Esthetic evaluation of timber harvesting in the northern Rockies : a progress report

Historic, archived document

Do not assume content reflects current
scientific knowledge, policies, or practices.

INTERMOUNTAIN FOREST & RANGE EXPERIMENT STATION
507 — 25th STREET, OGDEN, UTAH 84401

e
oe
iss

. oO 7
USDA Forest Service AP, Zz
Research Note INT-203 ees Z,
1976 i Se :
ESTHETIC EVALUATION OF TIMBER HARVESTING Keine oe
IN THE NORTHERN ROCKIES--A PROGRESS REPORT! =)

Dennis L. Schweitzer, James R. Ullrich, and Robert E. Bengon2

ABSTRACT

Panels of judges have been evaluating the esthetic
dimenston of harvested areas in the Northern Rocktés. Studtes
conducted in Wyoming and Montana agree with intuttton tin that
forest scenes are generally liked less as the evidence of
man's activities tnereases.

OXFORD: 615
KEYWORDS: harvesting (silvicultural) systems,
esthetics, visual impact

The appearance of forest areas that have been harvested is an important determinant
of the timber harvesting systems that will be used on the National Forests in the future.
At the present time, esthetics are considered in planning timber harvesting through
formal agency guidelines and through the judgment of landscape architects in laying out
harvesting units. Our modest research effort is intended to supplement these procedures.
This report is intended both to present our findings to date and to let the reader know
the nature of our work.

IMost of this work has been carried out under a cooperative agreement between the
University of Montana and the Intermountain Forest and Range Experiment Station. We
acknowledge the contributions of Maureen F. Ullrich and Roy F. Touzeau of the University
of Montana in collecting data and developing the methodology reported here.

The authors are, respectively, principal economist, Intermountain Station,
Associate Professor, Psychology Department, University of Montana; and research for-
ester, Intermountain Station. Schweitzer and Benson are stationed in Missoula, Montana,
at the Forestry Sciences Laboratory, maintained in cooperation with the University
of Montana.

Numerous questions need to be answered before our efforts will yield answers that

can be consistently used in complex decisions involving timber harvesting practices.
These questions include:

1. Do individuals and groups with differing backgrounds differ widely in their
perceptions of, and judgments about, the visual appeal of harvested areas? If so, how
do they differ?

2. Do judgments of visual appeal depend upon the physical perspective or vantage

point of the viewer?

3. Can judgments of case study areas be usefully quantified, are they reliable,
and can they be generalized for application elsewhere?

In cooperation with researchers of the Rocky Mountain Forest and Range Experiment
Station at Tucson and the University of Arizona, we are attempting to answer these
questions. For the immediate future, our major efforts will lie essentially in adapting
already-developed methodologies to the particular need of evaluating alternative harvest-
ing systems in the rugged topography of western Montana.

An individual's visual perception of a harvested area depends in part upon his
physical position relative to that area. We have divided our research efforts into
determining

1. The impressions received when near or within a stand which has recently been
harvested (that is, "near-view''), and

2. the impressions received when a stand is viewed from a distance, as from across
a valley ("far view").

To date we have concentrated on obtaining quantitative and reproducible evaluations of
the first, the near-view perspective, because a well-tested technique has been available
(Daniel and Boster, in press). We are beginning preliminary work to validate the use of
this technique or some substitute for evaluating far views; this may well be more
Significant in the Northern Rockies in light of the rugged topography.

TECHNIQUE USED

To date we have relied on the Scenic Beauty Estimation method (SBE) to define
preferences for harvested areas; the mechanics of this technique, its validation and
a wide range of applications have been thoroughly described by Daniel and Boster (in
press). Essentially, the technique amounts to taking color slides or photographs of
areas of interest and having groups of judges report on their degree of like/dislike for
each. The process can be broken down into the following steps:

1. Selecting areas to be evaluated.--In the next section we report on evaluations
of newly logged areas on the Teton National Forest in Wyoming and on the Bitterroot
National Forest and Coram Experimental Forest (Flathead National Forest), both in
Montana.

