


Sex 


Live 


Work 




Mean 


SD 


On-site/off-site 


< 30/30 + 


F/M 


Yes/no 


other 


wkl/< 


wkl/< 


Overall satisfaction 


4.1 


0.8 






+ + 


+ + + 




+ + + 




Seating 


4.1 


.9 


+ + + 




+ + 


+ + + 




+ + + 


+ + 


City's care of park 


4.1 


1.0 


+ 


■+ 


+ + + 


+ + 


X X 






People in the park 


3.8 


1.0 


+ + + 


- 











+ 


Activities 


3.2 


1.1 
















Having the park 


4.6 


.8 


+ 












t 



seating questions also. In addition, people in the on- 
site sample expressed greater satisfaction with seat- 
ing than did those in the off-site sample (4.6 and 4.3). 
The satisfaction items included the "the city's care 
of the park." This also received a mean of 4.1, 
expressing overall approval of park care as the first 
year of park activity neared its end. It is interesting 
that the "work downtown" group showed the lowest 
rating on care (3.9), while the downtown residents 
and "others" were more pleased (4.1 and 4.3). Here 
again, women were more favorably inclined. 

Even a beautifully designed park might not be well 
received if it attracts "the wrong kind" of people. The 
item dealing with "people who are in the park" 
received a mean rating of 3.8; people who neither live 
nor work downtown rated it highest, and local resi- 
dents rated it lowest (4.1 and 3.5, respectively, with 
3.9 for the "work downtown" group). People in the on- 
site sample expressed greater satisfaction with the 
people in the park than those in the off-site group. 
Interestingly, the older participants rated this item 
higher than did those under 30. 

The lowest ratings among the satisfaction items 
were given to those dealing with special events in the 
park: "planned activites, special programs," and 
"other activities that go on." The mean rating was 3.3 
and none of the analyses based on background varia- 
bles yielded any significant differences. (One excep- 
tion: the bikers rated these more favorably.) The 
"activities" issue received considerable mention in 
the open-ended responses, and appears to be the 
major sore point with respect to the park. Some local 
citizens complained of excessive noise and excessive 
duration of concerts. On the other hand, many resi- 
dents, as well as people who work in the area and 
others, expressed a desire for more activities. Well 
over a third of the participants indicated they would 
like to have musical events in the park. About 12 
percent said they wanted no sponsored activities of 



any kind. (Interestingly, two-thirds of these are 
people who work in the downtown area.) While the 
disgruntled are seriously affected, their numbers 
seem to be few indeed. Still, this is clearly an issue 
that needs to be discussed and handled carefully. 

While people seem pleased with the physical as- 
pects of the park, with its care, and with the people 
who frequent it, the biggest source of pleasure is the 
fact that the park is there. Having 355 people yield an 
average rating of 4.6 on "having the park there" must 
be considered an overwhelming vote of appreciation. 

The answers to the open-ended question "What 
kinds of things do you like most about the park?" 
further reflect these satisfactions. The most frequent 
comments (and the number of people who made 
them) were: a place to sit, benches (50); knowledge of 
it being there (46); natural atmosphere, trees, plants 
(43); attractive design (34); quiet and peaceful (28) 
accessible, good location (27); an oasis in the city (24) 
a sunny atmosphere (16); safe, clean, well-lit (13) 
watch people (13); place for people (12); private, 
different levels (10). 



Problems 

The items dealing with potential problems of the 
park revealed two noteworthy findings: first, the 
responses are not highly interrelated. This means 
that participants' concerns for any one issue do not 
have a bearing on their concern for other issues. 
Second, the problems were not generally considered 
serious. 

Of the problems listed in the questionnaire, the 
biggest by far is traffic, referring to the heavy use of 
both streets adjoining the park. The mean rating for 
the item was 2.5, half-way between "a little" and 
"somewhat" of a problem (table 4). The people re- 
sponding to the questionnaire while at the park rated 
traffic significantly more problematic — a mean of 2.7 
compared with 2.3 for the off-site group. 



Table 4. — Analyses pertaining to problems (see table 2 for instructions on interpreting table) 





Rating (1-5) 




Respondents' 




Relation to park 


Problem 


Location 
On-site/off-site 


Age 
<30/30 + 


Sex 
F/M 


Live/work/ 




Mean 


SD 


other 


Traffic 


2.5 


1.3 


+ + 


+ 






Maintenance 


2.3 


1.0 


- 






X 


Crowds 


2.0 


1.1 









XXX 


Noise 


1.8 


1.1 









+ 


Feeling unsafe: 














Daytime 


1.3 


.8 










Evening 


2.2 


1.3 


- 




+ + + 





The park maintenance and amount of litter were 
the next greatest problem, with a mean of 2.3 for the 
sample as a whole. The off-site people considered it a 
worse problem than did the on-site participants (2.4 
and 2.1), and the people who work downtown feel it 
is a worse problem than do the other two groups. 
These differences reflect the previously discussed 
satisfaction with the "city's care of the park" and 
taken together indicate a relatively high level of 
satisfaction. 

Not surprisingly, crowding during lunchtime was 
seen as a bigger problem by those who work in the 
downtown area than by the residents or others 
(means 2.1, 1.9, and 1.7, respectively). It was also 
judged a greater problem by the off-site group (which 
was more heavily represented by people who work 
downtown) than by the on-site participants. Here 
again, the ratings suggest that as problems go, this is 
not a big one. 

Some citizens had previously complained about 
noise at special events, and it was not clear how 
pervasive a problem it might be. The overall mean 
rating of 1.8 suggests that it is at most a minor 
nuisance. Interestingly, it is rated approximately the 
same by those who live downtown and those who 
work there, and least by others (1.8, 1.9, and 1.5, 
respectively). 

Finally, there is the issue of safety — both in the 
daytime and in the evening. In the pre-design survey, 
people living in the park's vicinity were concerned 
about the possibility of "muggers" lurking behind 
retaining walls and trees. For the participants in the 
current study at least, daytime safety is not an issue. 
With a mean rating of 1.3 it is clear that few 
respondents consider it to be a problem at all. "Feel- 
ing unsafe in the evening" is seen as a slightly more 
realistic problem, with a mean of 2.2. (Presumably 
the fact that only about three-fourths of the partici- 
pants answered this item indicates that many are not 



in the vicinity of the park in the evenings. ) For the 
off-site sample, this problem rated somewhat higher 
than for the on-site participants. In addition, as 
would be expected, women rated safety as being a 
bigger problem (2.5) than did men (1.7). 

Written comments about problems were widely 
scattered. The "bees and wasps" in great evidence 
that summer received 19 mentions, and overflowing 
trash cans received 13 comments. "Drunks, street 
people" were listed as a problem by 16 participants, 
mostly downtown residents. Eleven people expressed 
the desire for more trees. This was also by far the 
most frequently mentioned item under "What kinds 
of things do you wish could be changed about the 
park?". The desire for more trees, more plants and 
flowers, more grass, and less concrete was mentioned 
between 12 and 20 times, reminiscent of some of the 
comments in the pre-design survey. 



Participation 

Citizens can express their interest and concern for 
the well-being of the park in different ways. They 
were asked about a few of these. For example, the 
questionnaire inquired whether participants were 
likely to weed the plantings. The results confirm 
what one would expect. By and large, this is some- 
thing people are unlikely to do: mean 1.5 for the 
sample as a whole (table 5). Picking up litter, how- 
ever, is a different matter. Here the overall mean was 
2.6, and those who frequent the park more often are 
much more likely to do it. 

Another expression of interest involves a more 
vicarious concern for the plantings and curiosity 
about "what the city workers are doing in the park". 
Here the group mean is 2.3, with the people who see 
the park from their place of work and the participants 
who have lived downtown longer than a year express- 
ing far greater interest than others. Interestingly, 



8 



Table 5.- 


-Summary of participation questions (see table 2 for 


instructions 


on interpreting table) 






Respondents' Access 








Relation to park 






Live/ 
work/ 
other 


Time 
Live 
wkl/< 


in park 
Other 
wkl/< 


Live downtown 
yr/yr + 


See park 


Participation 
item 


Rating (1-5) Location Age Bike Foot 
Mean SD On-site/off-site <30/30+ Yes/no Yes/no 


Car 

Yes no 


from work 
Yes/no 


Weeding 
Litter 
Plantings 
Own/share 


1.5 1.0 

2.6 1.4 

2.3 12 — + 

2.4 1.2 + + + + + 


+ 


« 
+ 


+ 
+ 


+ 
+ + + 





+ + + 
+ 



Table 6. — Preferred places analyses (see table 2 for instructions on interpreting table; the last column indicates the 
number of votes for each area) 





Rating (1-5) 


Respondents' 
Sex 


Access 






Relation to park 






See park 
Live/work/ from work 


"Favorite 


Preferred 


Bike 


Foot 


place'' 


area 


Mean 


SD 


F/M 


Yes/no 


Yes/no 


other 


Yes/no 




A 


2.7 


1.4 


+ 


+ + + 




X 


- 


16 


B 


3.1 


1.3 












35 


C 


3.4 


1.3 




+ 


+ 






32 


D 


3.3 


1.3 








> 


+ 


43 


E 


3.3 


1.4 


+ 


+ + + 








69 


F 


3.0 


1.4 




+ + + 








10 



people in the off-site sample rated this item higher 
than those in the on-site, and those who pass the park 
either on foot or by car say they are more likely to 
check to see how the plants are doing. 

Perhaps the strongest sense of participation is 
expressed by the items reflecting a "sense of. 
ownership toward the park" and a desire to "share it 
when people come to visit". Clearly, those who spend 
more time in the park reflect a greater sense of this, 
and the downtown residents also express more of this 
"own/share" characteristic. 



Preferred Places 

The park has several levels with stairs and ramps 
connecting them. Partly as a result of the levels, and 
partly because of its access from both streets as well 
as a nearby parking lot, there are several distin- 
guishable settings within the park. There are numer- 
ous ways for pedestrians to enter the park. 

In terms of places to sit, six areas were marked on 
the questionnaire (fig. 2). The responses reflect dif- 
ferent preferences for these settings with mean 
ratings between 2.7 and 3.4 (table 6). None of the six 
places showed any noteworthy differences between 



the on-site and off-site groups, nor for the younger as 
opposed to the older participants. The diversity and 
distinctiveness of the different areas within the park 
met a wide range of desires. 

Students seem to prefer Area A, a little "nook" that 
affords privacy in terms of seating, though not in 
terms of passers-by. This area is also preferred by 
women in the sample, but not people who work 
downtown and can see the park from their working 
place. 

Women also seem to prefer Area E, a fairly large 
area on the lower level. This area gives a relatively 
enclosed feeling, providing room for quite a few 
gatherers. 

The area nearest the street corner (C), easily 
accessible from the corner or from the shopping 
street, won the favor of those who are most likely to 
come "on foot." 

Area D, by contrast, a corner seating arrangement 
nearest the entry from the shopping corridor, was 
highly favored by those who can see the park from 
their place of work (mean 3.8), and was rated higher 
by those who work downtown than by local residents 
(means 3.5 and 3.0). Of people who live nearby, those 
who frequent the park more often greatly preferred 
Area D. 



CONCLUSIONS AND 
IMPLICATIONS 

The park serves different groups of users, and each 
group derives different satisfactions. Its primary 
users are people who work downtown, but many are 
also people who are downtown to do their shopping 
and run errands. The nearby residents are less likely 
to use the park as a place to sit and rest, to have 
lunch, or meet friends. 

Nearby residents were less satisfied with the type 
of people in the park than the downtown workers and 
others. One would expect nearby residents to see 
more problems with the park than people who use it 
during the working day, but this was not generally 
the case. The three groups surveyed did not differ in 
their feelings toward safety, but nearby residents 
and downtown workers found noise more of a problem 
than did nonlocals. Downtown workers regarded 
maintenance and lunchtime crowding as greater 
problems than the other two groups. This suggests 
that those who work downtown use the park more 
than nearby residents, which, in fact, is the case. 

Little background information on participants was 
obtained in the pre-design study, so it is difficult to 
judge whether the current sample and the earlier one 
are comparable. Several concerns expressed in the 
earlier study were not voiced in the post-design 
survey. For example safety was seen as a problem 
before the park was built, but not after. The park 
seems to have satisfied workers who stressed the 
need for a pleasant place to sit and have lunch. The 
general satisfaction with "having the park" seems to 
be common with all user groups, regardless of resi- 
dence and employment. 

This post-design evaluation serves two purposes. 
The first is its applicability to other small downtown 
parks. A park that serves a residential clientele and 
one that serves a daytime working group must meet a 
variety of demands and must avoid a variety of 
problems. Where the residents are likely to use the 
park less extensively than people who work in the 



area, it is appropriate for the park to be designed with 
the needs of the working group in the forefront — as 
long as the concerns of the other group are also met. 

The study also has conceptual importance (Kaplan 
1978b), in that it is just as important for many people 
to have the park there as a resource as it is to 
physically partake of it. This type of enjoyment and 
satisfaction cannot be assessed by a user count. While 
the satisfaction of "having it there" was greater for 
the workers who use it more often, this was the source 
of greatest satisfaction for users and nonusers alike. 
The park is passed by hundreds of people each day as 
they drive along the major thoroughfare at its 
perimeter; hundreds more pass it on foot as they go to 
the bank, the library, the shops, or the federal 
building a block away. No doubt many of these 
passersby notice the changes in season as they are 
reflected in the park — benches high with snow, trees 
showing signs of life. Even people who rarely see or 
use the park may derive some satisfaction from 
knowing the city has provided such a place and that it 
is there should one want to visit it. 

Another conceptual facet of the study revolves 
around people's reaction to the fact that the public 
had been involved in the planning of the park. The 
response to an open-ended questionnaire item on this 
topic was overwhelmingly enthusiastic, with "good 
idea," "great," and "excellent" the most common 
responses. The knowledge that the public had been 
involved was warmly received, which should be a cue 
to park planners in many situations. 



LITERATURE CITED 

Kaplan, R. 1978a. Participation in environmental 
design: some considerations and a case study. In 
Humanscape: environments for people, p. 427-438. 
S. Kaplan and R. Kaplan, eds. Duxbury, Belmont, 
California. 

Kaplan, R. 1978b. The green experience. In Human- 
scape: environments for people, p. 186-193. Kaplan 
and R. Kaplan, eds. Duxbury, Belmont, California. 



10 



APPENDIX 



SAMPLE SIZE FOR ANALYSES 

Each of the analyses presented in tables 2-6 was 
based on a somewhat different number of partici- 
pants, depending on how many people answered a 
particular question. The maximum and minimum 
number of people involved in each specific analysis is 
presented here. (Some of these analyses did not 
appear in the tables but were mentioned in the 
discussion.) 



Categories 

Sample On-site/Off-site 

Age Under 30/30 & Over 

Sex Women/Men 
Time in park: 

Live downtown Weekly/Less 

Work downtown Weekly/Less 

"Other" Weekly/Less 

Bike Occasionally/Less 

Foot Often/Less 

Car Often/Less 

Live/Work Live/Work/Other 
Student 

(age in 20's) Yes/No 
See Park 

(Work downtown) Yes/No 



Maximum 


Minimum 


160-222 


124-177 


240-126 


209-92 


208-141 


165-116 


43-39 


38-33 


90-57 


77-43 


42-60 


31-47 


68-248 


61-202 


206-167 


170-130 


173-177 


145-140 


90-177-108 


76-141-87 


95-119 


82-96 


61-119 


50-90 



11 



ACKNOWLEDGMENTS 

The author would like to thank the following organizations and individu- 
als for their contributions to this study: The Ann Arbor Parks and 
Recreation Department, for providing staff assistance and supplies; 
George R. Owers and Thomas Raynes of that department for their interest 
and cooperation; the Urban Forestry Unit of the USDA Forest Service's 
North Central Forest Experiment Station, for financial support; John 
Dwyer, of the Urban Forestry Unit, for his help and encouragement; and 
Charles Cares and Terry Brown, the designers of Liberty Plaza, for their 
continuing participation. 



12 



Kaplan, Rachel. 

1981. Evaluation of an urban vest-pocket park. U.S. Department of 
Agriculture Forest Service, Research Paper NC-195, 12 p. U.S. 
Department of Agriculture Forest Service, North Central Forest 
Experiment Station, St. Paul, Minnesota. 

