Full text of "Snake River birds of prey research project : annual report, 1989"

BUM LIBRARY

88026754

1 RIVER BIRDS OF PREY ARE

i/o

1989 ANNUAL REPORT

Research and Monitoring

684

.12

LOO

1989

f <m

/ffijz

11%

™i

H/f

s, ^

L J

U.S. Department of the Interior

BUREAU OF LAND MANAGEMENT
Boise District, Idaho

As the Nation's principal conservation agency, the Department of the
Interior has responsibility for most of our nationally owned public lands
and natural resources. This includes fostering the wisest use of our land
and water resources, protecting our fish and wildlife, preserving the
environmental and cultural values of our national parks and historical
places, and providing for the enjoyment of life through outdoor recreation.
The Department assesses our energy and mineral resources and works to assure
that their development is in the best interests of all our people. The
Department also has a major responsibility for American Indian reservation
communities and for people who live in Island Territories under U.S.
administration .

1 BLM LIBRARY ^ XL

SC-653, BLDG. 50 * ?s

DENVER FEDERAL CiNTIR ^

P. 0. BOX 25047
DENVER, CO 80225-0047

SNAKE RIVER BIRDS OF PREY RESEARCH PROJECT
ANNUAL REPORT
1989

Karen Steenhof
Editor

NOT FOR PUBLICATION

The data presented herein are preliminary and may be inconclusive.
Permission to publish or cite any of these materials is therefore
withheld pending specific authorization of the Boise District, BLM,
and the specific Principal Investigator.

Michael N. Kochert J. David Brunner

Research Leader Boise District Manager

Delmar Vail
Idaho State Director

PREFACE

This report summarizes research and monitoring activities in the Snake
River Birds of Prey Area during calendar year 1989.

In 1989, the BLM Birds of Prey Research staff, in consultation with the
Idaho Army National Guard, completed a research plan to investigate the
effects of habitat alteration on the raptors, their prey, and their habitat
in the SRBOPA. Wildfire and National Guard activities are the 2 main
activities that will be addressed by the proposed research, though some
information will be collected on grazing and seedings. In 1989 study design
teams were formed and 4 workshops were held, including an integration
workshop. Work began on specific designs for each study and a Statement of
Work was completed for the TGS study in 1989. As a result of an identified
information need Drs. Mark Fuller (Patuxent Wildlife Research Center) and Kim
Titus conducted a pilot radio telemetry study in the OTA.

Work continued on copperative studies in 1989. Two graduate studies by
students from Boise State University continued in 1989. Helen Ulmschneider
continued an investigation of long-eared owl post-fledging activity, and Dawn
McAnnis Gerhardt completed the second year of a study on ferruginous hawk
habitat use. The ferruginous hawk study was supported jointly by Boise State
University and the Bureau of Land Management. Studies of the effects of fire
on soil microbial communities were also continued by Boise State University
faculty members, Marcia Wicklow-Howard and Robert Rychert. Claire Fuller of
Oregon State University initiated a study of Townsend's ground squirrel
mating systems and behavior.

Two cooperative studies with utility companies were completed in 1989.
Final reports were completed for the Idaho Power /Pacific Gas and Electric
Companies cooperative study on prairie falcons and the Pacific Power and
Light Company (PP&L) raven roosting study. Allison Beck was contracted to
monitor the response of roosting ravens to installation of new shields and
pegging at the Initial Point roost. The PP&L nesting study continued at the
same levels as in past years.

Other 1989 field activities included continuing investigations of common
barn-owl feeding ecology by Dr. Carl Marti of Weber State College, assessment
of nest box use by American kestrels by Karen Steenhof , and studies of
western screech-owl and northern saw-whet owls using nest boxes in the SRBOPA
by Jeff Marks and John Doremus.

Monitoring of prey was carried out by the Bruneau Resource Area staff and
Research Staff volunteers in 1989. One new jackrabbit spotlighting transect
was established within the Orchard Training Area with the assistance of Idaho
Army National Guard personnel. Monitoring of winterfat seeding projects was
continued by the U.S. Forest Services Laboratory and the Idaho BLM State
Office and greenstrip monitoring was initiated by the Idaho BLM State Office
in 1989.

Although the main thrust for 1989 was development of the BLM/IDARNG
research plan and continuation of cooperative research projects, work still

continued on Technology Transfer. Fourteen scientific papers were published
and/or accepted for publication in 1989, and Birds of Prey Research Project
staff members and associates made 14 technical presentations at meetings.

ACKNOWLEDGMENTS

The Bureau of Land Management wishes to thank those
agencies and individuals who assisted with the project.
Appreciation is extended to the Idaho Department of Fish
and Game and the U.S. Fish and Wildlife Service for special
permits and assistance. Special thanks go to all other
individuals who volunteered their services.

COOPERATING AGENCIES AND INSTITUTIONS

Boise State University, Raptor Research Center

College of Idaho

Idaho Army National Guard

Idaho Department of Fish and Game

Idaho Department of Transportation

Idaho Department of Water Resources

Idaho Power Company

Idaho State University

Pacific Gas & Electric Company

Pacific Power and Light Company

University of Idaho

University of Wisconsin

USDA Agricultural Research Service

U.S. Fish and Wildlife Service

U.S. Forest Service Intermountain Research Station

Weber State College

World Center for Birds of Prey

TABLE OF CONTENTS

PART I. TECHNOLOGY TRANSFER

Page

Summary of Project Publications 2

1989 Technical Presentations 8

PART II. PROGRESS REPORTS

Nesting Density and Productivity of Golden Eagles in and near the

Snake River Birds of Prey Area. 13

Monitoring Long-eared Owl Occupancy in the Snake River Birds

of Prey Area. 19

Raptor and Raven Nesting on the PP&L Malin to Midpoint 500-kV

Transmission Line. 22

Implications of Communal Roosting by Common Ravens to Operation and

Maintenance of the Malin to Midpoint 500-kV Transmission Line. 37

Use of Nest Boxes by American Kestrels in Southwestern Idaho. 47

Feeding Ecology of the Barn Owl in the Snake River Birds

of Prey Area. 50

Nest Box Use, Reproduction, Movements, Food Habits, and Annual
Cycle of Body Mass of Western Screech-Owls in the Snake River
Birds of Prey Area. 55

Post-nesting Ecology of Long-eared Owls in the Snake River Birds

of Prey Area, Idaho. 58

Home Range and Habitat Use of Ferruginous Hawks in the Snake River

Birds of Prey Area. 78

Morphometric and Genetic Variation in Ferruginous Hawk Populations. 82

Testing Performance of Wildlife Telemetry Equipment in the Orchard

Training Area. 84

Abundance of Lagomorphs and Rodents in the Snake River Birds

of Prey Area. 89

Townsend's Ground Squirrel Relative Abundance in the Snake River

Birds of Prey Area. 96

Townsend's Ground Squirrel Diets in the Snake River Birds

of Prey Area. 104

Factors Affecting the Mating System in Townsend's Ground Squirrels. 123

Parasites and Population Ecology in Kangaroo Rats. 127

Effect of Wildfires on Soil Microbial Communities. 130

Winterfat Seeding Projects in the Snake River Birds of Prey Area. 135

Environmental Monitoring in the Idaho Army National Guard

Orchard Training Area. 142

Establishment of Monitoring Studies on Greenstrip Projects

in the Snake River Birds of Prey Area. 148

PART I
TECHNOLOGY TRANSFER

Snake River Birds of Prey Area
Technical Articles Published And/Or Accepted for Publication

Allen, G.T., R.K. Murphy, K. Steenhof, and S.W. Piatt. 1986. Late fledging
dates, renesting, and large clutches of prairie falcons. Wilson Bull.
98:463-465.

Collopy, M.W. 1983. A comparison of direct observations and collections of
prey remains in determining the diet of golden eagles. J. Wildl. Manage.
47:360-368.

Collopy, M.W. 1983. Foraging behavior and success of golden eagles. Auk
100:747-749.

Collopy, M.W. 1984. Parental care and feeding ecology of golden eagle
nestlings. Auk 101:753-760.

Collopy, M.W. 1986. Food consumption and growth energetics of nestling
golden eagles. Wilson Bull. 98:445-458.

*Collopy, M.W., and T.C. Edwards, Jr. 1989. Territory size, activity

budget, and role of undulating flight in nesting golden eagles. J. Field
Ornith. 60:43-51.

Craig, T.H., E.H. Craig, and J.S. Marks. 1982. Aerial talon-grappling in
northern harriers. Condor 84:239.

*Diller, L.V. 1990. A field observation on the feeding behavior of Crotalus
viridis lutosus. J. Herpetology. Accepted 1989.

Diller, L.V., and D.R. Johnson. 1982. Ecology of reptiles in the Snake
River Birds of Prey Area. U.S. Dep. Inter., Bur. Land Manage., Boise,
Idaho. 107pp.

Diller, L.V., and D.R. Johnson. 1988. Food habits, consumption rates, and
predation rates of western rattlesnakes and gopher snakes in southwestern
Idaho. Herpetologica 44:228-233.

Diller, L.V. , and R.L. Wallace. 1981. Additional distribution records and
abundance of three species of snakes in southwestern Idaho. Great Basin
Nat. 41:154-157.

Diller, L.V. , and R.L. Wallace. 1986. Aspects of the life history and
ecology of the desert night snake, Hypsiglena torquata deserticola :
Colubridae in southwestern Idaho. Southwestern Nat. 31:55-64.

Doremus, J.H., and A.R. Bammann. 1982. The Snake River Birds of Prey Study
Area bird list U.S. Dep. Inter., Bur. Land Manage., Boise, Idaho.

Edwards, T.C, Jr., and M.W. Collopy. 1983. Obligate and facultative brood
reduction in eagles: an examination of factors that influence
fratricide. Auk 100:630-635.

*Edwards, T.C. , Jr., and M.W. Collopy. 1989. Response to G.R. Bortolotti
and T.W. Arnold. Auk 106:523-524

Edwards, T.C, Jr., M.W. Collopy, K. Steenhof, and M.N. Kochert. 1988. Sex
ratios of fledgling golden eagles. Auk 105:793-796.

Edwards, T.C, Jr., and M.N. Kochert. 1986. Use of body weight and length
of footpad as predictors of sex in golden eagles. J. Field Ornithol.
57:317-319.

*Engel, K.A., and L.S. Young. 1989. Evaluation of techniques for capturing
common ravens in southwestern Idaho. North Amer. Bird Bander. 14:5-8

*Engel, K.A. , and L.S. Young. 1989. Spatial and temporal patterns in the
diet of common ravens in southwestern Idaho. Condor 91:372-378.

Evans, D.L. 1982. The wingloading method of weight control. Hawk Chalk
21:55-59.

Groves, CR. , and K. Steenhof. 1988. Responses of small mammals and
vegetation to wildfire in shadscale communities of southwestern Idaho.
Northwest Sci. 62:205-210.

*Holthuijzen, A.M. A. 1990. Prey delivery, caching, and retrieval rates in
nesting prairie falcons. Condor. Accepted 12/15/89.

*Holthuijzen, A.M. A. 1990. Behavior and productivity of nesting prairie
falcons in relation to construction activities at Swan Falls Dam. Final
Report. Idaho Power Co., Boise, Idaho. 77pp.

Holthuijzen, A.M. A. , P. A. Duley, J.C Hagar, S.A. Smith, and K.N. Wood.
1987. Piracy, insectivory, and cannibalism of prairie falcons ( Falco
mexicanus ) nesting in southwestern Idaho. J. Raptor Res. 21:32-33.

Holthuijzen, A.M. A. , P. A. Duley, J.C Hagar, S.A. Smith, and K.N. Wood.
1987. Bathing behavior of nesting prairie falcons ( Falco mexicanus ) in
southwestern Idaho. Wilson Bull. 99:135-136.

Hyde, W.F. , A. Dickerman, and D. Stone. 1982. Development versus

preservation in the Snake River Birds of Prey Conservation Area. Am. J.
Agric. Econ. 64:756-760.

Innis, G. , J. Wiens, M. El-Bayoumi, L. Nelson, R. Ryel, K. Smith, J.

Wlosinski, G. Worthen. 1979. Birds of Prey Study Area simulation model,
Vol. 1. Final Report. Utah State Univ. Wildlife Science Rep. 7. Logan,
Utah. 277pp.

Johnson, D.R., N.C Nydegger, and G.W. Smith. 1987. Comparison of
movement-based density estimates for Townsend ground squirrels. J.
Mammal. 68:689-691.

Knopf, F.L. , and T.E. Olson. 1984. Naturalization of Russian olive:

implications to Rocky Mountain wildlife. Wildl. Soc. Bull. 12:289-298.

Kochert, M.N. 1980. Golden eagle reproduction and population changes in
relation to jackrabbit cycles: implications to eagle electrocutions.
Pages 71-86 in R.P. Howard and J.F. Gore, eds. Proceedings of a workshop
on raptors and energy developments U.S. Dep. Inter., Fish and Wildl.
Serv. , Boise, Idaho. 125pp.

Kochert, M.N. 1986. Raptors. Chapter 16 in A.Y. Cooperrider, R.J. Boyd,
and H.R. Stuart, eds. Inventory and monitoring of wildlife habitat. U.S.
Dep. Inter., Bur. Land Manage . , Denver, Colo. 858pp.

*Kochert, M.N. 1989. Responses of raptors to livestock grazing in the
western United States. Pages 194-203 in B.A. Giron Pendleton, et al.,
eds., Proceedings of the Western Raptor Management Symposium and
Workshop. Natl. Wildl. Fed., Washington D.C.

Kochert, M.N. , B.A. Millsap, and K. Steenhof. 1988. Effects of livestock
grazing on raptors with emphasis on the southwestern U.S. Pages 325-334
in R. L. Glinski et al., eds. Proceedings of the Southwestern Raptor
Management Symposium and Workshop. Natl. Wildl. Fed., Washington, D.C.

Kochert, M.N. , and M. Pellant. 1986. Multiple use in the Snake River Birds
of Prey Area. Rangelands 8:217-220.

Kochert, M.N. , K. Steenhof, and M.Q. Moritsch. 1983. Evaluation of

patagial markers for raptors and ravens. Wildl. Soc. Bull. 11:271-281.

Marks, J.S. 1983. Prolonged incubation by a long-eared owl. J. Field
Ornithol. 54:199-200.

Marks, J.S. 1984. Unusual nest sites of a western screech-owl and an
American kestrel. Murrelet 64:96-97.

Marks, J.S. 1984. Feeding ecology of breeding long-eared owls in
southwestern Idaho. Can. J. Zool. 62:1528-1533.

Marks, J.S. 1985. Yearling male long-eared owls breed near natal nest.
J. Field Ornithol. 56:181-182

Marks, J.S. 1986. Nest-site characteristics and reproductive success of
long-eared owls in southwestern Idaho. Wilson Bull. 98:547-560.

Marks, J.S., and J.H. Doremus. 1988. Breeding season diet of northern
saw-whet owls in southwestern Idaho. Wilson Bulletin. 100:690-694.

Marks, J.S., J.H. Doremus, and A.R. Bammann. 1980. Black-throated sparrows
breeding in Idaho. Murrelet 61:112-113.

*Marks, J.S., J.H. Doremus, and R.J. Cannings. 1989. Polygyny in the
northern saw-whet owl. Auk 106:732-734.

Marks, J.S., and V.A. Marks. 1981. Comparative food habits of the

screech owl and long-eared owl in southwestern Idaho. Murrelet 62:80-82.

Marks, J.S. , and CD. Marti. 1984. Feeding ecology of sympatric barn owls
and long-eared owls in Idaho. Ornis Scandinavica 15:135-143.

Marks, J.S., and R.L. Redmond. 1987. Parent-offspring conflict and natal
dispersal in birds and mammals: comments on the Oedipus hypothesis. Am.
Nat. 129:158-164.

Marks, J.S., and E. Yensen. 1980. Nest sites and food habits of long-eared
owls in southwestern Idaho. Murrelet 61:86-91.

Marti, CD. 1988. A long-term study of food-niche dynamics in the common
barn-owl: comparisons within and between populations. Can. J. Zool.
66:1803-1812.

Martin, J.W. 1987. Behavior and habitat use of breeding northern harriers
in southwestern Idaho. J. Raptor Res. 21:57-66.

Messick, J. P., and M.G. Hornocker. 1981. Ecology of the badger in
southwestern Idaho. Wildl. Monogr. 76. 53pp.

Messick, J. P., G.W. Smith, and A.M. Barnes. 1983. Serologic testing of
badgers to monitor plague in southwestern Idaho. J. Wildl. Diseases
19:1-6.

Messick, J.P. , M.C Todd, and M.G. Hornocker. 1981. Comparative ecology of
two badger populations. Pages 1290-1303 in J. A. Chapman and D. Pursley,
eds. Proceedings of the World Furbearer Conference. Frostburg,
Maryland.

Moritsch, M.Q. 1983. Photographic guide for aging nestling prairie
falcons. U.S. Dep. Inter., Bur. Land Manage . , Boise, Idaho. 15pp.

Moritsch, M.Q. 1983. Photographic guide for aging nestling red-tailed
hawks. U.S. Dep. Inter., Bur. Land Manage., Boise, Idaho. 14pp.

Moritsch, M.Q. 1985. Photographic guide for aging nestling ferruginous
hawks. U.S. Dep. Inter., Bur. Land Manage., Boise, Idaho. 21pp.

*Nydegger, N.C, and D.R. Johnson. Size and overlap of Townsend ground
squirrel home ranges. Great Basin Nat. 49:108-110.

Nydegger, N.C, and G.W. Smith. 1986. Prey populations in relation to
Artemisia vegetation types in southwestern Idaho. Pages 152-156 in E.D.
McArthur and B.L. Welch, eds. Proceedings of a Symposium on the biology
of Artemisia and Chrysothamnus . U.S. Dep. Agric, For. Serv. ,
Intermountain Res. Sta. , Ogden, Utah. 398pp.

Olendorff, R.R., and M.N. Kochert. 1977. Land management for the

conservation of birds of prey. Pages 294-307 in R.D. Chancellor, ed.
Proceedings of the World Conference on Birds of Prey. Int. Counc. for
Bird Preservation, Great Britain. 442pp.

*01endorff, R.R. , D.D. Bibles, M.T. Dean, J.R. Haugh, and M.N. Kochert.
1989. Raptor habitat management under the U.S. Bureau of Land
Management Multiple-Use Mandate. Raptor Res. Rep. 8:1-80.

Powers, L.R., T.H. Craig, and J. Martin. 1984. Nest defense by northern
harriers against the coyote in southwestern Idaho. Raptor Res. 18:78-79.

Smith, G.W., and D.R. Johnson. 1985. Demography of a Townsend ground
squirrel population in southwestern Idaho. Ecology 66:171-178.

Smith, G.W., and N.C. Nydegger. 1985. A spotlight line-transect method
for surveying jack rabbits. J. Wildl. Manage. 49:699-702.

Smith, G.W., N.C. Nydegger, and D.L. Yensen. 1984. Passerine bird
densities in shrubsteppe vegetation. J. Field Ornithol. 55:261-264.

Steenhof, K. 1982. Use of an automated geographic information system by
the Snake River Birds of Prey Research Project. Comput. Environ. Urb.
Systems 7:245-251.

Steenhof, K. 1983. Prey weights for computing percent biomass in raptor
diets. Raptor Res. 17:15-27.

Steenhof, K. 1984. Book review: behavior of fledgling peregrines.
Murrelet 65:31.

Steenhof, K. 1984. Use of an interspecific communal roost by wintering
ferruginous hawks. Wilson Bull. 96:137-138.

Steenhof, K. 1987. Assessing raptor reproductive success and productivity.
Pages 157-170 in B.A. Giron Pendleton, B.A. Millsap, K.W. Cline, and D.M.
Bird, eds. Raptor management techniques manual. Natl. Wildl. Fed.,
Washington, D.C.

Steenhof, K. 1988. Book review: raptor conservation in the next 50 years.
Wilson Bull. 100:338-339.

Steenhof, K. 1988. Identifying potential bald eagle nesting habitat: a
review of the state of the art. Pages 31-59 in D. Garcelon and G.W.
Roemer, eds., Proceedings of the international symposium on raptor
reintroduction, 1985. Institute for Wildlife Studies, Areata,
California.

Steenhof, K. , and M.N. Kochert. 1982. Nest attentiveness and feeding rates
of common ravens in Idaho. Murrelet 63:30-32.

Steenhof, K. , and M.N. Kochert. 1982. An evaluation of methods used to
estimate raptor nesting success. J. Wildl. Manage. 46:885-893.

Steenhof, K. , and M.N. Kochert. 1985. Dietary shifts of sympatric buteos
during a prey decline. Oecologia 66:6-16.

Steenhof, K. , and M.N. Kochert. 1988. Dietary responses of three raptor
species to changing prey densities in a natural environment. J. Animal
Ecol. 57:37-48.

Steenhof, K. , M.N. Kochert, and J.H. Doremus. 1983. Nesting of subadult
golden eagles in southwestern Idaho. Auk 100:743-747.

Steenhof, K. , M.N. Kochert, and M.Q. Moritsch. 1984. Dispersal and

migration of southwestern Idaho raptors. J. Field Ornithol. 55:356-368.

*Wicklow-Howard, M. 1989. The occurrence of vesicular-arbuscular

mychorrhizae in burned areas of the Snake River Birds of Prey Area,
Idaho. Mycotaxon 34:253-257.

Yensen, D.L., and G.W. Smith. 1984. Big sagebrush-winterfat and big
sagebrush-Nuttall saltbush mosaic vegetation in southwestern Idaho.
Pages 28-33 in A.R. Tiedemann, E.D. McArthur, H.C. Stutz, R. Stevens, and
K.L. Johnson, eds. Proceedings of the symposium on the biology of
Atriplex and related chenopods. Gen. Tech. Rep. INT-172. U.S. Dep.
Agric, For. Serv., Intermountain For. Range Exp. Sta. , Ogden, Utah.

Yensen, D. 1980. A grazing history of southwestern Idaho with emphasis on
the Birds of Prey Study Area. U.S. Dep. Inter., Bur. Land Manage.,
Boise, Idaho. 82pp.

Yensen, D.L. 1981. The 1900 invasion of alien plants into southern Idaho.
Great Basin Nat. 41:176-183.

*Young, L.S. 1989. Effects of agriculture on raptors in the western United
States an overview. Pages 209-218 in B.A. Giron Pendleton et al., eds.,
Proceedings of the Western Raptor Management Symposium and Workshop.
Natl. Wildl. Fed., Washington D.C.

Young, L.S., J.G. Crenshaw, and L.L. Crenshaw. 1988. Food caching by a
short-eared owl . Murrelet 69 : 39 .

*Young, L.S., and K.A. Engel. 1988. Implications of communal roosting by
common ravens to operation and maintenance of Pacific Power and Light
Company's Malin to Midpoint 500 kV transmission line. Final Report.
Pacific Power Env. Serv. Dep. Portland, OR.

Young, L.S., and M.N. Kochert. 1987. Marking techniques. Pages 125-156 in
B.A. Giron Pendleton, B.A. Millsap, K.W. Cline, and D.M. Bird, eds.
Raptor management techniques manual. Natl. Wildl. Fed., Washington, D.C.

Published and/or accepted for publication in 1989. Reprints of most of
these articles are available from the Birds of Prey Research Project,
Boise District, BLM, 3948 Development Avenue, Boise, ID 83709.

1989 Birds of Prey Research Technical Presentations

1/28/89

2/21/89

2/24/89

3/14/89
3/17/89

Holthuijzen, A.M. A. , R.D. Williams . M.N. Kochert, and L.S. Young.
Mate guarding in the Prairie Falcon: with particular reference to
the sperm competition hypothesis. Annual Meeting of the Western
section of the Wildlife Society. Redding, California.

Young, L. Bald eagle habitat management: an international
perspective. Oregon Eagle Foundation Bald Eagle Habitat Management
Workshop. Hood River, Oregon.

Holthuijzen, A.M. A. Copulation frequency and timing in the Prairie
Falcon. Annual Meeting of the Idaho Chapter of The Wildlife
Society. Boise, Idaho.

Ulmschneider, H. Post-nesting movements of long-eared owls in
southwestern Idaho. Annual Meeting of the Idaho Chapter of the
Wildlife Society. Boise, Idaho.

Steenhof, K. Bald eagle recovery. Graduate seminar on Eagle
Biology. Boise State University. Boise, Idaho.

Young, L. , K.A. Engel, K. Steenhof, M.N. Kochert, and J. A. Roppe.
Interactions between ravens and transmission lines. 14th Annual
Meeting and Symposium of the Desert Tortoise Council, Mesquite,
Nevada.

3/21/89

4/15/89

4/25/89
4/27/89

6/23/89

6/25/89

McAnnis, P., and M. J. Bechard. Home range and habitat use of
Ferruginous Hawks in the Snake River Birds of Prey Area. Northwest
Science Meetings. Richland, Washington.

fog 61 * K - Communal roosting of Common Ravens in southwestern
Idaho. Graduate Seminar. University of Wisconsin, Madison.

Kochert, M. N. Reproductive performance, food habits, movements,
and subadult breeding of Golden Eagles. Graduate seminar on Eagle
Biology. Boise State University. Boise, Idaho.

Marti, CD. Feeding, Reproduction, movements, and survival of the
common barn-owl in the Great Basin. Symposium on Long-term
Ecological Studies of Birds. 59th Annual Meeting of the Cooper
Ornithological Society. Moscow, Idaho.

Steenhof, K^ Variations in Golden Eagle reproductive rates:
effects of prey and weather. 59th Annual Meeting. Cooper
Ornithological Society. Moscow, Idaho.

The

10/12/89 Spahr, R. , M. Bechard, and K. Steenhof . Factors affecting

distribution of wintering bald eagles on the Boise River in Idaho.
Joint meeting, Raptor Research Foundation and ICBP World Working
Group of Birds of Prey. Vera Cruz, Mexico.

10/13/89 Ulmschneider , H. Post-nesting movements of long-eared owls in
southwest Idaho. Joint Meeting, Raptor Research Foundation and
ICBP World Working Group of Birds of Prey. Vera Cruz, Mexico.

PART II
PROGRESS REPORTS

TITLE: Nesting Density and Productivity of Golden Eagles in and near
the Snake River Birds of Prey Area.

INVESTIGATORS: Karen Steenhof , BLM Birds of Prey Research

Michael N. Kochert, BLM Birds of Prey Research

OBJECTIVES :

1. To determine occupancy, nesting success, and productivity at
traditional golden eagle ( Aquila chrysaetos ) nesting territories.

2. To compare occupancy and productivity of golden eagle pairs in the
Snake River Birds of Prey Area (SRBOPA) with those of eagle pairs in
the Comparison Area.

ANNUAL SUMMARY

In 1989 the number of occupied golden eagle territories in the SRBOPA was
lower than in 1988, but percent of eagle pairs breeding and percent of
breeding pairs successful were both higher in 1989 than in 1988. Overall,
eagle productivity was higher than it has been since 1983 . As in the past 5
years, eagle productivity was lower in the SRBOPA than in the Comparison
Area. Eagles that nested on a transmission line were more productive than
eagles that nested on cliffs.

METHODS

Golden eagle territories in the SRBOPA and Comparison Area (Fig. 1) were
surveyed for occupancy from a Hiller/Soloy jet helicopter on 16 March. Some
territories on or near the Pacific Power and Light Company (PP&L) 500-kV
transmission power line were also checked from a Bell 206 Jet Ranger on 12
April, a Hiller/Soloy on 4 May, and/ or a Messerchmidt Boelkow on 22 May.
Territories where breeding status could not be determined during the March
helicopter flight were subsequently surveyed from the ground for signs of
occupancy or breeding (see below for criteria). Territories containing
breeding pairs were checked for productivity from a Bell 206 Jet Ranger on 8
and/or 21 June.

We attempted to ascertain the breeding status and nesting success of all
eagle pairs. We climbed to 1 golden eagle nest to band young. Eagle pairs
that showed no evidence of egg laying after repeated observations or after
climbing into and examining potential nests were categorized as
"nonbreeding . " A "breeding attempt" was confirmed if an occupied territory
contained an incubating adult, eggs, young, or any other indication that eggs
were laid (e.g., fresh eggshell fragments in fresh nesting material). A
breeding attempt was "successful" if it produced 1 or more young that reached
fledging age (i.e., 51 days for golden eagles (Steenhof 1987)). Eagle nests
discovered after young had fledged were considered successful if: 1) a
platform decorated this season was worn flat and contained fresh prey

LEGEND

BIRDS OF PREY STUDY AREA

COMPARISON AREA

BIRDS OF PREY NATURAL AREA

SE R HE

Fig. 1. LOCATION OF THE BIRDS OF PREY STUDY AREA (BPSA) AND COMPARISON AREA

remains, 2) fresh fecal matter covered the back and extended over the edge of
the nest, and 3) no dead young birds were found within a 50-m radius of the
nest.

Percent of pairs breeding in 1989 and earlier years was calculated from
preselected pairs (Steenhof and Kochert 1982). Percent of nesting attempts
successful was based on attempts found during incubation. Nests manipulated
by fostering , disease treatment , shade devices , etc . , were excluded from
productivity calculations.

RESULTS

Nesting Density

In 1989 golden eagle pairs occupied 30 (79%) of the 38 traditional eagle
nesting territories in the SRBOPA. The total number of occupied territories
in the SRBOPA was 2 less than in 1988 (Table 1). All but 2 of 29 territories
surveyed outside the SRBOPA (Hammett GN and Malad) were occupied in 1989. Of
the 10 vacant sites, 6 have not been occupied since 1982 or earlier. Three
vacant sites (Strike II, Swan Dam, and PP&L 125) had been occupied by eagles
in 1988. The Upper Sinker nesting territory was occupied after having been
vacant in 1988. The Pump Station nesting territory was vacant for the 2nd
consecutive year.