2. Representing the areas as photographic sltdes.--Because it is impractical to
transport many viewers to sites that require evaluation, we have chosen to represent
those sites through photographic slides. In our work, slides have been selected to
represent the views that would be seen by an observer walking near or through the
treated areas. Esthetic judgments seem to be unaffected by a rather wide variation in
photographic quality, perhaps because of the radical differences between clearcut,
partially cut, and uncut stands that we have compared. Presumably, photoquality would
be more critical if we were attempting to measure more subtle differences, but this
point has not yet been validated.

3. Selecting panels of judges to evaluate the sltdes.--Typically, evaluations are
collected from a group of 25 or more judges at one time. Most of our work has been done
with volunteer undergraduate students attending the University of Montana. Ullrich and
others (1975) found that for a single set of slides of harvested and unharvested
old-growth stands in Wyoming and Montana, essentially identical preferences were
assigned by:

--psychology undergraduates at the University of Montana,
--psychology undergraduates at the University of Michigan at Flint,
--Montana elementary school teachers, and

--Montana Forest Service researchers.

Daniel and Boster (in press) conducted extensive studies with a wide variety of
interested groups and found that, while numeric ratings vary somewhat by groups, average
group evaluations ranked different landscapes in a consistent manner.

4. Colleeting judgments from the panels.--The judges have been shown series of
Slides and asked to rate their preferences on a categorical scale from 0 to 9. There
are a number of slides of each of several treatment areas. From all the photographic
replications for a given area, one slide from each area is randomly selected to form a
block; the order of presentation of the slides within each block is also randomized.
Each block then consists of one picture of each area. Blocks are then shown until all
of the photographic replications are used. This procedure usually requires a total of
about 100 slides and 20 minutes of time.

5. Reducing judgments to averages for selected areas on the ground.--The judgments
provide a set of numbers reflecting the panel members' perceptions of esthetic attrac-
tiveness. Daniel and Boster (in press) have pointed out that working with raw scores
alone can be misleading if different panels use the 0-to-9 rating scale in different
ways, especially if some tend to use the upper end and others the lower end of the
scale. The authors have developed a method of mathematically transforming raw scores
to avoid this problem. In addition, their transformed numbers meet all theoretical
requirements for calculating means and variances and applying standard statistical
tests of differences. Because, in our work to date, the results of working with such
transformations are no different than the results obtained directly from the raw
scores, no transformations are included in the present paper. (The transformed scores
are available upon request. )

6. Confirming that the average evaluattons of treated areas from most- to least-
preferred would be the same tf made on the ground.--To date we have not independently
confirmed that this is true. Research carried out in the southwest and reported by
Daniel and Boster (in press) strongly suggest the same esthetic judgments would be made
based on slides as would be made by the same observers in the forest.

We have applied this general technique in the three areas described in the next
section. For convenience of presentation and ease in reading, we have adopted the
convention that all raw data scores are presented as if judges had used a 0-to-9 cate-
gorical scale from ''dislike" to "like.'' In fact, in some instances the judges were
instructed that 0 represented "like" and 9 represented "dislike." (The literature of
psychological testing suggests judges would be only slightly influenced by scale
reversals in ranking treatments.) We have used this convention in our most recent
work and intend to use it in the future.

AAR A IND MING oF SANE AINOY LETINUYUUISL YVUINL LI/ ODO

We have evaluated logging practices in three different areas: 3

Area Prinetpal Compartson
Teton National Forest Alternative treatments of logging
northwestern Wyoming residues in old-growth lodge-
pole pine
Bitterroot National Forest Logging by horses, crawler-tractor
western, Montana skidders and wheeled skidders

in old-growth lodgepole pine

Coram Experimental Forest (Flathead Clearcutting and shelterwood
National Forest), northwestern cutting in old-growth fir-larch
Montana

TETON STUDY

In 1971, a study was begun to evaluate a system of harvesting mature lodgepole pine
in which virtually all the logging residue--branches, tops, and dead and cull material--
was yarded and chipped at the logging site (reported in detail in Benson 1974). The
study area is located in Wyoming near the Union Pass area of the Teton National Forest.