Evaluates the effectiveness of a downtown vest-pocket park in Ann 
Arbor, Michigan. 

KEY WORDS: Urban forestry, urban parks, recreation, public 
involvement. 



•& U.S.GPO:1981-766-923/197-6 



/ 



IA FOREST SERVICE 
EARCH PAPER NC-196 

3, ■.'.'/ £- /96> 




GOVT. DOCUMENTS 
DEPOSITORY ITEII 

JUN \ 1981 

CLEMSOftf 

L«KAfl& 



An economic and energy analysis of 

poplar intensive cultures 

in the Lake States 

Dietmar Rose, Karen Ferguson, David C. Lothner, and J. Zavltkovskl 



Mh Central Forest Experiment Station 

F est Service, U.S. Department of Agriculture 



PESTICIDE PRECAUTIONARY STATEMENT 

This publication reports research involving pesticides. It does not contain 
recommendations for their use, nor does it imply that the uses discussed here 
have been registered. All uses of pesticides must be registered by appropriate 
State and/or Federal agencies before they can be recommended. 

CAUTION: Pesticides can be injurious to humans, domestic animals, desir- 
able plants, and fish or other wildlife — if they are not handled or applied 
properly. Use all pesticides selectively and carefully. Follow recommended 
practices for the disposal of surplus pesticides and pesticide containers. 




(/upv&uabga^ 



■ i Mf'iiam or niicvLivit 



ACKNOWLEDGMENT 

The authors extend a special word of thanks and acknowledge 
the support of Michael Morin, Packaging Corporation of Amer- 
ica, Filer City, Michigan. He gave generously of his time, pro- 
vided detailed production information, reviewed our work as it 
progressed, and provided excellent counsel. 



North Central Forest Experiment Station 

Forest Service — U.S. Department of Agriculture 

1992 Folwell Avenue 

St. Paul, Minnesota 55108 

Manuscript approved for publication March 3, 1980 

1981 



AN ECONOMIC AND ENERGY ANALYSIS OF POPLAR 
INTENSIVE CULTURES IN THE LAKE STATES 1 



Dietmar Rose, Professor, 

College of Forestry, University of Minnesota, 

St. Paul, Minnesota, 

Karen Ferguson, Graduate student in Business Administration, 

College of Forestry, University of Minnesota, 

St. Paul, Minnesota, 

David C. Lothner, Economist, 

North Central Forest Experiment Station, 

Duluth, Minnesota, 

and J. Zavitkovski, Principal Ecologist, 

North Central Forest Experiment Station, 

Rhinelander, Wisconsin 



Interest in the use of intensive silviculture to pro- 
duce high yields of wood fiber in short rotations has 
been increasing. This trend is not really a result of 
foreseen shortages in hardwoods — nationally, 
hardwood removals are still well below growth 
(USDA Forest Service 1973 ) — but instead reflects an 
increased awareness of the potential economic ad- 
vantages of growing wood fiber intensively. Through 
concentration of high yields in small areas close to 
the mill or plant, intensive short-rotation culture 
might remove many of the uncertainties connected 
with fiber supply from small private woodland and 
public land. It would also reduce hauling costs, an 
increasingly important consideration due to the en- 
ergy situation. The actual application of short rota- 
tion intensive wood production systems to supply 
biomass for energy, fiber for pulp mills, or both, will 
depend on the economic and energy efficiencies of 
alternative production systems available for produc- 
ing wood biomass. 

The purpose of this paper is to present a close look 
at the economic and energy efficiencies of intensively 
growing hybrid poplar in the Lake States in a man- 
ner likely to be used by industrial users of wood fiber. 
Realistic information with respect to all the relevant 
production costs and biological growth response in- 
formation are becoming increasingly available. Thus 
our intent is to make as specific an analysis as possi- 
ble without having a particular firm or location in 



Study was funded through Cooperative Research 
Agreement No. 13-607 with the North Central Forest 
Experiment Station. 



mind, in an attempt to help bridge the gap between 
theory and practice in intensive silviculture. Cash 
costs and returns of specific production systems are 
estimated in both dollar amount and relative uncer- 
tainty and are evaluated using cash flow techniques. 
The sensitivity of the investment performance mea- 
sures (e.g., internal rate of return, net present worth ) 
to various factors of production and yields is carefully 
evaluated. Detailed sensitivity tables are provided 
that will permit the manager or analyst to change 
values for any of the production factors and 
determine their impact on the financial performance. 
Thus we believe that this study may be of great 
utility in forest management planning and in further 
research. Energy requirements and outputs are esti- 
mated and contrasted with cash flows to identify 
critical cost/energy trade-offs. 



RECENT ECONOMIC 

INVESTIGATIONS OF 

INTENSIVE CULTURE SYSTEMS 

An industry-wide survey was conducted in 1975 to 
evaluate trends of and needs for intensive culture on 
forest industry land and to determine potential im- 
pacts on future wood supplies (DeBell 1976, DeBell et 
al. 1977, Gansner et al. 1977). The survey showed 
that intensively cultured wood from industrial land 
is not expected to increase total annual wood harvest 
in the North by more than 2 percent in the next 
decade. Lack of knowledge on intensive culture and 



the great uncertainties related to intensive manage- 
ment systems were major reasons that industries had 
not adopted intensive culture. Furthermore, many 
industries were not experiencing shortages of poplar 
supplies. 

Several conferences on intensive silviculture dur- 
ing the last few years have brought together existing 
knowledge on the many different facets of intensive 
silviculture. But many uncertainties remain and 
only actual operational testing of specific alterna- 
tives can provide the final answer to many questions 
(Iowa State University 1975, 1976). 

Several studies have dealt with economic ques- 
tions of intensive cultures. DeBell and Harms (1976) 
identified cost factors associated with intensive cul- 
ture of short-rotation forest crops. Their study 
showed that intensive silviculture will be expensive 
and therefore, must be evaluated carefully. Dutrow 
and Saucier (1976) reassessed the economic implica- 
tions of short-rotation systems of coppicing sycamore 
for silage. They concluded that only industrial land- 
owners would find production profitable. 

Rose and DeBell (1978) in their analysis of inten- 
sive cultures emphasized the sensitivity of major in- 
vestment performance measures to wide changes in 
costs and yields to obtain insight into the economic 
feasibility of intensive culture. Although the study 
revealed critical areas such as land cost, site prepara- 
tion, planting, harvesting, and fertilization, that 
should be addressed before implementing an inten- 
sive culture system, it was not designed to provide 
conclusive answers. In other studies, Rose found that 
long-rotation alternatives offer better investment 
opportunities than short ones but that irrigation does 
not appear economical (Rose and Kallstrom 1976, 
Rose 1977). 



Intensive silviculture has received attention re 
cently as a possible way to produce large quantities o: 
biomass for energy (Inman et al. 1977, Fege et al. 
1979, Rose 1977, Zavitkovski 1979). However, much, 
of the economic work in this area must be viewedj 
with caution because the analyses deal with untested; 
systems. Typically, assumptions concerning yield ap-: 
pear far too optimistic and various costs, especially; 
harvesting, are probably much too low. One otheri 
major criticism is the lack of adequate sensitivity, 
analyses to identify factors critical for economic pro-; 
duction and to deal with questions of uncertaintyi 
that surround these untried production systems* 
(Rose 1977). 



STUDY DESCRIPTION 
Scope 

Many alternative methods for the intensive pro- 1 
duction of hybrid poplar are possible. The major vari- 
ables of alternative systems are spacing, rotation: 
length, and cultural practices (including site prepa- 
ration, weed control, irrigation, and fertilization). 
Spacings have been proposed that range from 1 foot 
by 1 foot (0.3 m by 0.3 m) to 12 feet by 12 feet (3.6 m by 
3.6 m). Proposed rotations range from 4 to 15 years. 

In this study we evaluated four specific production 
systems that represent a range of spacings, rotations, 
and cultural practices. Two spacings, 4 feet by 4 feet 
(1.2 m by 1.2 m) and 8 feet by 8 feet (2.4 m by 2.4 m), : 
and three rotations, 5, 10, and 15 years, were chosen; j 
irrigation and fertilization were treated as options, 
whereas site preparation and weed control were as- • 
sumed the same for all four alternatives (table 1). 



Table 1. — Specifications for four intensive -culture alternatives 



Alternative 


Spacing 


Rotations 


Origin of 
stands 


Irrigation & 
fertilization 


Yield 1 














(Dry tons/acre/year) 


mt/ha/year 


1 


4 by 4 ft 


(1) 10 yrs 


cutting 


yes 


6.3 


14.1 




(1.2 by 1.2 m) 


(4) 5 yrs 


coppice 




7.2 


16.2 


2 


4 by 4 ft 


(1) 10 yrs 


cutting 


no 


3.2 


7.2 




(1.2 by 1.2 m) 


(4) 5 yrs 


coppice 




3.6 


8.1 


3 


8 by 8 ft 
(2.4 by 2.4 m) 


(1) 15 yrs 
(1) 15 yrs 


cutting 
coppice 


yes 


6.3 


14.1 


4 


8 by 8 ft 
(2.4 by 2.4 m) 


(1) 15 yrs 
(1) 15 yrs 


cutting 
coppice 


no 


3.2 


7.2 


1 Stem and branchwood, including bark. 


(See table 1 , Appendix 1 


)■ 









The size of operation for each alternative is 1,000 
acres of cleared, marginal agricultural land arranged 
in 10 tracts of from 80 to 120 acres each. All 1,000 
acres are put into production at the same time, i.e., 
the analysis does not deal with a sustained yield 
operation. The methods of site preparation, planta- 
tion establishment, and weed control are a combina- 
tion of chemical and mechanical means and are the 
same for each alternative. The method of irrigation is 
a traveling gun system (one system per tract) that 
applies 10 inches per acre annually. Fertilization 
includes only nitrogen additions applied in liquid 
form through the irrigation water at an annual rate 
of 110 kg per hectare (100 lbs per acre). Harvesting 
methods are whole-tree chipping for the 10- and 
15-year rotations, and forage-type mechanized har- 
vesting for the 5-year coppice rotations. Financial 
considerations common to all alternatives are ad- 
ministrative costs, insurance, land purchase and 
sale, equipment costs, and taxes (property and in- 
come). An annual inflation rate of 5 percent and a 
discount rate of 10 percent are applied to all costs and 
returns (Appendices 1 and 2 give a complete explana- 
tion of the inputs and outputs of the alternative pro- 
duction systems). 



Methods 

Specific inputs and outputs (both physical and 
monetary) of each productive system were identified 
through consultation with the USDA Forest Service 
researchers at the North Central Forest Experiment 
Station, Forestry Sciences Laboratory, in Rhinelan- 
der, Wisconsin, and with the forester in charge of 
what is currently the only industrial large-scale ap- 
plication of hybrid poplar intensive culture in the 
Lake States, in Filer City, Michigan. Along with 
dollar estimates of each cost and return (Appendices 
1 and 2), an estimate was made of the relative uncer- 
tainty of each cost and return item (Appendix 1). 

Costs and returns for each alternative system were 
evaluated for a 30-year period using simple cash flow 
techniques (Appendix 2). 

Direct energy inputs and outputs were evaluated 
in two ways: (1) using cash flow methods, substitut- 
ing energy units for dollars and discounting future 
energy flows at 10 percent annually, and (2) directly 
comparing inputs and outputs without discounting 
the future. Our rationale for discounting energy 
flows is that it represents a means for comparing the 
timing and risk involved in using energy inputs of 



known practical value (petroleum, electricity) for dif- 
ferent energy production schemes. For example, a 
barrel of oil can be invested today into producing 
more energy in the form of equipment for mining or 
for growing trees. Certainly the timing and risks of 
energy outputs for the same energy inputs are differ- 
ent and need a common basis for comparison. For the 
benefit of those who do not agree with this rationale, 
energy inputs and outputs are also compared without 
discounting (Appendices 3 and 4). 



RESULTS 

Investment Performance 
Measures 

At the assumed 10 percent discount rate, none of 
the alternatives have a positive net present worth 
(table 2). The two systems using irrigation and fertil- 
ization have negative internal rates of return, and 
the two systems that do not use irrigation or fertiliza- 
tion have after-tax rates of return of 8.1 percent (Ap- 
pendix 2). Economically, the difference between 
short (5 to 10 years) and long (15 years) rotation 
alternatives is small but the difference between irri- 
gated and nonirrigated alternatives is large. 



Sensitivity Analysis 

Sensitivity analysis is a valuable tool for predict- 
ing the potential effect of changes in uncertain 
estimates of costs and returns (Appendix 1 gives a 
complete explanation of how to interpret sensitivity 
analyses). 

The sensitivity analyses in the cash flows (Appen- 
dix 2) can be used to identify the uncertain factors 
that will critically affect net present worths (NPW) 
and internal rates of return (IRR). It can also be used 
to identify conditions under which intensive cultures 
might be economically attractive. 

In each of the alternative systems, the most impor- 
tant estimate affecting investment performance 
measures is product sale value. For the irrigated 
systems, a 10 to 12 percent change in product sale 
value could change the IRR by 2 percent; for the 
nonirrigated systems, an 18 to 22 percent change 
would do the same. Product sale value has two com- 
ponents — yield and market price. A change in either 
one or both from what is estimated could substan- 
tially change the economic attractiveness of an in- 
vestment in hybrid poplar, and such changes are 



Table 2. — Investment performance of four intensive -culture alternatives 



Alternative 



Description 



Internal rate 
of return 



Net presen 
worth 



4- by 4-foot (1.2-m by 1.2-m) spacing, 

irrigated and fertilized, short 

rotations (5 to 10 years) 

4- by 4-foot (1.2-m by 1.2-m) spacing, 

not irrigated or fertilized, short 

rotations (5 to 10 years) 

8- by 8-foot (2.4-m by 2.4-m) spacing, 

irrigated and fertilized, long 

rotations (15 years) 

8- by 8-foot (2.4-m by 2.4-m) spacing, 

not irrigated or fertilized, long 

rotations (15 years) 



(percent) 
-0.4 



8.1 



-1.6 



8.1 



($/acre) 
-2003.82 



-236.78 



-2149.51 



-200.30 






3 
h 



likely. The most likely direction of change in yield is 
downward (see the discussion on risks, below). The 
direction of change in the market prices for whole- 
tree chips 10 to 30 years from now is a matter of 
speculation. 

Irrigation operating costs are next in importance 
for the irrigated systems. They would have to be 
substantially reduced, however, to make these proj- 
ects look even remotely attractive. Because a large 
part of the cost of operating a traveling gun irrigation 
system is due to fuel, a substantial cost reduction in 
this irrigation system in the future is unlikely. The 
only way reduction in irrigation costs appears likely 
is to use different irrigation technology (trickle irri- 
gation), only irrigate during the first few years of 
each rotation, or irrigate a lesser amount each year 
which would likely reduce yields and perhaps cancel 
out the effect of cost reductions through reductions in 
product sale value. Irrigation equipment costs are of 
some importance but to a lesser degree than operat- 
ing costs. 

Our cost estimates for irrigation were derived in 
coordination with the University of Minnesota Agi- 
cultural Extension specialist for irrigation and are 
believed to be valid for Minnesota operations. It is 
known, however, that large differences can occur 
from one region to another and between different 
irrigation systems and power units. Sheffield (1979) 
compared both the fixed and variable costs for the 
four major types of power units used for irrigation 
pumping plants and found that natural gas is the 
most economical source of irrigation pumping. Elec- 
tricity is the second most economical source. Shef- 
field ( 1979) points to a recent shift to propane or LPG 



• 



t 
in 

(Liquid Propane Gas), a by-product of the oil refining 
process, that has shown relatively stable prices ir 
recent months. Diesel fuel is the most expensive waj 
to irrigate, but nonetheless is the most widely used ir' 
the Lake States. 

Reductions in the variable cost of up to 20 percen 
appear possible by switching to electric and propan 
fuel sources and reductions of possibly 50 percent by 1 
switching to natural gas. Fixed costs also might be 
reduced as much as 50 percent by switching to elec- 
tric power, if power lines are available at the irriga- 
tion site. 

If costs could be reduced 50 percent both in the. 
variable as well as fixed cost components of irriga- 
tion, it would not make NPW positive although it 
would increase NPW for both irrigation alternatives' 
by $1,135.50. A thorough analysis of the circum- 
stances for irrigation is essential in each specific 
production situation. 