Reproduction

Golden eagle reproduction in the SRBOPA increased slightly from 1988
levels (Table 1). Percent of pairs breeding increased from 74% in 1988 to
80% in 1989, and percent of attempts successful increased from 55% in 1988 to
63% in 1989. Young fledged per pair increased from 0.63 in 1988 to 0.69 in
1989 (Table 1). As in the past 5 years, golden eagle reproduction was higher
in the Comparison Area than in the SRBOPA (Table 2). Number of young eagles
raised per pair in the Comparison Area decreased from 0.99 in 1988 to 0.81 in
1989. Eighty-six percent of eagle pairs in the Comparison Area laid eggs.
Although the percent of attempts successful in the Comparison Area (61%) was
slightly lower than in the SRBOPA (63%), Comparison Area pairs fledged more
young per successful attempt (1.55) than pairs in the SRBOPA ( 1 . 36 ) .
Throughout the southwestern Idaho study area, power line nesters fared better
than canyon nesters in 1989 (1.00 and 0.70 young fledged per pair,
respectively; Table 2).

ACKNOWLEDGMENTS

Pacific Power and Light Company is acknowledged for providing flight time
to conduct the eagle helicopter surveys. K. Troxel and P. Kochert assisted
with golden eagle surveys. We thank R. Holman, Idaho Department of Fish and
Game Region IV, for making observations at eagle nests in the Comparison

Area.

Table 1. Nesting success and productivity of golden eagles in the SRBOPA,
1971-1989. Sample sizes are in parentheses.

Year # of % of Pairs % of # Fledged/ # Fledged Total

Occupied Breeding Attempts Successful Per Pair # Fledged
Territories Successful Attempt

1971

100%

(31)

60%

(20)

1.89

19)

1.13

1972

—

1.64

11)

—

1973

65%

(34)

44%

(18)

1.37

0.39

1974

73%

(30)

56%

(18)

1.42

12)

0.58

1975

75%

(32)

56%

(18)

1.43

14)

0.60

1976

70%

(33)

47%

(15)

1.62

13)

0.53

1977

82%

(33)

59%

(17)

1.53

15)

0.74

1978

80%

(30)

70%

(10)

1.71

17)

0.96

1979

97%

(30)

61%

(23)

1.53

19)

0.91

1980

87%

(31)

72%

(18)

1.77

22)

1.11

1981

100%

(30)

74%

(23)

1.73

22)

1.28

1982

87%

(30)

80%

(25)

1.95

19)

1.36

1983

96%

(27)

72%

(18)

1.56

16)'

1.07

1984

—

61%

(18)

1.55

11)

0.60*

1985

39%

(31)

42%

(12)

1.00

0.16

1986

54%

(28)

29%

(14)

1.33

0.21

1987

78%

(32)

32%

(25)

1.38

0.34

1988

74%

(31)

55%

(22)

1.54

13)

0.63

1989

80%

(30)

63%

(19)

1.36

14)

0.69

* based on % of pairs successful

Table 2. Nesting success and productivity of golden eagles in southwestern
Idaho, 1989. Sample sizes are in parentheses.

Study Area % of Pairs % of Attempts # Fledged/ # Fledged

Breeding Successful Succ. Attempt Per Pair

SRBOPA* 80% 63% 1.36 0.69

(30) (19) (14)

Comparison 86% 61% 1.55 0.81

(22) (19) (11)

PP&L 500 kV 100% 50% 2.00 1.00

(4) (4) (2)

Canyon ** 81% 62% 1.39 0.70

(48) (34) (23)

* includes the pair at PP&L 119.

** includes all cliff nests in the SRBOPA and Comparison Area.

LITERATURE CITED

Steenhof, K. 1987. Assessing raptor reproductive success and productivity.
Pages 157-170 in B. A. Giron Pendleton, B. A. Millsap, K. W. Kline, and
D. M. Bird, eds. Raptor Management Techniques Manual. Natl. Wildl.
Fed. , Washington, D.C.

Steenhof, K. , and M. N. Kochert. 1982. An evaluation of methods used to
estimate raptor nesting success. J. Wildl. Manage. 46:885-893.

TITLE: Monitoring Long-eared Owl Occupancy in the Snake River Birds of
Prey Area.

INVESTIGATOR: John Doremus, Boise District, BLM

OBJECTIVES: To locate all occupied long-eared owl territories on public
lands in the Snake River Birds of Prey Area (SRBOPA) .

INTRODUCTION

The first effort to systematically locate long-eared owls (Asio otus) in
the SRBOPA took place in 1980, when Jeffrey Marks located 64 breeding pairs
(Marks 1980). In 1981 Marks located 41 pairs of breeding long-eared owls
(Marks 1981). During both years Marks searched all suitable riparian habitat
along the Snake River and its tributaries from the Indian Cove Bridge to the
Mouth of Sinker Creek. During 1985 Doremus and Car la Schroer visited 72
traditional long-eared owl nest sites and found 19 of the sites occupied by 1
or more owls (Doremus and Schroer 1985). No systematic effort has been made to
locate all occupied long-eared owls sites since 1985. In 1989 I planned to
visit all potential long-eared owl habitat on public land in the SRBOPA.
Conflicts in scheduling reduced the time available for the long-eared owl
surveys. I was not able to survey all potential long-eared owl habitat in the
SRBOPA in 1989 (Fig. 1).

METHODS

Riparian areas where there were sufficiently developed shrubs and trees
were entered and searched for long-eared owls or signs of long-eared owl
occupancy from mid-March to mid-April. Corvid nests and other structures that
could serve as a long-eared owl nest site were observed for owls or signs of
owl eggs, young, or feathers and down. I also looked for roost sites with
fresh casts and fecal material. The number of passes through a riparian area
depended on the width of the area and the density of the vegetation. A site
was considered occupied if long-eared owls attempted to nest, behaved in a
defensive manner, or if 1 or 2 owls were seen at a traditional site 2 or more
times during the breeding season.

RESULTS

Sixty-nine of 80 traditional sites (sites found prior to 1989) were
visited during the 1989 survey. Thirty-eight of the 69 traditional sites were
visited frequently enough that occupancy of the site could be determined.
Twenty- three of the 38 traditional sites were occupied by one or more owls .
Seven new sites were located. Single birds were found at 3 traditional sites,
but occupancy was not determined during my visit to these sites and no follow
up visits were made. Long-eared owls were found at 8 other locations where
they have not been found in the past. Follow up visits found no owls at two of
these locations. There were no follow up visits at the other 6 locations. The
density of the vegetation made complete surveys of the Barber Islands, Pump
Draw Islands and Flat Iron Island impossible.

« Melba

w
o

V, Walters Ferry
Bridge

^L Swan Falls Dam

Scale 1 inch = 10 miles
'/."/ Search Area

Hammett

Fig. 1. Location of long-eared owl surveys, 1989.

Breeding attempts were observed at 20 sites. Ten pairs of owls were
successful at fledging young, 5 pairs were unsuccessful, and nestxng success
was not determined at the breeding sites of 5 pairs. No effort was made to
locate successful breeding sites so it is likely that other pairs were
successful at fledging young.

DISCUSSION

The temporal overlap of winter roosts and the breeding season of
long-eared owls may lead an investigator to count winter roost stragglers as
owls occupying a breeding site. To conduct a complete survey for long-eared
owls in the SRBOPA it is necessary for the investigator to search all
potential owl habitat, not just sites that have been occupied in the past.

ACKNOWLEDGMENTS

Marks J. S. 1980. Productivity, nest site characteristics, and food habits
of Long-eared Owls in the Snake River Birds of Prey Study Area. Page
36 in Snake River Birds of Prey Res. Proj. Annu. Rep. U.S. Dep.
Inter., Bur. Land Manage., Boise, Idaho.

Marks, J. S. 1981. Productivity, nest site characteristics, and food habits
of Long-eared Owls in the Snake Rive Birds of Prey Study Area. Pages
28-36 in Snake River Birds of Prey Res. Proj. Annu. Rep. U.S. Dep.
Inter., Bur. Land Manage., Boise, Idaho.

Doremus, J. H., and C. Schroer. 1985. Occupancy and Breeding Success of

Long-eared Owls and Northern Harriers in the Snake River Birds of Prey
Area. Pages 145-148 in K. Steenhof and M.N. Kochert, eds., Snake River
Birds of Prey Res. Proj. Annu. Rep. U.S. Dep. Inter., Bur. Land
Manage., Boise, Idaho.

TITLE: Raptor and Raven Nesting on the PP&L Malin to Midpoint 500-kV
Transmission Line.

INVESTIGATORS: Karen Steenhof , BLM Birds of Prey Research

Michael N. Kochert, BLM Birds of Prey Research
Jerry Roppe, Pacific Power & Light Company
Mike Mulrooney, Pacific Power & Light Company

COOPERATOR: Pacific Power and Light Company

OBJECTIVES:

1. To identify all occupied raptor and raven nests on the first 371
miles of the PP&L Malin to Midpoint 500-kV transmission line.

2. To ascertain nesting success and productivity of raptors and ravens
nesting on the first 133 miles of the line.

3. To ascertain nesting success and productivity of golden eagles and
ferruginous hawks nesting on natural substrate near the intensively
studied portion of the line for comparative purposes.

4. To identify preferred nesting locations on the towers and to
document how these locations relate to nesting success and possible
contamination .

5. To assess physiographic features that may influence use of towers by
nesting raptors and ravens.

INTRODUCTION

Construction of a 500-kV transmission line across southern Idaho and
Oregon in 1980-81 provided government and industry biologists with an
opportunity to investigate the biology of raptors and ravens nesting on
transmission lines. Pacific Power & Light Company (PP&L), in
cooperation with the Bureau of Land Management (BLM), has surveyed the 500-kV
transmission line between Midpoint (Jerome), Idaho and Malin, Oregon (Fig. 1)
since 1981, and intensive research has been carried out since 1984. This
report presents findings of the 6th and final year of intensive studies of
raptor and raven nesting density and productivity on the line.

METHODS

Surveys were conducted in 3 study areas in 1989. The "intensive survey
area" consisted of Miles 0-133 of the transmission line, and the "extensive
survey area" consisted of Miles 134-371 (Fig. 1). To compare nesting success
and productivity, golden eagles ( Aquila chrysaetos ) were surveyed in a third
"canyon study area" extending along the Snake River and its major tributaries
from Hagerman, Idaho to Mile 135.

400

Kilometers

Fig. 1. Location of the Pacific Power and Light 500 kV
transmission line. Intensive survey stretch
is shaded.

The entire line was surveyed from a Bell 206 Jet Ranger helicopter
between 10 and 12 April to locate incubating pairs. Some eagle nests on
cliffs were also checked during the April survey. Follow-up surveys of the
entire intensive study area were conducted from a Hiller/Soloy helicopter on
4 May and from a Messerschmidt Boelkow helicopter on 22 May. Individual
nests in the intensive area were spot-checked from a Bell 206 Jet Ranger on 8
and 21 June. The extensive area was re-surveyed on 23 May from a
Messerschmidt Boelkow. Traditional golden eagle nesting territories in the
Snake River Canyon were checked from a Hi Her Soloy on 16 March, and all were
re-checked either from the ground or during helicopter flights in May or
June. During all surveys helicopters were flown at speeds of 70-95 km/hr; we
usually hovered approximately 10 m from nests for 5-25 sec to view nest
contents. Some nests were photographed from the helicopter, and some were
subsequently observed from the ground.

Pairs were considered "breeding" if they laid at least 1 egg; this was
confirmed by observing eggs, young, or an incubating adult. Because aerial
surveys covered only the power line structures, some nonbreeding pairs could
have been missed. We considered a breeding attempt successful if 1 or more
young reached 80% of the average age when most young normally leave the nest.
Nestlings were aged by comparison with photographs of known-age chicks
(Moritsch 1983, 1985; BLM, unpubl. data).

Data from 1984 to 1989 were reviewed to evaluate survey timing and to
make recommendations for scheduling a single monitoring flight in future
years. Data from each complete survey flight (1984-1989) were reviewed to
tabulate the number of raptor and raven breeding attempts that were confirmed
during that particular survey. Totals from each flight were then compared
with the total number of breeding attempts identified in that particular
year. Data were tabulated separately for the intensive survey area (Miles
1-133) and the extensive survey area (Miles 133-371).

RESULTS
Nest Density and Distribution

In 1989, 133 pairs of raptors and ravens occupied nests on the 500-kV
transmission line. Breeding attempts were confirmed for 129 of the 133
nesting pairs on the line. Breeding could not be confirmed at 3 occupied
red- tailed hawk ( Buteo jamaicensis ) nests and 1 common raven ( Corvus corax)
nest. There were no confirmed re-nesting attempts in 1989.

Ravens were still the most numerous species nesting on the line.
Eighty-one raven, 33 red- tailed hawk, 11 ferruginous hawk ( Buteo regalis ) ,
and 8 golden eagle pairs occupied nesting territories associated with nests
on the line in 1989 (Table 1). The 133 occupied nesting territories
represent the highest number of nesting pairs observed on the line since
studies began. The total, however, is only 2 more than the previous high of
131, recorded in 1987 and represents only a 4% increase from 1988 levels.
Numbers of red- tailed hawk and raven pairs were the same as in 1987, their
previously recorded high levels. Golden eagle numbers were unchanged from
1988; ferruginous hawk pairs declined by 1; and great horned owls ( Bubo
virginiaus ) did not nest on the line in 1989.

Table 1 . Number of occupied raptor and raven nesting territories found on the PP&L
Malin to Midpoint 500-kV transmission line, 1981-89.

Species

1981

1982

1983

1984

1985

1986

1987

1988

1989

Golden Eagle

Ferruginous Hawk

Red- tailed Hawk

Common Raven

Great Horned Owl

TOTAL

113

131

128

133

The golden eagle pair that nested at Mile 125 in 1988 did not nest on the
line in 1989. As in the past 2 years, the golden eagle pair at Little Canyon
Creek (49/3) nested on one of the traditional cliff nests, approximately 400
m from the line. A new eagle territory was established in Mile 352 in 1989.
Eagles nested on a platform that had been used by red-tailed hawks in 1988 .

The number of raptor and raven pairs decreased by 2 from 1988 levels in
the intensive study area and increased by 7 in the extensive area (Table 2).
Red-tailed hawks within the extensive area increased from 22 in 1988 to 27 in
1989, and numbers in the intensive area increased from 5 to 6. The extensive
area had 1 more raven pair than in 1988, and numbers in the intensive area
were unchanged. As in past years, ferruginous hawks nested only in the
intensive area.

Nesting densities of all species combined were higher in the intensive
study area (0.33 per km; 0.54 per mi) than in the extensive area (0.16 per
km; 0.26 per mi). For the entire line, densities averaged 0.22 nests per km
or 8 nests for every 100 towers.

Significantly more breeding pairs occurred between Miles 59 and 127,
within or just north of the Snake River Birds of Prey Area (SRBOPA), than in
the remaining half of the intensive study area (X 2 = 4.63, P < 0.05).
However, numbers of pairs near the SRBOPA decreased in 1989, while numbers in
other parts of the intensive study area increased. As in past years, nesting
densities were particularly high along one stretch of line within the SRBOPA
(Miles 113-119). In 1989, this stretch contained more than 0.7 nesting pairs
per km (1 per mile), more than 3 times the average density for the line.

Marked Birds

One of the golden eagles nesting on the platform at Tower 68/3 was
wearing a blue and yellow wing marker on its right wing. The markings
suggest that the bird was marked as a nestling at a cliff nest in or adjacent
to the SRBOPA in 1980. This is the first sighting of a marked adult nesting
on the transmission line. The marked bird was incubating on 4 May 1989, but
the nesting attempt was unsuccessful.

Nest Site Selection

Raptors and ravens nested in all tower types on the line except the F
type in 1989. The frequency with which the tower types were used, however,
differed significantly from the frequency with which tower types were
available (X z = 5.97; P < 0.05). Although A towers (Fig. 2) were the most
commonly used towers, they were by far the most common tower type available.
Only 7% of the A towers were used in 1989, and 16 of the 80 breeding pairs
that nested on A towers nested on platforms and not on the tower
latticework. In contrast, raptors and ravens nested on 30% of the available
E towers, 11% of the C towers, and 8% of the T towers. B towers were used at
the same rate (7%) as the A towers. The sturdier and more extensive
latticework on the less common T and E towers (Fig. 2) may provide more
suitable nesting substrate and therefore account for the raptors' and ravens'
apparent preferences for them.

Table 2. Number of occupied raptor and raven nesting territories found in the intensive and extensive
study areas, 1984-89.

Intensive E xtensive

1984 1985 1986 1987 1988 1989 1984 1985 1986 1987 1988 1989

Golden Eagle

Ferruginous Hawk

Red-tailed Hawk

Common Raven

Great Horned Owl

* Does not include Little Canyon Creek (49/3) because the pair nested on a cliff from 1987 to 1989.

Fig. 2. Tower types used by raptors and ravens for nesting. The "E" (center)
and "T" (right) towers were preferred over the more common "A" towers
(left) apparently because of their sturdier and more extensive latticework.

Red-tailed hawks used more positions on the tower for nesting than the
other 3 diurnal species (Fig. 3). Ravens appeared to be the least versatile
in nest site selection, nesting in the x-position in every case but one in
1989. The x-position of the tower (Fig. 3) was the most frequently used
position on the towers, used by 79 of 80 breeding raven pairs, 13 of 29
red- tailed hawk pairs, 2 of 8 golden eagle pairs, and 7 of 11 ferruginous
hawk pairs.

More than 50% of the golden eagle and ferruginous hawk pairs nested on
the specially designed artificial platforms that were installed at the time
of line construction. A preference by hawks and eagles for the platforms is
apparent because only 2% of the towers on the line contain platforms.
Seventeen of 37 available platforms were used by breeding raptors in 1989 (6
by golden eagles , 4 by ferruginous hawks , and 7 by red- tailed hawks ) . Nine
of 12 platforms in the intensive area were used compared with only 8 of 25 in
the extensive area. More platforms were used in 1989 than in any other
single year. However, all platforms used in 1989 had been used in a previous
year. Only 2 platforms that had been used in previous years were not used in
1989. The Little Canyon Creek golden eagle pair nested on the cliff instead
of on the 49/3 platform for the 3rd consecutive year. The ferruginous hawk
pair which has nested on the 104/4 platform from 1982 to 1988 moved to the
X-south position of the same tower in 1989. For several years, nesting
ferruginous hawks had been bringing an unusually high number of sticks to the
104/4 platform nest, and by 1988, the stick nest was taller than the platform
itself.

The 79 raven pairs that nested in the x-position showed no significant
preference for side of tower. Forty nested on the south end of the tower,
and 39 nested on the north side. However, 5 of 7 ferruginous hawks that
nested in the x-position used the south side of the tower. Red-tailed hawks
showed only a slight preference for x nests on the south side of the tower (9
of 13 attempts in the x-position).

Nesting Success and Productivity

Reproductive rates of raptors in the intensive study area were generally
higher in 1989 than in 1988. All golden eagle, ferruginous hawk, and common
raven pairs associated with nests on the line laid eggs . Two of 4 eagle
pairs were successful and raised 2 young each (Table 3). Percent of nesting
attempts successful declined from 1988 levels for eagles and ravens but
increased for ferruginous and red- tailed hawks. For each species of diurnal
raptor, number of young fledged per pair in the intensive area was higher in
1989 than in 1988. Number of young fledged per raven pair declined slightly
from 1988 (Table 3).

Wind was the only confirmed cause of nest failure in 1989. Two raven
nests were blown from the tower between 4 May and 8 June. In addition, a
fledged raven was found dead below its nesting tower in Mile 105.

Golden Eagle (N = 8)

x 25

P 75

Ferruginous Hawks (N = 11)

X 36

p 64

Red-tailed Hawk (N = 29)

C —
x —
p —

w —

D —
B —

— 3

— 45

— 24

— 21

— 3

Common Raven (N = 80)

x 99

D 1

Fig. 3 Percent of raptor and raven breeding attempts in relation
to tower position for the PP&L Malin to Midpoint 500kV
transmission line, 1989. Sample sizes are shown in
parentheses. Summaries do not include "T" towers.

Table 3. Nesting success and productivity of raptors and ravens on the
intensive survey stretch of the PP&L Malin to Midpoint 500-kV
transmission line, 1989.

% of

No.

% of

Breeding

Fledged

No. of

Pairs

Attempts

Per Succ.

Per

Pairs

Breeding

Successful

Attempt

Pair

Golden Eagle

100%

50%

2.0

1.0

Ferruginous Hawk

100%

91%

2.6

2.4

Red- tailed Hawk

100%*

100%**

2.0

Common Raven

100%

79%***

3.9

3.1

* breeding was unconfirmed for 2 pairs

** success was unknown for 1 pair

*** based on all nesting attempts (including re-nests) where outcome was
known. The Mayfield (1961) estimate of nesting success was 72%.

Survey Timing

Each year from 1985-1989, the early May survey identified the highest
percentage (85-96%) of all known breeding attempts in the intensive survey
area (Table 4), and the April survey identified the fewest (46-76%). Late
May surveys were always intermediate (81-87%). In 1984, when there were no
early May flights, a mid-May flight identified the highest percentage of
breeding attempts (90%), a late March flight identified only 12%, and a late
April flight identified 81%.

When plotted as a function of time (Fig. 4), survey accuracy (as defined
by % of total known breeding attempts confirmed) clearly peaked between 3 and
17 May. April surveys missed a significant number of pairs that had not yet
initiated breeding attempts. Late May surveys missed pairs that had either
failed or already successfully fledged young. The early May surveys appeared
to minimize both of these sources of errors.

In the extensive area (Table 5), late May surveys always identified more
nesting pairs (82-100%) than early April surveys (34-83%). The analysis for
the intensive area is more meaningful because multiple surveys ensured more
complete counts of breeding pairs. The high percentage of attempts
identified in late May surveys of the extensive area does not necessarily
indicate that late May surveys are more complete than early May surveys.
Instead, it may suggest that the number of breeding attempts in the extensive
area has been consistently underestimated because nests are missed in both
the early April and late May surveys.

RECOMMENDATIONS AND PLANS FOR NEXT YEAR

We plan to prepare a manuscript for publication that summarizes 6 years
of data collected on raptors and ravens nesting on the transmission line.
Only one aerial survey of nesting raptors and ravens on the line is planned
for 1990. Results suggest that early May is the optimal period to schedule a
single survey flight, assuming that the objective of the flight is to
identify all nesting pairs on the line. If surveys are carried out using the
same procedures used from 1985-1989, they should be expected to identify
85-95% of all breeding attempts on the line. Data on reproduction of golden
eagles and ferruginous hawks nesting in the canyon will be collected as part
of a 4-year cooperative effort to assess the effects of National Guard
activity and habitat alteration on birds of prey.

ACKNOWLEDGMENTS

We thank M. Garrett and S. Wilder who assisted with data collection and
compilation. We are also indebted to our pilots: R. McLellan, T. Miller
(Idaho Helicopters), D. Day, (Pacificorp) , and M. Pearson (Reeder Flying
Service ) .

Table 4. Percent of known breeding attempts identified on individual survey
flights of the intensive survey area, 1984-1989.

Flight # 1984 1985 1986 1987 1988 1989

12% 54% 46% 73% 53% 76%

first (20 March) (9 April) (9 April) (14 April) (7 April) (12 April]

81% 90% 85% 93% 87% 96%

second (23 April) (6 May) (6 May) (7 May) (4 May) (4 May)

90% 85% 80% 87% 83% 81%

third (16 May) (24 May) (27 May) (22 May) (19 May) (22 May)

120

110

100

580
*70
1 60

£50

ES40

30
20
10

80 98 99 gg 102 104 114 125 124 126 126 127137 140 142 142 144 147

JULIRN DATE

Figure 4. Percent of Breeding Attempts Identified on Helicopter Survey Flights 1985-1989

Table 5. Percent of known breeding attempts identified on individual survey
flights of the extensive survey area, 1985-1989.

Flight # 1985 1986 1987 1988 1989

34% 60% 74% 81% 73%

first (8/9 April) (10 April) (13 April) (8 April) (10 April)

100% 85% 89% 88% 92%

second (20 May) (21 May) (21 May) (18 May) (23 May)

LITERATURE CITED

Kochert, M. N. , K. Steenhof, J. Roppe, and M. Mulrooney. 1984. Raptor and
raven nesting on the PP&L Malin to Midpoint 500 kV transmission line.
Pages 20-39 in Snake River Birds of Prey Research Project Annual Report,
1984. U.S. Dep. Inter., Bur. Land Manage., Boise, Idaho. 145pp.

Mayfield, H. F. 1961.
Bull. 73:255-261.

Nesting success calculated from exposure. Wilson

Moritsch, M. Q.
hawks . U.S.

1983. Photographic guide for aging nestling red-tailed
Dep. Inter., Bur. Land Manage., Boise, Idaho. 14pp.

hawks.

. 1985. Photographic guide for aging nestling ferruginous

U.S. Dep. Inter., Bur. Land Manage., Boise, Idaho. 21pp.

Steenhof, K. , M. N. Kochert, J. Roppe, and M. Mulrooney. 1987. Raptor and
raven nesting on the PP&L Malin to Midpoint 500 kV transmission line.
Pages 19-33 in K. Steenhof, ed. , Snake River Birds of Prey Research
Project Annual Report, 1987. U.S. Dep. Inter., Bur. Land Manage., Boise,
Idaho. 159pp.

TITLE: Implications of Communal Roosting by Common Ravens to

Operation and Maintenance of the Malin to Midpoint 500-kV
Transmission Line.

INVESTIGATOR: Alison L. Beck, BLM Birds of Prey Research

COOPERATOR: Pacific Power and Light Company

OBJECTIVES :

1. To monitor occupancy of the Initial Point communal raven roost
and trends in numbers of roosting ravens on the Malin to Midpoint
500-kV transmission line.

2. To evaluate the effectiveness of shields and pegging in
preventing insulator contamination at the Initial Point Roost.

INTRODUCTION

Research to determine the implications of communal roosting by common ravens
( Corvus cora x) to operation and maintenance of the Malin to Midpoint 500-kV
transmission line was completed in 1987. The final research report identified 3
monitoring objectives that were to be made part of the contamination management
program: evaluating shields and pegging, monitoring contamination, and
determining locations and causes of outages.

In May 1989, shields and pegging were installed on 24 towers at the Initial
Point roost, between miles 107 and 112. Contamination estimates were conducted
immediately after installation and washing of insulators and again 5 months
later. The roost was monitored for occupancy and trends in numbers of roosting
ravens from the first date of installation until 1 November. This report
presents the results of the 1989 monitoring effort.

METHODS

Fiberglass shields and plastic pegging were installed on 24 towers at the
Initial Point roost, from tower 2/107 to 4/112. Tower 1/110, a turn ("E")
tower, was excluded because it was not vulnerable to contamination (Young and
Engel 1988a). Shields and their installation are described by Young and Engel
(1988a). Pegging consisted of 0.95-cm (3/8-in) plastic tubing, cut in 30-cm
(12-in) lengths and glued into holes drilled in 5-cm (2-in) PVC pipe. Each
piece of PVC pipe was fitted with 36 pegs, 2 attached to the outer half of north
and south C sections on all treated towers with wire. Installation began on 21
May and was completed on 24 May. Insulators were washed at the time of
installation at towers with contamination determined to be severe by the line
patrolman (M. Mulrooney, pers commun. ) . Insulator strings on towers 111/4,
112/1, and 112/2 were treated by washing.

Contamination estimates were made at each tower fitted with shields and
pegging on the day following completion of installation. Estimates were made
according to procedures described by Young and Engel (1988a). Estimates were

made daily from 22 to 25 May. Contamination estimates were not made for towers
110/2 and 110/3 where insulators are made of an opaque ceramic material. Cloud
cover ranged from 10% to 90% on days when contamination estimates were conducted
in May. Contamination was scored for the same 22 towers 5 months later, on 1
November. Cloud cover was 0% at that time. Estimates were made for all 6
insulator strings at every tower. Individual bell scores for the 2 center phase
and for each of the outermost strings (directly under pegging on C sections)
were used for analysis. Two-sample median tests were used to determine whether
shields and pegging are effective in preventing insulator contamination (Young
and Engel 1988a) .

At the time of the final contamination estimates, shields and pegging were
visually inspected from the ground for damage.

Evening observations of roosting ravens began on 22 May and were conducted
nightly until 29 May. Beginning 6 June, Tuesday evening observations were done
biweekly until 1 November. The counts were done from an observation point
approximately 500 m south of the transmission line. The 24 treated towers were
monitored from 30 min before sunset until it was too dark to accurately count
ravens, usually 30 min after sunset. Every 10 min, an observer with a spotting
scope counted the number of ravens on each tower section (Young and Engel
1988a: Fig. 3-1). The highest total number of ravens counted during a 10-min
interval was considered the night's count for the roost.

Pacific Power and Light Company monitored for line trips and outages.

RESULTS

Occupancy of Roost and Numbers of Roosting Ravens

Twenty of the 24 treated towers were occupied by roosting ravens during the
study period. No roosting birds were observed at either the extreme western
(112/4 and 112/3) or eastern (107/2 and 107/3) ends of the roost. In May and
June, towers 111/3 through 112/2 were most frequently used. From early June
through September, birds occupied the center portion of the roost, from 108/2 to
111/1. Use of towers was restricted to the area from 109/3 to 111/2 in late
September and October. The maximum number of evenings any 1 tower was observed
to be occupied through the study period was 9. In the early part of the study
period, tower 112/1 was occupied by roosting ravens on 9 of 11 evenings
observation. Tower 110/3 was occupied during each of 9 evening observations
from 11 July to 1 November.

