Plumas Lassen
Study
2008 Annual
Report
USDA Forest Service, Pacific Southwest Research Station, Sierra Nevada Research Center
1731 Research Park Drive, Davis, CA 95618 530-759-1700
Table of Contents
Introduction…………………………………………………………………..3
Purpose of Study……………………………………………………..4
Objectives of Study…………………………………………………..4
Specific Management Questions to be Addressed…………………...5
Summary…………………………………………………………….14
Response Variable Modules
Chapter 1: Fuels and Fire Module…………………………………..16
Chapter 2: Vegetation Module………………………………………28
Chapter 3: Small Mammal Module………………………………….32
Chapter 4: Terrestrial Bird Module…………………………………104
Chapter 5: Spotted Owl Module…………………………………….188
2
Introduction
The Pacific Southwest Region and the Pacific Southwest Research Station agreed in 2002
to jointly develop and fund an administrative study to fill management information needs
concerning the relationship between management-caused changes in vegetation and their
effects on spotted owl habitat and population dynamics. The detailed discussions
explaining how this program was started is provided in previous Annual Reports. Copies
of previous Annual Reports for this program are available on the Sierra Nevada Research
Center web site (www.fs.fed.us/psw/programs/snrc) or upon request.
This is the seventh such Annual Report that we have compiled. The primary purpose of
this is to provide a periodic synopsis of what we have been learning so all interested
parties can remain abreast of the progress. Research products resulting from this effort
will be disseminated as they are ready and this will vary from module to module, project
to project, and from year to year. We expect that there will be a continuous flow of
findings documented primarily with publications in both refereed journals and other
publication outlets. The cadre of scientists, support staff, students, and others contributing
to this effort will also be making oral presentations and providing other kinds of outreach
materials to help inform interested parties and our peers on the results of this work.
We provide some review information here to reinforce the intent of our work. This
background information provides a general overview on the purpose of this research
program and helps set the context for the report. We have had to remind many interested
parties and in particular our own program administrators that we embarked on the project
virtually from square one. A project of this magnitude and ambition is difficult to initiate
under the best of circumstances. When a research program begins work in a new area,
addressing large geographic areas with complex questions on a busy landscape that is
already subject to many other demands, it is not easy to establish all the field activities
and produce results quickly.
However, we now believe we have emerged from the initiation phase and we have
collected an impressive amount of information. Many publications are in development
and we expect to provide useful information in the immediate future. Of course much of
our research purpose depends on forest management treatments to be put in place and
then observe short and even long term response to those treatments. Such treatments are
now being executed in some locations and thus some of our potentially most significant
work has only recently begun. Observations of response after treatments will logically
take place in the ensuing years. If funding can be sustained we intend to continue to
follow up with further data collection, field observations and insights addressing the
questions we have posed.
We recognize that response of different elements of the forest can occur immediately
after treatments however it is also possible that response can occur slowly and not be
recognized for some period of time depending on the response variable of interest.
Alternatively it is also possible that some response variables exhibit a notable initial
response and then return to a state similar to that of before the treatments. Thus we
believe it is prudent to look at a fairly long period of post treatment response if possible,
3
even if funding limitations require scheduling follow-up work in stages over time with
periods of inactivity.
Purpose of the Study
This study is interdisciplinary by design, examining at least five groups of response
variables (spotted owls, small mammals, terrestrial birds, vegetation, and fuels
conditions) through collaboration between researchers of the USDA Forest Service
Pacific Southwest Research Station (PSW) and cooperators from the Universities of
California, Berkeley and Davis, and the PRBO Conservation Science. The study
addresses some of the most significant uncertainties that confound management decisions
in the Sierra Nevada today, including in the HFQLG Pilot Project Area. How do oldforest-dependent species respond to vegetation management over space and time? Do
fuels management approaches effectively address fuels loadings without negatively
affecting species viability? How effective are landscape level fuels management
strategies in modifying fire behavior and reducing the extent and severity of wildland
fire? These and related questions are the focus of the work being done in this study.
Objectives of Study
The original overarching objective of this proposed research was to address an array of
related ecological questions in a coordinated, integrated effort, thereby providing
empirical data to inform future management decisions. The landscape scale of this design
was both the driving force addressing the key questions as well as the largest impediment
to successful construction of a scientifically credible experimental design and
implementation in the field. Our research team believes that assessing many of the key
elements of forest ecosystems should be done over larger spatial and temporal scales than
has typically been investigated in past research. The important difference we are
investigating is the response to changes in forest structure and composition over space
and time rather than simply site specific and immediate response. We believe this
difference is especially relevant to forest management practices that are designed for
large landscapes, executed over relatively long time frames, such as landscape level fuels
treatment strategies.
This research program is designed to address the three principal issues described
below. These issues are specifically addressed through research questions and
attending investigational approaches tailored for five different research components of
this research program. These specific questions are detailed in the individual study
plans for each module. Here we simply highlight the main objectives of the integrated
research program and summarize the primary research questions that we plan to
pursue.
• Wildland Fire Behavior and Protection. How do landscape level fuels and
silvicultural treatments affect potential fire behavior and effects? Are specific
combinations of defensible fuel profile zones (DFPZs) and subsequent
individual tree selection or area treatments to thin the matrix effective in
reducing the extent and severity of wildland fires? Are realized fire management
4
benefits consistent with hypothesized results in reducing fire risk and altering
fire behavior?
• Landscape Dynamics. How do combinations of DFPZs, subsequent
individual tree selection or area treatments to thin the matrix, group selection,
riparian protection standards, and species-specific protection measures affect
landscape dynamics such as forest structure, composition, and succession at
multiple scales of space and time?
• Species Viability. Induced by a forest management regime, how will oldforest dependent species, particularly the California spotted owl and its prey
base comprised of various species of small mammals, respond to changes in
vegetation composition, structure, and distribution over space and time? How
is response to treatments manifested at the individual and population levels of
biological organization?
Below we provide brief summary statements that capture the essence of the questions
we are pursuing under this research agenda. Once again we direct you to the detailed
study plans for further information on each module of this research program.
The specific management questions that are being addressed within the five
different research components are:
Fuels and Fire Module
1. Current conditions: measurement of vegetation and fuels at the
landscape scale
2 Fire modeling: how might current conditions (above) affect fire behavior
and effects?
3. Effects of treatments: how might landscape-scale treatments change fire
behavior and effects (as measured by using simulation programs such as
FlamMap)?
4. Fire and habitat model integration (how can we address fuels
management objectives in ways compatible with sensitive species
conservation?).
Vegetation Module
1. What are the effects of canopy reduction due to thinning treatments on
understory microclimate and shrub cover? How do we accurately
measure changes in canopy cover to meet management prescriptions?
2. What are the appropriate ecological conditions to induce regeneration of
shade-intolerant conifer species?
3. How does ecosystem resilience to forest harvesting, particularly group
selection silviculture, vary across landscape gradients of precipitation and
soil type?
5
Small Mammal Module
1. What are the habitat associations of the different taxa of small mammals
found in coniferous forests in the northern Sierra Nevada (objective of
developing refined yet functional models of habitat associations)? What is
the relative abundance and distribution of these taxa with respect to forest
structure and composition?
2. Estimate values of the demographic parameters (for example,
population size, reproductive output, survivorship, and mortality rates) of
these taxa.
3. Estimate values for spatial patterns (for example, home range area and
configuration) for these taxa.
Bird Community Module
1. Do current forest management practices promote an ecologically
balanced forest ecosystem that supports sustainable populations of the
breeding bird community over time?
2. What are the critical local-scale habitat components and landscape-scale
composition elements that should be managed for in order to sustain the
avian community over time (20 to 50 years)? Can we predict species
composition, abundance, and distribution in response to future landscape
treatments?
3. How do, or will, a suite of avian species that are associated with a wide
range of forest conditions respond to fuels treatments, at the local and
landscape scales in the short (one to five years) and long term (five to 20
years)?
4. Do Spotted Owl protected activity centers provide high quality habitat
for the broader avian community? What are the differences in the avian
community composition within owl territories compared to the
surrounding landscape?
California Spotted Owl Module
1. What are the associations among landscape fuels treatments and CSO
density, distribution, population trends and habitat suitability at the
landscape-scale?
2. What are the associations among landscape fuels treatments and CSO
reproduction, survival, and habitat fitness potential at the core area/home
range scales?
3. What are the associations among landscape fuels treatments and CSO
habitat use and home range configuration at the core area/home range
scale?
4. What is the population trend for CSOs in the northern Sierra Nevada
and what factors account for variation in population trend?
5. Are barred owls increasing in the northern Sierra Nevada, what factors
are associated with their distribution and abundance, and are they
associated with reduced CSO territory occupancy?
6. Does West Nile Virus affect the survival, distribution and abundance of
California spotted owls in the study area?
6
Progress to Date
Given that we have completed a seventh year of work we are beyond the initiation phase
and many findings are beginning to take shape. Some results, based on primarily
pretreatment data, are crystallizing and findings are being reported. Some of the work
described here includes activities from other locations but are potentially relevant to the
Plumas and Lassen National Forest landscape, thus they are included in this summary. A
preliminary list of completed and anticipated publications is summarized below:
FIRE AND FUELS MODULE
Menning, K.M., and S.L. Stephens. (2008: draft complete, being submitted to Landscape
Ecology). "Potential forest fire behavior as a function of three weather scenarios and two
landscape fuels treatments based on a fuels and vegetation landscape derived from finegrain IKONOS satellite imagery, Sierra Nevada (USA)." Draft being submitted to
Landscape Ecology.
Menning, K.M., and S.L. Stephens. 2007. Fire Climbing in the Forest: a semiqualitative, semi-quantitative approach to assessing ladder fuel hazards, Western Journal
of Applied Forestry 22(2): 88-93.
Menning, K. M. and S. L. Stephens (2006). Modeling Landscape Fire Behavior and
Effects in the Northern Sierra Nevada. 3rd International Fire Ecology and Management
Congress, San Diego, CA.
Menning, K. M. and S. L. Stephens (2006). Landscape-scale Fire Risk Wildlife Habitat
Considered Jointly. 21st Annual Symposium of the United States Regional Chapter of the
International Association for Ecology (US IALE), San Diego, CA.
Menning, K. M. and S. L. Stephens (2006). Assessing Ladder Fuels in Forests. 3rd
International Fire Ecology and Management Congress, San Diego, CA.
Menning, K.M., and S. L. Stephens (2005) Fire rising in the forest: Ladder fuel hazard
assessment using a mixed qualitative and quantitative approach, Ecological Society of
America, August 7-12, 2005, Montreal Canada. (Abstract attached to end of report).
Menning, K. M. and S. L. Stephens (2005). (Invited speaker:) Linking fire and wildlife
habitat in California: Spectral entropy canopy diversity analysis. UK Centre for Ecology
and Hydrology, Monks Wood, Cambridgeshire, England, UK. November 21, 2005.
Menning, K. M. and S. L. Stephens (2005). (Invited speaker:) Spatial Ecological Links
Between Fire, Forests and Habitat in the Plumas-Lassen Administrative Project.
Geographic Information Centre Seminar: City University, London, London, England UK.
November 22, 2005.
7
Menning, K. M. and S. L. Stephens (2005). (Invited speaker:) Forest Structural Diversity:
Spectral Entropy Canopy Diversity Analysis. Swiss Federal Institute for Forest, Snow
and Landscape Research, Birmensdorf, Switzerland. December 5, 2005.
Planned Publications
Menning, K. M. and S. L. Stephens. "Spectral Entropy Canopy Diversity Analysis
(SpECDA) used to Assess Variability in Forest Structure and Composition" to be
submitted to Photogrammetric Engineering and Remote Sensing.
Menning, K. M., S. L. Stephens, J. Keane, D. Kelt, and others. "Integrated modeling of
fire and California Spotted Owl habitat conditions given different weather and landscape
treatment scenarios" To be submitted to a journal mutually agreed upon.
Menning, K. M. and S. L. Stephens. "Fire Behavior and Effects as a Result of Defensible
Fuel Profile Zones" To be submitted to International Journal of Wildland Fire.
Menning, K. M. and S. L. Stephens. "Landscape Forest Variability across the Northern
Sierra Nevada" To be submitted to Landscape Ecology.
VEGETATION MODULE
Publications (in Review)
Bigelow, S. W., and S. A. Parks. Predicting altered forest connectivity due to group
selection silviculture. In review at Landscape Ecology.
Bigelow, S. W., M. P. North, and W. R. Horwath. Resource-Dependent Growth Models
for Sierran Mixed-Conifer Saplings. In review at the Open Forest Science Journal.
Papers planned
Understory light in mixed-conifer forest: effects of fuels treatments and group selection
silviculture.
Seth Bigelow, Carl Salk, Malcolm North.
Summary: We show the effects of thinning to various cover levels on the understory light
environment, and infer probable effects on tree species composition (particularly
comparing shade-tolerant and intolerant species) with data on seedling height growth in
response to light.
Fuels Treatment and Group Selection Effects on Fire Climate.
Malcolm North, Seth Bigelow.
Summary: Air temperature and humidity, wind speed, and fuel moisture, duff and
mineral soil moisture were measured for 3 yrs prior thinning to 1 yr after (end of 2008
season). Did treatments affect fire climate and soil moisture dynamics?
8
Group selection harvest impacts in an ecotonal environment.
Seth Bigelow, Malcolm North.
Summary: We measured soil water dynamics in natural gaps, group selection openings,
and closed canopy stands in patchy east-side forest. Natural gaps had rocky soil which
prevented tree establishment. Despite large differences in canopy cover inside and
outside group openings there were no measurable differences in soil water dynamics.
SMALL MAMMAL MODULE
Theses
Coppeto, S. A. 2005. Habitat associations of small mammals at two spatial scales in the
northern Sierra Nevada, California. M.S. Thesis, University of California, Davis, 39 pp.
Innes, R.J. 2006. Habitat selection by dusky-footed woodrats in managed, mixed-conifer
forest of the northern Sierra Nevada. M.S. Thesis, University of California, Davis, 31 pp.
Smith, J.R. In Prep. Home range and habitat selection of the northern flying squirrel
(Glaucomys sabrinus) in northeastern California. M.S. Thesis, University of California,
Davis. Winter 2009.
Peer-reviewed
Coppeto, S. A., D. A. Kelt, D. H. Van Vuren, J. A. Wilson, S. Bigelow, and M. L.
Johnson. 2006. Habitat associations of small mammals at two spatial scales in the
northern Sierra Nevada. Journal of Mammalogy 87:402-416.
Innes, R. J., D. H. Van Vuren, D. A. Kelt, M. L. Johnson, J. A. Wilson, P. A. Stine.
2007. Habitat selection by dusky-footed woodrats in managed, mixed-conifer forest of
the northern Sierra Nevada. Journal of Mammalogy 88(6): 1523-1531.
Innes, R. J., D. H. Van Vuren, D. A. Kelt. 2008. Characteristics and use of tree houses by
dusky-footed woodrats in the northern Sierra Nevada. Northwestern Naturalist 89(2):
109-112.
Wilson, J. A., D. A. Kelt, and D. H. Van Vuren. 2008. Home range and activity of
northern flying squirrels (Glaucomys sabrinus) in the Sierra Nevada. Southwestern
Naturalist.
Wilson, J. A., D. A. Kelt, D, H, Van Vuren, and M. Johnson. 2008. Population
dynamics of small mammals in relation to cone production in four forest types in the
northern Sierra Nevada, California. The Southwestern Naturalist 53(3): 346-356.
9
Submitted
Mabry, K.E., and Wilson, J. A. Submitted. Trapping rodents in a cautious world: the
effects of disinfectants on trap success. American Midland Naturalist.
In Preparation
Innes, R. J., D. H. Van Vuren, D. A. Kelt, M. L. Johnson, and J. A. Wilson. In Prep.
Spatial organization of the dusky-footed woodrat (Neotoma fuscipes) in mixed-conifer
forests of the northern Sierra Nevada. Journal of Mammalogy. Winter 2009.
Coppeto, S. A., D. A. Kelt, D. H. Van Vuren, J. Sullivan, J. A. Wilson, and N. Reid. In
Prep. Different scales tell different tales: niche conservatism vs. niche differentiation in
chipmunks in the northern Sierra Nevada. To be determined. Spring 2009.
Jesmer, B. J., D. A. Kelt, and D. H. Van Vuren. In Prep. Asociality: home range and
dispersal of golden-mantled ground squirrels (Spermophilus lateralis). To be determined.
Spring 2009.
Presentations
Coppeto, S. A., D. A. Kelt, J. A. Wilson, D. H. Van Vuren, and M. L. Johnson. 2004.
Habitat selection by small mammals in the northern Sierra Nevada, California. Poster to
the American Society of Mammalogists, Annual Meeting, Arcata, CA.
Coppeto, S. A., D. A. Kelt, D. H. Van Vuren, J. A. Wilson, S. Bigelow, and M. L.
Johnson. 2005. Spatial scale and habitat use of small mammals in the northern Sierra
Nevada, California. Poster to the American Society of Mammalogists, Annual Meeting,
Springfield, MO.
Innes, R. J., D. H. Van Vuren, J. A. Wilson, D. A. Kelt, and M. B. Johnson. 2004.
Factors affecting the distribution and use of dusky-footed woodrat (Neotoma fuscipes)
houses. Poster to the American Society of Mammalogists, Annual Meeting, Arcata, CA.
Innes, R. J., D. H. Van Vuren, J. A. Wilson, D. A. Kelt, and M. B. Johnson. 2005. Space
use and social organization of dusky-footed woodrats (Neotoma fuscipes) in mixedconifer forests of the northern Sierra Nevada. Poster to the American Society of
Mammalogists, Annual Meeting, Springfield, MO.
Innes, R. J., D. H. Van Vuren, D. A. Kelt, M. B. Johnson, J.A. Wilson. 2006. Habitat
relations of dusky-footed woodrats (Neotoma fuscipes) in mixed-conifer forests of the
northern Sierra Nevada. Poster to the American Society of Mammalogists, Annual
Meeting, Amherst, MA.
10
Smith, W. 2006. Ecology of Glaucomys sabrinus: habitat, demography, and community
relations. Presentation to the American Society of Mammalogists, Annual Meeting,
Springfield, MO.
Wilson, J.A., and K.E. Mabry. 2005. Trap disinfection to reduce Hantavirus risk: does it
also reduce small mammal trapability? Presentation to the American Society of
Mammalogists, Annual Meeting, Springfield, MO.
Wilson, J. A., D. A. Kelt, and D. H. Van Vuren. 2005. Effects of maternal body
condition on offspring dispersal in golden-mantled ground squirrels (Spermophilus
lateralis). Presentation to the American Society of Mammalogists, Annual Meeting,
Springfield, MO.
Wilson, J. A., D. A. Kelt, and D. H. Van Vuren. 2005. Effects of maternal body
condition on offspring dispersal in golden-mantled ground squirrels (Spermophilus
lateralis). Presentation to the IX International Mammalogical Conference, Sapporo,
Japan.
Wilson, J. A., D. A. Kelt, and D. H. Van Vuren. 2006. Home range and activity of the
northern flying squirrel (Glaucomys sabrinus) in the northern Sierra Nevada. Poster to
the American Society of Mammalogists, Annual Meeting, Amherst, MA.
TERRESTRIAL BIRD MODULE
Publications in Preparation
Landscape effects on songbird abundance in the Northern Sierra – submitted March 2008
– Journal of Wildlife Management.
Avian community composition in the context of Spotted Owl management in the Sierra
Nevada – submitted April 2008 – Forest Ecology and Management.
Habitat use and productivity of two shrub dependent bird species in clear cut plantations
in the Sierra Nevada – submitted spring 2008 – The Condor.
Short-term response of the avian community to Aspen enhancement timber harvest
treatments – submitted summer 2008 – Restoration Ecology.
Presentations
Listening to the Birds: An Ecosystem Approach to Sierra Nevada Management – invited
oral presentation – Society of American Foresters – California Chapter Annual Meeting –
1/31/2009 - Sacramento, CA.
11
Avian Monitoring in the HFQLG Area – 2007. Plumas-Lassen Study Symposium –
3/28/2008 - Quincy, CA.
Using Birds to Guide National Forest Management in the Sierra Nevada – oral
presentation – International Partner’s in Flight Conference – 2/16/08 – McAllen, TX.
Managing Disturbance Associated Habitats for Birds in the Sierra Nevada – invited oral
presentation – Region 5 Forest Management Conference – 2/6/08 – Reno, NV.
Managing Aspen Habitat for Birds in the Sierra Nevada– invited oral presentation at:
Aspen Delineation Project – Aspen Workshop – 9/12/2007 – Lassen National Forest.
Ecological Significance of Lake Almanor Meadows to Birds – oral presentation at
Almanor Basin Watershed Advisory Committee Workshop on meadow management –
8/7/07 - Chester, CA.
Using Birds to Guide Forest Management in the HFQLG Area: Results from 2002 – 2006
– invited oral presentation – USFS Region 5 biologist conference – 5/23/07 - Sacramento,
CA & PLAS symposium 3/2007.
Other Outreach
“Birds in the Park” – presentation on managing coniferous forest for birds and bird
banding demonstration in collaboration with Lassen Volcanic National Park – over 200
park visitors participated 7/20/08.
Sierra Institute Bird Field Tour – Lead field tour to discuss the importance of meadows to
Sierra Nevada birds including presentation and banding demonstration. – 6/28/2008.
Bird Banding Field Trip – coordinated outreach field trips with the Lassen National
Forest to view bird banding and discuss the use of birds as indicators in forest
management, PLAS study, and PRBO –7/16/2008.
Participated in several Forest Service Field Trips on the Almanor and Eagle Lake Ranger
Districts. 2008.
“Birds in the Park” – presentation on managing coniferous forest for birds and bird
banding demonstration in collaboration with Lassen Volcanic National Park – over 200
park visitors participated 7/22/07.
Sierra Nevada Conservancy Field Trip – 5/1/2007 – Westwood, CA.
Aspen Workshop – invited to participate in the event co-sponsored by the Lassen
National Forest, Aspen Delineation Project, and Sierra Forest Legacy – 9/13/2007.
Led Plumas Audubon Society Field Trip – 10/3/2007 – Chester, CA.
12
Bird Banding Field Trip – coordinated outreach field trips with the Lassen National
Forest to view bird banding and discuss the use of birds as indicators in forest
management, PLAS study, and PRBO – 7/25/2007, 8/8/2007.
Integration with Management
We provided input to several important Forest Service projects in 2007 in an effort to
integrate our results to help guide forest management in the Sierra Nevada. In addition
we:
1. Updated the “Interactive GIS Project” with 2007 avian monitoring data. This
product can be used by forest planners in the region to determine the
presence/absence or abundance of all species detected in the study area.
2. Updated the Lassen National Forest interactive GIS CD with
presence/absence data of each woodpecker species at every point count station
ever surveyed by PRBO in the district. We also conducted a tutorial of its
application and use with ARD biologist Mark Williams.
3. Continued distribution with positive feedback for our white papers integrating
avian monitoring data into science based recommendations for managing four
important Sierra habitat types for birds.
OWL MODULE
Keane, J.J., J.A. Blakesley, C.V. Gallagher, D.L. Hanson, P.A. Shaklee, and D.W.H.
Shaw. Status and Distribution of the Barred Owl in the Sierra Nevada. To be submitted
to the Condor.
Keane, J.J., J.A. Blakesley, C.V. Gallagher, D.L. Hanson, P.A. Shaklee, and D.W.H.
Shaw. Nest-site habitat characteristics of California spotted owls in the northern Sierra
Nevada. To be submitted to Journal of Wildlife Management.
Keane, J.J., J.A. Blakesley, C.V. Gallagher, D.L. Hanson, P.A. Shaklee, and D.W.H.
Shaw. Landscape nesting habitat characteristics of California spotted owls in the
northern Sierra Nevada. To be submitted to the Journal of Wildlife Management.
Keane, J.J., J.A. Blakesley, J.R. Dunk, and S.A. Parks. Predictive habitat suitability
models of California spotted owls for assessing effects of forest management and fuels
treatments. To be submitted to Ecological Applications or Forest Ecology and
Management.
13
Keane, J.J., J.A. Blakesley, C.V. Gallagher, D.L. Hanson, P.A. Shaklee, and D.W.H.
Shaw. Diets of California spotted owls in the northern Sierra Nevada. To be submitted to
Forest Ecology and Management.
Dunk, J.R., J.J. Keane, and S.A. Parks. Predictive habitat suitability models of northern
goshawks for assessing effects of forest management and fuels treatments in the northern
Sierra Nevada. To be submitted to Ecological Applications or Forest Ecology.
J.J. Keane , J.R. Dunk, and S.A. Parks. Landscape habitat patterns and predictive habitat
suitability models for northern goshawks in the Lake Tahoe Basin, Sierra Nevada. To be
submitted to Journal of Wildlife Management or Forest Ecology and Management.
J.J Keane, J.R. Dunk, and T. Gaman. Nest-site characteristics of northern goshawks in the
southern Sierra Nevada. To be submitted to Condor.
J.J. Keane, B.Woodbridge, and S.A. Parks. Conservations status and distribution of the
northern goshawk in California. To be submitted to the Journal of Biogeography or
Biological Conservation.
J.J Keane and J.R. Dunk. Predictive habitat modeling of California spotted owl and
northern goshawk habitat in the Sierra Nevada. To be submitted to Ecological
Applications.
B. Woodbridge, J.J. Keane, J.R. Dunk, and J. Hawley. Habitat conservation assessment
for northern goshawks in California. To be published as a GTR.
J.J. Keane. Effectiveness of artificial great horned owls for capturing northern goshawks.
To be submitted to the Journal of Raptor Research or Journal of Field Ornithology.
J.J. Keane and B. Woodbridge. Effectiveness of broadcast surveys for detecting northern
goshawks. To be submitted to the Wildlife Society Bulletin.
J.J. Keane, E.B. Jepsen, L.A. Tierney and C.V. Gallagher. Effectiveness of survey
techniques for detecting great gray owls. To be submitted to the Journal of Wildlife
Management.
Summary
This work represents some significant scientific study that has occurred over the last
seven years. Our original expectation was to continue for up to another two years
within the HFQLG Pilot Project area to capture adequate post-treatment data.
However, when we began this study the pilot project was scheduled to end in 2005 and
since then it has been extended twice, now to 2012 to enable the complete pilot project
to be implemented. If funding support persists we will continue to pursue filed work
for perhaps one to two more field seasons. Upon completion of the field work the
remainder of the effort will be devoted to data analysis and reporting.
14
At the conclusion of the pilot project the HFQLG Act requires the Forest Service to
commission a team of scientists to evaluate the pilot project and provide the Forest
Service with guidance on the efficacy of the work and what were the environmental
consequences on the natural resources of the geographic region. The results of these
studies will contribute valuable, objective scientific insights that managers can use to
develop subsequent management direction for the Plumas and Lassen National Forests, as
well as other National Forest lands in the northern Sierra Nevada such as the portions of
the Tahoe National Forest that contain similar ecological conditions. A team, lead by the
Gifford Pinchot Institute, has been assembled for this purpose and they have begun their
work as of the fall of 2007. Our research team will assist and cooperate with the Pinchot
Team in every way we can.
We cannot ignore or deny the fact that designing a credible and useful research program
in this area has been challenging. We want to be clear to all interested parties that the
Pacific Southwest Research Station was asked to become involved in this project and for
the purposes stated in the introduction above and we responded with the intent to provide
as much new scientific learning as would be possible. PSW knew that we would be
entering into efforts that would have many more challenges than research projects
typically encounter. Our goal was to contribute as much as we could to the better
understanding of forest ecosystem response to fuels and other forest management
practices as they are manifested at a landscape scale.
We understand there is some uncertainty and sometimes controversy over how various
forest elements will respond to planned forest management practices. This is likely to be
the case under any chosen management regime. The objective of PSW was to tackle the
difficult scientific challenges derived from the salient management questions. PSW, as a
research organization, remains wholly objective in executing this charge. We have
assembled an excellent team of scientists with the appropriate areas of expertise and we
have done the best we can to design our work to address the important questions. Many
of these questions present significant challenges to experimental design of field ecology
experiments and management constraints further constrain our ability to test questions
with traditional hypothesis testing approaches. We expect to make the most of these
opportunities in advancing our scientific understanding of forest ecosystem response to
management practices.
15
Plumas-Lassen Administrative Study: Fuels and Fire at the Landscape Scale
Annual Report for 2008
3/23/2009
Written by:
Dr. Scott Stephens, Associate Professor of Fire Science
Ecosystem Sciences Division
Department of Environmental Science, Policy, and Management
137 Mulford Hall
University of California, Berkeley, CA. 94720-3114
510-642-7304 e-mail stephens@nature.berkeley.edu
Research Team (all affiliated with my lab at UCB)
Mr. Jason Moghaddas
Dr. Brandon Collins
Dr. Kurt Menning
Dr. Emily Moghaddas
16
2008 Annual Report: Fuels and Fire at the Landscape Scale
Project Background:
Past management activities including fire suppression, timber harvesting, and
livestock grazing have changed the structure and composition of many coniferous forests
in the Sierra Nevada, particularly those that once experienced frequent, low-moderate
intensity fires (Parsons and Debenedetti 1979; SNEP 1996; van Wagtendonk 1996,
Stephens 1998, Stephens and Moghaddas 2005a:b, Stephens et al. 2009). These changes
in vegetation have altered habitat for a variety of species. Correspondingly, changes in
vegetation and fuel loads have increased the probability of high severity fire (Miller et al.
2009).
The USDA Forest Service is currently actively managing vegetation with the goal
of reducing the probability of large, intense, or severe fires while minimizing negative
effects on wildlife habitat and ecosystem stability. Proposed treatments in portions of the
Plumas and Lassen National Forests include group selections and defensible fuel profile
zones (DFPZs). Group selection treatments involve the harvest of all trees smaller than
30” diameter at breast height (DBH) over a one to two acre area (Stine et al. 2002).
DFPZs are areas with extensive forest thinning and activity fuel treatment intended to
reduce surface and canopy fuel loads. They are also known as shaded fuel breaks and are
designed to allow access for active fire suppression and burn-out operations. DFPZs are
spatially-extensive, covering hundreds to thousands of acres (Stine et al. 2002).
Currently, there is limited information on the effects of landscape fuels treatments
on reducing severe fire behavior and effects, especially at the landscape scale (Agee et al.
2000; Fites-Kaufman et al. 2001, Finney et al. 2007). Elsewhere in the Sierra Nevada,
group selections have been shown to have little effect on the landscape-level behavior of
fire (Stephens 1998); the proposed group selections in the Plumas, however, retain more
large trees per acre than typical group selections. To date, the modeled effects of group
selections with large tree retention have not been published for this forest type.
Assessing the effects of these vegetation management strategies—group
selections and DFPZs—across the forested ecosystems of the Plumas and Lassen
National Forests is the goal of the Plumas-Lassen Administrative Study (Stine et al.
2002). The study is composed of five research teams with distinct focuses: California
spotted owls, small mammals, songbirds, fuels and fire, and vegetation. Due to practical
considerations of a study as spatially extensive as this, we have to mix research with
monitoring. The overall study does not comprise a formal scientific experiment in that the
scientists involved had no control over actual treatments. The study amounts to far more
than monitoring, however, in that we are independently assessing a large landscape and
modeling changes to that landscape given a set of prescriptive treatments.
For the Fuels and Fire Module we investigated the landscape-scale effects of the
proposed forest treatments by answering a suite of questions: First, what are current
conditions, in terms of fuel loads and vegetation, measured directly in the field? Second,
what is the current potential fire behavior and effects given these measured fuel and
vegetation conditions? Third, how would landscape fuels treatments affect vegetation
condition and fire behavior and effects?
The current work is focused on the mixed conifer forests in the Spanish Creek
(Meadow Valley and surrounding areas) watershed in the Plumas National Forest. This
area has received actual DFPZ’s and group selection openings whereas other areas in the
Plumas and Lassen National Forests are still in the planning phase of project
17
2008 Annual Report: Fuels and Fire at the Landscape Scale
implementation. The Meadow Valley area has real DFPZ’s installed, other areas are still
in the planning phase and DFPZ’s are normally outlined on maps as continuous lines
versus separate units that were actually treated.
Study Area:
Our study area is a subset of the Plumas National Forest in Northern California,
USA. The original research plan was to focus on the forests in the study area’s treatment
units (TU) 2, 3 and 4 (Stine et al. 2002), which present widely varying topographical
conditions and contain a variety of owl habitat quality. The total area of these three TUs
is about 60,000 ha (150,000 ac) (Keane 2004). Vegetation varies widely through this
region, presenting a good opportunity to examine fire behavior and end effects across a
spectrum of conditions. As written previously, we have now focused our work on the
only region to actually receive the prescribed treatments which is largely the area covered
by TU 4. The town of Quincy lies directly eastward of Meadow Valley Research Area
(core area encompasses approximately 46,000 acres, buffer around this area of 18,600
acres).
Vegetative cover in this area is primarily mixed conifer forest. The mixed conifer
forest community comprises a mix of three to six conifers and several hardwoods
(Barbour and Major 1995; Holland and Keil 1995; Sawyer and Keeler-Wolf 1995).
Common conifers include ponderosa pine (Pinus ponderosa), Jeffrey pine (P. jeffreyi),
sugar pine (P. lambertiana), incense-cedar (Calocedrus decurrens), Douglas-fir
(Pseudotsuga menziesii) and white fir (Abies concolor). Red fir (Abies magnifica) is
common at higher elevations where it mixes with white fir (Holland and Keil 1995;
Sawyer and Keeler-Wolf 1995). At mid to lower elevations, common hardwoods include
California black oak (Quercus kelloggii) and canyon live oak (Q. chrysolepis) (Rundel et
al. 1995).
In addition, a number of species are found occasionally in or on the edge of the
mixed conifer forest: western white pine (P. monticola) at higher elevations, lodgepole
pine (P. contorta) in cold air pockets and riparian zones, western juniper (Juniperus
occidentalis) on dry sites, California hazelnut (Corylus cornuta), dogwood (Cornus spp.)
and willow (Salix spp.) in moister sites, California bay (Umbellularia californica) and
California nutmeg (Torreya californica) in lower, drier areas (Griffen and Critchfield
1976; Holland and Keil 1995; Rundel et al. 1995).
A variety of vegetation types currently comprise the matrix of covers in which the
mixed conifer forest is arrayed. Vegetation in the matrix ranges from chaparral on
exposed, shallow soils on south and west facing slopes to oak woodlands and riparian
meadows. At higher elevations, particularly toward the Bucks Lake Wilderness, red fir
may be found in pure stands.
Methods:
Plot Layout and Design
Data on forest cover and fuels has been collected in 0.05ha (0.125 ac) plots 12.6m
(41.3 ft) in radius. Plot locations were established using a stratified-random approach.
Strata of elevation, aspect and vegetation type were defined using the layers previously
18
2008 Annual Report: Fuels and Fire at the Landscape Scale
supplied by the contractor VESTRA (Stine et al. 2002). This process identified over 700
plot locations in TU’s 2, 3 and 4. In addition to the randomly-stratified plot locations
described above, similar data was collected at locations identified by the other modules:
plots are located at each owl nesting site and mammal study grid in the three treatment
units. There is a total of 615 inventory plots in TU’s 2, 3, and 4 and 255 of these plots are
in the Meadow Valley Research Area.
We collect data on tree species, diameter at breast height (DBH), categorical
estimate of height, and height to lower crown. Site data collected include location (using
high-precision GPS), slope, and aspect. Canopy cover is assessed at 24 points (every 1
meter) along two linear fuels transects using a site tube.
Surface and ground fuels were sampled in each plot using the line intercept
method (Brown 1974; Brown et al. 1982). Ground and surface fuels were sampled along
two transects radiating from plot center. The first transect is located along a random
azimuth and the second falls 90 degrees clockwise from it. We sampled 1 and 10 hour
fuels from 10-12 meters along each transect, 100 hour fuels from 9-12 meters, and 1000
hour fuels data from 1-12 meters. Duff and litter depth (cm) were measured at 5 and 8
meters along each transect. Maximum litter height is additionally sampled at three
locations from 7 to 8m (Brown 1974; Brown et al. 1982). Total fuel loads were estimated
using equations derived for Sierra Nevada tree species (van Wagtendonk et al. 1996,
1998). Ladder fuels in all plots were assessed using a standard approach (Menning and
Stephens 2007).
Remote Sensing
High resolution IKONOS imagery covering part of the study area was acquired
from Space Imaging in 2003 and another, overlapping section, in 2004. This high spatial
resolution imagery was used to provide information on continuous forest pattern,
structure, cover and variability using methods developed by Menning (2003). Both
acquisitions had identical prescriptions: 1 m panchromatic and 4 m multispectral imagery
collected with an upgraded and narrowed field of view (72-90 degrees from azimuth).
Delivered products were not radiometrically or geometrically corrected but were sent in a
GeoOrtho kit. We completed radiometric corrections in our lab to minimize backscatter
and distortion due to atmospheric moisture and haze. We used PCI Geomatica 9.1’s EASI
modeler module to apply sun angle corrections. Dark target haze removal corrections
were completed using lakes in the scenes as targets. These radiometrically-corrected
images were spatially corrected—orthorectified—using Geomatica 9.1’s Orthoengine
module. To support this effort, ground control points (GCPs) had been collected in the
field using a Trimble GeoXT Global Positioning System (GPS) with hurricane antenna
with sub meter accuracy using wide-angle area support (WAAS). After the
orthorectification was completed we evaluated the results using twelve independent
ground reference points. The analysis indicated the five scenes of the imagery were
accurate within 2.0, 2.6, 2.8, 3.4 and 3.6 m with an overall average of 2.9 m. Each of
these measures is within a single 4 m pixel of the multispectral imagery and so the
resulting orthorectification was deemed precise and consistent enough to use.
Fuel characteristics were mapped from the IKONOS mosaic using supervised
classification cross referenced with pre-treatment data from our PLAS vegetation data
plots and HFQLG pre-treatment monitoring data. Five layers were created as inputs to
19
2008 Annual Report: Fuels and Fire at the Landscape Scale
FLAMMAP and FARSITE: vegetation and fuel type, canopy cover, crown base height,
crown height, and crown bulk density (Finney 1998; 1999). We mapped vegetation and
fuel types applying fuel types described by Burgan and Scott (2005). The national
Landfire project uses these fuel types and we were able to apply a reduced set drawing on
extensive inventory plots and field time in the area.
