Historic, Archive Document

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

United States
Department of
Agriculture

Forest Service

>r&'

Forestry

Research

April 1987

A report for land managers on recent
developments in forestry research at
the four western Experiment Stations
of the Forest Service, U.S. Depart¬
ment of Agriculture.

Forestry

Research

West

n This Issue

page

The heli-stat legacy: A new

model for timber harvest

planning 1

A new program for riparian
research 7

Hurdygurdy— California’s first
basin-wide study of fish
habitat aims to improve
stream management 1 1

Helping bald eagles keep a
foothold in the Southwest 16

New publications 21

Cover

Despite the crash of the Heli-Stat
airship last July, the USDA Forest
Service has not lost interest in using
airships and helicopters for yarding
timber. Scientists at the Pacific North¬
west Station have developed a
computer model that permits a com¬
prehensive look at timber harvesting
operations and offers information on
what could become a new way of
evaluating and coordinating all
aspects of timber harvesting. Here, a
helicopter is being used in validation
tests where logs are weighed on the
scales at left. Read more about this
exciting research on page 1 .

To Order Publications

Single copies of publications referred
to in this magazine are available with¬
out charge from the issuing station
unless another source is indicated.
See page 26 for ordering cards.

Each station compiles periodic lists
of new publications. To get on the
mailing list, write to the director at
each station.

To change address, notify the maga¬
zine as early as possible. Send mail¬
ing label from this magazine and new
address. Don’t forget to include your
Zip Code.

Permission to reprint articles is not
required, but credit should be given
to the Forest Service, U.S.D.A.

Mention of commercial products is
for information only. No endorsement
by the U.S.D.A. is implied.

Western Forest
Experiment Stations

Pacific Northwest Research Station
(PNW)

PO. Box 3890
Portland, Oregon 97208

Pacific Southwest Forest and Range
Experiment Station (PSW)

1960 Addison St.

Berkeley, California 94704

Intermountain Research
Station (INT)

324 25th Street
Ogden, Utah 84401

Rocky Mountain Forest and Range
Experiment Station (RM)

240 West Prospect Street

Fort Collins, Colorado 80526-2098

The heli-stat
legacy: A new
model for
timber harvest
planning

by Dorothy Bergstrom
Pacific Northwest Station

Preparations by Forest Service re¬
searchers to evaluate an airship for
possible use in yarding timber have
produced an unexpected bonus: a
computer model that permits a more
comprehensive look at timber harvest
operations than has been possible in
the past and provides the foundation
for what could become a new way of
evaluating and coordinating all parts
of the timber harvest operation.

The model was developed under the
direction of Engineer Michael Lam¬
bert of the Pacific Northwest Station
in his capacity as program analyst
for evaluation trials of the Heli-Stat,
the prototype yarding airship that
crashed and burned in July 1986.
Initial plans were to bring the Heli-
Stat to the West Coast and evaluate
its aerodynamic performance in
logging trials.

The Heli-Stat, consisting of an
aerostat (balloon) attached to four
helicopters, was intended to demon¬
strate the feasibility of ferrying logs
from remote roadless areas in larger
loads, over longer distances, than
has been possible with other yarding
methods.

The Heli-Stat during hover test in New Jersey
July 1, 1986, just before it crashed.

Although the testing opportunity was
lost along with the Heli-Stat, the in¬
vestment in planning for the evalua¬
tion was not; a computer submodel
is now available for evaluating the
aerodynamic capabilities of the next
generation of yarding airships. Per¬
haps more important, a new plan¬
ning simulation model provides the
conceptual framework for integrating
data from many sources, including
existing timber harvest models.
Although it is oriented to airship and
helicopter yarding, the model pro¬
vides a common basis for comparing
a range of combinations of timber
harvest processes, from initial plan¬
ning to reforestation. It is based on
information from specific situations—
not on averages. When supplied with
information about a particular logging
operation, the model can help the
planner, logging engineer, or logging
operator identify the combination of
equipment, workers, and procedures
that will achieve the greatest produc¬
tivity at the least cost.

Evaluating an airship
for logging

In 1981, Lambert was given an
assignment: analyze the technical
and economic feasibility of logging
with large airships; develop methods
for estimating the weight of logs and
trees; and prepare for field trials of
the Heli-Stat.

The assignment meant translating
into reality the intriguing idea of an
airship lifting logs out of otherwise
inaccessible timber stands. It meant
examining every detail of what would
be a new type of timber harvest
operation. It required not only de¬
scribing the capabilities needed in an
airship that could yard timber effici¬
ently; it meant anticipating how the
usual logging operation would be
changed. In addition to new possi¬
bilities, yarding with an airship would
introduce new requirements for sup¬
port equipment and workers.

Although a major problem was to
find an efficient method for assem¬
bling loads of logs for lifting by the
airship; there would be other con¬
siderations. For one thing, workers
would need to get to the logging
site. For another, landings would
have to be bigger to accommodate
not only the airship but all the equip¬
ment needed for its support. Some
operations might be done better at
locations other than the usual ones.
For example, if it became feasible to
ferry whole trees from the logging
site, bucking and delimbing could
probably be done more efficiently at
the landing.

Mechanical Engineer Michael Lambert.

Assembling loads would require
accurate estimates of log weight-
made in the field. It would also re¬
quire a machine to move trees and
logs on steep slopes. Crawler trac¬
tors, rubber-tired log skidders, and
other conventional skidding equip¬
ment cannot maneuver well enough
on steep slopes, or are likely to
cause serious soil disturbance or
other environmental damage.

Processing large turns of logs and
trees at the landing, without delaying
the airship, would mean figuring out
the rates of all steps in log handling.
How, for example, would frequency
of turns affect the type of equipment
and number of workers needed at
the landing? How would distance
between the cutting area and the
landing affect the arrival times of
turns?

In short, it would be impossible to
evaluate the performance of a log¬
ging airship without both visualizing
the complete operation and examin¬
ing each step in detail to find out
how it relates to and affects all the
other steps and their costs.

Because logging equipment is ex¬
pensive to own, operate, and move
around, Lambert decided to figure
out as much as possible in the office
before any trials on the ground or in
the air. Doing this would also avoid
setting up support equipment by trial
and error, risking expensive delays,
and possibly having to run tests
more than once.

To do this complicated job in the
short time before logging trials were
then scheduled for the West Coast,
Lambert decided to use the com¬
puter to simulate all steps in the
aerial logging operation. This meant
using the computer on a broader
scale, with more precise data and
greater emphasis on weight, than
previous computerized logging
routines had required. There were no
simulation models that could be used
or extrapolated to project the perfor¬
mance of large airships. Much of the
earlier logging planning had been
based on averages, past experience,
and rules of thumb.

2

Green density is measured at the logging site
with a portable xyiodensimeter invented by
Gary Bergstrom, Rogue River National Forest.
(1) Cores are taken at various heights from

LOGPAC: A conceptual
formulation

LOGPAC (for Logging Productivity
and Costs) is the name of the model
that began as a way to describe the
aerodynamic performance required
of a logging airship but expanded to
provide the framework for a family of
submodels that together express
mathematically all phases of logging,
from sale preparation to slash treat-
nent and reforestation. The model
provides the framework for accom¬
plishing two important functions: (1)
simulating the performance of an
aerial vehicle with the desired fea¬
tures and capability before it is built,
and (2) comparing the productivity
and costs of different logging plans,
regardless of yarding method.

several sample trees. (2) Each core is trimmed
to a fixed length, with the correct proportion of
sapwood to heart wood, and inserted into a
calibrated hydrometer. (3) The height at which

LOGPAC offers a systematic, uni¬
form, reliable method of tracking pro¬
ductivity and costs. It provides a
standard means of communicating
about logging production and costs
by people of different backgrounds
and roles in the timber harvest
operation.

The advantages of LOGPAC over
other planning and accounting
methods are: (1) It figures produc¬
tivity and costs on the basis of the
logs being handled, on a specific
site, under the environmental condi¬
tions at hand, by machines and peo¬
ple whose performance is quantified.
Productivity and costs are not based
on average values. (2) It contains a
simulation clock that makes it possi¬
ble to analyze the effects of relative
production rates on successive ac¬
tivities, so material handling will flow
efficiently, without expensive bottle¬
necks. (3) Data the user supplies are
standardized but flexible; they can

the hydrometer floats in distilled water in¬
dicates the density of the green wood.

be easily modified and improved. (4)
Numerous analyses can be made
and compared by making simple
changes in the input data. (5) The
model contains automated reminders
for the user and makes sure that all
items are considered in the same
way for each analysis.

How the model works

A person using the model carries on
a dialogue with the computer. The
first step is to supply data from the
timber cruise on the size and num¬
ber of trees in a cutting unit.

The model then asks what log
lengths are preferred, what is the
maximum weight, and which is the
more important constraint.

3

After getting this information, the
model will "cut" all the trees into
logs (unless whole-tree yarding is
selected), report the length and
weight of logs in each tree, and
recommend a bucking strategy. It
will tell how many pieces are to be
yarded and how many trips the
yarder will have to make. It will also
tell how long felling will take with a
given number of fallers.

The dialogue continues through a
series of questions and answers that
cover the range of details about the
operation, down to and including
such items as the cost of owning the
logging equipment (investment,
depreciation period, taxes, insurance)
and weather (altitude, wind speed
and direction, temperature, humidity)
at both the loading and landing sites.

Based on information and relation¬
ships built into the model and data
added by the user, the computer
simulates, integrates, and reports on
all phases of the logging operation,
finishing with a summary of costs,
tons of logs removed, and the bot¬
tom line— cost per ton of product.