Maximum numbers of roosting ravens were highest from July through October
(Fig. 1). Approximately 100 birds were counted at the roost in May and June
(mean = 103.00 ± 91.33). The mean maximum count from 11 July through 3 October
was 556 (± 107.8). Numbers of ravens observed on 17 October and 1 November were
59 and 113. The highest maximum count was 675 on 5 September.

700

Julian Date

Fig. 1 . Numbers of roosting ravens, Initial Point roost, 1 989.

Effectiveness of Shields and Pegging

Immediate Response of Ravens

Installation of shields and pegging took place over a period of 5 days, with
2 crews working from opposite ends of the roost toward the middle. Throughout
the period, towers 111/4, 112/1 and 112/2 were used by roosting ravens. On 22
May, towers 112/1 and 112/2 were treated, ravens occupied them that evening
without apparent hesitation. The following day, 111/4 was among the towers
treated; it was occupied that evening.

Use of Towers

Ravens roosted on 20 of the 24 treated towers. None of the towers east or
west of the treated segment and visible from the observation point were occupied
during the study period. Tower 110/1, the only untreated tower of the Initial
Point roost, was consistently used by roosting ravens after the end of July.

Distribution of Ravens Among Tower Sections

Significantly fewer ravens roosted above insulators during the week of
installation and the first 3 weeks following installation than after that period
(1-way ANOVA, F = 11.585, df = 2, P < 0.01) (Table 1). During installation and
the 3 weeks following, more ravens used the D sections than any other, while
after that time the B section was most heavily occupied. Throughout the study
period, fewer than 10% of roosting ravens occupied the C sections. Birds were
not observed roosting on the portions of C sections fitted with pegging. Ravens
occupied the inner half of the C section, and on 2 occasions the horizontal bar
below the pegging. Even where individual pegs were lost from C sections, ravens
did not roost in the resulting gaps.

Use of tower sections during the period after 3 weeks following installation
in 1989 differed from use at the Initial Point Roost in 1984 and 1985 (Young and
Engel 1988a). The proportion of birds roosting above insulators was slightly
higher in years prior to installation (82%) than in 1989 (78.9%). Use of C
sections was lower in 1989 (8.4%, n = 330) than in previous years (23%, n -
17,846). More ravens roosted on X and A sections and on A wires in 1989 than
reported for 1984 and 1985.

Contaminat i on

Contamination estimates were made for 22 treated towers in May, immediately
following treatment and washing, and again 5 months later in November (Table
2). Towers 110/3 and 110/4 have opaque porcelain insulators and were not scored
for contamination. Scores for the 2 innermost insulator strings and for the 2
outermost strings for May and November were compared. Calculated Spearman rank
coefficient resulted in an r value of 0.362, indicating no significant
correlation between changes in contamination scores and number of tower nights
of occupancy (P > 0.05).

Table 1. Distribution of roosting ravens among tower sections at the Initial Point roost during 1984 and 1985,
during and 3 weeks following installation of shields and pegging (22 May-27 June 1989) and after 3
weeks following installation (11 July-1 November 1989).

£»

Period

No . tower-

Above

Tower Sections

evenings

insulators

1984, 1985«

75948

17846

25»

3017

22959

11382

460

1559

82.0%

23.0%

25.0%

4.0%

30.0%

15.0%

1.0%

2.0%

Installation and

1235

108

216

261

390

3 weeks following

62.0%

7.9%

8.7%

17.5%

6.8%

21.1%

31.6%

3.0%

3.4%

After first

4085

330

248

978

413

1289

698

3 weeks

78.9%

8.4%

6.3%

24.9%

10.5%

32.9%

17.8%

2.1%

1.1%

« Young and Engel 1988

b sections A and X combined

Table 2. Use of towers and contamination trends at the Initial Point Roost following installation
of shields and pegging, 22 May - 1 November, 1989.

Mean number of ravens/night
Tower Above insulators On C Sections (n)

Contamination trend (May-November)
Center Insulators Cuter insulators

107/2

107/3

6.617,

0.05)

0.543,

0.05)

107/4

■-

0.332,

0.05)

6.382,

0.05)

108/1

24.50

0.50

-:•

0.50

(1)

1.010,

0.05)

1.010,

0.05)

108/3

49.00

26.52

7.00

9.30

(4)

108/4

35.25

•:•

20.84

6.50

7.37

(4)

108/5

63.00

17.01

18.25

■!-

12.52

(4)

7.526,

0.05)

7.562,

0.05)

109/1

18.00

-!•

7.00

(2)

5.827,

0.05)

2.040,

0.05)

,fc>

109/2

38.50

10.01

5.00

0.71

(4)

0.378,

0.05)

109/3

54.57

35.80

7.14

5.25

!7)

2.777,

0.05)

3.092,

0.05)

109/4

65.67

14.51

13.83

9.25

16)

100.0,

0.05)

72.41,

0.05)

110/2

27.20

14.22

2.040,

0.05)

2.343,

0.05)

110/3

41.89

18.86

0.22

0.42

'9)

110/4

52.86

-!-

23.28

1.00

1.69

110/5

64.00

27.311

1.80

2.23

3.368,

0.05)

0.440,

0.05)

111/1

26.60

-:-

28.67

0.20

0.40

21.76,

0.05)

111/2

11.50

5.50

1.214,

0.05)

111/3

2.00

1.00

45.01,

0.05)

0.543,

0.05)

111/4

67.33

21.31

11.33

7.36 1

15.85,

0.05)

3.092,

0.05)

112/1

43.11

19.15

5.55

7.37

0.584,

0.05)

™

1.000,

0.05)

112/2

42.66

-)■

54.74

4.33

6.13 <

7.428,

0.05)

112/3

11.96,

0.05)

112/4

8.695,

0.05)

14.94,

0.05)

D = significant decrease, I = significant increase, N = no significant change
b = all insulator scores equivalent, test statistic could not be calculated
c = porcelain insulators, contamination could not be estimated

At 3 towers, decreases in contamination were observed for center insulator
strings (107/3, 109/1, 112/3). Contamination of both center and outer insulator
strings decreased at tower 112/4. All but one of those towers were not used by
ravens throughout the study period. Tower 109/1 was occupied on 2 evenings of
observation; no ravens roosted on the tower's C sections. There were increases
in contamination at 8 towers, 3 showed increases only on the center strings
(111/3, 111/4, 112/2), 3 on outermost strings only (107/4, 108/2, 111/1).
Increased contamination was shown on both center and outermost strings for 2
towers (108/5, 109/4). Towers showing increased contamination were occupied on
from to 6 nights observations. Towers showing overall increase were occupied
on 4 (108/5) and 6 (109/4) evenings. There was no significant difference
between contamination scores for May and November for the remaining 10 towers.
Evenings of observed use of those towers ranged from to 9.

Durability of Shields and Pegging

At the time of contamination estimation each tower was checked for broken
and missing pegs. Each tower had been fitted with 4 sets of 36 pegs, total of
144 per tower. Twenty-one towers had from 1 to 13 missing or broken pegs (mean
= 2.7).

Outages

No outages were reported by Pacific Power and Light during the 1989 study
period.

DISCUSSION

Maximum numbers of roosting ravens were lower in 1989 than in previous
years. The peak count of 2,103 ravens in 1984 is more than 3 times the highest
count for 1989. Coordinated roost watches were not conducted in 1989, so it
cannot be determined whether overall numbers are lower or whether alternate
roosts are being used more heavily.

Patterns of roost use over a period of months are similar for 1989 and
previous years. There is a general trend of increasing maximum count beginning
in early summer, with a decrease beginning in late September. The date of the
highest maximum count was slightly later in 1989 than in previous years. From
1984 to 1987, maximum numbers of roosting ravens were observed between mid-July
and mid-August. In 1989, the highest maximum count occurred on 5 September.

Without 1989 data for roost use prior to installation of shields and
pegging, assessment of raven response to treatment of roost towers is limited.
The birds' use of 2 towers on the day of treatment indicates that, as previously
observed, raven response to installation is negligible. Significantly more
birds roosted below the insulators during the time of installation and the 3
weeks following than in previous years or in the latter part of the 1989
season. This pattern changed after that time. Young and Engel (1988b)
documented a change in use of tower sections associated with installation of
shields and pegging at the Marsing Southwest Roost in 1988. Prior to and after
installation, ravens were more likely to roost above insulators, but in the time

immediately following installation, more birds roosted below the insulators.
This year's data seem to corroborate previous years' findings that ravens are
more likely to roost below insulators around the time of installation of shields
and pegging.

Changes in contamination over the study period show no clear patterns.
There was no strong correlation between changes in contamination levels and
frequency of use by roosting ravens. Of 4 towers showing decreases in
contamination, all but 1 were not observed to have been occupied during the
study period. Tower 109/1 showed decreased contamination of the center
insulator strings and was occupied twice during the observation period. Numbers
of ravens roosting on 109/1 were relatively low (mean = 18 ± 7). These results
indicate that over time, existing contamination may decrease on insulators that
are protected from further contamination. Not all towers that were not observed
to be occupied during the period showed decreasing contamination, however. One
(107/4) showed an increase in contamination of the outermost strings; another
(107/2) showed no change.

Increases in contamination were observed for 8 towers, 6 on either the
center or outermost strings of insulators, 2 on both. Included were towers
never observed occupied by roosting ravens (107/4), one used twice (111/3), 3
used 3 times (108/2, 111/4, 112/2), and one used 5 times (111/1). Of the 2
towers which showed increased contamination on both center and outer insulator
strings, one was occupied 4 times (108/5) and one 6 times (109/4). Towers
showing no change in contamination also varied in the number nights they were
observed occupied by roosting ravens, ranging from to 9.

The inconsistent patterns of levels of tower use and contamination changes
may indicate that towers could have been occupied during the study period, on
evenings when no observation was conducted. Further, varying light conditions
may be a complicating factor in estimating contamination. Estimations in May
were conducted on the days following installation of shields and pegging and
cloud cover ranged from 10% to 90% over the period of 4 days. End of the season
estimates were all done 1 November when cloud cover was 0%. This may have
affected consistency in scoring contamination of insulators. Observer
inconsistency is also a possible factor, though this was minimized by having the
same observer do all scoring.

While not quantified, levels of contamination on shields is important to
note. Throughout the study period, accumulations of waste on shields increased
visibly. In November when insulator contamination was estimated, extremely high
levels of accumulated waste were observed on shields on towers which were most
frequently used for roosting. This indicates that, if the shields had been
absent, insulator contamination might have been greater.

Visual counts of pegs at each tower showed an average of 2.7 pegs missing or
broken from an original total of 144. Pieces of pegging were collected below
several towers. All pieces had one unevenly broken end, suggesting that they
had been broken off. Most appeared from the ground to have broken off at the
point of attachment to the PVC pipe, though at 3 towers pegs were only partially
broken off. The 36-peg units of PVC appeared to have remained securely in place
throughout the study period. The materials and attachment methods used at
Initial Point in 1989 were developed in response to problems at the Wilson Creek

iMBur—"""™*"™

roost. There, plastic material used to attach lengths of pegging deteriorated,
and strips of pegging were lost (L. Young, pers commun.). Five months after
installation, no such problem had occurred at the Initial Point roost.

No problems were observed with shields during the study period. Attachments
appear to be secure, gaps between shields are minimal. Observations during
heavy winds showed no indication of instability of shields.

Patterns of raven occupancy of towers indicates that a sufficient number of
towers were treated at the Initial Point roost.

MANAGEMENT RECOMMENDATIONS
Results of 1989 monitoring give rise to the following recommendations:

1 . Communal raven roosts should continue to be monitored in order to
determine what other areas should be treated with shields and pegging. Monthly
coordinated roost observations in 1990 at the 5 major roosts identified by Young
and Engel (1988a) would permit assessment of use of roosts. During annual line
surveys conducted by Pacific Power and Light Company contamination should be
noted. Evening observations of possible new roosts may be done on an as
recommended basis. Occurrences of outages may also provide opportunities to
identify roost sites vulnerable to high levels of contamination by roosting
ravens as possible candidates for application of shields and pegging.

2. Pegging material used this year should be observed further. Its
ability to withstand cold temperatures, as well as its durability over a period
of months and years should be evaluated. The low level of loss and breakage
indicates that the materials and attachment methods used at Initial Point may be
more effective than other treatments.

3. Durability of shields over time should be checked. As of November, the
fiberglass and installation method both appeared to be in good condition. An
effort was made to minimize the widths of gaps between shield on individual
towers. Similar efforts should be made in future installation efforts.

4. Effectiveness of shields and pegging at preventing accumulation of
contamination, and, ultimately, power outages should be evaluated over a period
of years. This should include monitoring contamination at treated roosts during
annual line surveys and investigation of causes of line trips and outages.

5. Long-term evaluation of numbers of roosting ravens would provide
information about fluctuations in raven populations and roost site use over a
period of years. A solid base of data has been built since this study began in
1984, and with continued low-level monitoring, important information about
long-range population trends could be gained.

A<3KN0WLEDGMENTS

Leonard Young, Mike Kochert, and Karen Steenhof provided guidance for the
1989 monitoring effort. Support and logistical assistance from Mike Mulrooney,
Monte Garrett, and Jerry Roppe are greatly appreciated. Pat Haas was an

invaluable field assistant throughout the study period. Roost observations were
conducted by Eric Atkinson, Dave Beig, and Dawn McAnnis in the early part of the
season. This work was conducted under BLM contract No. ID-010-PH9-093.

LITERATURE CITED

Young, L. S. and K. A. Engel. 1988a. Implications of communal roosting by
coiranon ravens to operation and maintenance of Pacific Power and Light
Company's Malin to Midpoint 500-kV transmission line. Pacific Power and
Light Co., Portland, Oreg. 154pp.

Young, L. S. and K. A. Engel. 1988b. Implications of communal roosting by
common ravens to operation and maintenance of Pacific Power and Light
Company's Malin to Midpoint 500-kV transmission line. 1988 Annual Report.
Pacific Power and Light Co., Portland, Oreg. 13pp.

TITLE: Use of Nest Boxes by American Kestrels in Southwestern Idaho.

INVESTIGATORS: Karen Steenhof , BUI Birds of Prey Research

CO0PERAT0RS: Idaho Department of Fish and Game, Nongame Program
Idaho Department of Transportation

OBJECTIVES :

1. Determine kestrel occupancy rates and nesting success at boxes
erected in southwestern Idaho.

2. Mark nestling and adult kestrels to accumulate baseline data for
future studies on dispersal and population dynamics.

INTRODUCTION

This study began in 1985 when nest boxes were erected on abandoned power
poles in the Snake River Birds of Prey Area in an effort to learn more about
the food habits and breeding biology of American kestrels ( Falco sparverius )
in southwestern Idaho (Steenhof et al. 1985). The limited use of those boxes
prompted curiosity about factors that influence occupancy rates. In 1986,
the Idaho Fish and Game Department's Nongame program joined the study, and
the study area expanded to include agricultural and suburban habitat in the
Kuna area as well as Interstate 84 between Simco Road and Caldwell.

METHODS

In 1989, 69 boxes were available to kestrels for nesting in the
southwestern Idaho study area. Seven boxes had fallen, deteriorated, or been
vandalized since the nesting season of 1988. An eighth box was destroyed
after a kestrel had fledged from the box in 1989. Of the 69 boxes available,
30 were on trees within agricultural habitat in the vicinity of Kuna, Idaho
(hereafter referred to as the Kuna Route), 29 were on highway signs along
Interstate 84 between Simco Road and Caldwell (hereafter referred to as the
1-84 Route), and 10 were on abandoned power poles along the north rim of the
Snake River Canyon (hereafter referred to as the BOPA route).

Each box on the Kuna and 1-84 route was checked at least twice during the
nesting season. The BOPA boxes were checked only once. Nest checks began in
April and continued through July, with all boxes being checked at least once
between 15 May and 15 June. All boxes with viable nesting attempts were
re-visited as necessary to band young and ascertain nesting success. Nests
with young that were 75% feathered were considered successful. Investigators
checked boxes from a ladder or while standing on a vehicle. When checking
boxes, investigators inserted a hole stuff er into the nest box entrance to
trap any adult birds in the box. Adults and young were weighed with a 300 g
Pesola balance. Kestrel eggs that did not hatch after the normal period of
incubation were collected, stored in a refrigerator, and transferred to
Lowell McEwen (Colorado State University) for chemical analysis. Nests of
European starlings (Sturnus vulgaris ) were removed in an effort to keep all
boxes suitable for kestrel nesting.

RESULTS

Sixteen of the 69 boxes (23%) were used for nesting by kestrels in 1989.
In addition, kestrel pairs exhibited defensive behavior in the vicinity of 2
other boxes (Kuna #5 and Kuna #15) and were probably nesting in nearby tree
cavities. The 1989 box occupancy rate was lower than the rates observed in
1987 and 1988 (30% and 28%, respectively). Occupancy rates on the 1-84 and
Kuna routes were similar (28% and 27%). Eight of the occupied boxes were on
the Kuna route, and 8 were on the 1-84 route. No kestrels nested on the BOPA
route in 1989.

Three boxes were used by kestrels for the first time in 1989. Ten of the
boxes used by kestrels had been used in 1988, and 3 were last used in 1987.

Overall nesting success of 16 kestrel pairs in 1989 was 50%, the same as
that of 22 pairs studied in 1988 (Steenhof and Groves 1988). However,
success rates on the Kuna route in 1989 (75%) were significantly higher (X 2
= 4.0, P >0.05) than rates on the 1-84 route (25%). In the past 2 years,
success rates have been similar on the 2 routes (Steenhof and Groves 1987,
1988), and in 1986 success rates were higher on the 1-84 route (Steenhof et
al. 1986).

Clutch sizes ranged from 4 to 6 in 1989. This was the first year that a
clutch of 6 had been observed in this area. All 6 young fledged from 1 of
the 2 nests with 6 eggs. Overall, brood size at fledging ranged from 1 to 6
and averaged 3.38 (n = 8). Five of the 8 successful nests experienced some
loss of eggs or nestlings. Investigator disturbance was responsible for at
least 4 egg losses at 3 nests and 1 nestling loss at another nest. Two young
at Kuna Box #106 reached fledging age but did not survive because of physical
deformities. The male and female siblings (the only 2 young known to have
hatched from a clutch of 5) had severely splayed feet. The bone deformities
were first observed when the young were 14 days of age.

I banded 32 kestrels in 1989 (24 nestlings and 8 adults). Of the adults
banded, 4 were females, and 4 were males. On 18 May, I captured a banded
female in a box with 5 eggs. This bird had been banded as an adult on 22
January 1988 by Ted Swem within 8 km of Kuna. It was nesting in a box that
had never been used previously. All 4 males were captured in boxes where
males had never been previously caught. However, 3 of the 5 females were
from boxes where different females had been captured in either 1988 (n =2) or
1987 (n = 1). These findings, along with similar findings in 1988, suggest
that kestrel nest site fidelity is low in this study area.

Starling eggs or young were found in 29 of the 69 boxes, and house
sparrow ( Passer domesticus ) eggs were found in 1 box. Unidentified nest
material was found in an additional 11 boxes. Starling occupancy rates were
highest along the 1-84 route (62%). House sparrows also nested in the space
between the nest box and the highway sign at 1-84 #7. A deer mouse
( Peromyscus maniculatus ) was found in 1 of the BOPA boxes. Twelve boxes had
no confirmed use by any species during the nesting season. One box was used
by both kestrels and starlings; kestrels nested in a box that had starling
eggs on 20 April.

Weights of adult females ranged from 119 to 151 g and averaged 135 g (n :
5; s.d. = 13). The lightest female was captured while brooding small young;
the lightest incubating female weighed 127 g. Weights of adult males ranged
from 99 to 115 g and averaged 110 g (n = 4; s.d. = 8).

ACKNOWLEDGMENTS

This year's study would not have been possible without the assistance of
several volunteers who accompanied me on nest checks. I would like to thank
M. Luque, L. Langelier, M. McCoy, W. Heck, D. Parrish, A. Okamoto, B. Rose,
and B. Tripp, for their assistance. The National Geographic Society donated
a ladder to the study. We are especially indebted to the landowners in the
Kuna area who graciously allowed us to set up and check boxes on their
property.

LITERATURE CITED

Steenhof, K. , D. Ramirez, and J. Doremus. 1985. Use of nest boxes by

American kestrels in the Snake River Birds of Prey Area. Pages 138-144
in K. Steenhof and M.N. Kochert, eds. Snake River Birds of Prey Research
Pro j . Annu. Rep., Bur. Land Manage. Boise, Idaho.

Steenhof, K. , C. Groves, and D. Ramirez. 1986. Use of nest boxes by

American kestrels in Southwestern Idaho. Pages 130-132 in K. Steenhof
and M.N. Kochert, eds. Snake River Birds of Prey Research Proj. Annu.
Rep., Bur. Land Manage. Boise, Idaho.

Steenhof, K. , and C. Groves. 1987. Use of nest boxes by American kestrels
in Southwestern Idaho. Pages 94-97 in K. Steenhof, ed. Snake River Birds
of Prey Research Proj. Annu. Rep., Bur. Land Manage. Boise, Idaho.

Steenhof, K. , and C. Groves. 1988. Use of nest boxes by American kestrels
in Southwestern Idaho. Pages 53-55 in K. Steenhof, ed. Snake River Birds
of Prey Research Proj. Annu. Rep., Bur. Land Manage. Boise, Idaho.

TITLE: Feeding Ecology of the Barn Owl in the Snake River Birds of Prey
Area,

INVESTIGATOR: Carl D. Marti, Department of Zoology, Weber State College

COOPERATOR: Weber State College

OBJECTIVES :

1. Determine food habits and other food niche parameters of
nesting barn owls ( Tyto alba ) .

2. Determine food niche variation (a) among sites and (b)
among years.

3. Determine the barn owl's position in the raptor feeding
guild.

INTRODUCTION

Field studies for this project began in 1978 and have continued through
1989. All data reported here were obtained in the Snake River Birds of Prey
Area (SRBOPA) which is described in U.S. Dep. Inter. (1979). Analysis and
subsequent statistical treatments were done at Weber State College, Ogden,
Utah. Reports on previous years are also available (Marti 1979, 1981, 1982,
1983, 1984, 1985, 1986, 1987, 1988).

This report summarizes field activities and preliminary analysis for 1989
data. An index to trends in barn owl numbers in the SRBOPA is continued from
previous years .

ANNUAL REPORT

I made only one collecting visit (26-28 April) to the SRBOPA in 1989
because low barn owl densities did not warrant further visits. I collected two
samples of regurgitated pellets from 2 sites (Table 1). Two additional samples
of pellets were collected from Upper Lower Black Butte by Helen Ulmschneider.
A summary of the prey content of these pellets is in Table 2.

Dietary frequencies of major prey changed little from 1988. The dominant
prey species, Microtus montanus , declined slightly from the previous year.
Birds showed the greatest change in dietary composition from 1988 to 1989
increasing from 1.3 to 4.8%.

The apparent decline in number of nesting barn owls beginning in 1986
continued in 1989. Only two of 15 traditional nest sites were occupied and
nesting was not verified at either. Data from 1989 are compared with other
years in Table 3. Three traditional barn owl nest sites were occupied by other
species: prairie falcon ( Falco mexicanus ) , great horned owl ( Bubo virginianus )
and woodrat ( Neotoma sp. ) .

Table 1. Collection sites for barn owl food habits data in the
SRBOPA, 1989.

Lower Lower Black Butte
Upper Lower Black Butte

Table 2. Total prey identified for the barn owl in the SRBOPA,
1989.

Percent

Prey species

Number

number

MAMMALS

Sorex vagrans

0.2

Mus musculus

7.2

Peromyscus spp.

10.0

Reithrodontomys megalotis

4.5

Microtus montanus

307

48.0

Neotoma lepida

0.3

Perognathus parvus

9.4

Dipodomys ordii

7.0

Thomorays townsendii (juvenile)

8.3

unidentified leporid (neonate)

0.2

BIRDS

Sturnus vulgaris

0.2

unidentified icterid

0.5

unidentified medium bird

0.5

unidentified small bird

3.6

ARTHROPODS

unidentified scorpion 1 0-2

Totals 639 100

Table 3. Comparison of pellet collection data from nest sites of barn owls in
the SRBOPA.

Year of collection

1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989

Number of occupied

sit es 15 13 25 16 16 19 15 5 4 7 2

Number of pellet

collections 33 18 38 22 28 33 26 11 8 18 4

Total number of

prey identified 5426 3326 5359 5074 4142 7475 5559 996 1451 3552 639

Number annual

collecting visits 3332333223s

PLANS FOR 1990

Three trips to the SRBOPA are planned for the spring/ summer of 1990. The
goals for 1990 are to continue collecting food habits data for the analysis of
long-term predation trends by barn owls and to monitor trends in breeding numbers,

ACKMOWLEDGMENTS

I thank Michael Kochert, Karen Steenhof , John Doremus, Jeff Marks, Lenny Young
and all of the Snake River Birds of Prey Research staff for a variety of
assistance in carrying out this study. The assistance of Helen Ulmschneider in
making two collections of pellets is appreciated. The Bureau of Land Management
provided a vehicle for field use and living space in field camps. Weber State
College provided a Faculty Research Grant covering travel to the study area from
Ogden, Utah, and laboratory space and computer facilities for data analysis.

LITERATURE CITED

Marti, CD. 1979. Ecology of nesting barn owls in the Snake River Birds of
Prey Area. Pages 32-37 in Snake River Birds of Prey Res. Proj. Annu. Rep.
U.S. Dep. Inter., Bur. Land Manage., Boise, Idaho.

Marti, CD. 1981. Feeding ecology of the barn owl in the Snake River Birds

of Prey Area. Pages 37-40 in Snake River Birds of Prey Res. Proj. Annu. Rep.
U.S. Dep. Inter., Bur. Land Manage., Boise, Idaho.

Marti, CD. 1982. Feeding ecology of the barn owl in the Snake River Birds
of Prey Area. Pages 39-42 in Snake River Birds of Prey Res. Proj. Annu.
Rep. U.S. Dep. Inter., Bur. Land Manage., Boise, Idaho.

Marti, C D. 1983. Feeding ecology of the common barn owl in the Snake River

Birds of Prey Area. Pages 53-55 in Snake River Birds of Prey Res. Proj. Annu.
Rep. U.S. Dep. Inter., Bur. Land Manage., Boise, Idaho.

Marti, C D. 1984. Feeding ecology of the common barn owl in the Snake River
Birds of Prey Area. Pages 121-123 in Snake River Birds of Prey Res. Proj.
Annu Rep. U.S. Dep. Inter., Bur. Land Manage., Boise, Idaho.

Marti, C D. 1985. Feeding ecology of the common barn owl in the Snake River
Birds of Prey Area. Pages 131-134 in Snake River Birds of Prey Res. Proj.
Annu. Rep. U.S. Dep. Inter., Bur. Land Manage., Boise, Idaho.

Marti, C D. 1986. Feeding ecology of the common barn owl in the Snake River
Birds of Prey Area. Pages 123-126 in Snake River Birds of Prey Res. Proj.
Annu. Rep. U.S. Dep. Inter., Bur. Land Manage. , Boise Idaho.

Marti, C D. 1987. Feeding ecology of the common barn owl in the Snake River
Birds of Prey Area. Pages 101-105 in Snake River Birds of Prey Res. Proj.
Annu. Rep. U.S. Dep. Inter., Bur. Land Manage., Boise, Idaho.

Marti, C. D. 1988. Feeding ecology of the common barn owl in the Snake River

Birds of Prey Area. Pages 56-61 in Snake River Birds of Prey Res. Proj. Annu.
Rep. U.S. Dep. Inter., Bur. Land Manage., Boise, Idaho.

U.S. D.I. 1979. Snake River Birds of Prey Special Research Report. Bur. Land
Manage., Boise Distr., Boise, Idaho.

TITLE: Nest Box Use, Reproduction, Movement, Food Habits,

and Annual Cycle of Body Mass of Western Screech-Owls
in the Snake River Birds of Prey Area.

INVESTIGATORS: Jeffrey Marks, University of Montana, Missoula, Montana
John Doremus, BLM Bruneau Resource Area

OBJECTIVES: To determine patterns of nest box use, fidelity to site and

mate, reproductive success, and the annual cycle of body mass
in Western Screech-Owls (Otus kennicotti ) .

METHODS

Nest box locations available for roosting/reproduction of western screech
owls in 1989 are the same as in 1988 (Table 1). All locations were visited at
least once a month, January- April and November and December. Known breeding
locations were visited in May to band nestlings. During each visit (excluding
incubation and brood rearing) we collected prey remains and pellets, banded all
unbanded owls, and recorded the body mass of all owls encountered. The sex of
adult owls was determined during the breeding season by the presence or absence
of an incubation patch. Owls encountered between November and April could be
aged by the color of the underwing. Birds less than one year of age have a
buffy color on portions of the primaries, secondaries, and greater wing
coverts. Birds older than one year have a pinkish wash on their new primaries,
secondaries and greater wing coverts.

The nest boxes were not visited frequently enough to determine the number
of eggs laid or hatched or the number of young fledged.

RESULTS AND DISCUSSION

Western screech-owls were found at 19 locations during 1989. Ten of the
locations were not used during the breeding season. Three of the locations
were occupied during the breeding season but there was no evidence of a
breeding attempt. Owls attempted to breed at 6 sites. The nest attempt at
Rabbit Springs failed during incubation. The remaining 5 sites were successful
at hatching young. At banding 2 boxes contained 1 young, 1 contained 4 young
and 1 contained 5 young. At the other box the young were fledged, 2 were
captured and banded but there may have been other young in the vicinity that
were not found.