Harvesting, especially whole tree harvesting, is a 
significant cost for all systems. Roughly a 20 to 40 
percent change in harvesting costs would change the 
rate of return by 2 percent for the long rotation sys- 
tems because they would use only the whole-tree 
method. Forage-type mechanized harvesting costs, 
though much more uncertain, are not nearly as im- 
portant to the financial appearance of the short rota- 
tion alternatives. Both types of harvesting employ 
new technology and their costs depend on many vari- 
able stand and terrain factors, which makes future 
cost predictions difficult. However, even though 
these costs are uncertain, they are important to the 
economic performance of any hybrid poplar invest- 
ment. 






The other uncertain estimates — fertilization, land 
sale value, and income tax — do not significantly af- 
fect investment performance. 



both conditions, i.e., need for fertilization in nonirri- 
gated crops and lower irrigation energy inputs, irri- 
gated and nonirrigated systems would be about 
equally energy efficient. 



Energy Flows 

Energy output was measured in terms of gross 
heat value, which does not account for losses during 
conversion of the biomass into other forms of energy. 
Discounted and nondiscounted energy flows were 
used to calculate energy output/input ratios (table 3). 
The two irrigated systems have lower output/input 
ratios than the two nonirrigated systems. (Details on 
the energy analysis in terms of energy inputs and 
outputs are found in Appendix 3.) 

These comparisons are valid for the specific as- 
sumptions of this analysis. If, for example, fertiliza- 
tion and associated energy inputs are considered 
necessary for nonirrigated systems, the nondis- 
counted energy output/input ratios would drop. Nat- 
urally, yields or energy outputs might also increase 
and counteract a decline in the ratios. For irrigated 
systems energy inputs for irrigation might be lower 
because of an alternative system used or a less inten- 
sive irrigation schedule. For example, a 50 percent 
reduction in irrigation energy inputs would increase 
the energy output/input ratios by about 0.45. Under 



DISCUSSION 

A Proper Perspective in the 

Economic Evaluation of Intensive 

Culture Investments 

From the viewpoint of an industrial user of wood, 
the economic performance of an investment in hybrid 
poplar does not mean much in isolation, but must 
instead be compared with alternative investments in 
other sources of supply. The real value of intensive 
culture is not its return on investment (though for 
nonirrigated systems, an 8 percent after-tax return is 
not bad compared to other forestry investments), but 
its ability to provide a secure source of supply to a 
mill or plant that would be very costly to shut down. 
For a particular firm, its location, access to wood 
supplies (including present or potential environmen- 
tal regulations and restrictions), and the amount of 
competition it must face for the wood supply are more 
important factors to consider than the economic per- 
formance measures of intensive culture investments, 
although these can be used as guides in choosing 
between particular investment opportunities. 



Table 3. — Net present worths (NPW) and output/input (Oil) ratios of energy flows under two discount rates (10 
percent and percent), when energy output is the gross energy value of wood 







Net present worth 


Output/input 


Alternative 


Description 


(MMBTU's/acre) 1 


ratio 2 






10 percent 


percent 


10 percent 


percent 


1 


4- by 4-foot spacing, irrigated 
and fertilized, short rotations 
(5 to 10 years) 


453.42 


2,346.82 


2 62 


3.08 


2 


4- by 4-foot spacing, not irrigated 
or fertilized, short rotations 
(5 to 10 years) 


285.08 


1,362.00 


4.50 


4.61 


3 


8- by 8-foot spacing, irrigated 
and fertilized, long rotations 
(15 years) 


251.60 


2,129.80 


2.15 


3.04 


4 


8- by 8-foot spacing, not irrigated 
or fertilized, long rotations 
(15 years) 


184.74 


1,254.10 


4.64 


4.76 



1 Net present worth at percent discount rate equals the sum of the returns minus the sum of the costs. Net present worth at 10 percent discount rate obtained 
from energy flow analysis in Appendix 4. 

2 0utput/input ratio at percent discount rate equals the sum of the returns divided by the sum of the costs. Output/input ratio at 1 percent discount rate obtained 
from investment performance measures in Appendix 4. 






Risks and Uncertainties of 
Intensive Fiber Production 

High yields are the most attractive feature of in- 
tensive culture systems. Any reduction in yield is 
therefore significant and deserves careful considera- 
tion in any decision regarding investment in inten- 
sive culture systems. Yields may be lower than we 
have predicted for three major reasons: (1) yield data 
have been reported only for carefully tended research 
plots grown for short periods, not for long-term opera- 
tions; (2) risks from insects and disease damage are 
significant — they may reduce annual growth or even 
destroy entire portions of a crop; (3) yields for nonirri- 
gated, nonfertilized crops are speculative because lit- 
tle data are available. The lack of irrigation and 
fertilization may reduce annual growth, as we have 
assumed, or may make the difference between suc- 
cess or failure of the crop during the establishment 
period. Yields from nonirrigated crops are no more or 
less certain than those from irrigated crops; insects 
and disease are probably the greatest sources of risk, 
whether the crop is irrigated and fertilized or not. 

According to some experts in the field, fertilization 
is a must for short-rotation intensive culture. If we 
applied the fertilizer regime described for the irri- 
gated alternatives to the nonirrigated production al- 
ternatives, NPW would be reduced by about 
$200/acre. An offsetting increase in yield may occur, 
however. With all other assumptions unchanged, 
this would not change the ranking of the alterna- 
tives. In combination with other changes such as the 
discussed lower cost irrigation alternatives, the 
ranking of the alternatives could conceivably change 
in favor of the irrigation alternatives. This, however, 
does not make them financially attractive. 

Even small yield reductions can have substantial 
impacts on returns and the overall economic perfor- 
mance of an intensive culture project. Uncertainty 
about yields combined with other financial uncer- 
tainties (irrigation and harvesting costs and market 
value of the product) that can affect economic perfor- 
mance, means that an investment in the intensive 
culture of hybrid poplar must be regarded as having 
substantial risk and evaluated accordingly. 



Irrigation and Fertilization 

Even with optimistic yield estimates, irrigation 
and fertilization are economically unattractive due 
largely to the high cost of operating irrigation equip- 
ment. Fertilization may be an economical way to 



increase yields, though from our analyses this is diffi- 
cult to evaluate because the method of application is 
not included in its cost. Irrigation, however, seems to 
be clearly uneconomical. 

Nor is irrigation energy efficient. The net energy 
value when energy output is measured as gross or 
usable energy is higher for irrigated systems, but the 
output/input ratio, a measure of efficiency, is lower. 
In terms of using wood fuel to produce electrical en- 
ergy, irrigated systems produce only a little more 
energy than they use in the production process. 

Not irrigating means lower yields. However, other 
things being equal, nonirrigated yields would have to 
be reduced to less than 10 percent of irrigated yields 
(roughly V2 dry ton/acre/year) before the NPW of non- 
irrigated systems would decline to that of irrigated 
systems. This is not likely. On the other hand, not 
irrigating would make site selection more important 
to avoid losing an entire crop due to drought during 
the establishment period. And such sites might be 
more expensive. 

If irrigation costs can be reduced 50 percent or 
more by switching to other types of power or by irri- 
gating less frequently, the irrigated systems would 
more closely compete economically with nonirrigated 
systems. 

Situations might exist in which larger blocks of 
land can be obtained. Each traveling irrigation gun 
system can handle up to about 300 acres so the fixed 
cost could possibly be reduced by 67 percent. How- 
ever, this reduction, even in combination with a sub- 
stantial reduction in the variable cost of irrigation, 
would still not make the irrigated alternatives more 
attractive than the nonirrigated ones. 



Short vs. Long Rotations 

The short- (5 to 10 years) and long- (15 years) 
rotation alternatives we have looked at differ little in 
terms of economic return and energy efficiency. Each 
has advantages and disadvantages. Short-rotation 
production systems return revenues sooner and more 
frequently. The wood can be harvested with forage 
harvesting methods that are as yet undeveloped but 
may be less expensive than traditional methods. The 
crop is carried for shorter periods of time so the risk ol 
losing a crop is not as great nor would the loss be as 
severe. Short rotations also allow managers to more 
quickly incorporate yield improvements, resulting 
from new genetically improved hybrids, into the 
plantation operation. However, the type of wood pro- 
duced may not be usable for all purposes because it 



contains more juvenile wood and bark than conven- 
tional chips. Long-rotation crops produce wood of a 
more conventional form that can be harvested with 
proven methods. Planting costs are much lower, but 
the crops must be carried for long periods so revenues 
are returned later and less often than with short- 
rotation crops. 

Any initial decision about short or long rotations 
can be changed as questions are answered about in- 
sect and disease risks, uses for wood fiber from young 
trees, and harvesting technology. Flexibility in rota- 
tion length is one advantage of growing wood because 
it is a crop that can be stored on the stump. 



made than irrigated systems so no trade-off is in- 
volved. If gross energy is of concern, cost/energy 
trade-offs could be made (table 3). 

Our view is that energy efficiency is the more im- 
portant criterion. Producing wood biomass without 
irrigation (and fertilization) is more energy-efficient 
and economical. The short- (5 to 10 years) and long- 
(15 years) rotation alternatives (both nonirrigated) 
differ little either in economic performance or energy 
efficiency. 



CONCLUSIONS 



Economies of Scale 

The cost estimates used in our analyses were for 
the most part variable, which does not allow us to 
make a quantitative estimate of how the overall in- 
vestment might look on a different scale of operation. 
We examined investment alternatives as solitary 
projects, not as sustained yield operations, as would 
be the most likely practice. For a sustained yield 
operation of this size (planting 1,000 acres per year), 
overhead costs such as administrators, full-time em- 
ployees, a nursery, equipment storage and repair, 
etc., would certainly increase. However, the most 
important costs and returns in terms of overall eco- 
nomic performance (product value and harvesting) 
would not change. 

Another important consideration, whether the op- 
eration is viewed as sustained yield or not, is the size 
and distribution of tracts. One large block of 1,000 
acres would be less costly to prepare, plant, harvest, 
and irrigate, though such a block of land near a mill 
and for sale would be difficult to find in the Lake 
States. The location of tracts, in relation to each other 
and to the mill, is probably more important than 
their individual sizes, because this would affect costs 
of moving equipment (site preparation, planting, and 
harvesting) and administration. Economical tract 
sizes would depend primarily on harvest costs be- 
cause ( 1 ) these are significant in the overall economic 
outlook of a project, and (2) certain fixed costs of 
putting in landings and skid roads are necessary for 
every tract regardless of size. 



Cost/Energy Trade-offs 

Nonirrigated systems are more energy efficient 
and also more economical under the assumptions 



Intensive culture of hybrid poplars in the Lake 
States with our estimates carries substantial risks 
and does not yield high monetary returns. Intensive 
culture projects are primarily of interest to industrial 
users of wood fiber, who can compare them with other 
sources of supply before making investment deci- 
sions. 

Nonirrigated production strategies are recom- 
mended as long as users find our assumptions about 
irrigation regimes, costs, and obtainable land tract 
sizes acceptable. Under the conditions specified, irri- 
gating hybrid poplar does not appear economical nor 
energy efficient. 

Short- (5 to 10 years) and long- ( 15 years) rotations 
differ little in terms of economics and energy effi- 
ciency. In view of the uncertainties of some costs and 
returns that may dramatically affect the economic 
outlook of a project (specifically, product value, irri- 
gation, need for fertilization, and harvest costs), any 
initial decision about rotation length for a particular 
project should be regarded as tentative. If a careful 
site-specific investigation into irrigation technology 
and costs and available tract sizes can reveal a more 
favorable cost picture than we assumed, irrigation 
alternatives could be more attractive. 



LITERATURE CITED 

Blankenhorn, P. R., T. W. Bowersox, and W. K. Mur- 
phy. 1978. Recoverable energy from the forests, an 
energy balance sheet. TAPPI 61(4):57-60. 

Bowersox, T. W., and W. W. Ward. 1976. Economic 
analysis of a short-rotation fiber production system 
for hybrid poplar. Journal of Forestry 74:750-753. 

DeBell, D. S. 1976. Intensive culture on industrial 
forest lands and future wood supplies. TAPPI 
59:13. 



DeBell, D. S., and J. C. Harms. 1976. Identification of 
cost factors associated with intensive culture of 
short-rotation forest crops. Iowa State Journal of 
Research 50:295-300. 

DeBell, D. S., A. P. Burnette, and D. L. Schweitzer. 
1977. Expectations from intensive culture on in- 
dustrial forest lands. Journal of Forestry 75:10-13. 

Dutrow, G. F., and J. R. Saucier. 1976. Economics of 
short-rotation sycamore. U.S. Department of Agri- 
culture Forest Service, Research Paper SO-114, 16 
p. U.S. Department of Agriculture Forest Service, 
Southern Forest Experiment Station, New Or- 
leans, Louisiana. 

Eidman, V., C. Dubbins, and H. Schwartz. 1975. The 
impact of changing energy prices on net returns, 
production methods, and kilo-calories of output for 
representative irrigated farms. Professional Paper 
230, 14 p. Agricultural Experiment Station, Okla- 
homa State University. 

Ek, A. R., and D. H. Dawson. 1976. Yields of inten- 
sively grown Populus — actual and projected. In 
U.S. Department of Agriculture Forest Service, 
General Technical Report NC-21, p. 5-9. U.S. De- 
partment of Agriculture Forest Service, North 
Central Forest Experiment Station, St. Paul, Min- 
nesota. 

Fege, A. S., R. E. Inman, and D. J. Salo. 1979. Energy 
farms for the future. Journal of Forestry 77:358- 
361. 

Gansner, D. A., O. W. Herrick, and D. W. Rose. 1977. 
Intensive culture on northern forest-industry 
lands: Trends, expectations, and needs. U.S. De- 
partment of Agriculture Forest Service, Research 
Paper NE-371, 8 p. U.S. Department of Agriculture 
Forest Service, Northeastern Forest Experiment 
Station, Broomall, Pennsylvania. 



Inman, R. E., D. J. Salo, and B. I. McGurk. 1977. 
Silvicultural biomass farms. Mitre Technical Re- 
port 7347, 123 p. Vol. IV: site-specific production 
studies and cost analyses. Mitre Corporation, 
Washington, DC. 

Iowa State University. 1975. Conference on intensive 
culture of forest crops. Iowa State Journal of Re- 
search 49:263-352. 

Iowa State University. 1976. Conference on intensive 
culture of short-rotation forest crops. Iowa State 
Journal of Research 50:267-315. 

Rose, D. W. 1977. Cost of producing energy from wood 
in intensive cultures. Journal of Environmental 
Management 5:23-35. 

Rose, D. W., and D. S. DeBell. 1978. Economic assess- 
ment of intensive culture of short-rotation 
hardwood crops. Journal of Forestry 76:706-711. 

Rose, D. W., and R. D. Kallstrom. 1976. Economic 
feasibility of intensive culture. In Intensive planta- 
tion culture: 5 years research. U.S. Department of 
Agriculture Forest Service, General Technical Re- 
port NC-21, p. 96-108. U.S. Department of Agricul- 
ture Forest Service, North Central Forest Experi- 
ment Station, St. Paul, Minnesota. 

Sheffield, L. I. 1979. Energy.. .is it the Achilles heel 
for irrigated agriculture? Irrigation Age 14:32- 
33,43,50. 

University of Minnesota. 1978. Water sources and 
irrigation economics. Miscellaneous Report 150- 
1978, 76 p. Agricultural Experiment Station, St. 
Paul, Minnesota. 

U.S. Department of Agriculture, Forest Service. 
1973. The outlook for timber in the United States. 
Forest Resource Report 20, 367 p. Washington, DC. 

Zavitkovski, J. 1979. Energy production in irrigated, 
intensively cultured plantations of Populus 'Tristis 
1' and jack pine. Forest Science 25:383-392. 



APPENDIX 1.— DETAILS OF CASH FLOW ANALYSIS 



The following discussion details and documents 
the assumptions and numbers used in our analysis of 
alternative investments in hybrid poplar planta- 
tions. After the discussion of cash flow methodology 
and basic assumptions about the projects, an expla- 
nation of each item in the cash flow printouts is given 
(Appendix 2). Those interested in these details may 
not agree with every estimate, but should find it easy 
to understand our reasoning, to make changes, and to 
assess the effect of these changes through the sensi- 
tivity analysis tables. 