Following are examples of the type
of information the model will report
when yarding is done aerially:

— Total number and weight of logs
in the harvest unit

— Time and pounds of fuel used per
turn, based on airship configura¬
tion, load weight, and flight profile

— Number of turns per refueling
cycle

— Number of fuel cycles per harvest
unit

— Payload per turn, number of prob¬
able aborts, and overall load fac¬
tor (actual weight carried divided
by theoretical carrying capacity)

— Weights of unmerchantable
residue (bole and crown)

4

— Number of days needed to
harvest the unit

— Ownership and operating costs for
each activity and for the total
operation

How the model
can be helpful

One big advantage of computer
simulation is that it allows the user to
check out all components of a timber
harvest operation and identify poten¬
tial bottlenecks before logging
begins. The size of potential back¬
logs aids logging planners in specify¬
ing sizes of equipment, crews, and
landing areas to increase production
and reduce costs. In aerial logging,
for example, queues of logs may
build up on the landing as a result of
differences in turn arrival times, crew
sizes, and production rates. Such
backlogs may suggest shifting idle
time to less costly equipment;
queues on the landing may be
preferable to delays for the airship.

Another comparison that is important
to simulate when yarding is done by
air is the cost of transporting workers
to and from the logging site. If the
fallers and choker-setters must be
flown by helicopter to their work site,
one set of commuting expense
based on distance is used; if they go
by truck and walk part way to the
harvest site, the expense is probably
different.

Simulation also permits examining
the advantages and disadvantages
of landing sites. The costs of driving
log trucks can be compared with
yarding costs for different locations.
The effects of changing the driving
or yarding distance a few miles, or a
few hundred feet in elevation, are
easily identified.

LOGPAC can help a planner use
existing equipment to the best ad¬
vantage. It can become a compre¬
hensive spread sheet that shows
how changes in any part of the sys¬
tem affect other parts. For example,
it can indicate the effects of down
time and other delays and illustrate
how stretching out a job increases
costs. It can project costs in relation
to the amount of product produced
and help the planner relate produc¬
tion costs to expected market values
and risk. Perhaps most important, it
can make comparisons and help
identify the logging system that will
minimize costs.

Flexibility is built in

LOGPAC is a generic model,
capable of expansion, change, and
refinement. The user can change
any item and substitute better infor¬
mation. For example, the cost per
hour of owning and operating a
front-end loader or its production rate
can be entered as known data, or as
a calculating submodel inserted
when better information becomes
available or conditions change.
Models already available for different
types of equipment can be incorpor¬
ated as, for example, a software pro¬
gram for skidder yarding.

Users can also adapt the LOGPAC
family of models to different data
processing systems and provide new
productivity algorithms or their own
cost accounting subroutines.

Where the subroutines
and data in model
came from

In developing LOGPAC, Lambert
pulled together information already
available from a number of sources,
some of it from the Pacific Northwest
Station Other Station scientists and
several cooperators assisted in con¬
ducting individual field studies on
activities for which data were lacking.

Using chokers the same length saves time,
fuel, and money over using chokers of differ¬
ent lengths. Less energy is required to drag
the logs free of the ground, less vertical ascent
is required to clear the ground, there is less
aerodynamic drag, and load-release is quicker
at the landing. This method requires pre-

Cooperators included Intermountain
Station, the Pacific Northwest and
Pacific Southwest Regions of the
Forest Service, Montana State
University, University of California,
and Oregon State University. Tech¬
nical assistance with the computer
studies for the aerodynamic perfor¬
mance model came from a private
aerospace firm.

Estimating weight

The foundation of LOGPAC is infor¬
mation from the timber stand and the
weight estimating routine. Weight
assumes vital importance in aerial
yarding. If the equipment is not used
to capacity, the operation is more ex¬

bunching turns before yarding. LOGPAC
simulation reveals how one such improvement
in efficiency affects the overall efficiency of a
logging operation. Planners can analyze the
costs and benefits of different processing
steps.

pensive than necessary; if the equip¬
ment is overloaded, loads will be
aborted, wasting time and adding to
costs.

Research at the Pacific Northwest
Station has produced methods for
estimating tree and log volume and
developed related tools and com¬
puting capability to support methods
of estimating weight in the field. This
research involved taking trees apart
and weighing the parts after they
were graded in size classes. It also
used a device to measure green
density that was invented by a Forest
Service employee in the Pacific
Northwest Region. The result is a
submodel that can predict the weight
of logs and recommend bucking

strategies that will optimize the num¬
ber of preferred length logs within
the constraints of equipment and
handling capabilities.

The number and sizes of logs to be
handled form the basis for simula¬
tions of the number of workers
needed and their production rates.
Other studies provided information
that helps quantify the most efficient
and cost-effective combination of
people, equipment, and operations,
based on costs of assembling loads,
lifting and transporting loads, and
handling logs at the landing— all ex¬
pressed in costs per ton of yarded
logs.

Validation trials

The data in most of the submodels
were validated and quantified in
several logging operations in Oregon
and California between May 1981
and November 1983.

The most recent trial was a logging
operation on the Rogue River Na¬
tional Forest in late 1983 that was
planned as a dress rehearsal for the
Heli-Stat evaluation. The yarding
vehicle was a large helicopter. Data
collected were used to validate the
helicopter part of the airborne perfor¬
mance model. The trial also built
confidence in the validity of the
model as an approximation of the
yarding capacity of an airship that
combines an aerostat with one, two,
or four helicopters.

Tests on the Gifford Pinchot National
Forest in 1982 and the Mount Hood
National Forest in 1983 validated the
log weight estimating methods that
were developed during studies on
the Wind River Experimental Forest
in 1981-82. Other validation studies
included: load factors on the El
Dorado National Forest, load assem¬
bly methods for small timber on the
Gallatin National Forest, landing
operations on the Rogue River and

5

AERODYNAMIC MODEL

PROGRAM INPUTS

PROGRAM OPE RATIONS

PROGRAM APPLICATIONS

THESE TWO BLOCKS CAN BE USEO
INOEPENOENTLY AS A PRELIMINARY
DESIGN TOOL FOR VEHICLE OESlGN
SENSITIVITY STUDIES AND FOR
DEVELOPMENT OF NEXT GENERATION
QUAD ROTOR CONFIGURATIONS

THE COMPLETE PROGRAM IS USEO TO
EXAMINE THE ECONOMIC VIABILITY
OF NEW QUAD ROTOR DESIGNS IN
THE TIMBER HARVESTING OPERATION
ANO THE SENSITIVITY OF THE
ECONOMICS TO VARIATIONS IN VEHICLE
OESlGN PARAMETERS ANO OPERATIONAL
STRATEGIES

THIS BLOCK CAN BE USEO ALONE TO
STUDY THE ECONOMICS OF PREOEFINEO
CONFIGURATIONS ANO TO OEVELOP
OPTIMUM STRATEGIES FOR TIMBER
HARVESTING

Olympic National Forests, and pro¬
duction rates for the Spyder (the
machine for assembling logs on
steep slopes) on the Olympic
National Forest.

Present status
of the model

In its current form, parts of the model
are ready to analyze harvesting
systems and help agency people
protect costs and help bidders come
up with realistic bid prices. Not all
submodels have been validated.

With a little more work, there could
be routines for other operations—
skyline yarding, for example.

Lambert says the model needs to be
refined as people use it with different
species of timber, under varied en¬
vironmental conditions, with new
types of equipment, or new config¬
urations of machines and workers.

One of the improvements needed is
to incorporate road costs. The frame¬
work for adding these is in place. If
the purpose of yarding with an air¬
ship or helicopter is to reach places
where it would be difficult or environ¬
mentally unacceptable to build
roads, then adding the cost of build¬
ing and maintaining roads to the
economic evaluations will give a
clearer picture of the cost of various
types of aerial yarding.

LOGPAC can also evaluate ways of
treating fuel or residue and plans for
multi-product marketing.

Lambert has made practical use of
the model. For example, in June
1986 it was used to help evaluate a
salvage sale in a municipal water¬
shed on the Mount Hood National
Forest. The sale was designed as a
two-stage operation— part helicopter,
part cable— to protect water quality.
But there were no bids on the sale
until after LOGPAC was used to ana¬
lyze the costs of various new combi-
6

nations of logging processes and
slash treatment that were proposed.

Lambert expects that a yarding air¬
ship will be developed within the
next few years. In fact, two are cur¬
rently in the flying prototype stage in
the United States. In other countries,
there is a good deal of interest in
aerial vehicles for moving people,
ship cargo, and prefabrications.
Machines are being developed in
Russia, Japan, Great Britain, Ger¬
many, and Canada.

Lambert has written about evaluating
airships for yarding and about the
family of models. For reprints of the
following journal articles, write
Michael B. Lambert at the Pacific
Northwest Station or call him (FTS
423-2030; or 503/231-2030). Reprints
available inciude: Development and
trials of a heavy-lift airship used for
logging, ASAE Pap. 81-1585, In Pro¬
ceedings, Winter Meeting, American
Society of Agricultural Engineers,
1981; Airship logging: parameters af¬
fecting load factors, In Transactions
of the ASAE, Vol. 28, No. 5, pp.
1363-1366 and 1370, 1985; Model¬
ing load assembly methods for heli-
stat logging, In Transactions of the

This drawing illustrates how the LOGPAC
aerodynamic model opens up a new frontier in
planning. It permits manipulation of the design
features of an aerial yarding vehicle before it is
built and shows how the performance capabil¬
ities of various design configurations would
translate into economic factors— the fuel and
time required to yard a given amount of timber.