Twenty-three western screech-owls were banded during 1989 prior to 1
December. Thirteen nestlings, 2 HY, 6 SY, 1 ASY, and 1 AHY owl were banded.
One nestling was injured during banding and died from the injury. There have
been 171 western screech-owls (97 nestlings and 74 adult sized owls) banded in
the study area since 1980.

In 1989, a owl hatched in the study area was found as a breeding adult.
Owl #865-37590 was banded as a nestling at the Strike Reservoir site on 10 May
1987. This owl was found in a owl box at Crane Falls Sturgeon, 12.5 km from

its natal nest, on 24 January 1988. On 25 March 1989 it was in a owl box at
Crane Falls Sturgeon with a female owl. There were 3 eggs in the box, and the
female had an egg in her oviduct.

Owl #885-02427 was banded as a nestling on 15 May 1989 at the Delta South
site. It was found in a box at the Strike Camp Treeline II site on 24 November
1989. The natal and recovery sites are 3.36 km apart.

We obtained 51 weights from 23 adult size owls (28 female weights, 10 male
weights and 13 weights from owls of unknown sex) . Female weights ranged from
195-296 g. Male weights ranged from 186-212 g. The weights of owls of unknown
sex ranged from 170-256 g. The female that weighed 296 g was weighed during
laying and had an egg in her oviduct when weighed.

American kestrels ( Falco sparverius ) nested in boxes at 4 locations (Table
1) •

Table 1. Location names and occupancy status at roosting/nesting boxes
available to western screech-owls in the Snake River Birds of
Prey Area in 1989.

Location name Status Nesting success

Black Butte, Lower Lower roosting

Black Sands III roosting

Bruneau Boat Launch roosting

Bruneau Marsh vacant

Bruneau Marsh North vacant

Bruneau Marsh South roosting

Bruneau Marsh West roosting 3

Bruneau Rest Stop vacant b

Cabin Lower breeding unknown success

Cabin, Upper roosting"

Castle Creek Mouth North IV breeding unknown success

Cellar Hole breeding unknown success

Crane Falls Sturgeon breeding successful

Delta East vacant

Delta South breeding unknown success

Delta West roosting

Dunes Entrance Pond vacant

Flat Iron Marsh vacant

Harris vacant b

Little Valley School roosting

Lover idge Bridge vacant

Orcutt Treeline vacant

Rabbit Springs breeding unsuccessful

River Road roosting"

Strike Camp vacant

Strike Dam vacant

Strike Reservoir roosting 3

Strike Camp Treeline II roosting

Three Forks roosting

Tom Draw roosting 3

Two Forks vacant

3 Western screech-owls present during the nesting season but not nesting in

the boxes.
b American kestrels nesting in box.

TITLE: Post-nesting Ecology of Long-eared Owls in the Snake River Birds
of Prey Area, Idaho.

INVESTIGATOR: Helen Ulmschneider, Raptor Research Center, Boise State
University

COOPERATOR: Raptor Research Center, Boise State University

OBJECTIVES:

1. To determine the timing, distances, and directions of movements of
long-eared owls when they leave their nest groves.

2. To compare prey deliveries by male and female parents during the
post-fledging period.

3. To determine whether adult and juvenile hunters select different prey.

INTRODUCTION

This paper summarizes results from the second year of a 2-year master's
study at Boise State University. The basic objective of the study this year was
to discover the distances and directions of the owls' initial movements when
they left the desert after nesting. A second objective was to determine
relative numbers of prey deliveries by male and female parents during the time
from branching until the females left. In my first season I discovered that
females deserted their families before the young reached independence.

A third objective was to determine whether juvenile hunters selected
Perognathus prey in greater proportion than did adult hunters. This was
suggested by Marks (1984) from data collected in the SRBOPA in 1980-81. Lastly,
I collected data on roosting heights of adult and juvenile owls.

METHODS

Trapping . I caught adult owls using a mist net with a plastic great horned
owl statue as a lure. I placed the net near the nest grove, in an open area or
on a rise where the statue would be silhouetted to the long-eared owls. The
owls reacted to the statue by dive-bombing it. I trapped at dusk or night, so
the owls could not see the net. Position of the set was critical: if the
statue was not silhouetted, the long-eared owls did not react. Changing the
position to silhouette the statue caused the long-eared owls to immediately
begin dive-bombing and "yowling" at the statue.

Radio- tagging ■ I used backpack harnesses to attach 2-stage transmitters
(Biotrack, Wareham, Dorset, England) to the owls. These transmitters had at
least twice the range of the single-stage ones I used last year, though they had
a slightly shorter life of 2.5-3.5 months. I tagged 13 owls from 4 nests,
including 4 adult females, 4 adult males, and 5 juveniles (Table 1 and Fig. 1).

The transmitter plus backpack harness weighed 10 g, which averaged <4% of
an owl's body weight. For the harness, I used either 6-mm elastic or flat
teflon tubing. I recaptured 2 owls after tagging them, to check fit, and found

Table 1. Nest sites and radio- tagged Long-eared Owls in the Snake River Birds
of Prey Area, Idaho, 1988 & 1989.

Year

Site

Radio-Tagged

Total

male

female

juvenile

Fledged

1988

SDM-G

—

NP-S

51 SDR 1 1

2
Total

1989

NP-NW

SDR-E

ETS-U

ETS-L
Total

Abbreviations :

SDM-G Strike Dam Marsh Green

BA Bruneau Arm

NP-S Nicholson Pond South

NP-NW Nicholson Pond Northwest

SDR Strike Dam Rd

SDR-E Strike Dam Rd East

ETS Emigrant Trail South, upper and lower

en
o

o
*

i i

i — j *

5 km

NP*™

1988

1989

BA Bruneau Arm

ETS Emigrant Trail South, (upper, lower)
NP Nicholson Pond, (south, northwest)
SDM Strike Dam Marsh
SDR Strike Dam Rd, (east, west)

ETS ^r

Figure 1: Locations of study nests of Long-eared Owls for
1988 and 1989 in the Snake River Birds of Prey
Area, Idaho.

that I had been making the back length too short and the owls were developing a
crusting on the harness where it rubbed behind the wing. When fitting a harness
it is important to realize that the fit seems much looser while the birds are
lying on their backs than when they are upright in a natural posture.

I determined that the harness should be about 330 mm long, allowing 15 mm
for overlap of the ends to form a continuous loop. The 2 side loops should each
be 145 mm, allowing 50 mm length in front of the wing from the radio to the
breast sewing, and 95 mm at the back. This varied somewhat depending on the
size of the owl, however, I found this to be a good proportion for an even fit
front to back. I needed 0.5-0.75 hr. to sew the harness on an owl by this
method.

Alternatively, I tried sewing several harnesses using 4 separate ends
instead of 2 loops, sewing them together at the breast. The purpose of doing it
this way was to allow the harness to drop off when the transmitter had died and
the stitching (cotton) had rotted. However, it was much more difficult to fit,
and required over an hour to sew.

Aerial tracking. I hired the State Bureau of Aeronautics Cessna 182 to
search for owls. The plane was equipped with 2 H-antennas mounted on the wing
struts, pointing sideways and down at a 30 degree angle. I flew lines about
24-32 km apart, paralleling the lay of the front range. When I found an owl,
location was determined using both maps and a Loran, with accuracy within 2-4
km.

Night watches. I conducted night watches at each site from dusk to
1200-0100, 1-2 times a week. The purpose of these was to track the development
of the young and to compare prey deliveries by male and female parents. During
a nightwatch I recorded flying, visits by adults, and changes in numbers and
intensity of young food-begging. I determined when owls were flying by
fluctuations in the volume and directions of radio signals.

I could not usually see prey deliveries, so I had to determine what
happened by listening. I defined a prey delivery as a visit to the young by an
adult, accompanied by a change in number or intensity of young food begging
either before or after the visit.

Food Habits. I collected pellets throughout the season, recording roost
site and what owl(s) used it, if known. Pellets were analyzed by soaking them
in a 7% solution of NaOH to dissolve the hair, and using skulls, dentaries, or
( for larger species ) , long bones to identify species and numbers of prey items .
Within a sample of pellets, I counted the number of a prey species as the number
of skulls, right or left jaws, or (for Dipodomys , Thomomys , and leporids) right
or left femurs, humeri, or tibia-fibulas, whichever was greatest.

Roost sites. I visited nest groves every 1-2 days, and tried to spot
roosting owls without flushing them. I sketched the position of owls, and
returned later to take measurements. The position of whitewash and pellets also
helped relocate roost sites. The owls reuse the same few trees over and over,
causing much auto-replication within the roost site data I collected in 1988.
Because of indications from the first season's data that newly branched owlets
roost higher than adults and older juveniles, roost height was the variable I
measured this year.

RESULTS

The results from this study are being written as a Master's Thesis at Boise
State University. I briefly summarize them here.

Prey Delivery Rates. Males delivered. 2.45 as many prey items as females,
during the period from branching to the departure of the females. This factor
varied among sites, however (Table 2, Fig. 2). The males then continued to
feed the young for a mode of 19 more days, until both they and the young left
the desert (Fig. 3). Prey delivery rates for the same owls on different nights,
and among different owls varied greatly (Fig. 3). Within each pair of owls, the
male had the greatest overall prey delivery rate (Table 2), although on 2
occasions the female surpassed the male in a particular night (Fig. 3).

Timing of Movements. In both years, all the females left their nest groves
before the males and juveniles, when the young were about 8 weeks old (Fig. 4).
However, 2 females left at an earlier stage, when the young were about 6.5 weeks
old; both these females were from late nests, about 2 weeks behind the other
nests. Most of the males and juveniles then left when the young were about
10-11 weeks old.

Local movements. Most of the owls roosted in their nest groves until
leaving the desert; however there were some local movements (Figs. 5 and 6).
Five of the successful females from both years started to roost away from their
nest groves just prior to departing the desert. Except for 1 owl, when the
females began roosting away from their young, either in separate clumps within
the same grove, or in different groves, it was a sign they would soon desert
their families.

Three of the 5 successful radio-tagged males roosted in their nest groves
until leaving the desert. Two males roosted up to 2.7 km away from the nest
grove on scattered days, but continued to feed their young.

At 2 sites in 1989, I observed mixing of broods of young before they left
the desert . At ETS ( see Table 1 for abbreviations ) , where there were 2 nests
6 m apart, the broods roosted together from the time the younger brood branched
until the older brood left the desert. At night, both broods food-begged
together. All 4 parents fed the young, however, I could not tell which parents
fed which young.

The second case of brood mixing occurred after the SDR brood reached
independence. The 3 young from SDR moved to Nicholson Pond N (1.5 km away) and
roosted for about 2 weeks with the younger brood there.

Long-range Movements. In 1988, I did not find any owls once they left the
desert. I searched 13,000 km by truck and 18 hrs from the air, in SW Idaho
(Fig. 7).

In 1989, I found 7 of the 13 owls after they left the desert: 3 females, 2
males, and 2 juveniles (Fig. 8). All of the these owls moved northward into the
mountains. Males, females, and juveniles from the same family went to widely
separated areas. Directions ranged from 342 to 64 degrees, with 4 clustering
between and 14 degrees east of north (Fig. 9). My initial locations of owls,

Table 2. Summary of night watches and prey deliveries at 4 nest sites of

long-eared owls in the Snake River Birds of Prey Area, Idaho, 1989

NP-NW

SDR-E

ETS U

ETS L

TOTAL

# Nightwatches

# Hrs obs m & f

# Hrs obs m only

10
12

7
9.5
4.5

9.5

10
12

27*
33.5*
26.5*

# PD m (f present)

# PD m (alone)

# PD f

16
4

10
4
7

54
33
22

Av, PD/hr m
Av. PD/hr f

1.9
.2

1.4
.7

1.2
.6

1.3
.5

Ratio m/f PD

(f present)

9.5

2.3

1.5

1.4

2.5

* Total reflects the fact that at ETS U & L, nightwatches were done
simultaneously .

m - male f - female PD - prey delivery

CX>.

NP-NW

SDR-E

ETS1

ETS2

CT
LU

>-
LU
CC
a.

_i
<
h-
O

MALE
FEMALE

NP-NW

SDR-E

ETS1

ETS2

Figure 2: Average prey delivery rates and total number of prey
deliveries by male and female parents during the post-
fledging period. Data from night watches (dusk to 12 00-
0100) at 4 nest sites of Long-eared Owls in the Snake
River Birds of Prey Area, Idaho, 1989.

^^■■[HK:'

EMGRANT TRAIL SOUTH 1

15 35

AGE OF YOUNG IN DAYS

2.8
2.4
2-
1.6-

1.2
0.8
0.4

EMIGRANT TRAILS 2

NICHOLSON FOND

tr
x
tr

a.
in

tr
\n

D
>-

AGE OF YOUNG IN DAYS

STRIKE DAM RD E

■ MALE
+ FEMALE

Figure 3: Prey delivery rates for male and female parent Long-
eared Owls during the post-fledging period, in the Snake
River Birds of Prey Area, Idaho, 1989.

nannm^

(f)

10-|
9-
8-
7-
6-
5-
4-
3-
2-
1 -

o-J-

1988

■ Adult females
I I Adult males
^ Juveniles

, Spans period

of departure

9 10 11 12
Weeks of Age

14 15

8n
7-
6-
5-
4-
3-
2-
1 -
0--

S §

I I

1989

■ Aduit females
I I Adult males
£§§ Juveniles

10 11 12 13 14 15

Weeks of Age

Figure 4: Timing of departures of adult male, female, and juvenile
Long-eared Owls from their nest groves in the Snake
River Birds of Prey Area, Idaho, 1988 and 1989.

i i j 1 1 1

5 km

2 wks

1-2 wks ! day

^ 1988 nest

^L. 1989 nest
9 adult feraa

O 1988 roost

■ 1989 roost

le adult male

^see fig 6
for these nests

Bruneau

Figure 5: Local movements of Long-eared Owls to day roosts away
from their nest groves in the Snake River Birds of Prey
Area, Idaho, 1988 and 1989. Length of days a roost was
used is indicated near it. Arrows indicate direction of
movement.

yC nest
u upper

I lower

o ad. male
j juvenile

X roost used 1-4 times
▲ roost used >4 times
* indicates last roost

used before leaving
^ ad. female

1 KM

<}+\ f'

A*
J

J s :

/ '■

x* x

J ' nests

Figure 6: Local movements of Long-eared Owls away from their

nest grove at Emigrant Trail South, in the Snake River
Birds of Prey Area, Idaho, 1989. These movements were
made after nesting and before leaving the desert.

Figure 7: Aerial search routed for radio-tagged Long-eared Owls
after they disappeared from their nest groves in the'
Snake River Birds of Prey Area, Idaho, in 1988

^SSBnHHBHHB^HKME

Nevada

Figure 8: Locations (•) of 7 radio-tagged Long-eared Owls after
they left their nest groves (•) in the Snake River Birds
of Prey Area, Idaho. Number of days owls were tracKea
after leaving is indicated near last locations. Area
searched is approximately that of the map.

f: adult female m: adult male j: juvenile

Figure 9: Azimuths (in degrees) of initial locations of 7 Long-
eared Owls after they left the Snake River Birds of Prey
Area, Idaho, in 1989.
f: adult female m: adult male j: juvenile

usually within a week of their departure, ranged from 72-96 km from their nest
groves (Table 3). The farthest subsequent location was 125 km from the nest
grove .

I conducted the first aerial search 2-8 days after the 4 females left, then
once a week for 4 more weeks, then 1 last time after a delay of 4 weeks (Table
3). I located the 2 juveniles only once each after they left the desert. I
found 1 of the ETS 2 owlets 12 days after it disappeared from the desert, on an
open mountain ridge 80 km north. I found the NP-NW owlet 3 days after it left,
in a box-elder maple ( Acer negundo ) in a residential backyard in Boise. I
followed this owl by truck that evening. It flew due north, going over 30 km in
less than 1 hr. When last heard it was veering NE into forested mountains.

I was able to track the 5 adults for 21-38 days after they left the desert,
relocating individuals from 3-8 times by air and ground. After their initial
movements into the mountains, the adults moved shorter distances, from 0-30 km,
and settled in 1 general area (Fig. 8). My last contact with them was on 14
July. I flew again on 10 August, going to the last locations of each owl and
also searching farther north and east, without finding any signals. I think the
radios were dead, as they were supposed to last 2.5-3.5 months, and it had been
3+ months for most of them.

Food Habits. The pellets from 1989 are still being analyzed. I present
here the data from 1988, representing 201 prey items. The complete list of prey
is presented in Table 4. Peromyscus (33%), Perognathus (30%), Dipodomys (14%),
Microtus (9%), Thomomys (6%), Reithrodontomys (3.4%), and Mus (3.1%) were the
major prey species by frequency. Young leporids, Neotoma , and birds constituted
<1% each of the diet. Fig. 10 depicts the overall percent frequency of the
major prey species.

Fig. 11 depicts the changes in percent of prey species in the diet from
April thro ugh July. The percent of Peromyscus declined over this period, while
that of Perognathus increased. This varied among sites, however (Fig. 12). My
pellet samples were not equally distributed among the different sites over time,
and therefore the changes in percentages of prey species may be affected by the
distribution of samples. However, the high % of Perognathus in the diet during
late summer holds for 4 of the 6 sites.

Marks (1984) also found an increase in Perognathus during the post-fledging
period (June and July), in conjunction with a decrease in Dipodomys . He
suggested that newly fledged hunters may choose smaller prey. However, I found
that a lone adult male which stayed at Crane Falls Sturgeon from June-Sept
consumed 80-90% Perognathus (Fig. 12). Also, during nightwatches , fledged
owlets seemed to do little flying (hunting), and food begged for most of the
night until the time they disappeared. Therefore, most of the prey eaten by
post-fledged owlets before they leave the desert is probably captured by their
parents. This suggests that the increase in Perognathus in June and July may be
owing to changes in population or availability of the prey species rather than
prey choice by the owls.

Table 3,

Pio^ 8110 ^ ^ km ) long-eared owls moved from last location, in southwest Idaho,
1989. Transmitters were probably dead on 10 August.

June

July

August
10

NEST

SEX/ Date left

Age

nest area

air

grnd

air

grnd

air

grnd

air

grnd

air

grnd

air

ETS U

07 June

;•;

SDR-E

10 June

;;:

NP-NW

12 June

SDR-E

21 June

;':

ETS U

25 June

32?

>:;

;■;

ETS U

25 June

■>

NP-NW

12 July

73,32+

- not searched for

* searched for but not found

? location approximate

Table 4,

Numbers and percents of prey species (by frequency) in the diet of
long-eared owls in the Snake River Birds of Prey Area, Idaho, during
April -November, 1989. > "s

PEMA

PEPA

DIOR

MIMO

THTO

REME

MUMU

LEP

BIRD

NELE

TOTAL
%

APR
NO.

196
69
51

S5
4

18
2
9
5
2

46.6

16.4

12.1

15.4

4.3

0.5

2.1

1.2

0.5

421
21

MAY
NO.

241

119

581
28.9

41.5

20.5

13.8

5.9

3.3

2.2

0.9

0.5

JUN
NO.

185

282

133

23.3
35.6
16.8

4
5
3
5.9
0.9

793
39.4

PEMA
MIMO
MUMU

Peromyscus

Microtus

Mus

JUL
NO.

10
72
6
11
3
3
1

107
5.3

9.3

67.3

5.6

10.3

2.8

0.9

AUG-NOV
NO.

30
61

1
5

2
i

27.5

8.3

0.9

4.6

1.8

0.9

PEPA = Perognathus DIOR

THTO = Thomomys (neonates) HEME
LEP = Leporid (neonates) NELE

109

5.4

TOTAL

NO.

662

603

279

172

125

2011

Dipodomys
Reithrodontomys
Neotoma (neonates)

32.9
30

13.9
8.6
6.2
3.4
3.1
1
0.5
0.3

LJJ
D

lu
U

LJJ
OL

RAB NEO

MUS BIRD

PM Peromyscus
MM Microtus
MUS Mus

PP Perognathus DO
TT neonate Thomomys RM
RAB neonate leporids NEO

Dipodomys
Reithrodontomys
neonate Neotoma

Figure 10 :

1988: Percent frequency of the major prey species in
the diet of Long-eared Owls from April to September in
the Snake River Birds of Prey Area, Idaho.

LU
CL

DIPODOMYS

MICROTUS

PERCMYSCUS

PERCDGNATHUS

THOMOMYS

REITHRODONTOMYS

APRIL

MAY

JUNE

JULY

Figure 11:

1988: Percent frequency of the major prey species in
the diet of Long-eared Owls during the nesting and
post-fledging months of April-July, Snake River Birds
of Prey Area, Idaho.

80-
70

60
50
40
30
20
10

NICHOLSON POND

100

BQ<

30-
20

* o
o

STRIKE DAM RD

MAY

JUNE

APR

MAY

JUNE

JULY

BRUNEAUAHM

% 80!
70
60

SO
40
30

2D
10

J^_

CASTLE MOUTH NORTH

WAY

JUNE

JULY

APR

■

MAY

JUNE

% 60-

STRIKE DAM MARSH

96100

30-
80-
70
60 -I

50
40
30
20
10

CRANE FALLS STURGEON

rvq aiiaM

APR

MAY

JUNE

■:■■

JULY

AUG

SEPT

Figure 12: 1988: Percent frequency of the major

prey species in the diet of Long-eared
Owls, by nest site and month, in the
Snake River Birds of Prey Area, Idaho.

H DIPODCMYS
§1 MICROTUS

PEROGNATHUS
PEROMYSCUS

□ THOMGMYS

LITERATURE CITED

Marks, J. S. 1984. Feeding ecology of breeding long-eared owls in southwestern
Idaho. Can J. Zool. P. 1525-1533.

'.., :...'■■■

TITTLE:

Home Range and Habitat Use of Ferruginous Hawks in the Snake
River Birds of Prey Area.

INVESTIGATORS: Dawn McAnnis Gerhardt, Raptor Research Center, Boise

State University
Marc Bechard, Raptor Research Center, Boise State

University
Richard Gerhardt, Raptor Research Center, Boise State

University
COOPERATOR: Raptor Research Center, Boise State University

OBJECTIVES:

1. To determine home ranges of 4 nesting male ferruginous hawks.

2. To determine habitat use within home ranges.

3. To determine effects of range fires on home range size and habitat
use.

INTRODUCTION

This is a continuation of a 2-year study begun in 1988 to determine the
effect burning of native shrub steppe habitat has on habitat use of nesting
ferruginous hawks.

METHODS

Four adult male ferruginous hawks were trapped near their nests between
15 and 26 May 1989, and equipped with radio-transmitters. Males at the
following nests were trapped: PP&L 104/4, PP&L 102/1, PP&L 96/3, and Indian
Creek .

The observation schedule was divided into 3 blocks. Two different males
were followed each day during 2 of 3 possible observation periods. Block A
began 30 min before to 1 hr after sunrise and ended 5-5.5 hrs later. Block B
began between 1100 and 1200 hrs MST (between 30 and 60 min after Block A
ended) and lasted 3 hrs. Block C began at 1500 to 1700 hrs MST and
terminated 4.5-5.5 hrs later. This allowed each bird to be observed for
12-14 hrs every 7 days.

Locations, habitat use and activity information were recorded every 15
min or each time the birds perched or made a foraging strike. Observations
on 2 radioed males were discontinued when they molted the tail feathers
holding their radio transmitters before fledging began. Observations on the
other 2 males continued until after fledging.

Minimum convex polygon and harmonic mean estimations of home range area
were computed for each bird using the computer program Home Range (Samuel et
al 1985). A chi-square goodness-of-fit test was used to determine if

habitat composition of actual home ranges differed from that which was
available in circles drawn around the nests using as the radius the farthest
point the bird was observed from the nest. A second Chi-square test and
Bonferroni-Z confidence intervals were used to determine if habitats found
within the home ranges were used in proportion to their availability in the
home ranges (Neu et al. 1974, Byers et al. 1984).

RESULTS

The results of this study will be included in a Master's thesis to be
completed in 1990. The following is a summary of the results for 1989:

1. Home range size for the 4 ferruginous hawks monitored in 1989
averaged 5 . 6 km 2 ( range 4 . 8-6 . 4 km 2 ) .

2. The range of PP&L 104/4 male consisted of 86% grasses, 13% native
shrub and 1% shrub/grass mosaic. Of 265 total locations obtained for this
bird, 91% were in grasses, 9% in native shrub and 0.4% in mosaic. Habitats
were used in proportion to their availability (X 2 = 5.8, 0.10< P <0.05).

The range of PP&L 102/1 male consisted of 69% native shrub, 20%
grasses, 8% mosaic, and 4% other (an old quarry) . A total of 236 locations
were obtained for this bird. Forty percent of these locations were in
mosaic, 37% in grasses and 24% in native shrub. This bird used mosaic and
grasses more than expected and native shrub less than expected (X 2 = 390.1,
P <0.0001).

The range of PP&L 96/3 male consisted of 67% mosaic, 30% grasses and
3% native shrub. Of 155 total locations, 64% were in grasses, 32% in mosaic,
and 4% in native shrub. This bird used grasses more than expected, mosaic
less than expected, and native shrub in proportion to its availability in the
home range (X 2 = 89.0, P <0.0001).

The range of Indian Creek male consisted of 93% grasses and 7%
mosaic. Of 153 locations obtained for this bird, 98% were in grasses and, 2%
in mosaic. This bird used mosaic less than expected and grasses in
proportion to availability (X 2 = 6.9, 0.0K P <0.25).

3. There did not seem to be any correlation between amount of home
range burned in the past and fledging success of ferruginous hawks in this
study (Table 1).

LITERATURE CITED

Byers, C. R. , R. K. Steinhorst, and P. R Krausraan. 1984. Clarification of a
technique for analysis of utilization-availability data. J. Wildl.
Manage. 48:1050-1053.

Neu, C. W. , C. R. Byers, and J. M Peek. 1974. A technique for analysis of
utilization-availability data. J. Wildl. Manage. 38:541-545.

Table 1. Percent of past-burned habitats in home ranges of ferruginous
hawks (now represented by grasses) and number of young fledged.

# Young
Bird % Burned Fledged

PPL 104/4 86% 2

PPL 102/1 20% 4

PPL 96/3 30% 3

Indian Creek 93% 4

Samuel, M. D. , D. J. Pierce, E. 0. Garton, L. J. Nelson, and K. R. Dixon.

1985. User's manual for Program Home Range, Second ed. For., Wildl.,
and Range Exp. Sta. , Tech. Rep. 15. Univ. Idaho, Moscow. 70pp.

"* iiiiiihii

TITLE: Morphometric and Genetic Variation in Ferruginous Hawk
Populations .

INVESTIGATORS: Dale K. Ward, Dep. Biology, Boise State University
Marc Bechard, Dep. Biology, Boise State University

COOPERATOR: Boise State University

A study of genetic and morphometric variation in Ferruginous Hawk ( Buteo
regalis) populations was initiated in cooperation with Boise State University's
Master's Program in Raptor Biology.

Three ferruginous hawk nests on Bureau of Land Management property were
sampled on 13 and 14 June 1989. These nests were located to the south of
Indian Creek Reservoir, at Bryans Run, Slaters Flat and Indian Creek. Each
nest contained 4 nestlings, each of approximately 40 days old or older
(Moritsch 1985).

A 1.5 ml. blood sample was withdrawn from the branchial vein of each
nestling for genetic analysis. The nestlings were banded with United States
Fish and Wildlife Service bands, weighed, and measurements made of the
following lengths: wing chord, footpad, tarsal, hallux, middle tail feather
and primary feather #4 (Table 1).

One sample from each nest was sent to Sheila Schmutz at the Department of
Biology, University of Saskatchewan, Saskatoon, Canada S7N OWO for genetic
analysis utilizing DNA fingerprinting. Results of this analysis are still
pending. The remainder of the blood samples are stored with Dr. Marc J.
Bechard at the Biology Department, Boise State University, Boise, Idaho
83725. Further inquiries concerning this project should be directed to Dr.
Bechard.

LITERATURE CITED

Moritsch, Marc Q. 1985. Photographic guide for aging nestling ferruginous
hawks. Dep. Inter., Bur. Land Manage. Boise, Idaho 21pp.

Table 1. Characteristics of ferruginous hawks sampled in 1989.

(a)

Bird!