Supervised classification of vegetation and fuel models was completed in Erdas
Imagine 9.0. Training sites for were chosen using the high resolution panchromatic
imagery as well as the multispectral IKONOS mosaic. Between five and ten training sites
were chosen for each class with emphasis on minimal intermixing of other vegetation
types in the training sample. Four additional data layers were created for input into
FLAMMAP and FARSITE (Finney 2006). Canopy cover was linked to the vegetation
and fuel type (Table 1). To create a more realistic setof continuous values for the canopy
cover, we smoothed the canopy cover values (7x7 pixel FAV filter, PCI Geomatica). The
resulting canopy cover across the landscape ranges from zero, where no trees are
classified, to 90% for pure, almost completely overlapping stands that occasionally
occurred on northern aspects. As a result of the smoothing, however, patches of
forest usually average a more realistic and variable 30-80% canopy cover, depending on
tree density. Predictably, the densest stands grow on northern aspects and this is where
the canopy cover is highest. Canopy height and crown base height were assigned as set
values for each vegetation and fuel class (Table 1).
As we were unable to differentiate different species of conifers, we assigned a
standard bulk density for each class and made it respond to the canopy cover. Thus,
where canopy cover is high, bulk density is assumed to be high (up to 0.25 kg/m3) and
where canopy cover is low, so is bulk density. To create the post-treatment landscape
files we altered a copy of the original vegetation by changing the vegetation and fuel in
areas where DFPZs and group selection units were created.
Post treatment DFPZ treatments and locations are based on refined maps provided
by the Plumas National Forest (in 2009). These maps reflect changes to the proposed
DFPZ treatments in Meadow Valley. Specifically, these maps show which treatments
were implemented after areas were "dropped" for access issues. As post treatment
vegetation data was not collected within DFPZ units under the PLAS project, post
treatment conditions were defined using HFQLG post treatment monitoring data. This
data was used to define a range of scenarios that reflect post treatment stand conditions in
the DFPZ treatment areas. This range of scenarios was randomly applied to DFPZ
treatments across the landscape. Within group selection units, previously published
young plantation vegetation characteristics (Stephens and Moghaddas 2005b) were used
to populate the post treatment FLAMMAP layers.
To verify accuracy of the pre and post treatment FLAMMAP layers, fires were
modeled and compared to actual fire data from the 2008 Rich Complex (Figure 1).
Crown fire potential generally cross referenced well with actual fire severity maps
provided by Jay Miller (unpublished, 2009). If the vegetation and fuel value was
originally grassland or woodland, we left the value the same. Thus, we did not “create” a
coniferous forest where none was previously; these areas retained their non-forest
characteristics.
Weather data were drawn from the remote access weather station (RAWS) in
Quincy and Cashman, CA., from a recent ten year period and processed in Fire Family
20
2008 Annual Report: Fuels and Fire at the Landscape Scale
Plus (Main et al.1990). We chose this ten-year period rather than a longer duration as we
wanted to simulate conditions given the likely continuing warming and drying this region
has experienced in the last decade. Data were collected for 2 weather scenarios—severe
and extreme.
Topographic variables—slope, elevation and aspect—are mapped across the study
area using pre-existing Digital Elevation Models (DEM) on a 10x10m grid. Fire
modeling was conducted in two major phases: first, we evaluated fire behavior and
potential using current condition and post-treatment conditions.
Simulations: Potential fire behavior
Potential fire behavior is being estimated using a similar technique developed by
Stephens (1998) but at much broader spatial scales. The effectiveness of the different
treatments was assessed with computer model FlamMap (Finney 2006) and FARSITE
(1998). Weather scenarios use data from 90th (severe) and 97th (extreme) percentile
conditions collected from local weather stations. Outputs from the fire simulation include
GIS files of fire line intensity (kW/m), heat per unit area (kW/square meter), rate of
spread (m/s), area burned (ha), and if spotting and crowning occurred. This information
will be used to compare the effects of the different landscape level restoration treatments
on altering fire behavior (including no treatment).
A critical response variable focuses on escapements of fire across the landscape
during a longer time period. We will report the flame characteristics near DFPZ’s as a
proxy for fire suppression effectiveness. This will be defined at 90th and 97.5th percentile
fire conditions. This will be an important measure of the effectiveness of the DFPZs at
reducing the chance of fire spreading across the landscape.
Surface and canopy fires are dramatically different in behavior, severity, intensity
and likelihood to spread across a forested landscape. Surface fires are often beneficial,
reducing fuel from the ground and surface, and reducing competition for small trees.
Another response variable, therefore, is a ratio of the area of canopy fire to total fire
extent.
FARSITE will be used to model fire spread and behavior under 90th and 97.5th
percentile weather conditions. Problem winds based on the analysis of historic wind
records are from the southwest. The effectiveness of the DFPZ network and group
selection units will be assessed using fire size, flame lengths, and the number of spot
fires. Conditional burn probability maps will also be created using FLAMMAP for the
pre and post treatment Meadow Valley landscape.
The general approach is to determine pre and post treatment crown fire potential
and average flame lengths across the study area (Figures 2 and 3). These values will be
analyzed across the different land allocations within the study area to determine fire risk
by land use allocation. Specifically, modeled flame length and crown fire potential values
will compared between Protected Activity Centers ("PAC'S"), Riparian Habitat
Conservation areas ("RHCA'S), Off base and Deferred areas (OBD), Defensible Fuel
Profile Zones ("DFPZ's"), and Group Selection ("GS") land allocations.
Complete project outputs will be done by the in early April and a journal
publication will be produced and submitted after the public meeting. During the
presentation in early April a summary of the fire behavior effects of the DFPZ and group
selections will be provided.
21
2008 Annual Report: Fuels and Fire at the Landscape Scale
Literature Cited
Agee, J. K., B. Bahro, et al. (2000). "The use of shaded fuelbreaks in landscape fire
management." Forest Ecology and Management 127(1-3): 55-66.
Barbour, M. G. and J. Major, Eds. (1995). Terrestrial Vegetation of California: New
Expanded Edition. Davis, California Native Plant Society.
Brown, J. K. (1974). Handbook for inventorying downed woody material. Ogden, Utah,
Intermountain Forest and Range Experiment Station Forest Service U.S. Dept. of
Agriculture.
Brown, J. K., R. D. Oberheu, et al. (1982). Handbook for inventorying surface fuels and
biomass in the interior West. Ogden, Utah, U.S. Dept. of Agriculture Forest
Service Intermountain Forest and Range Experiment Station.
Burgan R.E., and J.H. Scott. (2005). Standard fire behavior fuel models: A
comprehensive set foruse with Rothermel’s surface fire spread model. USDA
Forest Service, RMRS-GTR-153.
Finney, M. A. (1998). FARSITE: Users Guide and Technical Documentation. Missoula,
Montana (USA), USDA Forest Service Research Paper: 47.
Finney, M. A. (1999). Some effects of fuel treatment patterns on fire growth: A
simulation analysis of alternative spatial arrangements. Appendix J: Fire and
fuels. Herger-Feinstein Quincy Library Group Forest Recovery Act Final EIS,
USDA Forest Service, PSW.
Finney, M.A. (2006). An overview of FlamMap modeling capabilities. P. 213-220 in
Proc. Of conf. on Fuels management - how to measure success, Andrews, P.L.,
and B.W. Butler (eds.). USDA For. Serv. RMRS-P41.
Finney, M.A., R.C. Selia, C.W. McHugh, A.A. Ager, B. Bahro, and J.K. Agee. (2007).
Simulation of long-term landscape-level fuel treatment effects on large wildfires.
Int. J.Wildland Fire 16(6):712-727.
Fites-Kaufman, J. A., D. Weixelman, et al. (2001). Forest Health Pilot Monitoring
Report:Vegetation, Fuels, Wildlife Habitat, USFS Adaptive Management Services
EnterpriseTeam.
Griffen, J. R. and W. B. Critchfield (1976). The Distribution of Forest Trees in
California.Berkeley, CA, USDA Forest Service.
Holland, V. L. and D. J. Keil (1995). California Vegetation. Dubuque, Iowa,
Kendall/HuntPublishing Company.
Keane, J. (2004). Annual Report on the Spotted Owl Module (Appendix E) in the
Plumas-Lassen Administrative Study, Sierra Nevada Research Center, USDA
Forest Service, Davis, CA.
Main, W.A., D.M. Paananen, and R.E. Burgan. (1990). Fire Family Plus. USDA Forest
Service General Technical Report, NC-138. USDA Forest Service, North Central
ForestExperiment Station, St. Paul, MN.
Menning, K. M. (2003). Forest Heterogeneity: Methods and Measurements From
ExtensiveField Plots, Fire Modeling, and Remote Sensing of the Mixed Conifer
Forest of theSouthern Sierra Nevada, USA (Ph.D. Dissertation). Wildland
Resource Science, University of California, Berkeley.
22
2008 Annual Report: Fuels and Fire at the Landscape Scale
Menning, K.M., and Stephens, S.L. (2007). Fire climbing in the forest: a semi-qualitative,
semiquantitative approach to assessing ladder fuel hazards. Western Journal of
Applied Forestry 22(2): 88-93.
Miller, J.D., H.D. Safford, M. Crimmins, and A.E. Thode. (2009). Quantitative evidence
forincreasing forest fire severity in the Sierra Nevada and southern Cascade
Mountains, California and Nevada, USA. Ecosystems 12(1):16-32.
Parsons, D. J. and S. H. Debenedetti (1979). "Impact of fire suppression on a mixedconifer forest." Forest Ecology and Management 2: 21-33.
Rundel, P. W., D. J. Parsons, et al. (1995). Montane and Subalpine Vegetation of the
Sierra Nevada and Cascade Ranges. Terrestrial Vegetation of California: New
Expanded Edition. M. G. Barbour and J. Major. Davis, California Native Plant
Society.
Sawyer, J. O. and T. Keeler-Wolf (1995). A manual of California vegetation.
Sacramento, CA,California Native Plant Society.
SNEP (1996). Sierra Nevada Ecosystem Project, Final Report to Congress, Vol. II.
Assessments and Scientific Basis for Management Options. Center for Water and
Wildland Resources, University of California, Davis, CA.
Stephens, S. L. (1998). "Evaluation of the effects of silvicultural and fuels treatments on
potential fire behaviour in Sierra Nevada mixed-conifer forests." Forest Ecology
andManagement 105(1-3): 21-35.
Stephens, S.L. and J.J. Moghaddas. (2005a). Experimental fuel treatment impacts on
forest structure, potential fire behavior, and predicted tree mortality in a mixed
conifer forest. Forest Ecology and Management 215:21-36.
Stephens, S.L. and J.J. Moghaddas. (2005b). Silvicultural and reserve impacts on
potential fire behavior and forest conservation: 25 years of experience from Sierra
Nevada mixed conifer forests. Biological Conservation 25:369-379.
Stephens, S.L., J.J. Moghaddas, C. Ediminster, C.E. Fiedler, S. Hasse, M.Harrington, J.E.
Keeley, J.D. McIver, K. Metlen, C.N. Skinner, and A.Youngblood. (2009). Fire
treatment effects on vegetation structure, fuels, and potential fire severity in
western U.S. forests. Ecological Applications 19: 305-320.
Stine, P., M. Landram, et al. (2002). Fire and Fuels Management, Landscape Dynamics,
and Fish and Wildlife Resources: An Integrated Research Plan on the Plumas and
Lassen National Forests, Sierra Nevada Research Center, USDA Forest Service,
Davis, CA.
van Wagtendonk, J. W. (1996). Use of a deterministic fire growth model to test fuel
treatments. Pages 1155-1166 in Assessments and scientific basis for management
options. Sierra Nevada Ecosystem Project, final report to congress, volume II.
University of California, Centers for Water and Wildland Resources, Davis,
California, USA.
van Wagtendonk, J. W., J. M. Benedict, et al. (1996). "Physical properties of woody fuel
particles of Sierra Nevada conifers." International Journal of Wildland Fire 6(3):
117-123.
van Wagtendonk, J. W. and W. M. Sydoriak (1998). "Fuel Bed Characteristics of Sierra
Nevada Conifers." Western Journal of Applied Forestry 13(3): 73-84.
23
2008 Annual Report: Fuels and Fire at the Landscape Scale
Table 1: Values used in development of fuel and canopy layers needed to model fire in
FARSITE and FLAMMAP.
Occurrence in
study area
Initial
Canopy
Cover
(%)
Canopy
Bulk
Density
(kg/m3)
Canopy
Height
(m)
Canopy
Base
Height
(m)
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0.25
tracks
canopy
coverage
from 00.25
tracks
canopy
cover 00.25
tracks
canopy
cover 00.25
tracks
canopy
cover 00.25
tracks
canopy
cover 00.25
24
0.6
28
1.8
14
2.2
21
1.2
31
1.2
#
Burgan
& Scott
Fuel
Model
Description
98
99
NB8
NB9
Water
Bare ground
•
•
102
GR2
Grass – Low
load dry grass
•
144
SH4
Low load
shrub
•
•
147
SH7
Shrub –
chaparral
•
145
SH5
Shrub with low
forest cover
•
•
188
TL8
Moderate load
needle litter
•
Red fir, and higher
white fir areas
0.9
189
TL9
Hardwood with
fuel understory
•
•
Aspen stands Oak
stands in riparian
areas
0.75
163
TU3
Moderate fuel
load timbershrub
•
Extensive
0.9
165
TU5
High fuel load
timber-shrub
•
Northern aspects
only
0.9
Major water bodies
Bare ground, talus,
roads, urban areas
Extensive
grasslands in
American & Indian
Valleys
South facing slopes
Recovering timber
harvest areas
Chaparral type,
dense, south and
west aspects
South aspects only
Forest present but
canopy cover low
24
2008 Annual Report: Fuels and Fire at the Landscape Scale
Figure 1. Canopy cover change resulting from the 2008 Rich fire, derived from an initial
assessment of burn severity using Landsat TM imagery (severity estimates here may be
10-20% higher than actual; a one-year post fire image will be collected and processed
later to more accurately estimate wildfire severity). The Rich fire occurred just north of
the Meadow Valley study area and was largely within the greater PLAS study area. As a
result, we were able to simulate the Rich fire to test fuel and canopy layers we develop in
order to run FARSITE and FLAMMAP. We adjusted fuel and canopy layers such that
our simulations yielded fire types reasonably consistent with the observed canopy cover
change.
25
2008 Annual Report: Fuels and Fire at the Landscape Scale
Figure 2. Modeled flame lengths output from FlamMap under 97th percentile fire weather
conditions (fuel moisture and wind speed) for Meadow Valley study area (outlined in
orange/light line in black and white). The left image represents the Meadow Valley
landscape prior to implementation of DFPZ and group selection treatments. The right
image represents the post-treatment landscape, using only those treatments actually
implemented (as of 2008).
Flame Length
26
2008 Annual Report: Fuels and Fire at the Landscape Scale
Figure 3. Modeled fire type using FlamMap under 97th percentile fire weather conditions
(fuel moisture and wind speed) for Meadow Valley study area (outlined in purple/dark
line in black and white). The left image represents the Meadow Valley landscape prior to
implementation of DFPZ and group selection treatments. The right image represents the
post-treatment landscape, using only those treatments actually implemented (as of 2008).
Fire Type
No fire
27
Plumas/Lassen Administrative Study Vegetation Module
Forest Restoration in the Northern Sierra Nevada:
Impacts on Structure, Fire Climate, and Ecosystem Resilience.
Report of Activities during 2008
Project Staff
Dr. Seth Bigelow, Biologist. Phone: 530-759-1718. sbigelow@fs.fed.us
Dr. Malcolm North, Research Plant Ecologist. Phone: 530-754-7398. mnorth@ucdavis.edu
Sierra Nevada Research Center, Pacific Southwest Research Station U.S. Forest Service,
1731 Research Park drive, Davis, CA 95618
OBJECTIVES
The vegetation module of the Plumas-Lassen Administrative Study studies how changes
in the forest canopy affect ecosystem functioning, including microclimate, growth and
competition of shrubs and juvenile trees, understory diversity, and landscape continuity.
Research approaches include stand-level experimental manipulations, measurement of
plant growth and survival along existing environmental gradients, and assessment of
impacts of routine (i.e., non-experimental) forest management activities.
Research Activities 2008
We monitored microclimate conditions (air temperature and humidity, wind speed), fuels
dryness, and soil temperature in twelve plots subsequent to experimental thinning and
group selection in 2007.
Results
We viewed the microclimate data on the day when a major fire began (Moonlight fire,
Sept. 3-15 2007, 65, 000 acres) to see how the three treatments (untreated control, fuelsreduction thinning, and group selection) behaved. Maximum wind gust speed was
slightly higher in the thinned stands than in the controls, and at least twice as high (up to
15 mph) in group selection openings as in controls (Figure 1). In contrast, maximum
daily air temperature and minimum relative humidity did not differ among the treatments,
perhaps because of the increased ventilation in the thinned stands and group openings
(Figure 2 and 3).
28
Figure 1. Wind gust speed at 8 feet above ground in experimentally treated mixed
conifer stands in Meadow Valley on the day the Moonlight fire began, 3 Sept. 2007.
Treatments were control, thinning for fuels reduction (thin1 and thin2), or harvest
via group selection (Group). Each row denotes an area where a group of 4
treatments were applied.
29
Figure 2. Air temperature at 6 feet about the ground on the day of the Moonlight
Fire (3 Sept. 2007), measured in stands in the Meadow Valley area that were either
untreated (Control), thinned for fuels reduction (Thin1 and Thin2), or harvested via
group selection (Group). Maximum air temperature did not differ among
treatments.
30
Figure 3. Relative humidity at 6 feet above the ground on 3 September, 2007, in
experimental stands in the Meadow Valley area. There is no difference in minimum
relative humidity among treatments. See legend of Figure 1 for explanation of
treatment and row headings.
Outreach
Plumas-Lassen study symposium, Quincy, CA, April 2008.
Publications (in Review)
Bigelow, S. W., and S. A. Parks. Predicting altered forest connectivity due to group
selection silviculture. In review at Landscape Ecology.
Bigelow, S. W., M. P. North, and W. R. Horwath. Resource-Dependent Growth Models
for Sierran Mixed-Conifer Saplings. In review at the Open Forest Science Journal.
31
Plumas-Lassen Administrative Study Module on Small Mammal
Distribution, Abundance, and Habitat Relationships
2008 Annual Report
Submitted March 2009
Project Leader:
Brett R. Jesmer
University of California Davis
Department of Wildlife, Fish, and Conservation Biology
One Shields Avenue
Davis, CA 95616
brjesmer@ucdavis.edu
Principal Investigators:
Douglas A. Kelt1, Dirk H. VanVuren1, and Michael L. Johnson2
1
University of California Davis
Department of Wildlife, Fish, and Conservation Biology
One Shields Avenue
Davis, CA 95616
2
University of California Davis
John Muir Institute of the Environment
One Shields Avenue
Davis, CA 95616
32
TABLE OF CONTENTS
EXECUTIVE SUMMARY .........................................................................................................................35
INTRODUCTION.......................................................................................................................................36
OBJECTIVES .............................................................................................................................................40
MATERIALS & METHODS......................................................................................................................40
Live-trapping...........................................................................................................................................40
Long-term grids ..................................................................................................................................40
Northern flying squirrels.....................................................................................................................43
Dusky-footed woodrats.......................................................................................................................44
Chipmunks..........................................................................................................................................45
Animal handling......................................................................................................................................45
Radiotelemetry ........................................................................................................................................46
Radiotransmitter application...............................................................................................................46
Homing ...............................................................................................................................................46
Triangulation.......................................................................................................................................47
Home range analysis ...........................................................................................................................47
Vegetation ...............................................................................................................................................49
Long-term grids ..................................................................................................................................49
Northern flying squirrels.....................................................................................................................50
Dusky-footed woodrats...............................................................................................................51
RESULTS AND DISCUSSION..................................................................................................................51
2008 Field Season ...................................................................................................................................52
Long-term grids ..................................................................................................................................52
Publications #1 and #4: Habitat associations of small mammals at two spatial scales in the
northern Sierra Nevada .......................................................................................................................53
Publication #8: Population dynamics of small mammals in relation to cone production in
four forest types in the northern Sierra Nevada ..................................................................................53
Publication #9: Trapping rodents in a cautious world: the effects of disinfectants on trap
success. ...............................................................................................................................................54
Northern flying squirrels .........................................................................................................................54
Publication #7: Home range and activity of northern flying squirrels in the northern Sierra
Nevada ................................................................................................................................................55
Publication #3: Home range and habitat selection of northern flying squirrels in the northern
Sierra Nevada......................................................................................................................................55
Dusky-footed woodrats ...........................................................................................................................55
Publication #2 and #5: Habitat selection by dusky-footed woodrats in managed mixedconifer forest of the northern Sierra Nevada ......................................................................................56
Publication #6: Characteristics and use of tree houses by dusky-footed woodrats in managed
mixed-conifer forest of the northern Sierra Nevada ...........................................................................56
Publication #11: Spatial organization of dusky-footed woodrats in managed mixed-conifer
forest of the northern Sierra Nevada...................................................................................................57
Golden-mantled ground squirrels............................................................................................................57
Publication #8: Effects of maternal body condition on offspring dispersal in golden-mantled
ground squirrels. .................................................................................................................................58
Chipmunks ..............................................................................................................................................58
Publication #10: A multiple spatial scale perspective of the habitat affinities of sympatric
long-eared and Allen’s chipmunks. ....................................................................................................58
2009 Field Season ...................................................................................................................................59
COLLABORATION ...................................................................................................................................59
PUBLICATIONS ........................................................................................................................................59
Theses .....................................................................................................................................................59
33
Peer-reviewed..........................................................................................................................................60
Submitted ................................................................................................................................................60
In Preparation..........................................................................................................................................60
PRESENTATIONS .....................................................................................................................................60
PERSONNEL..............................................................................................................................................62
ACKNOWLEDGEMENTS ........................................................................................................................62
REFERENCES............................................................................................................................................63
FIGURES LEGENDS .................................................................................................................................72
FIGURES ....................................................................................................................................................74
TABLE LEGENDS .....................................................................................................................................96
TABLES......................................................................................................................................................97
34
EXECUTIVE SUMMARY
In this document we report on the Mammal Module of the Plumas-Lassen Administrative
Study (PLAS). A pilot study was conducted September-November 2002, the study
design was incorporated in 2003, and 2008 marked the sixth year of implementation of
the study. As of the end of the 2007 field season, all of the proposed treatments have been
implemented; thus, everything we report in 2008 reflect post-treatment conditions,
whereas all data reported prior to 2007 reflect pre-treatment conditions.
The information provided in this report is intended to provide background information on
the pre-treatment and post-treatment status of small mammals in a variety of forested
habitat types, determine habitat associations of many small mammal species, particularly
the principle prey of the California spotted owl (i.e., dusky-footed woodrat, Neotoma
fuscipes; northern flying squirrel, Glaucomys sabrinus), and provide resource managers
with important habitat attributes to manage for to ensure a sustainable mammalian
community.
In 2008 Brett Jesmer succeeded Robin Innes as Project Leader of the Mammal Module of
the PLAS. Robin Innes continues to develop and improve upon manuscripts initiated
during their time with the PLAS, although she currently works with the U.S.D.A. Forest
Service in Missoula, MT. To date, we have had three graduate students at the University
of California, Davis successfully complete their graduate work with the PLAS. In 2005,
Stephanie Coppeto completed her graduate work on the habitat associations of small
mammals at multiple spatial scales. In 2006, Robin Innes completed her graduate work
on habitat selection by dusky-footed woodrats. Jaya Smith will complete his graduate
work on home range and habitat use of the northern flying squirrel in 2009.
35
INTRODUCTION
Small mammals play vital roles in forest ecosystems, serving as important consumers and
dispersers of seeds, fruits, and fungi (Carey et al. 1999; Gunther et al. 1983; Maser and
Maser 1988; Pyare and Longland 2001), and as prey for mammalian and avian predators,
including many species of concern in the Sierra Nevada (e.g., California spotted owl,
Strix occidentalis occidentalis; northern goshawk, Accipiter gentilis; fisher, Martes
pennanti; and marten, M. americana; Carey et al. 1992; Forsman et al. 1984; Zielinski et
al. 1983). Given their essential interactions with flora and fauna across multiple trophic
levels (e.g., Carey et al. 1992; Forsman et al. 1984), changes in the distribution and
abundance of small mammals could substantially affect the dynamics of forest
communities. This makes small mammals valuable subjects for the integrative research
necessary to fully understand the ecological responses of spotted owls and other species
to forest management practices.
Here we report on the Mammal Module of the PLAS, one of five integrated study
modules intended to evaluate land management strategies within the area covered by the
Herger-Feinstein Quincy Library Group Forest Recovery Act (HFQLG) Pilot Project.
Forest management practices implemented during the HFQLG Pilot Project aim to
“…promote ecologic and economic health for federal lands and communities of the
northern Sierra Nevada” (Title IV-HFQLG Forest Recovery Act- Sec.401).
Understanding how small mammal communities respond to different forest management
regimes at macrohabitat (i.e., stand-level, landscape) and microhabitat (trap-level, home
range) scales would provide valuable feedback to other PLAS modules. Managers
manipulate fuel loads through mechanical thinning of the forest in an effort to prevent
catastrophic wildfire events and to create more suitable conditions for prescribed burning.
Such management activities are becoming more frequent throughout western North
America (Covington et al 1997, Dodge 1972, Long et al. 2008). Exactly how these
management practices effect small mammal populations, particularly in the Sierra
Nevada, is not well understood (Carey 2000, Meyers et al 2007, Suzuki and Hayes 2003).
As with the current status of the Mammal Module, most studies focus on the short-term
responses of small mammal communities to habitat manipulation (e.g., Carey 2000,
Carey and Wilson 2001, Converse et al. 2006, Klenner and Sullivan 2003, Meyers et al.
2007). Long-term studies have the potential to record a broader range of variability in
community dynamics which will lead to a greater understanding of ecosystem processes
(Beche and Resh 2007); i.e. distilling natural fluctuations in population, climatic effects,
and predator responses from the effects of management practices. The value of long-term
studies has been exemplified in many field studies across taxa (eg. Beche and Resh 2007,
Brook and Bradshaw 2006, Elias et al 2006, Jackson and Furender 2006, Pettorielli and
Durant 2007, Tsuji et al 2006). When considering the goals of the HFQLG forest
recovery act, the long-term study approach is most appropriate for the mammal module.
We will continue to develop predictive small mammal habitat models to forecast how
individual species will respond to forest management treatments and test these models by
assessing the impacts of forest management treatments on long term responses of small
mammal abundance and species diversity. This is done by monitoring several
independent populations of small mammals for multiple years before and after forest
36
management treatments are applied, developing demographic profiles (e.g., survival,
reproduction) of species, and obtaining detailed measurement of habitat characteristics.
To sample and monitor these small mammal populations, we have established permanent
(long-term grids) and temporary (land bird transects) live-trapping grids throughout the
Plumas National Forest (PNF).
Key non-hibernating small mammals in the northern Sierra Nevada include the northern
flying squirrel (Glaucomys sabrinus) and dusky-footed woodrat (Neotoma fuscipes).
Northern flying squirrels and dusky-footed woodrats are the principle prey of the
California spotted owl (Carey et al. 1992; Rosenberg et al. 2003), a species of concern in
California due to its dependence upon late-seral forest ecosystems (United States
Department of the Interior 2003), which are among the most highly altered ecosystems in
the Sierra Nevada (Beardsley et al. 1999; Franklin and Fites-Kaufman 1996). For
example, some populations of northern flying squirrel may be depressed by the intensity
of spotted owl predation (Carey et al. 1992), and high woodrat biomass may reduce the
area requirements of the spotted owl (Carey et al. 1990; Zabel et al. 1995). Thus,
northern flying squirrels and dusky-footed woodrats are an important focus of our study
module.
Northern flying squirrels are nocturnal, arboreal rodents located throughout the northern
latitudes of the United States and Canada (Wells-Gosling and Heaney 1984), and
frequently associated with forests with high densities of large trees (Smith et al. 2004,
2005). Northern flying squirrels act as a major dispersal agent for hypogeous fungal
spores, which are important for nutrient and water uptake by host trees (Fogel 1980).
Although they are typically associated with mesic northern forests, northern flying
squirrels are also found throughout the Sierra Nevada where they experience a much
more xeric landscape than elsewhere in their range; as a result, populations of northern
flying squirrel inhabiting the Sierra Nevada may be quite different from those inhabiting
the more mesic forests of Oregon, Washington, and Alaska. Specifically, northern flying
squirrels may be more sensitive to moisture availability in the Sierra Nevada since this
influences the distribution of truffles, their primary food source. This disjunctive
distribution of food resources may drive differences in northern flying squirrel biology,
suggesting that northern flying squirrels may exhibit a more clumped distribution, lower
overall densities, increased competition for suitable nest trees, and larger individual home
ranges; thus, northern flying squirrels in the Sierra Nevada may be affected differently by
forest management practices than in other parts of their range. We are using live-trapping
and radiotelemetry techniques to determine the abundance and distribution, habitat use,
and home range of northern flying squirrels in the Sierra Nevada, comparing this with
data from other parts of their distribution, and evaluating the effects of forest
management practices on this species within the area covered by the HFQLG Pilot
Project.
The dusky-footed woodrat is a nocturnal, semi-arboreal rodent found throughout northern
California and Oregon that inhabits a wide variety of densely vegetated habitats,
including chaparral, juniper woodland, streamside thickets, and deciduous or mixed
forests with well-developed undergrowth (Carraway and Verts 1991). Dusky-footed
37
woodrats play an important role in community dynamics. As mentioned previously, they
are prey for many avian and mammalian predators, including the California spotted owl.
Additionally, the availability of woodrat houses may influence species richness for small
mammals, reptiles, amphibians, and invertebrates (Cranford 1982; M’Closkey et al. 1990;
Merritt 1974; Vestal 1938). Thus, promoting quality habitat for the dusky-footed
woodrat may provide a variety of ecological values in managed forests, for example in
the form of increased biodiversity, with important consequences for forest conservation
(Carey et al. 1999). We used live-trapping and radiotelemetry to determine the abundance
and distribution, habitat use, and home range of dusky-footed woodrats in the Sierra
Nevada, and evaluate the effects of forest management practices on this species.
Specifically, our first objective was to test for an association between woodrat abundance
and abundance of California black oak (Quercus kelloggii), an important food source
(Atsatt and Ingram 1983; Cameron 1971; Meserve 1974). Our second objective was to
evaluate the importance of microhabitat variables to dusky-footed woodrats at 2 levels -placement of houses within mixed-conifer habitat, and use of houses within home ranges.
Dusky-footed woodrats construct conspicuous houses on the ground using sticks, bark,
and plant cuttings, and sometimes also on limbs or in cavities of trees (Fargo and
Laudenslayer 1999). Given the investment involved in building, maintaining, and
defending a house, we predicted that houses should be distributed such that they
minimize energetic costs in movement, yet maximize individual fitness components
(Manley et al. 1993), such as access to food, protection from predators, and a thermally
suitable microclimate (Atsatt and Ingram 1983). Thus, we evaluated ground and tree
house-site selection of houses by dusky-footed woodrats by comparing house sites with
nearby random sites. Since only a subset of available houses is used by woodrats at any
one time (Carey et al. 1991; Cranford 1977; Lynch et al. 1994), some houses may be
more suitable than others. We evaluated house suitability by comparing characteristics of
used and unused ground houses and availability and use of house trees. Because
woodrats defend their house against conspecifics, subadults might be forced to settle in
lower quality houses (Vestal 1938), thus, we also evaluated whether subadults selected
houses differently from those selected by adults. Our third objective was to examine the
spatial organization of dusky-footed woodrats. The spatial organization of a population
has important implications for population dynamics, as well as the genetic structure of a
population (e.g., Dunning et al. 1992; Lambin and Krebs 1991; Sugg et al 1996).
Two commonly occurring species of mice, the deer mouse (Peromyscus maniculatus) and
the brush mouse (Peromyscus boylii), inhabit PNF. These species display different
behavior and morphological characteristics yet both posses similar foraging habits. Due
to the ubiquitous nature of deer mice, there is a great overlap in habitat affinity with the
brush mouse (Jameson 1951). Deer mice are nocturnal, non-hibernating, terrestrial to
semi-arboreal species, which are distributed throughout the Sierra Nevada (King 1968;
Martell and Macaulay 1981; Jameson and Peters 2004) and forage on a variety of seeds,
berries, insects, and hypogeous fungi. Deer mice are by far the most commonly occurring
Peromyscus encountered within the study areas of the Mammal Module in PNF (98% of
6252 Peromyscus individuals captured). We will group these two similar species and
herein refer to them collectively as Peromyscus species as is commonly done when
referring to spotted owl forage in situations when more than one species of Peromyscus is
38
present. Variability among spotted owl subspecies diet and regional differences within
spotted owl subspecies diet indicate foraging habits of the spotted owl may differ based
on habitat and prey availabilities. Little is known of the California spotted owl diet in
PNF. Diet studies of the spotted owl (Strix occidentalis) indicate that Peromyscus
species are a frequently taken prey item but do not dominate the percent of biomass
ingested by the spotted owl (Block et al. 2005, Bravo-Venaja 2005, Forsman et al. 2004,
Rosendberg et al. 2003, Smith et al. 1999, Trailkill and Bias 1989). However,
reproductive success in northern spotted owls is positively correlated with trends in
Peromyscus abundance (Rosenburg et al. 2003), and Munton and others (2002) recorded
an increase in Peromyscus taken during the breeding season of California spotted owls in
the Sierra National Forest. Coppeto et al (2006) reported that trends of Peromyscus in
PNF are explained primarily by forest type and year. Wilson et al (2008) further
supported the findings of Coppeto et al. by correlating yearly trends, such as mean cone
production, with Peromyscus abundance in PNF. Such factors appear to strongly
influence annual Peromyscus abundance.
Other key small mammals include the golden-mantled ground squirrel (Spermophilus
lateralis) and chipmunks (Tamias sp.), which also are important prey species of the
northern goshawk, a species of increasing concern to resource managers due to the
species’ sensitivity to the effects of forest management. Chipmunks are forest-associated,
semi-arboreal rodents that constitute a considerable portion of the small-mammal
biomass in an area, making them important prey for a variety of mammalian and avian
predators (Vaughan 1974). Additionally, chipmunks are important consumers and
dispersers of seeds (Briggs and Vander Wall 2004; Vander Wall 1992), and may
contribute to the natural regeneration of some species of plants by caching seeds (Aldous
1941). Small mammals may cache seeds beneath the layer of decaying vegetation on the
forest floor, where they stand a better chance of germinating than those remaining on the
surface litter (Sumner and Dixon 1953); alternatively, they may deposit seeds in
underground burrows where seeds cannot establish seedlings. If soil-moisture levels have
been altered due to fire, logging, or weather patterns, the ability of chipmunks to retrieve
cached seeds may be reduced, thus promoting germination of a larger proportion of seeds
after disturbance (Briggs and Vander Wall 2004;Vander Wall 2000). However, if
chipmunks are very abundant, they can prevent normal regeneration of some plants,
particularly pines, by eating their seeds, which may contribute to the generation of dense
brushfields that could further hinder the return of timber (Smith and Aldous 1947, Tevis
1953). We were particularly interested in the long-eared (T. quadrimaculatus) and
Allen’s (T. senex) chipmunks, which occur commonly throughout PNF. These sympatric
species are similar in body mass, diet, and general resource utilization, and thus are likely
to compete locally. Similar species often coexist by partitioning habitat. However,
detecting differences in habitat affinities is influenced by spatial scale. Our objective was
to investigate the abundance, distribution, and habitat associations of the long-eared and
Allen’s chipmunks at three spatial scales in PNF and evaluate the affect of forest
management practices on these species.
39
OBJECTIVES
The primary objective of the Mammal Module is to evaluate small mammal responses to
different forest management practices, and to model these responses in terms of
demography, spatial distribution, and habitat associations at local and landscape scales.
To meet the primary objective, we will address the following:
1.
Determine small mammal habitat associations at macro- and microhabitat scales.
2.
Develop demographic profiles of small mammal populations inhabiting a variety of
habitat types.
3.
Develop predictive small mammal habitat models, based on the results of objectives
1-2, to forecast how individual species will respond to forest management
treatments.
4.
Quantitatively assess the impacts of forest management treatments on small
mammal abundance and species diversity.
5.
Determine small mammal population trends, evaluate how populations are changing
temporally, and assess the factors responsible for the observed trends.
6.
Evaluate the spatial distribution (i.e., home range), social organization (i.e., home
range overlap), and habitat selection (i.e., den use, house use) of the principle prey
of the California spotted owl, the northern flying squirrel and dusky-footed
woodrat.
7.
Evaluate habitat affinities of two sympatric and semi-cryptic chipmunks, the longeared and Allen’s chipmunks, at multiple spatial scales.
MATERIALS AND METHODS
Live-trapping
Capture-recapture data obtained from the live-trapping methods described herein allow us
to measure population parameters such as abundance, density, and frequency of
occurrence of individual small mammal species and small mammal species richness and
diversity, and permit the measurement of habitat use, availability and selection (Lancia et
al. 1996, Litvaitis et al. 1996). Live-trapping methods are useful for making comparisons
of small mammal communities across time, locations, habitats, and land-use treatments.
We established several different live-trapping designs, each appropriate to the small
mammal community or species of interest.
Long-term grids
To provide base-line information on small mammal populations inhabiting major forest
types, and to quantitatively assess the impacts of forest management treatments on small
mammal abundance and species diversity, we established 21 long-term grids using
controls and pre- and post-treatment data. In 2007, all of the proposed treatments were
implemented. All data collected between 2003 and 2006 were collected prior to any
treatments to determine baseline conditions. In 2003, we established 18 permanent, livetrapping grids (Fig. 1a); we established 3 additional long-term grids in 2005. Twenty
40
grids consist of a 10 x 10 array of Sherman traps (Model XLK, 7.6 x 9.5 x 30.5 cm, H. B.
Sherman Traps, Inc., Tallahassee, FL, USA) with 10 m spacing, nested within a larger 6
x 6 grid of 72 Tomahawk traps (Model 201, 40.6 x 12.7 x 12.7 cm, Tomahawk Live
Trap, Tomahawk, WI, USA; 1 ground, 1 arboreal) with 30 m spacing (Fig. 1b). The
remaining long-term grid was constrained by road configuration such that the array of
Sherman traps was nested within a 4 x 9 grid of 72 Tomahawk traps (30 m trap spacing; 1
ground, 1 arboreal). Arboreal traps were placed approximately 1.5 to 2 m above the
ground on a randomly-selected tree located <10 m from the grid point; arboreal traps may
or may not be placed on the same tree each trapping session. Ground traps were placed
within 1 m of the grid point under protective cover, such as a shrub or log, at small
mammal burrow entrances, and along small mammal runways, when possible.