ASAE, Vol. 26, No. 2, pp. 357-362,
a983; Simulating changes to heli¬
copter logging operations, In Trans¬
actions of the ASAE (in process);
Optimizing productivity, costs, and
products through complete harvest¬
ing planning, In Proceedings, 1986
Council on Forest Engineering
Annual Meeting, (in process).

Several Station publications describe
studies that contributed information
incorporated in the model. For
copies of the following, write to Pub¬
lications, Pacific Northwest Station:
Estimating tree bole and log weights
from green densities measured with
the Bergstrom xylodensimeter, Re¬
search Paper PNW-322, 1984; Wood
density-moisture profiles In old-
growth Douglas-fir and western
hemlock, Research Paper PNW-397,
1986; and Estimating the weight of
Douglas-fir tree boles and logs with
an iterative computer model,

Research Paper (in press).

A new program
for riparian
research

Where water is king

Water is the lifeblood of the vast
country lying West of the 100th meri¬
dian. It didn't take long for the
pioneers to understand the crucial
importance of water in the arid West.
Where springs, streams, and lakes
were found, so was lush vegetation
that provided food and shelter for a
variety of critters, large and small. By
claiming ownership of a few key
acres of springs or along streams, an
individual exercised de facto owner¬
ship of thousands of surrounding
acres. No wonder then, that the
range wars between ranchers,

Indians, and farmers often involved
access to water and control of irri¬
gation rights.

Although the days when disagree¬
ments were settled with a 6-shot
revolver are history, conflict over,
and abuse of, riparian areas has
continued to the present. Human
activity in riparian areas— including
home building, crop production, road
construction, mining, timber harvest¬
ing, and grazing— has jeopardized
the ability of these areas to provide
all the things demanded of them.

Although over 90,000 miles of
streams and rivers on land admin¬
istered by the Forest Service and the
Bureau of Land Management (BLM)
provide nearly 3 million acres of
riparian-stream areas, there simply
isn’t enough to go around. Man
depends on these acres to provide:

• water for cities and towns

• water and forage for livestock and
wildlife

• quality shade and shelter for
livestock

• prime habitat for birds and other
wildlife

• fish habitat

• preferred locations for recreation
activities

• an important part of the "scenic”
value of western landscapes. 7

“What would you do with a riparian,
if you could catch one?” Personnel
of the Intermountain Research Sta¬
tion’s new riparian research program
often begin talks to civic and com¬
munity groups with such a question
—recognizing that their research
topic isn’t exactly a household word.

Although the term may not be well
known, many public resource
managers in the West believe that
conflicts surrounding the manage¬
ment and use of the productive and
sensitive land adjacent to lakes,
streams, seeps, and springs is— next
to timber management— the most
potentially explosive land manage¬
ment issue of the day.

by Mike Prouty
Intermountain Station

New information resulting from the Riparian-
Stream unit will lead to livestock grazing
regimens that ensure protection of riparian
areas on public land.

Destructive chain reaction

Riparian areas are complicated
natural systems that haven’t held up
well under the pressure of these con¬
flicting uses. Streambank vegetation
provides an important stabilizing fac¬
tor in riparian areas. This plant cover,
and the bank itself, can be damaged
in a number of ways— including ex¬
cessive livestock grazing, off-road
vehicle use, and even by excessive
trampling by fishermen. Often one
destablizmg factor sets in motion a
series of events that all lead to the
same result— loss or degradation of
the riparian area.

The destruction of a streambank by over-
grazing leads to reduced fish habitat, poor
forage, and loss of wildlife habitat.

When vegetation on a streambank—
or the bank itself— is damaged,
several changes can occur. Com¬
position of the streambank plants
can change or the amount of vege¬
tation can diminish, resulting in a
loss of shade and nutrients for the
stream and a reduction in fish
habitat. As vegetation is lost, so are
soil-holding roots, making a stream-
bank more unstable. With weakened
banks, a stream is likely to flex its
muscle in several ways— all with
negative consequences.

Streams can begin bank cutting, thus
becoming wider and shallower. Or
streams can begin to scour the chan¬
nel, making the streambed deeper
than the surrounding area. This
causes the water table in the adja¬
cent riparian area to drop— with
drastic consequences for existing
vegetation. Whether streams move
sideways or downward, the result is
the same— poor water quality, re¬
duced fish habitat, or poor range
conditions. Nation-wide, over 80
percent of these areas have been
altered. In the West, hundreds of
thousands of acres of what were
once lush riparian areas now support
only sagebrush and other dry site
plants— a result of disturbed stream
systems and resulting lowered water
tables.

Addressing the problem

The Intermountain Station has re¬
sponded to the growing interest and
controversy over competing uses for
riparian areas. The new “Riparian-
Stream Ecology and Management’’
research work unit, based at the
Forestry Sciences Laboratory in
Boise, ID, is chartered “to improve
the understanding of riparian and
stream habitats and improve
methods of managing them for the
conservation of resources and pro¬
duction of livestock, wildlife, and
fish."

The project leader for the unit is
range scientist Warren Clary. Other
members of the unit include research
fisheries biologist William Platts, wild¬
life biologist Dean Medin, botanist
Nancy Shaw, fisheries biologist
Richard Torquemada, range techni¬
cian John Kinney, and part-time help
from economist Fred Wagstaff.

The new unit will build upon a foun¬
dation of work established during 10
years of research by Platts. Platts
pioneered the study of riparian-
stream ecology. Establishing re¬
search plots in Nevada, Idaho, and
Utah, he studied the effects of dam¬
aged riparian areas on fish produc¬
tion. His work has provided evidence
that livestock grazing, road construc¬
tion, timber harvest, and mining in
riparian areas can reduce the capa¬
city of streams to produce fish. When
these streams occur within the head¬
waters of the Columbia River system,
and represent critical spawning and
rearing areas for salmon and steel-
head, the impact of damaged
riparian areas can be severe and far
reaching.

With the concentration of expertise
from a variety of scientific disciplines
now focused on this issue, the new
unit has become a center of Forest
Service riparian research. Members
of the unit bring different perspec¬
tives to bear on a variety of aspects
to the problem. In response to the
recreational values of nongame wild¬
life, Medin will use his wildlife back¬
ground to study how livestock
grazing in riparian areas affects the
habitat of small mammals and birds.

8

Fisheries biologists assess and describe the
conditions conducive to productive aquatic
habitat.

Shaw will utilize her botany training
to develop handling and planting
techniques for woody plants such as
willow, cottonwood, and alder. These
species can help rehabilitate dam¬
aged riparian areas, by helping
stabilize streambanks, by providing
shade and shelter for wildlife, and by
furnishing organic material to
streams. Wagstaff provides an impor¬
tant economic focus to the problem
of competing uses on riparian areas.
He’ll address such questions as,
“How many added recreation days
of fishing justify the expense of
fencing a stream?” He’ll also study
the cost effectiveness of various
management practices and
techniques.

Their combined talents will address
these research objectives:

—improve knowledge of the struc¬
ture and function of riparian-stream
ecosystems— including fish and wild¬
life— so that improved management
practices can be developed,

—develop a scientific basis for
cost-effective rehabilitation practices
on riparian areas,

—develop improved grazing
management methods,

—determine the effects of sedi¬
ment on fish.

The project will not be without help.
According to Clary, the National
Forests and the BLM will be impor¬
tant partners in the work of project
personnel. “We have worked closely
with National Forests in Idaho and
Nevada, and with BLM folks in
Idaho, Nevada, and Utah in estab¬
lishing numerous study plots.”

The Pacific Northwest Research Sta¬
tion and the Rocky Mountain Forest
and Range Experiment Station also
have riparian studies under way. In
addition to the Forest Service,
several universities are conducting
riparian research, including the
University of Nevada-Reno, Oregon
State University, the University of
Wyoming, and Montana State
University.

Walking the fine line

Clary and his associates must be
careful to keep from becoming iden¬
tified with the interest groups contest¬
ing use of riparian areas in the West.
The main conflict boils down to
ranchers— who want to continue to
graze livestock on public lands—
versus environmental and wildlife
groups— who see the impact of live¬
stock on riparian areas as reason for
eliminating livestock grazing on
public lands.

While the emphasis of the research
work unit is related to the effects of
grazing on riparian areas (as
opposed to other impacts such as
urban development, road construc¬
tion, or timber harvesting), Clary
wants to avoid involvement in the
emotional debate surrounding the
issue of grazing on public lands.

“The best contribution we can
make,” says Clary, “is to provide
unbiased, biological facts about the
ecological function of these areas, so
land-use policy and management
practices can be based on sound in¬
formation about the resource, not on
emotion. I want our program to pro¬
vide credible facts that have a posi¬
tive influence on natural resource
management."

9

Project personnel will use interim
reports to transfer information result¬
ing from their studies to resource
managers, interest groups, and com¬
munity organizations. “We think it’s
important to get the facts out quickly,
especially in a situation where feel¬
ings run high,” says Clary. “So while
we'll go through the careful review
process associated with technical re¬
search publications, we also want to
release periodic ‘progress reports' as
new information is developed. This is
an important way we can con¬
tribute.”

No two alike

One problem facing the group is the
variability of riparian areas. Some
areas are extremely tough and
durable— able to withstand heavy
use by man and his animals with
little damage to streambanks, loss of
forage and wildlife cover, or reduced
fish production. Other areas are very
fragile, and will be severely dam¬
aged by even moderate use. Some
areas recover quickly from impacts,
others will show the scars of distur¬
bance for many years.