Cash Flow Methodology 

The various intensive culture alternatives were 
analyzed using cash flow analysis and charted in 
cash flow tables (Appendix 2). The cash flow table is 
primarily a listing of the amounts of costs and 
benefits for each year in the production period. In 
addition, the cash flow presents the total costs, total 
benefits, and net benefits for each year. The table, 
therefore, contains the data necessary to calculate 
several measures of project performance. Given a 
specified discount rate, four useful intermediate 
measures can be calculated and included in the cash 
flow table: discounted net benefits, cumulative dis- 
counted net benefits, discounted costs, and cumula- 
tive discounted costs. The above measures are listed 
in the bottom four rows of the cash flow tables gener- 
ated by the program. 

These four intermediate measures are used to cal- 
culate several of the investment performance mea- 
sures described below. Each discounted net benefit 
equals the net benefit figure for that year (year y) 
divided by 
(1 + i) y - b , where: 

i = the rate of discount, 

y = the year in which the net benefit occurs, and 

b = the beginning year of the project (initial year 
of investment). 
Each cumulative discounted net benefit equals the 
sum of the discounted net benefits for all years up to 
and including that year (year y). Each discounted 
cost and cumulative discounted cost is determined in 
exactly the same way as discounted net returns and 
cumulative discounted net benefits, respectively, ex- 
cept that the total cost row is used instead of the net 



benefit row. Cumulative discounted net benefits 
show the incremental yearly totals of discounted net 
benefits. The final yearly total equals the net present 
worth. Cumulative discounted costs show the present 
value of all costs up through the column year. When 
using the cost-price analytical approach described 
below, the value of the cumulative discounted costs is 
important. 

Investment performance measures 

The four investment performance measures calcu- 
lated by the program are defined as follows: 

(1) Internal rate of return (IRR) is the rate of 
discount that makes the present value of benefits 
equal the present value of costs, i.e., the rate that 
makes NPW equal zero. 

e e 



2 

y = b 



B, 



(l + IRR) y - b 



2 

y = b 



(1 + IRR) V b 



(2) Net present worth (NPW) is the sum of the 
discounted net benefits for one production period. 

NFW= ' lNB ' 



2 

y = b 



;i + i)- v ~ b 



(3) Net future worth (NFW) is the sum of the 
compounded net benefits for one production period. 



NFW = 



(NBL (l + ir-^NPW (l + i) 



y = b 



(4) Benefit-cost (B/C) is the ratio of the total 

discounted (gross) benefits and total discounted costs. 

e 

B v 



2 



(l + i) y - b 



B/C = 



C, 



y=b 



(l + i 



,y-b 



Where: 



b = the beginning year of the project 
(initial year of investment), 

e = the ending year of the project (last 
year of benefits or costs), 

i = the rate of discount, 



y = the year designation represent- 
ing the individual years between 
year b and year e, inclusive, 
B y = total benefits in year y, 
C y = total costs in year y, and 
(NB) V = net benefit in year y, equals B y - 

a. 



Future cost (FC) equals the future value of all 
project costs and is also calculated by the program 
although it is not directly an investment perfor- 
mance measure. Future cost is calculated by multi- 
plying the cumulative discounted costs for the last 
year (year e) by (l + i) e_b . Another method of calcu- 
lating FC is as follows: 
e 



FC - 



2 

y = b 



C y (l + ir y 



Sensitivity analyses tables 

A desirable procedure in any investment analysis 
is to examine how sensitive various measures of proj- 
ect performance are to changes in costs, prices, inter- 
est rates, and other inputs, e.g., machine production 
rates, time constraints on a silvicultural activity, etc. 
The reason for such sensitivity analyses is that most 
estimtes of inputs and outputs are interval rather 
than point estimates. In other words, individual esti- 
mates have errors associated with them that might 
be expressed by putting limits of confidence around 
them. Knowledge of the sensitivity of an investment 
to the various factors is an essential part of the as- 
sessment of the risk associated with the investment. 
It gives valuable insights into what might happen if 
yields, prices, and/or costs turned out differently than 
expected. Two types of sensitivity analyses are car- 
ried out within the program automatically. Various 
points will be illustrated on the nonirrigated 4- by 4- 
foot plantation. 

First sensitivity analysis table (Appendix 2) 

The first sensitivity analysis table shows changes 
in NPW due to a percentage change (increase or 
decrease) in each benefit and cost. The magnitudes of 
the changes in NPW indicate the impacts of changes 
in the costs and benefits. The larger the number in a 
given column, the greater the impact resulting from 
a given percentage change. In our example the great- 
est impact would result from changes in product sales 
and the least impact from changes in administration 
and site preparation costs. In addition to impacts on 
the investment performance measures, specific im- 
pacts can also be calculated. For example, NPW for 



the nonirrigated 4- by 4-foot plantation was 
-$236.78. If forage harvesting cost decreased by 10 
percent, NPW would be -$215.97 (-$236.78 
+ $20.81). On the other hand, if forage harvesting 
cost increased by 10 percent, NPW would be 
-$257.59 (-$236.78 - $20.81). Cost increases and 
benefit decreases of 10 percent decrease the four 
investment performance measures by the amounts 
indicated in the table, and cost decreases and benefit 
increases of 10 percent increase the measures by the 
same amounts. Furthermore, changes in NPW re- 
sulting from changes in costs and benefits other than 
10 percent can be calculated directly from the table. 
For example, a 50 percent increase in land cost would 
lower NPW by five times the amount of a 10 percent 
land cost increase (five times $40 or $200). 

Combinations of any number of changes in the 
costs and benefits can be calculated. For example, if 
all costs were assumed to be 10 percent higher and all 
benefits 20 percent higher, NPW would be -$128.20 
(-$236.78 - $155.94 + $264.52). In other words, even 
if the above cost and benefit changes occurred, the 
investment decision would remain the same because 
the decision switching value occurs when NPW 
equals zero. 

Sometimes it is useful to know how much change is 
necessary in one or more costs and benefits to change 
the investment decision. In the example, one might 
wish to know by how much product sales values must 
increase to cause the project to be accepted. This is 
determined by dividing NPW by the corresponding 
change in that measure (from the first sensitivity 
table), and then multiplying the result by the percent 
change specified in the table ( 10 percent in the exam- 
ple). Using NPW the result is the following: 

— : — = 2.06 or a 20.6 percent increase of product 

$114.81 

sales value would make NPW = and IRR = 10 

percent. 

Second sensitivity analysis table (Appendix 2) 

The second sensitivity analysis table shows the 
percent changes in activities (costs and benefits) nec- 
essary to raise and lower IRR and NPW by specified 
amounts. This sensitivity table is of most interest 
when the IRR is close to the specified discount rate. In 
this case, it is valuable to know how likely it would be 
that the investment return could be above or below a 
desired rate of return. For the nonirrigated 4- by 4- 
foot plantation case presented in Appendix 2, the 
specified changes were 2 percent in IRR and $236.78 
in NPW. As in the case of the first sensitivity table, 
the activities are listed in the first column. Columns 



10 



2 and 3 show the percent changes in each activity 
necessary to raise and lower IRR by 2 percent. Col- 
umns 4 and 5 similarly show the percent changes in 
each activity necessary to raise and lower NPW by 
$236.78. It is apparent in this case that changes in 
product sales and land cost have the greatest influ- 
ence on IRR and NPW. For NPW the same percent 
change is always required to raise and lower the 
measure by an equal amount — only the sign changes. 
However, the percent changes necessary to raise and 
lower IRR by the same percentage amount are sel- 
dom equal. Similarly, the additive and multiplicative 
advantages associated with sensitivity testing of 
NPW are not applicable to sensitivity testing of IRR. 
Just because a 22 percent increase in product sales 
increases the IRR by 2 percent in the example, a 44 
percent increase will not necessarily increase IRR by 
4 percent. Yet, because a 21 percent increase in prod- 
uct sales will increase NPW by $236.78, a 42 percent 
increase will increase NPW by $473.56. Therefore, it 
is clear that such sensitivity testing for simultaneous 
changes in several activities is better performed us- 
ing NPW than IRR. 



Basic Assumptions 

Inflation 

We chose a 5 percent annual inflation rate, because 
economic forecasters have predicted this as an aver- 
age long-term rate. You will note three exceptions to 
this rate in the cash flow printouts: (1) land resale 
value, which we estimate will increase more rapidly 
than other prices (we have said 7 percent annually), 
(2) income tax, which already takes inflation into 
account because it is calculated from income and 
costs in the years they occur, and (3) irrigation equip- 
ment insurance, which we assumed to remain con- 
stant for a given group of equipment thus declining in 
real terms to reflect the declining value of aging 
equipment. 

Discount rate 

A 10 percent discount rate was used in each cash 
flow analysis based on an estimated cost of capital. 
We think a discount rate based on opportunity cost or 
risk should be accounted for at other stages of the 
decision making process. 

Rotations 

For the 4- by 4-foot spacing, the first rotation is 10 
years and the four subsequent coppice rotations are 5 



years. For the 8- by 8-foot spacing, both the initial 
and coppice rotations are 15 years. Total project 
length is 30 years. 

Yields 

Stem and branch wood dry biomass measured in 
tons is the unit of product yield. Biomass rather than 
roundwood is considered because the expected end 
uses of intensively grown wood are pulp and fuel. 
Yields for irrigated, fertilized, first-rotation crops 
were adapted from Ek and Dawson (1976) and are 
optimistic. Yields for nonirrigated, nonfertilized 
crops will likely be at least 50 percent less. Yields for 
the 5-year coppice rotations will be roughly l 3 A times 
the yields for 5-year first-rotation crops. We expect 
that this growth increase will be notable only in the 
first few years. Thus, the yield for the 15-year coppice 
rotation will be similar to the first rotation (table 4). 
Total project acreage is larger than the total acreage 
planted in most cases because some will always be 
lost to roads, powerlines, swamps, etc. For the analy- 
sis, the yields per planted acre had to be spread over 
total project acreage and are, therefore, smaller for 
the latter. 

We did not include insect and disease control in the 
costs and returns. Although control measures would 
almost certainly be needed at some point in an opera- 
tion of this size and length, we did not include their 
costs because they are difficult to estimate, both in 
amount and timing, and because we feel they are best 
evaluated as risks in the same way as fire and other 
uncertainties by making adjustments in the harvest- 
able yields. 



Explanation of Cash Flow Cost 
and Return Items 

Site preparation and establishment activities 
considered 

Site preparation of already cleared land for this 
hypothetical plantation includes plowing, disking, 
and applying a preplanting herbicide (Round-up and 
Simazine). All these activities take place the late 
summer or fall prior to spring planting. For the 4- by 
4-foot spacing, 3 lbs. of Round-up and 2 lbs. of Sima- 
zine/acre are applied in a broadcast application using 
a tractor-pulled sprayer. For the 8- by 8-foot spacing, 
the chemical is applied in 3-foot strips using the same 
equipment. Trees are later planted in these strips. 



11 



Table 4. — Yields of stem and branch wood for four production alternatives 



Spacing 



Irrigated 



Rotation 



Yield of stem and branchwood at end of rotation 



4- by 4-ft 
(1.2 by 1.2 m) 



4- by 4-ft 
(1.2 by 1.2 m) 



8- by 8-ft 
(2.4 by 2.4 m) 

8- by 8-ft 
(2.4 by 2.4 m) 



Yes 



No 



Yes 
No 



1st (10 yrs) 
2nd (5 yrs) 
3rd (5 yrs) 
4th (5 yrs) 
5th (5 yrs) 

1st (10 yrs) 
2nd (5 yrs) 
3rd (5 yrs) 
4th (5 yrs) 
5th (5 yrs) 

1st (15 yrs) 
2nd (15 yrs) 

1st (15 yrs) 
2nd (15 yrs) 



Dry tons/ 
planted acre 
70 
40 
40 
40 
40 

35 
20 
20 
20 
20 

105 
105 

52.5 
52.5 



mtl 
planted ha 
157 
90 
90 
90 
90 

78 
45 
45 
45 
45 

235 
235 

118 
118 



Dry tons/ 
project acre 
63 
36 
36 
36 
36 

31.5 

18 

18 

18 

18 



94 
94 



47.25 
47.25 



mtl 

project ha 

141 

81 

81 

81 

81 

71 
40 
40 
40 
40 

212 
212 

106 
106 









Postplanting establishment activities are cultiva- 
tion (3 times) using a 20-foot Lilliston 2 cultivator, 
and a fall and spring treatment of Simazine, applied 
in the same way as the preplanting Round-up and 
Simazine treatment. 



These methods are not put forth as the one best 
prescription to follow in raising hybrid poplar but 
only to explain the derivation of site preparation 
costs. This sequence was recommended by research- 
ers at the North Central Forest Experiment Station, 



2 Mention of trade names does not constitute en- 
dorsement of the products by the USD A Forest Serv- 



Forestry Sciences Laboratory in Rhinelander, Wis- 
consin, and by the Intensive Forestry manager at 
Packaging Corporation of America as being one 
method of growing poplar. 3 In practice, a variety of 
treatments would be used, depending on site condi- 
tions. Some additional or substitute treatments we 
did not consider are liming, spring disking, addi- 
tional cultivations, cover crop seeding for controlling 
vegetation, and different types of chemicals. Also, we 
assume that no additional treatments will be neces- 
sary after the first rotation because the rapid initial 
growth of a coppice crop should be sufficient to stay 
ahead of the weeds. 



ice. 



^Personal communication with M. Morin, Packag- 
ing Corporation of America, Filer City, Michigan. 



12 



As shown in the following tabulation we have rated 
each cost and return item as to how accurate we feel 
our estimates are. 



Cost/return item 

Site prep., establishment 
(operating costs) 

Planting (operating costs) 

Irrigation (operating 
costs) 

Fertilization 



Whole-tree harvest 



Forage harvest 

Land cost 
Administrative 
irrigation equipment 



Site prep., planting, 
estab. equipment 

Insurance 

Income tax 

Property tax 

Investment tax credit 

Product sale 

Land sale 



Uncertainty of estimate 

Fairly certain 

Fairly certain 
Fairly uncertain 

Fairly certain to 
uncertain 



Uncertain 



Comment 



Extremely uncertain 

Fairly certain 

Fairly certain 

Fairly certain to 
uncertain 

Fairly certain 

Fairly certain 
Uncertain 

Fairly certain 

Fairly certain 

Very uncertain 

Uncertain 



Estimates are reliable 
for the near future only. 

Depends on the type of 
fertilizer and method of 
application. Estimates 
are reliable for near future. 

Uncertainty primarily due 
to distance in the future 
and yield. No harvest 
experience with intensive 
culture stands. 

Uncertainty due to both 
lack of information and 
distance in future. 

Depends on location. 

Depends on scale of operation. 

Good estimates available 
for initial investment only, 
not future replacement. 

Depends on how cost is 
accounted for. 

Depends on interaction of 
all cost and return items. 

Estimates reliable for near 
future only. 

Depends on price of irrigation 
equipment (see above). 

Uncertainty due to yields and 
future prices. 

Rate of inflation, future 
markets unknown. 



13 



Operating costs 

Only variable costs — labor, fuel, repair, and mate- 
rials — are included. Fixed equipment costs are 
treated separately. Because only 90 percent of the 
available acreage in this hypothetical plantation is 
workable, 900 acres are used as a basis for calculat- 
ing costs. Total site preparation costs are calculated 
as follows: for each operation, labor, fuel, and repair 
costs per hour are multiplied by the time needed to 
work 900 acres. (Time = 900 acres divided by the 
production rate in acres/hour.) Time spent moving 
between tracts and for employee breaks is included in 
the production rate. Material costs/acre are multi- 
plied by 900 acres and added to the total labor, fuel, 
and repair costs to give the total cost for the opera- 
tion. This is divided by 1,000 project acres to give a 
cost/acre (table 5). 

Labor costs are $6/hour for equipment operators 
and $4/hour for planting crews. Repair costs are 
based on the American Society of Agricultural Engi- 
neers' estimates of total lifetime repair costs for farm 



equipment (table 6). We estimated diesel fuel con- 
sumption for tractors at 0.044 gal/hp and cost at 
$l/gal. Current prices for Round-up and Simazine are 
$60/gal (4 lbs/gal) and $2.40/lb, respectively. 