We trapped all long-term grids (n=21) in 2008. All grids had 120 trap stations and
covered 2.25 ha (3.24 ha with a ½ inter-trap distance buffer) of contiguous forest.
Arboreal Tomahawk traps were removed from all grids on August 1, 2004 because of
consistently poor capture rates; however, arboreal Tomahawk traps were again used in
2005 and thereafter, and capture rates were improved by placing the trap entrance flush
against the tree bole, fastening the trap more securely to the tree, and switching to a more
desirable bait mixture, following Carey et al. (1991).
Prior to August 2005, all traps were baited with crimped oats and black oil sunflower
seeds lightly coated in peanut butter; thereafter, traps were baited with a mixture of rolled
oats, molasses, raisins, and peanut butter which was formed into a small, sticky ball
(Carey et al. 1991). Small nest boxes made from waxed-paper milk cartons were placed
behind the treadle in Tomahawks to minimize stress and provide thermal and protective
cover (Carey et al. 1991); in addition, synthetic bedding material (nonabsorbent
polyethylene batting), and natural cover (e.g., bark, moss) or cover boards, were provided
as needed for thermal insulation. After the trap session was completed, bait was
discarded on the ground at the grid point and all traps were removed.
The 18 long-term grids established in 2003 were placed in 5 principal forest types
(Coppeto et al. 2005, 2006; these are Pub. #1 and #4). Forest types were defined by the
dominant live tree species representing ≥ 70% of total tree composition, and included
white fir (Abies concolor, n = 4), red fir (A. magnifica, n = 3), mixed fir (co-dominant
mix of white fir and Douglas-fir, Pseudotsuga menziesii, n = 5), mixed conifer (n = 3),
and pine-cedar (co-dominant mix of yellow pine [ponderosa pine–Pinus ponderosa and
Jeffrey pine–P. jeffreyi], and incense cedar, Calocedrus decurrens, n = 3). In 2005,
sampling grids were established in group selects located in white fir (n=2) and mixedconifer (n=1) habitats. In an effort to more fully integrate our module with those of other
research modules of the PLAS, Wilson et al. (Publication #5) used alternative forest type
classes for these grids, as follows: white fir (n=9), red fir (n=3), Douglas fir (n=3), and
ponderosa pine (n=3). According to this classification, the 3 group selects established in
2005 were placed within white fir habitat. Overall, PNF is dominated by white fir and
Douglas fir so these forest types had proportionally more trapping grids placed within
them. Common shrubs in the region include mountain rose (Rosa woodsii), Sierra
gooseberry (Ribes roezlii), serviceberry (Amelanchier utahensis), bush chinquapin
41
(Chrysolepis sempervirens), green- and white-leaf manzanita (Arctostaphylos patula and
A. viscida), mountain dogwood (Cornus nuttallii), mountain whitethorn and deer brush
(Ceanothus cordulatus and C. intigerrimus), bitter cherry (Prunus emerginata), willow
(Salix spp.), Fremont silk tassel (Garrya fremontii), and huckleberry oak (Quercus
vaccinifolium). Pinemat manzanita (Arctostaphylos nevadensis) occurred almost
exclusively in red fir forests, and buck brush (Ceanothus cuneatus) predominantly in
pine-cedar/ponderosa pine forests.
GRID
#
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
SITE NAME
BLACKOAK
BOA
CABIN
DOGWOOD
GIMP
GREENBOTTOM
GULCH
MONO
NOGO
NONAME
OASIS
RALPH
RIPPER
RUTT
STEEP
SWARM
TEEPEE
THICK
TOAST
TRIANGLE
VIEW
LOCATION
SNAKE LAKE
SCHNEIDER CREEK
MILLER FORK
LITTLE SCHNEIDER CREEK
DEAN'S VALLEY
GRIZZLY MOUNTAIN
BEAN CREEK
GRIZZLY MOUNTAIN
SNAKE LAKE
DEAN'S VALLEY
BEAN CREEK
GRIZZLY MOUNTAIN
MILLER FORK
SNAKE LAKE
TAYLOR CREEK
BEAN CREEK
DEAN'S VALLEY
DEAN'S VALLEY
MILLER FORK
MILLER FORK
SNAKE LAKE
FOREST TYPE AS
PRESENTED IN
COPPETO ET AL.
MIXED-CONIFER
MIXED-FIR
WHITE FIR
MIXED-FIR
MIXED-FIR
RED FIR
PINE-CEDAR
RED FIR
MIXED-CONIFER
WHITE FIR
PINE-CEDAR
RED FIR
WHITE FIR
MIXED-CONIFER
MIXED-FIR
PINE-CEDAR
WHITE FIR
WHITE FIR
WHITE FIR
WHITE FIR
MIXED-CONIFER
FOREST TYPE AS
PRESENTED IN
WILSON ET AL.
WHITE FIR
MIXED-FIR
WHITE FIR
MIXED-FIR
WHITE FIR
RED FIR
PINE-CEDAR
RED FIR
WHITE FIR
WHITE FIR
PINE-CEDAR
RED FIR
WHITE FIR
WHITE FIR
MIXED-FIR
PINE-CEDAR
WHITE FIR
WHITE FIR
WHITE FIR
WHITE FIR
WHITE FIR
EXPERIMENT
TYPE
LIGHT THIN
HABITAT
CONTROL
HABITAT
HEAVY THIN
HABITAT
HABITAT
HABITAT
CONTROL
LIGHT THIN
HABITAT
HABITAT
HEAVY THIN
GROUP SELECT
HABITAT
HABITAT
CONTROL
GROUP SELECT
GROUP SELECT
LIGHT THIN
HEAVY THIN
Twelve of the long-term grids were placed within the experimental management plots
established by the Vegetation Module of the PLAS. These 12 study plots were placed in
3 groups (Miller Fork, Dean’s Valley, and Snake Lake) of 4 study plots, consisting of 1
control plot and 3 experimental plots (1 group select plot, 1 light thin, and 1 heavy thin).
The remaining 9 study plots were not established in groups. Minimum distance among
long-term grids (n=21) was 1 km with the exception of 4 grids that were 700-900 m
apart. Movement between grids was extremely rare; we documented a single goldenmantled ground squirrel that moved from red-fir grid #12 to red-fir grid #6 in 2006.
Long-term grids were trapped monthly (May-October) during 2003-2004 and biannually
(June, Oct) during 2005-2006. We sampled once in 2007 (July-August) because logging
and prescribed burn activities at treatment grids restricted sampling activities. Grids were
42
again sampled biannually in 2008. Trapping sessions consisted of 4 consecutive trapnights. Sherman and Tomahawk traps were set and baited every evening just before
dusk, and checked just after dawn; Sherman traps were then closed until dusk whereas
Tomahawk traps were re-baited and checked again at mid-day, a minimum of 2 hours
after the first trap check, at which point they were closed until dusk. This resulted in all
traps remaining closed from 12:00 – 16:00, but enabled us to sample both diurnal and
nocturnal species while reducing deaths that result from thermal stress during the hottest
part of the day. Field technicians were thoroughly trained and rotated among grids each
trapping session, to reduce the variability in capture success due to differences among
technicians.
Demographic profiles.—Population demographics will be modeled by species
using program MARK. Species without sufficient captures to generate detailed capture
history will be modeled using the minimum number known alive (MNKA) parameter.
Monthly or seasonal survival and population densities will be modeled for each species
by habitat type using the Cormack-Jolly-Seber data type in program MARK. Suitable
habitat parameters, such as cone production, will be incorporated into population models
and can be used to identify habitat variables that are linked to population parameters
using multivariate analyses.
Northern flying squirrels
We captured and radiocollared northern flying squirrels at long-term grids, land bird
transects, and at areas predicted to have moderate and high suitability for northern flying
squirrels, hereafter flying squirrel transects (Fig. 3). At long-term grids and land bird
transects, northern flying squirrels were collared only in areas where triangulation was
feasible, which required fairly large areas of habitat with one or two roads bisecting the
area. In 2004, animals were captured and radiocollared at 3 long-term study grids
located in upper elevation (2,100 m) red-fir habitat. Additional transects bisecting or
parallel to original transects were established during 2005-2007 to increase the area
covered and increase capture success. The 3 long-term grids and intervening habitat are
hereafter referred to as study site FS-1. In 2005, we established a second study site (FS2) in mixed-conifer forest at 1,500 m elevation; in 2006 and 2007, additional transects
bisecting or parallel to original transects were established. FS-2 was selected using a GISbased northern flying squirrel habitat-relations model based on known habitat affinities of
this species, and which predicted poor, moderate, and high suitability habitat for northern
flying squirrels. Although we established many live-trapping transects (>10) in areas
predicted to have high and moderate suitability, study site FS-2 was the only study site to
yield captures where triangulation also was feasible; FS-2 was predicted to have
moderate suitability for northern flying squirrels.
We established flying squirrel transects primarily along riparian areas, due to the
importance of this habitat type to northern flying squirrels (Meyer and North 2005). If
habitat, road configuration, and topography were suitable, we used a live-trapping grid
(i.e., several parallel transects) to maximize the number of captures. We used a
combination of Sherman and Tomahawk traps, typically 1 Sherman and 2 Tomahawk (1
ground, 1 arboreal) traps, spaced 40-50 m apart. Sherman and Tomahawk traps were set
43
and baited every evening just before dusk, and checked with the same schedule as our
long-term trapping grids. As indicated above, we changed out trapping bait and protocol
for flying squirrels in August 2005.
Dusky-footed woodrats
We established four study sites (WR1 – WR4, 1,450–1,750 m elevation; Fig. 2) in earlyseral forest (30–40 years post-logging), representative of Sierra Nevada westside mixedconifer forest type and characterized by California black oak, sugar pine (Pinus
lambertiana), yellow pine, white fir, Douglas fir, and incense cedar. All study sites had a
brushy understory consisting primarily of deer brush, buck brush, and mountain
whitethorn, with lesser coverage by greenleaf and whiteleaf manzanita and mountain
dogwood. Each study site included 2–4 habitat types, which varied in composition of
overstory and understory dominants, canopy closure, and aspect. Habitat type was
defined by GIS data layers provided by the USDA, Forest Service. WR-1 and WR-2 had
moderately sloping topography, whereas WR-3 and WR-4 had mixed terrain or
undulating topography. Because woodrat activities extended somewhat into adjacent
habitats, we trapped woodrats at all houses located within approximately 3 home range
diameters (ca. 180 m— Cranford 1977, Lynch et al. 1994, Sakai and Noon 1997) of each
study site to ensure that all woodrats potentially influencing the spatial structure at each
study site were identified. Historic logging activities and fire suppression practices
contributed to heterogeneity within study sites, with abundant dead wood as well as
shrubby gaps interspersed with patches of closed canopy forest. Recent (<5 yr)
management activities (e.g., prescribed burns, logging) have created open understory and
overstory conditions in areas between study sites. Study sites lay 1.2–2.8 km apart, and
no woodrats were recorded moving between study sites.
We systematically searched for woodrat houses in the spring and fall of 2004-2006 by
walking overlapping belt transects that covered each study site. In addition, woodrat
houses were opportunistically located at all study sites during a concurrent radiotelemetry
study of woodrat movements. Each house was marked and its location mapped (≤1 m)
using a GPS unit (Trimble Navigation, Ltd., Sunnyvale, California; GeoExplorer,
GeoXT), and volume was estimated as a cone using measurements of length, width, and
height.
Woodrats are active year-round, but our study was limited to the snow-free period (MayOctober). We documented house use by livetrapping in the spring (May–June) and late
summer-early fall (August–September) of 2004-2006 with 4 Sherman live-traps (H.B.
Sherman Traps, Inc., Tallahassee, Florida; 7.6 × 9.5 × 30.5 cm) set at the base of each
house for 4 consecutive nights; longer trapping efforts (>4 consecutive nights) do not
yield higher success (Carey et al. 1999; Laudenslayer and Fargo 1997; Willy 1992).
Traps were baited with raw oats and sunflower seeds coated with peanut butter and
opened at dusk and checked at dawn. Synthetic batting was provided for thermal
insulation. Traps were set at all houses within each study site.
44
Chipmunks
Chipmunk species in PNF display considerable overlap in habitat requirements, diet, and
activity. Both long-eared and Allen’s chipmunks are captured frequently in our livetrapping efforts. These species overlap greatly in external characteristics and
consequently can be challenging to identify in the field (Clawson et al. 1994; Gannon and
Forbes 1995). To date, the only sure means to identify some individuals appears to be
with skeletal features, which are obtained only by sacrificing animals. We collected a
sample of reference chipmunks throughout PNF by salvaging trap mortalities at longterm grids as well as euthanizing a small portion of animals from land bird transects (≤3
chipmunks per census transect). To avoid affecting capture-recapture data, animals were
only collected on the last day of the trapping session. All specimens were frozen and
submitted to the University of California, Davis, Museum of Wildlife and Fish Biology,
where they have been prepared as standard museum specimens (full skeleton plus skin)
and tissues (e.g., liver, heart, muscle, and kidney) have been preserved for use in
molecular analyses. We have collaborated with Dr. Jack Sullivan (University of Idaho) to
develop molecular markers for non-lethal identification of chipmunk species in the
future. We are collecting tissue samples (small sections (< 1 mm) of ear pinna, frozen in
95% ethanol) from all chipmunks captured at long-term grids and land bird transects.
During 2005-2008, we recorded the presence of six external morphological
characteristics that have been suggested to visually distinguish between the two species.
These are: ear patch size and color; face stripe color and curvature; length and shape of
the ear; and body color. We will use these data to determine characteristics that reliably
distinguish these species in the field, thereby allowing us to proceed with analyses of
habitat use.
Animal handling
Captured animals were transferred to a mesh handling bag, identified to species,
individually marked with numbered Monel ear tags (National Band & Tag Co., Newport,
Kentucky), weighed, aged, measured (e.g., ear length, hind foot length), examined for
reproductive status, and released at the point of capture. Total processing time for an
experienced technician was generally <2 minutes. Reproductive condition for males was
noted as either scrotal (enlarged and scrotal testes) or non-scrotal (reduced and abdominal
testes); for females, the vagina was noted as either perforate (thereby receptive) or
imperforate (not receptive), the vulva as either swollen or not, and the animal as lactating
(nipples were enlarged and/or reddened, reflecting nursing offspring), or not. Animals
were aged based upon a combination of weight, pelage (juvenile: gray, subadult:
intermediate, and adult: brown), and reproductive condition (juvenile/subadult:
nonreproductive, adult female: pregnant/lactating, and adult male: scrotal).
At initial capture, a tissue sample was collected from each animal. Tissue samples were
collected by snipping the terminal 1 mm of ear tissue using sterile surgical scissors and
placing the tissue in a cryovial with 95% ethanol. Samples were placed in a freezer for
long-term storage. Prior to 2006, we collected tissue samples from dusky-footed
woodrats and chipmunks. In 2006, we collected tissue samples from all captured
animals. In 2007, we collected tissue samples from chipmunks and northern flying
45
squirrels. During the field season of 2008 tissue samples were collected from chipmunks
only.
Any trap mortality, including incidental trap deaths, is thoroughly documented, and
specimens are frozen and submitted to the University of California, Davis, Museum of
Wildlife and Fish Biology, in accordance with the permitting requirements of the
California Department of Fish and Game. All field work and handling procedures are
approved by the University of California, Davis Animal Use and Care Administrative
Advisory Committee protocol (#10394), and meet guidelines recommended by the
American Society of Mammalogists (Animal Care and Use Committee 1998; Gannon et
al. 2007).
Radiotelemetry
Movement data obtained from the radiotelemetry methods described herein allow us to
measure home range, movement patterns, and social organization of individuals, permit
the detailed measurement of habitat use and selection, and document the location and
frequency of use of denning, nesting, and resting sites (Lancia et al. 1996, Litvaitis et al.
1996). Radiotelemetry methods are useful for making comparisons of small mammal
movements and space use across time, locations, habitats, and land-use treatments. We
applied radiocollars to a subset of dusky-footed woodrats and northern flying squirrels
and radiolocated them during daytime resting activities and at night during foraging
activities.
Radiotransmitter application
During 2003-2006, we applied radio transmitters to northern flying squirrels and duskyfooted woodrats. In 2007 only northern flying squirrels were radiotagged. A 4.0 g collartype radio transmitter (Holohil Systems Ltd., Model PD-2C) was placed on the neck of
individuals. Woodrats were lightly sedated with ketamine hydrochloride (100mg/ml)
injected into the thigh muscle to facilitate application of radio-collars. Prior to being
released at the point of capture, woodrats were allowed to fully recover from anesthesia
(4-5 hours). Northern flying squirrels were not anesthetized prior to radiocollaring in
earlier years; however in 2007, the use of anesthesia was implemented to avoid handlingrelated stress and mortality. A mild anesthetic plane was achieved by administering 0.81.0ml of a low concentration mixture of ketamine hydrochloride (100mg/ml) and
xylazine (100mg/ml) (1.4mg ketamine, 0.06 mg xylazine) in a saline solution. Flying
squirrels were then released after application of the radiocollar and full recovery at their
point of capture. Radiotelemetry activities of newly collared individuals were initiated
after a 24-hour acclimation period succeeding their release.
Homing
To document the location and frequency of use of denning, nesting, and resting sites we
used homing techniques. For northern flying squirrels, diurnal locations were determined
once per day, sporadically in 2003-2005 and 1-2 days per week in 2006, and 1 day per
week in 2007. For dusky-footed woodrats, diurnal locations were determined once per
day, sporadically in 2003 and 3 days per week in 2004 and 2005, and 1-2 days per week
46
in 2006. Locations were marked and accurately (≤ 1 m) mapped using a Trimble GPS
unit.
Triangulation
Nocturnal telemetry sessions using triangulation techniques occurred during 5 nights per
month in 2003 and 8-10 nights per month during 2004-2007. We used a Yagi® antenna
and a hand-held radiotelemetry receiver (Model R-1000, Communications Specialists,
Orange, CA, USA) to obtain the location of radiocollared animals. Compass bearings for
the radio-collared animal were obtained by using a hand-held compass and bisecting the
signal drop-offs. Fixed telemetry stations, mapped to within 1 m accuracy using a
Trimble GPS unit were located remotely from the transmitter’s position to avoid
disturbance of the radio-tagged animal. Technicians worked in synchronized teams to
achieve 3-6 directional bearings within as short of a time interval as possible (typically
<10 minutes). Radiolocations were obtained for each animal 2-3 times per night at a
minimum of 2.5 hours and 2 hours apart for dusky-footed woodrats and northern flying
squirrels, respectively, to avoid serial correlation (Swihart and Slade 1988, Taulman and
Smith 2004). The timing of nightly telemetry was varied from dusk until dawn to ensure
that radiolocations were sampled at different times of activity. To reduce the error due to
differences among field technicians, technicians were thoroughly trained and rotated
among stations and study sites each radiotelemetry session. To ensure the accuracy of
the triangulation method, triangulation systems were tested each night during regular
radiotelemetry activities using 1-2 “dummy” collars placed within each study area;
technicians did not know dummy collar locations, and the dummy collars were moved
once per week. To assess bearing error rates, dummy collar locations were determined
through triangulation and compared to their actual location previously mapped with a
Trimble GPS unit.
Program Locate III (Nams 2006) was used to calculate northern flying squirrel locations
from bearing data obtained during triangulation. We used several criteria to evaluate
bearing data and determine animal locations. These included convergence of bearings,
presence of outliers, number of bearings (≥ 4), and signal quality. Signal quality was
reported for each reading as good, intermediate, or poor by technicians during
triangulation. All poor readings were excluded from analysis. Bounce was an issue that
contributed to error. To address this issue, we removed the two most divergent bearings
until no fewer than 4 bearings were used for triangulation. This was possible because in
2007 we typically took 6 simultaneous bearings for each animal. Accepted locations
were analyzed in Ranges VI. We estimated home range (95%) and core range (50%)
using the Minimum Convex Polygons (MCP) (Mohr 1947) and Fixed Kernel (FK)
methods (Kenward 2001).
Home range analysis
Northern flying squirrels.--To help describe home range area and identify
important habitat features for the northern flying squirrel, we employed the use of
radiotelemetry. Using simultaneous bearings (triangulation), a pinpoint of an animal’s
location may be achieved (see Fuller et al. 2005). Data-points generated from
triangulation were subsequently further scrutinized. To determine if a stable home range
was reached through incremental area analysis, data-points were added to a home range
47
in the order in which they were collected until an asymptote was observed (for example,
see Fig. 24). Home ranges stabilized after an average of 34 points were collected. This
was used to justify using data from flying squirrels that had fewer than 50 locations for
kernel analysis.
We then constructed home range models using RangesVI (Kenward et al. 2003). One
such model used is the minimum convex polygon (MCP- Mohr 1947) method (Fig. 22).
In this model, a home range can be defined by drawing lines that connect the outermost
points to form a contained area. The second model generates a home range using kernel
densities. The fixed kernel (FK- Worton 1989) method uses a utilization distribution, or
focal point of activity to define the home range (Fig. 23). This added characteristic
weighs areas with a high density of points more heavily than areas with fewer points to
generate “core use areas” (Silverman 1986, Worton 1989). Sampling in the core use area
reveals which areas within a home range are important to the northern flying squirrel with
respect to habitat type. Both 95% FK and 95% MCP home range models were generated
for 2006 and 2007 (Table 6). Mean home range area across both years was 8.85 +/- 1.03
hectares (95% MCP) and 11.00 +/- 1.51 hectares (95% kernel). MCP was reported to
compare with previous studies, and kernel estimators will be used for all subsequent
analyses. Additionally, core use areas were generated using a 50% kernel estimator.
Home ranges are currently being used to determine habitat preference of the flying
squirrel through the implementation of compositional analysis (Aebischer et al. 1993).
Compositional analysis tests for differences between use and availability of habitat. A
two–scaled approach is used; the broad scale compares the study area (availability) to the
home range of each animal (use), and a second and finer scale comparison utilizes the
home range (availability) and points within the home range (use). Statistical tests in
compositional analysis use Multivariate Analysis of Variance (MANOVA) to compare
the proportions of available habitat to the proportions of used habitat (Aebischer et al.
1993).
Dusky-footed woodrats.— Radiotelemetry techniques were employed to
determine the space use of woodrats during 2004 to 2006 at study sites WR-1 and WR-2
(Fig. 2). We used the maximum-likelihood estimator method (Lenth 1981) in the
software program Locate III (Nams 2006) to estimate locations and error ellipses for
triangulations. We excluded all triangulations for which >50% of bearings received a
rank of low confidence. Locations ≥1 km from the station to the transmitter also were
excluded (Schmutz and White 1990). We used RangesIV (Kenward et al. 2003) to
calculate incremental area analysis, home range, core area, and overlap among
individuals. All analyses used a combination of nocturnal movement locations and
diurnal locations obtained from trapping and homing. Because a given woodrat was often
found multiple times at 1 house, we used only 1 diurnal location per house to avoid
biasing core area estimates towards house locations, resulting in about 80% of locations
being nocturnal.
MCP and FK methods were used to calculate home range and core area. MCP home
range (95%) and core area (50%) were calculated using the arithmetic mean (Nams
2006). Incremental plots of home range size versus number of locations were inspected
48
for each individual using Ranges6 to check that the range area reached an asymptote; if
an asymptote was not observed, then that individual was excluded from further analysis
(Kenward 2001). We found that a minimum of 16 (mean=24.5±1.3) locations was
required to reach an asymptote in home range area using MCP. In addition, woodrats that
were radiocollared for <30 days were also excluded because of the short duration.
Application of these criteria resulted in the exclusion of 12 collared woodrats (2 in 2004;
7 in 2005; 2 in 2006), all of which appeared to have been killed by predators shortly after
collaring. In addition, 5 collared woodrats (2 in 2004; 3 in 2005; 0 in 2006) were
excluded from analyses because they were transient or resided outside of the study areas.
Fixed kernel volume contours (95% home range, 50% core area) were calculated utilizing
the least-squares cross-validation method in Ranges6 for those animals with ≥30
locations (Seaman et al. 1999, Millspaugh et al. 2006); application of this criteria resulted
in the exclusion of 17 additional individuals for FK analyses.
We calculated an index of overlap (OI; Minta 1992), with possible values ranging from 0
(no overlap) to 1 (100% overlap). For each study site and year, we calculated OI for
home ranges and core areas for each male-male, male-female, and female-female pair.
We only included woodrats whose home ranges overlapped with ≥1 other home range.
All overlap calculations were based on MCP home ranges and core areas, rather than FK,
because we did not want to exclude any potentially interacting individuals from overlap
calculations, and to facilitate comparison with previous studies (e.g., McEachern 2005).
We assessed synchronous and asynchronous sharing and successive occupancy of houses
by all radiocollared woodrats based on diurnal locations, when woodrats are inactive
within their houses. Duration of house sharing was determined by assuming sharing
occurred between successive radiolocations. We examined placement of houses within
core areas using FK because it relies on probability distributions, which indicate areas of
intense use (Seaman and Powell 1996).
All statistical tests were performed using JMP IN 5.1.2 (SAS Institute 2004) and
significance was set at α=0.05 and Bonferroni-corrected for multiple comparisons, when
appropriate. Only 10% of woodrats were radiocollared for two consecutive years, thus we
considered data from different years to be different samples. Differences among groups
were analyzed using analysis of variance (ANOVA) after transformation to meet
assumptions of normality (Kutner et al. 2005). The Wilcoxon rank scores test was used to
test for differences between groups when data could not be transformed to meet
assumptions of normality.
Vegetation
Long-term grids
Coppeto et el. (Publications #1 and #4) provides a detailed analysis of the macro- and
microhabitat associations of the full compliment of small mammal communities within
18 long-term grids established within 5 habitat types in PNF during 2003-2004. A subset
of the data that Stephanie A. Coppeto collected represents pre-treatment vegetation
conditions within the 9 of the 12 experimental grids. In 2008 the Mammal Module staff
49
conducted a full resample of the 12 experimental grids. By strictly adhering to the
macro- and microhabitat vegetation sampling protocols developed in Coppeto et al.
(2006) we are able to fully analyze the effects of different forest management practices
performed in PNF on the short-term responses of small mammal communities,
vegetative cover, and their associations.
Vegetation Sampling. — In order to investigate immediate short-term responses
of small mammal abundance in the fuel reduction treatments of the PNF, we
systematically sampled for 23 micro and 27 macrohabitat variables within the
experimental grids during the 2008 field season. 1440- 1 meter radius (3.14m2) ground
cover plots, 216 point centered quarter plots, and 1440 canopy photos, with generous help
from the vegetation module, were collected. The suite (Table 1.) of microhabitat
variables were collected using ocular estimations of vegetative cover in a layered fashion
at or below breast height (1.4 meters) at each trap station (n=120) within a grid. Any
understory vegetation occurring above breast height was sampled as canopy cover and
included in a photo analysis. Canopy cover data was collected using a Nikon® digital
camera affixed with a hemispherical lens mounted on an adjustable tripod and analyzed
using Gap Light Analyzer, Version 2.0 (Frazer et al. 2000). The assemblage totaled 1.4
meters in height. Point centered quarter tree sampling was conducted in a stratified
manner within each grid (n=18). The nearest tree (> 10 cm. diameter at breast height
(DBH)) in each quadrant was sampled. Species, slope, aspect, distance, DBH, and height
of each individual were collected. Slope was determined through the use of a clinometer,
aspect with the use of a compass, distance using a transect tape, and DBH through the use
of a pre-calculated DBH tape. Heights were determined by first training technicians,
offsite, for ocular estimation. This was accomplished by measuring tree heights using a
clinometer and meter tape and having technicians estimate heights prior to revealing
actual tree heights to them. Tree heights were placed into five meter interval groups.
Cone Counts.—To evaluate the effects of conifer seed production on small
mammal abundance, we measured cone production during fall of 2003, 2004, 2006,
2007, and 2008 using 10 randomly selected individual trees of each species on each longterm grid. For this we selected mature dominant or codominant trees with pointed
crowns, as tall as or taller than the surrounding canopy, sufficiently far apart that their
crowns did not touch. For grids with <10 individual trees of a given species, additional
trees were found as close to the grid as possible (<500 m). The same trees were counted
in each year within the same 2-wk period to prevent confounding temporal factors.
Counting was performed by standing at a distance of ≥1.5x the tree height and visually
counting cones using binoculars. For each tree we recorded tree height, diameter at
breast height (DBH), species, and crown class. Temporal differences in cone production
were determined using repeated measures analysis of variance (rmANOVA) with year,
habitat type, and species as treatments, and individually counted trees as the repeated
measure.
Northern flying squirrels
Den use.—We documented northern flying squirrel den locations during homing
activities, and a number of measurements were taken at these dens to determine the
50
micro-habitat preferences of flying squirrels. These data were used to test for tree use
versus availability. DBH, species, condition (live tree, snag), den height, and type (cavity
or external) of each den tree were recorded. We measured habitat characteristics at den
locations and paired random points. Den plots were centered on the den tree, and paired
with a plot whose outer edge intersected the outer edge of the den plot. All trees ≥10 cm
DBH within an 18 m radius (0.1 ha) were measured and species recorded. Additionally,
decay characteristics (fungi present, cavities) were noted and epiphyte loads estimated
according to the methods of Bakker and Hastings (2002) to see if northern flying
squirrels showed any preferential selection of den trees within sites. All trees <10 cm
DBH were tallied. Estimates were taken of ground cover to the nearest percent.
Dominant over- and understory trees were recorded as well. Spherical densiometers were
used to take canopy measurements in a randomly selected direction at the edge of the
plot, with 3 successive measurements at 90° from the first. Canopy readings were also
taken at the plot center. Two randomly chosen transects were used to estimate coarse
woody debris. Degree of decay, length, diameter at both ends, and species were
recorded. All woody debris ≥10 cm diameter at the largest end were measured and
recorded. Percent slope at each site was estimated using a clinometer. In 2006, 38 flying
squirrel dens and paired comparison plots had measurements taken. In 2007,
measurements were taken at 40 dens and comparison plots.
Dusky-footed woodrats
Tree house characteristics and use.— We examined tree house characteristics and
use during 2004 to 2006 at 2 study sites WR-1 and WR-2 (Fig. 3). Ground houses were
those located on the soil surface or on downed wood. Tree houses were characterized as
either built within a tree cavity or externally on limbs. For all tree houses, we recorded
whether the tree was alive or a snag and the species of live trees. We measured diameter
at breast height (dbh; cm) of a random sample of the trees in which houses were found
(88% and 83% of house trees at study sites 1 and 2, respectively). We determined tree
availability by counting all trees and snags (≥5 cm dbh) in randomly located, 4-m radius
circular plots (72 at site 1, 77 at site 2), and recorded tree and snag characteristics for
each plot. We based house use analyses on radiotelemetry locations during the daytime
period of inactivity determined using homing. For each woodrat, we calculated the
proportion of radio locations occurring at each house type (ground or tree), then averaged
across individuals and years by sex. We tested for differences in tree house use between
sexes each month using the Wilcoxon rank scores test. Because we found no difference in
proportional availability and use of houses between sites, results from the 2 study sites
were combined for all analyses.
RESULTS AND DISCUSSION
We have been making steady progress towards our objectives. In 2008, we completed
several projects. In addition to successfully completing an extensive (1 May-1 October)
field season, our study module has produced quality peer-reviewed publications and other
products. In 2008, we had 2 manuscripts in publication, 1 manuscript in press, and
several more in preparatory stages. We have chosen to present the abstracts of our
published, submitted, or in preparation manuscripts herein as a representation of the work
that we have completed to date.
51
2008 Field Season
During the 2008 field season we marked a total of 543 individuals over 1103 captures of
9 species. Predominant species in the study area were deer mice, brush mice, long-eared
and Allen’s chipmunks, California ground squirrels (Spermophilus beecheyi), goldenmantled ground squirrels, Douglas squirrels (Tamiasciurus douglasii), long-tailed voles
(Microtus longicaudus), California red-backed vole (Clethrionomys californicus), and
northern flying squirrels. Incidental mammals captured included shrews (Sorex sp.),
striped skunk (Mephitis mephitis), snowshoe hare (Lepus americanus), and western gray
squirrels (Sciurus griseus).
In 2008, abundance of Peromyscus sp. and Tamias sp. fell from levels observed in 2007
(Fig. 9 and 11 respectively). We noticed a marked increase in capture rate of northern
flying squirrels at long-term grids in 2005, 2006, 2007, and 2008 (Fig. 13). However,
during this time woodrat abundance at long-term grids steadily declined steadily (Fig.
15), and no woodrats were captured in 2008. The 2008 field season represents our first
full year of post-treatment data collection. Trends in abundance of Peromyscus sp.,
Tamias sp., northern flying squirrel, and dusky-footed woodrat within treatment plots
have mimicked trends observed within control plots (Fig. 10,12,14, and 16 repectively).
Year continues to be a strong determining factor in small mammal abundance (e.g., Fig.
17). Long term monitoring of the experimental plots should distill the effects of
treatment from the effects of year. Across all years sampled, biomass of PNF small
mammal communities was greatest within red fir habitat (Fig. 18). Large abundances of
large small mammal species, such as golden mantled ground squirrels and chipmunks,
accounted for 42% and 49%, respectively, of the biomass in red fir habitat which
contained 24% -36% greater total biomass than other forest types sampled (Table 3).
Total biomass of key California spotted owl prey species was greatest in mixed fir habitat
(Fig. 19) by 4%-8% over other forest types (Table 4). However, during 2008, in pinecedar forest type all northern flying squirrels were captured on grid #7 in a gulch
consisting of a mixed fir habitat. An increase in overall abundance trends of larger
species, such as chipmunks and northern flying squirrels, has been observed (Fig. 8) and
assumedly contributed significantly to the marked increase in biomass observed in all
experimental plots post treatment (Table 5). Light thin and heavy thin plots increased at
similar proportions to control sites. Conversely, group select sites exhibited a 30%
increase in biomass compared to control plots. However, 75% of group select site
biomass was contributed from a single white fir habitat site, “Toast” (grid #19) (Fig. 20).
Figure 21 illustrates community abundance of closely clustered groups of control and
experimental plots. The “Miller Fork” group, which contains group select “Toast”,
displayed 52% greater abundance in 2008 than group clusters “Dean’s Valley” and
“Snake Lake”. The skew in 2008 group cluster abundances could explain a great deal of
the marked increase in abundance and biomass in group selection plots.
Long-term grids
One of our objectives for the long-term grid data is to determine small mammal habitat
associations at macro- and microhabitat scales (Objective #1). We have examined this at
our long-term grids and include this summary herein (Publications #1 and #4). Another
objective for our long-term grid data was to determine small mammal population trends,
52
evaluate how populations are changing temporally, and assess the factors responsible for
the observed trends (Objective #5). We have documented the dynamics of small mammal
abundance at long-term grids since 2003, and we have currently evaluated trends using
data from 2003-2004, and include this summary herein (Publication #8). In 2007, the
planned treatments were implemented and data on small mammals were collected
immediately after the treatments were completed. During the 2008 field season we
collected the first full year of post-treatment data. We will analyze data obtained at longterm grids pre-treatment (2003-2006) and post-treatment (2007-funding availability) to
assess the impacts of forests management treatments on small mammal abundance and
species diversity (Objective #4).
Publications #1 and #4: Habitat associations of small mammals at two spatial scales
in the northern Sierra Nevada
Effective management strategies require an understanding of the spatial scale at which
fauna use their habitat. Towards this end, small mammals were sampled in the northern
Sierra Nevada, California, over 2 years (2003-2004) at 18 live-trapping grids among 5
forest types (Fig. 1a). Macrohabitats were defined by overstory tree composition, and 19
microhabitat variables were measured at all trap stations (Table 1). Macrohabitat and
year explained 93% of variation in abundance of deer mice (Peromyscus maniculatus),
whereas 69% was explained by microhabitat and year. Variation in abundance of Tamias
sp. (long-eared and Allen’s chipmunk) was slightly better explained by microhabitat and
year (70%) than by macrohabitat and year (67%). Red fir forests supported significantly
more mice and chipmunks than mixed conifer and pine-cedar forests, and more
chipmunks than mixed fir forests. Five of 6 uncommon species were significantly
associated with macrohabitat type; golden-mantled ground squirrels, northern flying
squirrels, and Microtus sp. (long-tailed vole–M. longicaudus; Mountain vole–M.
montanus) were captured almost exclusively in red fir forests, whereas dusky-footed
woodrats and California ground squirrels were found in pine-cedar, mixed fir, and mixedconifer forests. The first 2 axes of a canonical correspondence analysis on microhabitat
variables explained 71% of variation in combined small mammal abundance.
Microhabitat associations varied among species but were driven primarily by canopy
openness, shrub cover, and shrub richness. Although much of the small mammal fauna
appeared to select habitat at both spatial scales studied, CCA using macrohabitat as a
covariate revealed that microhabitat explained much less of the variation in small
mammal abundance than did macrohabitat. Still, the strongest scale of association may
be species-dependent and hierarchical in nature.
Publication #8: Population dynamics of small mammals in relation to cone
production in four forest types in the northern Sierra Nevada
We studied the small mammal assemblage in 4 forest types (white fir, red fir, Douglas fir,
and ponderosa pine) in the Sierra Nevada of California for 2 consecutive field seasons
(2003-2004). We also assessed cone production by dominant conifer species in both years.
Cone production was greater overall in fall 2003, but varied within forest type and
between conifer species (Fig. 4). Parallel to this, mean maximum densities of deer mice
increased in 2004 (from 0.7 - 7.3 ind./ha to 65.7 - 112.7 ind./ha; Fig. 5). Numbers of
golden-mantled ground squirrels were similar in both years, and displayed the typical
53
pattern of a hibernating species, with low densities in May (6.6  0.2), peak densities in
September (24.5 – 32.5 ind./ha), and declines in October (9.2  4.8; Fig. 6). Long-eared
chipmunks reached higher densities in red fir (48.2  13.4 ind./ha) and Douglas-fir forests
(36.0  13.5 ind./ha) than in white fir forests (7.6  2.7 ind./ha), and all populations
peaked in September. Allen’s chipmunk remained at lower densities than long-eared
chipmunks except during September 2004, when populations of the former reached high
densities (54.6  26.8 ind./ha; Fig. 7). Survival of deer mice was dependant on an
interaction between forest type and month with additive effects of winter and 2003 fall
mean cone production. Golden-mantled ground squirrel survival varied by month
whereas survival in both species of chipmunk varied by an interaction of forest type and
month + winter (Table 2). Dusky-footed woodrats were present at lower elevations and
reached greatest densities in ponderosa pine forests. Northern flying squirrels were
uncommonly captured and found predominantly in red fir forests.