“Because riparian areas are not all
alike, we simply can’t generalize our
research findings,” says Clary. “We
need to establish study plots on
many different sites to develop a feel
for the range of conditions— and then
we can develop associated manage¬
ment practices that apply in different
settings.”

Why the variability? The structure of
streams varies— from small, fast-
moving, relatively straight streams, to
larger, older, slower-moving, mean¬
dering streams. Some streams run
through highly erodible material while
others lie on very hard, durable rock.
In addition, the character of streams
and riparian areas is influenced by
topography, soil types, weather pat¬
terns, and vegetative cover.

Useful information

With their research mission estab¬
lished, personnel in place, problems
selected for study, and action plans
written, the unit has set out on an
aggressive 5-year program of work.
Studies are being designed to pro¬
vide information regarding different
types of riparian areas. But what
specifically will result from this work?

“After we understand more about
what’s going on in riparian areas—
how the various components of
water, soil, topography, weather,
animals, and plants interact— then we
can devise management practices to
ensure that man’s activities won’t
damage the system,” explains Clary.

Clary stresses the practical appli¬
cation of the work. “We want to
develop an understanding of when,
where, and how grazing of riparian
areas is appropriate. With this knowl¬
edge, managers can develop guide¬
lines for livestock grazing specific to
their unique situations. Research by
Bill Platts will tell us what the effect
of sediment will be on fish popula¬
tions. Finally, we want to develop
cost-effective ways to repair and
restore riparian areas that are
damaged.”

Debate over the use of riparian areas
is likely to continue. The demand for
water, grazing, fish production, and
wildlife habitat will increase. But be¬
cause of the attention Clary and
others are focusing on these areas,
the future of riparian areas— and all
that they mean to the West— looks
brighter.

10

Hurdygurdy —
California’s first
basin-wide
study of fish
habitat

by R.B. Pearce, for
Pacific Southwest Station

Current studies on the habitat likes
and dislikes of fish should help forest
and fishery managers alike. Results
should enable them to maintain and
perhaps even boost the populations
of steelhead trout and Chinook salm¬
on in Northern California’s streams.
Habitats the fish prefer, either for
egg-laying, shelter, or feeding, could
then be protected or new ones pro¬
vided where none exist presently.

But first, biologists and geomorphol¬
ogists must correctly identify those
habitats these commercially impor¬
tant fish covet and why.

Although researchers from the Forest
Service and other agencies have
been conducting year-round, basin¬
wide habitat studies in Oregon,
Washington, and British Columbia for
some time, only a few studies, most
of them restricted to small reaches
explored only in the summer, have
been carried out in California. At the
southern limit of the range of anad-
romous salmonids in North America,
California’s climate with large winter
storms and hot, dry summers limits
survival of fish in streams. The com¬

bination of climate, logging, erodible
landscape, and intense fishing is like¬
ly to have a unique and strong im¬
pact on fish productivity and survival.
That’s why scientists from the Pacific
Southwest Station’s Redwood Sci¬
ences Laboratory in Areata have
launched a two-part study along a
12.5 mile channel in the Klamath
Mountains. One part will relate the
various types of habitat in the stream
to the distribution and abundance of
anadromous fish; the other will seek
to discover whether physical and
geological characteristics in and
around the channel can be used to
predict the frequency and total size
of habitat types.

The study is under the direction of
Hydrologist Thomas E. Lisle, and
Fisheries Biologist Lynn M. Decker,
and is scheduled to run at least for
the next few years and probably
longer.

Fish are directly observed underwater using
snorkel and mask to learn their relative prefer¬
ence for different types of habitats.

The goal: To maximize
fish habitat

According to Lisle, coordination of
research of anadromous fish habitat
originated in 1976 when a group of
forest researchers and managers
saw the need to learn more about
how natural habitat affects fish repro¬
duction and survival. Such research
should have many practical benefits,
they reasoned. For one thing, the
physical enhancement of the stream
environment might prove an effective
way of maximizing habitat availability.
The placement of boulders, weirs,
woody debris and other structures,
for instance, may significantly in¬
crease a stream’s productivity by
providing more pools, side channels,
and other habitats in which fish thrive
[See Sidebar: Debris— Not the
Menace Once Thought}. The alterna¬
tive— artificially propagating fish into
a system— though a direct solution to
the problem of dwindling popula¬
tions, is costly and the outcome un¬
certain when variables relating to
habitat are poorly understood.

In round-table discussions with state
and Federal researchers and forest
managers in 1982 and 1983, it was
decided that California needed a
long-term, basin-wide study of anad¬
romous fish and their habitat.

“Only after the relationships between
stream structure and fish productivity
and diversity have been determined
for an entire basin,’’ says Lisle, “can
we expect to verify positive and/or
negative impacts of management on
aquatic systems.”

Debris: Not the menace once thought

Conventional wisdom is often proved wrong when scientists finally get
around to applying the appropriate litmus test. No exception to this is the
long-held belief that removing woody debris and other sizeable material
from a streambed will enhance its productivity.

“Debris improves both the quality and quantity of fish habitat by providing
cover and by varying stream velocity and depth,” says Thomas E. Lisle,
whose studies in Alaska and near Mount St. Helens have allowed him to
conclude such structure as fallen trees, branches, rootwads, and boulders
are a valuable commodity.

Lisle has also found that prey often shares the same reaches with preda¬
tors because cover can be easily found in the patchy environment of a
debris-strewn creek. During low flow in the summer, debris builds living
space by damming the stream and by scouring pools.

In the Alaska study, Lisle's biologist co-workers compared fish populations
in streams where logging debris was either left in or removed, to those
from forested regions. They found that there were more fish and a greater
diversity of fish in streams containing debris. Surprisingly, forested streams
actually had fewer fish than those that had been logged.

Lisle speculates that debris helps sustain fish populations by keeping more
water in the stream during the dry summers and by partitioning habitats
so the fish do not see their neighbors.

“Debris jams also trap sediments and gravels that are important for
spawning,” he said.

Following the May 18, 1980 eruption of Mount St. Helens the question
arose whether trees that had fallen across area streams in the wake of the
blast but that still had commercial value should be harvested or left in. To
answer the question, Lisle and his collaborators from PNW in Corvalis,
studied 4 reaches and 12 subreaches. In some subreaches all material
was removed, in others only merchantable trees were logged, while in still
others the streams were left untouched. Just as in Alaska, they found that
the reaches with the most debris had the greatest diversity and volume of
habitat.

“ It's hard to find a situation where there's too much debris , " he said.

Lisle cautioned, however, that the short-term positive effects of these large
introductions of debris into streams need to be viewed “ with several
grains of salt. "

“The material that’s in these streams now will begin to rot and wash away
or get buried, ” he said. “At the same time the new trees growing up
around the stream will not be large enough to produce replacement
debris. ” He said that because of the later downturn in volumes of debris,
he wouldn't be surprised to see a downturn in fish densities in these
streams in the future, even below levels that are typical for forested
streams.

Lisle sees merit in leaving debris in streambeds affected by logging or
natural processes. “People have always had an urge to ‘ cleanse ' a
stream, ” he says, “but our studies suggest that such a move could be
counter-productive. ”

12

One of Decker’s jobs is to classify
Hurdygurdy’s many habitats.

Researchers map habitat units by plotting fish,
boulders, water depth, and cover on grids.

Selecting the site

Choosing a suitable site in northern
California was step one towards this
goal.

“The basin had to have extensive
habitat and enough fish to permit
useful study,” Lisle said. “We also
wanted it to be typical of a large
area in California.”

Hurdygurdy Creek, a tributary of the
South Fork of the Smith River in Six
Rivers National Forest, satisfied most
of the criteria.

“It's not in the Klamath-Trinity river
basin so maybe it’s not the best
political choice,” Lisle admitted, re¬
ferring both to the perennial debate
over who might be responsible for
depleting fish populations in that
basin and to the recent large influx
of federal restoration money targeted
for the Klamath and Trinity Rivers.

But Lisle thinks Hurdygurdy is a
good choice nonetheless, noting that
it has the same geology and climate
as the Klamath-Trinity basin. Besides,
he said, “the large number of hatch¬
eries in the Klamath-Trinity basin
wouldn't make for a very clean
study.”

Assessing the
stream habitats

The objectives for the PSW study are
two-fold:

(1) to find out what sorts of
habitats fish prefer at different stages
of their life cycles and at different
seasons of the year, and

(2) to define the physical pro¬
cesses that account for the formation
of the various types of habitat.

“There are three types of riffles, two
types of runs, and nine different
types of pools. And those are all
divided up by how flowing water cre¬
ates the pools,” she said, explaining
the complexity of the scheme.

Decker has identified 908 habitat
types— or “units”— along Hurdy-
gurdy’s entire length. Representative
habitats are then selected for de¬
tailed analyses and fish counts.

To do their field work Lisle and
Decker, accompanied by three to
twelve assistants, hike into the differ¬
ent reaches and measure the physi¬
cal attributes of the selected units.
Each unit is marked and measured
for length, width, area, and volume.
Some units are more thoroughly
characterized by measuring thalweg
and depth profiles, determining sub¬
strate type and relief, and by pro¬
viding a detailed description of the
site.

Lisle sees three types of channels in
Hurdygurdy Creek. In order of in¬
creasing width, confined channels

are bordered by bedrock, lack flood-
plains, and tend to be straight and
narrow; semi-alluvial channels

have floodplains along one bank, at
least, and have single threads,
braided channels split into multiple
threads. These three types may be
useful for classifying streams be¬
cause they are common to many
streams with gravel beds and could
be easily recognized by fishery
managers in the field or on aerial
photographs.