Date

Nest

Age

Band

Weight

Footpad

Tarsus

Talon

Beak

Tail

Primary 4

Chord

(days)

Number

(kg)

(m)

(mm)

flat (mm)

13 June 89

Bryans Run

87724318

1.15

75.1

82.3

24.3

22.3

150

230

324

327

13 June 89

8ryans Run

87724319

73.4

86.5

23.7

22.2

158

241

317

336

13 June 89

Bryans Run

87724320

62.0

83.7

23.7

21.7

141

214

314

318

13 June 89

Bryans Run

120723457

73.3

86.8

28.3

26.8

146

231

315

321

13 June 89

Slaters Flat

120723458

83.9

90.7

28.6

25.5

152

215

305

322

13 June 89

Slaters Flat

120723459

78.7

94.5

29.6

25.4

142

231

314

319

13 June 89

Slaters Flat

120723460

80.4

85.3

27.3

25.3

141

205

288

302

14 June 89

Slaters Flat

87724321

72.3

84.6

24.4

23.2

151

208

300

301

14 June 89

Indian Creek

87724322

75.8

83.6

24.3

23.2

136

197

274

290

14 June 89

Indian Creek

120723461

88.1

90.1

26.5

24.6

121

192

291

292

14 June 89

Indian Creek

120723462

86.9

85.1

26.9

23.8

120

162

262

263

14 June 89

Indian Creek

87724323

66.9

79.7

24.0

20.9

131

125

283

284

Average

1.07

61.1

68.8

20.8

19.0

113

163

239

245

Standard Deviation

0.60

32.4

35.8

10.9

10.0

125

128

TITLE:

INVESTIGATORS:

Testing Performance of Wildlife Telemetry Equipment in the
Orchard Training Area

Mark R. Fuller, U.S. Fish and Wildlife Service, Patuxent

Wildlife Research Center
Brett A. Hoover, U.S. Fish and Wildlife Service, Patuxent

Wildlife Research Center
Susan S. Klugman, U.S. Fish and Wildlife Service, Patuxent

Wildlife Research Center
Linda S. Schueck, U.S. Army Chemical Research, Development and

Engineering Center, Aberdeen Proving Ground
Kimberly Titus, Dep. of Fisheries and Wildlife Science,

Virginia Polytechnic Inst, and State University
Diane S. Chu, U.S. Fish and Wildlife Service, Patuxent Wildlife

Research Center

C00PERATORS:

OBJECTIVE:

Idaho Army National Guard

U.S. Army Chemical Research, Development and Engineering Center
U.S. Fish and Wildlife Serv. , Patuxent Wildlife Research Center
Virginia Polytechnic Institute and State University

Test reception range, accuracy, and function of telemetry
equipment for potential application to studies of raptors in the
Snake River Birds of Prey Area.

INTRODUCTION

Research is proposed for learning about the movements, habitat selection,
and general behavior of raptors using the Idaho Army National Guard Orchard
Training Area (OTA) and adjacent area within the Snake River Birds of Prey
Area. Radio telemetry is a potential research technique because of the
limitations to repeated direct observation of the rapid-moving, wide-ranging
raptors and to researchers' mobility in the study area. Furthermore, 1 goal of
the research is to study raptor movements during the military training that
takes place on the OTA. During periods when tank, artillery, or helicopter
weapons training is underway, about 60% of the OTA is closed to ground and air
access. Therefore, "location" of raptors must be accomplished by telemetry
conducted from the Range Road (perimeter of the impact zone) or beyond. This
report summarizes some of the telemetry test results obtained to date.
Additional analyses may provide some different interpretations, and allow for
more distinction among the effects of some variables.

METHODS

We assessed the feasibility of a telemetry study on the OTA using 2
criteria, (1) evaluation of bearing error defined to be the known bearing minus
the observed bearing (Chu et al. 1989), and (2) evaluation of location estimates
based on distance of the point estimate from known transmitter location, area of
the confidence ellipse associated with the point estimate, and the coverage that
the confidence ellipse gave (i.e., whether the transmitter was within the
ellipse) .

We obtained bearings towards transmitters placed 0.3m and 1.5m
above-ground at USGS benchmarks on the OTA. These transmitter locations were
unknown to the observers obtaining the bearings. The first (summer) test period
started on 23 August and ended 28 August. The second (fall) test period was 31
October and 1 November. We used 3 sizes of 164-mHz transmitters (in order of
increasing size these were "merlin", "hawk", and "eagle" transmitters) and 3
antenna positions: hand-held, and 3.05-m and 6.1-m towers with metal or PVC
plastic masts. Towers were erected on buttes accessible from Range Road on the
OTA. The antennas were a 4-element Yagi configuration, and were connected to
scanning receivers: Advanced Telemetry Systems Model 2000 (ATS) and LOTEK Model
SRX-400 (use of product names does not imply endorsement) .

During the summer observations, we used 3.05-m and 6.1-m towers constructed
from metal masts, and ATS receivers only. During the fall observations, merlin
transmitters were not used, and we used 3.05-m towers equipped with upper-masts
constructed of PVC pipe rather than metal. For the fall observations, we used
both ATS and LOTEK receivers. The LOTEK receiver contained a built-in meter
that digitally displayed signal strength. Not all signals were powerful enough
to be recorded on the meter; therefore the receivers were classified into 3
groups: 1) ATS receiver (no signal strength indicator), 2) LOTEK receiver,
signal not strong enough to register on meter, and 3) LOTEK receiver, signal
registered on meter.

RESULTS
Analysis of Bearing Error

Five observers obtained 628 bearings during the summer. The overall
bearing error was -5.83 degrees (sd = 17.56 degrees). Signals were received
from up to 30 km away, although no merlin transmitter signal was received at
distances greater than 10 km. The height of the transmitter did not have any
effect on bearing error.

Four factors: observer, antenna position (hand-held, 3.05 or 6.1-m towers),
distance class (in 5-km intervals), and type of transmitter (merlin, eagle, and
hawk) were used to assess variables that influence bearing error. All factors
and 2-way interactions were significant (P _< 0.0433). Almost all of the
observers were significantly different from one another in mean bearing error.
Bearings obtained using the hand-held antennas proved significantly less
accurate than bearings from the 3.05-m towers. There was no difference between
the 3.05-m and 6.1-m towers or between the hand-held and 6.1-m tower. Although
distance class was significant, no single distance class emerged from the
analysis as being the best. Bearings obtained for the merlin transmitters were
significantly less accurate than for both the hawk and eagle transmitters.
Bearings taken on the eagle and hawk transmitters were not significantly
different in accuracy.

Of the interactions, the following results were most relevant to our
assessment of error. Mean error for merlin transmitters detected with hand-held
antennas and 3.05-m tower systems was -25.0 and -12.3 degrees, respectively.
For the merlin transmitters detected with 6.1-m towers, the mean error was -3.0
degrees. Distance class and transmitter type interactions also were revealing.
For the 5-km and 10-km distance categories in which merlin transmitters were
detected, mean error was -15.1 and -15.3 degrees, respectively.

The overall mean error for the fall period was -3.37 degrees (s.d. = 15.06)
from a total of 278 obtained bearings. Antenna method, observer- transmitter
type, method-distance class, and distance class-receiver class interactions were
significant (P < 0.0813). The 3.05-m tower data were more accurate than the
data for the hand-held antenna using the LOTEK (both categories; we gathered no
fall data using hand-held antennas with ATS receivers ) . The mean error of the
ATS using 3.05-m towers was -4.24 degrees (s.d. = 16.49, n = 98). At distances
greater than 25-km, it was not possible to get a reading on the power indicator
of the LOTEK receivers. The mean error of the LOTEK receivers with 3.05-m
towers was 1.65 degrees (s.d. = 14.58, n = 58) with the power indicator and 1.45
degrees (s.d. = 26.86, n = 36) without the power indicator. In pairwise
comparisons, mean bearing error was not significantly different for the 2 LOTEK
classes, but the LOTEK with the power indicator functioning was more precise (P

< 0.00001). The LOTEK with the power indicator was more accurate than the ATS
receiver (P < 0.0257), but testing of variance between the 2 was not
significant. In the comparison of the ATS to the LOTEK without the power
indicator, the means were not significantly different, but the LOTEK was less
precise than the ATS (P < 0.00001).

Because the telemetry studies might involve using both tower and hand-held
systems at the same time, we compared the LOTEK receivers to the ATS receivers
for all methods together. Again the LOTEK without use of the power indicator
was less precise than the LOTEK with the power meter (P < 0.00001) or the ATS (P

< 0.0034). However, there was no significant difference in precision between
the LOTEK using the power indicator and the ATS when data from the towers and
hand-held antennas were analyzed together.

If the assumption is made that the summer and fall test periods do not
differ, then comparison of the 2 3.05-m tower systems (using the same observer)
indicates that the 3.05-m system with PVC mast (u = -3.64) is more accurate than
the 3.05-m tower with a metal mast (u = -8.66, P < 0.0618).

Analysis of Location Estimates

Locations were estimated using a least squares estimator (Pace 1988).
Ninety location estimates were obtained for the summer period. The overall
average distance error was 3381 m with an average associated 95% confidence
ellipse of 2541 ha. Actual coverage was 36.7%. Distance class (the average of
the sum of the receiver- to- transmitter distances classified in 5-km intervals)
and the antenna method-distance class interaction had an effect on the distance
error. These, in addition to antenna method, also had significant effects on
the size of the 95% confidence interval. For the 95% confidence regions, both
the hand-held antennas and 6.1-m towers gave more useful estimates than the
3.05-m towers (P _< 0.05). For distance error, only the hand-held antenna and
3.05-m tower were significantly different (P < 0.05), with the hand-held antenna
being more accurate. For both 95% confidence area and distance errors, the
20+km class was significantly different (P < 0.05), and always gave poor
results .

Because only 2 observers were used during the fall, location estimates were
not obtained. However, given that the bearing errors for fall were similar to
those from the summer period, similar results for location estimates would be
expected.

DISCUSSION

The reception range of the larger hawk and eagle transmitters will be
adequate for triangulation throughout the area within the Range Road. However,
the inclusion of the longer distances of reception increases the error between
the estimated point location and the actual location. The data from the
shorter-range merlin transmitters illustrate this and the advantage of elevating
the receiving antenna; the shorter-distance bearings from the higher antenna
position provided the most accurate bearings.

To achieve coverage of the OTA, bearings must be taken over comparatively
long distances, and inclusion of all data resulted in large bearing error (-5.83
degrees, s.d. = 17.56 degrees), and very large distance errors (3381 m) and 95%
confidence ellipses (2541 ha). These values would not be very useful for
understanding raptor movements. Therefore, we tested some ways of potentially
reducing the error.

Plastic masts were tested to reduce interference with the antenna reception
characteristics that a metal mast can cause under some circumstances. Use of
plastic masts did produce less bearing error than we obtained with metal masts.
However, rigidity and ease of construction and assembly were reduced with the
design we employed for the plastic masts. Greater antenna elevation would
probably improve accuracy, but that will require a more elaborate stand and
stronger structure (less "portable"), to withstand the force of wind on the
antenna.

We tested the signal power indicator on the LOTEK receiver as a means of
more carefully determining the direction from which the strongest signal was
coming. The indicator is not activated if signal strength is not of a threshold
level. Usually, a lack of a reading on the power indicator was indicative of a
transmission from a longer distance. The combination of a closer transmitter
plus use of the power indicator did reduce error.

Additional reduction of error can be obtained with other changes in
equipment, but the funding available and reduced mobility are 2 important
limiting factors in pursuing alternative equipment. Another strategy to control
error is to discard some estimates involving very long distances, or very large
or small angles between 2 receiving points and the transmitter (e.g., Chu et al.
1989 ) . Access restricted to the area along the Range Road limits the extent to
which this strategy can be employed. We continue to explore options for
reducing error. However, these preliminary results suggest that the specific
research objectives will be shaped to some (potentially large) extent by the
precision of raptor location estimates that can be achieved.

ACKNOWLEDGMENTS

This pilot study was made possible by funds from the Idaho Army National
Guard through Col. J. Kane, Chief of Staff, and by support and advice from W. S.
Seegar, U.S. Army Chemical Research, Development and Engineering Center. The
National Guard also assisted in field communications, and we are especially
grateful to staff in Range Control and Fire Suppression. M. Blew and M. Drake
(National Guard), and J. Doremus (BLM) took us to benchmarks; this was not an

enviable task! J. D. Fraser and P. Schempf loaned us transmitters. Mobility,
communication and coordination in the field are essential for triangulation
telemetry. Our tests would not have been possible without the vehicles and
2-way radio communications provided by the BLM. M. Kochert and K. Steenhof were
instrumental in arranging these. Our sincere thanks are extended to S. Senchak
and D. Clugston, who conducted fieldwork with us.

LITERATURE CITED

Chu, D. S., B. A. Hoover, M. R. Fuller, and P. H. Geissler. 1989. Telemetry
location error in a forested habitat. Pages 188-194 in C. J. Amlaner,
Jr., ed. Biotelemetry X: Proc. of the 10th Internatl. Symp. on
Biotelemetry. Univ. of Arkansas Press, Fayetteville.

Pace, R. M. , III. 1988. Radiotracking wildlife by triangulation: an

evaluation of directional measurement errors and location estimators.
Ph.D. Thesis, Purdue Univ., West Lafayette, IN.

TITLE: Abundance of Lagomorphs and Rodents in the Snake River Birds

of Prey Area.

INVESTIGATORS: John Doremus, BLM Bruneau Resource Area
Marjorie Blew, Idaho Army National Guard
Steve Knick, BLM Birds of Prey Research

OBJECTIVE: To monitor changes in the Black- tailed jackrabbit ( Lepus
californicus ) population in the Snake River Birds of Prey
area.

INTRODUCTION

We conducted spotlight surveys along 13 previously established transects
and one new transect (Range 3). The transects run through major cover types
within the study area (Table 1). The Range 3 transect is 6.78 km in length and
runs through big sagebrush ( Artemisia tridentata ) , grass (native and exotic),
and big sagebrush/grass.

METHODS

Each transect was sampled 3 times from 15 May to 3 June, using the method
of Smith and Nydegger (1985). A total of 709 km was sampled in 1989. The
Range 3 transect was selected using the guidelines from the black-tailed
jackrabbit spotlight survey instruction memo on file at the Boise District
Office. Data were analyzed using the computer program "TRANSECT" (Burnham et
al. 1980). The location of all kangaroo rats ( Dipodomys prdii, D. mi crops ) ,
Nuttall's cottontails ( Sylvilagus nuttalli) , pygmy rabbits ( Brachylagus
idahoensis ) , voles ( Micro tus spp. and Lagurus curtatus ) and mice
( Re i throdontomys megalotis , Onychomys leucogaster , Perognathus parvus and
Peromyscus maniculatus ) seen during the jackrabbit surveys were recorded.

Vegetation was recorded along the length of all transects in April and
early May prior to the rabbit surveys. With the same vehicle used to survey
rabbits, all transects were driven during the daylight hours. The mileage at
the start of the transect and at each change in vegetation was recorded to the
nearest hundredth of a mile. Vegetation classifications were ocular and based
on the observer's subjective assessment of the vegetation changes.

The mileage at each small mammal sighting was recorded to the hundredths
of a mile. The location of a small mammal sighting was compared with the
vegetation mileage to determine the vegetation type where the mammal was seen.

RESULTS AND DISCUSSION

The number of jackrabbits observed on the 11 baseline transects increased
from 175 in 1988 to 265 in 1989. The annual density index for all cover types
increased from 0.41 jackrabbits/ha in 1988 to 0.59 jackrabbits/ha in 1989
(Table 2). The density of jackrabbits/ha decreased from 1.33 to 1.05 in the
big sagebrush types from 1988 to 1989 (Table 3).

Table 1. Kilometers for the black- tailed jackrabbit transects by habitat type
1989.

Habitat type Kilometers

Grass 179.47

Big sagebrush 114.86

Big sagebrush/Grass 108.96

Big sagebrush/Winterfat 46.32

Shadscale 43 - 10

Crested wheatgrass 42.00

Shadscale/Grass 30.91

Big sagebrush/Shadscale 25.39

Winterfat/Shadscale 22.94

Winterf at 18.72

Winterf at/Spiny hopsage 13.30

Greasewood 11.57

Winterf at/Grass 10.03

Greasewood/Shadscale 9 . 22

Greasewood/Halogeton 7.92

Big sagebrush/Crested wheatgrass 5.71

Rabbi tbrush/Grass 5 . 18

Bare ground 3.74

Big sagebrush/Spiny hopsage 2.78

Big sagebrush/Nuttall saltbush 1«?8

Halogeton l* 78

Grass/Bare ground 1.63

Big sagebrush/Bare ground 1.15

Spiny hopsage/Grass 0^58

Total 709.04

Table 2. Line transect results for jackrabbit censuses from 1977 through
1989. Transects include all habitat types. Density index was
determined by the exponential power series estimator from the
program TRANSECT.

Density

Distance

No.

Index

Year

(km)

Jackrabbit s

n/km

n/ha

95%

C.I.

c.v.

1977

270.68

218

0.81

0.231

0.200

- 0.261

6.77

1978

262.74

103

0.39

0.322

0.260

- 0.385

9.85

1979

562.52

701

1.25

0.747

0.692

- 0.803

3.78

1980

562.52

807

1.43

0.654

0.471

- 0.837

14.29

1980a

562.52

807

1.43

0.480

0.473

- 0.527

4.98

1981"

534.76

870

1.63

0.778

0.516

- 1.040

17.17

1982

541.17

282

0.52

0.196

0.105

- 0.287

23.74

1983

538.61

194

0.36

0.183

0.157

- 0.209

7.18

1984

536.54

0.11

0.082

0.061

- 0.103

13.02

1985

536.36

0.10

0.030

0.022

- 0.038

13.48

1986

549.34

0.05

0.057

0.036

- 0.078

18.57

1987

613.69

0.14

0.078

0.061

- 0.094

10.85

1988

674.00

190

0.28

0.406

0.348

- 0.464

7.25

1989

709.04

281

0.40

0.591

0.522

- 0.660

5.97

a / Density index determined by negative exponential estimator.
b / Data truncated (w < 100 m) .

Table 3. Line transect results for jackrabbit censuses from 1977 through

1989. Transects include only sage brush habitats (Types 01, 02, 03,
22, 23, 24, 41, 42, 43). Density index was determined by the
exponential power series estimator from the program TRANSECT.

Density
Distance No . Index

Year (km) Jackrabbits n/km n/ha 95% C.I. C.V.

60 1.54 0.312 0.233 - 0.391 12.91

54 0.73 0.797 0.585 - 1.010 13.61

412 1.88 1.208 1.091 - 1.325 4.93

445 2.00 1.046 0.670 - 1.422 18.34

455 2.04 1.067 0.664 - 1.471 19.30

466 2.50 1.323 0.829 - 1.817 19.05

158 0.87 0.385 0.146 - 0.624 31.64

163 0.90 0.400 0.162 - 0.637 30.30

94 0.54 0.458 0.366 - 0.551 10.31

99 0.57 0.465 0.373 - 0.556 10.05

36 0.23 0.127 0.085 - 0.168 16.67

25 0.17 0.056 0.034 - 0.079 20.00

14 0.13 0.087 0.042 - 0.133 26.73

45 0.43 0.152 0.108 - 0.196 14.91

86 0.86 1.327 1.046 - 1.607 10.78

64 0.56 1.047 0.791 - 1.304 12.50

a / Data set from file on disk.

b / Data set adjusted by reconstruction from frequency distribution
of distances in raw data.

c / Data truncated (w < 100 m) .

1977

39.07

1978

73.98

1979

219.14

1980*

222.51

1980b

222.51

1981=

186.40

1982*

181.40

1982b

181.40

1983*

174.20

1983b

174.20

1984

155.40

1985

147.30

1986

105.96

1987

103.97

1988

99.77

1989

114.61

The number of kangaroo rats seen per unit (n/km) varied from 0.00 to 5 17
depending on cover type (Table 4). Kangaroo rats were found in all cover types
in the survey area except big sagebrush/bare ground, grass/bare ground and big
sagebrush/crested wheatgrass. For all cover types the number of kangaroo rats
increased from 0.50 in 1988 to 0.89 in 1989. The number of kangaroo rats
observed increased from 362 in 1988 to 629 in 1989 (Table 5).

The total number of Nuttall's cottontails seen decreased from 24 in 1988
to 13 in 1989 (Table 5). The number of Nuttall's cottontails seen per unit
effort (n/km) varied from 0.01 to 0.09 in those cover types where they were
found in 1989 (Table 6).

The number of pygmy rabbits seen per unit effort (n/km) decreased from
0.01 in 1988 to 0.004 (Table 7) for all cover types. The total number of pygmy
rabbits seen decreased from 9 in 1988 to 3 in 1989 (Table 5).

The number of mice seen per unit effort (n/km) varied from 0.04 to 0.19 in
those cover types where mice were observed (Table 8). The total number of mice
observed increased from 28 in 1988 to 43 in 1989 (Table 5). The total number
of voles seen increased from in 1988 to 1 in 1989 (Table 5). This is the
first sagebrush vole seen during the jackrabbit transects.

The drought that began in the fall of 1986, continued through the summer
of 1989. However, March 1989 had precipitation that ranged from 0.88 inches
above normal at Grandview, Idaho to 2.43 inches above normal at the Boise,
Idaho airport, National Oceanic and Atmospheric Administration (1989). The
temperatures averaged from 2.2 to 2.6 degrees Fahrenheit above average for this
area for the same time period. These conditions were suitable for the growth
of annual grasses and forbs and might have hidden some small mammals that would
have been seen under 1989 conditions.

The information on Nuttall's cottontails, pygmy rabbits, and kangaroo rats
could be analyzed on program "TRANSECT" (Burnham et al. 1980), but low numbers
seen in past years would make comparison of the data very difficult. The
information on mice and voles was not collected in a manner that allows it to
be analyzed on program "TRANSECT" (Burnham et al. 1980).

LITERATURE CITED

Burnham, K.P., D.R. Anderson and J.L. Laake. 1980. Estimation of density from
line transect sampling of biological populations. Wildl. Monogr. 72.
202pp.

Smith, G.W. and N.C. Nydegger. 1985. A spotlight, line-transect method for
surveying jack rabbits. J. Wildl. Manage. 49:699-702

National Oceanic and Atmospheric Administration. 1989 Climatological Data,
Idaho, March 1989. 92(3) :4.

Table 4. Number of kangaroo rats seen per unit effort (n/km) on the spotlight
transects within all cover types within the Birds of Prey Area, 1989.

Cover Type

1989

u v. 1-51

Big sagebrush Q ^

Big sagebrush/Nuttall saltbush

Big sagebrush/Grass

Big sagebrush/Winterfat

Shadscale

Shadscale/Grass

Big sagebrush/Shadscale

Winterfat/Shadscale

Winterfat/Spiny hopsage

Winterfat/Grass

Halogeton

Rabbi tbrush/Grass

Grass

Bare ground

Big sagebrush/Bare ground

Grass/Bare ground

Winterfat

Spiny hopsage/Grass

Greasewood/Shadscale

Big sagebrush/Crested wheatgrass

Crested wheatgrass

Greasewood/Halogeton

Greasewood

Big sagebrush/Spiny hopsage

All cover types pooled

0.51

0.91

1.23

0.00

1.10

1.22

0.90

0,00

4.49

2.70

1.20

0.80

0.00

3.42

5.17

0.98

0.00

1.10

1.39

1.90

0.36

0.86

Table 5. Number of small mammals seen on 709 km of spotlight transects, 1989.

Species

1989

Nuttall's cottontails ^

Pygmy rabbits

Mice

Voles

Kangaroo rats

629

Table 6. Number of Nuttall's cottontails seen per unit effort (n/km) on the

spotlight transects within selected cover types in the Birds of Prey
Area, 1989.

Cover type 1989

Winterf at/Spiny hopsage 0.09

Big sagebrush/Grass 0.04

Big sagebrush 0.02

Big sagebrush/Winterfat 0.02

Grass 0.01

All cover types pooled 0.06

Table 7. Number of pygmy rabbits seen per unit effort (n/km) on the spotlight
transects within selected habitat types within the Birds of Prey Area,
1989.

Cover types 1989

Big sagebrush 0.01

Big sagebrush/Winterfat 0.02

Big sagebrush/Grass 0.01

All cover types pooled 0.00

Table 8. Number of mice seen per unit effort (n/km) on the spotlight transects
within selected cover types in the Birds of Prey Area, 1989.

Cover type 1989

Winterf at/Shadscale 0.04

Grass 0.12

Big sagebrush/Winterfat 0.11

Big sagebrush/Grass 0.06

Big sagebrush 0.05

Rabbi tbrush/Grass 0.19

Crested wheatgrass 0.05

Big sagebrush/Shadscale 0.04

All cover types pooled 0.06

TITLE: Townsend's Ground Squirrel Relative Abundance in the Snake
River Birds of Prey Area.

INVESTIGATORS: Susan Wheeler, BLM Birds of Prey Research
John Doremus, BLM Bruneau Resource Area
Kristina Timmerman, BLM Birds of Prey Research
Karen Steenhof , BLM Birds of Prey Research

OBJECTIVE: To continue monitoring the relative abundance of the

Townsend's ground squirrel in the Snake River Birds of Prey
Area.

ANNUAL SUMMARY

Active Townsend's ground squirrel ( Spermophilus townsendii or TGS) burrow
densities increased slightly from 1988 levels, but badger { Taxidea taxus )
burrow numbers declined between 1988 and 1989. As in past years, winterfat
habitat supported the highest mean TGS and badger burrow densities (b/ha) in
the Snake River Birds of Prey Area (SRBOPA). Exotic annuals supported the
lowest mean TGS and badger densities in 1989. Habitats with the highest TGS
burrow densities showed declines in burrow numbers from 1988 to 1989, while
habitats with low densities showed increases. In range seedings, both TGS
and badger burrow densities declined from 1988 levels.

INTRODUCTION

Townsend's ground squirrels are a major food source for raptors, badgers,
and other predators. The abundance of prey is a crucial factor contributing
to the abundance and diversity of raptors in the Snake River Birds of Prey
Area (Nydegger and Smith 1986). Within the SRBOPA the TGS population can
vary between years (Smith and Johnson 1985). This variation may be caused by
predation, dispersal, weather, and disease, and could affect the populations
of raptors and other predators in the area (U.S. Dep. Inter. 1979).

Wildfire is a primary cause of change in species composition and
structure within habitats. In the SRBOPA, over 75,000 ha burned and were
converted to exotic annuals between 1980 and 1988 (U.S. Dep. Inter,
unpublished data). Nydegger and Smith (1986) reported that big
sagebrush/ winter fat ( Artemisia tridentata / Ceratoides lanata) and winterfat
habitats support higher numbers of TGS than other plant communities in the
SRBOPA. Because the TGS population appears to have a high association with
specific plant communities, it was considered important to monitor TGS
population trends and food habits in altered and relatively unaltered habitat
types.

PREVIOUS WORK

In 1982, 190 hole count transects were established to monitor population
trends in altered and relatively unaltered habitat types (Johnson et al. .
1982). In 1986, Peterson and Yensen (1986) re-sampled 60 of these transects
and established a vegetation canopy coverage transect (Daubenmire 1959) on

each hole count transect. All 60 transects were permanently marked with
fence posts in 1986. In 1987, 54 of the 60 transects were re-sampled.
Hereafter, these 54 transects (sampled in 1982, 1986, 1987, 1988, and 1989)
are referred to as "core" transects. In addition to the 54 core transects,
J° ^f?f ect ?u Were est ablished and permanently marked with fence posts in 1986
^T \Z herB TT e ° f Sma11 raaJimals < incl "ding TGS) and vegSation to a
!!!";! S in /hadscale (Atriplex cpn ferti foli a) and winterfat habitat
types (Groves and Steenhof 1988), and 11 transects were established and
marked with wooden lathes in 1987 to monitor TCS population and vegetation
trends in burned areas that had been reseeded. In 1988, 81 TCS hole count

wltT^r^l^^?' A11 transects not Permanently marked were staked
with rebar in 1988, and fence posts were replaced on transects where they
were missing. Metal tags with the transect number were attached to fence
posts on the 54 transects sampled in 1982, 1986, and 1987. Peterson and
Yensen (1986) concluded that hole count transects seem to be reliable
estimators of relative TGS density in the SRBOPA. For a detailed explanation
of the methods used in 1982, 1986, and 1987, see Peterson and Yensen (1986).

METHODS

In 1989, 81 TCS hole count transects were sampled. Seventy-nine of the
transects had been sampled in 1988 (Timmerman et al. 1988). However, 2 new

iTtefllf 0?\f 4 ^T e eStaMished ^™ the old transect posts "uld
not be found. Of the 81 transects sampled, 54 were part of the "core"

habitat ri ^nf r ° UP ^ n big , sa ^ ebrush ' winterfat, shadscale, and exotic annual
habitat types. Eleven transects, established in 1987, were located in range
seedings. The remaining 16 transects had been established in 1986 in the
vicinity of the Black Butte burn and were located in burned, partially
burned, burned-reseeded, and native vegetation ( shadscale/winterf at ) .

As in previous years, the belt transect method (Peterson and Yensen 1986)
was used on all 81 transects. Each transect was a linear str" 40oTlon^ ^
and 5-m wide (2000 m* ) . A 5-m flexible pole with lightweight chains
approximately 1-m long at each end was used to define the transect width.
One observer held the pole at its center and walked the center line of the
transect. The observer counted all active and inactive TGS and badger

tZ° WS ST X i ne - - ACtiVe h ° leS Were ^termined by the censusers as
those used by ground squirrels or badgers during the year of the census as
evidenced by tracks, claw marks, and feces, and by sightings oftnTaSmaL
forth ^Ttt ? nC tl\, A ^u^ ° bserver ^ed systematically back and
burrows ThV ? R searchl ^ for active and inactive TGS and badger
burrows. The 2 observers were m constant communication with each other to
prevent duplication of the burrow count.

oonnS a i 1 8 + 1 . tra f s f cts ' abundance of TGS and badgers was determined by
counts of active holes. Inactive holes were also recorded. All TCS and
badger burrows were converted to burrows per ha to facilitate comparisons

RESULTS

The number of active TGS burrows increased from 67.7 per ha in 1988 to
73.2 per ha in 1989. The total number of burrows counted on 54 transects was
3955, 300 more than in 1988 (Table 1). Numbers of burrows increased on 25
transects, decreased on 18, and showed no change on 11. Overall, the
increase was not statistically significantly different (Wilcoxon signed ranks
test T = -545.5, n = 43, P = > 0.10).

Among the 54 "core" transects sampled in 1989, the winterfat vegetation
type supported the highest mean number of TGS b/ha (Table 2) . Winterfat
transects have had the highest mean burrow densities in every year since
1986. However, 1989 burrow densities in winterfat were just slightly over
half the levels recorded in 1982 and 1987 (Table 2). Sagebrush habitats had
the next highest burrow densities, followed by exotic annuals and shadscale.