Publication #9: Trapping rodents in a cautious world: the effects of disinfectants on
trap success.
Recommendations for hantavirus prevention include disinfecting traps that have captured
small mammals. However, the potential effects of disinfection on small mammal
trappability have not been thoroughly investigated. We conducted an experiment to
compare the effects of 2 disinfectants (Lysol and household bleach) on trappability of 3
small mammal species (deer mice, chipmunks, and golden-mantled ground squirrels).
We established triplicate trap grids in 2 forest types (red fir and mixed conifer), each
consisting of a 6 x 6 array of Sherman live traps placed at 10 m intervals. Traps were
given 1 of 3 treatments: control (water), Lysol, or bleach; and were placed such that the 3
treatments alternated in a regular pattern. Traps were run for 4 consecutive nights with
application of each treatment daily. We found a difference in the trappability of deer
mice between years; however we did not detect a statistically significant difference in
trappability due to disinfection for any of the 3 study species. Within deer mice,
disinfectant effects on capture probability were not supported by model selection in
Program MARK. These results indicate that although populations may fluctuate
temporally and spatially, trap disinfection does not have a significant effect on small
mammal trappability.
Northern flying squirrels
We have captured and radiotracked northern flying squirrels since 2004 in an effort to
evaluate the abundance and distribution, habitat use, and home range of this important
species (Objective #6). We have examined data from 2004-2005 and include this
summary herein (Publication #7). We continued these efforts during 2006 and 2007 to
increase our sample size and improve our statistical power; 2007 marks the final year of
northern flying squirrel radiotracking. Data from 2006 and 2007 will be included in an
additional publication (Publication #3).
54
Publication #7: Home range and activity of northern flying squirrels in the northern
Sierra Nevada
We studied the northern flying squirrel in PNF using radiotelemetry. Fourteen northern
flying squirrels from 2 forest types (mixed conifer and red fir) were fitted with
radiocollars and provided sufficient locations for home range analysis. We used 95%
adaptive kernel and 95% minimum convex polygon (MCP) analysis to determine home
ranges. No sex differences and no differences in forest type were observed for home
range size. Mean kernel home range size was 25.7 ha for all squirrels. Mean distance to
the nearest nest tree did not vary throughout the night; however, females tended to travel
greater distances from nest trees.
Publication #3: Home range and habitat selection of northern flying squirrels in the
northern Sierra Nevada
Average home range size for northern flying squirrels during 2006 using 95% Minimum
Convex Polygon (MCP) was 9.12 ha ± 2.41 and using 95% Fixed Kernel (FK) was 13.89
ha ± 4.18. In 2007, average home range size for northern flying squirrels using 95%
MCP was 8.73 ha ± 1.13 and using 95% FK was 9.83 ha ± 1.32. Mean home range size
across years (2006-2007) using 95% MCP is 8.84 ha ± 1.02 and 10.99 ± 1.51 using FK.
Each year, females were larger than males (2006: fem = 122.2 g, male = 102.0 g; P
<0.0001; 2007: fem = 129.7 g, male = 103.6 g; P = 0.0039); however, home ranges of
females and males were similar (P = 0.41).
Most dens (n=53) were located in cavities (49%), but some were external stick nests
located on the limbs of trees (12%); 39% could not be identified because they were not
visible to the observer. Preliminary results obtained using 53 dens and 53 paired random
plots indicate that dens were distributed amongst various tree species and size classes.
Most den trees were located in white fir (28%) and California black oak (26%; Table 7).
However, comparison of use and availability indicate that California black oak may be
used preferentially for den sites (Fig. 25). Many den trees were located in large
sawtimber (≥53.4 cm dbh, 44%), but poletimber (10-27.9 cm dbh, 35%) and small
sawtimber (28-53.3 cm dbh, 21%) were also used. Comparison of use and availability
indicate that northern flying squirrels are using larger trees than those available (Table 7).
Dusky-footed woodrats
We have captured and radiotracked dusky-footed woodrats since 2003 in an effort to
evaluate the abundance and distribution, habitat use, and home range of this important
species (Objective #6). To date, we have examined vegetation data obtained during
2004-2005 and include this summary herein (Publications #2, #5, and #6). In 2007, we
prepared a manuscript on the spatial organization of dusky-footed woodrats (Publication
#11). The 2006 field season marked the final year of data collection, so that we might
focus our efforts on northern flying squirrel ecology during 2007 and analyze data
obtained on woodrats from previous years.
55
Publication #2 and #5: Habitat selection by dusky-footed woodrats in managed
mixed-conifer forest of the northern Sierra Nevada
Dusky-footed woodrats are important components of forest communities, including
serving as a primary prey of the California spotted owl, a species of concern in
California. We examined the macro- and microhabitat associations of the dusky-footed
woodrat at 4 study sites within mixed-conifer forest of the northern Sierra Nevada,
California, during 2003–2005. We investigated the importance of California black oak as
a macrohabitat component for woodrats, and we examined microhabitat selection at 2
levels, house location and house use, by comparing house-site (n = 144) characteristics to
random sites (n = 144) and characteristics of used and unused houses, respectively. We
found a strong trend towards a positive relationship between woodrat density and large
(≥33 cm diameter at breast height) oak density, suggesting that large oaks are an
important macrohabitat component for woodrats, probably because of their value as a
food resource. At the microhabitat scale, house location was strongly influenced by the
presence of large (≥30 cm diameter at root collar) stumps, but also by abundance of logs,
steeper slopes, and lack of bare ground and mat-forming shrub cover. Houses used by
adults were not distinguishable from unused houses on the basis of microhabitat
variables, suggesting that woodrats make decisions about microhabitat conditions at the
time a house is built. Adult and subadult woodrats selected houses with different
microhabitat characteristics, but this pattern was not consistent between years. In 2005,
adults chose larger houses that were characterized by more logs and less poletimber, but
we detected no such differences in 2004. Dusky-footed woodrats in the northern Sierra
Nevada would benefit from management techniques that promote the growth and
retention of large California black oaks and create abundant dead wood within a stand.
Publication #6: Characteristics and use of tree houses by dusky-footed woodrats in
managed mixed-conifer forest of the northern Sierra Nevada
Dusky-footed woodrats are important components of forest communities, including
serving as a primary prey of the California spotted owl, a species of concern in
California. Because previous studies have focused on the more “typical” ground houses,
little is known about tree houses, perhaps because their inconspicuous nature makes them
difficult to locate (Fargo and Laudenslayer 1999). Our objective was to describe locations
of tree houses and determine if dusky-footed woodrats used these houses preferentially.
Most houses (n=252) were located on the ground (58%), but many were also located in
cavities of trees or snags (27%) or on the limbs of live trees (15%). Three houses were
located aerially in shrubs (hence neither ground nor tree), and were excluded from
analyses. Tree houses were located primarily in white fir, Douglas-fir, California black
oak, and snags (Table 8). Comparison of use and availability suggests that white fir were
preferred as locations for houses constructed on limbs. White fir were mostly smaller,
understory trees with splayed branches suitable for supporting the woody debris used in
house construction. Large California black oaks and snags were strongly preferred as
sites for cavity houses, probably because their size and tendency to decay resulted in
formation of cavities of sufficient size for constructing houses.
56
Individual woodrats used as many as 3 tree houses and 8 ground houses, and use of tree
houses was common, with 70% of males and 73% of females using at least 1 tree house.
We expected that tree houses might provide increased protection from predation, because
houses on the ground were vulnerable to destruction by black bears, or provide better
access to arboreal food sources. However, woodrats did not spend more time at tree
houses than expected on the basis of availability (Table 9). Among tree houses, cavity
locations seemed preferred to limb locations when compared with availability, perhaps
because cavity locations were more protected, and there was some evidence that females
used cavity locations more frequently than did males (Table 9). Use of tree houses
increased during the late summer with a peak in October (Fig. 26), possibly because mast
availability in the fall increased arboreal foraging opportunities. Females used tree houses
more frequently than did males during June (Z = −2.13, P = 0.032) and July (Z = −0.2.22,
P = 0.026), coincident with the time of reproduction, perhaps because tree houses offer
enhanced protection for unweaned offspring.
Our results suggest that tree houses are a prevalent and frequently used resource for
dusky-footed woodrats in mixed-conifer forest of the northern Sierra Nevada. Tree house
use is most prevalent during late summer and fall, and large California black oaks and
snags are the most important forest elements for tree house location because of the
protected sites provided by their cavities.
Publication #11: Spatial organization of dusky-footed woodrats in managed mixedconifer forest of the northern Sierra Nevada
We studied the spatial organization of dusky-footed woodrats (Neotoma fuscipes) in
mixed-conifer forest of the northern Sierra Nevada, California by radiotracking 63 adult
woodrats at 2 study sites (Fig. 2.) during May–Oct, 2004–2006. Home range and core
area estimates differed between study sites, but they were within the range reported
elsewhere; variability in home range size was explained in part by density (Table 10.).
Woodrat home ranges overlapped with multiple neighboring woodrats (Fig. 27.), both
same-sex and opposite-sex, suggesting that foraging areas were shared. However, core
areas showed little overlap between same-sex neighbors. Woodrats occupied multiple
houses and frequently moved among them, and sharing of houses (either simultaneously
or nonsimultaneously) with neighboring woodrats was common but occurred mostly
between male-female pairs (Fig. 28.). Females typically shared their core area and
houses with 1 male, whereas males shared core areas and houses with multiple females;
further, males moved more than females. Our results suggest that dusky-footed woodrats
are semi-territorial, maintaining near-exclusive use of their core area and houses against
same-sex conspecifics, and that the mating system likely is polygynous.
Golden-mantled ground squirrels
We captured and radiotracked golden-mantled ground squirrels during 2003-2005. Data
analysis and manuscript preparation took place in 2009; no additional data has been
collected since 2005. The following summary (Publication #8) represents the culmination
of this work.
57
Publication #8: Spatial-organization and dispersal of golden-mantled ground
squirrels (Spermophilus lateralis): questioning asociality in a small spermophiline.
Social-organization and dispersal patterns indicate that Spermophilus lateralis exhibit
single family female kin clustering rather than complete ascociality. A great deal of
home range overlap (FK- mean IO 0.31 ± 0.20 SD, range 0.00-0.69; MCP- mean IO 0.28 ±
0.22, range 0.00-0.81) and core area territoriality (FK- mean IO 0.03 ± 0.08 SD, range
0.00-0.43; MCP- mean IO 0.01 ± 0.03, range 0.00-0.17) is observed. 40% of juveniles
remained philopatric in their first year. Juvenile males dispersed more often (75%)
whereas juvenile females tended to remain philopatric (62%). Dimorphic dispersal
behavior in a population where males are dispersing further and more often than females
results in increased female kin proximity, or “clustering”, promoting social interaction.
Further clarification of kin and non-kin spatial organization and interaction will be
necessary to determine whether S. lateralis should truly be placed at Michener’s (1983)
social grade of 1-“asocial” or 2-“single family female kin clusters”.
Chipmunks
We have live-trapped chipmunks at long-term grids, land bird grids, and flying squirrel
transects since 2003. One of our objectives was to evaluate the habitat affinities of 2
species found commonly in PNF, long-eared and Allen’s chipmunks, using data obtained
from long-term grids during 2003-2004 (Objective #8). The following (Publication #10)
is a summary of these results.
Publication #10: A multiple spatial scale perspective of the habitat affinities of
sympatric long-eared and Allen’s chipmunks.
Sympatric species that are similar in body mass, diet, and general resource utilization are
likely to compete locally. Similar species often coexist by partitioning habitat. However,
detecting differences in habitat affinities is influenced by spatial scale. We investigated
the habitat associations of two ecologically similar chipmunk species – the long-eared
chipmunk and the Allen’s chipmunks – at three spatial scales in the northern Sierra
Nevada, California. Locally, we censused these species over two years at 18 trapping
grids, and recorded 19 microhabitat metrics at all trap stations. At a macrohabitat scale,
we assessed relative abundances at different study sites as a function of forest type.
Finally, at a landscape (e.g., geographic range) scale we examined digital vegetation
information and calculated extent of range overlap. At this largest spatial scale, both
species showed similar habitat affinities, with extensive overlap in distribution within the
Sierra Nevada. At the macrohabitat scale, both the species reached their highest mean
abundance in red fir forests but showed divergent secondary affinities. At the
microhabitat scale, however, habitat affinities differed significantly. Logistic regression
models indicate that microhabitat presence of long-eared chipmunks was associated
positively with open canopies, cover by rocks, and multiple sapling species, and
negatively with east and south facing, steep slopes. Allen’s chipmunks shared the
affinity for open canopies but differed in exhibiting a preference for traps on south facing
slopes with multiple shrub species, and aversion to traps on hard substrates covered by
litter and vegetation mats (e.g., Mahala mat—Ceanothus prostratus). Affinities at microand macrohabitat scales varied between sampling years, indicating that these species
58
retain a degree of flexibility in habitat associations while maintaining segregation and
minimizing the potential for competition.
2009 Field Season
We continued to capture, record external characteristics, and collect tissue samples from
chipmunks while performing live-trapping duties at long-term grids. In future analyses
we hope to evaluate our technique of determining chipmunk species using external
characteristics.
COLLABORATION
We have continued to maintain and improve collaborative efforts with all PLAS
Modules. Most notably, we improved collaboration with the Land bird Module in 2006
and 2007 by establishing temporary trapping grids at songbird census stations. We plan to
begin live trapping on the land bird transects once again in 2009 based on the availability
of funding. Vegetation and Fuels Modules have collected and continue to collect
vegetation, fire and fuels, and microclimate data within some portion of our long-term
and land bird trapping grids. In 2008 the Vegetation Module provided the Mammal
Module with training, equipment, and data in our effort to collect post-treatment
vegetative data in the long term grids. We are currently coordinating an effort in which
the Mammal Module will provide valuable feedback to the remote sensing analyses and
resultant models developed by the Fire and Fuels Module. In 2008, we continued to
collaborate closely with the directors of the University of California Davis McLaughlin
Reserve, Cathy Koehler and Paul Aigner, who provided space to train our field crew prior
to our housing becoming available at the University of California, Berkeley Forestry
Camp. In exchange for housing and training facilities, we provided information on the
abundance and distribution of small mammal species within a long-term study grid
established on the reserve. We collaborate with the University of Idaho for molecular
analyses to determine chipmunk species identification and worked together with them to
secure outside funding for these analyses. Lastly, we work closely with the University of
California Davis Museum of Wildlife and Fish Biology to preserve specimens for
research and educational purposes.
PUBLICATIONS
Theses
1. Coppeto, S. A. 2005. Habitat associations of small mammals at two spatial scales in
the northern Sierra Nevada, California. M.S. Thesis, University of California, Davis,
39 pp.
2. Innes, R.J. 2006. Habitat selection by dusky-footed woodrats in managed, mixedconifer forest of the northern Sierra Nevada. M.S. Thesis, University of California,
Davis, 31 pp.
3. Smith, J.R. In Prep. Home range and habitat selection of the northern flying squirrel
(Glaucomys sabrinus) in northeastern California. M.S. Thesis, University of
California, Davis. Winter 2009.
59
Peer-reviewed
4. Coppeto, S. A., D. A. Kelt, D. H. Van Vuren, J. A. Wilson, S. Bigelow, and M. L.
Johnson. 2006. Habitat associations of small mammals at two spatial scales in the
northern Sierra Nevada. Journal of Mammalogy 87:402-416.
5. Innes, R. J., D. H. Van Vuren, D. A. Kelt, M. L. Johnson, J. A. Wilson, P. A. Stine.
2007. Habitat selection by dusky-footed woodrats in managed, mixed-conifer forest
of the northern Sierra Nevada. Journal of Mammalogy 88(6): 1523-1531.
6. Innes, R. J., D. H. Van Vuren, D. A. Kelt. 2008. Characteristics and use of tree
houses by dusky-footed woodrats in the northern Sierra Nevada. Northwestern
Naturalist 89(2).
7. Wilson, J. A., D. A. Kelt, and D. H. Van Vuren. 2008. Home range and activity of
northern flying squirrels (Glaucomys sabrinus) in the Sierra Nevada. Southwestern
Naturalist.
8. Wilson, J. A., D. A. Kelt, D, H, Van Vuren, and M. Johnson. 2008. Population
dynamics of small mammals in relation to cone production in four forest types in the
northern Sierra Nevada, California. The Southwestern Naturalist 53(3): 346-356.
9. Innes, R. J., D. H. Van Vuren, D. A. Kelt, M. L. Johnson, and J. A. Wilson. In Prep.
Spatial organization of the dusky-footed woodrat (Neotoma fuscipes) in mixedconifer forests of the northern Sierra Nevada. Journal of Mammalogy. Winter 2008.
Submitted
10. Mabry, K.E., and Wilson, J. A. Submitted. Trapping rodents in a cautious world: the
effects of disinfectants on trap success. American Midland Naturalist.
In Preparation
11. Coppeto, S. A., D. A. Kelt, D. H. Van Vuren, J. Sullivan, J. A. Wilson, and N. Reid.
In Prep. Different scales tell different tales: niche conservatism vs. niche
differentiation in chipmunks in the northern Sierra Nevada. To be determined. Spring
2009.
12. Jesmer, B. J., D. A. Kelt, and D. H. Van Vuren. In Prep. Asociality: home range and
dispersal of golden-mantled ground squirrels (Spermophilus lateralis). To be
determined. Spring 2009.
PRESENTATIONS
1. Coppeto, S. A., D. A. Kelt, J. A. Wilson, D. H. Van Vuren, and M. L. Johnson. 2004.
Habitat selection by small mammals in the northern Sierra Nevada, California. Poster
to the American Society of Mammalogists, Annual Meeting, Arcata, CA.
60
2. Coppeto, S. A., D. A. Kelt, D. H. Van Vuren, J. A. Wilson, S. Bigelow, and M. L.
Johnson. 2005. Spatial scale and habitat use of small mammals in the northern Sierra
Nevada, California. Poster to the American Society of Mammalogists, Annual
Meeting, Springfield, MO.
3. Innes, R. J., D. H. Van Vuren, J. A. Wilson, D. A. Kelt, and M. B. Johnson. 2004.
Factors affecting the distribution and use of dusky-footed woodrat (Neotoma fuscipes)
houses. Poster to the American Society of Mammalogists, Annual Meeting, Arcata,
CA.
4. Innes, R. J., D. H. Van Vuren, J. A. Wilson, D. A. Kelt, and M. B. Johnson. 2005.
Space use and social organization of dusky-footed woodrats (Neotoma fuscipes) in
mixed-conifer forests of the northern Sierra Nevada. Poster to the American Society
of Mammalogists, Annual Meeting, Springfield, MO.
5. Innes, R. J., D. H. Van Vuren, D. A. Kelt, M. B. Johnson, J.A. Wilson. 2006.
Habitat relations of dusky-footed woodrats (Neotoma fuscipes) in mixed-conifer
forests of the northern Sierra Nevada. Poster to the American Society of
Mammalogists, Annual Meeting, Amherst, MA.
6. Smith, J.R.. 2006. Ecology of Glaucomys sabrinus: habitat, demography, and
community relations. Presentation to the American Society of Mammalogists,
Annual Meeting, Springfield, MO.
7. Wilson, J.A., and K.E. Mabry. 2005. Trap disinfection to reduce Hantavirus risk:
does it also reduce small mammal trapability? Presentation to the American Society
of Mammalogists, Annual Meeting, Springfield, MO.
8. Wilson, J. A., D. A. Kelt, and D. H. VanVuren. 2005. Effects of maternal body
condition on offspring dispersal in golden-mantled ground squirrels (Spermophilus
lateralis). Presentation to the American Society of Mammalogists, Annual Meeting,
Springfield, MO.
9. Wilson, J. A., D. A. Kelt, and D. H. VanVuren. 2005. Effects of maternal body
condition on offspring dispersal in golden-mantled ground squirrels (Spermophilus
lateralis). Presentation to the IX International Mammalogical Conference, Sapporo,
Japan.
10. Wilson, J. A., D. A. Kelt, and D. H. Van Vuren. 2006. Home range and activity of
the northern flying squirrel (Glaucomys sabrinus) in the northern Sierra Nevada.
Poster to the American Society of Mammalogists, Annual Meeting, Amherst, MA.
61
PERSONNEL
This project is currently coordinated and supervised by Brett Jesmer. Field work in 2008
was conducted by Brett Jesmer, Sarah Chinn, Sean Bogle, and Jessica Wright. This study
was carried out under the guidance of Dr. Douglas Kelt, Dr. Dirk Van Vuren, and Dr.
Michael Johnson, professors at the University of California Davis.
ACKNOWLEDGEMENTS
We thank the dedicated field crews of 2003–2008. We would like to extend our
appreciation to the contributions and collaborative efforts of PLAS module project
leaders and principal investigators, S. Bigelow, J. Keane, and M. North of the U.S.D.A
Forest Service, Pacific Southwest Research Station, Sierra Nevada Research Center, R.
Burnett of the Point Reyes Bird Observatory, and S. Stevens and K. Manning of the
University of California Berkeley. We would also like to thank R. Tompkins of the Mt.
Hough Ranger District. We thank J. Schaber of the University of California Berkeley
Forestry Camp for providing housing for our field crew. We would also like to thank the
Quincy Library Group and the communities of the Plumas National Forest, particularly
that of Quincy, California. This work was supported by the Joint Fire Sciences Program
and the United States Department of Agriculture, Forest Service (Region 5).
62
REFERENCES
AEBISCHER, N.J., P.A. ROBERTSON, AND R.E.KENWARD. 1993. Compositional analysis of
habitat use from animal radio-tracking data. Ecology74:1313-1325.
ALDOUS, S.E. 1941. Food habits of chipmunks. Journal of Mammalogy 22:18-24.
ANDERSON D.J. 1982. The home range: a new nonparametric estimation technique.
Ecology 63: 103-112.
ANIMAL CARE AND USE COMMITTEE. 1998. Guidelines for the capture, handling, and
care of mammals as approved by the American Society of Mammalogists. Journal of
Mammalogy 79:1416-1431.
ARMITAGE, K. B. 1989. Sociality as a life-history tactic in ground squirrels. Oecologia
48:36-49.
ATSATT, P. R., AND T. INGRAM. 1983. Adaptation to oak and other fibrous, phenolic-rich
foliage by a small mammal, Neotoma fuscipes. Oecologia 60:135-142.
BAKKER V.J., AND K. HASTINGS. 2002. Den trees used by northern flying squirrels
(Glaucomys sabrinus) in southeastern Alaska. Canadian Journal of Zoology 80: 16231633.
BARASH, D. P. 1974. The evolution of marmot societies: a general theory. Science
185:415-420.
BECHE, L. A. and V. H. RESH. 2007. Short-term climatic trends affect the temporal
variabiligy of macroinvertebrates in California ‘mediterranean’ streams. Freshwater
Biology 52: 2317-2339.
BELL, J. F., AND J. R. DILWORTH. 1993. Log scaling and timber cruising. OSU Book
Stores, Inc., Corvallis, Oregon.
BENNETT, R. P. 1999. Effects of food quality on growth and survival of juvenile
Columbian ground squirrels (Spermophilus columbianus). Canadian Journal of
Zoology 77:1555-1561.
BLOCK, W. M., J. L. GANEY, P. E. SCOTT., AND R. KING. 2005. Prey ecology of mexican
spotted owls in pine-oak forests of northern Arizona. Journal of Wildlife Management
69(2): 618-629.
BRAVO-VINAJA, M. G., L. A. TARANGO-ARAMBULA, F. CLEMENTE-SANCHEZ, G. D.
MENDOZA-MARTINEZ, J. L. ALCANTARA-CARBAJAL, AND H. V. SOTO-AQUINO. 2005.
Mexican Spotted Owl’s (Strix occidentalis lucida) diet composicion and variation at
Vlaparaiso, Zacatecas, Mexico. Agrociencia 39: 509-515.
BRIGGS, J.S., AND S.B. VANDER WALL. 2004. Substrate type affects caching and
pilferage of pine seeds by chipmunks. Behavioral Ecology 15:666-672.
BROADBOOKS, H.E. 1970. Populations of the yellow-pine chipmunk, Eutamias amoenus.
American Midland Naturalist 83:472-488.
BROOK, B. W. and C. J. A. BRADSHAW. 2006. Strength of evidence for density
dependence in abundance time series of 1198 species. Ecology 87(6): 1445-1451.
63
BURNHAM, K. P., AND D. R. ANDERSON. 1998. Model selection and inference: a practical
information-theoretic approach. Springer, New York, NY.
CAMERON, G. N. 1971. Niche overlap and competition in woodrats. Journal of
Mammalogy 52:288-296.
CAREY, A. B.. 2000. Effects of new forest management strategies on squirrel populations.
Ecological Applications 10(1): 248-257.
CAREY, A. B., B. L. BISWELL, AND J. W. WITT. 1991. Methods for measuring
populations of arboreal rodents. United States Department of Agriculture, Forest
Service, General Technical Report PNW-GTR-273.
CAREY, A. B., S. P. HORTON, AND B. L. BISWELL. 1992. Northern spotted owls:
influence of prey base and landscape character. Ecological Monographs 62:223-250.
CAREY, A. B., J. KERSHNER, B. BISWELL, AND L. DOMINGUEZ DETOLEDO. 1999.
Ecological scale and forest development: squirrels, dietary fungi, and vascular plants in
managed and unmanaged forests. Wildlife Monographs 142:1-71.
CAREY, A. B., J. A. REID, AND S. P. HORTON. 1990. Spotted owl home range and habitat
use in southern Oregon Coast Ranges. Journal of Wildlife Management 54:11-17.
CAREY, A. B. and S. M. WILSON. 2001. Induced spatial heterogeneity in forest canopies:
responses of small mammals. Journal of Wildlife Management 65(4): 1014-1027.
CARRAWAY, L. N., AND B. J. VERTS. 1991. Neotoma fuscipes. Mammalian Species
386:1-10.
CLAWSON, R.G., J.A. CLAWSON, AND T.L. BEST. 1994. Tamias quadrimaculatus.
Mammalian species 469:1-6.
CONVERSE, S. J., G. C. WHITE, K. L. FARRIS, and S. ZACK. 2006. Small mammals and
forest fuel reduction: national-scale responses to fire and fire surrogates. Ecological
Applications 16(5): 1717-1729.
COOPER, L.D., AND J.A. RANDALL. 2007. Seasonal changes in home ranges of the giant
kangaroo rat (Dipodomys ingens): A study of flexible social structure. Journal of
Mammalogy 88:1000-1008.
COPPETO, S. A., D. A. KELT, D. H. VAN VUREN, J. A. WILSON, and S. BIGELOW. 2006.
Habitat associations of small mammals at two spatial scales in the northern Sierra
Nevada. Journal of Mammalogy 87(2):402-414.
COVINGTON, W. W., P.Z. FULE, M. M. MOORE, S. C. HART, T. E. KOLB, J.N. MAST, S. S.
SACKETT, M. R. WAGNER. 1997. Restoring Ecosystem Health in Ponderosa Pine
Forests of the Southwest. Journal of Forestry 95: 23-29.
CRANFORD, J. A. 1977. Home range and habitat utilization by Neotoma fuscipes as
determined by radiotelemetry. Journal of Mammalogy 58:165-172.
CRANFORD, J. A. 1982. The effect of woodrat houses on population density of
Peromyscus. Journal of Mammalogy 63:663-666.
DOBSON, F. S., T. S. RISCH, & J. O. MURIE. 1999. Increasing returns in the life history of
Columbian ground squirrels. Journal of Animal Ecology 68:73-86.
64
DODGE, M. 1972. Forest fuel accumulation- a growing problem. Science 177:139-142.
DUNNING, J.B., B.J. DANIELSON, AND H.R. PULLIAM. 1992. Ecological processes that
affect populations in complex landscapes. Oikos 65: 169-175.
ELIAS, S. P., J. W. WITHAM, and M. L. HUNTER, Jr.. 2006. A cyclic red-backed vole
(Clethrionomys gapperi) population and seedfall over 22 years in Maine. Journal of
Mammalogy 87(3): 440-445.
FARGO, R., AND W. F. LAUDENSLAYER. 1999. Are house counts reliable estimators of
dusky-footed woodrat population size? Transactions of the Western Section of the
Wildlife Society 35:71-75.
FOGEL, R. 1976. Ecological studies of hypogeous fungi. II. Sporocarp phenology in
Oregon Douglas-fir stand. Canadian Journal of Botany 54:1152-1162.
FORSMAN, E. D., R. G. ANTHONY, E. C. MESLOW, AND C. J. ZABEL. 2004. Diets and
foraging behavior of northern spotted owls in Oregon. Journal of Raptor Research
38(3): 214-230.
FORSMAN, E. D., E. C. MESLOW, AND H. M. WIGHT. 1984. Distribution and biology of
the spotted owl in Oregon. Wildlife Monographs 87:1-64.
FRANKLIN, J. F., AND J. FITES-KAUFMAN. 1996. Assessment of late-successional forests
of the Sierra Nevada. Pp. 627-662 in Sierra Nevada Ecosystem Project. Final report to
Congress. Volume II. Assessments and scientific basis for management options.
University of California, Centers for Water and Wildland Resources, Davis, California.
FRAZER, G. W., C. D. CANHAM, AND K. P. LERTZMAN. 2000. Gap light analyzer (GLA),
version 2. Bulletin of the Ecological Society of America 81: 191-197.
FULLER, M.R., J.J. MILLSPAUGH, K.E. CHURCH, AND R.E. KENWARD. 2005. Wildlife
radiotelemetry. Pages 377-414 in C.E. Braun, editor. Techniques for wildlife
investigations and management. Sixth edition. The Wildlife Society, Bethesda,
Maryland, USA.
GANNON, W.L., AND R.B. FORBES. 1995. Tamias senex. Mammalian species 502:1-6
GANNON, W.L., R.S. SIKES, AND ANIMAL CARE AND USE COMMITTEE OF
THE AMERICAN SOCIETY OF MAMMALOGISTS. 2007. Guidelines of the
American Society of Mammalogists for the use of wild mammals in research. Journal
of Mammalogy 88:809–823.
GARRISON, B.A., R.L. WACHS, J.S. JONES, AND M.L. TRIGGS. 1998. Visual counts of
acorns of California black oak (Quercus kelloggii) as an indicator of mast production.
Western Journal of America Forestry 13(1): 27-31.
GUNTHER, P. M., B. S. HORN, AND G. D. BABB. 1983. Small mammal populations and
food selection in relation to timber harvest practices in the western Cascade Mountains,
Washington, USA. Northwest Science 57:32-44.
GREENWOOD, P.J. 1980 Mating system, philopatry, and dispersal in birds and mammals.
Animal Behavior 28:1140-1162.
65
HASKELL, S.P., AND W.B. BALLARD. 2007. Accounting for radiotelemetry signal flux in
triangulation point estimation. European Journal of Wildlife Research 53:204-211.
HOLLANDER, J.L., AND S.B. VANDER WALL. 2004. Effectiveness of six species of
rodents as dispersers of singleleaf pinon pine (Pinus monophylla). Oecologia 138:5765.
HOSMER, D. W., AND S. LEMESHOW. 1989. Applied logistic regression. 2nd ed. John
Wiley & Sons, Inc., New York, NY.
JACKSON, J. K. and L. FUREDER. 2006. Long-term studies of freshwater
macroinvertebrates: a review of the frequency, duration and ecological significance.
Freshwater Biology 51: 591-603.
JAMESON, Jr, E. W. 1951. Local distribution of white-footed mice, Peromyscus
maniculatus and P. boylei, in the northern Sierra Nevada, California. Journal of
Mammalogy 32(2): 197-203.
JAMESON, Jr., E. W. and H. J. PEETERS. 2004. Mammals of California. University of
California Press. Berkeley, CA.
KEATING, K. A., AND S. CHERRY. 2004. Use and interpretation of logistic regression in
habitat-selection studies. Journal of Wildlife Management 68:774-789.
KENWARD, R.E. 2001. A manual for wildlife radio tagging.
Press, San Francisco, U.S.A.
Academic
KENWARD, R.E., A.B. SOUTH, and S.S. WALLS. 2003. Ranges6 v1.215: For the analyses
of tracking and location data. Online manual. Anatrack Ltd. Warehame, UK.
KING, J. A. 1968. Biology of Peromyscus (Rodentia). “Habitats and Distribution” (98122). American Society of Mammalogists, U.S.A..
KLENNER, W. and T. P. SULLIVAN. 2003. Partial and clear-cut harvesting of highelevation spruce-fir forests: implications for small mammal communities. Canadian
Journal of Forest Research 33: 2283-2296.
KOTEJA, P. 1996. The usefulness of a new TOBEC instrument (ACAN) for investigating
body composition in small mammals. Acta Theriol. 41:107-112.
KREBS, C. J. 1966. Demographic changes in fluctuating populations of Microtus
californicus. Ecological Monographs 36(3): 239-273.
KUTNER, M.H., C.J. NACHTSHEIM, J.NETER, AND W. LI. 2005. Applied linear statistical
models. 5th ed. McGraw-Hill/Irwin, New York, NY. 1396 pp.
LAMBIN, X., AND C.J. KREBS. 1991. Spatial organization and mating system of Microtus
townsendii. Behavioral Ecology and Sociobiology 28:353-363.
LANCIA, R.A., J.D. NICHOLS, AND K.H. POLLOCK. 1996. Estimating the number of
animals in wildlife populations. Pp 215-254, In T.A. Bookhout, (Ed.). Research and
Management Techniques for Wildlife and Habitats. The Wildlife Society, Bethesda,
MD. 740 pp.
66
LAUDENSLAYER, W. F., AND R. J. FARGO. 1997. Small nocturnal mammals in oak
woodlands: some considerations for assessing presence and abundance. U.S. Forest
Service General Technical Report PSW-GTR-160.
LENIHAN, C., & D. VAN VUREN. 1996. Growth and survival of juvenile yellow-bellied
marmots (Marmota flaviventris). Canadian Journal of Zoology 74:297-302.
LENTH, R.V. 1981. On finding the source of a signal. Technometrics 23(2):149-154.
LINSDALE J.M., L.P. TEVIS. 1951. The dusky-footed wood rat; a record of observations
made on the Hastings Natural History Reservation. Berkeley, California: University of
California Press. 664pp.
LITVAITIS, J.A., K. TITUS, AND E.M. ANDERSON. 1996. Measuring vertebrate use of
terrestrial habitats and foods. Pp 254-274, In T.A. Bookhout, (Ed.). Research and
Management Techniques for Wildlife and Habitats. The Wildlife Society, Bethesda,
MD. 740 pp.
Long, L. A., J. L. Rachlow, J. G. Kie. 2007. Effects of season and scale on response of
elk and mule deer to habitat manipulation. The Journal of Wildlife Management 72(5):
1133-1142.
LYNCH, M. F., A. L. FESNOCK, AND D. H. VAN VUREN. 1994. Home range and social
structure of the dusky-footed woodrat (Neotoma fuscipes). Northwestern Naturalist
75:73-75.
MANLEY, B.F.J., L.L. MCDONALD, AND D.L. THOMAS. 1993. Resource selection by
animals: statistical design and analysis for field studies. Chapman & Hall, Ney York,
USA.
MARTELL, A. M. AND A. L. MACAULAY. 1981. Food Habits of Deer Mice (Peromyscus
maniculaturs) in northern Ontario. Canadian Field-Naturalist 95(3): 319-324.
MASER, C., AND Z. MASER. 1988. Interactions among squirrels, mycorrhizal fungi,
and coniferous forests in Oregon USA. Great Basin Naturalist 48:358-369.
MCDONALD, P. M. 1969. Silvicultural characteristics of California black oak (Quercus
kelloggii Newb.). United States Department of Agriculture, Forest Service, General
Technical Report PSW-53.
MCEACHERN, M.B. 2005. Behavioral variation and genetic structure in dusky-footed
woodrats, Neotoma fuscipes. PhD Dissertation. University of California, Davis.
M’CLOSKEY, R. T., R. J. DESLIPPE, C. P. SZPAK, AND K. A. BAIA. 1990. Tree lizard
distribution and mating system: the influence of habitat and food resources. Canadian
Journal of Zoology 68:2083-2089.
MERRITT, J. F. 1974. Factors influencing the local distribution of Peromyscus
californicus in northern California. Journal of Mammalogy 55:102-114.
MESERVE, P. L. 1974. Ecological relationships of two sympatric woodrats in California
coastal sage scrub community. Journal of Mammalogy 55:442-447.
67
MEYER, M. D., AND M. P. NORTH. 2005. Truffle abundance in riparian and upland
mixed-conifer forest of California’s southern Sierra Nevada. Canadian Journal of
Botany 83:1015-1020.
MEYERS, M. D., D. A. KELT, and M. P. NORTH. 2007. Effects of burning and thinning on
lodgepole chipmunks (Neotamias speciosus) in the Sierra Nevada, California.
Northwestern Naturalist 88: 61-72.
MICHENER, G. R. 1983. Kin identification, matriarchies, and the evolution of sociality in
ground-dwelling sciurids. Vol. Special publication No. 7. The American Society of
Mammalogists.
MILLSPAUGH, J.J., R.M. NIELSON, L. MCDONALD, J.M. MARZLUFF, R.A. GITZEN, C.D.
RITTENHOUSE, M.W. HUBBARD, AND S.L. SHERIFF. 2006. Analysis of resource selection
using utilization distributions. Journal of Wildlife Management 70:386-395.
MINTA, S.C. 1992. Tests of spatial and temporal interaction among animals. Ecological
applications 2(2):178-188.
MOHR, C.O. 1947. Table of equivalent populations of North American small mammals.
American Midland Naturalist 37: 223-249.
MUNTON, T. E., K. D. JOHNSON, G. N. STEGER, AND G. P. EBERLEIN. 2002. Diets of
california spotted owls in the Sierra National Forest. U.S. Forest Service General
Technical Report PSW-GTR-183.