13

The relative abundance of habitat
types within the three channel types
has already been determined. Con¬
fined channels have the greatest
area of low-gradient riffles. Semi-
alluvial channels, having the greatest
areas in both cascades and pools,
show the greatest diversity of habitat
conditions. Braided channels have
intermediate distributions of habitat
units, but also have the greatest area
of side channels which are important
for winter habitat.

For analysis of fish distribution and
abundance, the workers rely on
bank-counting as well as underwater
tallies.

“The best way to count fish," says
Lisle, “is to put on a snorkle and
mask and sneak up on them from
downstream.”

Seasonal effects

Also completed are the summer and
fall counts for 1986. Although the
data are just now being analyzed, it’s
Lisle's impression that, in the sum¬
mer, young fish “are all over the
stream” without showing an obvious
preference for one area over
another. “I don’t know if it’s density
pressures or what,” he says, “but all
the fish are finding places to hang
out.”

As might be expected, the situation
in winter is quite different.

During a couple of benchmark
counts last winter, the researchers
couldn't find fish anywhere in the
main part of the stream; but after a
thorough search they found fish
“hiding” between rocks and other
large structures. Lisle said that num¬
erous fish were also spotted in back¬
water channels. Those beds, he
explained, are formed on the flood-
plain and may not carry water at all
in the summer, while in the winter
they probably offer sanctuary from
swift currents.

Decker said her first impression of
the winter count is that there are
fewer fish in general and a con¬
spicuous absence of all but large
Chinook. But she added that she will
“continue to look through the winter
to see whether the distribution and
abundance of fish within these
habitats changes.”

Formation of fish habitats

Lisle, trained as a fluvial geomorphol-
ogist, is particularly interested in de¬
fining the physical processes that
lead to habitat development.

Such things as landslides, structures
in the streambed, sediment quality,
stability of bars and banks, as well
as the overall topography and geo¬
logy of the creek basin may define
what habitats are possible and how
they are formed. Coming to terms
with the physical forces that shape
Hurdygurdy’s creek bed may allow
forest managers to anticipate habitat
diversity and abundance in other
similarly constituted streams without
actually having to do costly field
studies.

“Channels with common morpho¬
logies should have common relative
abundances of habitat units,” said
Lisle.

Lisle has already described several
principles that govern how obstruc¬
tions affect channel behavior and
hopes to study other streams in
Northern California to see how vari¬
ations in such factors as sediment
size interact with these structures to
form stream channels over various
widths.

Lisle predicts that these same
principles could then be used to
anticipate the effects of sediment
deposition in the stream from land¬
slides, logging, and road erosion.

The way such events affect the
stream channel may be determined
by the type of channel reached (e.g.,
braided, confined, obstructed, etc.).

14

Forest managers were faced with the choice

of whether or not to salvage large trees
toppled into channels around Mount St.
Helens as a result of the blast of May 18,
1980.

The geomorphic study at Hurdy-
gurdy Creek is being conducted in
two phases. Already underway, is
Phase One (part of the biological
study), which is to relate habitat
abundance to the various physical
characteristics of the channel. The
volume of pools, for instance, is likely
to depend both on the size of
obstructions, which intensify the
scouring force of water, as well as
sediment type, which controls the
depth of scour. The number of
pools, on the other hand, is ex¬
pected to relate to the frequency of
channel obstructions and to “sinuos¬
ity," a measure of a river’s twists
and turns.

Phase Two, to commence this sum¬
mer, will address possible cause and
effect relationships between the geo¬
morphic characteristics of Hurdy-
gurdy Creek and its habitat profile.

At the completion of the first two
phases of the study, the geomorphic
and biological data will be compared
in an attempt to divine the individual
and collective contributions of such
factors as channel width, bed
material size, stream gradient, and
the presence or absence of bars,
landslides, and debris on habitat
type and numbers of fish.

“There is no definite cut-off date for
the study," said Lisle, who antici¬
pates that as more is learned about
Hurdygurdy Creek, new questions
will arise that will beg to be an¬
swered. Already, some differences
between his and the basin studies in

Oregon and Washington have been
noted. For one thing, the Chinook
salmon in Hurdygurdy Creek migrate
farther into the basin and stay longer
in the smaller streams. “Before now
it has been commonly thought that
young Chinook migrate out to a
river’s estuary within a few months of
hatching,” Lisle said. “But we’re
finding that they stay in Hurdygurdy
Creek much longer than that and we
would like to find out what keeps
them there."

Eventually Lisle and Decker want to
place a trap at the bottom of the
stream so that fish can be systema¬
tically counted, weighed, measured,
and tagged. “In this way we should
be able to determine the number of
fish migrating to sea and, therefore,
the overall productivity of the basin.”
Lisle observed that it makes little dif¬
ference if fish are able to reach a
stream to spawn and lay eggs if the
young fail to grow up, migrate to
sea, and mature."

The Hurdygurdy study promises to
yield many such valuable insights
into the life of a Northern California
stream. Lisle and Decker invite scien¬
tists from other agencies, universities,
and industry to work with them in
this natural laboratory in the Klamath
mountain range.

15

Helping bald
eagles keep a
foothold in the
Southwest

by Rick Fletcher
Rocky Mountain Station

16

These 7-week-old eaglets are perched at a
typical cliff nest location along the Salt River in
Arizona.

‘‘Unfortunately this nest was very
poorly built. Loosely constructed on
a slanting ledge, it had a prickly
pear pad supporting the front of it.
The nest is very accessible and both
young were lost to predation. We
think it was probably a bobcat or a
coyote. During the non-breeding
season last summer, we went in and
constructed an artificial nest, though
you wouldn’t know from the outward
appearance. We found a column on
the same cliff that is basically im¬
penetrable to mammalian predation.
It's 12 feet from the top of the cliff
face down to the nest, and about 70
to the slope below. However, this
year, the birds went back to their
traditional nest. We hated to see that.
Presently the controlling agencies are
deciding if it would be wise to re¬
locate the young birds if hatching
occurs again. When they’re 2 or 3
weeks old, they could be placed in a
more secure, protected nest. At least
we could remove the possibility of
mammalian predation.”

Research Wildlife Biologist Terry
Grubb discusses some of the suc¬
cesses and disappointments of the
past 10 years of studying bald
eagles in the Southwest. Since 1983
the Rocky Mountain Station has had
the lead responsibility for coor¬
dinating and conducting bald eagle
research, most recently on the
breeding population, in central
Arizona. Early research efforts
focused primarily on winter habitat
needs— feeding habits, behavior,
roosts, perches, etc. However, a
3-year breeding study has just been
completed.

Although some eagles overwinter in
New Mexico, Arizona is the site of
several year-round breeding pairs.

‘‘It surprises most folks to think that
the cactus-covered desert Southwest
is home to bald eagles,” said Grubb.
‘‘Even after being here for a number
of years, I still find it hard to get
used to seeing a saguaro and eagle
side-by-side.”

" Eagle Beagles" are bald eagle nest watchers
that help provide nest site protection and
collect scientific data.

With a couple of exceptions, virtually
all of the nesting sites are on the Salt
and Verde River systems in the cen¬
tral part of the State. Eagles are
associated with the riparian areas of
the state, where there is often conflict
between livestock, recreation, and
wildlife interests. The birds tend to
nest on cliffs, ledges, or pinnacles.
Snags are also a favored spot. Egg-
laying occurs early, from mid-
January to mid-February. This allows
the young to fledge prior to the
summer heat.

“In 1984-85, we had 18 known
breeding areas within Arizona,” said
Grubb. “We found 4 new ones last
year, 2 of which were definitely new
nests and presumably new pairs.
There’s been a very significant in¬
crease in successfully fledged young.
In 1981, we doubled the average of
5 to 7 young per year out of 3 or 4
nests to 14. We dropped to 13 in
1982, went to 15 in 1984, and hit an
all time record of 22 young from 13
nests in 1985. There’s a variety of
possible reasons, but we would like
to think it is largely a result of our in¬
tensive coverage, surveillance, and
protective management actions that
led to the increased population,” he
said.

Eagle beagles

Perhaps one of the most important
reasons for the success of this re¬
search effort can be attributed to the
“eagle beagles”— a volunteer bald
eagle nest watcher program that
dates back to 1978. The program
was established on Arizona National
Forests to help prevent excessive
human disturbance at nest sites, and
later broadened to study eagle and
human activity in and around bald
eagle breeding areas. From a
modest beginning on a single Forest
with just 1 Audubon Society volun¬
teer to monitor 1 nest, the program
has expanded to other National
Forests, and into a large-scale site
monitoring operation with several
full-time, well-trained volunteers.
“Today,” said Grubb, “these people
not only assist with site protection,
but they collect extensive biological
data, from nest repair to post-
fledging dispersal.” Typically 1, or
preferrably 2, people are stationed
on a strategic ridge or cliff over¬
looking an area where they can
record eagle activities and move¬
ments. One of the main objectives is
to map geographic units for manage¬
ment purposes based on habitat use
by the nesting eagles. Another objec¬
tive is to identify and correct disturb¬
ing activities in these areas. Seasonal
closures to human activity have been
used in conjunction with the nest
watch program to achieve this goal.

“About three-fourths of the known
bald eagle breeding areas are on
National Forests, and all of these are
monitored under the nest watch pro¬
gram. I can’t emphasize enough the
importance of these volunteers to our
studies” said Grubb.