The 2 habitats with the highest burrow densities in both 1988 and 1989
showed declines in burrow densities from 1988 to 1989 (Table 2). Numbers of
burrows decreased on 59% of the winterfat transects and 57% of the sagebrush
transects. However, burrow numbers increased from 1988 to 1989 in habitats
with lower burrow densities. All 6 transects in shadscale habitat showed
increases, and 50% of transects in exotic annual vegetation had increased
burrow numbers. One transect in exotic annuals (#55) had a 30-fold increase
in burrow numbers (Table 1).

The badger population index decreased from 8.8 b/ha in 1988 to 8.5 b/ha
in 1989. The 1989 badger density was 87% lower than the 1982 density of 66.9
b/ha (Table 3). As in past years, there was a high positive correlation
(Spearman rank correlation r s = 0.62, n = 54, P < 0.001) between the number
of active TGS burrows and active badger burrows on transects in 1989.

In 1989, 5 transects in stands that have been dominated by exotic annuals
throughout the study period (1982 - 1989) had higher average TGS densities
(66 b/ha) than 19 stands that had been converted to exotic annuals since 1982
(29 b/ha).

Four years after the Black Butte burn, TGS densities were still higher on
the control transects (70.0 b/ha) than within the burn (Table 4). As in 1986
and 1988, transects in partial burns had the second highest densities (51.3
b/ha) . Densities within the burn increased from 1988 to 1989 in both
reseeded and untreated habitats. Untreated burned areas had higher densities
(25.0 b/ha) than seeded sites within the burn (18.8 b/ha). The 16 Black
Butte burn transects supported an average TGS density of 41.3 b/ha in 1989,
more than twice the 1988 density of 17.2 b/ha (Table 4).

In 1989, the 11 range seeding transects supported an average of 36.8 TGS
b/ha, 36% lower than the 1988 average density of 57.3 b/ha (Table 5). Badger
densities in range seedings decreased from 17.7 b/ha in 1988 to 3.2 b/ha in
1989.

Table 1. Townsend's ground squirrel burrows /hectare on the same transects in 1982
and 1986 - 1989 in the Snake River Birds of Prey Area, Ada and Elmore
Counties, Idaho (1982 and 1986 data from Petersen and Yensen 1986).
Vegetation acronyms are as follows: CELA = winterfat, ARTR = big
sagebrush , ATCO = shadscale , EXAN = exotic annuals ( cheatgrass , mustards ,
Sandberg ' s bluegrass ) .

Active

Transect

Vegetation

Burrows/ha

1982

1986

1987

1989

1982

1986

1987

1988

1989

CELA

735

315

505

455

410

CELA

EXAN

CELA

131

230

320

CELA

232

180

240

CELA

373

260

645

225

175

CELA

624

370

120

275

210

CELA

342

100

250

315

185

CELA

453

150

625

295

265

CELA

322

340

470

255

235

CELA

201

130

CELA

322

340

230

120

100

CELA

50i

CELA

105

115

ARTR

483

ARTR

131

EXAN

584

170

165

ARTR

EXAN

101

155

ARTR

242

ARTR

EXAN

322

120

195

135

ARTR

EXAN

393

105

115

ARTR

EXAN

111

ARTR

121

EXAN

201

ARTR

EXAN

262

145

150

CELA

363

270

560

170

200

CELA

201

180

330

170

230

121

ARTR

121

170

110

115

282

122

ATCO

EXAN

ATCO

123

ATCO

EXAN

124

ATCO

EXAN

312

125

ATCO

EXAN

ATCO

126

ATCO

EXAN

127

ATCO

EXAN

252

128

ATCO

EXAN

147

ATCO

EXAN

101

148

ATCO

EXAN

149

ATCO

111

150

ATCO

EXAN

151

ATCO

EXAN

152

ATCO

EXAN

161

ATCO

EXAN

162

ATCO

EXAN

232

165

ARTR

172

ATCO

141

115

100

182

ATCO

EXAN

183

ATCO

EXAN

200

EXAN

222

100

201

EXAN

332

830

202

EXAN

796

725

100

203

EXAN

CELA

100

204

ATCO

205

CELA

141

206

CELA

207

ATCO

Total

burrow entrances

10460

4130

8105

3655

3955

x burrows /ha

193.3

76.5

150.1

67.7

73.2

1 Change in vegetation classification from previous year.
2 New transect - couldn't find old post.

Table 2. Mean number of active Townsend's ground squirrel burrows per hectare
in the Snake River Birds of Prey Area in 1982 and 1986 - 1989.
Number of transects in parentheses.

Vegetation Type

Active Burrows/ha (# transects)

1982

1986

1987

1988

1989

Ceratoides lanata
Artemisia tridentata
Atriplex confertifolia
Exotic Annuals

x TGS b/ha

290.1(16)

198.9(12)

61.4(20)

367.5 (6)

185.6(16)
55.0 (7)
27.5 (4)
24.6(27)

290.0(16)
65.7 (7)
45.0 (4)

104.6(27)

175.3(15)
56.4 (7)
23.8 (4)
19.1(28)

146.5(17)
53.6 (7)
33.3 (6)
37.1(24)

193.3(54) 76.5(54) 150.1(54) 67.7(54) 73.2(54)

Table 3. Number of badger burrows in different vegetation types in the Snake
River Birds of Prey Area in 1982 and 1986 - 1989. Number of
transects in parentheses.

Vegetation Type

Active Burrows/ha (# transects)

1982

1986

1987

1988

1989

Ceratoides lanata

119.0(16)

53.2(16)

6.9(16)

19.7(15)

20.6(17)

Artemisia tridentata

62.8(12)

8.6( 7)

2.1( 7)

12. 1( 7)

3.6 (7)

Atriplex confertifolia

9.5(20)

11. 3( 4)

2.5( 4)

5.0( 4)

1.7 (6)

Exotic Annuals

127. 7( 6)

5.7(27)

1.3(27)

2.7(28)

3.1(24)

Total # burrows

3614(54)

1110(54)

170(54)

475(54)

460(54)

x Badger b/ha

66.9

20.9

3.1

8.8

8.5

100

Table 4. Density of active Townsend's ground squirrel burrows on the Black
Butte burn transects in 1986, 1988, and 1989. Minimum and maximum
densities per transect in parentheses.

Burrows per Hectare
Range Condition* 1986 1988 1989

Control 36.3 (20-60) 50.0 (20-85) 70.0 (30-110)

Partial Burn 23.8 (5-40) 10.0 (0-35) 51.3 (0-80)

Burn/Reseeded 8.8 (5-20) 8.8 (0-25) 18.8 (0-60)

Burn/No Treatment 11.3 (0-35) 0.0 (0-0) 25.0 (0-70)

x Four transects per range condition

Total 20.0 17.2 41.3

101

Table 5. Active Townsend's ground squirrel and badger

burrow densities in range seedings, 1987 - 1989
( 1 1 transects ) .

Transect
Number

1003
1005
1010

x b/ha

1985
1985
1981

TGS
burrows /ha

1987 1988 1989

1986

125

1986

100

1983

175

1986

1001

1985

1002

1985

180 175

5
45
60
60

20
40
10
10
15
5
135

Badger
burrows /ha

1987 1988 1989

67.7 57.3 36.8

3.1 17.7 3.2

102

PLANS FOR NEXT YEAR

The 81 transects will be monitored in 1990 as in past years. In
addition, a pilot study will be initiated to validate hole count procedures
and to test for variability among transects.

LITERATURE CITED

Daubenmire, R. A. 1959. A canopy coverage method of vegetational analysis.
Northwest Sci. 33:43-64.

Groves, C. R. , and K. Steenhof. 1988. Responses of small mammals and
vegetation to wildfire in shadscale communities of southwestern Idaho.
Northwest Sci. 62:205-210.

Johnson, D. R. , N. C. Nydegger, R. Feldman, S. S. Sundseth, and D. L.

Yensen. 1982. Abundance and demography of prey populations in the Snake
River Birds of Prey Area. Pages 20-37 in Snake River Birds of Prey
Research Project Annual Report. U.S. Dep. Inter. 53pp.

Nydegger, N. C, and G. W. Smith. 1986. Prey populations in relation to
Artemesia vegetation types in southwestern Idaho. Pages 152-156 in E. D.
McArthur and B. L. Welch, eds. Proceedings — Symposium on the biology of
Artemisia and Chrysothamnus . Gen. Tech. Rep. INT-200, U.S. Dept. Agric,
For. Serv., Intermountain Res. Sta. , Ogden, Utah. 389pp.

Peterson, K. , and E. Yensen. 1986. Comparison of 1982 and 1986 Townsend's
ground squirrel burrow entrance transects in the Snake River Birds of
Prey Area, southwestern Idaho. Pages 152-168 in K. Steenhof and M.N.
Kochert, eds. U.S. Dep. Inter. Snake River birds of prey research
project annual report. 172pp.

Smith, G. W. , and D. R. Johnson. 1985. Demography of a Townsend ground
squirrel population in southwestern Idaho. Ecology 66:171-178.

Timmerman, K. , K. Steenhof, J. Weaver, and R. Gerber. 1988. Townsend's
ground squirrel food habits and relative abundance in the Snake River
Birds of Prey Area. Pages 100-121 in K. Steenhof, ed. , Snake River Birds
of Prey Area 1988 Annual Report.

U.S. Department of the Interior. 1979. Snake River Birds of Prey Special
Research Report. Bur. Land Manage., Boise District, Idaho. 141pp.

103

TITLE: Townsend's Ground Squirrel Diets in the Snake River Birds of Prey
Area.

INVESTIGATOR: Eric Yensen, Department of Biology, The College of Idaho

00OPERATORS: The College of Idaho

Boise State University

OBJECTIVE: To investigate the food habits of the Townsend's ground squirrel
in burned and unburned, seeded and unseeded big
sagebrush/winterfat mosaic habitats in the Snake River Birds of
Prey Area.

INTRODUCTION

Townsend's ground squirrels ( Spermophilus townsendii ) are important prey of
raptors in the Snake River Birds of Prey Area (U.S. Dep. of Inter. 1979).
Continued existence of dense breeding populations of raptors depends upon
maintenance of Townsend's ground squirrel (TGS) populations. Recent
fire-induced vegetation changes in the Snake River Birds of Prey Area have
resulted in the replacement of native shrub and bunchgrass-dominated communities
by exotic annual-dominated communities (Yensen 1980, Kochert and Pellant 1986).
Townsend's ground squirrel burrow count transects have indicated that TGS
populations are much less stable in exotic annual -dominated communities (Yensen
et al. in prep.). Consequently, it is important to gather base-line data on TGS
diets and to determine if these exotic species will meet the nutritional
requirements of Townsend's ground squirrels.

Howell (1938), Davis (1939a), Davis (1939b), Alcorn (1940), Johnson (1961),
Johnson et al. (1977), Johnson et al. (1978), Rogers and Gano (1980), Johnson
(1980), Rickart (1982a), Rickart (1982b), and Smith and Johnson (1985) have
reported on TGS food habits in varying amounts of detail. However, information
on TGS food habits in southwestern Idaho sufficient to predict their response to
a diet of exotic annuals is needed. As a step toward that goal, this study was
designed to collect data on diets of Townsend's ground squirrels in the Snake
River Birds of Prey Area.

Dietary analysis using Sparks and Malechek's (1968) microhistological
technique is well-established and widely used (review in Holechek et al. 1982).
There is considerable information on the accuracy of the technique, but the
precision of the dietary estimates has received relatively little attention
(Holechek et al. 1982). Since the technique estimates the contents of a TGS
stomach (10-20 cc) from a series of microscopic fields several orders of
magnitude smaller, both the accuracy and precision of the sampling scheme are of
interest. A secondary goal of this study was to estimate the accuracy and
precision of the Townsend's ground squirrel food habits data.

METHODS

Vegetation analysis (Daubenmire 1959) was conducted at each site in early
June 1987 and late May 1988 while TGS were being collected. Percent cover and %
frequency were determined for each plant species at each site.

104

Study Areas

Townsend's ground squirrels were collected for stomach contents analysis at
4 sites in the Snake River Birds of Prey Area by Bureau of Land Management
personnel. The sites are located near Coyote Butte, approximately 19 km south
of Kuna, Ada Co., Idaho.

Site 1, native grass. This site is a burned big sagebrush ( Artemisia
tridentata ) -winterf at ( Ceratoides lanata) community now dominated by remnant
native Sandberg's bluegrass (Poa secunda ) . six-weeks fescue (Vulpia octoflpra),
and squirrel tail (Sitanion hystrix ). but it has been invaded by tumbleweed
( Salsola lberica), cheatgrass ( Bromus tectorum), and other exotic annuals
(Tables 1 and 2).

Site 2, exotic annuals. This site is adjacent to Site 1, and was similar
to it prior to the burn. However, it is now dominated by cheatgrass and tumble
mustard (Sisymbrium altissimum ) , with some remnant native grasses, especially
Sandberg's bluegrass.

Site 3, big sagebrush. This site is <1 km southeast of Site 2 and is in a
big sagebrush-winterfat mosaic, representing the unburned condition of Sites 1
and 2. Big sagebrush, winterf at, and native grasses (Sandberg's bluegrass,
squirreltail, six-weeks fescue) dominate the site; cheatgrass is the main exotic
annual.

Site 4, rehabilitation seeding. This former big sage-winterfat mosiac is
located 6 km east, 2.5 km south of Site 3. The area burned and was subsequently
reseeded with desert wheatgrass ( Agropyron desertorum) in 1981. The area is
currently dominated by Sandberg's bluegrass, desert wheatgrass, tumbleweed, and
other native and exotic forbs.

Diet Analysis

Squirrels were collected by BLM personnel at all 4 sites in May and June of
1987 (n = 75) and in March and May of 1988 (n = 42) by trapping and shooting,
except no squirrels were collected from Site 4 in May 1988. Squirrels were aged
in the field using body weight criteria (Bureau Land Management, unpubl. data).
Selected specimens were prepared as standard study skins with skulls and
deposited in the College of Idaho Museum of Natural History. Examination of
tooth wear patterns confirmed the age assignments. The 1987 specimens were
collected late in the active season while the adults were entering seasonal
torpor and all were juveniles; all 1988 specimens were either yearlings or
adults .

Stomachs were removed from the animals immediately post-mortem and
preserved in 70% ethanol. In the lab, stomach contents were removed from
ethanol, diluted 50% with water, and homogenized 1 min in a Waring blender to
produce fragments of uniform size. The homogenate was washed through a 1-mm
sieve (Hansen 1978) and collected in a 0.1-mm screen to remove tiny,
unidentifiable fragments. The material was then mounted on microscope slides
using Hertwig's and Hoyer's media (Sparks and Malechek 1968).

105

Plant species in the diet were identified by comparisons to a reference
collection of microscope slides using microhistological characters. All
reference slides were made from catalogued specimens in the College of Idaho
Harold M. Tucker Herbarium and were prepared using the technique described
above.

For food habits analysis, one slide was examined per stomach. Occurrence
of food categories (frequency) was recorded from each of 20 microscope fields
per slide using a phase-contrast microscope at 100X. Frequency/20 fields was
then converted to percent relative density (Sparks and Malechek 1968) using a
table developed for frequency-to-density conversion (Fracker and Brischle 1944).

Three measures of a dietary category's importance in TGS diets were used:
1) Percent relative density as described above was a dry weight conversion from
frequency data. 2) Percent frequency in stomachs was simply the percentage of
stomachs from a site with the item. 3) Percent frequency in microscopic fields
was the percentage of all the microscopic fields from a site with the item.

ACCURACY AND PRECISION OF THE DIETARY ANALYSIS

Eight stomachs were randomly selected from the stomachs collected in 1987
and 1988. Ten slides were prepared from each of these stomachs using the
techniques described above. Three test mixtures were made from known weights of
various mixtures of species occurring in the study areas. Twenty microscopic
fields were examined from each slide using a pre-determined pattern, and
frequency of occurrence of each species was recorded.

The frequency of each dietary category/20 fields on 1 slide was compared to
other slides (or replicate counts of the same slide) using the Kulcyznski Index
(Oosting 1956) (also well known as the "Bray-Curtis" similarity index [Bray and
Curtis 1957]),

2w/(a + b)

The index was calculated as a dissimilarity index,

1 - [2w/(a + b)]

using a BASIC microcomputer program provided by Ludwig and Reynolds (1988) and
converted to similarity when that was of interest.

Intra-slide Variation

Twenty is a convenient number of fields to locate on a standard microscope
slide, either randomly or using a predetermined search pattern. Holechek and
Vavra (1981) used correlations to show that more than 20-24 fields/slide did not
improve accuracy of the estimate, but fewer than 20 fields/slide gave poor
results. Consequently, they recommended 20 fields/slide.

What level of precision is associated with 20 fields/slide? Three slides
were randomly selected, and 20 fields were recorded. This procedure was
replicated 6 times for each slide, and the similarities of the frequency data
from each replication were compared. Mean similarities between replicates were

106

88% (range = 83-93%) for the 3 slides. The mean standard deviation within
slides was 2.09. Thus a given slide will be read with about 88% precision at 20
fields/slide .

Inter-slide variation

For this analysis, 10 slides each were made from stomachs 1987-14, 1987-16,
1988-4, and test mixture "A," and 20 fields were read from each slide. There
are 45 possible comparisons of 10 slides from a given stomach. Mean
between-slide similarities were 87.0% (s = 5.7), 86.6% (s = 7.3), 87.0 (s =
4.1), and 88.6% (s = 3.3), respectively for the 3 stomachs and test mixture.
Thus, there is approximately 87% precision from 1 slide to another prepared from
a single stomach when 20 fields per slide are examined. This is not appreciably
different from the precision of replicate readings of the same slide. I
interpret this as meaning that the food particles in the samples were well mixed
and that the variation in readings was due to sampling error resulting from the
number of fields examined, rather than differences in distribution of food items
among slides from a given stomach.

How many slides constitute an adequate sample?

Running means of the frequency of each species were calculated as the
number of slides from a stomach increased from 2 to 10. Kulcyznski's Index was
calculated for similarities among cumulative means at each increase in sample

size.

For stomach #1988-4, similarity values among running means increased from a
mean of 89.1% (2 slides per stomach) to 94.7% (3 slides) to 97.4% (4 slides) to
98.7% (5 slides) and then went asymptotic, finally reaching 99.0% (8-10
slides). Test mixture "A" with only 3 species increased from a mean of 91.5% (2
slides/stomach) to 97.3% (3 slides), then went asymptotic. Two other samples
(#1987-14 and #1987-16) followed intermediate patterns.

Holechek and Vavra (1981:338) found using a correlation approach that "Nine
slides per sample would estimate all major species in the diet within 10% of the
mean at the 95% confidence level. However, 60 slides per sample would be
required for minor species and 156 slides for trace species to achieve this
level of precision." However, for species that comprise >30% of the diet, 1
slide will yield this level of precision.

My results suggest that mean frequencies will become relatively stable
after counting 5 slides with 20 fields/slide, but that the means should change
<5% after the 3rd slide. These samples contained only a few common species and
thus fewer than 5 slides were required for relatively precise estimates.
Because of the small number of relatively abundant species in TGS diets in the
Snake River Birds of Prey Area, a relatively low number of slides yields
acceptably precise estimates of diet.

The level of precision in a given study will be a compromise between the
precision needed for the purposes of the study and the economic cost of
attaining that precision. For purposes of a diet survey such as this, 1 slide
per stomach is adequate to show general trends. For a detailed analysis in a

107

well-controlled study, the increased precision gained from examining 5 slides
per stomach will be desired. Future emphasis should also focus on sampling more
stomachs (or fecal pellets), since a stomach equals 1 meal only.

The data reported below will have ca. 10% error in precision. Thus for
example, 39% relative density of Bromus tectorum in Table 6 will imply a true
value between 35-43%.

Accuracy

Since only frequency is recorded in reading slides (presence or absence of
a species in a microscope field) , error results from incorrect recognition of a
fragment or error in deciding whether enough of a species is present to count
it. Not every fragment will be recognizable. The major problem involves
species that are easy to recognize, since they will be overrepresented.

Accuracy in microhistological diet analysis technique is gained by
utilizing test slides (Holechek and Gross 1982). Slides are prepared using
known mixtures of plant tissue, then the slide is counted and the results are
compared to the known frequencies. Counting techniques are then adjusted on a
trial and error basis until good results are consistently obtained.

There are 5 major sources of error:

1. Forbs with numerous, easily recognized trichomes (especially winterfat) tend
to be overrepresented.

2. Easily recognized grasses (especially cheatgrass) also tend to be
overrepresented .

3. Conversely, plants with few distinctive features tend to be underrepresented
in the counts.

4. Plants which fragment easily are overrepresented in microscope fields.

5. More digestable shrub and forb species will be underrepresented while
grasses will be overestimated (Vavra and Holechek 1980, Samuel and Howard
1983).

Vavra and Holechek (1980) found actual and estimated dietary frequencies
could vary significantly due to differential digestibility and/or partial
destruction of plant tissue in slide preparation. They recommended the use of
correction factors to compensate for these errors (cf. Leslie et al. 1983).
Others (Holechek and Gross 1982) advocate the use of "counting rules" and test
slides ( ignoring trichomes on certain species , etc . ) as the most efficient way
to compensate.

In this study, I used "counting rules" to minimize over- or
underestimation. Use of test mixtures indicated an additional 5% error
(Bray-Curtis dissimilarity) in the results reported below, mostly due to
overrepresentation of cheatgrass and winterfat. Total error (precision plus
accuracy) is thus estimated at 10-15%.

108

RESULTS

Results of the vegetation analysis (% cover and % frequency) are shown in
•+iTv i Percent cover was converted to % relative cover and averaged
•r 1 £T e< l uenc y to Sive an importance value for each plant species at each
site. There was variation in % cover, % frequency, and importance values of the
species at a site between years. Importance values averaged only 65% similar
(range 48-77%) at a site between years, using the Kulcyznski Index. Total %
?° V ™ decr ^ ased « a11 sites in 1988. The similarity among the 4 sites averaged
48.7/o similarity (range 27-73%) in 1987. Site 3 (unburned sagebrush "control")
was most similar (60%) to Site 1 (native grasses), and less similar to Sites 2
and 4 (burn and seeding, 44 and 47%, respectively). In 1988, when there was
aT^J^ rbaceous cover ' the sites were slightly more similar (mean = 61.3%, range
47-74%) . Thus the sites differed in floristic composition almost as much
between years as between each other.

Although the 3 measures of dietary importance (% relative density (= % dry
weight), % frequency in stomachs, % frequency in fields) gave different
numerical results, the rank orders among categories were generally consistent
(Tables 3-8). However, % frequency in stomachs (Tables 4 and 7) was very
sensitive to small sample sizes.

I found 1-9 food categories per stomach. Site means varied from 3.8 to 4.4
categories per stomach. The total number of food categories used by all TGS
sampled at a site varied from 4-17 on the 3 sampling occasions (May- June 1987,
March 1988, May 1988). However, if species used in trace amounts are
eliminated, only 4-10 categories were used. Furthermore, at each site only 2-4
species comprised >10% relative density in the diet. Species comprising >10% of
the diet at 1 or more study sites included cheatgrass, Sandberg's bluegrass,
six-weeks fescue, winterfat, big sagebrush, tumbleweed, Descurainia spp. , seeds
of bur buttercup ( Ranunculus testiculatus ) . and insects.

Grasses were important constituents of the diet in both 1987 and 1988, and
often comprised over 50% of the diet (37-88% relative density, Tables 3 and 6).
Cheatgrass and Sandberg's bluegrass were both heavily utilized, especially in
March 1988 (55-87% of diet). Late in the TGS active season (May and June) use
of grasses declined (except Site 2 in 1988). Most of the grass eaten in
May- June consisted of seeds, especially of cheatgrass. Sandberg's bluegrass
leaves were utilized slightly more than cheatgrass leaves (Tables 6-8, March)
and the 2 together were far more important than all other grasses combined.
Interestingly, squirreltail was little used, although it was the third most
abundant grass.

Winterfat (0-43% relative density) and big sagebrush (0-21%) were both
eaten, and winterfat was especially important at Site 3 where it was most
abundant. Winterfat was utilized at all sites in 1987, even though it was not
abundant enough to be sampled by the vegetation analysis in some cases. It was

Sfi 2 at Site 3 in 1988, and its use decline d between March and May 1988
(Tables 7 and 8). Big sagebrush was used in March at all sites in both years,
but was less important in May.

Tumbleweed and tumblemustard were the most important forb species
consumed. Tansymustards ( Descurainia sophia and D. pinnata ) . peppergrass
(L epidium perfoliate ) , seeds of bur buttercup, and leaves of prickly lettuce

109

Table 1. Vegetation cover and frequency in May 1987 at 4 study sites
near Coyote Butte in the Snake River Birds of Prey Area,
southwestern Idaho. Data are from 40 1-m* quadrats
(Daubenraire transect) per site. Abbreviations: %C = percent
cover, %F = percent frequency, "tr" = <0.1%, dash = not
present .

Species

Site
%C

1
%F

Site
%C

2
%F

Site
%C

4
%F

GRASSES

Bromus tectorum

0.4

14.6

100

2.0

_-_—

Poa secunda

21.9

3.9

13.7

13.9

93
15

Vulpia octof lora

1.2

0.3

1.4

Sitanion hystrix

1.2

0.3

0.7

1.0

Agropyron desertorum

"""

SHRUBS

Ceratoides lanata

0.1

—

7.0

1.1

Artemisia tridentata

—

■■* — ~~

~ —

11.6

PORBS

Salsola iberica

0.8

—

_- -—

— _

1.0

Descurainia sophia

0.1

—

__—

Sisymbrium altissimum

0.2

1.8

_» —

mmmm

Lactuca serriola

—

— .—

■"*"

"other forbs"

—

TOTAL COVER (%)

25.9

20.6

35.3

18.4

110

Table 2. Vegetation cover and frequency in May 1988 at 4 study sites
near Coyote Butte in the Snake River Birds of Prey Area,
southwestern Idaho. Data are from 40 1-m 2 quadrats
(Daubenmire transect) per site. Abbreviations: %C = percent
cover, %F = percent frequency, "tr" = <0.1%, dash = not
present .

Species

Site

GRASSES

Bromus tectorum

3.8

1.6

0.3

2.9

Poa secunda

5.8

3.6

6.2

5.4

Vulpia octoflora

—

0.3

0.2

0.1

Sitanion hystrix

1.2

2.9

0.8

3.6

Agropyron desertorum

—

SHRUBS

Ceratoides lanata

0.3

—

7.2

—MM

Artemisia tridentata

—

0.8

8.8

_ — —

Atriplex nuttallii

—

1.5

PORBS

Salsola iberica

0.9

0.1

___

0.4

Descurainia sophia

—

__ _

— _—

Sisymbrium altissimum

1.8

0.8

0.4

0.1

Lactuca serriola

—

— —

— —.

auj

"other forbs"

0.7

0.1

TOTAL COVER (%)

14.5

10.2

24.0

14.0

111

Table 3. Late season (25 May- 19 June 1987) percent relative
density of dietary categories in juvenile
Townsend's ground squirrel stomachs at 4 sites in
the Snake River Birds of Prey Area. Adults were
entering torpor and none were collected during this
period. A dash indicates the species was not
recorded, "tr" indicates <1%, and n s number of
stomachs .

Food Item

Site 1 Site 2 Site 3 Site 4

GRASSES

Bromus tectorum

Poa secunda

Sitanion hystrix

—

Oryzopsis hymenoides

—

grass seed

—

grass root?

—

Total grasses

SHRUBS

Ceratoides lanata

Artemisia tridentata

Atriplex nuttallii

—

Chrysothamnus viscidiflorus

—

Total shrubs

PORBS

Salsola iberica

Sisymbrium altissimum

—

Descurainia - 2 spp.

—

Lepidium perfoliatum

—

Cryptantha interrupta

—

Ranunculus testiculatus

—

Lactuca serriola

—

Chenopodiaceae

—

forb

—

Total forbs

MISC.

insects

fungi

—

unknown

—

seed

—

Total misc.

112

Table 4. Percent of stomachs containing a dietary category
in 1987. Samples of juvenile Townsend's ground
squirrels were collected late in the active season
(25 May- 19 June) at 4 sites in the Snake River
Birds of Prey Area. Adults were entering torpor
and none were collected during this period. A dash
indicates the dietary category was not recorded at
the site during the sampling period; n = number of
stomachs .

Food Item

Site 1 Site 2 Site 3 Site 4

GRASSES

Bromus tectorum

Poa secunda

Sitanion hystrix

Oryzopsis hymenoides

grass seed

grass root?

SHRUBS

Ceratoides lanata
Artemisia tridentata
Atriplex nuttallii
Chrysothamnus viscidiflorus

100

—

FORBS

Salsola iberica

Sisymbrium altissimum

Descurainia - 2 spp.

Lepidium perfoliatum

Cryptantha interrupta

Ranunculus testiculatus

Lactuca serriola

Chenopodiaceae

forb

—

5
21

MISC.

insects

fungi

unknown

seed

90
10

87
7

53
5
5

113

Table 5. Percent of microscope fields containing each

dietary category in 1987. Samples of juvenile
Townsend's ground squirrels were collected late in
the active season (25 May-19 June) at 4 sites in
the Snake River Birds of Prey Area. Adults were
entering torpor and none were collected during this
period. A dash indicates the dietary category was
not recorded at the site during the sampling
period; n = total microscope fields.

Food Item Site 1 Site 2 Site 3 Site 4

420 400 300 380

CLASSES

Bromus tectorum

Poa secunda

Sitanion hystrix

—

Oryzopsis hymenoides

—

~— -

~~— '

grass seed

_—

grass root?