Four units of approximately 20 acres each were harvested by clearcutting. Two of
the units were logged following conventional practices for that part of Wyoming. Saw
logs to a 6-inch top were removed; the remaining material was left for burning on the
site. The other two units had 'near-complete'' removal of slash; in addition to taking
out the merchantable saw logs, virtually all the remaining material was yarded and
chipped. On the near-complete units, a feller-buncher and rubber-tired grapple skidder
were used in connection with an in-the-woods chipper.

Esthetic evaluations of six different logging and residue treatments were made by
a panel of students. They rated a series of slides taken to represent the views of an
observer hiking or driving alongside the treated areas on a 0-to-9 scale; summary
statistics are presented in table 1. The median ratings show that half the judges rated
the treatment higher and half lower than the value shown.

From these data, the judges most preferred scenes of unharvested areas and least
preferred scenes of a recent clearcut where logging residues had recently been piled
and burned. Clearcut areas where vegetative regrowth had started to mask the partially
burned residues or where the residues had been converted to chips that were then spread
throughout the harvested area were judged intermediate in esthetic value. The panel's
judgments agreed closely with independent evaluations made of the area by a Forest
Service landscape architect.

3Detailed physical descriptions of all study areas have been reported in the
internal report, ''Forest residues utilization research and development program: progress
report I,'' March 1975, 126 p. USDA For. Serv., Intermt. For. and Range Exp: StnaetiOcden,
Utah.

Table 1.--Evaluattons of narvested ana unnarvesveda OLaA-growln
lodgepole pine stands on the Teton Nattonal Forest

Treatment : Median rating

Edge between meadow and unharvested

old-growth stand 8.00
Unharvested old-growth stand 5.85
5 years after clearcutting with residues piled

and burned AZS
1 year after clearcutting with chips spread over

area 55910)
1 year after clearcutting with near-complete

removal of residues 5.165
1 year after clearcutting with residues piled

and burned Peles)

BITTERROOT STUDY

In 1974, a number of small units averaging about 1-1/2 acres were clearcut in
small-diameter, old-growth lodgepole pine on the Bitterroot National Forest to determine
the productivity of several logging systems including rubber-tired skidding, crawler-
tractor skidding, horse skidding, and a combination of horse and rubber-tired skidding.
Photographs were taken before and after harvesting to determine whether near-view
esthetic values varied by the skidding method used.

Because of the small size of the clearcut areas, the photographic slides were taken
from the periphery of each cutting unit, oriented inward. These slides were then
evaluated by 29 students (summary results are presented in table 2).

Higher scores indicate those skidding methods that were relatively liked. Although
the uncut stands were all preferred to any of the stands after harvest, no statistically
significant differences (regardless of the level of probability) were found among the
areas either before or after harvesting.

It is important to recognize the dangers of trying to extrapolate the results
from this small experiment to generalizations about the esthetic dimension of the
examined skidding systems. For example, the woody material left on the ground after
logging averaged nearly 1,000 cubic feet per acre, quantities which might well have led
to different esthetic evaluations than if only a few pieces had remained. Further,

Table 2.--Evaluations of four dtfferent skidding methods
used in the Bitterroot Nattonal Forest

Skidding method Mean rating

Before harvest : After harvest
Combination horse and rubber-tired skidder Sats 1.60
Crawler-tractor skidder 3.69 1.98
Rubber-tired skidder 5455 il al
Horse skidder S243 G7

the evaluations probably were influenced by the small size of the clearcut areas,
which averaged about 1-1/2 acres each, and by every photograph having a background of
uncut forest. We can only state with assurance that our particular study did not
detect differences among the appearances of the small areas that were harvested.

CORAM STUDY

In 1974, old-growth stands in the Douglas-fir and western larch forest type in
northwestern Montana were harvested under several different silvicultural prescriptions
to test the ecologic, economic, and esthetic consequences. Two stands each were
harvested by clearcutting, group selection cutting, and shelterwood cutting techniques.
Each stand was divided further into subtreatments by the manner in which the logging
residues and understory were to be treated before, during, and after harvesting.