Study focus

“I’ve been studying bald eagles in
Arizona for about 10 years now,”
says Grubb, “and I must admit that
we feel like we have a pretty good
idea of what’s going on with the
population here. And yet, each year
something unusual, something un¬
expected, something that never hap¬
pened before comes up. Nonethe¬
less, such occurrences are one of
the things that make bald eagles
exciting to study.”

Major areas of research for Grubb
and his associates include:

1) food habits and foraging areas—
what the eagles are eating and
where they are getting their food;
and

2) human disturbances and nest
manipulations.

One of the surprise findings is that,
while 53 percent of the bird’s diet is
fish, a substantial part is made up of
mammals. And, the varability of the
diet during different months surprised
researchers. In February, about 80
percent of the diet at some nest sites
is mammals. By March, fish make up
the bulk of the menu, as mammals
account for only 15 to 20 percent.
“We believe this relates to a change
in streamflow and weather,” said
Grubb. “By late spring, streamflows
stabilize, the water clears, and fish
become more available.”

17

Since 1981 , productivity for Arizona bald
eagles has doubled. A significant portion of
this success is directly attributable to Forest
Service research and management efforts.

Bald eagles forage primarily in the
early morning and late afternoon.
However, scientists observed one
breeding pair that foraged only dur¬
ing the morning, and attribute it to a
high level of human activity along the
river in the afternoon. “One of the
important realizations to come out of
this is that one must look at the in¬
dividual nest and circumstance, and
not rely solely on population aver¬
ages," said Grubb.

Bird botherings

A great deal of research involved
how eagles react to human distur¬
bances. Different classes of distur¬
bances and the eagles’ response to
them— from nonrestless, to flight, to
leaving the area— were considered.
Scientists found 250 yards to be the
average distance to disturbances
which caused the birds to fly. At 200
yards, they leave the area.

“The most intriguing aspect of
disturbance and management re¬
quirements,” says Grubb, “is the
pre-season or pre-egg laying activity.
Observations so far indicate that
early in the season, when the eagles
spend the least time at the nest, may
be the most sensitive time— the
period when the birds are the most
easily disturbed off the nest.”

Fostering and nest
manipulation

Two efforts to help improve nesting
sites and fledging success involve
fostering young, and repairing or
building nests. Fostering has been a
real success for the researchers. The
first attempt occurred when two
young were rescued from a nest
being inundated by flood waters. The
one-week-old chicks were placed in
an incubator and fed through an
eagle puppet at the Phoenix, Arizona
zoo. Meanwhile, another pair of
eagles had incubated well past the
normal period with no signs of hatch¬
ing. Scientists replaced the 2 addled
eggs with the now 2-week-old chicks.
Within 20 minutes, the adults were
back at the nest feeding and brood¬
ing the young. Other fostering efforts
have met with similar success.

Over the past few years, several
man-made nests have been con¬
structed to replace fallen nests or to
encourage nesting in areas lacking
good natural nesting sites. “Eagles
don’t always build their nests in safe
and secure spots,” says Grubb.
“They may be in a flood plain and
subject to high water, or built in
rotten snags that are ready to topple.

18

Construction of artificial nests has been a
useful management tool in increasing nesting
success and productivity.

One inexperienced pair built where
predators had easy access to eggs
and chicks. What we’ve done is
build more secure nests using old
nesting and other natural materials—
a very tedious and time-consuming
task. Other nest manipulations have
involved designing and building
tripod-like artificial nest support struc¬
tures made of aluminum pipe, attach¬
ing shading structures to nests to
protect young from the searing sum¬
mer heat, and using explosives to
enlarge a too-small cliff cavity used
for nesting.”

Since 1977, 6 of 9 artificial or mani¬
pulated nests were actively used by
breeding adults the following year.
Four of these successfully produced
young.

Management concerns

Grubb says there are important
management implications concerning
studying bald eagles in Arizona.

“With such a small, isolated popula¬
tion, it seemingly makes little differ¬
ence whether the operative word in
management direction is mainte¬
nance or recovery; either way there
is justifiable cause for concern and
intensive management,” he said.

A variety of techniques have been
applied to these birds in the name of
increasing fledging success. Some
have been successful, while others
remain unproven. But always there
remains the question of “how much
is enough?”

Small, endangered populations pro¬
vide arguments for and against
manipulative management. Because
numbers are so small, some argue
that the birds should be left alone
and spared the risks of “hands-on”
research. Others argue that the small
numbers themselves justify innova¬
tive, sometimes drastic, manipulative
measures. The general philosophy
for bald eagle research/management
in Arizona has been somewhere be¬
tween the two philosophies, with
perhaps an inclination toward
manipulation.

A related question concerns the risks
and benefits of studying 100 percent
of a small population. “Passive, un¬
disturbing activity is one thing,” says
Grubb, “but hands-on, manipulative

This adult pair of breeding bald eagles are
believed to be year-round residents. Courtship
and nest repair typically occurs in late fall, and
eggs are laid by February.

work has been questioned and is
now limited. Leaving some breeding
areas alone provides for experimen¬
tal control, as well as allowing for a
natural safeguard in case the applied
techniques have a detrimental
effect.”

Grubb explains, “In the endeavor to
study or help a species, it is always
possible to overdo or lose perspec¬
tive. For instance, a particular mani¬
pulation may allow individuals to
survive, but fosters characteristics
detrimental to the population’s overall
fitness. For these reasons it is wise to
periodically step back to review past,
present, and future management and
research efforts for the recovery pro¬
gram, and constantly evaluate how
much of such efforts to permit before
allowing the species or population to
fend for itself.”

19

Despite the problems and ever¬
present reservations against exces¬
sive intervention, much of what the
scientists have done in behalf of
Arizona’s small breeding population
has been well worthwhile. The nest
watch program has not only been in¬
strumental in educating the public
and enforcing nest site closures, but
it has also contributed significantly to
the collection of useful biological
data.

The artificial nests and nest modifi¬
cations have been remarkably suc¬
cessful. Permanent, flood-resistant
structures are now available in a
territory that has always had prob¬
lems with nest tree loss. In addition,
as such techniques are tested and
refined, the potential for supplement¬
ing marginal habitat or moving the
eagles to more suitable habitat in¬
creases. It often takes bald eagles
several years to accept and use ar¬
tificial nest structures, so application
of these techniques requires both in¬
sight into the resource needs, and
patience.

In summation, Grubb states, “It is
hoped that through the careful pur¬
suit of such research programs, the
information necessary for effective
management can be obtained while
minimizing or eliminating any nega¬
tive impact on the eagles. There is
no definitive answer to ‘how much
intervention is enough?', but through
definition and prioritization of objec¬
tives, through careful, scientific appli¬
cation of proven field techniques,
and through continuing coordinated
program reassessment, it is possible
to achieve sound bald eagle man¬
agement in the Southwest."

The Station’s bald eagle research is
part of a cooperative, multi-agency
assessment of research and manage¬
ment priorities for Arizona’s breeding
bald eagles, under the direction of
the U.S. Fish and Wildlife Service,
and the Southwestern Bald Eagle
Recovery Team. The ultimate objec¬
tive is to recover the population to a
point that it can be delisted from its
current endangered status.

For more information on bald eagle
research in the Southwest, contact
Terry Grubb at the Forestry Sciences
Laboratory, Arizona State University,
Tempe, Arizona, 85287, (602) 261 -
4365, FTS-261 -4365.

Rappelling down to nests is often a necessary
feat in order to study the young, measure
nests, and collect prey remains, addled eggs
and eggshell fragments.

20

New

publications

The LOGS Study:
Twenty-year results

In 1962, representatives of State,
Federal, and industrial forestry organ¬
izations began an international coop¬
erative effort to determine how the
amount of growing stock retained in
repeatedly thinned young stands of
Douglas-fir affects cumulative wood
production, tree size, and ratios of
growth to growing stock. At the time
the study was started, it was com¬
monly thought that if amounts of
growing stock differed, volume
growth per acre would not vary
much, although the size of trees
produced would be affected.

Results now show that the number of
trees left per acre does affect the
total volume of wood produced, as
well as average tree size. Volume in¬
crements is strongly related to grow¬
ing stock, under the conditions
represented in the LOGS studies.
Relatively high stand density is
required for high cubic-volume
production. Conversely, diameter
increment declines with increasing
stand density.

The study plan, which examines
cumulative wood production, tree
size development and growth-to-
growing stock ratios under eight dif¬
ferent thinning regimes, was initially
developed at Weyerhaeuser Com¬
pany, Centralia, Washington. Pro¬
cedural details for maintaining
consistency among the cooperators
were developed by PNW Station,
according to report authors Robert
O. Curtis and David D. Marshall.

The LOGS study consisted of nine
field installations established in Ore¬
gon, Washington, and British Colum¬
bia. This 20-year report is a joint
effort by the cooperators and pro¬
vides a general description of the
LOGS program, some comparisons
of results between the installations,
and some generalizations and dis¬
cussions of implications. According
to the authors, these comparisons
are ‘‘necessarily incomplete, because

the studies have not yet run their full
course, and because not all analyses
of interest can be included in a
single report.”

The most productive future use of
the LOGS data— and the most effec¬
tive means of applying these results
to practical management— will prob¬
ably be in constructing and refining
stand simulators. The LOGS studies
provide a unique set of high quality
data from young stands maintained
at relatively low densities. Information
on stands of this type is crucial for
evaluating stand management
regimes for our future forests. Very
little other information of this sort is
currently available. The LOGS report
provides basic information on rela¬
tionships between growth and grow¬
ing stock in such stands.