MURIE, J. O., & D. A. BOAG. 1984. The relationship of body weight to overwinter
survival in Columbian ground squirrels. Journal of Mammalogy 65:688-690.
NAMS, V.O. 2006. Locate III user’s guide. Pacer Computer Software, Tatamagouche,
Nova Scotia, Canada.
NETER, J., M. H. KUTNER, C. J. NACHTSHEIM, AND W. WASSWEMAN. 1996. Applied
linear statistical models. Fourth edition. WCB McGraw-Hill Co., Inc., New York, NY.
NUNES, S., P. A. ZUGGER, A. L. ENGH, K. O. REINHART, AND K. E. HOLEKAMP. 1998.
Body fat and time of year interact to mediate dispersal behaviour in ground squirrels.
Animal Behavior 55:605-614.
NEUHAUS, P., D. R. BROUSSARD, J. O. MURIE, & F. S. DOBSON. 2004. Age of primiparity
and implications of early reproduction on life history in female Columbian ground
squirrels. J. Animal Ecology 73:36-43.
OSTFELD, R.S. 1986. Territoriality and mating system of California voles. Journal of
Animal Ecology 55:691-706.
OSTFELD, R.S., 1990. The ecology and territoriality in small mammals. Trends in
Ecology and Evolution 5:411-415.
PETTORELLI, N. and S. M. DURANT. 2007. Longevity in cheetahs: the key to success?.
Oikos 116: 1879-1886.
POST, D.M., O.J. REICHMAN, AND D.E. WOOSTER. 1993. Characteristics and significance
of the caches of eastern woodrats (Neotoma floridana). Journal of Mammalogy
74(3):688-692.
68
PULAWA, L. K., & G. L. FLORANT. 2000. The effects of caloric restriction on the body
composition and hibernation of the golden-mantled ground squirrel (Spermophilus
lateralis). Physiol. Bio. Zool. 73:538-546.
PUSEY, A.E. 1987. Sex-biased dispersal and inbreeding avoidance in birds and mammals.
Trends in Ecology Evolution 2:295-299.
PYARE, S., AND W. S. LONGLAND. 2001. Patterns of ectomycorrhizal-fungi consumption
by small mammals in remnant old-growth forests of the Sierra Nevada. Journal of
Mammalogy 82:681-689.
ROSENBERG, D. K., K. A. SWINDLE, AND R. G. ANTHONY. 2003. Influence of prey
abundance on northern spotted owl reproductive success in western Oregon. Canadian
Journal of Zoology 81:1715-1725.
SAKAI, H. F., AND B. R. NOON. 1993. Dusky-footed woodrat abundance in different-aged
forests in northwestern California. Journal of Wildlife Management 57:373-382.
SAKAI, H. F., AND B. R. NOON. 1997. Between-habitat movement of dusky-footed
woodrats and vulnerability to predation. Journal of Wildlife Management 61:343-350.
SAS INSTITUTE. 2000. Statistical analysis system (SAS) user’s guide: statistics, version
8 SAS Institute Inc. Clay, North Carolina 1686pp.
SAS INSTITUTE. 2001. SAS version 8.02. SAS Institute, Cary, North Carolina.
SAS INSTITUTE. 2004. JMP version 5.1.2. SAS Institute, Cary, North Carolina.
SCHMUTZ, J.A., AND G.C. WHITE. 1990. Error in telemetry studies: effects of animal
movements on triangulation. Journal of Wildlife Management 54:506-510.
SEAMAN, D.E., J.J. MILLSPAUGH, B.J. KERNOHAN, G.C. BRUNDIGE, K.J. RAEDEKE, AND
R.A. GITZEN. 1999. Effects of sample size on kernel home range estimates. Journal of
Wildlife Management 63:739-747.
SEAMAN, D.E., AND R.A.POWELL. 1996. An evaluation of the accuracy of kernel density
estimators fro home range analysis. Ecology 77(7):2075-2085.
SILVERMAN, B.W. 1986. Density estimation for statistics and data analysis. Chapman
and Hall, London, England.
SLOBODCHIKOFF, C.N., AND W.C. SCHULZ. 1988. Cooperation, aggression, and the
evolution of social behavior (C.N. Slobodchikoff, ed.) Academic Press, Inc., San
Diego, California.
SMITH, C.F., AND S.E. ALDOUS. 1947. The influence of mammals and birds in retarding
artificial and natural reseeding of coniferous forests in the United States. Journal of
Forestry 45:361-369.
SMITH, R. B., M. Z. PEERY, R. J. GUTIERREZ, and W. S. LAHAYE. 1999. The relationship
between spotted owl diet and reproductive success in the San Bernardino Mountains,
California. The Wilson Bulletin 11(1): 22-29.
SMITH, W. P., S. M. GENDE, AND J. V. NICHOLS. 2004. Ecological correlates of flying
squirrel microhabitat use and density in temperate rainforests of southeastern Alaska.
Journal of Mammalogy 85:663-674.
69
SMITH, W. P., S. M. GENDE, AND J. V. NICHOLS. 2005. The northern flying squirrel as an
indicator species of temperate rain forests: test of a hypothesis. Ecological
Applications 689-700.
STEELE, M.A. 1998. Tamiasciurus hudsonicus. Mammalian Species 586:1-9.
STOKES, M. E., C. S. DAVIS, AND G. G. KOCH. 2001. Categorical data analysis using the
SAS system. SAS Institute, Cary, North Carolina.
SUGG, D.W., R.K., CHESSER, F.S. DOBSON, AND J.L. HOOGLAN. 1996. Population
genetics meets behavioral ecology. Trends in Ecology and Evolution 11:338-342.
SUMNER, L., AND J.S. DIXON. 1953. Birds and mammals of the Sierra Nevada with
records from Sequoia and Kings Canyon National Parks. University of California Press,
Berkeley, CA, 484 pp.
Suzuki, N. and J. P. Hayes. 2003. Effects of thinning on small mammals in Oregon
coastal forests. Journal of Wildlife Management 67(2): 352-371.
SWIHART, R. K., AND N. A. SLADE. 1988. Relating body size to the rate of home range
use in mammals. Ecology 69:393-399.
TAULMAN, J. F., AND K. G. SMITH. 2004. Home range and habitat selection of southern
flying squirrels in fragmented forests. Mammalian Biology 69:11-27.
TEVIS, L., JR. 1953. Stomach contents of chipmunks and mantled squirrels in
northeastern California. Journal of Mammalogy 34:316-324.
THRAILKILL, J. and M. A. BIAS. 1989. Diets of bredding and nonbreeding california
spotted owls. Journal of Raptor Research 23(2): 39-41.
Title IV- The Herger-Feinstein Quincy Library Group Forest Recovery Act- Sec. 401.
<http://qlg.org/pub/act/act.htm>
TOPPING, M.G. AND J.S. MILLAR. 1996. Spatial distribution in the bushy-tailed woodrat
(Neotoma cinerea) and its implications for the mating system. Canadian Journal of
Zoology 74:565-569.
TSUJI, Y., S. FUJITA, H. SUGIURA, C. SAITO, and S. TAKATSUKI. 2006. Long-term
variation in fruiting and the food habits of wild japanese macaques on Kinkazan Island,
northern Japan. American Journal of Primatology 68: 1068-1080.
UNITED STATES DEPARTMENT OF THE INTERIOR. 2003. Endangered and threatened wildlife
and plants: determination of threatened status for the California spotted owl. Federal
Register 68:7580-7608.
VANDER WALL, S.B. 1992. The role of animals in dispersing a “wind-dispersed” pine.
Ecology 73:614-621.
VANDER WALL, S.B. 2000. The influence of environmental conditions on cache recovery
and cache pilferage by yellow-pine chipmunks (Tamias amoenus) and deer mice
(Peromyscus maniculatus). Behavioral ecology 11:544-549.
VAUGHAN, T.A. 1974. Resource allocation in some sympatric, alpine rodents. Journal of
Mammalogy 55:764-795.
70
VERNER, J., K.S. MCKELVEY, B.R. Noon, R.J. Gutierrez, G.I. Gould, Jr., T.W. Beck,
(technical coordinators). 1992. California spotted owl: A technical assessment of its
current status. U.S. Forest Service General Technical Report PSW-GTR-133. 286p.
VERTS, B.J., AND L.N. CARRAWAY. 2002. Neotoma lepida. Mammalian Species 669:1-12
VESTAL, E. H. 1938. Biotic relations of the woodrat (Neotoma fuscipes) in the Berkeley
Hills. Journal of Mammalogy 19:1-36.
WALLEN, K. 1982. Social organization in the dusky-footed woodrat (Neotoma fuscipes): a
field and laboratory study. Animal Behavior 30: 1171-1182.
WALSBERG, G. E. 1988. Evaluation of a nondestructive method for determining fat
stores in small birds and mammals. Physiological Zoology 61:153-159.
WEIGL, P. D., AND D. W. OSGOOD. 1974. Study of the northern flying squirrel,
Glaucomys sabrinus, by temperature telemetry. American Midland Naturalist 92:482486.
WELLS-GOSLING, N., AND L. R. HEANEY. 1984. Glaucomys sabrinus. Mammalian
Species No. 229:1-8.
WHISSON, D.A., J.H. QUINN, AND K.C. COLLINS. 2007. Home range and movements of
roof rats (Rattus rattus) in old-growth riparian forest, California. Journal of
Mammalogy 88(3):589-594.
WILLY, A. G. 1992. The habitat associations of the dusky-footed woodrat (Neotoma
fuscipes) in Marin County, California. Thesis, University of California, Berkeley,
California, USA.
WILSON, J. A., D. A. KELT, D. H. VAN VUREN, and M. L. JOHNSON. 2008. Population
dynamics of small mammals in relation to production of cones in four types of forests
in the northern Sierra Nevada, California. The Southwestern Naturalist 53(3): 346-356.
WORTON, B.J. 1989. Kernel methods for estimating the utilization distribution in homerange studies. Ecology 70:164-168.
ZABEL, C. J., K. S. MCKELVEY, AND J. P. WARD. 1995. Influence of primary prey on
home range size and habitat use patterns of spotted owls (Strix occidentalis). Canadian
Journal of Zoology 73:433-439.
ZIELINSKI, W. J., W. D. SPENCER, AND R. H. BARRETT. 1983. Relationship between food
habits and activity patterns of pine martens. Journal of Mammalogy 64:387-396.
71
FIGURE LEGENDS
Fig. 1. — Map of long-term grids in Plumas National Forest with a) locations of 18
long-term grids in 5 forest types and b) trap configuration within a long-term grid. Inset
shows the location of the Forest in California. Map extracted from Coppeto et al. (2006).
Fig. 2.— Map of 4 dusky-footed woodrat study areas in Plumas National Forest (PNF),
California. Numbers indicate study site location. Inset shows the location of PNF in
California. Map extracted from Innes et al. (2007).
Fig. 3-- Map of 4 northern flying squirrel study areas in Plumas National Forest (PNF),
California. Squares and circles indicate the location of study sites and major towns,
respectively.
Fig. 4.—Mean fall cone production by the major conifers at long-term grids (n=18).
Means were derived by counting cone production on 10 individual trees/species on each
grid and averaging across forest types. Cones were counted visually during the fall of
2003 and 2004. Statistically significant differences are represented by different letters
within each species and in each year.
Fig. 5.— Mean monthly density (A) and survival (B) of deer mouse populations
inhabiting four forest types in Plumas National Forest, California: white fir, Douglas-fir,
red fir, and ponderosa pine. Population estimates were obtained using long-term grid
data and program MARK. Populations were monitored from June 2003 to October 2004.
Fig. 6. — Mean monthly density (A) and survival (B) of golden-mantled ground squirrel
populations inhabiting red fir forests in Plumas National Forest, California. Population
estimates were obtained using long-term grid data and program MARK. Populations
were monitored from June 2003 to October 2004.
Fig. 7.—Mean monthly density of (A) long-eared chipmunk and (B) Allen’s chipmunk,
inhabiting three forest types (white fir, red fir, Douglas-fir) in Plumas National Forest,
California. Density estimates were obtained using long-term grid data and program
MARK. Populations were monitored from June 2003 to October 2004.
Fig. 8.—Overall mean abundance of mice, chipmunks, northern flying squirrels, and
dusky-footed woodrats regardless of habitat or treatment type in Plumas National Forest,
California, during 2003-2008.
Fig. 9.— Mean abundance of mice (Peromyscus sp.) across 5 forest types within Plumas
National Forest, California, during 2003-2008.
Fig. 10.—Mean abundance of mice (Peromyscus sp.) pre and post treatment within the
Plumas National Forest, California, during 2005-2008.
72
Fig. 11.— Mean abundance of chipmunks (Tamias sp.) across 5 forest types within the
Plumas National Forest, California, during 2003-2008.
Fig. 12.— Mean abundance of chipmunks (Tamias sp.) pre and post treatment within the
Plumas National Forest, California, during 2005-2008.
Fig. 13.— Mean abundance of northern flying squirrels across 5 forest types within the
Plumas National Forest, California, during 2005-2008.
Fig.14. — Mean abundance of northern flying squirrels (Glaucomys sabrinus) pre and
post treatment within the Plumas National Forest, California, during 2005-2008.
Fig. 15.— Mean abundance of dusky-footed woodrats across 5 forest types within the
Plumas National Forest, California, during 2003-2008.
Fig. 16.-- Mean abundance of dusky-footed woodrats (Neotoma fuciceps) pre and post
treatment within the Plumas National Forest, California, during 2005-2008.
Fig. 17.—General relationships between mean mouse (Peromyscus sp.) mean abundance,
mean annual cone production, and mean annual snow fall from 2003-2008. Note: lag in
response to mean conifer cone abundance and immediate response to annual snow fall.
Fig. 18-- Mean annual biomass of small mammal species from 2003-2008; summarized
across 5 forest types.
Fig. 19-- Mean annual biomass of key spotted owl prey species (mice, northern flying
squirrel, and dusky-footed woodrat) from 2003-2008; summarized across 5 forest types.
Table 5. -- Mean annual biomass (g) of all species in experimental plots pre and post
treatment.
Fig. 20— Mean annual post treatment biomass in the 3 experimental group select plots.
Fig. 21— 2008 MNKA summarized by experimental plot “clusters”, Dean’s Valley
(DV), Snake Lake (SL), and Miller Fork (MF).
Fig. 22. —95% minimum convex polygon (MCP) graphic output for home range analysis
for a single animal. Calculating the area within the polygon can be useful to determine
what type or area an animal needs to exist. Quality of foraging land can be elucidated as
well if MCP area varies between sites.
Fig. 23. —95% fixed kernel (FK) estimator from the same animal from Figure 22. Like
contours on a topographic map, each line represents the degree of usage of the home
range by the animal. As you move from the exterior to the interior, the animal utilizes
the area more frequently.
73
FIGURES
Fig. 1. — Map of long-term grids in Plumas National Forest with a) locations of 18
long-term grids in 5 forest types and b) trap configuration within a long-term grid. Inset
shows the location of the Forest in California. Map extracted from Coppeto et al. (2006).
74
Fig. 2.— Map of 4 dusky-footed woodrat study areas in Plumas National Forest (PNF),
California. Numbers indicate study site location. Inset shows the location of PNF in
California. Map extracted from Innes et al. (2007).
75
Fig. 3-- Map of 4 northern flying squirrel study areas in Plumas National Forest (PNF),
California. Squares and circles indicate the location of study sites and major towns,
respectively.
76
Fig. 4.—Mean fall cone production by the major conifers at long-term grids (n=18).
Means were derived by counting cone production on 10 individual trees/species on each
grid and averaging across forest types. Cones were counted visually during the fall of
2003 and 2004. Statistically significant differences are represented by different letters
within each species and in each year.
300
ABCO
PILA
PIPO
PSME
ABMA
PIMO
2003
a
Number of Cones
250
200
c
150
b
100
50
b
aa
a
0
Douglas Fir
b
a
b
a
b
a
b
Ponderosa
Red Fir
White Fir
Ponderosa
Red Fir
White Fir
250
2004
Number of Cones
200
150
100
50
0
Douglas Fir
Habitat
77
Fig. 5.— Mean monthly density (A) and survival (B) of deer mouse populations inhabiting four
forest types in Plumas National Forest, California: white fir, Douglas-fir, red fir, and ponderosa
pine. Population estimates were obtained using long-term grid data and program MARK.
Populations were monitored from June 2003 to October 2004.
160
White Fir
Red Fir
Douglas Fir
Ponderosa
140
A
100
Winter
Individuals/grid
120
80
60
40
20
0
Jul
Aug
Sep
Oct
May
Jun
Jul
Aug
Sep
Oct
B
1.0
Winter
Probability of Survival
0.8
0.6
0.4
0.2
0.0
Jul
Aug
Sep
Oct
May
Jun
Jul
Aug
Sep
Oct
78
Fig. 6. — Mean monthly density (A) and survival (B) of golden-mantled ground
squirrel populations inhabiting red fir forests in Plumas National Forest, California.
Population estimates were obtained using long-term grid data and program
MARK. Populations were monitored from June 2003 to October 2004.
40
A
Spermophilus lateralis
Hibernation
Individuals/grid
30
20
10
0
Jul
1.0
Aug
Sep
Oct
May
Jun
Jul
Aug
Sep
Oct
B
Hibernation
Probability of Survival
0.8
0.6
0.4
0.2
0.0
Jul
Aug
Sep
Oct
Jun
Jul
Aug
Sep
Oct
79
Fig. 7.—Mean monthly density of (A) long-eared chipmunk and (B) Allen’s chipmunk,
inhabiting three forest types (white fir, red fir, Douglas-fir) in Plumas National Forest,
California. Density estimates were obtained using long-term grid data and program
MARK. Populations were monitored from June 2003 to October 2004.
50
A
Hibernation
Individuals/grid
40
30
20
10
0
Jul
110
Aug
Sep
Oct
May
Jun
Jul
Aug
Sep
Oct
May
Jun
Jul
Aug
Sep
Oct
White Fir
Red Fir
Douglas Fir
B
100
90
Hibernation
Individuals/grid
80
70
60
50
40
30
20
10
0
Jul
Aug
Sep
Oct
80
Fig. 8.—Overall mean abundance of mice, chipmunks, northern flying squirrels, and
dusky-footed woodrats regardless of habitat or treatment type in Plumas National Forest,
California, during 2003-2008.
Overall Year Trends
25
Chipmunks
Mice
20
15
Mean Abundance
10
5
0
1.6
northern flying squirrel
dusky footed woodrat
1.4
1.2
1.0
0.8
0.6
0.4
0.2
0.0
2003
2004
2005
2006
2007
2008
Year
81
Fig. 9.— Mean abundance of mice (Peromyscus sp.) across 5 forest types within Plumas
National Forest, California, during 2003-2008.
Mice
40
Mixed Conifer
Mixed Fir
Pine-cedar
Red Fir
White Fir
35
Mean Abundance
30
25
20
15
10
5
0
2003
2004
2005
2006
2007
2008
Year
Fig. 10.—Mean abundance of mice (Peromyscus sp.) pre and post treatment within the
Plumas National Forest, California, during 2005-2008.
Mice
25
Experimental Controls
All Controls
Group Select
Heavy Thin
Light Thin
Mean Abundance
20
Treatment
15
10
5
0
2005
2006
2007
2008
Year
82
Fig. 11.— Mean abundance of chipmunks (Tamias sp.) across 5 forest types within the
Plumas National Forest, California, during 2003-2008.
Chipmunks
40
Mixed Conifer
Mixed Fir
Pine-cedar
Red-Fir
White-Fir
35
Mean Abundance
30
25
20
15
10
5
0
2003
2004
2005
2006
2007
2008
Year
Fig. 12.— Mean abundance of chipmunks (Tamias sp.) pre and post treatment within the
Plumas National Forest, California, during 2005-2008.
Chipmunks
25
Experimental Control
All Controls
Group Select
Heavy Thin
Light Thin
Mean Abundance
20
Treatment
15
10
5
0
2005
2006
2007
2008
Year
83
Fig. 13.— Mean abundance of northern flying squirrels across 5 forest types within the
Plumas National Forest, California, during 2005-2008.
Northern Flying Squirrels
2.5
Mixed Conifer
Mixed Fir
Pine-cedar
Red Fir
White Fir
Mean Abundance
2.0
1.5
1.0
0.5
0.0
2005
2006
2007
2008
Year
Fig.14. — Mean abundance of northern flying squirrels (Glaucomys sabrinus) pre and
post treatment within the Plumas National Forest, California, during 2005-2008.
Northern Flying Squirrels
3.5
Experimental Control
All Control
Group Select
Heavy Thin
Light Thin
3.0
Mean Abundance
2.5
Treatment
2.0
1.5
1.0
0.5
0.0
2005
2006
2007
2008
Year
84
Fig. 15.— Mean abundance of dusky-footed woodrats across 5 forest types within the
Plumas National Forest, California, during 2003-2008.
Dusky-footed Woodrats
2.0
Mixed Conifer
Mixed Fir
Pine-cedar
Red Fir
White Fir
Mean Abundance
1.5
1.0
0.5
0.0
2003
2004
2005
2006
2007
2008
Year
Fig. 16.-- Mean abundance of dusky-footed woodrats (Neotoma fuciceps) pre and post
treatment within the Plumas National Forest, California, during 2005-2008.
Dusky-Footed Woodrat
4.0
3.5
Experimental Control
All Control
Group Select
Heavy Thin
Light Thin
Treatment
Mean Abundance
3.0
2.5
2.0
1.5
1.0
0.5
0.0
2005
2006
2007
2008
Year
85
Fig. 17.—General relationships between mean mouse (Peromyscus sp.) mean abundance,
mean annual cone production, and mean annual snow fall from 2003-2008. Note: lag in
response to mean conifer cone abundance and immediate response to annual snow
fall.
Mice
80
mean conifer cone abundance
annual snow fall
mean mouse abundance
60
40
20
0
2003
2004
2005
2006
2007
2008
Year
86
Fig. 18-- Mean annual biomass of small mammal species from 2003-2008; summarized
across 5 forest types.
Mean Total Biomass
8000
Biomass (g)
6000
CLCA
GLSA
Microtus sp.
NEFU
Peromyscus sp.
Spermophilus sp.
Tamias sp.
TADO
4000
2000
0
MC
MF
PC
RF
WF
Habitat Type
87
Fig. 19-- Mean annual biomass of key spotted owl prey species (mice, northern flying
squirrel, and dusky-footed woodrat) from 2003-2008; summarized across 5 forest types.
Spotted Owl Prey Across Years
1000
Peromyscus sp.
northern flying squirrel
dusky-footed woodrat
Mean Biomass (g)
800
600
400
200
0
MC
MF
PC
RF
WF
Habitat
88
Fig. 20— Mean annual post treatment biomass in the 3 experimental group select plots.
Post Treatment
6000
5000
Biomass (g)
4000
3000
2000
1000
0
RU
TH
TO
Group Select
89
Fig. 21— 2008 MNKA summarized by experimental plot “clusters”, Dean’s Valley
(DV), Snake Lake (SL), and Miller Fork (MF).
2008 MKNA
180
160
# Individuals
140
120
100
clca
glsa
peromyscus sp.
nefu
spbe
tado
tamias sp.
Microtus sp.
80
60
40
20
DV
SL
MF
Experimental Group
90
Fig. 22. —95% minimum convex polygon (MCP) graphic output for home range analysis
for a single animal. Calculating the area within the polygon can be useful to determine
what type or area an animal needs to exist. Quality of foraging land can be elucidated as
well if MCP area varies between sites.
Fig. 23. —95% fixed kernel (FK) estimator from the same animal from Figure 22. Like
contours on a topographic map, each line represents the degree of usage of the home
range by the animal. As you move from the exterior to the interior, the animal utilizes
the area more frequently.
91
Fig. 24.-- Graphical output for incremental area analysis for northern flying squirrels,
generated by adding subsequent points to a home range. The area (percentage) is
generated by comparing the home range with “x” locations to the largest home range
generated with all locations. In this figure, the home range stabilizes at 21 locations.
92
Fig. 25.— Total number and percentage (n, %) of tree species available (A) and used (B)
by northern flying squirrels at den sites (n = 53) and paired random sites (n = 53)
in the Plumas National Forest, California, 2006 – 2007.
A
115, 3%
123, 4%
Yellow Pine
311, 9%
1118, 33%
355, 10%
Acer macrophyllum
Calocedrus decurrens
Abies magnifica
355, 10%
Quercus kelogii
Abies concolor
Pseudotsuga menziesii
1059, 31%
B
3, 6%
8, 15%
4, 8%
5, 9%
4, 8%
15, 28%
Yellow Pine
Acer macrophyllum
Calocedrus decurrens
Abies magnifica
Quercus kelogii
Abies concolor
14, 26%
Pseudotsuga menziesii
93
Fig. 26. —Proportional use of tree houses by dusky-footed woodrats, by month, in
Plumas National Forest, 2004 to 2006.
100%
Male
90%
Female
Proportional Use of Tree Houses
80%
70%
60%
50%
40%
30%
20%
10%
0%
Jun
Jul
Aug
Sep
Oct
Month
94
Fig. 27.—Core areas (a) and home ranges (b) of dusky-footed woodrats at study site 1 in
Plumas National Forest, California, during May-October 2004. The minimum convex
polygons for core area (50% MCP) and home range (95% MCP) are shown for graphical
simplicity. Solid lines indicate adult females and dashed lined indicate adult males.
Fig. 28.— Frequency of house sharing (%) by dusky-footed woodrats, by month, in the
Plumas National Forest, 2004 to 2006.
100%
2004
90%
2005
2006
Frequency of House Sharing
80%
70%
60%
50%
40%
30%
20%
10%
0%
Jun
Jul
Aug
Sep
Oct
Month
95
TABLE LEGENDS
Table 1.—Description of microhabitat variables measured in 1-m radius (3.14m2) plots at
all long-term and landbird grid trap stations.
Table 2.— Results of the Program MARK analyses for 4 species of rodent in Plumas
National Forest, California. All species were analyzed individually using the CormackJolly Seber data type. Best-fit models are shown for each species. Akaike's corrected
information coefficient (AICc), adjusted for overdispersion, and the model weight
relative to other less fit models is given. Data for other species were too sparse for
analysis with Program MARK.
Table 3.-- Mean annual biomass (g) of small mammal species from 2003-2008;
summarized across 5 forest types.
Table 4.-- Mean total biomass (g) of key spotted owl prey species (mice, northern flying
squirrel, and dusky-footed woodrat) across from 2003-2008; summarized across 5 forest
types.
Table 5. -- Mean annual biomass (g) of all species in experimental plots pre and post
treatment.
Table 6..-- Home range areas (in hectares) generated by minimum convex polygon
(MCP) and kernel home range estimation methods for all northern flying squirrels from
2006 and 2007.
Table 7.—Mean size (cm; dbh) of trees by species available and used by northern flying
squirrels at den sites (n = 53) and paired random sites (n = 53) in Plumas National
Forest, California, 2006 – 2007. Presence of an asterisk indicates significant differences.
Yellow pine includes ponderosa pine (Pinus ponderosa) and Jeffrey pine (P. jeffreyi).
Table 8.— Availability and use of trees for tree house locations by dusky-footed
woodrats, by species (%) and by mean size (cm), in the northern Sierra Nevada, 2004 to
2006. Other trees include mountain dogwood (Cornus nuttallii), green and white-leaf
manzanita (Arctostaphylos sp.), and willow (Salix sp.). Availability was calculated as the
mean proportion of trees and snags.
Table 9.— Proportional (%) availability and use of ground and tree houses by duskyfooted woodrats in Plumas National Forest, 2004 to 2006. Numbers in parentheses
indicate standard error.
Table 10.— Mean home range (95%) and core area (50%) estimates and associated
standard errors (±SE) of dusky-footed woodrats using minimum convex polygon (MCP)
and fixed kernel (FK) methods at 2 study sites in the Plumas National Forest, California.
96
TABLES
Table 1.—Description of microhabitat variables measured in 1-m radius (3.14m2) plots at
all long-term and landbird grid trap stations.
Rocks
Bare ground
Litter
Branches
Small logs
Large logs
Forbs/grasses
Live shrubs
Dead shrubs
Vegetation
mats
Woody
perennials
Moss
Saplings
Live tree boles
Snags
Stumps
Shrub richness
Sapling richness
Tree richness
Canopy closure
(%)
Degree slope
Aspect
Exposed large rocks and stones
Exposed soil
Dead leaves, pine needles, wood chips, sawdust like debris
Twigs with diameter <10 cm
Logs and stumps with diameter > 10 - 50 cm
Logs and stumps with diameter > 50 cm
Herbaceous and flowering vegetation and grasses
Woody vegetation not considered sapling; height < 2 m
As for live shrub but with no living foliage or no foliage
Near ground surface shrub cover (Ceanothus prostratus)
Shrub- and forblike vegetation lacking woody stems
Any plant resembling "moss" in appearance and growth form
Small trees with height < 2 m
Live standing tree
Dead standing tree
Remaining portion of tree still attached to root after falling or
being cut
Number of distinct, live shrub species
Number of distinct, sapling species
Number of distinct, live tree species
Percentage open sky above breast height (1.4 m)
Degree incline or decline of the forest floor
Probable direction in which water will flow
97
Table 2.— Results of the Program MARK analyses for 4 species of rodent in Plumas National Forest,
California. All species were analyzed individually using the Cormack-Jolly Seber data type. Bestfit models are shown for each species. Akaike's corrected information coefficient (AICc),
adjusted for overdispersion, and the model weight relative to other less fit models is given. Data
for other species were too sparse for analysis with Program MARK.
Species
Model
Peromyscus maniculatus (habitat*t+overwinter+mean cones)p(habitat*t)
Spermophilus lateralis
(t)p(t)
Tamias quadrimaculatus (habitat*t+overwinter+mean cones)p(habitat*t)
Tamias senex
(habitat*t)p(habitat*t)
(habitat*t+overwinter)p(habitat*t)
AICc Weight C-hat
1740.6 0.99
1.85
358.2 0.96
1.14
923.5 1.00
1.22
683.2 0.60
1.23
684.1 0.39
Table 3.-- Mean annual biomass (g) of small mammal species from 2003-2008;
summarized across 5 forest types.
Species
Clethrionomys
Glaucomys
Microtus
Neotoma
Peromyscus
Spermophilus
Tamias
Tamiasciurus
Total Biomass
% of Biomass
MC
0
270.2
0
176.75
146.853
312.5
139.0125
226
1271.316
0.081126
MF
34.5
226.5393
53.2
373.3
217.1592
676.8437
1177.346
234.4167
2993.305
0.191011
PC
0
229.125
0
335.2333
124.6859
672.55
0
185
1546.594
0.098692
RF
0
270.6
37.72222
0
283.9552
3358.914
2883.879
0
6835.071
0.436165
WF
74.8
215.694
29.5
197
205.264
664.6714
1284.306
353.3333
3024.568
0.193006
Table 4.-- Mean total biomass (g) of key spotted owl prey species (mice, northern flying
squirrel, and dusky-footed woodrat) across from 2003-2008; summarized across 5 forest
types.
Species
MC
MF
PC
RF
WF
Peromyscus
146.853
217.1592
124.6859
283.9552
205.264
Glaucomys
270.2
226.5393
229.125
270.6
215.694
Neotoma
176.75
373.3
335.2333
0
197
Total Biomass
593.803
816.9985
689.0443
554.5552
617.958
% Biomass
0.18146
0.249667
0.210565
0.169466
0.188842
98
Table 5. -- Mean annual biomass (g) of all species in experimental plots pre and post treatment.
Pre
Species
Clethrionomys
Glaucomys
Microtus
Neotoma
Peromyscus
Spermophilus
Tamias
Tamiasciurus
Total Biomass
Post
Control
Pre
Post
Light Thin
Pre
Post
Heavy Thin
168.3
0
37.6
0
0
34
263.25
382.3222
109.25
232.3333 116.8333 248.7917
17
0
0
0
0
42
0
0
149.5
0
214.125
0
166.5413 207.219
234.583 124.5587 150.1573 213.2201
364
468.25
523
0
430
172
499.865 1031.132 572.6054 1870.285 389.5691 705.8455
437
347.5
364.6667
241
0
216.25
1915.956 2436.423 1991.205 2468.177 1300.685 1632.107
Pre
Post
Group Select
18.5
118.5
0
0
140.74
982.1
368.1875
0
178.575
0
0
180.9655
659
2117.016
0
240
1628.027
3375.556
99
Table 6..-- Home range areas (in hectares) generated by minimum convex polygon
(MCP) and kernel home range estimation methods for all northern flying squirrels from
2006 and 2007.
Squirrel
ID
8
6
27
1
2
7
12
6
7
15
16
17
18
19
20
27
21
22
23
24
25
Number
of
Locations
50
50
35
51
33
60
68
67
54
63
55
59
61
53
55
66
55
57
61
51
65
MCP
(ha)
3.02
10.55
17.67
12.35
9.24
1.90
3.76
9.74
2.04
6.05
12.60
13.15
6.94
10.56
10.45
3.72
15.58
8.42
2.65
10.66
14.68
Fixed
Kernel
(ha)
3.87
10.16
29.68
21.58
14.34
3.77
3.35
12.36
2.33
3.88
11.34
17.51
7.61
13.78
13.21
6.80
15.15
9.18
3.72
9.83
17.46
Year
2006
2006
2006
2006
2006
2006
2007
2007
2007
2007
2007
2007
2007
2007
2007
2007
2007
2007
2007
2007
2007
Table 7.—Mean size (cm; dbh) of trees by species available and used by northern flying
squirrels at den sites (n = 53) and paired random sites (n = 53) in Plumas National
Forest, California, 2006 – 2007. Presence of an asterisk indicates significant differences.
Yellow pine includes ponderosa pine (Pinus ponderosa) and Jeffrey pine (P. jeffreyi).
Tree Type
Abies concolor
Abies magnifica
Calocedrus
decurrens
Yellow Pine
Pseudotsuga
menziesii
Quercus kelogii
Acer macrophyllum
Availablemean DBH
(cm)
26.76
32.42
# of
observations
1059
355
Used- mean
Den size DBH
(cm)
61.07*
58.25
# of
observations
15
4
26.38
36.17
311
115
73.60*
121.33*
5
3
30.45
17.59
18.96
1118
355
123
89.25*
29.61*
19.00
8
14
4
100
Table 8.— Availability and use of trees for tree house locations by duskyfooted woodrats, by species (%) and by mean size (cm), in the northern
Sierra Nevada, 2004 to 2006. Other trees include mountain dogwood
(Cornus nuttallii), green and white-leaf manzanita (Arctostaphylos sp.),
and willow (Salix sp.). Availability was calculated as the mean proportion
of trees and snags.
Tree houses
Species
Availability (%)
Cavity (%)
Limb (%)
White fir
30
3
56
Incense cedar
20
0
10
Ponderosa pine
7
0
0
Sugar pine
4
1
0
Douglas-fir
16
3
15
California black oak
13
72
15
Snag
7
21
0
Other trees
4
0
3
Size group
Availability (cm)
Cavity (cm)
Limb (cm)
Tree size
16.7
49.6
18.4
Snag size
10.5
58.2
-
101
Table 9.— Proportional (%) availability and use of ground and tree houses by
dusky-footed woodrats in Plumas National Forest, 2004 to 2006. Numbers in
parentheses indicate standard error.
Ground houses
Tree houses
Female (%)
61 (4.7)
Cavity
37 (4.6)
Male (%)
71 (5.2)
24 (5.3)
5 (1.7)
58
27
15
Availability (%)
Limb
2 (0.7)
102
Table 10.— Mean home range (95%) and core area (50%) estimates and
associated standard errors (±SE) of dusky-footed woodrats using minimum
convex polygon (MCP) and fixed kernel (FK) methods at 2 study sites in the
Plumas National Forest, California.
Site Year
Sex
1
2004 Male
Female
2005 Male
Female
2006 Male
Female
2
2004 Male
Female
2005 Male
Female
2006 Male
Female
MCP
FK
N Home range Core area N Home range
5
1.2±0.2
0.3±0.1
3
1.1±0.4
Core area
0.4±0.2
9
0.8±0.1
0.2±0.04
6
1.1±0.3
0.4±0.1
5
1.9±0.6
0.6±0.2
5
1.7±0.7
0.6±0.2
8
1.2±0.2
0.3±0.1
7
1.4±0.3
0.6±0.1
3
1.8±0.5
0.4±0.1
1
3.0
0.9
7
1.2±0.3
0.4±0.1
5
1.2±0.3
0.5±0.1
7
3.7±0.3
0.9±0.1
4
3.5±0.4
1.0±0.2
6
2.8±0.3
0.7±0.1
4
2.9±0.7
1.0±0.3
2
7.0±0.4
2.6±0.3
2
7.7±0.2
2.8±0.3
4
5.0±0.5
1.5±0.2
4
6.3±0.7
2.4±0.2
2
4.6±0.4
1.2±0.6
2
3.6±1.6
1.5±0.9
5
2.9±0.6
0.7±0.1
3
4.0±1.2
1.3±0.2
103
Avian Monitoring in the Lassen and Plumas
National Forest
2008 Annual Report
March 2009
Ryan D. Burnett, Dennis Jongsomjit, and Diana Stralberg
PRBO Conservation Science
3820 Cypress Drive # 11
Petaluma, CA 94970
www.prbo.org
PRBO Contribution Number 1684
104
PRBO Avian Monitoring in the Plumas and Lassen National Forests - 2008
EXECUTIVE SUMMARY .......................................................................................... 107
ACKNOWLEDGEMENTS ......................................................................................... 107
MANAGEMENT RECOMMENDATIONS .............................................................. 109
CHAPTER 1. SHORT-TERM RESPONSE OF AVIAN SPECIES TO HFQLG
FUEL TREATMENTS IN THE NORTHERN SIERRA NEVADA ................... 114
BACKGROUND AND INTRODUCTION ............................................................................... 115
METHODS........................................................................................................................ 116
RESULTS ......................................................................................................................... 121
DISCUSSION..................................................................................................................... 126
LITERATURE CITED........................................................................................................ 129
CHAPTER 2. RESIDENT AND NEOTROPICAL MIGRATORY BIRD
RESPONSE TO ASPEN ENHANCEMENT ON THE LASSEN NATIONAL
FOREST ........................................................................................................................ 134
BACKGROUND AND INTRODUCTION ............................................................................... 135
PROJECT AREA ............................................................................................................... 136
METHODS........................................................................................................................ 136
RESULTS ......................................................................................................................... 139
DISCUSSION..................................................................................................................... 147
CONCLUSIONS ................................................................................................................. 150
LITERATURE CITED........................................................................................................ 150
APPENDIX 1. GPS COORDINATES ................................................................................... 154
CHAPTER 3. PILEATED WOODPECKER MONITORING ON THE
LASSEN NATIONAL FOREST .............................................................................. 158
BACKGROUND AND INTRODUCTION ............................................................................... 159
RESULTS ......................................................................................................................... 166
DISCUSSION..................................................................................................................... 170
CONCLUSIONS ................................................................................................................. 172
LITERATURE CITED........................................................................................................ 172
105
PRBO Avian Monitoring in the Plumas and Lassen National Forests - 2008
CHAPTER 4. RESIDENT AND NEOTROPICAL MIGRATORY BIRD
MONITORING IN MOUNTAIN MEADOWS ON THE ALMANOR RANGER
DISTRICT..................................................................................................................... 174
BACKGROUND AND INTRODUCTION ............................................................................... 175
METHODS........................................................................................................................ 175
RESULTS ......................................................................................................................... 179
DISCUSSION..................................................................................................................... 184
CONCLUSIONS ................................................................................................................. 185
LITERATURE CITED........................................................................................................ 186
106
PRBO Avian Monitoring in the Plumas & Lassen National Forests - 2008
Executive Summary
PRBO Conservation Science (PRBO) has been conducting songbird monitoring in
the Northern Sierra since 1997. In this report we present results from monitoring efforts
of forest management activities within the Herger Feinstein Quincy Library Group
(HFQLG) project area.