SHRUBS

Ceratoides lanata

Artemisia tridentata

Atriplex nuttallii

—

Chrysothamnus viscidiflorus

—

~ ~"

FORBS

Salsola iberica

Sisymbrium altissimum

—

Descurainia - 2 spp.

—

■""

Lepidium perfoliatum

—

Crvntantha interrupta

—

«._

Ranunculus testiculatus

—

— -

— ~

Lactuca serriola

—

Chenopodiaceae

—

forb

—

_—

MISC.

insects

fungi

—

unknown

—

seed

114

Table 6. Percent relative density of dietary categories in Townsend's
ground squirrel stomachs at 4 sites in the Snake River Birds
of Prey Area in 1988. Site 4 was not sampled in May. A
dash indicates the species was not recorded; "tr" indicates
<1%.

Species Site 1 Site 2 Site 3 Site 4

Mar May Mar May Mar May Mar

n 415574 16

GRASSES
Poa secunda
Bromus tectorum
Vulpia octoflora
Sitanion hystrix
Agropyron desertorum
grass seed
grass root

Total grasses 87 57 83 88 55 37 79

SHRUBS

Ceratoides lanata — — — — 24 13

Artemisia tridentata 11 — 15 — 21 3 3

Atriplex nuttallii — 3 — 6 2 3

Chrysothamnus viscid if 1 . — — — — tr

—

"~~ —

—

Total shrubs 11 3 15 6 45 18 6

PORBS

Salsola iberica — 22

Sisymbrium altissimum — — 1 6 tr 1 2

Descurania spp. 2 — — 15 \o

Ranunculus testiculatus — 17 — — 28

Cryptantha interrupta — — — — __ 2

Halogeton glomeratus — — — \

Lepidium perfoliatum ? — — — tr

Crepis accuminata ? — — — tr

Lactuca serriola ? — — tr

Chenopodiaceae

forb root — — tr

Total misc. tr

Total forbs 2 39 1 6 tr 44 15

MISC.

insect — — tr

unknown plants — — — tr

115

Table 7. Percent of Townsend's ground squirrel stomachs containing a
dietary category. Specimens collected at 4 sites in the
Snake River Birds of Prey Area in March and May 1988. Site
4 was not sampled in May. A dash indicates the dietary
category was not found in stomachs at the site during the
sampling period.

Species Site 1 Site 2 Site 3 Site 4

Mar May Mar May Mar May Mar

GRASSES

Poa secunda

100

Bromus tectorum

100

Vulpia octoflora

100

Sitanion hystrix

100

—

Agropyron desertorum

100

grass seed

100

—

grass root

SHRUBS

Ceratoides lanata

Artemisia tridentata

100

Atriplex nuttallii

100

Chrysothamnus viscid.

PORBS

Salsola iberica

100

Sisymbrium altissimum

Descurania spp.

Ranunculus testiculatus

100

Cryptantha interrupta

—

Halogeton glomeratus

Lepidium perfoliatum?

—

Crepis accuminata?

—

Lactuca serriola?

Chenopodiaceae

f orb root

MISC.

insect

unknown plants

116

Table 8. Percent of microscope fields containing each dietary-
category in 1988. Specimens collected at 4 sites in the
Snake River Birds of Prey Area in March and May 1988. Site
4 was not sampled in May. A dash indicates the dietary
category was not recorded at the site during the sampling
period; n = total microscope fields.

Species Site 1 Site 2 Site 3 Site 4

Mar May Mar May Mar May Mar

n 80 20 100 100 140 80 320

GRASSES

Poa secunda

_ _

Bromus tectorum

Vulpia octoflora

—

Sitanion hystrix

—

Agropyron desertorum

—

grass seed

—

grass root

—

SHRUBS

Ceratoides lanata — — — — 55 25

Artemisia tridentata 41 — 35 — 46 8 6

Atriplex nuttallii — 15 — 14 — 4 8

Chrysothamnus viscid . — — — — — 1

FORBS

Salsola iberica — 60

Sisymbrium altissimum — — 5 18 14 4

Descurania spp. 6 — — — 33 25

Ranunculus testiculatus — 55 — — — 43

Cryptantha interrupta — — — — — 4

Halogeton glomeratus — — — — 7

Lepidium perfoliatum ? — — — — — tr

Crepis acuminata? — — — — tr

Lactuca serriola ? — — — tr

Chenopodiaceae — — — tr

forb root — — — 1

MISC.

insect — — ±

unknown plants — — — 7

117

(Lactuca serriola ) were of secondary importance. All of these are introduced
annuals. Common cryptantha ( Cryptantha intermedia ) was the only native forb
found in TGS stomachs. The importance of forbs in the diet increased between
March and May 1988, while the percentage of grasses and shrubs decreased (Table
6).

A surprising number of insects were eaten, especially in May- June 1987
(11-44% of microscope fields, Table 5). However, insects were not important in
1988 (1% of microscope fields at Site 3 only, Table 8). Insect remains were so
fragmentary that identification was not usually possible. However, abundant
lepidoptera larvae could be recognized by the soft exoskeleton and prolegs, and
recognizable fragments of beetle antennae and elytra were found.

The 1988 sampling indicated large differences between March and May diets
(Tables 6-8). Percent relative densities of dietary categories (Table 6) at
Sites 1, 2, and 3 were only 24%, 36%, and 20% similar, respectively, between
March and May, indicating a strong seasonal component in TGS diets.

In 1987, diets were 47% (range 36-73%, s = 13.8%) similar between sites.
Diets were most similar at Sites 2 and 4. Site 1 diets were least similar to
the other sites (37%, 41%, 36% similarity with Sites 2, 3, and 4,
respectively). In March 1988, TGS diets averaged 65% (range 40-75%, s = 15%)
similarity among all sites. By May 1988, TGS diets averaged only 47% (range =
39-51%, s = 7%) similarity among all sites. This decrease in similarity between
March and May corresponds to decreasing importance of grasses and increasing use
of other food categories.

Between-year dietary similarity was 47%, 40%, and 43% at SiteB 1, 2, and 3,
respectively, using the May-June 1987 and May 1988 stomachs. At these sample
sizes, differences between years at 1 site were approximately as large as
differences between sites in a given year.

DISCUSSION

Both the vegetation and TGS diets vary nearly as much between years as
among sites. However, the reasons for between-year differences in vegetation
importance values at a site are not clear. This could be the result of 1) less
September-May precipitation (192 mm vs. 170 mm in 1986-7 and 1987-8,
respectively, at Kuna ca. 20 km N), or 2) differences in timing of precipitation
(no fall rains in 1987), but possibly 3) the result of differences in observer
subjectivity in the Daubenmire technique, or 4) spatial heterogeneity in
vegetation. The Daubenmire quadrats were taken on the same transect in both
years by technicians with the same training. Consequently, differences in
amount and timing of precipitation are the most likely factors.

The substantial annual differences in TGS diets may be the result of 1)
vegetation differences between years, 2) the fact that juveniles were sampled in
1987 and adults and yearlings were collected in 1988, or 3) differences in
collecting dates (25 May-19 June 1987 and 16-19 May 1988).

The differences in age classes are probably not important. Fitch (1948)
found no differences in adult and juvenile diets in California ground squirrels
(S. beecheyi ) and Dyni and Yensen (in prep.) found no dietary differences
between adult/yearling and juvenile Idaho (S. brunneus ) and Columbian (S.

118

columbianus ) ground squirrels. Hansen and Johnson (1976) found no differences
in elegant (S. elegans ) ground squirrel diets by sex or age class. On the other
hand, the 1988 data do show a strong seasonal component. Thus, the observed
annual dietary differences are probably a result of later collecting dates in
1987, combined with annual vegetation differences.

At each site, several plant species were found in TGS stomachs that did not
appear in the vegetation analysis for that site. The Daubenmire (1959) method
of vegetation analysis gave an intuitively acceptable estimate of dominant
vegetation, but for establishing a close link between plant abundance and
herbivore diets, a finer scale method of resource analysis will be necessary.
For example, the Daubenmire method used a single transect with 40 1-m 2
quadrats at each site. However, individual TGS have large home ranges (mean =
1357 m 2 , Smith and Johnson 1985). Given the heterogeneous nature of
vegetation and the wide foraging radius of individual TGS, it is not surprising
that TGS were eating species not recorded by the vegetation analysis.

Although halogeton ( Halogeton glomeratus ) was not recorded by the
vegetation analysis, small amounts of it were found in 2 stomachs at Site 4 in
March 1988 (Tables 6-8) . Halogeton is poisonous to livestock, but sheep can eat
it with impunity in winter, probably because rains have leached the oxalates out
of the dried leaves (Cook 1977). Presumably, TGS were eating dried, rather than
fresh, leaves in March.

Johnson, et al. (1977) reported that Townsend's ground squirrels in the
Snake River Birds of Prey Area ate cheat grass preferentially, and Sandberg's
bluegrass received little use. This study found that both grass species were
heavily used, especially in March. This discrepancy could either be the result
of the methods used (microscopic examination vs. field observations), site
differences, or the result of annual variation in TGS diets.

The number of plant species in the diet not recorded by the vegetation
analysis precluded determining dietary preference indices for TGS. However,
examination of vegetation abundance (Tables 1 and 2) in comparison to
consumption (Tables 3-8) indicates that most abundant plant species were also
important in the diet, and that rare plants were being used only in trace
amounts. There were some interesting exceptions to this, however. Cheatgrass
on Site 4 was dietarily important (39% relative density) but not recorded in the
vegetation analysis.

Diets became more diverse in May, probably as a result of grasses curing and
seeds becoming available. Ground squirrels eat large amounts of seeds prior to
entering torpor (pers. obs.). Perhaps if insufficient seeds are available
during a drought year, TGS turn to insects as a fat source. However, at Site 2
where insect use was highest in 1987, cheatgrass (mostly seeds) was the major
constituent of the diet. This relationship should be explored further.

Idaho and Columbian ground squirrels have highly varied diets of 11-25
plant species per fecal pellet group (Dyni and Yensen, in prep. ) . However, in
that study only 2-4 plant species (usually grasses) contributed >10% to the
diet. Rogers and Gano (1980) found that only 3 plant species (Poa spp. ,
Descurainia pinnata , and Lupinus laxiflorus ) contributed >10% of the diet of TGS
in southeastern Washington. Hansen and Ueckert (1970) found 1-5 species that
contributed >10% in the diverse (47 plant species) diets of elegant ground

119

squirrels (S. elegans ) in Colorado. Hansen and Johnson (1976:750) concluded
that "Richardson [=elegant] ground squirrels graze on a variety of plants as
they fill their stomachs rather than selecting only preferred foods when their
stomachs are nearly empty. This may be an evolutionary strategy developed to
allow them to consume vetches. The dilution of toxic foods by non- toxic foods
decreases the probability of plant poisoning . "

The evolutionary strategy proposed by Hansen and Johnson (1976) may be
general in members of the subgenus Spermophilus . To carry their thinking a step
further, the strategy may be to specialize on 2-4 highly nutritional species but
to supplement them with a wide variety of other species as "poisoning
insurance." In this study, Townsend's ground squirrels similarly depended on
only a few species for the bulk of the diet, but the wide variety of trace
species was not available. The consequences of limited dietary variety on the
long term nutrition of TGS are unknown.

The question of whether TGS can adapt to exotic annuals as dietary staples
is answered in the affirmative by this study. Native forb species were of minor
importance in the diet, but this does not necessarily reflect preference. Native
forbs were so rare at the 4 sites that none were recorded by the vegetation
analysis (Tables 1 and 2), and thus apparently were not generally available for
consumption. We are left with 2 interesting finer-scale questions. How well
can exotic species supply TGS ' long-term energetic and nutritional needs? What
would happen to TGS populations if a highly toxic plant species that could not
be "diluted" in the stomach invaded the SRBOPA?

An understanding of the timing of resource availability and consumption
patterns should better uncover any resource limitations faced by Townsend's
ground squirrels. Since 1987 and 1988 were drought years, future comparative
work in non-drought years should help us understand how weather patterns and
resource availability influence Townsend's ground squirrel diets.

ACKNOWLEDGMENTS

D. L. Quinney designed the sampling and provided collection information on
the ground squirrel specimens and the vegetation analysis techniques. I thank
J. Weaver and other Bureau of Land Management personnel who collected the
squirrels and the vegetation data, B. Dyni and T. Foppe for discussion of the
microhistological analysis technique, and I especially appreciate M. P.
Luscher's assistance in preparing the slides. K. Steenhof and S. Knick made
helpful comments on an earlier draft of the manuscript.

LITERATURE CITED

Alcorn, J. R. 1940. Life history notes on the Piute ground squirrel. J.
Mammal. 21:160-170.

Bray, J. R. , and J. T. Curtis. 1957. An ordination of the upland forest
communities of southern Wisconsin. Ecol. Monogr. 27:325-349.

120

Cook, C. W. 1977. Effects of season and intensity of use on desert

vegetation. Utah Agric. Exp. Sta. Bull. 483, Utah State Univ. 57pp.

Daubenmire, R. F. 1959. A canopy-coverage method of vegetation analysis.
Northwest Sci. 33:43-66.

Davis, W. B. 1939a. The Townsend ground squirrels of Idaho. J. Mammal.
20:182-190.

Davis, W. B. 1939b. The recent mammals of Idaho. Caxton Printers, Ltd.,
Caldwell, ID. 400pp.

Fitch, H. S. 1948. Ecology of the California ground squirrel on grazing
lands. Am. Midi. Nat. 90:334-340.

Fracker, S. B., and J. A. Brischle. 1944. Measuring the local distribution of
Ribes. Ecology 25:283-303.

Hansen, R. M. 1978. Shasta ground sloth food habits, Rampart Cave, Arizona.
Paleobiology 4:302-319.

Hansen, R. M. , and M. K. Johnson. 1976. Stomach contents weight and food
selection by Richardson ground squirrels. J. Mammal. 57:749-751.

Hansen, R. M. , and D. N. Ueckert. 1970. Dietary similarity of some primary
consumers. Ecology 51:640-648.

Holechek, J. L. , and B. Gross. 1982. Training needed for quantifying simulated
diets from fragmented range plants. J. Range Manage. 35:644-647.

Holechek, J. L. , and M. Vavra. 1981. The effect of slide and frequency

observation numbers on the precision of microhistological analysis. J.
Range Manage. 34:357-358.

Holechek, J. L. , M. Vavra, and R. D. Pieper. 1982. Botanical composition
determination of range herbivore diets: a review. J. Range Manage.
35:309-315.

Howell, A. H. 1938. Revision of the North American ground squirrels, with a
classification of North American Sciuridae. North American Fauna
56:1-256.

Johnson, D. R. 1961. The food habits of rodents on rangelands of southern
Idaho. Ecology 42:407-410.

Johnson, D. R. , and G. W. Smith, R. M. Olson. 1977. Population ecology and

habitat requirements of Townsend ground squirrels. Pages 203-224 in Snake
River Birds of Prey Research Project, Annual Report, 1976.

Johnson, D. R. , G. W. Smith, E. M. Cobb, and C. W. Woodley. 1978. Population
ecology and habitat requirements of Townsend ground squirrels. Pages
143-153 in Snake River Birds of Prey Research Project, Annual Report,
1977.

121

Johnson, M. K. 1980. Food of Townsend ground squirrels on the Arid Land
Ecology Reserve (Washington). Great Basin Nat. 37:128.

Kochert, M. N. , and M. Pellant. 1986. Multiple use in the Snake River Birds
of Prey Area. Rangelands 8:217-220.

Leslie, D. M. , Jr., M. Vavra, E. E. Starkey, and R. C. Slater. 1983.

Correcting for differential digestibility in microhistological analyses
involving common coastal forages of the Pacific Northwest. J. Range
Manage. 36:730-732.

Ludwig, J. A., and J. F. Reynolds. 1988. Statistical ecology. John Wiley &
Sons, NY.

Costing, H. J. 1956. The study of plant communities, 2nd ed. W. W. Freeman,
San Francisco.

Rickart, E. A. 1982. The ecology of Townsend 's ground squirrei, Spermophilus
townsendii mollis . Ph.D. Dissertation, University of Utah, Salt Lake
City. 74pp.

Rickart, E. A. 1982. Annual cycles of activity and body composition in
Spermophilus townsendii mollis . Can. J. Zool. 60:3298-3306.

Rogers, L. E. , and K. A. Gano. 1980. Townsend ground squirrel diets in the
shrub-steppe of southcentral Washington. J. Range Manage. 33:463-465.

Samuel, M. J., and G. S. Howard. 1983. Disappearing forbs in

microhistological analysis of diets. J. Range Manage. 36:132-133.

Smith, G. W., and D. R. Johnson. 1985. Demography of a Townsend ground
squirrel population in southwestern Idaho. Ecology 66:171-178.

Sparks, D. R. , and J. C. Malechek. 1968. Estimating percentage dry weight in
diets using a microscopic technique. J. Range Manage. 21:264-265.

U.S. Department of Interior. 1979. Snake River Birds of Prey. Special
Research Report. U.S. Bur. of Land Manage., Boise Dist., Boise, ID.
142pp.

Vavra, M. , and J. L. Holechek. 1980. Factors influencing microhistological
analysis of herbivore diets. J. Range Manage. 33:371-374.

Yensen, D. L. 1980. A grazing history of southwestern Idaho with emphasis on
the Snake River Birds of Prey Area. Prog. Rep., U.S. Dep. Inter.,
Bur. of Land Manage., Boise, ID. 82pp.

122

TITLE: Factors Affecting the Mating System in Townsend's Ground
Squirrels .

INVESTIGATOR: Claire Fuller, Dep. Zoology, Oregon State University

COOPERATOR: Oregon State University

INTRODUCTION

This work was done as part of a pilot study to determine the effect of
population density and sex ratio on reproductive success in Townsend's ground
squirrel ( Spermophilus townsendii ) . Three study sites were established and
animals were marked so that breeding activity could be observed the following
winter .

METHODS

Three trapping grids for S_^ townsendii were established in the
northwestern portion of the Snake River Birds of Prey Area (SRBOPA). Traps
were set in winterfat communities in Sections 15, 20, and 21, Township 1 South,
Range 1 West. Grids were 200 x 200 m per side (4 ha) with quadrat markers at
20-m intervals. Animals on 1 of these grids {Grid 1) were trapped with
Pymatung and/or National live- traps using apple pieces as bait. Because my
primary objective was to trap and mark all resident squirrels, rather than
determine population density, I set 2-4 traps per quadrat near burrow
entrances. One quarter to one half of the grid was trapped at any one time
(approx. 200 traps /day) . Traps were first set early in the morning (usually @
0730) then checked every 1.5-2 hours until approx. 1400. Animals were
trapped on the grid from 4 May - 16 May 1989 and around the edges of the grid
on 17 May 1989.

At first capture, all animals were toe-clipped for permanent
identification, sexed, weighed and the capture location was recorded. In
addition, adults were anesthetized with ether and given an additional mark on
their back with Clairol black hair-dye. This mark will facilitate
identification of individuals from a distance. Pups were not dye-marked
because the first 4 anesthetized stopped breathing and had to be resuscitated
It is unclear why they were more susceptible to ether than adults. When
animals were recaptured, they were identified and their location was recorded.
Animals recaptured on 5, 9, 14 and 15 May were weighed again and these second
weights were used to calculate growth rates. All animals were released at the
point of capture.

Animals were separated into adults and juveniles based on weight and, for
females, on the presence or absence of enlarged teats or mammary glands. The
largest male pup weighed 150 g and the smallest adult male weighed 208 g. For
females, the largest pup weighed 137 g and the smallest adult weighed 142 g —
these animals were exceptionally large and small, respectively.

123

RESULTS

A total of 127 animals were captured in 11 days of trapping. Approximately
200 traps were set per day. Twenty- three adults (8 males and 15 females) and 104
juveniles (48 males and 51 females) were captured. Two pups were not sexed, and
3 died in traps as a result of extremely hot or cold weather.

Seventy-six of the animals were re-captured between 1 and 5 times (Fig. 1).
Because only a small portion of the grid was trapped on any one day, this figure
underestimates the trappability of Sj. townsendii . There were no differences in
the number of times captured between age classes or sexes, suggesting that age
and sex classes were equally trappable (Table 1).

Adult males were significantly heavier than adult females (t = 5.74, df =
21, P < 0.001; Table 1), and juvenile males were significantly heavier than
juvenile females (+ = 5.09, df - 96, P < 0.001; Table 1). Growth rates were
obtained for 16 juveniles which were re-weighed 2-11 days (mean = 6.3) after
initial capture. One pup which lost weight was excluded from the analysis. Male
pups grew significantly faster than female pups both in grams per day (t = 2.45,
df = 14, P < 0.05; Table 1) and as a percentage of initial body weight per day (t
= 2.13, df = 14, P < 0.05, Table 1).

124

O
LU

1 2 3 4 5 6

NUMBER OF CAPTURES PER ANIMAL

Figure 1 ,

Frequency of Captures for Individual Animals. This represents data pooled across sex
and age classes (see text). In addition, captures were pooled across a 2-week trapping
period. Animals which were first marked later in the study were not "at risk" as long
as animals marked early in the study Thus, these data can only be used to obtain a
general idea of trappability.

Table 1 . Comparison of mean capture rates and mean weights between sex and age
classes of Townsend's ground squirrel.

Adults

Male

Female

Juveniles

Male

Female

Number Caught

Number of Times
Individuals were
Captured (SD)

Weight (SD)*

Weight Gain in
G/Day (SD)**

2.67(1.37)
n = 6

248.9(24.9;
n = 8

2.67(1.32)
n = 9

186.4(24.7)
n = 15

2.66(1.55)
n = 38

115.5(12.2)

n = 47

2.38(1.16)
n = 11

2.42(1.19)
n = 42

104.3(9.5)
n = 51

1.03(0.60)

n = 5

Weight Gain in
% Original Body
Weight/Day (SD)**

7.87(2.21)
n = 11

5.53(1.55)
n = 5

* Within age classes, males are significantly heavier than females; P < 0.001,
Student's T-test.

** Differences are significant at 0.05 level. Student's T-test.

126

TITLE: Parasites and Population Ecology in Kangaroo Rats.

INVESTIGATOR: James C. Munger, Dep. of Biology, Boise State University

COOPERATOR: Boise State University
OBJECTIVES :

1. To identify in a local population of kangaroo rats ( Dipodomys spp. ) a
parasite species that is at reasonable prevalence and is relatively
easy to work with.

2. To establish a trapping grid to gather baseline data on the
prevalence of selected parasite species and on the reproductive
status of kangaroo rats.

3. To perform preliminary studies to determine the effect of selected
parasite species on the energy budget of host kangaroo rats.

INTRODUCTION

Little empirical evidence exists to allow a critical assessment of the
role of parasitism in host population ecology. Over the last 4 years I have
initiated studies that study the effects of parasites at two levels: (i) on the
energy budget of host individuals and (ii) on the reproduction and survival of
individuals. Effects at either level should have important ramifications for
host populations.

Research that I recently completed in Wisconsin documented effects of 2
parasites, a tapeworm and a fly larva, on host white-footed mice. We showed
that (i) both parasites have small but measurable effects on the host energy
budget, (ii) we were unable to detect compensations for these perturbations in
the energy budgets, leading us to conclude that effects on energy budget were
not of major significance, (iii) we were unable to detect effects of either
parasite on survival or reproduction (Munger and Karasov 1989 and in press,
Munger et al . 1989 ) .

The goal of the present research is to gather similar evidence on a
different system. It is hoped that the methods I am using will eventually
provide a means by which probable effects on host population dynamics can be
predicted by study of parasites on selected components on the host energy
budget .

METHODS

I trapped at several sites on the Snake River Plain before settling on the
Rabbit Creek area north of Murphy in Owyhee County as the most desirable site
for further study. It has substantial populations of kangaroo rats and there
are sites which are relatively isolated from human activity. On 12 July 1989 a
13 x 13 grid with 15 m between stations was established on the plain
approximately 500 m W of Con Shea road, at T2S R2W S14. This has been trapped
with Sherman live traps baited with millet on the occasions noted in Table 1.

127

Table 1. Number captured and prevalence (in parentheses) of Trichurus sp.
infection.

Date Dipodomys ordii Perognathus parvus Peromyscus maniculatus

13 July 8 (12.5%) 1 (0)

14 July 11 (27.3) 7 (0)
2 August 36 (29.4) 22 (0)

5 September 24 (20.8) 16 (0)

6 October 8 (37.5) 8 (0) 3 (0)

128

All animals captured were identified to species and eartagged with monel
fingerling tags. Reproductive status was assessed, and a fecal specimen was
taken. Fecal specimens were analyzed for the presence of parasite eggs by
standard centrifugal flotation techniques using saturated sucrose solution
(Pritchard and Kruse 1982).

RESULTS

Table 1 depicts the captures by species and gives the prevalence of the
only gut parasite ( Trichurus sp. , a nematode) both easily identifiable from its
eggs and at reasonable prevalences.

PLANS FOR 1990

I plan to recommence trapping in March, and continue monthly trapping at
least through July. In May, assuming prevalence of Trichurus is high enough, I
will begin a laboratory study to assess the effect of this parasite on
digestive efficiency of kangaroo rats. Thirty infected kangaroo rats will be
brought into the laboratory, and digestive efficiency will be measured by the
mass balance method (Munger and Karasov 1989). Then half the nematodes in half
of the rats will be killed with Ivomectin, and digestive efficiency will be
remeasured. If removal of the nematodes causes a significant increase in
digestive efficiency, I will pursue other aspects of the effect of this
parasite on the energy balance and population biology of kangaroo rats.

LITERATURE CITED

Munger, J. C. , and W. H. Karasov. 1989. Sublethal parasites and host energy
budgets: tapeworm infection in white-footed mice. Ecology 70:904-921.

Munger, J. C, and W. H. Karasov. (In Press). Sublethal parasites in

white-footed mice: impact on survival and reproduction. Can. J. Zoology

Munger, J. C, Karasov, W. H., and D. Chang. 1989. Host genetics as a cause
of overdispersion of parasites among hosts: how general a phenomenon. J. of
Parasitology 75:707-710.

Pritchard, M. , and G. Kruse. 1982. The collection and preservation of animal
parasites. University of Nebraska Press. 141pp.

129

TITLE: Effect of Wildfires on Soil Microbial Communities.

INVESTIGATORS: Marcia Wicklow-Howard, Dep. of Biology, Boise State University
Robert Rychert, Dep. of Biology, Boise State University

COOPERATOR: Boise State University

MICROBIAL ACTIVITY

Study sites were established within the Snake River Birds of Prey Area
(SRBOPA), Idaho. Sites were characterized as burned and unburned sagebrush,
and as burned and unburned winterfat ( Ceratoides lanata) . The unburned
sagebrush site is dominated by big sagebrush (Artemisia tridentata) and
cheatgrass ( Bromus tectorum ) . Root and soil samples were taken from these
sites immediately following the burns in August 1987 and then from April to
June 1988.

The sagebrush burn site exhibited reduced microbial activity and biomass-C
(August 1987) when compared with the unburned site. The converse was true for
the winterfat site. However, when dehydrogenase activity per unit biomass-C
was compared for burned and unburned sites, the differences between activity
and biomass on burned and unburned sites are not pronounced. By June 1988, the
activity per unit biomass for the unburned sites is not significantly greater
than the burned sites. In terms of overall microbial activity and biomass,
spring rainfall readily lends to a restoration of any lost microbial activity
and biomass following an intense summer rangeland fire.

SOIL MINERAL NITROGEN

The soil mineral nitrogen (N) status did exhibit pronounced differences on
burned versus unburned plots on both the sagebrush and winterfat sites (Figs. 1
and 2). While soils were analyzed for pH, organic-C, total phosphorous (P) ,
total N, and soluble P, only the NH4 + -N and NOr-N values were
different on burned versus unburned plots. These preliminary results would
suggest that the major impact of rangeland fires on soil microorganisms and
their activities is reflected in major effects on the N cycle. Fire appears to
enhance mineralization of N initially, and subsequently there is an increase in
the soil N0 3 --N. The latter may be due to enhanced nitrification or
reduced denitrification. Since biomass-C differences are not very great, the
data suggest an enhancement of nitrification.

MYCORRHIZAL ROOT INFECTION

Roots of plant species were microscopically examined for the presence of
mycorrhizal infection, and quantified to obtain percent root area infected
(Table 1). In the unburned sagebrush site, it appears that big sage has a
greater mycorrhizal infection at all samplings, as compared to cheatgrass and
other grass species. Additionally, the mycorrhizal infection appears to
increase from a low of 47% in April to a high of 80% in May. On the burned
site, little root infection is noted for early spring on any of the plants

130

Sagebrush

40 n

20 -

NO^-N/g soil

10 -

burned

unburned

1 1 1 1

Aug 87 Apr 88 May 88 Jun 88

Fig. 1. Concentration of Soil Nitrate/Nitrogen following burn.
(Mg/gr)

131

Winterfat

30 n

NO3- N/g soil

10 -

- - - burned

unburned

1 1 1 1 r

Aug 87 Apr 88 May 88 Jun 88

Fig. 2. Concentration of Soil Nitrate/Nitrogen following burn,
(/ig/gr)

132

mammma ^ stmmlll

Table 1. Percent mycorrhizal root infection. Percent of samples indicated in parentheses.