For a copy of The Level-of-Growing
Stock Cooperative Study in Douglas-
fir: Report No. 8, The LOGS Study;
Twenty-Year Results, request Re¬
search Paper PNW-356 from the
Pacific Northwest Station.

21

A succession model
for managers

A great many models have been
developed to depict forest succes¬
sion. While each of these has met a
particular need, none has allowed
resource managers to easily predict
vegetation response to alternative
management treatments.

Scientists at the Intermountain Sta¬
tion’s Forestry Sciences Laboratory
in Missoula, MT, have addressed this
need. In a new report, Research
Forester Steve Arno, and others, out¬
lines a method to model succession
within a habitat type, and to predict
the type of vegetation that will
develop after different management
activities.

Intended for use in land manage¬
ment and in planning, the approach
produces a general-purpose ecologi¬
cal classification of serai community
types on a given habitat type. It also
serves as a guide to help users con¬
firm the habitat type of serai stands.

Request Characterizing Succession
Within a Forest Habitat Type— An
Approach Designed for Resource
Managers , RN-INT-357.

Regenerating
Engelmann spruce

Prompt establishment of Englemann
spruce natural reproduction following
clearcutting is a major obstacle in
managing spruce-subalpine fir forests
in the central Rockies. However,
scientists at the Rocky Mountain Sta¬
tion have recently developed several
guidelines that will help foresters
overcome barriers to natural
regeneration.

Research findings indicate that in
the Engelmann spruce habitat type,
shaded, mechanically scarified, min¬

eral soil seedbeds on north aspects
can be adequately restocked natural¬
ly within a 5-year period in clearcut
openings 300 to 450 feet wide. With
shade or scarification alone, size of
opening that will restock is reduced
to 200 to 350 feet. The effective
seeding distance is so limited on
unshaded natural seedbeds on north
aspects and on south aspects re¬
gardless of the seedbed conditions
that the expectation of natural re¬
stocking in a reasonable period of
time with clearcutting is not a realistic
option.

Details on this study can be found in
Engelmann Spruce Seed Production
and Dispersal, and Seedling Estab¬
lishment in the Central Rocky Moun¬
tains, General Technical Report
RM-134. Copies are available from
the Rocky Mountain Station.

— Number of viable seeds required to produce 800 5-year-
old Engelmann spruce seedlings per acre in relation to seedbed
treatment and aspect.1

Seedbed

Aspect

treatments

North

South

Scarified

shaded

25,600

272,800

Scarified

unshaded

60,800

OC>2

Unscarified

shaded

57,600

249,600

Unscarified

unshaded

333,000

oo

1 Data presented are based upon the exclusion of seed-eating mam¬
mals. If these animals are not excluded, numbers of seed required
should be increased by 100%.

2°° = no survival.

22

All streams are not
created equal

All stream sections are not created
equal when it comes to accurately
estimating fish populations, accord¬
ing to David G. Hankin, author of
Sampling Designs for Estimating the
Total Number of Fish in Small
Streams.

A common two-stage sampling ap¬
proach often used in freshwater fish¬
eries begins with selecting a sample
of stream sections, usually of equal
length.

Next, the total number of fish in each
selected section is estimated and
extrapolated to the entire stream.

According to Hankin, this conven¬
tional practice of selecting stream
sections can result in errors. Errors in
estimating fish numbers within the
sample areas will be small, however,
compared to errors resulting from ex¬
panding the sampled areas to repre¬
sent a whole stream, he says. Fish
numbers are strongly related to
habitat type; the “equal stream
lengths” technique does not take this
correlation into account. Greater ac¬
curacy can be obtained by making
the sampled sections correspond to
natural habitat units such as pools or
riffles, so sampling designs can take
advantage of the natural correlation
between habitat and numbers.

Hankin describes and contrasts three
alternative two-stage sampling tech¬
niques that allow substantially in¬
creased precision in estimates of the
total number of fish in small streams.
He also discusses determination of
the best choice among the alterna¬
tive designs, considering relative
performance, cost, and cost-
effectiveness.

This paper also provides users with
some basic background information
on sampling design, but is designed
so readers with more sophisticated
knowledge can skip that section.

For a copy of Sampling Designs for
Estimating the Total Number of Fish
in Small Streams, request Research
Paper PNW-360 from the Pacific
Northwest Station.

Wilderness research
conference
proceedings
available

Scientists and managers from
throughout North America gathered
in Fort Collins, CO, in July of 1985 to
participate in a National Wilderness
Research Conference.

Over 70 reports of current wilderness
research presented at the conference
are contained in a new INT report.
The papers are grouped into nine
wilderness topics: fire, air quality, im¬
pacts to soil and vegetation, impacts
to fish and wildlife, impacts to water,
recreational use and user character¬
istics, attitudes and behavior, wilder¬
ness benefits, and wilderness
management concepts and tools.

23

The new report will assist those plan¬
ning future wilderness research in
selecting problems to study and in
developing research designs. Wilder¬
ness managers will be interested in
many of the reported research
results in formulating programs to
protect wilderness resources and in
providing quality visitor experiences.

Work is currently under way on a
second volume containing “state-of-
knowledge” review papers by wilder¬
ness agency heads and leading
scientists.

Request Proceedings— National
Wilderness Research Conference:
Current Research , GTR-INT-212.

Two new publications
on Alder

Forest managers need site-specific
information on site quality to make in¬
telligent decisions on species selec¬
tion and management practices. The
most commonly used measure of
potential site productivity is site index
(mean height of crown class trees
that have been free to grow in an
even-aged stand at a specified age).

In A Method of Site Quality Evalua¬
tion For Red Alder, PNW Station
Researcher Constance A. Harrington
has developed an accurate field
guide for predicting site index for red
alder in stands in western Washing¬
ton and northern Oregon. The guide
requires users to evaluate 14 soil-site
properties grouped in three general
categories: geographic and topo¬
graphic position; soil moisture and
aeration during the growing season;
and soil fertility and physical
condition.

The guide was developed using
techniques devised for site evaluation
for several southern hardwoods,
though it differs from those models in
placing considerable importance on
the actual location or position of the
site to be evaluated. One of the
goals of the project was to provide a
field guide for site evaluation that
could be used by practicing for¬
esters. Thus, when possible (without
reducing accuracy) soil-site proper¬
ties selected for use in the model
were those that could be determined
with a minimum of special
equipment.

The site evaluation tables in this pub¬
lication were developed from natural,
unmanaged stands west of the crest
of the Cascade Range in Washington
and northern Oregon, and should
not be used outside this geographic
area until local users determine the
applicability of the tables to their site
conditions.

The publication contains the site
evaluation tables and a sample field
worksheet. For a copy, request
General Technical Report PNW-192
from the Pacific Northwest Station.

24

Height Growth and Site Index
Curves for Red Alder by Constance
A. Harrington and Robert 0. Curtis
presents new polymorphic height
growth and site index curves devel¬
oped for red alder using a reference
age of 20 years.

The curves are applicable to natural
stands between 5 and 50 years of
age in western Washington and
northwestern Oregon. They provide a
better fit with observed patterns of
height growth than those previously
available. The new curves should be
an improvement, particularly for use
in short-rotation management.

The new curves for estimating height
growth and site index were devel¬
oped using stem-data analysis from
two sources: new data from 23
natural stands in western Oregon
and Washington, and previously
published data from western Wash¬
ington. The new curves are recom¬
mended over previous curves

because they are based on a much
larger sample of trees covering a
greater geographic range and a
somewhat greater range in site in¬
dex; the reference age of 20 years is
more appropriate to future short rota¬
tion management of the species than
the 50 years formerly used; and the
more flexible growth function used in
the new curves provides a better ex¬
pression of observed height growth
trends.

The site index estimation curves,
rather than the height growth curves,
are recommended for use in esti¬
mating site index (S20).

For more information, request Re¬
search Paper PNW-358 from the
Pacific Northwest Station.

PROGNOSIS—

version 5.0

PROGNOSIS, the computer model
intended to simulate the develop¬
ment of forest stands, was designed
so that it could be updated as im¬
provements were developed. Since
the model was released in 1981,
scientists at the Intermountain Sta¬
tion’s Forestry Sciences Laboratory
in Moscow, ID, have been develop¬
ing new routines and refining existing
models that make PROGNOSIS a
more powerful and accurate tool.

A new INT report describes changes
in existing features of the model, new
program options, and changes in the
program’s output. Version 5.0 of the
PROGNOSIS model includes:

® a regeneration establishment
model,

• small-tree increment model
revisions,

• an addition to the COVER model
that predicts shrub development
and total canopy cover,

• New management options,

• an Event Monitor for scheduling
management activities,

• more efficient growth prediction
methods,

• numerous improvements in the
biological models.

Request Supplement to the User’s
Guide for the Stand Prognosis Model
—Version 5.0 , GTR-INT-208, by
William Wykoff.

25

Managing wilderness
campsites

A new INT report suggests that
managers concerned about deterior¬
ating wilderness campsites concen¬
trate their efforts in areas where use
levels and impacts are currently low.
These sensitive areas will either
rapidly recover— or deteriorate— in
response to management actions,
according to ecologist David Cole.

Cole studied ecological changes on
22 campsites in the Eagle Cap Wild¬
erness. He found that heavily used,
deteriorated campsites require many
decades to recover— and may never
recover if compliance with closure is
poor. They are not likely to deterior¬
ate or improve substantially in a
short time, regardless what
managers do.