Chapter one investigates the short-term response of the avian community to a
suite of HFQLG treatments across the Plumas and Lassen National Forests. The response
to the different treatments varied among species with some showing positive effects and
others negative to most treatments. Prescribed fire had positive effects on the greatest
number of species while mastication and pre-commercial thinning affected the most
negatively.
The second chapter discusses results from monitoring aspen habitat on the Lassen
National Forest. Results show that treated aspen stands support greater total abundance
of birds and abundance of key species such as Mountain Bluebird, Chipping Sparrow,
and Red-breasted Sapsucker but these initial benefits may be short-lived for some
species. Avian abundance and richness indices have been increasing in aspen habitat in
the Lassen National Forest since 2005.
In Chapter three we discuss Pileated Woodpecker monitoring on the Lassen
National Forest. This project was focused on developing an effective monitoring plan
and spatially explicit habitat suitability model for this uncommon and elusive species.
We used a new landscape modeling technique (MaxEnt) to predict suitable habitat for
this species and targeted those areas for sampling using call back surveys. Results show
a far greater detection rate than from previous monitoring in the region and elucidate key
habitat components for the species. Our results suggest that suitable habitat for this
species occurs in areas with moderate to dense canopy closure and areas with large tree
components. We developed an interactive living GIS layer to help managers use up-todate information on detections of these species on the Lassen National Forest in project
planning.
In the fourth chapter we present results from monitoring of meadows in the
Almanor Ranger District of the Lassen National Forest from 2004 – 2008. We compared
results between sites and across years. Results suggest ARD meadows support diverse
and abundant bird populations including several species of conservation concern.
Meadow bird populations at these sites have been relatively stable from year to year and
across the five year period.
Acknowledgements
PRBO’s work in the Northern Sierra Nevada is a multi-project program with
several funding sources. Funding is provided by Region Five of the USFS through the
Pacific Southwest Research Station, Sierra Nevada Research Center as well as the
National Fire Plan. Additional funding is provided by the Lassen National Forest and
directly through Herger Feinstein Quincy Library Group Forest Recovery Act monitoring
funds. We wish to thank Peter Stine for his leadership and guidance with the PlumasLassen Study and staff of the Lassen and Plumas National Forest who support and
107
PRBO Avian Monitoring in the Plumas & Lassen National Forests - 2008
continue to advocate for our work, especially Mark Williams, Tom Rickman, Coye
Robbins, Tom Frolli, and Bobette Jones of the Lassen National Forest. We are indebted
to our hard working field crews who have spent the long hours in the field collecting the
massive amount of data required to produce such a report. The 2008 crew included crew
leaders Tim Guida and Paul Taillie and crew members Juan Caicedo, Pedro Costa, and
Ulla Kail. Diana Humple edited chapter’s 2 and 4 of this report.
108
PRBO Avian Monitoring in the Plumas & Lassen National Forests - 2008
Management Recommendations
General/Fuel Treatments








Manage for forest heterogeneity and diversity of habitat types and conditions placing
priority on those that exist in small quantities, have been significantly reduced in
quality or extent, or are disproportionately important to wildlife and ecosystem
function (e.g. aspen, shrub, pine-hardwood, meadows, late successional forest).
Restrict all activities that may disturb breeding bird habitat (e.g. timber harvesting,
grazing, burning, herbicide treatments, shrub treatments) to the non-breeding season
(August - April).
Maximize snag retention in all projects, including old snags ready to topple. Where
priority snags do not occur in high densities save senescing trees and shorter or
smaller snags than are currently in snag retention guidelines. Snags as small as eight
inches DBH and two meters tall are used by several species of cavity nesting birds
(e.g. White-headed Woodpecker). Snags ready to topple are the next generation of
down wood, important for many species including Pileated Woodpecker and Oregon
Junco.
Manage coniferous habitat for uneven aged stands with structural diversity including
multiple canopy layers and openings that supports shrub and herbaceous understory.
Focus DFPZ and other forest thinning in dense white fir dominated size class 3 stands
to develop more forest heterogeneity that is positively correlated with many avian
species.
Promote more late successional open forests conditions that support shrub and
herbaceous understory plant communities. Forests with large trees and 20-30%
canopy cover such as the shelter woods on the Swain Experimental forests support an
abundant and diverse bird community including declining species such as Olive-sided
Flycatcher and Chipping Sparrow.
Promote the development of forests with old-growth characteristics. Treatments in
these areas should focus on ensuring their persistence on the landscape and avoiding
impacts that alter their suitability for species such as Pileated Woodpecker.
Use prescribed fire and wildland fire use to achieve fuel reduction goals.
Aspen




Aspen habitat enhancement and expansion should be among the highest priorities as
aspen is rare on the landscape and the single most species rich avian habitat in the
Northern Sierra.
Promote aspen regeneration to increase overall aspen cover and an understory aspen
component. Aspen in the understory size classes were highly correlated with several
key bird indices in the ELRD.
Manage aspen habitat for multiple age and cover classes. Early successional open
canopy aspen habitat support a number of bird species of interest (e.g. Mountain
Bluebird, Chipping Sparrow).
Develop strategies for treating Aspen within riparian areas that support, or will
support, willows, alders, and other deciduous riparian vegetation. Aspen habitat with
109
PRBO Avian Monitoring in the Plumas & Lassen National Forests - 2008



these components, harbor a greater diversity and abundance of breeding birds than
any other habitat in the Northern Sierra.
Retain all snags over eight inch DBH in aspen treatments regardless of species,
though highest priority should be given to retaining aspen snags.
Reduce or eliminate over-browsing/grazing in regenerating Aspen stands through
fencing or removal of livestock from the area of concern to ensure long-term
continued regeneration and structurally diverse aspen stands.
Consider the potential negative impacts grazing adjacent to aspen treatments has on
the abundance of cowbirds and the potential ramifications on open cup nesting birds.
Pine Hardwood





Prioritize an inventory and delineation of all potential areas for pine-hardwood
enhancement at the district level.
Maximize snag retention focusing on retaining multiple decay classes. Retain all oak
and pine snags and where hazard trees are found top them to retain higher densities of
snags.
As both structural diversity and foliage volume are key avian habitat features,
restoring both should be a management priority for pine-hardwood enhancement.
Suckering of oaks would provide more mid-story foliage volume an important
foraging component for many insectivorous birds.
It is imperative to manage for understory habitat structure - including dense patches
of shrubs and herbaceous plant species - in pine-hardwood habitat enhancement
projects. Designing treatments that will create a mosaic of varying canopy covers
(e.g. 10 – 60%) across stands in combination with prescribed burning should promote
the establishment and enhance existing understory plant communities.
Develop Pine-Oak treatments to create greater mosaics of canopy cover than was
implemented at Brown’s Ravine. 40% canopy cover can be achieved across a stand
by creating dense clumps of conifers interspersed with semi-open pine-oak patches
and open canopies areas dominated by shrubs and regenerating oak and pine.
Montane Shrub




Consider the ecological value of shrubs within forested habitats and especially where
they occur in shrub fields in project planning and design and consider the long-term
viability of shrub habitats under the SNFPA.
Manage a portion (e.g. 50%) of group selections for natural regeneration, including
allowing for shrub communities to dominate some sites.
Allow some areas to regenerate naturally following stand replacing fire events rather
than salvaging, masticating, and re-planting for quick development of conifers. This
should promote greater diversity in habitat structure on the landscape, uneven aged
stands, and shrub and snag habitat for numerous avian and other wildlife species.
Prioritize sites that are, or have the potential to regenerate, mixed species shrub fields
(e.g. whitethorn, Manzanita, chinquapin, gooseberry, etc.). Mixed species shrub
habitats have higher diversity and abundance of shrub nesting bird species than
monotypic stands (e.g. Manzanita fields).
110
PRBO Avian Monitoring in the Plumas & Lassen National Forests - 2008





Retain high snag densities in group selections as snags in open areas are readily used
by numerous cavity nesting species (e.g. woodpeckers, bluebirds, swallows). Several
shrub study plots support up to five species of woodpecker within a 10 hectare area,
including Pileated, Hairy, White-headed, and Red-breasted Sapsucker.
Replant conifers in group selections not slated for natural regeneration in a clumped
design in order to create a mosaic with a semi-open canopy that invigorates shrub
development in the openings and reduces the need to re-enter sites for thinning in 10 20 years.
Design DFPZ plantation treatments and other thinning projects to create structural
diversity by thinning to create some open patches with little canopy cover. In these
openings promote lush and dense shrub communities.
Apply prescribed fire treatments in decadent shrub fields where growth and live
vegetative cover is now reduced. Manage these areas for regeneration of a newly
invigorated shrub community.
Greatly expand the use of under burns in thinning and group selection treatments to
allow herbaceous and shrub seeds access to mineral soils to promote their
regeneration.
Meadows





Manage for wet meadows with functional hydrology and long-term resiliency in
the face of changing climate patterns.
Foster partnerships with local government, state agencies, and non-profit
organizations to ensure meadow protection, enhancement, and long-term
management
Promote dense clumps of riparian deciduous shrubs and trees interspersed with
tall lush herbaceous vegetation.
Minimize non-natural disturbance such as livestock grazing and off-highway
vehicle use.
Manage for both low and high elevation meadows as they support different avian
assemblages.
Presentations
Using Birds to Guide National Forest Management in the Sierra Nevada – oral
presentation – International Partner’s in Flight Conference – 2/16/08 – McAllen, TX.
Listening to the Birds: An Ecosystem Approach to Sierra Nevada Management – invited
oral presentation – Society of American Foresters – California Chapter Annual Meeting –
1/31/2009 - Sacramento, CA.
Avian Monitoring in the HFQLG Area – 2007. Plumas-Lassen Study Symposium –
3/28/2008 - Quincy, CA.
111
PRBO Avian Monitoring in the Plumas & Lassen National Forests - 2008
Outreach
“Birds in the Park” – presentation on managing coniferous forest for birds and bird
banding demonstration in collaboration with Lassen Volcanic National Park – over 200
park visitors participated 7/20/08.
Sierra Institute Bird Field Tour – Lead field tour to discuss the importance of meadows to
Sierra Nevada birds including presentation and banding demonstration. – 6/28/2008.
Bird Banding Field Trip – coordinated outreach field trips with the Lassen National
Forest to view bird banding and discuss the use of birds as indicators in forest
management, PLAS study, and PRBO –7/16/2008.
Participated in several Forest Service Field Trips on the Almanor and Eagle Lake Ranger
Districts.
112
PRBO Avian Monitoring in the Plumas & Lassen National Forests - 2008
This page left intentionally blank.
113
PRBO Avian Monitoring in the Plumas & Lassen National Forests - 2008
Chapter 1. Short-term response of avian species to HFQLG fuel
treatments in the Northern Sierra Nevada
Sophie Webb
Ryan D. Burnett and Diana Stralberg
PRBO Conservation Science
114
Ch. 1Fuel Treatments
PRBO Avian Monitoring in the Plumas & Lassen National Forests - 2008
Background and Introduction
The Records of Decisions for the Sierra Nevada Forest Plan Amendment and
Herger Feinstein Quincy Library Group Forest Recovery Act direct the Forest Service to
maintain and restore old forest conditions that provide habitat for a number of plant and
animal species (HFQLG 1999, SNFPA 2001, 2004). Simultaneously, the direct the
Forest Service to take steps to reduce risks of large and severe fire by removing
vegetation and reducing fuel loads in overstocked forests (HFQLG 1999, SNFPA 2004).
Striking a balanced approach to achieving these potentially competing goals is a
significant challenge to effectively accomplish the various desired outcomes of forest
management (NFMA 1976).
Historically, fire was the primary force responsible for creating and maintaining
habitat diversity and landscape heterogeneity in the Sierra Nevada (Skinner and Chang
1996). Over the past century, fire return intervals have been lengthened and the area
affected by wildfire annually has been dramatically reduced in the interior mountains of
California (Taylor 2000, Taylor and Skinner 2003, Stephens et al. 2007). Thus, there is
little doubt fires role in influencing the composition of the Sierra Nevada landscape has
been reduced (Skinner and Chang 1996).
Fire suppression in concert with past silvicultural practices has resulted in
increased stand densities, loss of landscape heterogeneity, and increased fuel loads in
Sierra Nevada Forests (Vankat and Major 1978, Parsons and DeBenedetti 1979,
McKelvey and Johnston 1992, Minnich et al. 1995, Taylor and Skinner 2003). While the
ways in which these changes affect fire patterns and vegetation dynamics are frequently
discussed, they also undoubtedly impact the wildlife species that inhabit these forests. In
fact, many of the avian species now believed to be declining in the Sierra Nevada are
those associated with disturbance dependent habitat types and structure (Burnett et al. in
review).
Mechanical silvicultural treatments have the potential to fill some of fire’s historic
role in maintaining disturbance dependent habitats (Weatherspoon 1996, Arno and
Fiedler 2005). There has been considerable study of silvicultural treatments and their
effects on landbirds in eastern North American forests (Anand and Thompson 1997, King
et al. 2001, Fink et al. 2006, Askins et al. 2007) and the Cascades (Hansen et al. 1995,
Hagar et al. 2004, Chambers et al. 2007), but little published information exists on the
effects of mechanical fuel treatments on the avian community in the Sierra Nevada (but
see Siegel and DeSante 2003 and Garrison et al. 2006).
Forest Service management practices, primarily in the form of fuel reduction
treatments, are resulting in changes in habitat composition and structure across the
HFQLG area. By monitoring the populations of a suite of landbird species we can
measure the effectiveness of management actions in achieving a sustainable and
ecologically functional forest ecosystem. Specifically, we are interested in determining
the responses of landbirds to management practices intended to produce forests with
larger trees and high canopy cover along with more open-canopy, smaller size class forest
with reduced ladder and ground fuels.
In this chapter we investigate how a broad range of avian species respond to
changes in vegetation structure and composition that occur when forests are managed to
reduce fuels and generate timber products under the Herger Feinstein Quincy Library
Group Forest Recovery Act Pilot Project (HFQLG 1999). We investigated the short-term
115
Ch. 1Fuel Treatments
PRBO Avian Monitoring in the Plumas & Lassen National Forests - 2008
response of 17 breeding landbird species (e.g. passerines, woodpeckers) to a suite of
HFQLG treatments in the Lassen and Plumas National Forests between 2002 and 2008.
Methods
Study Location
The study occurred in the Lassen and Plumas National Forests within the
boundaries of the Herger-Feinstein Quincy Library Group Forest Recovery Act Pilot
Project (HFQLG). The study sites encompassed portions of Butte, Lassen, and Plumas
Counties at the intersection of the Sierra Nevada and Cascade mountains of Northeastern
California, USA (Figure 1). Survey sites ranged in elevation from 956m to 1896m within
the mixed conifer, true fir, and yellow pine zones.
Site Selection
We combined data across multiple projects on the Almanor and Eagle Lake
Ranger Districts of the Lassen National Forest and the Mt. Hough Ranger District of the
Plumas National Forest to investigate the effects of HFQLG treatments on landbirds
(Table 2). For the Plumas-Lassen study, three transects were established in each planning
watershed, (CalWater 1999), using a random starting point generated in a GIS
environment (ArcView 3.2a). For each transect, 11 additional points were added using a
random compass bearing from the starting point and spaced at approximately 250 m
intervals. If transects could not be established along a random bearing due to inaccessible
areas being encountered (e.g. private property, steep topography) we attempted a nonrandom bearing; if they still could not be established we placed the transect on or
adjacent to the secondary road nearest the starting point. A total of 876 stations along 73
transects were established in this manner across the 24 planning watersheds in the study
area.
116
Ch. 1Fuel Treatments
PRBO Avian Monitoring in the Plumas & Lassen National Forests - 2008
Table 1. Forest treatment types in the Northern Sierra Nevada for which the response of landbirds
was investigated.
Treatment
Defensible Fuel Profile Zones
Group Selection
Pre-commercial Thinning
Mastication
Prescribed Fire
Description
Shaded Fuel Break, generally linear in
shape, affects more acres than any
treatments in our study area
Removal of all overstory trees in 0.5 – 2
acre area, often embedded within a DFPZ
network
Removal of understory trees and shrubs,
often conducted prior to removal of
overstory trees but also used extensively as
independent treatment in Meadow Valley
(e.g. Waters project)
Mechanical shredding of shrubs that
sometimes uproots shrubs but often leaves
plant alive below ground that regenerate.
Generally low intensity human ignited
burning. Generally consumes understory
fuels and some middle story trees
A number of the sites that were intended to be part of the untreated sample were treated
either immediately before or during the course of this study (2002 – 2008) as part of
projects were unknown to us, or due to changes in treatment locations during the
planning process. We also established additional transects in areas slated to be treated as
part of the Meadow Valley project. For a more detailed description of site selection for
the Plumas-Lassen study see Stine et al. (2005).
DFPZ treatments monitored on the Eagle Lake Ranger District were established
in 2004 after consulting ranger district staff and available GIS layers. We selected 6 sites
that were slated for treatment in the next several years. At each treatment area we
established between 5 to 7 point counts inside of treatment boundaries and 5 to 8 sites in
similar habitat at least 100m outside the treatment but within 500m of the treated area
(see Burnett et al. 2004).
A similar protocol was used for the Brown’s Ravine Black Oak enhancement
DFPZ project in the Almanor Ranger District of the Lassen. In this project, treatment
units were larger so we filled each unit with points spaced 220m apart. Each unit
contained between 5 and 14 points. Control sites were established in adjacent units
where no treatment was planned (Burnett et al. 2004).
Survey Protocol
We used a standardized five-minute multiple distance band circular plot point
count census (Buckland et al. 1993, Ralph et al. 1993, Ralph et al. 1995) to sample the
avian community in the study area. In this method, points are clustered in transects, but
data were only collected from fixed stations, not along the entire transect.
117
Ch. 1Fuel Treatments
PRBO Avian Monitoring in the Plumas & Lassen National Forests - 2008
All birds detected at each station during the five-minute survey were recorded
according to their initial distance from the observer. These detections were placed within
one of six categories: within 10 meters, 10-20 meters, 20-30 meters, 30-50 meters, 50100 meters, and greater than 100 meters. The method of initial detection (song, visual, or
call) for each individual was also recorded. All observers underwent intensive 14 day
training in bird identification and distance estimation prior to conducting surveys. Laser
rangefinders were used to assist in distance estimation at every survey point.
Counts began around local sunrise, were completed within four hours, and did not
occur in inclement weather. Each transect was visited twice during the peak of the
breeding season from mid May through the first week of July in each year.
Analysis
Annual per-point species abundance and diversity metrics were summarized for
1,194 point-count locations surveyed between 2002 and 2008. For this analysis we
excluded detections beyond 50 m, as well as single surveys that were not repeated within
a season, resulting in a total sample size of 5,826 point-visits (Table 2). For each pointyear combination, the total number of detections for each of 17 species was calculated by
summing across two visits (each point was surveyed exactly twice). The 17 species were
comprised of all of the Coniferous Forest Focal species, (CALPIF 2002), for which we
had adequate detections to conduct meaningful analysis as well as eight additional
species that represented a range of habitat preferences and were relatively common in the
study area (Table 3). We also calculated overall species richness, conifer focal species
richness (CALPIF 2002), and Shannon and Simpson diversity metrics, for each point in
each year.
For each point-count location, we identified the treatment history with respect to
five distinct treatment types (Table 1). A given treatment was only considered to occur at
a point if the point fell inside the treatment polygon. An exception was made for group
selection treatments, due to their small size and their relatively extreme effects (removal
of all trees); a point was considered inside a group selection treatment if it was within a
group or was outside a group but within 25 m of the treatment edge. Of the 1194 points,
249 were treated in one or more ways; the remaining points were considered control sites.
For each point in each year, we then calculated the number of years since treatment for
each of the five treatments types. Since we did not have site specific historical treatment
and fire data we assigned all untreated sites an estimating average time since treatment or
fire. Pre-treatment and control sites were considered to be 35 years since timber removal
treatments, and 75 years since fire, based on estimates of fire exclusion in mixed conifer
habitat in the Sierra Nevada (Skinner and Chang 1996). The minimum time since
treatment was 1 year, as most treatments were implemented in the fall, after the the point
count survey season. If a treatment was performed in the spring (i.e., before survey
season), it was considered to have occurred the previous year (time = 1). If a site was
treated in the middle of the survey season, the surveys from that year were excluded from
analysis, as we were unable to determine whether surveys were conducted before or after
the treatment.
Other variables calculated from GIS layers for each study site included elevation,
slope, annual solar radiation, vegetation type (California Wildlife Habitat Relationships
118
Ch. 1Fuel Treatments
PRBO Avian Monitoring in the Plumas & Lassen National Forests - 2008
classification from the US Forest Service CalVeg layer), and presence of a riparian
habitat conservation area.
Table 2. The number of point count stations and total surveys conducted by treatment type in each
ranger district in PRBO’s Northern Sierra study area. Each point was visited twice in each year it
was surveyed.
Treatment Type
Total
DFPZ
Group Selection
Pre-commercial Thin
Mastication
Prescribed Burn
Number of points
Number of point visits
Number of points
Number of post-treatment point visits
Number of points
Number of post-treatment point visits
Number of points
Number of post-treatment point visits
Number of points
Number of post-treatment point visits
Number of points
Number of post-treatment point visits
Almanor
165
787
57
284
0
0
26
52
4
8
0
0
Eagle
Lake
71
264
29
112
0
0
0
0
0
0
0
0
Mt.
Hough
958
4775
30
148
19
78
24
208
32
242
40
344
For each species and diversity metric, we constructed a mixed-effects model
including five treatment effects (time since each of the five treatment types), five
covariates (described above, 1 categorical and 4 continuous), and a random site (point)
effect to account for the lack of independence within a given site across multiple years.
Models for species diversity metrics, which had nearly normal distributions, were
specified as linear models with a Gaussian distribution using the ‘nlme’ package for R (R
Development Core Team 2009). Individual species abundance, which were generally
approximated by the negative binomial distribution, were specified as generalized linear
models with a negative binomial distribution using a penalized quasi-likelihood approach
with the ‘MASS’ page for R. The dispersion parameters for the negative binomial mixed
models were estimated using a standard generalized linear model (‘glm.nb’ function) and
provided as inputs to the mixed models.
119
Ch. 1Fuel Treatments
PRBO Avian Monitoring in the Plumas & Lassen National Forests - 2008
Figure 1. Location of HFQLG treatment projects where landbirds were monitored in the Lassen and
Plumas National Forests with the Plumas-Lassen (PLAS) study units, treatment types, and point
count locations shown.
120
Ch. 1Fuel Treatments
PRBO Avian Monitoring in the Plumas & Lassen National Forests - 2008
The significance of treatment effects and covariates were evaluated at a 95%
confidence level (P<0.05). Model-predicted focal species abundance and species richness
were calculated for each of the five treatment types, as well as for the combined effect of
all treatments. For each of the treatment predictions, all other continuous model variables
were held constant at their mean values; vegetation type was assigned to Sierran Mixed
Conifer, the most common vegetation type in the dataset (occurring at 666 points). The
value of the treatment effect was set at one to indicate one year post-treatment.
Results
The mean abundance per point count station for the 17 species we investigated
ranged from 1.02 for Hermit Warbler to 0.03 for Olive-sided Flycatcher (Table 3). Fox
Sparrow and Hairy Woodpecker, the two new Management Indicator Species for the
Forest Service in the Sierra Nevada had an abundance of 0.33 and 0.06 respectively in
our study area.
Table 3. Common and scientific names of the 17 species investigated for effects of fuel treatments in
the Northern Sierra Nevada with the mean abundance per point count station per year (summed
across 2 visits) and standard deviation. California Partner’s in Flight Coniferous Forest Focal
Species are in bold.
Common Name
Hermit Warbler
Oregon Junco
Dusky Flycatcher
Audubon's Warbler
Mountain Chickadee
Nashville Warbler
Golden-crowned Kinglet
Red-breasted Nuthatch
Western Tanager
Fox Sparrow
Hammond's Flycatcher
Brown Creeper
MacGillivray's Warbler
Steller's Jay
Hairy Woodpecker
Chipping Sparrow
Olive-sided Flycatcher
Scientific Name
Dendroica occidentalis
Junco hyemalis oreganus
Empidonax oberholseri
Dendroica auduboni
Poecile gambeli
Vermivora ruficapilla
Regulus satrapa
Sitta canadensis
Piranga ludoviciana
Passerella iliaca
Empidonax hammondii
Certhia americana
Oporornis tolmiei
Cyanocitta stelleri
Picoides villosus
Spizella passerina
Contopus cooperi
Mean
1.02
0.71
0.66
0.63
0.62
0.59
0.44
0.39
0.35
0.33
0.23
0.23
0.20
0.12
0.06
0.04
0.03
SD
1.26
0.96
1.02
0.88
0.95
0.98
0.77
0.73
0.64
0.85
0.58
0.51
0.53
0.41
0.26
0.26
0.19
Of the 17 species we investigated, 14 showed a significant association with at
least one treatment type (Table 4, Figure 2). Seven species showed significant effects of
DFPZs, 5 of group selections, 6 of pre-commercial thinning, 5 of mastication, and 7 of
prescribed burning. Chipping Sparrow was the only species that had a significant effect
with each of the five treatments; no other species had more than three. Three species,
Olive-sided Flycatcher, Audubon’s Warbler, and Chipping Sparrow all increased in
121
Ch. 1Fuel Treatments
PRBO Avian Monitoring in the Plumas & Lassen National Forests - 2008
abundance following DFPZ treatment. Contrastingly, Dusky Flycatcher, Goldencrowned Kinglet, Nashville Warbler, and Hermit Warbler showed negative responses. Of
the five species with significant responses to group selection, Olive-sided Flycatcher,
Dusky Flycatcher, MacGillivray’s Warbler, and Chipping Sparrow responded positively,
while Hammond’s Flycatcher had a negative response. Hairy Woodpecker, Brown
Creeper, Audubon’s Warbler, Chipping Sparrow, and Western Tanager all showed a
negative relationship with pre-commercial thinning, while only Olive-sided Flycatcher
responded positively to this treatment. Hairy Woodpecker, Nashville Warbler,
Audubon’s Warbler, Fox Sparrow, and Chipping Sparrow all had negative associations
with mastication while no species showed a positive effect of this treatment. Of the 7
species that had a significant relationship with prescribed fire treatments, only Goldencrowned Kinglet’s was negative. Hairy Woodpecker, Dusky Flycatcher, Mountain
Chickadee, Fox Sparrow, Chipping Sparrow, and Western Tanager all responded
positively to prescribed fire.
For the four measures of species diversity examined, only the treatments of precommercial thinning and mastication had significant effects. Pre-commercial thinning
had a negative effect on all four measures and mastication negatively affected all but
conifer focal species richness (Table 4).
122
Ch. 1Fuel Treatments
PRBO Avian Monitoring in the Plumas & Lassen National Forests - 2008
Table 4. The effect of time since five separate treatments on the abundance of 17 species, and four diversity metrics in the Herger Feinstein Quincy
Library Group Pilot Project area. Negative coefficients represent negative associations with time since treatment, which means there was a positive
response to the treatment. DFPZ = Defensible Fuel Profile Zone, Group = Group Selection, PCThin = Pre-commercial Thin, Mast = Mechanical
Mastication, and Burn = Prescribed Burn. ELEV = elevation, VegType = California Wildlife Habitat Relationship Habitat Type, SolRAD = Solar
Radiation Index, RHCA = Riparian Habitat Conservation Area.
Metric
Focal Species Richness
Species Richness
Shannon Diversity
Simpson Diversity
Hairy Woodpecker
Olive-sided Flycatcher
Dusky Flycatcher
Hammond's Flycatcher
Steller's Jay
Mountain Chickadee
Red-breasted Nuthatch
Brown Creeper
Golden-crowned Kinglet
Nashville Warbler
Audubon's Warbler
Hermit Warbler
MacGillivray's Warbler
Fox Sparrow
Chipping Sparrow
Oregon Junco
Western Tanager
DFPZ
-0.0025
0.0033
0.0003
0.0002
0.0054
-0.0373***
0.0050*
0.0066
-0.0044
-0.0013
-0.0035
-0.0008
0.0142***
0.0260***
-0.0061*
0.0159***
0.0050
0.0015
-0.0555***
-0.0010
0.0016
Group
0.0005
-0.0131
-0.0037
-0.0014
0.7535
-0.0666***
-0.0150**
0.0207*
0.0009
0.0008
0.0016
0.0201
0.0042
-0.0030
0.0137
-0.0060
-0.0208*
0.0020
-0.0620***
-0.0013
-0.0033
PCThin
0.0135**
0.0283***
0.0069***
0.0024***
0.0254*
-0.0576***
0.0066
0.0104
-0.0070
0.0087
0.0048
0.0166*
0.0082
0.0061
0.0106*
0.0078
0.0140
0.0048
0.0385***
0.0028
0.0121*
Mast
0.0079
0.0175*
0.0039*
0.0012*
0.0415***
-0.0024
-0.0009
0.0148
-0.0087
0.0001
-0.0007
0.0082
0.0092
0.0392**
0.0079*
-0.0039
0.0086
0.0301***
0.0310*
-0.0054
-0.0061
Burn
-0.0047
-0.0086
-0.0017
-0.0005
-0.0129**
-0.0032
-0.0046*
-0.0062
0.0011
-0.0054**
0.0018
-0.0019
0.0086**
0.0048
-0.0026
0.0006
0.0037
-0.0078*
-0.0306***
-0.0027
-0.0060**
Other significant effects
Elev (+), VegType
Elev (+), VegType
Elev (+), VegType
Elev (+), VegType
Elev (+)
Elev (+)
Elev (+), Slope (-), RHCA (+), VegType
Elev (+), Slope (-), VegType
Elev (-), Slope (+), RHCA (-), VegType
Elev (+), VegType
Elev (+), VegType
Elev (-), VegType
Elev (+), VegType
Elev (-), SolRad (+), VegType
Elev (+), VegType
Elev (-), VegType
Elev (+), VegType
Elev (+)
Slope (-), SolRad (+), VegType
SolRad (+)
Elev (-), SolRad (+), VegType
* = P<0.05, ** = P<0.005, *** = P<0.0005
123
Ch. 1. Fuel Treatments
PRBO Avian Monitoring in the Plumas & Lassen National Forests - 2008
Figure 2. Predicted species abundance in the year following each of five treatments, as well as a
hypothetical combination of all five treatments (“all”). Predicted abundance (sum over two visits)
for each treatment was modeled for t=1 year since treatment, and all other variables were held
constant at their mean values (except VegType, which was assigned “Sierran Mixed Conifer” type).
Predicted values that were significantly different from the mean at untreated sites (dashed red line)
are indicated with asterisks. Treatments included Defensible Fuel Profile Zones (DFPZ), group
selections (Group), pre-commercial thin (PCT), mastication (Mast), and prescribed fire (Burn).
1.2
Olive-sided Flycatcher
***
0.05
Dusky Flycatcher
1
**
0.04
0.8
***
*
0.03
0.6
0.02
0.4
0.01
***
0.2
0
0
DFPZ Group PCT
0.2
*
Mast
DFPZ Group PCT
Burn
1.4
Hammond's Flycatcher
*
Mast
Burn
All
Mountain Chickadee
1.2
**
1
0.8
0.6
0.4
0.2
0
0
DFPZ Group PCT
0.4
Mast
Burn
All
DFPZ Group PCT
0.6
Brown Creeper
Mast
Burn
All
Golden-crowned Kinglet
0.4
0.2
***
*
**
0.2
0
0
DFPZ Group PCT
Mast
Burn
All
DFPZ Group PCT
Mast
Burn
All
124
Ch. 1. Fuel Treatments
PRBO Avian Monitoring in the Plumas & Lassen National Forests - 2008
Figure 2. continued
0.8
1
Nashville Warbler
Audubon's Warbler
0.8
0.6
*
0.6
*
*
0.4
***
0.4
**
0.2
0.2
0
0
DFPZ Group PCT
0.6
Mast
Burn
DFPZ Group PCT
All
0.2
MacGillivray's Warbler
Mast
Burn
All
Fox Sparrow
*
0.4
*
0.2
***
0
0
DFPZ Group PCT
Mast
Burn
DFPZ Group PCT
All
Chipping Sparrow
Mast
Burn
All
Western Tanager
1
***
0.1
0.8
**
0.6
0.4
***
***
*
0.2
***
*
0
0
DFPZ Group PCT
Mast
Burn
All
DFPZ Group PCT
Mast
Burn
All
125
Ch. 1. Fuel Treatments
PRBO Avian Monitoring in the Plumas & Lassen National Forests - 2008
Discussion
Overview
Fuel reduction treatments in our study area significantly influenced the abundance
of most of the species we investigated, with both positive and negative effects detected.
However, our results suggest prescribed fire benefits the greatest number of species while
negatively impacting the fewest while mastication and pre-commercial thinning benefited
the fewest species and had negative impacts on the most. Though there are several
limitations to this analysis, with its relatively large sample size and geographic scope it
fills a gap in information about the effects of fuel treatments on wildlife species in the
Sierra Nevada. Mechanical silvicultural treatments appear capable of providing habitat
for some disturbance dependent bird species but also may reduce the suitability for
species associated with higher canopy cover and later successional forests. Management
decisions should be made in the context of current trends in forest structure and
disturbance patterns in order to strike a balance that ensure the needs of the greatest
number of species are being met.
Limitations and Caveats
This study investigated the short-term effects (1 – 6 years post-treatment) of fuel
reduction activities, and thus provides an incomplete picture of treatment effects on
breeding landbirds. Post-treatment successional processes may result in considerable
change at these sites over longer time periods, though recent evidence suggests at least
DFPZ sites change little in vegetative structure in the first 15 years following treatment
(S. Stephens pers. comm.).
The results of this study should also be considered in the context of the conditions
that existed in the study area prior to implementation of these treatments. After over a
century of resource extraction and fire suppression, these forests should not be considered
natural as untreated sites have all been subjected to past timber harvest and a century of
fire suppression. We attempted to account for this in our estimates of time since treatment
at control sites (35 years for mechanical treatments and 100 years for burns), although
models would likely have been improved with site specific information about historic
timber management practices and fire occurrence.
Our analysis was focused primarily on species that are fairly common to
abundant. The species that are most sensitive to silvicultural treatments may already be
quite rare in these forests, which have been actively managed for over a century.
However, other studies in western forests have shown that few if any landbird species
appear to be negatively affected by fragmentation or habitat edges (McGarigal and
McCombs 1995, Scheick et al. 1995, Tewskbury et al. 1998, 2006, George and Dobkin,
2002).
Our analysis of pre-commercial thinning was limited to sites that received no
overstory treatment (e.g., DFPZ or group). Many of the group selection and DFPZ
treatments underwent pre-commercial thinning at the same time as these overstory
treatments were implemented. As a result we were unable to isolate the relative effects of
pre-commercial components within these treatments.
Finally, it is important to consider that this study only investigated the abundance
patterns of species and not demographic parameters (productivity or survival).
126
Ch. 1. Fuel Treatments
PRBO Avian Monitoring in the Plumas & Lassen National Forests - 2008
Abundance (or density), may not always be a good estimate of the suitability of habitat
for a species (VanHorne 1983, Bock and Jones 2004).
DFPZ and Group Selection: Promoting Heterogeneity
Group selection treatments, which are basically 0.5 – 2 acre clear cuts, had
predominantly positive short-term effects on the landbird species we investigated. Only
Hammond’s Flycatcher, a species associated with shaded mature forest, showed a
negative response. Similarly, Hagar et al. (2004) found evidence of this species being
sensitive to high intensity treatments in the Pacific Northwest.
Two species that have undoubtedly been declining in the Sierra Nevada for many
years, Olive-sided Flycatcher and Chipping Sparrow, both responded positively to group
selections. Given the Olive-sided Flycatcher’s strong associations with forest
heterogeneity and contrasting edges (McGarigal and McCombs 1995, Howell and
Burnett In review, Meehan and George 2003), group selection type treatments are likely
creating habitat for this species. However, mechanical silvicultural treatments that mimic
natural disturbance may be an ecological trap for this species as predation rates may be
higher and food availability lower in mechanically treated areas compared to those that
have burned (Robertson and Hutto 2007). Further investigation of the demographic
parameters of this declining species in mechanically- and naturally-created (e.g., windthrow, wildfire) edges is warranted. Chipping Sparrow, the only species to have a
significant response to all five treatments were significantly more abundant in Group
Selections. Forest openings that promote herbaceous vegetation and open ground for
foraging are likely to benefit this species.
Two shrub associated species, MacGillivray’s Warbler and Dusky Flycatcher also
responded positively to group selections. Unlike Fox Sparrow, which requires relatively
large patches of open shrub-dominated habitat (Howell & Burnett In review), these two
species readily occupy small shrub filled forest gaps. Thus it seems appropriate that they
would benefit from group selection treatments. The increased light and presumably soil
moisture within group selections may facilitate rapid establishment and growth of shrub
habitat preferred by these species. Further analysis of the changes in vegetation
following treatment will be necessary to conclusive link habitat changes to the observed
effects of treatments.