PREHARVEST AND POSTHARVEST EVALUATIONS

To determine the extent of preharvest differences in the stands, a series of near-
view pictures were taken in June 1974 of areas to be shelterwood-cut and clearcut. (The
group selection areas were omitted. )

In the subtreatments designated as "weed and bundle" areas (where trees to a l-inch
diameter were removed), all 5-inch and smaller trees had been cut and tied in bundles
which were lying on the ground (and which were later removed from those areas during
overstory harvest). Generally, these areas contained small, freshly cut stumps as well
as the ends of the bundled trees; in addition, the forest canopies were more open than
in the other subtreatments.

Two physical constraints dictated the method of obtaining the photographic slides.
Because the subtreatments consisted of long, narrow areas on the ground, all photographs
were oriented along the long axis to avoid confounding judgments with backgrounds of
different areas. Because we had little idea of how the steep topography would influence
esthetic judgments, half of the photographs were oriented uphill and half downhill.

The photographs were shown to a panel of 29 students. The mean scores are pre-
sented in table 3. Based on Tukey's HSD test (Winer 1971), when evaluations were
appropriately pooled the following differences were found to be statistically signif-
icant at the 0.05 probability level:

--the to-be-clearcut unit was preferred to either of the future shelterwood
units;

--the areas with undisturbed understories were preferred to those containing
bundles of small trees; and

--photographs looking uphill were preferred to those looking downhill.

A general analysis of variance (assuming fixed effects) revealed significant inter-
action between treatments and pnotograpnic replications, that is, there were some slides
from generally liked areas that were disliked and there were slides from generally
disliked areas that were liked. We hypothesized that if a particular slide happened
to present an unattractive object in the foreground of a generally liked area that
slide would be relatively disliked. Because we want to establish overall ratings of
like-dislike for each of the treatment conditions, we have ignored the statistically
significant interaction among individual slides.

Table 3.--Preharvest esthette evaluation of Coram areas destined to be
shelterwood-cut and clearcut

i

Harvest method Understory? panera neon
orientation rating

Shelterwood Undisturbed Uphill O. 22
(11- 4) Downhill 5.49

Weed-and-bundle Uphill S27:

(11-3) Downhill Sr59

Shelterwood Undisturbed Uphill 6.20
(21-4) Downhill 5.60

Weed-and-bundle Uphill 6.01

(21-3) Downhill 4.38

Clearcut Undisturbed Uphill ZZ
(23-4) Downhill 6.41

Weed-and-bundle Uphill 6.56

(23-3) Downhill 6.32

1, difference of 0.72 between any two numbers in this table is statistically
significant at the 0.05 level. Smaller differences between pooled judgments, as
reported in the text, are significant.

2Numbers in parentheses are treatment and subtreatment designations.

After harvest was completed in the fall of 1974, photographs were taken of clearcut
and shelterwood cut areas (a mixup led to not completely replicating the pretreatment
areas). A group of 32 students* then evaluated a mixed group of precut and postcut
slides; the results are presented in table 4. Using the same statistical testing pro-
cedures as before, we concluded that:

--both shelterwood areas before harvest were preferred to either the shelterwood or
clearcut area after harvest; and

--the harvested shelterwood area was preferred to the harvested clearcut area.

Although uphill photographs were again favored over downhill photographs, the differ-
ence is not statistically significant (the understory in three subareas had been slashed
when postharvest photographs were taken; the understory in one other subarea had been
inadvertently torn up during harvesting).

*The judgments of two student panels, of 18 and 14 members, were pooled because no
statistically significant differences could be found between the group scores, and the
experimental conditions were judged by the experimenters to bé similar. The data
underlying (preharvest) table 3 and (preharvest-postharvest) table 4 were not pooled
because the panel evaluations were conducted in the fall and spring, times having
substantially different implications for student participation.