This suggests, according to the
report, that in attempting to avoid
campsite deterioration, a high priority
should be placed on management of
campsites and destination areas that
are receiving relatively low but con¬
sistent amounts of use. Cole recom¬
mends that managers try to maintain
use levels on low-impact areas, avoid
consistent use of the same camp¬
sites, and discourage high-impact
types of use, such as large group
camping.

Copies of Ecological Changes on
Campsites in the Eagle Cap Wilder¬
ness, 1979 to 1984, RP-INT-368, by
David N. Cole are available from the
Intermountain Research Station.

Eat “Hobble Creek”
for an enriched
sagebrush diet

Mule deer and domestic livestock
need high-energy, nutritious food,
especially on winter ranges that tradi¬
tionally support forage plants having
nutrient levels below maintenance
requirements of animals.

Scientists at the Intermountain Sta¬
tion’s Shrub Sciences Laboratory, in
collaboration with the Utah Division
of Wildlife Resources, have released
a genetically improved low-elevation

26

mountain sagebrush— the ‘Hobble
Creek’ selection— that should help
prevent loss of animal weight in deer
and livestock that feed on winter
ranges. The 'Hobble Creek’ sage¬
brush exceeds typical winter forage
values in amount of energy-
producing compounds, crude pro¬
tein, phosphorus, and carotene. Two
other characteristics make this sage¬
brush a desirable winter range
shrub. It does not contain sub¬
stances that lower grass cell wall
digestion in ruminant animals. Better
yet, the grazing animals— both mule
deer and sheep— seem to prefer this
plant over others.

Information on the history, nutritive
profile, and methods of planting,
handling, and establishing this plant
is provided in a new INT research
paper. For your copy, request
“Hobble Creek''— A Superior Selec¬
tion of Low-Elevation Mountain Big
Sagebrush, Research Paper INT-370,
by Bruce Welch and others.

* U.S

GOVERNMENT PRINTING OFFICE 1987: 774-909/65110

To order any of the publications listed in this issue of Forestry Research West , use the order cards
below. All cards require postage. Please remember to use your Zip Code on the return address.

C/5

c

o

0

O

n

z>

Cl

c

o

03

CO

c

03

C

3

O

>

o

o

QC

CD

c

■Q

C

CD

CO

CD

CO

CD

_g

CL

CO

c

o

03

g

5

3

a

c

o

ro

co

to

CD

o

CD

CL

03

C

T3

C

CD

CO

CD

CO

CD

CD

cd -y

0.5

■a

Q A

o

P O

c sp

S A ^

rT £ a:

3D

c

CD

3 co

O CO

co i-

.£ cc

m „
o. O
Q_
CD 03

2 cc

CD

03
O
C\J C

<5-5

■O s £

QJ C g
CD C _

II s

Cl-O c
co p sz
in O
guj 0)

1 £2
ftps P

Old £

,C CD ®
Lj Co CD

□

CD p
OQ a)

s 8

co

03

03

03-
(tJ O
o 03

® £

PJ CD i_

g5 w ®
ISO

□ □

C/3

d o
03
co 03
-2 0)
§)cr
o

Q _

r- O
£ CD

CO CD
.1 1
sjj 5

<D ^

<3 -o

O co

0)0

£ ~j

CD

CD

3D

CD

CC

<D

CC

■Q

3D

CD

CC

§

LU

CD ,
o

Is

CD t-
-1 O
cd Cl
-c 03

fc Ct

□

o
O
><

■ 2
Sfl
d=T a
§1^0
O CL 3D

g °CC I

go

o-S O S’

CD (D 03

0- 3;

“O

c

CD

CO

CD

C

-C

co

a:

o

03

-Q

E

3

2

a §

.o CO

— CL
03 w_
C CD

I ™

.E CL

,« 3D

Uj O

(/) CD
m C CC

in -5* -

CO c/) CO

v CD E

§Q S

F o>£

cl . c 55

|e|I

CCJ CC3 fc ps

CL CO CO O

□ □ □ □

TD

C

CD

CO

C/5

C

o

0

g

n

3

CL

C

o

to

co

to

0

3

O

CO

o

CD

CL

CD

C

g

o

t5

0

SI

T3

C

0

C/5

0

C/3

0

0

Q_

□

3D

C

0

CO

CD

C

o

0

g

15

3

CL

C

o

0

CO

c

’0

c

3

o

E

0

C

CD

C

3D

C

0

C/5

0

C/5

0

0

CO

CO 00

o o

c CM
C3)|_l

"g o

p 8-

CO CC
03 —
_c~ 03
^ O
^ C

03

O

c

03 CM

■P CM
C ‘

03 03

CD 2

» c|

v»_ m
03
CO

O ^

O ?
-c tr
p o
cB g-

(33 CC
03 _
Cr 03
o

C/3 P

CO ^
03

c o

C CD

' CD
Q3

Cl co

03 CD
CJ co

03?

O ^

03

03 S-
r~ CD
Q_

03 _c

Ss

|C? co

h- 03
cn c/3
CD
CC
O .
p

03

2

£c

§ •
,1^
Lu CO

iz

-J —

O

03

C/3

O CD

So

^ C/3
£^
03 C

8 a

S 3

0. O

0) TO
-t= CD
m c/3

cl cd

E o:

03 -

(J ^

co

O 03

o

o ^

_ C/3
03 C^
CJ 03

'S>E

O CD

o 2

as

w p

OP
O CD
3 C
CO O)

c Q

) N

? 5

s ^

03

IS

C/5 5

□ □ □ □

P o ^

03 CL

^ Q-
o

co

CD
CL
d cd
O Q_

O -C
03 O
~03 fe
CO CD

v CO

g£

03 -

CL-C
C3 C/3

co 2

<r -O
I ^

I 03

Tc CO

8 p)

o®

2 i

■Q £

Sol

•50

□ □

CD

T3

c

0

CO

>

"0

c

gr

o'

0

o'

=3

CO

a

co‘

o'

c

o'

Z3

CD
CD
— 1
7T
CD

CD 0?
CD O
O =:

■=r O

CD

>
Q.
Cl
O Co'
CD O
— ZS

CO

O

c

o

CD

CD

4^

-si

O

CO 0
c^- co
CD ^
O ~n

0

o

0

Z3

Q.

ID

0

Z3

CO

0

m

X

"O

0

3

0

Z3

CO

0

o'

=3

0

>

ET

"0

c

CT

o'

0

o'

Z3

0

a

o'

o'

c

o'

=3

IF ^ J3

O-f^o

3. O o

O

o o' __
= 0

=3

CO x

o °

O -§
O CD

O

c

Z3

0

o'

0

Q_

O

00 0
O 0
cn '-H
ro

03

ro

o ti

—■ o

CO 0

0

0

o

CD

00

DO

0

=3

CQ

0

m

X

"O

0

t

0

=3

CO

0

o'

Z3

>

i I

=3

"□

C

CT

o'

0

o'

Z3

0

o

0'

o'

c

03 —
l\5

4^ 0

do 3

2 o

Z T C

CD =3

0

0

0

CO

0

o'

>

I I

Z3

"0

C

CT

o'

0

<"■ ♦

o'

=3

0

o

0‘

”~S

CJ

c:

o'

Z3

~p ro ti
o o &
o v o

Q3 co ^

=3 0°

Q_ X

r-, co o
U CD z

0

CQ

CD

O

CD

-sj

DO

O

00

0

0

0

CO

0

Z3

Q.

30

0

Z3

CQ

0

m

X

~o
0
— s

3'

0

0

CO

0

o'

=3

STAMP - STAMP

Range forage data
base now available

A new report has been issued by
the Rocky Mountain Station that de¬
scribes the content and accessibility
of a rangeland data set based on
the Soil Conservation Service (SCS)
Range Site Description.

A range site is an ecological subdivi¬
sion of the landscape based on the
ability of that site to produce a
characteristic natural plant commu¬
nity. Range Site Descriptions were
developed by the SCS to help plan
and carry out resource conservation
programs on rangeland and other
native grazing land.

Previously, Range Site Descriptions
were available within each state SCS
office in written form only. Now, the
information has been extracted and
placed on computer under the pro¬
gram name RANGE FORAGE. The
data set contains information taken
from all SCS Range Site Descriptions
for 20 states: Arkansas, Arizona, Cali¬
fornia, Colorado, Florida, Idaho, Loui¬
siana, Kansas, Montana, Nebraska,
Nevada, New Mexico, North Dakota,
Oklahoma, Oregon, South Dakota,
Texas, Utah, Washington, and Wyo¬
ming. All files are stored on tape at
the USDA Computer Center in Fort
Collins, Colorado.

For a copy of the publication, re¬
quest Range Forage Data Base for
20 Great Plains, Southern, and West¬
ern States, General Technical Report
RM-133, available from the Rocky
Mountain Station.

29

50

CD —t,
3 =*
0)

o'

o
&

CD
c

-r, £

5 S*
O o
=• CD

2 £
*— ►. C /)

c

c/3

CD

Co

o

o

T1

ro

T|

C

n

O

— i

-C^

o

O

CD

O

CD

3D

o

o

CD

C/3

a

CD

m

cd

c n

CD

"O

— t

m

* t

CD

03

3D

3

C/3

Tl

O

— *
<

o'

“ T

1 ♦

3

•<

3D

O

C/3

CD

0

m

o

"O

CD

r-+-

CD

o

O

o

m

— t

03

Cl

CD

>

J>

3D

O

» ♦

CQ

O

00

o

cn

CD

CD

t ♦

— t

o’

c

X

£

to

m

03

c

CD

CD

H