DFPZs, the treatments affecting the greatest number of acres in our study area,
had mixed effects on the avian species we investigated. Not surprisingly, several of the
species associated with more mature higher canopy-cover forest (Hermit Warbler and
Golden-crowned Kinglet) showed a negative response to this treatment, as did the
ground-nesting and middlestory-foraging Nashville Warbler. The Hermit Warbler is the
most abundant breeding landbird in our study area (Table 3). The increased canopy
cover and densification of white fir dominated forest has probably increased the available
habitat for Hermit Warbler and Golden-crowned Kinglet. Though the Golden-crowned
Kinglet is also quite abundant in our study area, unlike the Hermit Warbler, this species
has been declining in the Sierra Nevada according the Breeding Bird Survey (Sauer et al.
2008). Another declining species, Nashville Warbler, showed a strong negative
association with DFPZ and mastication. This species nests on the ground in dense patches
of vegetation with heavy leaf litter and forages in the middlestory (Williams 1996). The
short-term negative effects of DFPZ and mastication for this species may be a result of
the reduction in these habitat components within these treatments. However, because they
are closely allied with black oak (Quercus kelloggii) in our study area, (Burnett and
127
Ch. 1. Fuel Treatments
PRBO Avian Monitoring in the Plumas & Lassen National Forests - 2008
Geupel 2002, Burnett and Howell in review), DFPZ treatments that retain oak and reduce
canopy cover to increase oak vigor and regeneration are likely to have long-term positive
effect on this species.
Mechanical treatments that significantly reduce canopy cover and create canopy
gaps can result in increased abundance of middle and understory associated landbirds in
western forests and overall avian diversity (Hansen et al. 1995, Siegel and DeSante 2003,
Hagar et al. 2004). Additionally, many forest interior associated birds may benefit from
small gaps in mature forest as they utilize the unique resources they provide such as fruit
and nectar (Thompson et al. 1992, Vitz and Rodewald 2006, Greenberg et al. 2007).
None of the shrub dependent species we investigated showed a positive response to
DFPZ treatments. This is likely due to our analysis being limited to the short-term
response of treatments. However, based on our experience with most of these treated
areas, the retention of over 40% canopy cover is unlikely to allow for understory foliage
volume, especially of shade intolerant shrubs. In order to more effectively mimic the
mosaic patterns created through natural disturbance and benefit a greater number of
species dependent upon disturbance we suggest - where appropriate - DFPZ treatments
consider a greater reduction in canopy cover (Chambers et al. 1999). A mosaic pattern
with areas with reduced canopy cover can enhance shade intolerant understory plant
assemblages and promote landscape heterogeneity (McGarigal and McCombs 1995,
Siegel and DeSante 2003).
Prescribed Fire vs. Mechanical Understory Treatments: Understory Structure
The importance of forest structural diversity for landbirds in western forests is
well established (Beedy 1981, Verner and Larson 1989, Wilson and Comet 1996). Thus
fuel treatments that remove and inhibit understory habitat structure can have negative
impacts on a number of avian species while benefiting relatively few (Rodewald and
Smith 1998).
For landbirds in our study area, prescribed fire treatments had a far greater
positive effect than mastication or pre-commercial thinning. The effects of mastication
and pre-commercial thinning had almost unanimously negative effects on the avian
community while burning was almost always positive. While all three treatments are
primarily designed to reduce understory fuels, it is quite clear that their impacts on birds
are disparate.
Many of the factors believed to be driving the increased abundance of bird species
in burned habitat, such as high densities of snags, increased abundance of some insect
populations, and increased seed availability, may not be facilitated through mechanical
treatments alone. Prescribed fire in the Sierra Nevada generally results in a reduction in
surface fuels while mechanical treatments without fire generally increase surface fuels
(Stephens & Moghaddas 2005, Stephens et al. 2009). Reduction in surface fuels and
release of nutrients can promote an increase in herbaceous vegetation following
prescribed fire (Wayman and North 2007). Combining mechanical treatment with
prescribed fire can result in similar surface fuel loads and vegetative response as burn
only treatments (Collins et al. 2007). However, fire may be more beneficial than
mechanical treatments for shrub dependent birds as it often results in greater retention of
shrub cover than mastication treatments (Collins et al. 2007, Wayman and North 2007).
Our results suggest a reevaluation of the benefits of pre-commercial thinning and
mastication treatments as they clearly have negative impacts on a number of avian
species including a number that are declining in the Sierra Nevada.
128
Ch. 1. Fuel Treatments
PRBO Avian Monitoring in the Plumas & Lassen National Forests - 2008
Prescribed fire as well as mechanical treatments during the bird breeding season
can result in direct loss of nests and dependent young. All of the burns we monitored
were in the Mt. Hough Ranger District, with the majority carried out in April or after July
thus avoiding the peak of the bird breeding season. However, each of the other treatment
types was carried out at least in part during the middle of the bird breeding season (May –
July).
Conclusions
Fuel reduction treatments varied in their effects on landbirds in our study area.
Group selection and prescribed fire benefited the greatest number of species while
negatively impacting the least. Mechanical mastication and pre-commercial thinning
benefited the least while negatively impacting the greatest. However, the goal of land
management may not always be to maximize the number of species that benefit from a
treatment while minimizing those that do not. This approach may lead to more
homogenization of the landscape. We suggest a more landscape based ecological
approach is prudent. Promoting an increase in late successional habitat in some locations
while prescribing greater reductions in canopy cover that mimic natural disturbance
patterns in areas where biological diversity is relatively low (e.g. closed canopy size class
3 and 4 white fir stands). Under current management strategies being implemented on
National Forest lands in the Sierra Nevada, the loss of late seral forest, landscape
heterogeneity, and fire-dependent habitats appear to be the greatest threat to biodiversity
here. A balanced approach using a full range of management tools and prescriptions is
advisable to ensure biodiversity is sustained.
Literature Cited
Beedy, E.C. 1981. Bird communities and forest structure in the Sierra Nevada of
California. The Condor 83: 97-105.
Bock C.E. and Z.F. Jones. 2004. Avian habitat evaluation: should counting birds count?
Frontiers Ecol Environ 2:403–410
Buckland, S. T., D. R. Anderson, K. P. Burnham, and J. L. Laake. 1993. Distance
sampling: estimating abundance of biological populations. Chapman and Hall, London.
Burnett, R.D. and D.L. Humple. 2003. Songbird monitoring in the Lassen National
Forest: results from the 2002 field season with summaries of 6 years of data. A PRBO
report to the U.S. Forest Service. Contribution # 1069.
Burnett, R.D., D.Humple, T.Gardali, and M.Rogner. 2004. Avian Monitoring in the
Lassen National Forest. A PRBO report to the USFS. Contribution # 1242.
Burnett, R.D. and N. Nur. 2007. Plumas-Lassen Area Study module on landbird
abundance, distribution, and habitat relationships: 2006 annual report. A PRBO Report
to the U.S. Forest Service. Contribution # 1550.
129
Ch. 1. Fuel Treatments
PRBO Avian Monitoring in the Plumas & Lassen National Forests - 2008
Burnett, R.D., N.Nur, and C.A.Howell. In review. Implications of spotted owl
management for landbirds in the Sierra Nevada, CA, USA. Forest Ecology and
Management.
CALPIF (California Partners in Flight). 2002. Version 1.0. The draft coniferous forest
bird conservation plan: a strategy for protecting and managing coniferous forest habitats
and associated birds in California (J. Robinson and J. Alexander, lead authors). Point
Reyes Bird Observatory, Stinson Beach, CA. http://www.prbo.org/calpif/plans.html.
CalWater (The California Interagency Watershed Mapping Committee). 1999. California
Watersheds Version 2.2. (available at www.ca.nrcs.usda.gov/features/calwater/)
Chambers, C.L., W.C. McComb, and J.C. Tappeiner II. 1999. Breeding bird responses to
three silvicultural treatments in the Oregon Coast Range. Ecological Applications 9: 171185.
Collins, B.M., J.J. Moghaddas, and S.L. Stephens. 2007. Initial changes in forest
structure and understory plant communities following fuel reduction activities in a Sierra
Nevada mixed conifer forest. Forest Ecology and Management 239: 102-111.
ESRI 2000. Arc View GIS 3.2a. Environmental Systems Research Institute. Redlands,
CA.
George, T.L, and D.S. Dobkin (eds.). 2002. Effects of habitat fragmentation on birds in
western landscapes: Contrasts with paradigms from the eastern U.S. Studies in Avian
Biology 25, 270p.
Greenberg, C.H., A.L. Tomcho, J.D. Lanham, T.A. Waldrop, J. Tomcho, R.J. Phillips,
and D. Simon. 2007. Short-term effects of fire and other fuel reduction treatments on
breeding birds in a Southern Appalachian upland hardwood forest. The Journal of
Wildlife Management 71: 1906-1916.
Hagar, J., S. Howlin, and L.Ganio. 2004. Short-term response of songbirds to
experimental thinning of young Douglas-fir forests in the Oregon Cascades. Forest
Ecology and Management 199: 333-347.
Hansen, A.J., W.C. McComb, R. Vega, M.G. Raphael, and M. Hunter. 1995. Bird
habitat relationships in natural and managed forests in the west cascades of Oregon.
Ecological Applications 5: 555-569.
HFQLG (Herger-Feinstein Quincy Library Group Forest Recovery Act) 1999. Final
Environmental Impact Statement, U.S. Department of Agriculture, Forest Service,
Pacific Southwest Region, Vallejo, CA.
http://www.fs.fed.us/r5/hfqlg/publications/1999_feis/TOC.htm
Howell, C.A. and R.D. Burnett. In review. Landscape effects on songbird abundance in
the Northern Sierra Nevada, USA. Forest Ecology and Management.
130
Ch. 1. Fuel Treatments
PRBO Avian Monitoring in the Plumas & Lassen National Forests - 2008
King, D.I., R.M. Degraaf, and C.R. Griffin. 2001. Productivity of early successional
shrubland birds in clearcuts and groupcuts in an eastern deciduous forest. Journal of
Wildlife Management 65: 345-350.
McGarigal, K., and W.C. McComb. 1995. Relationships between landscape structure and
breeding birds in the Oregon Coast Range. Ecological Monographs 65: 235-260.
Meehan, T.D. and T.L. George. 2003. Short-term effects of moderate to high-severity
wildfire on a disturbance-dependent flycatcher in northwest California. Auk 120: 11021113.
Ralph, C.J., G.R. Geupel, P. Pyle, T.E. Martin, and D.F. DeSante 1993. Field Methods
for Monitoring Landbirds. U.S. Department of Agriculture, Forest Service, General
Technical Report PSW-144.
Ralph, C.J., S. Droege, and J.R. Sauer. 1995. Managing and monitoring birds using
point counts: standards and applications. In C. J. Ralph, J. R. Sauer and S.
Droege (eds.), Monitoring Bird Populations by Point Counts. USDA Forest
Service Publication, Gen. Tech. Rep. PSW-GTR-149, Albany, CA .
Roberston, B.A. and R.L. Hutto. 2007. Is selectively harvested forest an ecological trap
for olive-sided flycatchers? The Condor 109: 109-121.
Rodewald, P.G. and K.G. Smith. 1998. Short-term effects of understory and overstory
management on breeding birds in Arkansas oak-hickory forests. The Journal of Wildlife
Management 62: 1411-1417.
Sauer, J. R., J. E. Hines, and J. Fallon. 2008. The North American Breeding Bird Survey,
Results and Analysis 1966 - 2007. Version 5.15.2008. USGS Patuxent Wildlife Research
Center, Laurel, MD
Schieck, J.K., B. Lertzman, B., Nyberg, and R.Page. 1995. Effects of patch size on birds
in old-growth montane forests. Conservation Biology 9: 1072-1084.
Siegel, R.B. and D.F. DeSante. 2003. Bird communities in thinned versus unthinned
sierran mixed conifer stands. Wilson Bulletin 115: 155-165.
Skinner, C.N. and C.Chang. 1996. Fire regimes, past and present. Sierra Nevada
Ecosystem Project: Final Report to Congress. Vol. 2, Assessments and scientific basis for
management options, pp. 1041-1069. University of California Centers for Water and
Wildland Resources, Davis, CA, USA.
SNFPA 2001. Sierra Nevada Forest Plan Amendment Final Environmental Impact
Statement, Record of Decision. United States Department of Agriculture, Forest Service,
Pacific Southwest Region, Vallejo, CA.
http://www.fs.fed.us/r5/snfpa/library/archives/rod/rod.pdf
131
Ch. 1. Fuel Treatments
PRBO Avian Monitoring in the Plumas & Lassen National Forests - 2008
SNFPA 2004. Sierra Nevada Forest Plan Amendment, Final Supplemental
Environmental Impact Statement, Record of Decision. United States Department of
Agriculture, Forest Service, Pacific Southwest Region, Vallejo, CA.
http://www.fs.fed.us/r5/snfpa/final-seis/rod/
Stine, P., Landram, M., Keane, J., Lee, D., Laudenslayer, B., Weatherspoon, P., and
Baldwin, J. 2002. Fire and fuels management, landscape dynamics, and fish and wildlife
resources: An integrated research plan for the Herger-Feinstein Quincy Library Group
Pilot Study Area.
http://www.fs.fed.us/psw/programs/snrc/forest_health/plas_studyplan.pdf
Stephens, S.L. and J.J. Moghaddas. 2005. Experimental fuel treatment impacts on forest
structure, potential fire behavior, and predicted tree mortality in a California mixed
conifer forest. Forest Ecology and Management 215: 21-36.
Stephens, S.L., J.J. Moghaddas, C. Edminster, C.E. Fiedler, S. Haase, M.Harrington, J.E.
Keeley, E.E. Knapp, J.D. McIver, K. Metlen, C.N. Skinner, and A. Youngblood. 2009.
Fire treatment effects on vegetation structure, fuels, and potential fire severity in western
U.S. forests. Ecological Applications 19: 305-320.
Stephens, R.E. Martin, and N.E. Clinton. 2007. Prehistoric fire area and emissions from
California forests, woodlands, shrublands, and grasslands. Forest Ecology and
Management 251: 205-216.
Tewksbury, J.J., S.J. Hejl, and T.E. Martin. 1998. Fragmentation in a western landscape:
forest fragmentation does not reduce nesting success. Ecology 79: 2890-2903.
Thompson, F.R., W.D. Dijak, T.G. Kulowiec, and D.A. Hamilton. 1992. Breeding bird
populations in Missouri Ozark forests with and without clearcutting. The Journal of
Wildlife Management 56: 23-30.
Van Horne, B. 1983. Density as a misleading estimate of habitat quality. The Journal of
Wildlife Management 47: 893-901.
Verner, J. and T.A. Larson. 1989. Richness of breeding birds species in mixed-conifer
forests of the Sierra Nevada, California. The Auk 106: 447-463.
Vitz, A.C. and A.D. Rodewald. 2006. Can regenerating clearcuts benefit mature-forest
songbirds? An examination of post-breeding ecology. Biological Conservation 127:
477-486.
Wayman R.B. and M. North. 2007. Initial response of a mixed-conifer understory plant
community to burning and thinning restoration treatments. Forest Ecology and
Management 239: 32-44.
Williams, Janet Mci. 1996. Nashville Warbler (Vermivora ruficapilla), The Birds of
North America Online (A. Poole, Ed.). Ithaca: Cornell Lab of Ornithology; Retrieved
from the Birds of North America Online: http://bna.birds.cornell.edu/bna/species/205
132
Ch. 1. Fuel Treatments
PRBO Avian Monitoring in the Plumas & Lassen National Forests - 2008
Wilson, M..F. and T.A. Comet. 1996. Bird communities of northern forests: ecological
correlates of diversity and abundance in the understory. The Condor 98: 350-362.
133
PRBO Avian Monitoring in the Plumas & Lassen National Forests - 2008
Chapter 2. Resident and Neotropical Migratory Bird Response to
Aspen Enhancement on the Lassen National Forest
Ryan D. Burnett
PRBO Conservation Science
134
Chapter 2. Aspen Enhancement
PRBO Avian Monitoring in the Plumas & Lassen National Forests - 2008
Background and Introduction
Declines in numerous songbird populations throughout North America have been
well documented, particularly among Neotropical migrants – those species that breed in
the U.S. and Canada and migrate to Mexico or Central or South America (see Finch and
Stangel 1993). The Lassen area supports populations of many of these declining and
threatened species, including Warbling Vireo, Swainson’s Thrush, Willow Flycatcher,
Olive-sided Flycatcher, and Yellow Warbler. The area is home to 9 of the 14 Riparian
Focal Species and at least 12 of the 13 Coniferous Forest Focal Species listed by
California Partners in Flight (RHJV 2004, CalPIF 2002), as well as all of the species of
landbirds identified as declining or likely declining by the Sierra Nevada Ecosystem
Project Report (SNEP 1996).
The composition and structure of western North American forests have been
altered by fire-suppression, timber harvesting, grazing, and other forest management
policies (see Hejl 1994, SNEP 1996, and Siegel & DeSante 1999). Human mediated
shifts in the competitive balance of these vast and complex systems can result in
permanent loss of habitat types or conditions if steps are not taken to mitigate these
impacts.
In the Sierra Nevada, with extensive livestock grazing and the absence of regular
fire, aspen are often out-competed by conifers (Mueggler 1985). As a result, the health of
aspen has deteriorated and its extent throughout western North America has been reduced
by at least 50 and up to 96% (Bartos and Campbell 2001). In 2000, the Eagle Lake
Ranger District (ELRD) of the Lassen National Forest (LNF) began an aspen habitat
inventory and risk assessment project. This effort documented that nearly 80% of all of
the remaining stands had a high or highest risk rating, indicating that without immediate
action the future of aspen in the district was endangered. Henceforth, they began a
district-wide strategy to enhance and save aspen habitat by implementing conifer removal
and erecting grazing exclosures at all remaining stands (Jones et al. 2005). While the
study of birds in aspen habitat in the Sierra Nevada has only recently been a focus of
ornithological research, evidence from point count data from the Almanor Ranger District
of the LNF (Burnett and Humple 2003), the Mono Basin (Heath and Ballard 2003), and
the Lake Tahoe Basin (Richardson and Heath 2005), show that aspen habitat supports an
extremely rich and abundant avian community that includes several species of
conservation concern, such as Warbling Vireo and Red-breasted Sapsucker (Gardali et al.
2000, Rich et al. 2004).
The avian community in the Lassen National Forest occupies a diverse range of
niches with its members associated with a broad range of habitat types and features
(Siegel and DeSante 1999, Burnett and Humple 2003). Birds are relatively high on the
food chain and have been shown to be sensitive to environmental change. Using one
inexpensive standardized method, it is possible to acquire data on an entire community of
organisms. Thus, birds are an ideal candidate for use as ecosystem indicators as bird
monitoring can provide the necessary feedback at the appropriate breadth and scale
(Temple and Wiens 1989, Hutto 1998) to be a valuable tool to land managers.
In 2004, PRBO began monitoring bird response to aspen treatments on the Eagle
Lake Ranger District of the Lassen National Forest. With the recent attention the Forest
Service has place on monitoring and adaptive management (SNFPA 2004), this project
135
Chapter 2. Aspen Enhancement
PRBO Avian Monitoring in the Plumas & Lassen National Forests - 2008
will provide the necessary data to evaluate the efficacy of aspen treatments and provide
feedback to support and/or improve future aspen projects in the ELRD and throughout
western North America.
Project Area
All avian survey work was conducted on the Lassen National Forest in the Eagle
Lake and Almanor Ranger Districts at the junction of the Sierra Nevada and Cascade
Mountains of California (Lat 400 N, Long 1200 W). Sites ranged in elevation from
approximately 1500 – 2000 meters (Figure 1).
Methods
Aspen Sampling Design
For all aspen sites we used GIS layers containing polygons of known aspen stands
based upon aspen inventories conducted by Forest Service staff.
In the Eagle Lake Ranger District we selected sites non-randomly that represent
the range of conditions in which aspen are found throughout the District. We limited our
selection to areas that could be covered by one observer in a four hour morning count
window and that contained enough acres of aspen habitat to fit a minimum of 4 point
count stations with at least 220 meter spacing between points. We attempted to
maximize the number of post-treatment sites, which were limited in number, as they
could provide us with information on bird response to aspen treatments that were already
five to nine years old.
In the Almanor Ranger District we selected sites that were within proposed aspen
enhancement projects (e.g., Minnow, Creeks II, Brown’s Ravine, and Feather), and one
additional site that has been proposed for treatment (Robber’s Creek).
On both districts we attempted to maximize the number of points within the
delineated aspen stands in the areas selected. In some areas where stands were not in
high densities, we limited transect size to allow for the extra time to walk between stands
in order to allow for completion within the limited morning hours allowed by the
standardized protocol. Generally, the first stand chosen was the one closest to the nearest
road. Once the first stand was chosen, the next closest stand that was at least 200 meters
from the previous was selected, and so on. All sites were selected without previous
knowledge of the local micro habitat attributes.
Survey Protocol
Standardized five minute fixed radius multiple distance band point count censuses
(Ralph et al. 1993, Buckland et al. 1993) were conducted at 181 stations along 18
transects in 2008 (Table 1, Figure 1, and Appendix 1). Detections were placed within
one of six categories based on the initial detection distance from observer: less than 10
meters, 10-20 meters, 20-30 meters, 30-50 meters, 50-100 meters, and greater than 100
meters. Birds flying over the study area but not observed landing were recorded
separately. The method of initial detection (song, visual or call) for each individual was
136
Chapter 2. Aspen Enhancement
PRBO Avian Monitoring in the Plumas & Lassen National Forests - 2008
recorded. Counts began around local sunrise and were completed within four hours. All
birds detected at each station during the five-minute survey were recorded. Each transect
was surveyed twice between 15 May and 1 July in each year, including 2008 (Table 1).
An electronic range finder was used to assist with distance estimation at each point count
station.
Table 1. Aspen point count transects, ranger district, number of stations, and dates surveyed in 2008.
Site
Brown’s Ravine Aspen
Coon Hollow Aspen
Philbrook Aspen
Robber’s Creek Aspen
West Dusty Aspen 1
West Dusty Aspen 2
West Dusty Aspen 3
West Dusty Aspen 4
Willow Creek Aspen
Butte Creek Aspen
Crazy Harry Aspen
Feather Lake Aspen
Harvey Valley Aspen
Lower Pine Creek Aspen
Martin Creek Aspen
Pine Creek Aspen
Ruffa Aspen
Susan River Aspen
# of Stations
4
14
10
16
10
6
8
8
9
8
7
5
15
12
11
14
12
12
Ranger District
Almanor
Almanor
Almanor
Almanor
Almanor
Almanor
Almanor
Almanor
Almanor
Eagle Lake
Eagle Lake
Eagle Lake
Eagle Lake
Eagle Lake
Eagle Lake
Eagle Lake
Almanor
Eagle Lake
Date, 1st Survey
5/30/2008
6/5/2008
6/5/2008
5/28/2008
5/26/2008
5/25/2008
5/26/2008
6/03/2008
6/03/2008
6/06/2008
5/29/2008
5/29/2008
5/28/2008
6/6/2008
5/30/2008
5/28/2008
6/8/2008
6/5/2008
Date, 2nd Survey
6/23/2008
NS
6/23/2008
6/16/2008
6/12/2008
6/27/2008
6/26/2008
6/27/2008
6/21/2008
6/27/2008
6/21/2008
6/21/2008
6/20/2008
6/20/2008
6/18/2008
6/20/2008
6/23/2008
6/27/2008
Analyses
Avian community point count analysis was restricted to a subset of the species
encountered. We excluded species that do not breed in the study area as well as those
that are not adequately sampled using the point count method (e.g., waterfowl, kingfisher,
and raptors). We also excluded European Starling and Brown-headed Cowbird from
analysis of species richness and total bird abundance because they are invasive species
regarded as having a negative influence on the native bird community. However, we did
investigate the abundance of these two species separately.
Species richness
We define species richness is the average number of species detected within 50
meters per point across visits within a year of species adequately sampled using the point
count method.
137
Chapter 2. Aspen Enhancement
PRBO Avian Monitoring in the Plumas & Lassen National Forests - 2008
Figure 1. Location of PRBO Aspen point count stations in the Lassen National Forest surveyed in 2007.
138
Chapter 2. Aspen Enhancement
PRBO Avian Monitoring in the Plumas & Lassen National Forests - 2008
Total Bird Abundance
The index of total bird abundance is the mean number of individuals detected per
station per visit. This number is obtained by dividing the total number of detections
within 50 meters by the number of stations and the number of visits.
Relative Abundance of Species
The relative abundance of species is the total detections of a given species per
point summed across the two visits within a year. We used total detections instead of
detections per visit to allow for use of negative binomial regression – which requires raw
count data – to compare differences. For analysis that compare multiple years we
summed the total detections across years and divided by the number of years. Thus,
multiple-year analyses are directly comparable to those comparing single years.
Trends in Richness and Abundance
I investigated trends in species richness and total bird abundance at treated and
untreated aspen stands in the ELRD from 2004 – 2008. We included all sites surveyed on
the ELRD, and since treatment occurred at a number of sites during this four year period,
they may have been included in the untreated sample in one or more years and the treated
sample in later years.
Statistical Tests
I employed a suite of statistical tests in comparing treated aspen to untreated
aspen. Negative binomial regression was used to test for differences in indices of
abundance of individual species between treated and untreated aspen stands; while I used
linear regression to compare the community indices of species richness and total bird
abundance. The test statistic (F for linear & Likelihood Ratio for negative binomial) and
p-values are presented. For the analysis of trends I used linear regression with year as the
independent variable. To test the significance between the treated and untreated trends I
used a likelihood ratio test to compare linear regression models with and without a year x
treatment interaction. The likelihood ratio χ2 statistic and p-value from these tests are
presented. For all tests significance was assumed at an α = 0.05 level. Stata statistical
software was used to conduct all statistical analysis (Stata Corp 2008).
Results
In 2008, total bird abundance ranged from a high of 8.00 at Feather Lake to a low
of 1.29 at Crazy Harry, and species richness ranged from 8.42 at Ruffa Ranch to 2.25 at
Susan River (Table 2). The average total bird abundance by transect in 2008 was 4.42
while species richness was 6.08.
We compared the total bird abundance and species richness at untreated aspen
sites in the ARD to untreated aspen sites in the ELRD in 2008. Species richness was 6.07
139
Chapter 2. Aspen Enhancement
PRBO Avian Monitoring in the Plumas & Lassen National Forests - 2008
in the ARD and 5.54 in the ELRD. Total bird abundance in the ARD was 4.14 compared
to 3.49 in the ELRD (Figure 2); neither of these differences was statistically significant.
When sites in the ELRD that have been treated were included, ELRD mean per point
species richness increased to 6.32 while total bird abundance increased to 4.45.
Total bird abundance and species richness were higher at treated sites compared
to untreated sites in the ELRD between 2006 and 2008 (Figure 3). Across this three year
period, total bird abundance averaged 5.98 at treated sites and 4.50 at untreated sites
(F=29.40, p<0.01). Species richness at treated sites averaged 7.28 compared to 6.33 at
untreated sites (F=9.70; p<0.01).
Table 2. Mean per point total bird abundance (detections/point/visit) and species richness (within 50
m of observers) at aspen sites surveyed in the Lassen National Forest from 2004 – 2008. Sites not
surveyed are represented by double dashes. Coon Hollow and Philbrook transects were surveyed
only once in 2008 due to fire access restrictions, thus they were not included in 2008 figures.
Station
Ruffa Aspen
Brown’s Ravine
Butte Creek
Coon Hollow
Crazy Harry
Feather Lake
Harvey Valley
Lower Pine
Martin Creek
Philbrook Aspen
Pine Creek
Robber’s Creek
Susan River
West Dusty 1
West Dusty 2
West Dusty 3
West Dusty 4
Willow Creek
Total
Total Bird Abundance
2004
5.72
2.38
4.63
-4.50
4.60
3.47
4.00
3.78
-4.60
-3.67
-----4.16
2005
7.11
3.25
5.81
-4.00
7.40
3.03
2.67
4.18
-4.57
-3.13
-----4.67
2006
5.92
4.13
7.31
-5.43
5.30
5.93
4.04
3.91
-5.90
5.72
3.09
3.75
3.33
3.63
4.75
4.28
5.36
2007
6.88
3.75
5.69
4.75
3.64
9.50
4.17
4.67
6.32
3.65
5.04
5.78
4.92
4.30
3.67
3.81
5.25
5.44
5.32
Species Richness
2008
6.33
2.75
5.50
-3.57
8.00
2.43
3.96
5.86
-4.71
5.09
1.29
3.00
4.08
3.19
4.56
4.61
4.42
2004
7.56
2.75
5.75
-6.43
6.40
4.93
5.75
5.09
-5.93
-4.75
-----5.53
2005
7.33
5.25
8.00
-5.43
7.20
4.47
4.42
5.45
-6.43
-5.00
-----5.90
2006
7.50
6.25
9.63
-8.00
5.80
6.93
5.92
5.27
-7.21
7.63
4.50
5.5
4.00
5.50
6.75
5.33
6.68
2007
8.92
5.00
8.38
6.71
5.85
7.80
4.67
6.83
8.00
5.30
7.00
7.31
6.5
6.80
3.67
5.63
7.88
7.22
6.79
2008
8.42
4.25
7.75
5.71
7.80
3.47
6.17
8.36
-6.86
7.63
2.25
5.00
5.67
5.38
5.75
6.78
6.08
Species richness at treated sites has been increasing at a rate of 6.2% (f= 7.83,
p=0.01) per year between 2004 and 2008 while at untreated sites it was increasing at
4.6% (f=10.91, p=<0.01) per year (Figure 4). The rate of increase in treated stands was
not significantly greater than that in untreated stands (LR χ 2= 0.65, p=0.65). Total bird
abundance from 2004 through 2008 at treated sites was increasing at a rate of 5.2%
(f=3.27, p=0.07) per year while at untreated sites it was increasing at a rate of 3.2%
(f=3.57, p=0.06) per year (Figure 5). The difference in the trends between treated and
untreated was not significant (LR χ 2= 0.27, p=0.60).
140
Chapter 2. Aspen Enhancement
PRBO Avian Monitoring in the Plumas & Lassen National Forests - 2008
Figure 2. Mean per point species richness and total bird abundance (mean per visit) based on
detections within 50 meters of observers at untreated Aspen sites in the Almanor and Eagle Lake
Ranger Districts in 2008 with standard error bars (excluding Coon Hollow and Philbrook).
Almanor vs. Eagle Lake 2008
8
Almanor Untreated
7
Eagle Lake Untreated
Eagle Lake Total
6
5
4
3
2
1
0
Species Richnes s
Total Bird Abundance
Figure 3. Mean per point species richness and total bird abundance at treated and untreated aspen in
the Eagle Lake Ranger District from 2006 – 2008 compared to coniferous forest in the PlumasLassen study area from 2003 – 2006.
Treated vs. Untre ated Aspen
8 .00
Treated Aspen
# per point count station
7.00
Untreated As p en
Conifer Fores t
6.00
5.00
4.00
3.00
2.00
1.00
0.00
Species Richness
Total Bird Abundance
141
Chapter 2. Aspen Enhancement
PRBO Avian Monitoring in the Plumas & Lassen National Forests - 2008
We investigated an index of the abundance of ten of the twelve previously
identified aspen focal species (Burnett in press), at treated aspen, untreated aspen, and
conifer forest. We also included Mountain Chickadee, another potential focal species.
There were not adequate detections of Swainson’s Thrush and Olive-sided Flycatcher –
the remaining two focal species – to include them in the analysis.
Seven of the ten species were significantly more abundant in treated aspen than
untreated aspen each of these seven were also significantly more abundant in aspen of
any kind compared to coniferous forest in the region (Table 3, figure 6). Red-breasted
Sapsucker, Hairy Woodpecker, Warbling Vireo, Mountain Bluebird, Tree Swallow,
Mountain Chickadee and Chipping Sparrow were all significantly more abundant in
treated aspen than untreated aspen. Additionally, total bird abundance and species
richness were significantly greater in treated stands compared to untreated stands.
Western-Wood Pewee, showed a small non-significant difference between treated and
untreated aspen though it was far more abundant in either treated or untreated aspen than
conifer forest. Only two focal species, Dusky Flycatcher and MacGillivray’s Warbler,
remained more abundant in untreated than treated aspen. The difference was marginally
significant for Dusky Flycatcher and was not significant for MacGillivray’s Warbler.
Table 3. Species Richness, total bird abundance, and the detections per point count visit for
ten aspen focal species at treated and untreated aspen sites across the Lassen National Forest from
2006 -2008. P-value is from linear (species richness) or negative binomial regression (all other
metrics) comparing treated to untreated aspen. Means from conifer forest in the Plumas-Lassen
Administrative study area from 2003-2006 are also presented for comparison.
Species Richness
Total Bird Abundance
Red-breasted Sapsucker
Hairy Woodpecker
Western Wood-Pewee
Dusky Flycatcher
Warbling Vireo
Tree Swallow
Mountain Bluebird
Oregon Junco
Chipping Sparrow
MacGillivray's Warbler
Treated Aspen
7.28
5.98
0.24
0.20
0.18
0.13
0.59
0.48
0.19
0.55
0.21
0.10
Untreated Aspen
6.32
4.50
0.15
0.09
0.16
0.20
0.45
0.03
0.00
0.45
0.09
0.12
P
<0.01
<0.01
0.03
0.01
0.59
0.06
0.04
<0.01
<0.01
0.14
<0.01
0.38
Conifer Forest
5.47
4.08
0.03
0.03
0.02
0.26
0.09
0.01
0.00
0.36
0.01
0.11
142
Chapter 2. Aspen Enhancement
PRBO Avian Monitoring in the Plumas & Lassen National Forests - 2008
Figure 4. Mean per point species richness (with standard error) at treated and untreated aspen sites
from 2004 -2008 in the Lassen National Forest with fitted linear trend lines.
Species Richness
Treated Trend = 6.2%/year
Untreated Trend = 4.6%/year
9
8
# of Species/Point
7
6
5
4
3
Treated
Untreated
2
1
0
2004
2005
2006
2007
2008
Figure 5. Total bird abundance per point count visit (with standard error) by year at treated and
untreated aspen sites from 2004 -2008 in the Lassen National Forest with fitted linear trends.
Total Bird Abundance
Treated Trend = 5 .2%/year
Untreated Trend = 3.2%/year
8
# of Individuals/Point/ Visit
7
6
5
4
3
Treated
2
Untreated
1
0
2004
2005
2006
2007
2008
143
Chapter 2. Aspen Enhancement
PRBO Avian Monitoring in the Plumas & Lassen National Forests - 2008
Figure 6. Abundance per point count visit (with standard error) for the seven aspen focal species
with a significant difference in abundance (p<0.05) between treated and untreated aspen across the
Lassen National Forest from 2006-2008. Conifer habitat indices are shown for comparison using
data from the Plumas-Lassen Admin Study area from 2003 – 2006.
Treated vs. Untreated Aspen
Detections/Point Count Visit
0.80
Treated Aspen
0.70
Untreated Aspen
0.60
Conifer Forest
0.50
0.40
0.30
0.20
0.10
0.00
Redbreasted
Sapsucker
Hairy
Woodpecker
Warbling
Vireo
Tree
Sw allow
Mountain
Chickadee
Mountain
Bluebird
Chipping
Sparrow
We investigated the effect of time since treatment on total bird abundance and
species richness for all aspen sites on the Lassen National Forest while controlling for
year. When all treated and untreated sites are included (with those that have not been
treated coded as zero) there is a significant positive effect (F=18.5, p<0.01) of time since
treatment (Figure 7). When untreated sites were not included there was no effect of time
since treatment (F=1.68, p=0.19) on total bird abundance. For species richness the effect
of time since treatment was positive and significant when untreated sites were included
(F=14.26, p<0.01) but was not when they were excluded (F=1.74, p=0.19; Figure 8).
The time since aspen stands had been treated had a significant effect on the
abundance of six of the ten focal species (Figure 9). For Red-breasted Sapsucker and
Chipping Sparrow the effect was positive and the best fit was linear. For each of the
other five species the effect was more complex. For Hairy Woodpecker, Tree Swallow,
Mountain Bluebird, and Dusky Flycatcher, the best fit model was one with a quadratic
effect of treatment. For all of these except Dusky Flycatcher there was an increasing
trend peaking in the four to five year post treatment period followed by a significant
decrease after that. Dusky Flycatcher was the only species to show a negative effect of
time since treatment. It decreased in the years immediately following treatment but
showed an increase in abundance in the longest time since treatment interval.
144
Chapter 2. Aspen Enhancement
PRBO Avian Monitoring in the Plumas & Lassen National Forests - 2008
0
Detections/Point Count Visit
5
10
15
Figure 7. The effect of time since treatment on total bird abundance in aspen habitat on the Lassen
National Forest from 2004 – 2008. The black line is the predicted values including all sites that have
not been treated as zero years post treatment. The green line represents the predicted values if only
sites that have been treated are included. Within a year multiple identical values are only
represented with a single data point.
0
2
4
6
Years Post Treatment
8
10
145
Chapter 2. Aspen Enhancement
PRBO Avian Monitoring in the Plumas & Lassen National Forests - 2008
0
Species/Point Count Station
5
10
15
Figure 8. The effect of time since treatment on avian species richness in aspen habitat on the Lassen
National Forest from 2004 – 2008. The black line is the predicted values including all sites that have
not been treated as zero years post treatment. The green line represents the predicted values if only
sites that have been treated are included. Within a year multiple identical values are only
represented with a single data point.
0
2
4
6
Years Post Treatment
8
10
146
Chapter 2. Aspen Enhancement
PRBO Avian Monitoring in the Plumas & Lassen National Forests - 2008
Figure 9. The mean abundance per point count visit with standard error and predicted values for the
six focal species showing a significant effect of time since treatment from 2004 - 2008. Graphs show
time since treatment in intervals for illustrative purposes but regression was conducted with all data.
All aspen sites surveyed on the Lassen National Forest are included. All untreated sites were coded
as zero years post treatment.