Planting 

Ten inch unrooted cuttings purchased at $80/thou- 
sand are planted in the spring using Holland planters 
and large tractors. For the 4- by 4-foot spacing, 2,600 
trees/plantable acre are planted at a rate of 3 A acre/ 
hour; for the 8- by 8-foot spacing, 650 trees/plantable 
acre are planted at a rate of 2 acres/hour. We have 
estimated that about 95 percent of the plantable 
acreage would be planted, the 5 percent loss being 
due to row ends, rocks, low spots, etc. To finish plant- 
ing in a reasonable length of time (maximum 10 
weeks) would require 3 tractors and 6 planters for the 
close spacing and 2 tractors and 4 planters for the 
wide spacing. Overtime would probably be necessary 
for the close spacing, though we have not accounted 
for it. 



Table 5. — Operating costs for site preparation 

















Material cost/ 


Total cost 








r Operation 


Equi 


pment costs 


Production 
rate 


Total 

hours 4 


planted 


acre 


for operation 


Cost 


acre 


Yea 


Labor Fuel 1 


Repair 2 


4- by 4-ft 


8- by 8-ft 


4- by 4-ft 


8- by 8-ft 


4- by 4-ft 


8- by 8-ft 






Dollars/hour 


Acres/ hour 1 


No. 

















Plow 


$6.00 $9.90 $10.75 


6.55 


137.4 








$3,662 


$3,662 


$3.66 


$3.66 





Disk 


6.00 


9.90 


9.50 


10.04 


89.6 








2,276 


2,276 


2.28 


2.28 





Apply Roundup 


























and Simazine 


6.00 


9.90 


5.71 


14.18 


63.5 


51.00 


15.30 


47,272 


15,142 


47.27 


15.14 




Total year 




















53.21 


21.08 


1 


Cultivate 
(3 times) 


6.00 


9.90 


5.53 


9.09 


297.0 
(3 times) 








6,365 


6,365 


6.36 


6.36 


1 


Apply Simazine 
Total year 1 


6.00 


9.90 


5.71 


14.18 


63.5 


6.00 


1.80 


6,772 


2,992 


6.77 
13.13 


2.99 
9.35 


2 


Apply Simazine 
Total year 2 


6.00 


9.90 


5.71 


14.18 


63.5 


6.00 


1.80 


6,772 


2,992 


6.77 
6.77 


2.99 
2.99 



'Gal/hour = 0.044 gal/hp x 225hp = 9.9 gal/hour. 
At Si/gal for diesel fuel, cost/hour = $9.90. 

2 See table 6. 

3 Source: Benson, F. J. 1979. Machinery cost estimates. Unpublished data on file at Agricultural Extension Service, University of Minnesota, St. Paul, 
Minnesota. 

"Production rate computed on 900 workable acres. 

5 Current price for Roundup is $60/gal and for Simazine is $2.40/lb. 



14 



Table 6. — Repair costs for site preparation and establishment equipment 





Amount of repairs over 












lifetime in relation 




Available 


Repair 


Repair 


Equipment 


to new cost 1 


New cost 


hours/year 


cost/year 


cost/hour 




Percent 










225 hp tractor 


100 


$49,040 2 


1,200 


$4,904 


$4.09 


10 bottom plow 


120 


13,870 2 


250 


1,664 


6.66 


23-foot offset disk 


120 


11.290 2 


250 


1,355 


5.41 


30-foot sprayer 


100 


1,940 2 


120 


1,940 


1.62 


20-foot Lilliston 












cultivator 


120 


3,000 3 


250 


360 


1.44 


Planter 


120 


2,000 3 


250 


240 


.96 



1 Source: American Society of Agricultural Engineers, 1976 Agricultural Engineering Yearbook, p. 329. 

2 Source: Benson, F. J. 1979. Machinery cost estimates. Unpublished data on file at Agricultural Extension Service, University of Minnesota, St. Paul, 
Minnesota. 
3 Authors' estimate. 



Materials cost per planted acre comes to $208 for 
the close spacing and $52 for the wide spacing. Due to 
the slower rate of production, fuel, labor, and repair 
costs are also higher for the close spacing. As shown 
in the following tabulation, total operating costs for 
planting are $221.94/project acre for the 4- by 4-foot 
spacing and $59.83/project acre for the 8- by 8-foot 
spacing. 



Item 


4- by 4-foot 


8- by 8-foot 




spacing 


spacing 


Equipment 


3 225 hp 
tractors 


2 tractors 




6 planters 


4 planters 


Equipment cost/hour 






(for each equipment group) 






Labor 4 


$14.00 


$14.00 


Fuel 


9.90 


9.90 


Repairs 5 


5.05 


5.05 


Total equipment cost/hour 


28.95 


28.95 


Production rate (ac/hr) 


.75 


2.00 


Total hours (for all 


1,200 


450 


equipment groups) 6 






Total equipment cost 


$34,740 


$13,028 


Plant stock cost/acre 7 


$208 


$52 


Total stock cost 


$187,200 


$46,800 


Total cost 


$221 ,940 


$59,828 


Cost/acre 


$221.94 


$59.83 



4 One equipment operator and two planters. 

5 See table 6. 

6 Each equipment group (tractor and two planters) 
works simultaneously; total hours is the additive 
number of hours for each group. 

7 Stock cost estimated to be $80/M— 2,600 trees/acre 
for close spacing and 650 trees/acre for wide spacing. 



Many other types of stock could be used: longer 
cuttings, soaked cuttings, rooted cuttings, or even 4- 
foot to 5-foot rooted stock. The planting method and 
costs would, of course, depend on the type chosen. 
Most likely a large operation would produce its own 
stock, though we have assumed that it was purchased 
elsewhere. 

Irrigation 

Irrigation is probably the most important, and the 
most costly, cultural activity involved in intensively 
growing hybrid poplar. Research predicts high yields 
using it, but the large capital investment and high 
operating cost make it uneconomical. On the same 
site in the Lake States, yields without irrigation may 
be 50 percent less than those with irrigation. Irriga- 
tion may also make the difference between survival 
and failure in a dry year. In droughty years growth of 
established plantations may be reduced by 80 to 90 
percent without irrigation. 

One of the methods recommended for poplar is a 
traveling gun system. This involves a well and pump 
that sends water through an aluminum pipe to a 
rubber hose attached to a traveling sprinkler. The 
sprinkler propels itself down a lane, dragging the 
rubber hose as it travels. The sprinkler must be repo- 
sitioned when it reaches the end of a lane. It can be 
towed easily with a small tractor. The height of the 
gun can be adjusted up to 20 feet. 

We seriously question whether this system is prac- 
tical when trees reach 30 feet. For one thing, the gun 
may not be capable of spraying above the trees. For 
another, leaves may intercept much of the water and 
prevent it from reaching the ground. The high hu- 
midity may also increase susceptibility to leaf dis- 
eases. 



15 



For our analysis we will be optimistic and assume 
that these difficulties will be resolved and that an- 
nual irrigation is possible even for the 15 year rota- 
tions. Operating costs were based on a system with a 
100-foot well, 600 gal/min sprinkler, turbine pump, 
right-angle drive, diesel power unit, 3,000 feet of 6- 
inch aluminum pipe, and 660 feet of 4-inch hose (Uni- 
versity of Minnesota 1978). Each system would be 
sufficient to irrigate from 80 to 120 acres/year 10 
times for a total of 10 inches/planted acre/year. Ten 
such systems would be needed. One 40-horse tractor 
is considered sufficient to move the sprinklers when 
needed, though, of course, in practice this would de- 
pend on the closeness and accessibility of the tracts. 
Larger systems would be possible for larger tracts, 
but the operating costs would be similar for the same 
amount of water applied. 

As shown below the largest portion of the annual 
operating cost of irrigation is due to fuel — 
$67.65/acre/year or 68 percent of the total annual cost 
of $99.74, assuming diesel fuel is used at a cost of 
$l/gal (University of Minnesota 1978). 

Item Dollars/acre 8 

Fuel $67.65 

Pump and motor (lube and repairs) 9 12.07 

Distribution system 10 11.20 

Labor 8.82 

Total $99.74 

Fertilization 

One hundred pounds of nitrogen/acre/year are ap- 
plied in liquid form (28 percent solution) in the irri- 
gation water. We have considered only the annual 
cost of the material ($80/ton = $14.29/irrigated acre 
= $12.86/project acre) because the labor cost in- 
volved is probably small. Of course, other nutrients 
may be desirable in practice and other methods of 
application are possible. 

Whole -tree harvesting 

Harvesting intensively cultured wood is a new and 
untried concept. Whole-tree chipping, a highly pro- 
ductive, highly mechanized system, appears the most 



8 'Assuming 1 system per 100 acres. 

^Includes lubrication of pump, drive, power unit 
estimated as a percentage of fuel costs, and mainte- 
nance and repair of these units calculated as a per- 
centage of the initial investment. 

1 Includes the cost of the tractor required to move 
the gun, the cost of operating the power unit on the 
gun, and cost of the maintenance and repairs of the 
distibution system estimated as a percentage of initial 
investment. 



promising for wood larger than 6 inches d.b.h. We 
assume this method will be used for the 10- and 15- 
year rotations. In this system trees are severed and 
accumulated with feller-bunchers then grapple- 
skidded to a portable chipper that reduces them to 
chips. The chips are blown into vans for transport to 
the mill. 

Because field trials of whole-tree harvesting of in- 
tensively cultured wood are not yet possible, a com- 
puter simulation has furnished a rough estimate of 
harvesting costs (Mattson 1976). The simulation 
model uses nearly the same yield and spacing as- 
sumptions we have. For the 4- by 4-foot spacing, two 
70 hp skidders were used, and for the 8- by 8-foot 
spacing, a single 110 hp skidder was used. For both 
systems, a tracked feller-buncher and a small chipper 
were used. 

It has been concluded that harvesting costs are 
dependent on tree size. The 4- by 4-foot spacing prod- 
uces trees with an average diameter of 6 inches and 
harvest costs are estimated to be about $18/dry ton 
(including hauling). Twelve and one-half inch trees 
are expected from an 8- by 8-foot spacing in 15 years 
at an estimated cost of $14/dry ton (table 7). 

Predicting accurate harvesting costs for inten- 
sively grown poplar today would be difficult; 
predicting them 10 to 30 years from now is nearly 
impossible. Although they are one of the most signifi- 
cant costs of production, we consider them uncertain. 



Table 7. — Harvest costs 

(In dollars/acre) 
4- BY 4-FOOT SPACING (1.2- by 1.2-m) 



Year of Dry tons/acre 2 


Harvest cost/acre 


harvest Irrigated Nonirrigated 


Irrigated 


Nonirrigated 


10 63 31.5 
15 36 18 
20 36 18 
25 36 18 
30 36 18 


1,134.00 
288.00 
288.00 
288.00 
280.00 


567.00 
144.00 
144.00 
144.00 
144.00 


8- BY 8-F00T SPACING (2.4- by 2.4 


-m) 3 


15 94.5 47.25 
30 94.5 47.25 


1,323.00 
1,323.00 


661.50 
661.50 



'Harvest costs: whole tree = $18/dry ton (Source: Mattson. J. A. 1976. 
Harvesting research tor maximum yield systems. Unpublished manuscript on 
tile at the North Central Forest Experiment Station, Forestry Sciences Labora- 
tory, Houghton, Michigan) and forage = $8/dry ton (authors' estimate). 

2 From table 4. 

3 Harvest costs: whole tree = $14/dry ton (Source: Mattson, J. A. 1976. 
Harvesting research tor maximum yield systems. Unpublished manuscript on 
tile at the North Central Forest Experiment Station, Forestry Sciences Labora- 
tory, Houghton, Michigan). 



16 



Forage harvesting 

Forage-type mechanized harvesting of biomass is 
still in the development stage. Most proposed sys- 
tems involve multifunction machines that fell and 
chip the wood and blow the chips into containers for 
transport to the landing. We expect that this harvest- 
ing method would be appropriate for the 5 year cop- 
pice rotations. Our cost estimate of $8/dry ton is not 
based on any particular equipment but is simply a 
subjective estimate. This estimate is lower than har- 
vesting costs of conventional systems because of the 
higher degree of mechanization, but not as low as 
reported figures that are unjustifiably based on agri- 
cultural-type forage harvester operations. 

Land cost 

Cleared agricultural land in the Lake States close 
to the mill or plant, purchased at an average cost of 
$400/acre in 10 tracts of 80 to 120 acres, makes up the 
land base for this hypothetical plantation. Larger 
tracts would be more desirable but would be hard to 
find. 

Administrative 

A modest estimate of $7,500/year ($7.50/acre) for 
the first 2 years was based on a $15,000/year man- 
ager working during the growing season on the proj- 
ect, supervising employees and directing operations. 

Irrigation equipment 

The type of irrigation equipment has been de- 
scribed in the section on irrigation operating costs. 
The purchase price of this equipment is charged to 
the project when first purchased and when replaced 
(table 8). Most equipment will need to be replaced 
within 10 years, though in practice certain items will 
likely wear out sooner than others (the hoses for 
instance). Aluminum pipe can be expected to last 20 
years and the well 30 years (25 years is recom- 
mended). 

The entire purchase price is charged in 1 year 
rather than spread out in annual payments because a 
series of annual payments of both principal and in- 
terest (at 10 percent) would have the same present 
value as the purchase price for cash flow purposes. 

Fixed costs of irrigation equipment could be re- 
duced by a factor of 3 with tracts of 300 acres, the 
upper limit of acreage that can be served with one 
irrigation system. However, variable costs might in- 
crease if the system is used to capacity. 

Equipment for site preparation and planting 

Site preparation and planting equipment is leased 
rather than purchased because it is only used for a 



Table 8. — Irrigation equipment purchase schedule 



Equipment 


Quantity 


Year 


New 




needed 


purchased 


price 1 




No. 






40-hp tractor 


1 




$ 9,590 


Wells - 100-foot lift 


10 




119,700 


Turbine pumps 


10 




60,280 


R-angle drives 


10 




13,490 


Diesel power units 


10 




66,400 


Aluminum pipes 


10 




54,150 


Traveling guns 


10 




65,360 


Hoses 


10 




41,800 
$430, 770 2 


40-hp tractor 


1 


10 


9,590 


Turbine pumps 


10 


10 


60,280 


R-angle drives 


10 


10 


13,490 


Diesel power units 


10 


10 


66,400 


Traveling guns 


10 


1(1 


65,360 


Hoses 


10 


10 


41,800 
$256, 920 2 


40-hp tractor 


1 


20 


9,590 


Turbine pumps 


10 


20 


60,280 


R-angle drives 


10 


20 


13,490 


Diesel power units 


10 


20 


66,400 


Aluminum pipes 


10 


20 


54,150 


Traveling guns 


10 


20 


65,360 


Hoses 


10 


20 


41,800 
S311.070 2 



'Source: University of Minnesota. 1978. Water sources and irrigation 
economics. Miscellaneous Report 150-1978, 76 p. Agricultural Experiment 
Station, St. Paul, Minnesota. 

investment tax credit is 10 percent of the new price. 



couple of years. In an ongoing operation it would be 
used annually on different plantations, but the entire 
cost could not be fairly charged to one plantation and 
accounting for it as a lease is one method of allocating 
it. 

Lease payments were determined by equating the 
1979 price of the various pieces of equipment to a 
series of equal annual payments using a 10 percent 
interest rate (table 9). The full year's lease payment 
is charged. For certain operations the equipment is 
used such a short time that renting would make more 
sense than leasing but the difference to the project's 
overall return would be minimal. 

Insurance 

We assume the lessor would insure site prepara- 
tion and planting equipment. Therefore, only fire and 
storm insurance for the above-ground irrigation 



17 



Table 9. — Cost of site preparation and planting equipment 



Equipment 


Year(s) 


Useful life 


New price 2 


Lease payment 2 






Years 




■—Dollars 


225 hp tractor 


0,1,2 


7 


$49,040 


$10,073.08 


10 bottom plow 





7 


13,870 


2,848.97 


23-foot offset disk 





7 


11,290 


2,319.03 


30-foot sprayer 


0,1,2 


5 


1,940 


511.77 


20-foot Lilliston 










cultivator 


1 


7 


3,000 3 


616.22 


Planter 


1 


5 


' 2.000 3 


527.59 



1 Source: Benson, F. J. 1979. Machinery cost estimates. Unpublished data on file at Agricultural Extension Service, University of Minnesota, St. Paul, 
Minnesota. 
2 New price = present value of a series of equal annual payments at 10 percent interest rate. 
3 Authors' estimate. 



equipment is included. Insurance is estimated at 1 
percent of the new price (University of Minnesota 
1978). It will increase when equipment is replaced at 
a higher cost but not with inflation. 