S tudy Site

Sagebrush Sagebrush Winterfat Winterfat
Plant Species Date Burned Unburned Burned Unburned

Winterfat (Ceratoides lanata) APR 88 -- -- - (10) 0%

MAY 88 -- -- -- (10) 0%

JUN 88 -- -- -- (10) <o.i%

SEP 88 (2) 0% - -- ( 9) 0*

Big Sagebrush APR 88 -- (10) 47%

(i r i e |i?ia tridentata MAY 88 -- (8) 80%

var. wyoming ensis ) JUN 88 -- (8) 59%

SEP 88 -- (4) 44%

Cheat Grass (Brows tectoruj) MAY 88 -- ( 2) 34%

JUN 88 ( 6) 1.5% ( 9) 21%
SEP 88 ( 4) 7% ( 8) 8% ( 5) <0.1% ( 7) 0%

Squirrel Tail (Sitanion hystrix) JUN 88 ( 2) 21% ( 1) 10%

SEP 88 (1) <0.1%

Unidentified Grass Species APR 88 ( 9) 4% (10) 16% ( 7) 0.1% (10) 0%

MAY 88 (11) 14% ( 9) 29% ( 9) 3% (11) <0.1%

JUN 88 -- -- ( 8) 0.4% (10) 0%
SEP 88

Russian Thistle (Salsola kali) MAY 88 ( 1) 0%

JUN 88 ( 1) 0% - ( 5) 0%
SEP 88 ( 3) 0% -- ( 4 0%

Mustard (Descurainia p innate ) JUN 88 ( 1) 0%

Prickly lettuce (Lactuca serriola) JUN 88 (1) 30%

Onion (Allium sp.) APR 88 ( 1) 6%

Burr Buttercup

( Ranunculu s testiculatus) APR 88 -- ~ - ( i) o%

Kochia SEP 88 -- - ( 2) 0%

133

initially colonizing the area. Generally, plants colonizing the sites the
first year following a burn do not form mycorrhizal root infections. Species
such as Russian thistle ( Salsola kali ) and cheatgrass appear to have little
mycorrhizal dependence, and can colonize areas lacking mycorrhizal propagules.
Changes in mycorrhizal spore numbers in soil were noted from April through June
1988. On the winterfat site, spore numbers increase some during this period of
time. Winterfat does not normally form mycorrhizae even when propagules are
available. In contrast, the unburned sagebrush sites showed a decrease in
spore numbers in the soil, as mycorrhizal root infections increased throughout
the summer.

134

TITLE: Winterfat Seeding Projects in the Snake River Birds of Prey
Area.

INVESTIGATORS : Stephen B. Monsen, USDA-FS Shrub Sciences Laboratory
Mike Pellant, BLM Idaho State Office

C00PERATORS: USDA-FS Intermountain Research Station, Shrub Sciences Lab

OBJECTIVE: Determine the establishment and recruitment of winterfat on
reseeding projects.

INTRODUCTION

Revegetation plantings using winterfat ( Cera to ides l anata ) with various
perennial grasses were established at 3 separate burn sites within the Birds of
Prey area in the past 7 years. These sites include Melba, Kuna/Coyote, and
Trio. Plantings were conducted using different methods of broadcast seeding on
disturbed burns where winterfat and big sagebrush ( Artemisia tridentata ) had
occurred. These plantings were the first attempts to seed winterfat with
grasses on extensive areas of arid shrublands in southwestern Idaho. Different
methods of broadcast seeding were used to revegetate the burns (Table 1).
Methods of planting included hydroseeding , aerial seeding of pelleted and
nonpelleted seed. The 3 large areas were seeded at different dates, and
provide a useful number of sites to evaluate planting methods, initial
establishment, survival, and natural spread of seeded shrubs. Studies were
established in 1988 at the 3 planting locations to evaluate these factors.
This report summarizes the results of shrub seedling recruitment that has
occurred in 1988 and 1989. The previous year's progress report discussed the
initial establishment attained by different seeding methods. To date, studies
indicate that winterfat can be successfully established by different methods of
broadcast seeding and that natural recruitment can occur rather rapidly even
during years of low rainfall. However, the long-term results of these seedings
have yet to be determined.

METHODS

Winterfat plants established at all 3 planting sites from the broadcast
seedings conducted in the early 1980' s (Table 2). These plants matured and
began to produce seed crops in 1987. New seedlings appeared in the spring of
1988, and studies were initiated to record seedling establishment, survival,
and the influence of associated plants and soil factors upon shrub seedling
presence and growth. Permanent study plots established in 1988 were
reinventoried in 1989 at all 3 planting sites. The number of plants that
established in the spring of 1988 and 1989 were separately recorded.

At the Melba location, study plots were separately established on sites
that supported native stands of winterfat or Wyoming big sagebrush. Areas
within the burn that were occupied by winterfat or sagebrush could be
identified and separate study sites could be established. The occurrence of
different plant communities at the Trio and Kuna/Coyote locations could not be
determined following burning.

135

Table 1. Fire impacts and planting practices at 3 study sites in the Snake
River Birds of Prey Area

Melba

Study Sites
Kuna/Coyote

Trio

Date of Burn
Date Seeded
Ha Burned

Planting Methods

20 Aug 82

March 83

554

Hydro/harrow

19 Sept 81

Nov 82

10,664

Aerial and
hydro /harrow

12 July 85

March 86

2,024

Aerial pellet
drill

Table 2. Seeding success for 3 planting sites in the Snake River Birds of
Prey Area.

Melba Winterfat
No . /ha

Melba Sage
No. /ha

Kuna/Coyote
No. /ha

Trio
No. /ha

No. seeds sown 21,255
No. plants attained 3,138
% success 14.59

21,255

2,083

9.80

103,238 32,793

2,194 944

2.13 2.88

136

RESULTS AND DISCUSSION

At all 3 planting locations, approximately 50% of all mature winterfat
plants that established from broadcast seeding supported new seedlings in

1988. By 1989, seedling numbers diminished, and about 30% of all mature shrubs
at the Melba and Kuna/Coyote plantings supported seedlings (Table 3). Nearly
all seedlings that appeared at the Trio site in 1988 failed to survive until

1989. At the Melba location, 32% of the winterfat plants growing on the
original winterfat sites were producing seedlings compared to 28% of the plants
found growing on the original sagebrush communities.

Winterfat seedlings were near or beneath the crown of a mature shrub. Few
seedlings were further than 1-m from a mature plant. Winterfat seedlings often
occurred in groups near the base of a mature plant (Table 3). Seedlings were
usually spaced a few centimeters apart from each other. The winterfat
seedlings encountered in 1989 were vigorous and well established individuals.
Few seedlings would likely succumb, and most are expected to persist and
develop into mature plants.

Fewer plants became established at the Trio site from broadcast seeding
than at either of the other 2 locations. In addition, fewer seedlings were
encountered as natural recruitment. Mature plants at this location appeared to
produce less seed than at the other sites. However, the lack of seedling
survival cannot be explained from available data.

Winterfat seedlings were able to establish amid considerable perennial
herbaceous competition. Sandberg's bluegrass ( Poa secunda ) was the most
abundant understory species at all locations. However, the presence of
winterfat seedlings was not restricted by the density of this perennial grass
even if grass cover exceeded 30% (Fig. 1). Perennial grass cover occurred in a
rather normal distribution pattern at the Melba sage and Kuna/Coyote sites. At
the Melba winterfat site, perennial grass cover increased following the burn,
and a greater number of sites were occupied by Sandberg's bluegrass. A
significant increase of Sandberg's bluegrass was recorded in all cover
categories at this location. Yet, similar numbers of winterfat seedlings were
recorded in sites where grass cover ranged from to 30% ( Fig . 1 ) .

The recruitment of winterfat seedlings is occurring at a rapid rate,
particularly for an arid shrubland situation. Although little or no
recruitment was recorded prior to 1988, the seeded shrubs appeared to have
reached maturity in 3 to 6 years and have produced seed crops in 1988 and
1989. In addition, subsequent survival of new seedlings has occurred during
periods of low rainfall.

The presence of annual weeds, particularly cheatgrass ( Bromus tectorum ),
did have a negative affect upon winterfat seedling survival. Most winterfat
seedlings occurred on sites where the cheatgrass cover was lower than 10% (Fig.
2). No other plant had such a strong negative influence upon shrub seedling
survival. Although cheatgrass density adversely affected shrub seedling
survival, this annual weed did not dominate all portions of the burns.
Cheatgrass did not occur throughout the native winterfat communities prior to
burning. In addition, annual grasses did not invade these areas very quickly
after burning, particularly within sites where Sandberg's bluegrass was not
killed by the fire. Study sites that supported Wyoming big sagebrush contained

137

Table 3. Number of seeded winterfat plants producing new seedlings in 1989,

Planting Site

Kuna Winterfat Kuna Sage Kuna/Coyote Trio

% shrubs with
seedling

% shrubs with
multiple seedling

138

Melba Sage Site

m
§

Melba Winterfat Site

-2

S3
U

a
iS

L^fJfi

Cover Class

Coyote Site

H
Ik
•Ha

O 1 2

.Eft-

Fig. 1. Frequency of winterfat plants related to perennial grass
cover.

139

Melba Sage Site

£3

=*5

23430788

Cover Class

Melba Winterfat Site

«3
t

Q
I*

i i i i i ■__ ._

2 3 4 B 8

Cover Class

7 8 »

Coyote Site

Z££i Za s«a tZa. e«i_

O 1 2

3 4 S O

Cover Class

7 8

Fig. 2. Frequency of winterfat plants related to annual grass
cover.

140

a greater density of annual weeds than did the winterfat communities prior to
burning. The annual weeds increased rapidly in these areas, but their presence
did not restrict shrub establishment from the initial broadcast seedings or by
natural recruitment. However, it is apparent that burned sites should be
seeded before cheatgrass is able to establish or gain dominance.

It is apparent that Sandberg's bluegrass controlled the invasion of
annuals, but had less influence upon the recruitment of winterfat seedlings.
This is an important ecological relationship, particularly as it affects
restoration of native shrub communities. Burn sites are frequently seeded with
an understory of introduced grasses as a means to control invasion of
cheatgrass and other annual weeds. However, the presence of perennial grasses
often restricts the recovery of native shrub and herb seedlings. If Sandberg's
bluegrass could be used to selectively control annual weed invasion, yet allow
for recruitment of desirable species, natural plant succession could be better
employed to assure the recovery of diverse plant communities.

The establishment of winterfat by broadcast seeding is encouraging.
Numerous plants become established with minimal site preparation. In addition,
natural spread has occurred very quickly even during years of below normal
precipitation. Few shrub species are able to spread as well as winterfat,
particularly into a competitive understory of herbaceous species.

The winterfat seed that was planted on the 3 sites was acquired from a
source in western Utah. It has a more upright growth from than the native
ecotype, but it has grown well and has spread much better than the remaining
natives. The long-term implications of seeding an introduced strain is not
known. However, seeding winterfat on burn sites can be recommended, and
plantings can be designed to allow for natural spread to increase stand
density. Where possible, the presence of Sandberg's bluegrass should be
maintained to control cheatgrass and act as a nurse crop to aid shrub seedling
recruitment.

141

TITLE:

INVESTIGATORS I

COOPERATOR:

OBJECTIVES:

Environmental Monitoring in the Idaho Army National Guard
Orchard Training Area.

Dana Quinney, U.S. Army Corps of Engineers
Marjorie Blew, Idaho Army National Guard
Allison Brody, U.S. Army Corps of Engineers
Robert Gerber, U.S. Army Corps of Engineers
Jennifer Walrath, U.S. Army Corps of Engineers
Jay Weaver, U.S. Army Corps of Engineers
Jeff Courson, U.S. Army Corps of Engineers
Wiliam Whitworth, U.S. Army Corps of Engineers

U.S. Army Corps of Engineers CERL
Idaho Army National Guard

To assess and monitor the land condition and mammal populations
in the Orchard Training Area in order to minimize the impacts
of National Guard training.

ANNUAL SUMMARY

Two hundred ten permanent land condition-trend vegetation/soil monitoring
plots were established and inventoried, and a program of monitoring small
mammal populations was initiated.

METHODS

Vegetation/soils monitoring

A May 1988 SPOT satellite image of the Orchard Training Area (OTA) was
obtained, and the image was digitized using GRASS (Geographic Resources
Analysis Support System) , a computerized geographic information system. Twenty
land cover categories based upon reflectance values in the green, red, and
near-infrared wavelength bands were determined. The resulting land cover data
layer was superimposed on a digital soil series data layer derived from the
maps in the Soil Surveys of Ada County Area (U.S. Soil Conservation Service,
1960) and Elmore County Area (U.S. Soil Conservation Service, unpublished) Soil
Surveys. Each unique land cover/soil series combination was recognized as a
separate category.

Using a computer algorithm, every occurrence of each unique land
cover/soil combination was identified on a map of the Orchard Training Area as
one or more polygons. Another computer algorithm determined the area of each
polygon and eliminated all polygons smaller than 2 ha, leaving only those
polygons large enough to locate and inventory in the field. Two hundred plots
were to be established. Plot sites were chose by first determining the number
of plots to be established in each land cover/soil type by assigning a number
of plots proportional to the percent of land area occupied. For example, a
land cover/soil series category comprising 10% of the total land area of the
Orchard Training Area would be assigned 20 plot locations. The plot locations
within each category were then allocated to the polygons of that category by
random means.

142

The 200 plot locations were identified by map coordinates; these were the
preselected sites of the land condition- trend analysis (LCTA) sampling plots.
During field work, the locations of several LCTA plots were moved slightly
(within the land cover/soil category) from their preselected map-coordinate
locations in order to facilitate relocating them in later years, to minimize
off-trail travel in reaching them, or to avoid entering the core artillery
impact area. When establishing the plots in the field, no effort was made to
avoid trails, tracks, berms or other disturbances.

When a plot location was reached, a plot-midline azimuth was selected from
a random number table, and a sighting compass was used to establish the plot
midline on this azimuth. Each plot was given an identifying number.

Each LCTA plot consisted of a belt 6 m wide and 100 m long, extending 3 m
on either side of the plot midline. At every plot location, a midline was laid
out using a 100-m tape. Each midline was permanently marked with steel
stakes. At m on the midline, a 1.27-cm pipe (inside diameter) 46 cm long was
driven into the ground, and a steel stake 63 mm in diameter and 46 cm long was
dropped into it. At 25 m, 50 m, 75 m, and 100 m on the tape, another 46-cm
steel stake was driven into the ground.

Along the midline tape, a vegetation transect was sampled by dropping a
1-m steel rod marked in decimeters, to the ground at 1-m intervals. The type
of ground cover (or lack thereof) encountered by the pin at each drop-point
(ordnance, gravel, rock, lichen, moss, grass litter, forb litter, shrub litter,
dead wood, fecal material, plant species) was recorded; if the pin came to rest
in a vehicle or mammal track or trail, or a mammal mound or burrow, this was
recorded as well. At each sampling point, above-ground hits of litter, living
plants, and dead wood on the steel rod were recorded, along with hit-height in
decimeters. Litter was identified as grass, forb, or shrub litter. Living
plants were identified to species and recorded by genus-species acronym.

If one or more shrubs taller than . 1 decimeter were encountered on the
plot, a belt transect was sampled. Shrubs were identified to species, and the
location of each shrub in the belt was recorded (for example, 26 m, 1.2 m
left), as well as its height in decimeters. Dead shrubs (snags) were recorded
only if their height was equal to or greater than 1 m. However, the presence
of shorter snags on a plot was noted. If the total number of shrubs of a
species was fewer than 100 in the 6-m x 100-m belt, a 6-m wide belt was sampled
for that species. If the total number of shrubs in a 6-m wide belt was greater
than 100 but fewer than 200, a 4-m belt was sampled for that species, and if
there were 200-400 shrubs per 6 x 100-m belt, a belt 2 m wide was sampled. If
there were more than 400 shrubs of a species present in a 100 x 6-m belt, a 1-m
belt was sampled for that species. Plots having different numbers of several
species thus had several belt transects of different widths. Belt transects
were always of equal width on each side of the 100-m midline. Belt width and
shrub locations were measured using a collapsible, calibrated 6-m range pole,
and shrub heights were measured using a steel rod calibrated in decimeters.

Plant species not previously collected on the OTA, if encountered in the
6-m x 100-m plot, were collected and pressed. Evidence of livestock grazing,
troop bivouacing, vehicle tracking, fire, or other disturbance occurring on the
plot was noted.

143

On each plot at 25 m and 75 m, 1 m to the right of the tape (facing the
100-m rod) , 1-quart soil samples were collected. The soil samples were sent to
the Soil Conservation Service National Soils Laboratory in Lincoln, Nebraska,
for analysis. At m, 50 m, and 100 m on the plot midline, slope length and
gradient were recorded using a measuring rod and a clinometer. Color
photographs were taken of the view down the midline of each plot and of
approaches to each plot, using a 35-mm camera.

In addition, 10 LCTA plots were established as special-use plots, in order
to monitor small-patch or unique habitats or high-disturbance areas. A more
detailed explanation of the LCTA methods can be found in the U.S. Army Land
Condition Trend Analysis (LCTA) Field Methods Manual (U.S. Army Corps of
Engineers 1987).

Small mammal and jackrabbit monitoring

Sixty LCTA plots of the 200 were selected by unstratif ied random means for
snap trapping mammals. Beginning in early July and ending on 1 September, 57
of these plots were trapped for small mammals by William Whitworth and Jeff
Courson of the U.S. Army Corps of Engineers Research Laboratory. Three of the
60 preselected locations were not sampled because they fell inside the core
artillery impact area. Three special-use plots, 91 (rocky drainage), 92 (rocky
rim), and 113 (old sheep bedding ground in winter fat) were also sampled.

On each mammal-trapping plot, 50 Museum Special (Woodstream Corporation,
Lititz, Pennsylvania) snap traps baited with a peanut butter-oatmeal mixture
were set at 2-m intervals in a straight line 5 m to the right (facing the 100-m
rod) of the LCTA plot midline. Traps were set in late afternoon and checked
the following morning. Each plot was trapped for 2 consecutive nights for a
total of 100 trap-nights per plot. Species, sex, plot number, and date of
capture were recorded for each animal caught.

Abundance of black- tailed jackrabbits ( Lepus californicus ) in the OTA was
monitored in 1989 using the method of Smith and Nydegger (1985).

Townsend's ground squirrel and badger monitoring

In June and July, 37 of the 57 mammal-trapping locations plus 3
special -use plots (91, 92, and 113) were censused for Townsend's ground
squirrel ( Spermophi lus townsendii , hereafter TGS) and badger (Taxidea taxus )
burrows. Year-active burrows of both Townsend's ground squirrels and badgers
were distinguished from year-inactive burrows by the presence of one or more of
the following: tracks, claw marks, freshly excavated earth, trails from the
burrow entrance beaten through current-year vegetation, feces, exposed flexible
rootlets in the burrow entrance, and observation of an animal using the burrow.

TGS and badger burrow censusing was done by counting every year-active and
year-inactive burrow in a belt plot 5 m x 400 m. Each censusing plot began at
the LCTA-plot 0-m stake and continued along the plot midline for 100 m, then on
in a straight line for an additional 300 mm. While sampling a plot, the pole
technician carried a 5-m TGS censusing pole down a 400-m straight line. The
TGS censusing pole has chains or rods hanging from each end; the pole can be

144

lowered, and these can be dragged along the ground to mark the boundary when
there is a question of whether a burrow is inside or outside the belt
boundary. One or more recording technicians followed the pole technician. The
recorder (s) searched the belt for burrows as the pole carrier walked slowly
along the line ahead, recording all TGS burrows and badger burrows and noting
whether they were active or inactive.

RESULTS
Vegetation/soils monitoring

In 1989, 210 LCTA plots were established and surveyed in the Orchard
Training Area, 200 regular or "core" plots and 10 special -use plots. Sixty-six
of these plots fell inside the boundary of the artillery impact area. No plots
were established inside the boundary of the core artillery impact area. One
hundred and eighteen of the 210 plots were located north of the BLM Sunnyside
Winter/Sunnyside Spring-Fall grazing drift fence. LCTA plot data collected are
on file in the office of the U.S. Army Corps of Engineers Research Laboratory,
Champaign, Illinois, and at the Idaho Army National Guard Department of
Engineering, Gowen Field, Boise, Idaho. Plant species collected or recorded in
the OTA in 1989 are listed in Appendix 1; herbarium specimens have been
archived in the herbarium of Colorado State University, Fort Collins,
Colorado. Analysis of OTA soil samples by the U.S. Soil Conservation Service
National Soils Laboratory is now underway.

Small mammal and .jackrabbit monitoring

The preliminary mammal survey resulted in the capture of the following
species in the Orchard Training Area: Lepus californicus . Sylvilagus
nuttallii, Eutamias minimus . Dipodomys ordii , Onychomys leucogaster ,
Perognathus parvus . Peromyscus maniculatus , and Neotoma cinerea .
Mammal -capture data are on file in the office of the U.S. Army Corps of
Engineers Research Laboratory, Champaign, Illinois, and at the Idaho Army
National Guard Department of Engineering, Gowen Field, Boise, Idaho.

Black-tailed jackrabbit censusing was done as part of the BLM Snake River
Birds of Prey Area monitoring effort, and OTA results are included in the
section of this annual report that reports jackrabbit abundance in the Snake
River Birds of Prey Area.

Townsend's ground squirrel and badger monitoring

Forty burrow-count transects were sampled during June and July 1989 (Table

145

Table 1. 1989 Townsend's ground squirrel and badger burrow counts.

Plot

Active

Inactive

Active

Inactive

Comments

TGS

Badger

IMPACT AREA

108

IMPACT AREA

SPECIAL USE

IMPACT AREA

106

108

IMPACT AREA

109

IMPACT AREA

111

112

113

SPECIAL USE

115

116

117

118

119

120

121

122

123

124

125

130

IMPACT AREA

132

146

LITERATURE CITED

Smith, G.W. , and N.C. Nydegger. 1985. A spotlight, line- transect method for
surveying jack rabbits. J. Wildl. Manage. 49:699-702.

U.S. Army Corps of Engineers. 1987. U.S. Army land condition trend analysis
(LCTA) Field Methods. 38pp.

U.S. Soil Conservation Service. 1960. Soil Survey of Ada County Area, Idaho.
U.S. Dep. of Agric, Soil Conserv. Serv. in cooperation with U.S. Dep. of
the Inter., Bur. of Land Manage., Univ. of Idaho Coll. of Agric, Idaho
Agric. Exp. Stn. ; Idaho Soil Conserv. Comm. 327pp. plus maps.

U.S. Soil Conservation Service. Undated. Soil Survey of Elmore County Area,
Idaho. U.S. Dep. of Agric, Soil Conserv. Serv. in cooperation with U.S.
Dep. of the Inter., Bur. of Land Manage., Univ. of Idaho Coll. of Agric,
Idaho Agric. Exp. Stn.; Idaho Soil Conserv. Comm. Unpublished manuscript
and maps.

147

TITLE: Establishment of Monitoring Studies on Greenstrip Projects in
the Snake River Birds of Prey Area.

INVESTIGATOR: Mike Pellant, BLM Idaho State

OBJECTIVES:

1. Provide information to select drought affected greenstrips for
reseeding .

2. Quantify the effects of pretreatment vegetation, seedbed
preparation/seeding techniques and year of treatment on initial
establishment of greenstrip species.

3. Establish baseline monitoring studies to evaluate .long term
effects of grazing, climate and natural succession

on persistence of seeded species and reestablishment of
unwanted species.

INTRODUCTION

In 1984, Idaho BLM initiated a fire presuppression program, termed
greenstripping to reduce the rate of spread of wildfires. Greenstripping
involves establishing strips of fire-resistant vegetation at strategic
locations to protect important resource values and reduce fire suppression and
rehabilitation costs. Greenstripping is similar to past efforts to seed
vegetative fuel breaks; however an intensive research and technology transfer
effort has been added to the current program.

Greenstripping has been employed within the Snake River Birds of Area
(SRBOPA) to protect shrub stands from wildfires and to reduce fire frequency,
thereby accelerating shrub recruitment in previously burned areas (Kuna Normal
Fire Rehabilitation and Greenstripping Plan 1987).

A thorough monitoring and evaluation program is being implemented to
monitor progress in meeting the program objectives. Inspections of greenstrip
projects are conducted throughout each growing season to record "greenness" and
vigor of seeded species.

Pace frequency transects were established on 4 greenstrip projects within
the SRBOPA in 1987. Selected study sites were resampled in 1988; all studies
were reread in 1989. Study locations and data are on file in the Boise
District's Bruneau Resource Area and with the author.

Intensive monitoring studies were established on 2 greenstrip projects in
1988 (Monsen and Pellant 1988) . Density, frequency and cover measurements were
recorded on 5 replicated transects at each study site.

The studies described in this report were established on 8 greenstrip
projects in the SRBOPA (Table 1). Field work was accomplished by the Boise
District's monitoring crew during the summer of 1989.

148

Table 1. Study

site locations

, year greenstri

p establ

ished,

and site

preparation/seeding techniques

used

to establish selected SRBOPA

greenstrip projects.

Project

Legal Locatior

Site Pn

& Seed Tec

Name

Year

SEC

1/4

Veg.2

Burn

Disk

Drill

D-chn

Milepost 96

1984

BRTE

Bennett/Interst . :

1984

Site 1

ARTR

Site 2

BRTE

RRMP:

1986

Site 1

BRTE

Site 2

BRTE

Site 3

BRTE

Dorsey:

1986

Site 1

BRTE

Site 2

BRTE

Site 3

ARTR

Site 4

ATCO

Site 5

ATCO

Site 6

ATCO

XA«

Site 7

BRTE

Site 8

BRTE

Sites 9-11

BRTE

Site 12

ATCO

Milepost 101:
Site 1

1987

BRTE

Site 2

BRTE

Cinder Cone:

1987

Site 1

ARTR

Site 2

ARTR

Site 3

ARTR

Crater II:

1987

Site 1

BRTE

Site 2

BRTE

Site 3

BRTE

White Sage:

1987

Site 1

ARTR

Site 2

ARTR

Initial Point:

1988

Site 1

ARTR

Site 2

ARTR

Site 3

BRTE

Legal Location 1 - Study site locations are presented as T. (Township), R.
(Range), Sec. (Section), and 1/4 (quarter section study site is in).

Veg. 2 - Indicates dominant vegetation occurring on the study site immediately
preceding greenstrip implementation.

Site Prep, and Seed Tech. 3 - Provides information on whether site was burned
during the year that greenstripping was done and the type of mechanical
equipment used to complete seedbed preparation and/or seeding. "D-chn"
indicates that a disk chain was used to prepare the seedbed and to distribute
seed.

XA 4 - "A" indicates that a cultipacker was used to firm the seedbed prior to
drill seeding.

149

METHODS

The "Freqdens" (acronym for frequency-density) technique was developed
specifically for measuring initial establishment and persistence of seeded
species on Idaho greenstrip projects. This technique is relatively quick
and inexpensive thereby stretching limited funds available for greenstrip
monitoring. Replicated transects are permanently located to facilitate
statistical analysis.

Study site locations are selected by resource specialists on
representative portions of selected greenstrips. A reference post is
installed, and a 30-m (100-foot) baseline is established. Five belt
transects are randomly established within 6-m (20-foot) divisions
perpendicular to the baseline. Ten 50 x 50-cm plots are read at 1.5-m
(5-foot) intervals (5, 10, 15... 50 feet) on each belt transect.

Three sampling techniques are utilized to collect data on each belt
transect:

No. of samples Total No. of samples
Sampling technique per belt transect per study site

Nested frequency 10 50

Density 10 50

Point Cover 40 200

Density and nested frequency data are recorded for all seeded species
(seedling and mature plants are recorded separately). Dominant annual
species are recorded separately; less common annual species are combined
as "annual grasses" or "annual forbs".

Point cover data are recorded from 4 points on the sampling frame
within the categories of bare ground, litter, vegetation and rock. One
1/100 acre plot is also established at each point where the belt transect
crosses the baseline. All seedling and mature shrubs occurring in the
plot are counted and recorded separately.

The final study component is a "Vegetation Condition Worksheet" which
documents (qualitatively) vigor, height, utilization, phenology and
relative greenness of all seeded and dominant annual and perennial
species. Two general view and one close-up photograph are also taken at
each study site.

RESULTS

All data collected during the 1989 field season have been summarized
and are available for inspection at Boise District or with the author.
Data are currently being analyzed at the Range Sciences Department of the

150

University of Idaho. Findings from this analysis will be incorporated
into a presentation for the 1990 Society for Range Management Annual
Meeting in Reno, Nevada (February 1990).

A BLM progress report on findings from greenstripping research and
evaluations, including these studies, will be prepared in 1990. This
report will focus on the desirability of burned versus unburned seedbeds,
effectiveness of three types of equipment treatments (rangeland drill,
disk and rangeland drill and disk chain) , year of treatment and
pretreatment vegetation community (Table 1).

ACKNOWLEDGMENTS

I thank Jack LaRocco and the Boise District Monitoring Crew for their
assistance in fine tuning the study procedures and completing the field
work. They toiled under harsh environmental conditions and efficiently
and accurately completed these studies.

LITERATURE CITED

Monsen, S., and M. Pellant. 1988. An evaluation of selected

greenstripping and winterfat seeding projects in the Snake River
Birds of Prey Area. Pages 141-150 in K. Steenhof , ed. Snake River
Birds of Prey Research Proj. Annu. Rep., Bur. Land Manage. Boise,
Idaho.

U. S. Department of the Interior. 1987. Kuna normal fire rehabilitation
and greenstripping plan. Bureau of Land Manage. Boise Dist., Idaho.
71pp.

I'KiN'l'lNU OPl'MCili I yy<) ■ --'/y J -789/23047

151

BLM LIBRARY
SC-653, BLDG. 50
DENVER FEDERAL CENTER
p. 0. BOX 25047
DENVER, CO 80225-0047

SuiSa£Jiiiaih;Siiii

■ ■ ' '. . ' ■■.■■■

NORTH HASKiiT'%
MOUNTAIN HOME ID 8 3
' USA