¢ 1
Harvest method Understory ae Me eee AUN
orientation Preharvest Postharve:
Shelterwood Weed-and-bundle Uphill Se26 =
(11-3) Downhill 5246 --
Undisturbed Uphill SAAS) aa
(11-4) Downhill Sls --
Shelterwood Weed-and-bundle Uphill 6.40 4.33
(21-3) Downhill 4.70 5185)
Undisturbed Uphill 6.06 4.60
(21-4) Downhill Biel) Syeellail
Slashed Uphill - 32-24
(21-1) Downhill - 3.54
Slashed Uphill - 5;..99
(21-2) Downhill - 3.48
Clearcut Weed-and-bundle Uphill = =
(13-3) Downhill - 1.89
Unprotected Uphill ~ 5105
(13-4) Downhill - ANNES)
Slashed Uphill - 2.04
(13-1) Downhill - AWS eFAS)

eut and clearcut areas at Coram

Table 4.--Preharvest and postharvest esthetic evaluation of shelterwood

1a difference of 1.12 between any two numbers in this table is statistically
Significant at the 0.05 level. Smaller differences between pooled judgments, as reported
in the text, are significant.

OTHER CORAM STUDIES

Several other exploratory tests have been run on the Coram Experimental Forest as
preliminary steps in developing techniques for assessing the esthetic values of leave
strips between forest roads and harvested areas, of forest roads themselves, and of
distant views of harvested areas. These tests were based on adaptations of the pre-
viously discussed near-view methodology.

Leave strtp analysts.--A study was undertaken to determine the effects of leave
strips between the observer and the cutting units after harvesting. Photographs were
taken defining five subjectively determined treatment conditions: no leave strip, light
leave strip, and heavy leave strip next to a clearcut block (block 13), and presence or
absence of a leave strip next to a shelterwood block (block 21). Ratings of these
slides were obtained from 28 students; mean ratings are presented in table 5.

For the clearcut block, the greater the amount of leave strip the greater the
preference. This was not true for the shelterwood block, probably because a large
number of the slides with the leave strip contained views of a road in the foreground,

Table 5.--Evaluations of leave strips adjacent to clearcut
and shelterwood cut areas

: : ; : Mean
Treatment : Leave Strip : ge
Clearcut Heavy leave strip 4.46

Block 13 Light leave strip 2.98
No leave strip Hg JUS)

Shelterwood leave Strip present S92
Block 21 No leave strip AOD

while those without the leave strip did not. The subsequent comparisons were correlated
almost perfectly with the presence or absence of the road in the photographs: we
hypothesize the presence of the road led to lowered ratings.

Near-vitew road study.--Photographs were taken of four different road sections by
first shooting along the road and then pivoting the camera 45° between shots until a
circle was completed. Evaluations indicated the most preferred road segment was a
section of an old road with established vegetation. As more bare earth and rock and
less vegetation were included in the slides, ratings decreased. This conclusion held
both for individual slides taken at one point and for groups of slides taken on dif-
ferent road segments.

Far-view study.--To this point, all reported work was based on photographs taken in
or at the edge of treated areas. However, distant views predominate in the mountainous
regions of northwestern Montana. This study was an exploratory attempt to determine how
to best evaluate the esthetic values of such views of the Coram logging treatments.

Photographs were taken from the ground and from a helicopter using a 35-mm camera
(with a normal lens). From this pool of slides a sample of 39 was shown to a panel
of 16 students. These slides were selected as follows:

1. Show at least three different perspectives of each unit both before and after
cutting. (Some photographs were made from the air because it was not possible to get
enough different perspectives from the ground.)

2. Show only a single treatment, excluding other cutting units and irrelevant
scenic backgrounds and foreground foliage.

3. Use only slides of high photographic quality.

All three criteria were to some extent violated, most notably in our inability to
exclude areas outside the logged area.

While the perception of a harvested area is influenced by the kind of forest and
prominent land features surrounding the area, our interest was centered on evaluating
particular harvesting systems. To define the area to be evaluated, supplementary line
drawings were made of each photograph showing prominent terrain features; the area of
interest was enclosed in a bold, dotted box. The judges were instructed to rate only
that portion of the slide that corresponded to the area within the enclosure and to
ignore the surrounding land. To provide a reference point, slides taken before harvest
as well as after harvest were included. The mean ratings are given in table 6.