Property tax 

We assumed a property tax of $4/acre/year, which 
is a rough estimate of tax rates in the Lake States for 
this type of land (University of Minnesota 1978). 

Product sale 

Current market prices for whole tree chips vary 
with locality. We used $12.50/green ton delivered, or 



$25/dry ton, inflated at 5 percent annually, as a rep- 
resentative Lake States price. 

Income tax 

The federal capital gains tax on timber income is 
charged at 30 percent of the taxable income, which is 
the product value less costs for harvesting, site prep- 
aration, establishment, planting, irrigation, fertil- 
ization, and administration. For all rotations after 
the first, only irrigation, fertilization, and harvesting 
costs since the last harvest were deducted from prod- 
uct sale returns. For the last year, capital gains from 
the sale of land were also taxed. No losses were car- 
ried forward or tax benefits from losses computed. 



18 



APPENDIX 2— CASH FLOWS 

CASH FLOU -- 4' X 4 SPACING — IRRIGATED 

FINANCIAL ANALYSIS FOR A 1000 ACRE OPERATIONAL HYBRID POPLAR PLANTATION 

TOTAL PROJECT AREA IS DIVIDED INTO 10 TRACTS OF 80-120 ACRES EACH. 

90 PER CENT OF THE TOTAL PROJECT AREA IS PLANTED. 

TREES/PLANTED ACRE: 2600 10" UNROOTED CUTTINGS 

ROTATIONS: (1) 10 YEAR AND (4) 5 YEAR COPPICE ROTATIONS 

YIELD: 7 DRY TONS/PLANTED ACRE/YR. FOR THE 10 YR. ROTATION 

8 " " 5 " ROTATIONS 

IRRIGATION: 10 EFFECTIVE IN. /PLANTED ACRE/YEAR 
FERTILIZATION: 100 LBS. OF NITROGEN/PLANTED ACRE/YEAR 
NOTE: ALL COST FIGURES BELOU ARE IN DOLLARS/PROJECT ACRE 

ANALYSIS INPUTS 



UNIT OF CURRENCY 

LAND AREA 

PROD. PERIOD 

DISCOUNT RATE 

ANNUAL CHANGE IN COSTS 

ANNUAL CHANGE IN BEN. 



DOLLARS 
1.00 ACRE 
30 

10.00 PERCENT 
PERCENT 
PERCENT 



AMOUNT AND TIMING OF EXPENDIT AND RECEIPTS 
DOLLARS/ACRE 



INPUT NC 


1. NAME 


AMOUNT ANN. RATE OF INFL. 


YEAR(S) 


1 


SITE PREP 


53.21 


5.00 









1 


SITE PREP 


13.13 


5.00 


1 






1 


SITE PREP 


6.77 


5.00 


2 






2 


PLANTING 


221.94 


5.00 


1 






3 


IRRIGATION 


99.74 


5.00 


1 


TO 


30 


4 


FERTILIZAT'N 


12.86 


5.00 


1 


TO 


30 


5 


UT HARVEST 


1134.00 


5.00 


10 






6 


FORG. HARVEST 


288.00 


5.00 


15 


20 


25 


7 


LAND COST 


400.00 


5.00 









8 


ADMINISTR'TV 


7.50 


5.00 





TO 


1 


9 


IRR.EGUIPH'T 


430.77 


5.00 


1 






9 


IRR.EQUIPH'T 


256.92 


5.00 


10 






9 


IRR.EQUIPH'T 


311.07 


5.00 


20 






10 


SP EQUIPM'T 


15.75 


5.00 









10 


SP EGUIPH'T 


33.90 


5.00 


1 






10 


SP EQUIPM'T 


10.58 


5.00 


2 






11 


INSURANCE 


2.57 





1 


TO 


9 


11 


INSURANCE 


4.18 





10 


TO 


19 


11 


INSURANCE 


6.82 





20 


TO 


30 


12 


INCOME TAX 


58.28 





15 






12 


INCOME TAX 


74.39 





20 






12 


INCOME TAX 


94.94 





25 






12 


INCOME TAX 


861.74 





30 






13 


PROPERTY TAX 


4.00 


5.00 





TO 


30 


14 


PRODUCT SALE 


1575.00 


5.00 


10 






14 


PRODUCT SALE 


900.00 


5.00 


15 


20 


25 


15 


LAND SALE 


400.00 


7.00 


30 






16 


INV TAX CRDT 


43.08 


5.00 


1 






16 


INV TAX CRDT 


25.69 


5.00 


10 






16 


INV TAX CRDT 


31.11 


5.00 


20 







30 



30 



19 



CASH FLOU -- 4' X 4' SPACING — IRRIGATED 

FINANCIAL ANALYSIS FOR A 1000 ACRE OPERATIONAL HYBRID POPLAR PLANTATION 

TOTAL PROJECT AREA IS DIVIDED INTO 10 TRACTS OF 80-120 ACRES EACH. 

90 PER CENT OF THE TOTAL PROJECT AREA IS PLANTED. 

TREES/PLANTED ACRE: 2600 10" UNROOTED CUTTINGS 

ROTATIONS: <1) 10 YEAR AND (4) 5 YEAR COPPICE ROTATIONS 

YIELD: 7 DRY TONS/PLANTED ACRE/YR. FOR THE 10 YR. ROTATION 

8 " " " " 5 " ROTATIONS 

IRRIGATION: 10 EFFECTIVE IN. /PLANTED ACRE/YEAR 
FERTILIZATION: 100 LBS. OF NITROGEN/PLANTED ACRE/YEAR 
NOTE: ALL COST FIGURES BELOU ARE IN DOLLARS/PROJECT ACRE 



INVESTMENT PERFORMANCE MEASURES 



DISCOUNT RATE DISCOUNTED NET RECEIPTS 
PERCENT 

-.3807 -.93 

-.3818 -.00 

-.3812 (INTERNAL RATE OF RETURN) 



(10.00 PERCENT DISCOUNT RATE) DOLLARS)ACRE 

NET PRESENT UORTH -2003.82 

NET FUTURE UORTH -34965.47 

FUTURE COSTS 79290.68 

PRESENT BENEFITS 2540.21 

PRESENT COSTS 4544.03 

BENEFITS/COSTS .56 



20 



CASH FLOU -- 4' X 4' SPACING -- IRRIGATED 

FINANCIAL ANALYSIS FOR A 1000 ACRE OPERATIONAL HYBRID POPLAR PLANTATION 

TOTAL PROJECT AREA IS DIVIDED INTO 10 TRACTS OF 80-120 ACRES EACH. 

90 PER CENT OF THE TOTAL PROJECT AREA IS PLANTED. 

TREES/PLANTED ACRE: 2600 10" UNROOTED CUTTINGS 

ROTATIONS: (1) 10 YEAR AND (4) 5 YEAR COPPICE ROTATIONS 

YIELD: 7 DRY TONS/PLANTED ACRE/YR. FOR THE 10 YR. ROTATION 

8 " " " " " " " 5 " ROTATIONS 
IRRIGATION: 10 EFFECTIVE IN. /PLANTED ACRE/YEAR 
FERTILIZATION: 100 LBS. OF NITROGEN/PLANTED ACRE/YEAR 
NOTE: ALL COST FIGURES BELOU ARE IN DOLLARS/PROJECT ACRE 

SENSITIVITY ANALYSIS 

(10.00 PERCENT DISCOUNT RATE) 
HOLLARS/ACRE 

CHANGE. IN NPU 
DUE TO A 10.00 
PERCENT CHANGE IN 



1 


SITE PREP 


7.19 


2 


PLANTING 


21.19 


3 


IRRIGATION 


157.58 


4 


FERTILIZAT'N 


20.32 


5 


UT HARVEST 


71.22 


6 


FORG. HARVEST 


41.83 


7 


LAND COST 


40.00 


8 


ADHINISTR'TV 


1 .47 


9 


IRR.EQUIPM'T 


69.52 


10 


SP EQUIPH'T 


5.77 


11 


INSURANCE 


3.29 


12 


INCOME TAX 


8.32 


13 


PROPERTY TAX 


6.72 



14 PRODUCT SALE 229.62 

15 LAND SALE 17.45 

16 INV TAX CRDT 6.95 



21 



CASH FLOU — 4' X 4' SPACING — IRRIGATED 

FINANCIAL ANALYSIS FOR A 1000 ACRE OPERATIONAL HYBRID POPLAR 
TOTAL PROJECT AREA IS DIVIDED INTO 10 TRACTS OF 80-120 ACRES 
90 PER CENT OF THE TOTAL PROJECT AREA IS PLANTED. 
TREES/PLANTED ACRE: 2600 10" UNROOTED CUTTINGS 
ROTATIONS: (1) 10 YEAR AND (4) 5 YEAR COPPICE ROTATIONS 
YIELD: 7 DRY TONS/PLANTED ACRE/YR. FOR THE 10 YR. ROTATION 

8 " " " " 5 " ROTATIONS 

IRRIGATION: 10 EFFECTIVE IN. /PLANTED ACRE/YEAR 
FERTILIZATION: 100 LBS. OF NITROGEN/PLANTED ACRE/YEAR 
NOTE: ALL COST FIGURES BELOU ARE IN DOLLARS/PROJECT ACRE 



PLANTAT 
EACH. 



ION 



SENSITIVITY OF INTERNAL RATE OF RETURN 

PERCENT CHANGE IN INPUT OR OUTPUT TO RAISE/LOUER 

ROR BY 2.000 PERCENT AND NPU BY 2003.82 DOLLARS/ACRE 







RATE 


OF RETURN 


PRESENT 


NET UORTH 






UPPER 


LOUER 


UPPER 


LOUER 






THRESHOLD 


THRESHOLD 


THRESHOLD 


THRESHOLD 






1.62 


-2.38 





-4007.64 




INPUT/OUTPUT 










1 


SITE PREP 


-1620.88 


3317.91 


-2786.50 


2786.50 


2 


PLANTING 


-523.07 


1044.70 


-945.86 


945.86 


3 


IRRIGATION 


-23.19 


22.23 


-127.17 


127.17 


4 


FERTILIZAT'N 


-179.90 


172.44 


-986.28 


986.28 


5 


UT HARVEST 


-76.25 


106.10 


-281.37 


281.37 


6 


FORG. HARVEST 


-49.03 


38.71 


-479.08 


479.08 


-> 


LAND COST 


-299.88 


623.48 


-500.96 


500.96 


8 


ADHINISTR'TV 


-7866.01 


16020.45 


-13669.47 


13669.47 


9 


IRR.EQUIPH'T 


-85.67 


106.92 


-288.23 


288.23 


10 


SP EQUIPH'T 


-1932.43 


3869.32 


-3469.87 


3469.87 


1 1 


INSURANCE 


-1144.64 


1118.31 


-6083.96 


6083.96 


12 


INCOME TAX 


-172.44 


115.82 


-2409.68 


2409.68 


13 


PROPERTY TAX 


-567.42 


549.50 


-2982.12 


2982.12 


14 


PRODUCT SALE 


12.20 


-10.66 


87.27 


-87.27 


15 


LAND SALE 


63.78 


-39.75 


1148.33 


-1148.33 


16 


INV TAX CRDT 


856.68 


-1069.19 


2882.16 


-2882.16 



22 



CASH FLOW -- 4' X 4' SPACING — NON-IRRIGATED 

FINANCIAL ANALYSIS FOR A 1000 ACRE OPERATIONAL HYBRID POPLAR PLANTATION 

TOTAL PROJECT AREA IS DIVIDED INTO 10 TRACTS OF 80-120 ACRES EACH. 

90 PER CENT OF TH TOTAL PROJECT ACREAGE IS PLANTED. 

TREES/PLANTED ACRE: 2600 10" UNROOTED CUTTINGS 

ROTATIONS: (1) 10 YEAR AND (4) 5 YEAR COPPICE ROTATIONS 

YIELD: 3.5 DRY TONS/PLANTED ACRE/YR. FOR THE 10 YR. ROTATION 

4 " " " " " " " 5 " ROTATIONS 
IRRIGATION: NONE 
FERTILIZATION: NONE 
NOTE: ALL COST FIGURES BELOU ARE IN DOLLARS/PROJECT ACRE. 

ANALYSIS INPUTS 



UNIT OF CURRENCY 

LAND AREA 

PROD. PERIOD 

DISCOUNT RATE 

ANNUAL CHANGE IN COSTS 

ANNUAL CHANGE IN BEN. 



DOLLARS 
1.00 ACRE 
30 

10.00 PERCENT 
PERCENT 
PERCENT 



AMOUNT AND TIMING OF EXPENDIT AND RECEIPTS 
DOLLARS/ACRE 



INPUT 



NO. 
1 



1 

2 
3 
4 
5 
6 
7 
7 
7 
8 
8 
8 
8 
8 
9 

10 
10 
11 



NAME 
SITE PREP 
SITE PREP 
SITE PREP 
PLANTING 
UT HARVEST 
FORG. HARVEST 
LAND COST 
ADMINISTR'TV 
SP EQUIPM'T 
SP EQUIPM'T 
SP EQUIPM'T 
INCOME TAX 
INCOME 
INCOME 
INCOME 
INCOME 



TAX 
TAX 
TAX 
TAX 



PROPERTY TAX 
PRODUCT SALE 
PRODUCT SALE 
LAND SALE 



AMOUNT ANN. RATE OF INFL 

53.21 5.00 

13.13 5.00 

6.77 5.00 

221.94 5.00 

567.00 5.00 

144.00 5.00 

400.00 5.00 

7.50 5.00 

15.75 5.00 

33.90 5.00 

10.58 5.00 

10.16 

178.12 

227.33 

290.14 

1110.88 

4.00 5.00 

787.50 5.00 

450.00 5.00 

400.00 7.00 



YEAR(S) 



1 

2 

1 
10 
15 20 25 30 



TO 1 



1 

2 
10 
15 
20 
25 
30 

TO 30 
10 

15 20 25 30 
30 



23 



CASH FLOU -- 4' X 4' SPACING — NON-IRRIGATED 

FINANCIAL ANALYSIS FOR A 1000 ACRE OPERATIONAL HYBRID POPLAR PLANTATION 

TOTAL PROJECT AREA IS DIVIDED INTO 10 TRACTS OF 80-120 ACRES EACH. 

90 PER CENT OF TH TOTAL PROJECT ACREAGE IS PLANTED. 

TREES/PLANTED ACRE: 2600 10" UNROOTED CUTTINGS 

ROTATIONS: (1) 10 YEAR AND (4) 5 YEAR COPPICE ROTATIONS 

YIELD: 3.5 DRY TONS/PLANTED ACRE/YR. FOR THE 10 YR. ROTATION 

4 " " ' " 5 " ROTATIONS 

IRRIGATION: NONE 

FERTILIZATION: NONE 

NOTE: ALL COST FIGURES BELOU ARE IN DOLLARS/PROJECT ACRE. 



INVESTMENT PERFORMANCE MEASURES 



DISCOUNT RATE DISCOUNTED NET RECEIPTS 
PERCENT 

8.1174 -.07 

8.1169 .00 

8.1171 (INTERNAL RATE OF RETURN) 



(10.00 PERCENT DISCOUNT RATE) HOLLARS)ACRE 

NET PRESENT UORTH -236.78 

NET FUTURE UORTH -4131.60 

FUTURE COSTS 27210.08 

PRESENT BENEFITS 1322.59 

PRESENT COSTS 1559.37 

BENEFITS/COSTS .85 



24 



CASH FLOU -- 4- X 4' SPACING -- NON- IRRIGATED 

FINANCIAL ANALYSIS FOR A 1000 ACRE OPERATIONAL HYBRID POPLAR PLANTATION 

TOTAL PROJECT AREA IS DIVIDED INTO 10 TRACTS OF 80-120 ACRES EACH. 

90 PER CENT OF TH TOTAL PROJECT ACREAGE IS PLANTED. 

TREES/PLANTED ACRE: 2600 10" UNROOTED CUTTINGS 

ROTATIONS: (1) 10 YEAR AND (4) 5 YEAR COPPICE ROTATIONS 

YIELD: 3.5 DRY TONS/PLANTED ACRE/YR. FOR THE 10 YR. ROTATION 

4 " " ' ' 5 " ROTATIONS 

IRRIGATION: NONE 

FERTILIZATION: NONE 

NOTE: ALL COST FIGURES BELOW ARE IN DOLLARS/PROJECT ACRE. 

SENSITIVITY ANALYSIS 

(10.00 PERCENT DISCOUNT RATE) 
DOLLARS/ACRE 

CHANGE IN NPU 
DUE TO A 10.00 
PERCENT CHANGE IN 



1 


SITE PREP 


7.19 


1 


PLANTING 


21 .19 


3 


UT HARVEST 


35.61 


4 


FORG. HARVEST 


20.91 


5 


LAND COST 


40.00 


6 


ADhlNISTR'TV 


1.47 


7 


SP EQUIPM'T 


5.77 


8 


INCOME TAX 


17.08 


9 


PROPERTY TAX 


6.72 


10 


PRODUCT SALE 


114.81 


11 


LAND SALE 


17.45 



25 



CASH FLOU -- 4' X 4' SPACING — NON-IRRIGATED 

FINANCIAL ANALYSIS FOR A 1000 ACRE OPERATIONAL HYBRID POPLAR PLANTATH 

TOTAL PROJECT AREA IS DIVIDED INTO 10 TRACTS OP 80-120 ACRES EACH. 

90 PER CENT OF TH TOTAL PROJECT ACREAGE IS PLANTED. 

TREES/PLANTED ACRE: 2600 10" UNROOTED CUTTINGS 

ROTATIONS: (1) 10 YEAR AND (4) 5 YEAR COPPICE ROTATIONS 

YIELD: 3.5 DRY TONS/PLANTED ACRE/YR. FOR THE 10 YR. ROTATION 

4 " " " " " " " 5 " ROTATIONS 
IRRIGATION: NONE 
FERTILIZATION: NONE 
NOTE: ALL COST FIGURES BELOU ARE IN DOLLARS/PROJECT ACRE. 



SENSITIVITY OF INTERNAL RATE OF RETURN 

PERCENT CHANGE IN INPUT OR OUTPUT TO RAISE/LOUER 

ROR BY 2.000 PERCENT AND NPU BY 236.78 DOLLARS/ACRE 







RATE OF RETURN 


PRESENT 


NET UORTH 






UPPER 


LOWER 


UPPER 


LOUER 






THRESHOLD 


THRESHOLD 


THRESHOLD 


THRESHOLD 






10.12 


6.12 





-473.55 




INPUT/OUTPUT 










1 


SITE PREP 


-345.44 


553.83 


-329.26 


329.26 


9 


PLANTING 


-117.34 


183.68 


-111.76 


111.76 


3 


UT HARVEST 


-70.48 


79.08 


-66.49 


66.49 


4 


FORG. HARVEST 


-121.43 


88.70 


-113.22 


113.22 


5 


LAND COST 


-62.08 


100.84 


-59.19 


59.19 


6 


ADMINISTRATE 


-1694.84 


2703.28 


-1615.22 


1615.22 


-> 


SP EQUIPH'T 


-430.41 


676.19 


-410.01 


410.01 


8 


INCOME TAX 


-149.00 


100.61 


-138.64 


138.64 


9 


PROPERTY TAX 


-374.05 


379.55 


-352.37 


352.37 


10 


PRODUCT SALE 


22.01 


-18.94 


20.62 


-20.62 


11 


LAND SALE 


146.92 


-78.65 


135.69 


-135.69 



26 



CASH FLOU -- 8' X 8' SPACING — IRRIGATED 

FINANCIAL ANALYSIS FOR A 1000 ACRE OPERATIONAL HYBRID POPLAR PLANTATION 

TOTAL PROJECT ACREAGE IS DIVIDED INTO 10 TRACTS OF 80-120 ACRES EACH. 

90 PER CENT OF THE TOTAL PROJECT ACREAGE IS PLANTED. 

TREES/PLANTED ACRE: 650 10" UNROOTED CUTTINGS 

ROTATIONS: (2) 15 YEAR 

YIELD: 7 DRY TONS/PLANTED ACRE/YR. FOR EACH ROTATION 

IRRIGATION: 10 EFFECTIVE IN. /PLANTED ACRE/YEAR 

FERTILIZATION: 100 LBS. OF NITROGEN/PLANTED ACRE/YEAR 

NOTE: ALL COST FIGURES BELOU ARE IN DOLLARS/PROJECT ACRE. 

ANALYSIS INPUTS 



UNIT OF CURRENCY 

LAND AREA 

PROD. PERIOD 

DISCOUNT RATE 

ANNUAL CHANGE IN COSTS 

ANNUAL CHANGE IN BEN. 



DOLLARS 
1.00 ACRE 
30 

10.00 PERCENT 
PERCENT 
PERCENT 



AMOUNT AND TIMING OF EXPENDIT AND RECEIPTS 
HOLLARS/ACRE 



INPUT NO 


NAME 


AMOUNT ANN. RATE OF IN 


: L. Y( 


:ar<j 


i) 


1 


SITE PREP 


21.08 


5.00 









1 


SITE PREP 


9.35 


5.00 


1 






1 


SITE PREP 


2.99 


5.00 


2 






2 


PLANTING 


59.83 


5.00 


1 






3 


IRRIGATION 


99.74 


5.00 


1 


TO 


30 


4 


FERTILIZAT'N 


12.86 


5.00 


1 


TO 


30 


5 


UT HARVEST 


1323.00 


5.00 


15 


30 




6 


LAND COST 


400.00 


5.00 









7 


ADMINISTR'TV 


7.50 


5.00 





TO 


1 


8 


IRR.EQUIPM'T 


430.77 


5.00 


1 






8 


IRR.EQUIPM'T 


256.92 


5.00 


10 






8 


IRR.EQUIPM'T 


311.07 


5.00 


20 






9 


SP EQUIPM'T 


15.75 


5.00 









9 


SP EQUIPM'T 


22.77 


5.00 


1 






9 


SP EQUIPM'T 


10.58 


5.00 


2 






10 


INSURANCE 


2.57 





1 


TO 


9 


10 


INSURANCE 


4.18 





10 


TO 


19 


10 


INSURANCE 


6.82 





20 


TO 


30 


11 


INCOME TAX 


740.57 





30 






12 


PROPERTY TAX 


4.00 


5.O0 





TO 


30 


13 


PRODUCT SALE 


2362.50 


5.00 


15 


30 




14 


LAND SALE 


400.00 


7.00 


50 






15 


INV TAX CRDT 


43.08 


5.00 


1 






15 


INV TAX CRDT 


25.69 


5.00 


10 






15 


INV TA,< CRDT 


31 .11 


5.00 


20 







27 



CASH FLOU -- 8' X 8' SPACING — IRRIGATED 

FINANCIAL ANALYSIS FOR A 1000 ACRE OPERATIONAL HYBRID POPLAR PLANTATION 

TOTAL PROJECT ACREAGE IS DIVIDED INTO 10 TRACTS OF 80-120 ACRES EACH. 

90 PER CENT OF THE TOTAL PROJECT ACREAGE IS PLANTED. 

TREES/PLANTED ACRE: 650 10" UNROOTED CUTTINGS 

ROTATIONS: (2) 15 YEAR 

YIELD: 7 DRY TONS/PLANTED ACRE/YR. FOR EACH ROTATION 

IRRIGATION: 10 EFFECTIVE IN. /PLANTED ACRE/YEAR 

FERTILIZATION: 100 LBS. OF NITROGEN/PLANTED ACRE/YEAR 

NOTE: ALL COST FIGURES BELOU ARE IN DOLLARS/PROJECT ACRE. 



INVESTMENT PERFORMANCE MEASURES 



DISCOUNT RATE DISCOUNTED NET RECEIPTS 
PERCENT 

-1.6122 .53 

-1.6117 -.00 

-1.6120 (INTERNAL RATE OF RETURN) 



(10.00 PERCENT DISCOUNT RATE) DOLLARS)ACRE 

NET PRESENT UORTH -2149.51 

NET FUTURE UORTH -37507.75 

FUTURE COSTS 72492.83 

PRESENT BENEFITS 2004.94 

PRESENT COSTS 4154.46 

BENEFITS/COSTS .48 



28 



CASH FLOU -- 8'' X 8' SPACING — IRRIGATED 

FINANCIAL ANALYSIS FOR A 1000 ACRE OPERATIONAL HYBRID POPLAR PLANTATION 

TOTAL PROJECT ACREAGE IS DIVIDED INTO 10 TRACTS OF 80-120 ACRES EACH. 

90 PER CENT OF THE TOTAL PROJECT ACREAGE IS PLANTED. 

TREES/PLANTED ACRE: 650 10" UNROOTED CUTTINGS 

ROTATIONS: (2) 15 YEAR 

YIELD: 7 DRY TONS/PLANTED ACRE/YR. FOR EACH ROTATION 

IRRIGATION: 10 EFFECTIVE IN. /PLANTED ACRE/YEAR 

FERTILIZATION: 100 LBS. OF NITROGEN/PLANTED ACRE/YEAR 

NOTE: ALL COST FIGURES BELOU ARE IN DOLLARS/PROJECT ACRE. 

SENSITIVITY ANALYSIS 

(10.00 PERCENT DISCOUNT RATE) 
DOLLARS/ACRE 

CHANGE IN NPU 
DUE TO A 10.00 
PERCENT CHANGE IN 



1 


SITE PREP 


3.27 


2 


PLANTING 


5.71 


3 


IRRIGATION 


157.58 


4 


FERTILIZAT'N 


20.32 


5 


UT HARVEST 


98.61 


6 


LAND COST 


40.00 


7 


ADhlNISTR'TV 


1.47 


8 


IRR.EQUIPM'T 


69.52 


9 


SP EQUIPM'T 


4.71 


10 


INSURANCE 


3.29 


1 1 


INCOHE TAX 


4.24 


12 


PROPERTY TAX 


6.72 


13 


PRODUCT SALE 


176.09 


14 


LAND SALE 


17.45 


15 


INV TAX CRDT 


6.95 



29 



CASH FLOW -- 8' X 8" SPACING — IRRIGATED 

FINANCIAL ANALYSIS FOR A 1000 ACRE OPERATIONAL HYBRID POPLAR PLANTATIO 

TOTAL PROJECT ACREAGE IS DIVIDED INTO 10 TRACTS OF 80-120 ACRES EACH. 

90 PER CENT OF THE TOTAL PROJECT ACREAGE IS PLANTED. 

TREES/PLANTED ACREi 650 10" UNROOTED CUTTINGS 

ROTATIONS: (2) 15 YEAR 

YIELD: 7 DRY TONS/PLANTED ACRE/YR. FOR EACH ROTATION 

IRRIGATION: 10 EFFECTIVE IN. /PLANTED ACRE/YEAR 

FERTILIZATION: 100 LBS. OF NITROGEN/PLANTED ACRE/YEAR 

NOTE: ALL COST FIGURES BELQU ARE IN DOLLARS/PROJECT ACRE. 



SENSITIVITY OF INTERNAL RATE OF RETURN 



PERCENT CHANGE IN INPUT OR OUTPUT TO RAISE/LOUER 

ROR BY 2.000 PERCENT AND NPU BY 2149.51 DOLLARS/ACRE 







RATE 


OF RETURN 


PRESENT 


NET UORTH 






UPPER 


LOUER 


UPPER 


LOUER 






THRESHOLD 


THRESHOLD 


THRESHOLD 


THRESHOLD 






.39 


-3.61 





-4299.03 




INPUT/OUTPUT 










1 


SITE PREP 


-4753.20 


10811.18 


-6567.54 


6567.54 


2 


PLANTING 


-2592.41 


5774.79 


-3763.79 


3763.79 


3 


IRRIGATION 


-25.09 


25.73 


-136.41 


136.41 


4 


FERTILIZAT'N 


-194.57 


199.52 


-1057.99 


1057.99 


5 


UT HARVEST 


-21.11 


17.10 


-217.98 


217.98 


6 


LAND COST 


-405.57 


940.94 


-537.38 


537.38 


7 


ADMINISTR'TV 


-10572.44 


24018.70 


-14663.36 


14663.36 


8 


IRR.EQUIPM'T 


-100.33 


134.59 


-309.18 


309.18 


9 


SP EQUIPH'T 


-3172.23 


7086.79 


-4561.30 


4561.30 


10 


INSURANCE 


-1244.96 


1302.11 


-6526.31 


6526.31 


11 


INCOME TAX 


-246.05 


168.56 


-5064.71 


5064.71 


12 


PROPERTY TAX 


-616.03 


637.11 


-3198.94 


3198.94 


13 


PRODUCT SALE 


11.82 


-9.57 


122.07 


-122.07 


14 


LAND SALE 


59.84 


-41.00 


1231.82 


-1231.82 


15 


INV TAX CRDT 


1003.23 


-1345.84 


3091.71 


-3091.71 



30 



CASH FLOU -- 8' X 8" SPACING — NON-IRRIGATED 

FINANCIAL ANALYSIS FOR A 1000 ACRE OPERATIONAL HYBRID POPLAR PLANTATION 

TOTAL PROJECT ACREAGE IS DIVIDED INTO 10 TRACTS OF 80-120 ACRES EACH. 

90 PER CENT OF THE TOTAL PROJECT ACREAGE IS PLANTED. 

TREES/PLANTED ACRE: 650 10" UNROOTED CUTTINGS 

ROTATIONS: (2) 15 YEAR 

YIELD: 3.5 DRY TONS/PLANTED ACRE/YR. FOR EACH ROTATION 

IRRIGATION: 10 EFFECTIVE IN. /PLANTED ACRE/YEAR 

FERTILIZATION: 100 LBS. /PLANTED ACRE/YEAR OF NITROGEN 

NOTE: ALL COST FIGURES BELOU ARE IN DOLLARS/PROJECT ACRE. 

ANALYSIS INPUTS 



UNIT OF CURRENCY 


DOLLARS 


LAND AREA 


1.00 ACRE 


PROD. PERIOD 


30 


DISCOUNT RATE 


10.00 PERCENT 


ANNUAL CHANGE IN COSTS 


PERCENT 


ANNUAL CHANGE IN BEN. 


PERCENT 



AMOUNT AND TIMING OF EXPENDIT AND RECEIPTS 
DOLLARS/ACRE 



INPUT NO 


NAME 


AMOUNT ANN. RATE OF INFL. 


YEAR(S) 


1 


SITE PREP 


21.08 


5.00 







1 


SITE PREP 


9.35 


5.00 


1 




1 


SITE PREP 


2.99 


5.00 


2 




2 


PLANTING 


59.83 


5.00 


1 




3 


UT HARVEST 


661.50 


5.00 


15 


30 


4 


LAND COST 


400.00 


5.00 







5 


ADMINISTR'TV 


7.50 


5.O0 





TO 


6 


SP EQUIPM'T 


15.75 


5.00 







6 


SP EQUIPM'T 


22.77 


5.00 


1 




6 


SP EQUIPM'T 


10.58 


5.00 


2 




7 


INCOME TAX 


271.08 





15 




7 


INCOME TAX 


1369.54 





30 




8 


PROPERTY TAX 


4.00 


5.00 





TO 


9 


PRODUCT SALE 


1181.25 


5.00 


15 


30 


10 


LAND SALE 


400.00 


7.00 


30 





30 



31 



CASH FLOU -- 8" X 8' SPACING — NON-IRRIGATED 

FINANCIAL ANALYSIS FOR A 1000 ACRE OPERATIONAL HYBRID POPLAR PLANTATION 

TOTAL PROJECT ACREAGE IS DIVIDED INTO 10 TRACTS OF 80-120 ACRES EACH. 

90 PER CENT OF THE TOTAL PROJECT ACREAGE IS PLANTED. 

TREES/PLANTED ACRE: 650 10" UNROOTED CUTTINGS 

ROTATIONS: (2) 15 YEAR 

YIELD: 3.5 DRY TONS/PLANTED ACRE/YR. FOR EACH ROTATION 

IRRIGATION: 10 EFFECTIVE IN. /PLANTED ACRE/YEAR 

FERTILIZATION: 100 LBS. /PLANTED ACRE/YEAR OF NITROGEN 

NOTE: ALL COST FIGURES BELOU ARE IN DOLLARS/PROJECT ACRE. 