Table 6.--Evaluattons of far views of Coram harvesting areas by panels of
students and professional foresters (nwnber of slides in parentheses)

Mean rating

Harvest area 3 Preharvest : Postharvest
Students : Professionals : Students. : Professionals

Shelterwood

Block 11 6.91 (8) Ua ASS 2 1GS)) 1220
Shelterwood

Block 21 6.55 (8) T3226 S382, 165) 6.72
Group selection

Block 12 Aad i359) 1355 : -- --
Group selection

Block 22 -- -- 2260) (5) Sie:
Clearcut

Block 13 -- -- .81 (4) 455
Clearcut

Block 23 -- -- 100: (5) 4.14

Two groups of judges independently evaluated the slides: university undergraduates
and a group of foresters attending a professional meeting. While both groups assigned
relatively high ratings to the preharvest slides, the professional foresters assigned
substantially higher ratings to the postharvest slides than did the students. We
hypothesize this was primarily due to the professionals having undergone a 3-hour
introduction to many aspects of the experimental work at Coram just prior to their
acting as judges. We do not know whether their general technical expertise influenced
their evaluation.

The mean scores indicate an esthetic preference for the shelterwood cuttings and
a relative dislike for the clearcuts, with the group selections being intermediate. We
conducted no statistical testing, however, because of the small number of slides and
because of the unknown extent to which the several uncontrolled influences noted above
may have altered the ratings.

10

SUMMARY AND CONTINUING RESEARCH

The results reported in this paper have been obtained from several case-study areas.
Because most of the comparisons made between cut and uncut stands represent radically
different conditions of the forest, our test results generally agree with our intuition:
as the amount of downed wood or the evidence of man's activities increases, forest
scenes are liked less by most observers. As we explore more subtle differences among
scenes, it is likely that the testing procedure will lead to less "obvious" conclusions.

We are beginning a study to determine the manner in which esthetic judgments change
as vegetation changes after timber harvesting. The study will include periodic evalua-
tions of the harvested areas on the Coram Experimental Forest and on the Teton National
FOresits

Within any particular scenic view, we are trying to define the functional relation-
ship between like-dislike evaluations and specific features of those views. The current
experimental technique utilizes choices among simultaneously projected pairs and triads
On pLecures’.

Finally, distant-view evaluation techniques are being developed. Given the rugged
topography in the Northern Rockies, we feel such techniques will be the most useful of
all in providing information for the further development of generalized land manage-
ment guidelines.

PUBLICATIONS CITED

Benson, Robert E.
1974. Lodgepole pine logging residues: management alternatives. USDA For. Serv.
Res. Pap. INT-160, 28 p. Intermt. For. and Range Paresh: Ogden, Utah.

Dannicie erry Genwand Ron S. Booster.
Measuring scenic beauty: the SBE Method. USDA For. Serv., Rocky Mountain
fOmmdicduhange Exp. Stn., Fort Collins, Colio. (im press) .

UlirtcheeJancseR- sa Maureen Fo Ulinach, Dennis LU. Schweitzer, ‘Roy F: Touzeau, and
Harriet M. Braunstein.
1975. Aesthetic evaluation of forest scenes in the northern Rockies by different
groups of judges. Proc. Symp. on Environ. Effects on Behavior, William D. Bliss
(compiler), Human Factors Soc. and Mont. State Univ., p. 119-126.

Water. sBin td
1971. Statistical principles in experimental design. McGraw-Hill, New York. 907 p.

ial

a
bEseC olsip a Aaa i aie e
> 107 Meee
1 Vanier Car Atted avipe %
. gor 724 ELST SA ead,

alll + wen 1fo-Whe Hy eee
wey air you Fae AF kl le OF s
ee thei ab Ault
ry ' A PIM Gide GARE CNN AS DIOR SORE STIG, a
Me To ee he VO yeu Weiter Gt via, inte bas gi
90; clea > oa? yar ghining wuerngleee: hl
ae ; AC? bile anne Lae Wy bees whit < ;
Te Mica vests ae eter mai
‘

yi ee r ater a) ' JPA a Sheas th a 7s
shee fete a shel Lar aed. es
reeb La? WoT QO oi ae agit be

9

- 1 [3%
r-* “
itt OW Pte 5 yet pete
A PEE Meese. |
: 3% R Sutilfn» is St

ye hae es )

7 ' a -" ow ais Iae a
“gee io sept mined Ppa