INVESTMENT PERFORMANCE MEASURES 



DISCOUNT RATE DISCOUNTED NET RECEIPTS 



PERCENT 



8.0555 -.12 

8.0545 .00 

8.0550 (INTERNAL RATE OF RETURN) 



(10.00 PERCENT DISCOUNT RATE) DOLLARS)ACRE 

NET PRESENT UORTH -200.30 

NET FUTURE UORTH -3495.08 

FUTURE COSTS 21903.48 

PRESENT BENEFITS 1054.96 

PRESENT COSTS 1255.26 

BENEFITS/COSTS .84 



32 



CASH FLOW -- 8- X 8-' SPACING — NON-IRRIGATED 

FINANCIAL ANALYSIS FOR A 1000 ACRE OPERATIONAL HYBRID POPLAR PLANTATION 

TOTAL PROJECT ACREAGE IS DIVIDED INTO 10 TRACTS OF 80-120 ACRES EACH. 

90 PER CENT OF THE TOTAL PROJECT ACREAGE IS PLANTED. 

TREES/PLANTED ACRE: 650 10" UNROOTED CUTTINGS 

ROTATIONS: (2) 15 YEAR 

YIELD: 3.5 DRY TONS/PLANTED ACRE/YR. FOR EACH ROTATION 

IRRIGATION: 10 EFFECTIVE IN. /PLANTED ACRE/YEAR 

FERTILIZATION: 100 LBS. /PLANTED ACRE/YEAR OF NITROGEN 

NOTE: ALL COST FIGURES BELOU ARE IN DOLLARS/PROJECT ACRE. 

SENSITIVITY ANALYSIS 

(10.00 PERCENT DISCOUNT RATE) 
DOLLARS/ACRE 

CHANGE IN NPU 
DUE TO A 10.00 
PERCENT CHANGE IN 



1 


SITE PREP 


3.27 


2 


PLANTING 


5.71 


3 


UT HARVEST 


49.31 


4 


LAND COST 


40.00 


5 


ADHINISTR'TV 


1.47 


6 


SP EQUIPH'T 


4.71 


7 


INCOME TAX 


14.34 


8 


PROPERTY TAX 


6.72 


9 


PRODUCT SALE 


88.05 


10 


LAND SALE 


17.45 



33 



CASH FLOW -- 8' X 8' SPACING -- NON-IRRIGATED 

FINANCIAL ANALYSIS FOR A 1000 ACRE OPERATIONAL HYBRID POPLAR PLANTATION 

TOTAL PROJECT ACREAGE IS DIVIDED INTO 10 TRACTS OF 80-120 ACRES EACH. 

90 PER CENT OF THE TOTAL PROJECT ACREAGE IS PLANTED. 

TREES/PLANTED ACRE: 650 10" UNROOTED CUTTINGS 

ROTATIONS: (2) 15 YEAR 

YIELD: 3.5 DRY TONS/PLANTED ACRE/YR. FOR EACH ROTATION 

IRRIGATION: 10 EFFECTIVE IN. /PLANTED ACRE/YEAR 

FERTILIZATION: 100 LBS. /PLANTED ACRE/YEAR OF NITROGEN 

NOTE: ALL COST FIGURES BELOU ARE IN DOLLARS/PROJECT ACRE. 



SENSITIVITY OF INTERNAL RATE OF RETURN 



PERCENT CHANGE IN INPUT OR OUTPUT TO RAISE/LOUER 

ROR BY 2.000 PERCENT AND NPU BY 200.30 DOLLARS/ACRE 







RATE OF RETURN 


PRESENT 


NET UORTH 






UPPER 


LOUER 


UPPER 


LOUER 






THRESHOLD 


THRESHOLD 


THRESHOLD 


THRESHOLD 






10.06 


6.06 





-400.60 




INPUT/OUTPUT 










1 


SITE PREP 


-625.42 


1034.46 


-611.98 


611.98 


2 


PLANTING 


-358.52 


580.98 


-350.72 


350.72 


3 


UT HARVEST 


-41.92 


32.48 


-40.62 


40.62 


4 


LAND COST 


-51.16 


86.04 


-50.07 


50.07 


5 


ABhINISTR'TV 


-1396.41 


2305.76 


-1366.37 


1366.37 


6 


SP EQUIPH'T 


-434.46 


707.18 


-425.03 


425.03 


7 


INCOME TAX 


-144.40 


99.18 


-139.70 


139.70 


8 


PROPERTY TAX 


-306.30 


321.15 


-298.09 


298.09 


9 


PRODUCT SALE 


23.48 


-18.19 


22.75 


-22.75 


10 


LAND SALE 


119.05 


-65.93 


114.78 


-114.78 



34 



APPENDIX 3— DISCUSSION OF ENERGY ANALYSIS 



Only direct energy expenditures (fuel and chemi- 
cals) were considered in the energy analysis. Energy 
expended in manufacturing equipment and in labor 
was not considered because it is an insignificant part 
of the total energy picture. 

Preparing Site and Establishing 
Trees 

Energy inputs for preparing the site and establish- 
ing the trees include fuel and herbicides. Fuel energy 
was calculated from the total hours spent and the fuel 
consumption rates/hour (table 5), using a conversion 
factor of 0.138 MMBTU's/gal. of diesel. Herbicides 
were estimated to contain 11,000 kcal/lb (Eidman et 
al. 1975). 

Planting, Irrigating 

Fuel was the only energy input accounted for in 
planting and irrigating and was calculated in the 



same way as above, using information from Appen- 
dix 1. 



Fertilizing 

Energy content of nitrogen fertilizer was estimated 
to be 8,400 kcal/lb, or for 100 lbs/acre, 3.33 MMBTU's 
/acre (Eidman et al. 1975). 



Whole Tree and Forage Harvesting 

Fuel consumption/dry ton was estimated using in- 
formation from the simulation of a whole tree har- 
vesting system for intensively grown poplar 11 . For 
lack of a better estimate, we assumed that forage 
harvesting would take the same amount of energy 
(table 10). 



Table 10. — Harvesting energy expenditures 1 





Trees harvested 




Diesel fuel used 


Machine 


4- by 4-foot 


8- by 8-foot 


4- by 4-foot 


8- by 8-foot 




spacing 


spacing 




spacing 


spacing 




Green tons/hr 


Gal/hr 


Gal/green 


ton 


Medium skidder 


— 


14.7 


3.30 


— 


0.224 


Small skidders(2) 


12.4 


— 


4.20 


0.339 


— 


Feller-buncher 


21.2 


70.3 


4.71 


.222 


.067 


Chipper/baler 


12.4 


14.7 


7.65 
Total 


.617 


.520 




1.178 


.811 








Gal/dry ton 


2.356 


1.622 



'Source: Mattson, J. A. 1976. Harvesting research for maximum yield systems. Unpublished report on file at North Central Forest Experiment Station, Forestry 
Sciences Laboratory. Houghton, Michigan. 



n Mattson, J. A. 1976. Harvesting research for 
maximum yield systems. Unpublished report on file at 
the North Central Forest Experiment Station, For- 
estry Sciences Laboratory, Houghton, Michigan. 



35 



Hauling 

A 50-mile round trip over 10 miles of good gravel 
road and 15 miles of average paved road using a 40- 
foot van holding 12 dry tons of chips was used to 
estimate fuel consumption/dry ton. Hauling energy 
expenditures were as follows 12 : 
25 mile haul (one way) — 

loaded: 10 miles class II county roads 

x 0.276 gal/mile = 2.76 gal 
unloaded: 10 miles class II county roads 

x 0.110 gal/mile = 1.10 gal 
loaded: 15 miles class I paved road 

x 0.224 gal/mile = 3.36 gal 
unloaded: 15 miles class I paved road 

x 0.120 gal/mile - 1.80 gal 

9.02 gal 
Given 12 dry tons/van then 9.02 gals ■*■ 12 tons = 

.75 gal of diesel/dry ton 



Drying 

An estimated 3.184 MMBTU's/dry ton is used to 
dry wood chips (Blankenhorn et al. 1978). 

Wood energy 

The gross heat content of hybrid poplar is 16.8 
MMBTU's/dry ton, a weighted average of the heat 
content for stem and branch wood (Zavitkovski 
1979). Blankenhorn et al. (1978) estimate that 86 
percent of this gross heat energy is usable, and that 
only 35 percent of this is converted into electrical 
energy. We used gross energy in the energy flow 
analysis. 



12 



Source: Aube, P. J. 1979. University of Minne- 



sota. 



36 



APPENDIX 4.— DISCOUNTED ENERGY FLOWS 



ENERGY BUDGET — 4' X 4' SPACING — IRRIGATED 

ENERGY ANALYSIS FOR A 1000 ACRE OPERATIONAL HYBRID POPLAR PLANTATION 

TOTAL PROJECT ACREAGE IS DIVIDED INTO 10 TRACTS OF 80-120 ACRES EACH. 

90 PER CENT OF THE TOTAL PROJECT ACREAGE IS PLANTED. 

TREES/PLANTED ACRE: 2600 10" UNROOTED CUTTINGS 

ROTATIONS: (1) 10 YEAR AND (4) 5 YEAR COPPICE ROTATIONS 

YIELD: 7 DRY TONS/PLANTED ACRE/YR. FOR THE TEN YEAR ROTATION 

8 " " " ' FIVE YEAR ROTATIONS 

IRRIGATION: 10 EFFECTIVE INCHES/PLANTED ACRE/YEAR 

FERTILIZATION: 100 LBS. OF NITROGEN/PLANTED ACRE/YEAR 

NOTE: ALL ENERGY INPUTS AND OUTPUTS BELOU ARE EXPRESSED IN MMBTU'S/ACRE 

ANALYSIS INPUTS 



UNIT OF CURRENCY 

LAND AREA 

PROD. PERIOD 

DISCOUNT RATE 

ANNUAL CHANGE IN COSTS 

ANNUAL CHANGE IN BEN. 



MHBTU'S 
1.00 ACRE 
30 

10.00 PERCENT 
PERCENT 
PERCENT 



AMOUNT AND TIMING OF COSTS 
MHBTU'S/ACRE 



AND BENEFITS 



INPUT NO. 
1 
1 
1 
2 
3 
4 
5 
6 
7 
7 
8 
8 
9 
9 



NAME 
SITE PREP 
SITE PREP 
SITE PREP 
PLANTING 
IRRIGATION 
FERTILIZAT'N 
UT HARVEST 
FORG. HARVEST 
HAULING 
HAULING 
DRYING 
DRYING 
UOOD ENERGY 
UOOD ENERGY 



AMOUNT ANN. RATE OF INFL. YEAR(S) 

.52 

.59 1 

.19 2 

1.64 1 

9.34 1 TO 30 

3.33 1 TO 30 

20.47 10 

11.70 15 20 25 30 

6.52 10 

3.72 15 20 25 30 

200.59 10 

114.62 15 20 25 30 

1058.40 10 

604.80 15 20 25 30 



37 



ENERGY BUDGET -- 4" X 4' SPACING — IRRIGATED 

ENERGY ANALYSIS FOR A 1000 ACRE OPERATIONAL HYBRID POPLAR PLANTATION 

TOTAL PROJECT ACREAGE IS DIVIDED INTO 10 TRACTS OF 80-120 ACRES EACH. 

90 PER CENT OF THE TOTAL PROJECT ACREAGE IS PLANTED. 

TREES/PLANTED ACRE: 2600 10" UNROOTED CUTTINGS 

ROTATIONS: (1) 10 YEAR AND (4) 5 YEAR COPPICE ROTATIONS 

YIELD: 7 DRY TONS/PLANTED ACRE/YR. FOR THE TEN YEAR ROTATION 

8 FIVE YEAR ROTATIONS 

IRRIGATION: 10 EFFECTIVE INCHES/PLANTED ACRE/YEAR 

FERTILIZATION: 100 LBS. OF NITROGEN/PLANTED ACRE/YEAR 

NOTE: ALL ENERGY INPUTS AND OUTPUTS BELOU ARE EXPRESSED IN MHBTU'S/ACRE 



(10.00 PERCENT DISCOUNT RATE) 



MHBTU'S)ACRE 



NET PRESENT UORTH 

NET FUTURE UORTH 

FUTURE COSTS 

PRESENT BENEFITS 

PRESENT COSTS 

BENEFITS/COSTS 



453.42 

7911.98 

4882.34 

733.22 

279.80 

2.62 



ENERGY BUDGET — 4' X 4' SPACING — IRRIGATED 

ENERGY ANALYSIS FOR A 1000 ACRE OPERATIONAL HYBRID POPLAR PLANTATION 

TOTAL PROJECT ACREAGE IS DIVIDED INTO 10 TRACTS OF 80-120 ACRES EACH. 

90 PER CENT OF THE TOTAL PROJECT ACREAGE IS PLANTED. 

TREES/PLANTED ACRE: 2600 10" UNROOTED CUTTINGS 

ROTATIONS: (1) 10 YEAR AND (4) 5 YEAR COPPICE ROTATIONS 

YIELD: 7 DRY TONS/PLANTED ACRE/YR. FOR THE TEN YEAR ROTATION 

8 u " " " " " " FIVE YEAR ROTATIONS 
IRRIGATION: 10 EFFECTIVE INCHES/PLANTED ACRE/YEAR 
FERTILIZATION: 100 LBS. OF NITROGEN/PLANTED ACRE/YEAR 
NOTE: ALL ENERGY INPUTS AND OUTPUTS BELOU ARE EXPRESSED IN MhBTU'S/ACRE 



SENSITIVITY ANALYSIS 

(10.00 PERCENT DISCOUNT RATE) 
MHBTU'S/ACRE 



CHANGE IN NPU 



DUE TO A 10.00 
PERCENT CHANGE IN 



1 


SITE PREP 


.12 


2 


PLANTING 


.15 


3 


IRRIGATION 


8.80 


4 


FERTILIZAT'N 


3.14 


5 


UT HARVEST 


.79 


6 


FORG. HARVEST 


.63 


7 


HAULING 


.45 


8 


DRYING 


13.90 


9 


UOOD ENERGY 


73.32 



38 



ENERGY BUDGET -- 4' X 4' SPACING — NON-IRRIGATED 

ENER6Y ANALYSIS FOR A 1000 ACRE OPERATIONAL HYBRID POPLAR PLANTATION 

TOTAL PROJECT ACREA6E IS DIVIDED INTO 10 TRACTS OF 80-120 ACRES EACH. 

90 PER CENT OF THE TOTAL PROJECT ACREA6E IS PLANTED. 

TREES/PLANTED ACRE: 2600 10 M UNROOTED CUTTINGS 

ROTATIONS: (1) 10 YEAR AND (4) 5 YEAR COPPICE ROTATIONS 

YIELD: 7 DRY TONS/PLANTED ACRE/YR. FOR THE TEN YEAR ROTATION 

8 " " »»..» FIVE yEAR ROTATIONS 

IRRIGATION: NONE 

FERTILIZATION: NONE 

NOTE: ALL ENERGY INPUTS AND OUTPUTS BELOU ARE EXPRESSED IN HHBTU'S/ACRE 

ANALYSIS INPUTS 





UNIT OF CURRENCY 








HMBTU 


S 












LAND AREA 


1. 


00 


ACRE 


















PROD. PERIOD 


30 






















DISCOUNT RATE 


10. 


00 


PERCENT 














ANNUAL 


CHANGE IN COSTS 







PERCENT 














ANNUAL CHANGE IN BEN. 







PERCENT 
















AMOUNT AND TIHIN6 OF 


COSTS 


AND 


BENEFITS 










NMBTU S/ACRE 
















PUT NO 


NAHE 


AMOUNT 


ANN. 


RATE 


OF 


INFL. 


YEAR(S) 




1 


SITE PREP 


.52 






















1 
1 


SITE PREP 
SITE PREP 


.59 

.19 














1 
2 








2 


PLANTING 


1.64 













1 








3 


UT HARVEST 


10.24 













10 








4 


FORG. HARVEST 


5.85 













15 


20 


25 


30 


5 


HAULING 


3.26 













10 








5 


HAULING 


1.86 













15 


20 


25 


30 


6 


DRYING 


100.30 













10 








6 


DRYING 


57.31 













15 


20 


25 


30 


7 


UOOD ENERGY 


529.20 













10 








7 


UOOD ENERGY 


302.40 













15