Hairy Woodpecker
Detections/Point Count
Visit
Detections/Point Count
Visit
Red-breasted Sapsucker
0.4
0.3
0.2
0.1
0
0
1 to 2
3 to 4
5 to 6
0.6
0.5
0.4
0.3
0.2
0.1
0
>6
0
Years Post Treatment
Detections/Point Count
Visit
Detections/Point Count
Visit
1
0.5
0
3 to 4
5 to 6
0.4
0.3
0.2
0.1
0
>6
0
Detections/Point Count
Visit
Years Post Treatment
3 to 4
5 to 6
>6
Chipping Sparrow
0.3
0.25
0.2
0.15
0.1
0.05
0
1 to 2 3 to 4 5 to 6
1 to 2
Years Post Treatment
Dusky Flycatcher
Detections/Point Count
Visit
>6
0.5
Years Post Treatment
0
5 to 6
Mountain Bluebird
1.5
1 to 2
3 to 4
Years Post Treatment
Tree Sw allow
0
1 to 2
>6
0.5
0.4
0.3
0.2
0.1
0
0
1 to 2
3 to 4
5 to 6
>6
Years Post Treatment
Discussion
Aspen habitat on the Lassen National Forest harbors greater total bird abundance,
species richness, and abundance of almost all of the aspen focal species compared to
147
Chapter 2. Aspen Enhancement
PRBO Avian Monitoring in the Plumas & Lassen National Forests - 2008
conifer-dominated forest in the region. On average, aspen habitat on the Almanor and
Eagle Lake Ranger Districts have comparable avian community indices though across
both districts there is considerable site to site variation in these indices as well as in the
abundance of individual species. In general, community indices at all sites were lower in
2008 than in 2007 a pattern that was expected following record high avian indices across
habitats in the northern sierra in 2007 (PRBO data). The largest decrease from 2007 to
2008 was observed at Harvey Valley, a site where half of the stations we sampled had
been treated over the last winter. This decrease immediately following treatment appears
contrary to the majority of our results investigating the effects of aspen treatments on the
avian community. We did not include Harvey Valley in the following analyses of
treatment effects because it was only partially treated in 2008 and piles of logs were still
stacked within our sampling area.
Treated vs. Untreated
In the ELRD the short term response of the avian community to aspen treatments
has been decidedly positive. Over the five year period of monitoring bird populations in
aspen habitat on the ELRD, there have been significant increases in species richness and
a marginally significant increase in total bird abundance at both treated and untreated
aspen. We have not observed similar patterns in conifer forest during this period (Burnett
and Nur 2007). While the difference in the rate of increase between treated and untreated
aspen was not significant, the rate was greater in treated aspen for both species richness
and total bird abundance. It is not clear what would be leading to an increase in these
metrics at untreated aspen sites; however, we have several hypotheses. First, over this
time period we added new sites to our sample – particularly on the Almanor Ranger
District where untreated aspen sites have shown slightly higher levels of these two
indices. Second, recovery of habitat following the removal of grazing can result in
significant increases in the majority of aspen associated birds in the west (Earnst et al.
2006). Many of the aspen sites have seen a reduction or cessation of grazing over the last
five to 15 years which may be allowing for some improvement in aspen bird habitat.
Aspen treatments appear to be benefiting passerine species that are rare,
declining, or both. All of the seven focal species that were significantly more abundant in
treated aspen compared to untreated aspen were all also significantly more abundant in
treated aspen than conifer forest. Chipping Sparrow has shown a consistent increasing
trend as treated aspen sites mature. It has been significantly declining at a rate of 3.4%
per year from 1968-2007 in the Sierra Nevada (Sauer et al. 2008); however, they are
increasing significantly in treated aspen stands. This species often nests in understory
trees in areas with a substantial herbaceous layer where it forages on insects and seeds
(Middleton 1998). Thus, treated aspen stands appear to be ideal habitat for this species
that is very rare in conifer dominated forest in the region. Likewise, Mountain Bluebird
and Tree Swallow are all but absent from conifer forest and untreated aspen but are fairly
common to abundant (respectively) in treated aspen. Mountain Bluebird has been
declining over the past 40 years at a rate of 2.5% per year, though due to their rarity this
trend is not significant (Sauer et al. 2008).
Warbling Vireo, which from 2004-2005 was more abundant in untreated aspen,
continued to increase in treated aspen where it is now significantly more abundant than in
untreated aspen. Treated sites such as Butte Creek and Martin Creek with two to three
148
Chapter 2. Aspen Enhancement
PRBO Avian Monitoring in the Plumas & Lassen National Forests - 2008
meter tall aspen regeneration are being used by this species for foraging as well as
nesting. As total aspen cover was the best predictor of this species abundance (Burnett et
al. 2005), we should expect to see a continued increase in this species in treated stands as
they mature; a positive sign for a species that may be declining in at least part of its range
in the west (Gardali et al. 2000).
Aspen habitat often supports a diverse and abundant guild of cavity nesting
species, with many studies showing cavity nesters disproportionately select aspen trees
for nesting (Li and Martin 1991, Dobkin et al. 1995, Martin and Eadie 1999, Martin et al.
2004). While aspen often contain relatively high numbers of natural cavities, secondary
cavity nesting species have been found to nest predominantly in woodpecker created
holes in both live aspen and aspen snags (Li and Martin 1991, Dobkin et al. 1995, Martin
and Eadie 1999). At numerous treated aspen - including those at Feather Lake, Butte
Creek, Pine Creek, and Martin Creek - we confirmed active woodpecker nest cavities
within treated stands, and a myriad of previously excavated cavities. Removing
encroaching conifers from within and surrounding aspen stands, resulting in the
expansion of stands and increased density of large diameter aspen stems over time,
should increase habitat for woodpeckers. There is little doubt that aspen supports far
greater abundance of woodpeckers than coniferous forest and that treating aspen results
in even greater increases in these species of management interest. In turn, woodpeckers
are a critical component of the aspen community as the source of cavities for an abundant
and diverse group of secondary cavity nesting birds, many of which use these aspen areas
in relatively high numbers (e.g., Mountain Bluebird, Tree Swallow, and Mountain
Chickadee).
Time Since Treatment
The time since aspen stands had been treated had a generally positive but complex
effect on many of the focal species. The best fit models for four of the six species
showing a significant effect of time since treatment included a quadratic term. For three
of these species their abundance peaked in the three to four years post-treatment time
period and then declined in the following time intervals. For Dusky Flycatcher, the only
species showing an overall negative effect of time since treatment, its abundance
decreased through the five to six year post treatment period and then showed a marked
increase at sites that were treated more than six years prior. For the remaining two
species the effect of time since treatment was best represented by a linear increase.
It is important to remember that that the post-treatment sample is relatively small
(28 sites in 2008) and any inherent biases in how sites were chosen for treatment could
easily be magnified. Furthermore, the sites that have been the longest time since
treatment were treated using a hand thin prescription that left a greater number of conifers
than the more recent prescriptions. Thus, this pattern may be at least partially a result of
the different prescriptions utilized in older compared to younger treatments. With those
cautions in mind, these results should not be entirely dismissed.
These patterns suggest that no one aspen condition or post-treatment time period
is ideal for all species, that the habitat conditions created in the first four years following
treatment are important for a number of bird species, and for several species the benefits
of aspen treatments may be rather short lived. The conditions created immediately
149
Chapter 2. Aspen Enhancement
PRBO Avian Monitoring in the Plumas & Lassen National Forests - 2008
following aspen treatments may be mimicking the structure found in natural postdisturbance habitat that often supports greater numbers of some of these species (Raphael
1987). Though Hairy Woodpecker, Tree Swallow, and Mountain Bluebird showed
marked declines at sites over four years post-treatment each was more abundant in these
older sites than they were in untreated aspen. These results continue to support the notion
that management of aspen habitat should consider the importance of disturbance and the
early successional habitat it results in.
Conclusions
Our results from 2008 continue to suggest that aspen treatments employed on the
ELRD are having a positive effect on the aspen breeding bird community. Key species
such as Red-breasted Sapsucker, Mountain Bluebird, and Chipping Sparrow all appear to
have had a short-term positive response to treatment. Based on these and previous results
we believe that treatments that increase the size and health of aspen stands will be highly
beneficial to key breeding bird species in the Lassen National Forest and should be a top
priority of land managers here. We also recognize the value of continuing the monitoring
of landbird communities in treated aspen habitat in order to better understand the
complex patterns we have started to see as time since treatment increases.
Literature Cited
Bartos, D.L. and R.B. Campbell, Jr. 2001. Landscape dynamics of aspen and conifer
forest. In Sustaining aspen in Western Landscapes: Symposium Proceedings. Grand
Junction, CO: Rocky Mountain Research Station. USDA Forest Service. RMRS -18:5-14.
Buckland, S. T., D. R. Anderson, K. P. Burnham, and J. L. Laake. 1993. Distance
sampling: estimating abundance of biological populations. Chapman and Hall, London.
Burnett, R.D. and G.R. Geupel. 2001. Songbird monitoring in the Lassen National Forest:
Results from the 2001 field season. PRBO report to the US Forest Service. Contribution
Number 1003.
Burnett, R. D., and D. L. Humple. 2003. Songbird monitoring in the Lassen National
Forest: Results from the 2002 field season with summaries of 6 years of data (19972002). A PRBO report to the U.S. Forest Service.
Burnett, R.D. and N. Nur 2007. Plumas-Lassen Area Study Module on Landbird
Abundance, Distribution, and Habitat Relationships. PRBO report to the US Forest
Service. Contribution Number 1550.
Burnett, R.D. In press. Integrating Avian Monitoring into Forest Management: Pine-Oak
and Aspen Enhancement on the Lassen National Forest. USFWS Technical Report.
CALPIF (California Partners in Flight). 2002. Version 1.0. The draft coniferous forest
bird conservation plan: a strategy for protecting and managing coniferous forest habitats
150
Chapter 2. Aspen Enhancement
PRBO Avian Monitoring in the Plumas & Lassen National Forests - 2008
and associated birds in California (J. Robinson and J. Alexander, lead authors). Point
Reyes Bird Observatory, Stinson Beach, CA. http://www.prbo.org/calpif/plans.html.
Dobkin, D. S., A. C. Rich, J. A. Pretare, and W. H. Pyle. 1995. Nest-site relationships
among cavity-nesting birds of riparian and snowpocket aspen woodlands in the
northwestern Great Basin. Condor 97:694-707.
Earnst, S.L., J.A. Ballard, and D.S. Dobkin. 2005. Riparian songbird abundance a decade
after cattle removal on Hart Mountain and Sheldon National Wildlife Refuges. PSWGTR 191:550-558.
Gardali, T., G. Ballard, N. Nur, and G. Geupel. 2000. Demography of a declining
population of Warbling Vireo. Condor 102:601-609.
Heath, S.K. and G. Ballard. 2004. Patterns of breeding songbird diversity and occurrence
in riparian habitats of the Eastern Sierra Nevada. In California Riparian Systems:
Processes and Floodplain Management, Ecology, and Restoration. 2001 Riparian
Habitats and Floodplains Conf. Proc. (P. M. Faber, ed.). Riparian Habitat Joint Venture,
Sacramento, CA.
Hejl, S. J. 1994. Human induced changes in bird populations in coniferous forests in
western North America during the past 100 years. Studies in Avian Biology 15:232-246.
Hutto, R.L. 1998. Using landbirds as an indicator species group. Pages 75-92 in J. M.
Marzluff and R. Sallabanks, editors. Avian conservation: research and management.
Island Press, Washington, D.C.
Jones, B.E., T.H. Rickman, A. Vasquez, Y. Sado, K.W. Tate. In press. Removal of
invasive conifers to regenerate degraded aspen stands in the Sierra Nevada. Restoration
Ecology 13:373-379.
Li, P., and T. E. Martin. 1991. Nest-site selection and nesting success of cavity-nesting
birds in high elevation forest drainages. Auk 108:405-418.
Martin, K. and J.M. Eadie. 1999. Nest webs: A community wide approach to the
management and conservation of cavity nesting forest birds. Forest Ecology and
Management 115: 243-257.
Martin, K., K. E. H. Aitken, and K. L. Wiebe. 2004, Nest-sites and nest webs for cavitynesting communities in interior British Columbia: nest characteristics and niche
partitioning: Condor. 106 5–19.
Mueggler, W.F. 1985. Forage. In Aspen: Ecology and management in the Western
United States. USDA Forest Service General Technical Report RM-119:129-134.
151
Chapter 2. Aspen Enhancement
PRBO Avian Monitoring in the Plumas & Lassen National Forests - 2008
Middleton, Alex L. 1998. Chipping Sparrow (Spizella passerina), The Birds of North
America Online (A. Poole, Ed.). Ithaca: Cornell Lab of Ornithology; Retrieved from the
Birds of North America Online: http://bna.birds.cornell.edu/bna/species/334
Ralph, C. J., G. R. Geupel, P. Pyle, T. E. Martin, & D. F. DeSante. 1993. Field Methods
for Monitoring Landbirds. USDA Forest Service Publication, PSW-GTR 144, Albany,
CA.
Raphael, M.G., Morrison, M.L., Yoder-Williams, M.P., 1987. Breeding bird populations
during twenty five years of post-fire succession in the Sierra Nevada. The Condor 89,
614-626.
Rich, T.D., C.J. Beardmore, H. Berlanga, P.J. Blancher, M.S.W. Bradstreet, G.S.
Butcher, D.W. Demarest, E.H. Dunn, C. Hunter, E.E. Inigo-Elias, J.A. Kennedy, A.M.
Martell, A.O. Panjabi, D.N. Pashley, K.V. Rosenberg, C.M. Rustay, J.S. Wendt, T.C.
Will. 2004. Partners in Flight North American Landbird Conservation Plan. Cornell Lab
of Ornithology, Ithaca, NY.
RHJV (Riparian Habitat Joint Venture). 2004. Version 2.0. The riparian bird
conservation plan: a strategy for reversing the decline of riparian associated birds in
California. California Partners in Flight. http://www.prbo.org/calpif/pdfs/riparian.v2.pdf.
Richardson, T.W. and S.K. Heath. 2005. Effects of conifers on aspen breeding bird
communities in the Sierra Nevada. Transactions of the Western Section of the Wildlife
Society 40: 68 – 81.
Sauer, J. R., J. E. Hines, and J. Fallon. 2008. The North American Breeding Bird Survey,
Results and Analysis 1966 - 2007. Version 10.13.2007. USGS Patuxent Wildlife
Research Center, Laurel, MD.
Siegel, R.B. and D.F. DeSante. 1999. Version 1.0. The draft avian conservation plan for
the Sierra Nevada Bioregion: conservation priorities and strategies for safeguarding
Sierra bird populations.
SNEP (Sierra Nevada Ecosystem Project) 1996. Sierra Nevada Ecosystems. Volume 1,
Chapter 1. Regents of the University of California.
http://ceres.ca.gov/snep/pubs/web/PDF/v1_ch01.pdf
SNFPA (Sierra Nevada Forest Plan Amendment). 2004. Final Supplemental
Environmental Impact Statement and Record of Decision.
http://www.fs.fed.us/r5/snfpa/final-seis/
Stata Corp. 2005. Intercooled Stata 8.2 for Windows. Stata Corp. LP College Station,
TX.
Temple, S. A. and J. A. Wiens. 1989. Bird populations and environmental changes: can
152
Chapter 2. Aspen Enhancement
PRBO Avian Monitoring in the Plumas & Lassen National Forests - 2008
birds be bio-indicators? American Birds 43:260-270
153
Chapter 2. Aspen Enhancement
PRBO Avian Monitoring in the Plumas & Lassen National Forests - 2008
Appendix 1. GPS coordinates (UTM NAD 27) for all aspen point count locations
surveyed in the Lassen National Forest in 2007 & 2008.
STATION
Ruffa Aspen
Ruffa Aspen
Ruffa Aspen
Ruffa Aspen
Ruffa Aspen
Ruffa Aspen
Ruffa Aspen
Ruffa Aspen
Ruffa Aspen
Ruffa Aspen
Ruffa Aspen
Ruffa Aspen
Butte Creek Aspen
Butte Creek Aspen
Butte Creek Aspen
Butte Creek Aspen
Butte Creek Aspen
Butte Creek Aspen
Butte Creek Aspen
Butte Creek Aspen
Brown’s Ravine Aspen
Brown’s Ravine Aspen
Brown’s Ravine Aspen
Brown’s Ravine Aspen
Coon Hollow Aspen
Coon Hollow Aspen
Coon Hollow Aspen
Coon Hollow Aspen
Coon Hollow Aspen
Coon Hollow Aspen
Coon Hollow Aspen
Coon Hollow Aspen
Coon Hollow Aspen
Coon Hollow Aspen
Coon Hollow Aspen
Coon Hollow Aspen
Coon Hollow Aspen
Coon Hollow Aspen
Crazy Harry Aspen
Crazy Harry Aspen
Crazy Harry Aspen
Crazy Harry Aspen
Crazy Harry Aspen
Crazy Harry Aspen
Crazy Harry Aspen
Feather Lake Aspen
Feather Lake Aspen
CODE
ASPN
ASPN
ASPN
ASPN
ASPN
ASPN
ASPN
ASPN
ASPN
ASPN
ASPN
ASPN
BCA
BCA
BCA
BCA
BCA
BCA
BCA
BCA
BRAS
BRAS
BRAS
BRAS
COHO
COHO
COHO
COHO
COHO
COHO
COHO
COHO
COHO
COHO
COHO
COHO
COHO
COHO
CHA
CHA
CHA
CHA
CHA
CHA
CHA
FLA
FLA
SITE
1
2
3
4
5
6
7
8
9
10
11
12
1
2
3
4
5
6
7
8
1
2
3
4
1
2
3
4
5
6
7
8
9
10
11
12
13
14
1
2
3
4
5
6
7
1
2
X_COORDINATE
634087
633993
633909
633842
633746
633746
635118
635203
635411
634306
634612
634683
644638
644550
644760
644952
645027
645194
645272
645346
628386
628624
627589
628428
634428
634323
634065
633839
633622
633412
633149
632946
632771
632562
632367
632123
631951
631864
682820
682688
682703
681773
681857
682098
682189
667437
667620
Y_COORDINATE
4447622
4447459
4447283
4447102
4446885
4447193
4447923
4447725
4447925
4447661
4447680
4447371
4498553
4498065
4495527
4495285
4495074
4494831
4494654
4494398
4432142
4432262
4433429
4432429
4433745
4433988
4434058
4434016
4434061
4434120
4434111
4434228
4434429
4434400
4434376
4434511
4434663
4434894
4475480
4475240
4474972
4473900
4473575
4473532
4473220
4488993
4488996
154
Chapter 2. Aspen Enhancement
Feather Lake Aspen
Feather Lake Aspen
Feather Lake Aspen
Harvey Valley Aspen
Harvey Valley Aspen
Harvey Valley Aspen
Harvey Valley Aspen
Harvey Valley Aspen
Harvey Valley Aspen
Harvey Valley Aspen
Harvey Valley Aspen
Harvey Valley Aspen
Harvey Valley Aspen
Harvey Valley Aspen
Harvey Valley Aspen
Harvey Valley Aspen
Harvey Valley Aspen
Harvey Valley Aspen
Lower Pine Creek Aspen
Lower Pine Creek Aspen
Lower Pine Creek Aspen
Lower Pine Creek Aspen
Lower Pine Creek Aspen
Lower Pine Creek Aspen
Lower Pine Creek Aspen
Lower Pine Creek Aspen
Lower Pine Creek Aspen
Lower Pine Creek Aspen
Lower Pine Creek Aspen
Lower Pine Creek Aspen
Martin Creek Aspen
Martin Creek Aspen
Martin Creek Aspen
Martin Creek Aspen
Martin Creek Aspen
Martin Creek Aspen
Martin Creek Aspen
Martin Creek Aspen
Martin Creek Aspen
Martin Creek Aspen
Martin Creek Aspen
Philbrook Aspen
Philbrook Aspen
Philbrook Aspen
Philbrook Aspen
Philbrook Aspen
Philbrook Aspen
Philbrook Aspen
Philbrook Aspen
Philbrook Aspen
PRBO Avian Monitoring in the Plumas & Lassen National Forests - 2008
FLA
FLA
FLA
HVA
HVA
HVA
HVA
HVA
HVA
HVA
HVA
HVA
HVA
HVA
HVA
HVA
HVA
HVA
LPA
LPA
LPA
LPA
LPA
LPA
LPA
LPA
LPA
LPA
LPA
LPA
MCA
MCA
MCA
MCA
MCA
MCA
MCA
MCA
MCA
MCA
MCA
PHAS
PHAS
PHAS
PHAS
PHAS
PHAS
PHAS
PHAS
PHAS
3
4
5
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
1
2
3
4
5
6
7
8
9
10
11
12
1
2
3
4
5
6
7
8
9
10
11
1
2
3
4
5
6
7
8
9
667803
667477
668080
663482
663608
663820
664353
664447
665382
666678
666994
667246
667540
667974
669088
668861
668631
668785
660456
660334
660216
657955
658237
658449
658711
658995
659287
659286
659595
659793
672919
673274
673697
673905
674067
673832
671981
672235
673517
672833
672888
627678
627656
627857
628732
628544
632371
632352
631700
634153
4489035
4488439
4488016
4502834
4502617
4502901
4503212
4503537
4503145
4504026
4504055
4503973
4503942
4503901
4502928
4503100
4503130
4502703
4490845
4491146
4490936
4489672
4489822
4489995
4490186
4490395
4490252
4490494
4490602
4490770
4494467
4494078
4493728
4493440
4493319
4493247
4494288
4494142
4492496
4493680
4494725
4432335
4432684
4432784
4432132
4431953
4431587
4431246
4431187
4431618
155
Chapter 2. Aspen Enhancement
Philbrook Aspen
Pine Creek Aspen
Pine Creek Aspen
Pine Creek Aspen
Pine Creek Aspen
Pine Creek Aspen
Pine Creek Aspen
Pine Creek Aspen
Pine Creek Aspen
Pine Creek Aspen
Pine Creek Aspen
Pine Creek Aspen
Pine Creek Aspen
Pine Creek Aspen
Pine Creek Aspen
Robber’s Creek Aspen
Robber’s Creek Aspen
Robber’s Creek Aspen
Robber’s Creek Aspen
Robber’s Creek Aspen
Robber’s Creek Aspen
Robber’s Creek Aspen
Robber’s Creek Aspen
Robber’s Creek Aspen
Robber’s Creek Aspen
Robber’s Creek Aspen
Robber’s Creek Aspen
Robber’s Creek Aspen
Robber’s Creek Aspen
Robber’s Creek Aspen
Robber’s Creek Aspen
Susan River Aspen
Susan River Aspen
Susan River Aspen
Susan River Aspen
Susan River Aspen
Susan River Aspen
Susan River Aspen
Susan River Aspen
Susan River Aspen
Susan River Aspen
Susan River Aspen
Susan River Aspen
West Dusty Aspen 1
West Dusty Aspen 1
West Dusty Aspen 1
West Dusty Aspen 1
West Dusty Aspen 1
West Dusty Aspen 1
West Dusty Aspen 1
PRBO Avian Monitoring in the Plumas & Lassen National Forests - 2008
PHAS
PCA
PCA
PCA
PCA
PCA
PCA
PCA
PCA
PCA
PCA
PCA
PCA
PCA
PCA
ROCA
ROCA
ROCA
ROCA
ROCA
ROCA
ROCA
ROCA
ROCA
ROCA
ROCA
ROCA
ROCA
ROCA
ROCA
ROCA
SRA
SRA
SRA
SRA
SRA
SRA
SRA
SRA
SRA
SRA
SRA
SRA
WDA1
WDA1
WDA1
WDA1
WDA1
WDA1
WDA1
10
1
2
3
4
5
6
7
8
9
10
11
12
13
14
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
1
2
3
4
5
6
7
8
9
10
11
12
1
2
3
4
5
6
7
634369
660374
660524
660297
660175
659873
660075
660132
659993
660365
660627
660746
660931
660698
660328
669942
669793
669593
669486
669344
665405
665306
665115
663507
663373
663310
663106
663091
663513
663540
663579
677245
675682
675445
675110
674827
674932
674883
674697
675795
676097
676339
676609
634004
633923
634639
634539
634497
634387
634873
4431371
4492311
4492546
4492538
4492348
4492702
4492809
4493134
4493476
4493446
4493377
4493133
4493315
4493566
4492835
4468779
4468956
4468975
4469442
4469591
4475553
4475774
4475967
4478021
4478266
4478598
4478822
4479042
4478985
4478747
4478488
4477578
4477640
4477816
4477746
4478047
4478384
4478663
4478626
4477426
4477220
4477123
4477077
4469806
4469600
4469394
4468874
4468542
4468347
4468129
156
Chapter 2. Aspen Enhancement
West Dusty Aspen 1
West Dusty Aspen 1
West Dusty Aspen 1
West Dusty Aspen 2
West Dusty Aspen 2
West Dusty Aspen 2
West Dusty Aspen 2
West Dusty Aspen 2
West Dusty Aspen 2
West Dusty Aspen 3
West Dusty Aspen 3
West Dusty Aspen 3
West Dusty Aspen 3
West Dusty Aspen 3
West Dusty Aspen 3
West Dusty Aspen 3
West Dusty Aspen 3
West Dusty Aspen 4
West Dusty Aspen 4
West Dusty Aspen 4
West Dusty Aspen 4
West Dusty Aspen 4
West Dusty Aspen 4
West Dusty Aspen 4
West Dusty Aspen 4
Willow Creek Aspen
Willow Creek Aspen
Willow Creek Aspen
Willow Creek Aspen
Willow Creek Aspen
Willow Creek Aspen
Willow Creek Aspen
Willow Creek Aspen
Willow Creek Aspen
PRBO Avian Monitoring in the Plumas & Lassen National Forests - 2008
WDA1
WDA1
WDA1
WDA2
WDA2
WDA2
WDA2
WDA2
WDA2
WDA3
WDA3
WDA3
WDA3
WDA3
WDA3
WDA3
WDA3
WDA4
WDA4
WDA4
WDA4
WDA4
WDA4
WDA4
WDA4
WICA
WICA
WICA
WICA
WICA
WICA
WICA
WICA
WICA
8
9
10
1
2
3
4
5
6
1
2
3
4
5
6
7
8
1
2
3
4
5
6
7
8
1
2
3
4
5
6
7
8
9
635297
635469
636174
639420
639502
639619
640654
640951
641089
636449
637197
636961
637049
637181
637412
636864
636248
630461
630615
630501
630663
630154
629921
629708
629797
640030
640219
640837
641354
641541
641956
641999
642215
643562
4468584
4468617
4468629
4469076
4468483
4468179
4467742
4467632
4467671
4469388
4468745
4468828
4468527
4468351
4468346
4468309
4468425
4468307
4468421
4468560
4468939
4468780
4468724
4468657
4468887
4473252
4473149
4472266
4470754
4470368
4470077
4469674
4469538
4468519
157
PRBO Avian Monitoring in the Lassen and Plumas National Forests - 2008
Chapter 3. Pileated Woodpecker Monitoring on the Lassen
National Forest
Dennis Jongsomjit, Ryan D. Burnett, and Diana Stralberg
PRBO Conservation Science
158
Ch.3 Pileated Woodpecker
PRBO Avian Monitoring in the Lassen and Plumas National Forests - 2008
Background and Introduction
The National Forest Management Act (NFMA) of 1976 was created to help guide
management of National Forest lands in the United States. In 1982 planning regulations
were adopted that guided the establishment of Management Indicator Species (MIS)
under NFMA. The MIS approach was adopted in order to use a suite of species that can
elucidate the most appropriate management approaches by guiding resource management
plan revisions and forest plan project implementation. As part of this process the Lassen
National Forest identified Pileated Woodpecker (among other species) as a MIS (LRMP
1992).
Pileated Woodpecker is the largest extant woodpecker in United States (Bull and
Jackson 1995). While its distribution includes a variety of forested habitats across the
eastern United States, in the west it is associated almost exclusively with mid- to late
seral conifer-dominated forests (Mellen et al. 1992, Bull and Holthausen 1993). Its home
range size is large and extremely variable compared to other North American
woodpeckers, with a reported range from 53 to 1,056 hectares (Bull and Jackson 1995).
In the Western U.S., studies in Oregon found average home range sizes between 407- 478
hectares (Bull and Holthausen 1993, Mellen et al. 1992).
Due to their retiring nature, habitat specialization, and large territory sizes,
standard bird monitoring techniques such as point counts (Ralph et al. 1995) are unlikely
to detect sufficient numbers of this species for meaningful analysis of population trends.
In 2007 PRBO began a comprehensive forest wide monitoring program for Pileated
Woodpecker with four primary objectives:
1. Determine its spatial distribution across the forest
2. Provide baseline data for determining long-term trends
3. Identify key habitat features
4. Develop an appropriate monitoring protocol for the species in the Sierra
Nevada.
In order to adequately sample this species, we developed a GIS-based predictive
model of suitable habitat in the Lassen National Forest (LNF) using existing point-count
data. In 2007 we used this model to identify survey locations and conducted standard
point counts followed by call playbacks if no individuals were detected. Compared with
previous Plumas-Lassen random surveys, the 2007 surveys had significantly more
detections of Pileated Woodpeckers when looking at unlimited-distance point counts
before playbacks (Burnett et al. 2008), suggesting that targeting survey areas based on
distribution model predictions could be an effective way to locate suitable habitat. When
no individuals were detected, call playbacks resulted in a 37% increase in detections
above point counts alone, suggesting this method could be an effective way to increase
detections. However, some findings were inconclusive: Comparing point count
detection rates prior to call playbacks at distances of 100 m and 50 m, the Plumas-Lassen
random sampling actually had higher detection rates than the 2007 MIS surveys. Also,
several survey locations failed to detect any Pileated Woodpeckers despite being located
within very suitable habitat as identified by the model. Finally, we found that detection
rates during our playback surveys may have been inflated as a result of individual birds
159
Ch.3 Pileated Woodpecker
PRBO Avian Monitoring in the Lassen and Plumas National Forests - 2008
following observers and thus being detected at two or more consecutive survey points.
Thus there was room for improvement of both the distribution modeling and survey
techniques.
Evaluating the performance of a model is crucial to assessing a model’s
usefulness and effectiveness, especially when tested within a real-world application
(Pearce and Ferrier 2000, Vaughan and Ormerod 2005). So in conjunction with a new
model and survey method we created an entirely new set of transects designed to provide
a sufficient sample size to assess model performance while representing the anticipated
use of the model. We discuss the implications of our results on the use of distributional
modeling as a tool to focus monitoring efforts and inform management decisions.
Methods
Predictive Model
In order to maximize detections of Pileated Woodpeckers we developed a new
model to predict areas most likely to support this species prior to selecting sites to
monitor in 2008 (Figure 1). We used a powerful machine learning algorithm called
Maxent (Phillips et al. 2006) to predict Pileated Woodpecker distributions based on
presence and absence data from PRBO’s Northern Sierra point count survey database.
Maxent is based on the principle of maximum entropy, and uses information about a
known set of species occurrence points, compared with environmental background data,
to develop parsimonious models of species occurrence.
Although Maxent is typically used with presence-only data, we incorporated
species absence data in place of random environmental background data to constrain the
models to the environmental space that was sampled. Including absence data can serve to
improve the model by providing information on where the species does not occur
(Brotons et al. 2004).
In 2008 we developed a set of bioclimatic variables (Hijmans 2008) which are
thought to influence species distributions and are used widely in such modeling exercises.
These variables were derived from the monthly temperature and precipitation average
values used in our previous models. Using survey data collected from 1997-2007, we
updated our distribution model using these bioclimatic variables as well as a variety of
classified vegetation types from USDA Forest Service CALVEG vegetation shapefiles
that we converted to grids at a 50m resolution
(http://www.fs.fed.us/r5/rsl/clearinghouse/gettiles.shtml). Vegetation data were
manipulated to create local and several landscape scale variables of hypothesized
importance for Pileated Woodpecker (Table 1). By including these new bioclimatic
variables and results from the 2007 surveys, the 2008 model outperformed the 2007
model. Thus, we used this new model to inform our selection of survey sites in 2008.
Model Testing
Model performance was assessed using the area under the curve (AUC) of
receiver operating characteristic (ROC) plots (Fielding and Bell 1997). AUC values
represent the predictive ability of a distribution model and are derived from a plot of true
160
Ch.3 Pileated Woodpecker
PRBO Avian Monitoring in the Lassen and Plumas National Forests - 2008
positive against false positive fractions for a given model. Models using presence only
were tested using true positive against random background data points. Higher values
(up to 1.0) characterize higher accuracy models. An AUC value of 0.5 is the equivalent
of a random prediction. As a general guideline, AUC values of 0.6 – 0.7 indicate poor
accuracy, 0.7 – 0.8 is fair, 0.8 - 0.9 is good, and values greater than 0.9 represent
excellent accuracy (Swets 1988).
Table 1. GIS-based environmental predictors of species distribution. Habitat types, size classes, and
density classes are California Wildlife Habitat Relationship classifications (Mayer and Laudenslayer
1988).
Variable Name
Red fir
Sierran Mixed Conifer
White fir
Point Vegetation
Size class 4 and 5
High Density Forest
Bio_1
Bio_2
Bio_3
Bio_4
Bio_10
Bio_12
Bio_15
Bio_17
Description
Percent red fir habitat within a 1km radius
Percent mixed conifer forest within a 1km radius
Percent white fir vegetation within a 1km radius
Vegetation type at point count location
Vegetation size classes within a 1 km radius. Class 4 equals 11.0” –
23.9” DBH. Class 5 equals >24.0” DBH and was combined with
class 6 (large trees multi-storied)
Calculated within a 1km radius. Combines class M (40 – 59.9 %
canopy closure) and Class D (>60% canopy closure).
Annual mean temperature (˚C)
Mean diurnal range (˚C) (mean of monthly (max temp - min temp))
Isothermality ((Bio_2/Bio_7) * 100)
Temperature seasonality (standard deviation *100)
Mean temperature of warmest quarter (˚C)
Annual precipitation (mm * 10)
Precipitation seasonality (coefficient of variation)
Precipitation of driest quarter (mm * 10)
The model prediction was cross-validated using a subset of the data points (25%)
selected at random by the Maxent program. While this cross-validation technique is
considered to be a robust method of model testing, unbiased estimates of a model’s
performance are best attained when tested against data that are independent of the data
used to build the model (Fielding and Bell 1997). New surveys can provide an
independent dataset because they differ from the training data in several ways. These
include different collection methods, different geographic space, and different time
periods (Vaughan and Ormerod 2005). Testing with independent data can help identify
potential problems and provides a good assessment of a models ability to predict suitable
Pileated Woodpecker habitat into areas outside of those used to build the model (Araújo
et al. 2005). Testing a group of models allows us to identify more specifically how our
models can be improved.
161
Ch.3 Pileated Woodpecker
PRBO Avian Monitoring in the Lassen and Plumas National Forests - 2008
Figure 1. Map of model predicted probability of occurrence for Pileated Woodpecker. Presence and
absence points used to build the model are shown as is the boundary of the Lassen National Forest.
We followed a basic framework to create a set of models and tested their
performance using new survey data collected in 2008. We hypothesized that lower test
scores could be caused by a model overfitting the results due to a high number of
162
Ch.3 Pileated Woodpecker
PRBO Avian Monitoring in the Lassen and Plumas National Forests - 2008
variables. We tested this by creating two sets of models with fewer variables (vegetation
only and climate only). We also hypothesized that the absence data used to inform the
model could be unreliable due to the natural history of this species. Thus, each set of
variables was also used to run models using only presence data. This resulted in 5 total
models, in addition to the original model used to select survey locations in 2008.
Site Selection
Because one of our objectives was to locate areas with Pileated Woodpecker
within the LNF, we did not select transect starting points naively. We used spatially
explicit model output from previous point count and call back surveys for this species in
the LNF to select transect starting points (Burnett et al. 2008). We clipped the distribution
model output to the LNF boundary and filtered the data to show only those areas
considered to have a greater than 47% probability of occurrence for this species. This
percentage was chosen because it represented a threshold that maximized the models true
positive and true negative predictions based on our presence/absence data. A random
point generator was used to create 60 starting points within this filtered boundary. We
then randomly selected among starting points and established transects using a GIS road
layer and LNF atlas to establish 19 additional points. We only allowed a maximum of
eight transects in any of the three ranger districts to ensure adequate coverage of each
administrative unit (Table 2). Thus, once this limit was reached if a subsequent starting
point was selected that fell within that district we rejected it and moved on to the next
random point. Additionally, a few starting points initially selected were rejected if the
surrounding terrain was not appropriate (large tracts of private land or non forested
habitat surrounding a selected site) or if another transect had already been established
within 1km. When given a choice between two otherwise equal transect routes to follow,
we followed the one that lead further away from existing points, dead ends, and towards
areas that had higher model prediction values. All points were spaced approximately
500m apart, twice the normal distance between point counts to minimize the chances that
an woodpecker near one point could hear the playback at the next nearest point. All
transects were established on secondary unpaved roads (Figure 2).
163
Ch.3 Pileated Woodpecker
PRBO Avian Monitoring in the Lassen and Plumas National Forests - 2008
Table 2. Pileated Woodpecker playback survey transects, transect codes, ranger district and dates
surveys conducted on the Lassen National Forest in 2008.
Transect Name
Aspen Flat
Bogard Buttes
Box Canyon
Bunchgrass Valley
Crater Mountain Road
Elam Creek
Grey’s Flat
Humboldt Road
Jennie Creek
Jellico
Little Davis Creek
Little Grizzly
Logan Mountain Road
Mineral Summit
North Battle Creek
Pratville
Stump Ranch
Suicide Cabin
Tamarack Swale
Upper Yellow Creek
Transect Code
ASFL
BOBU
BOCA
BUVA
CRMR
ELCR
GRFL
HURO
JECR
JELL
LDCR
LIGR
LMRO
MISU
NBCR
PRAT
STRA
SUIC
TASW
UYCR
Ranger District
Eagle Lake
Eagle Lake
Hat Creek
Hat Creek
Eagle Lake
Almanor
Eagle Lake
Almanor
Almanor
Hat Creek
Hat Creek
Almanor
Eagle Lake
Almanor
Hat Creek
Almanor
Almanor
Hat Creek
Hat Creek
Almanor
1st Survey
6/11/2008
6/09/2008
6/10/2008
6/12/2008
6/06/2008
6/08/2008
6/07/2008
6/17/2008
6/10/2008
6/14/2008
6/13/2008
6/06/2008
6/08/2008
6/07/2008
6/16/2008
6/07/2008
6/07/2008
6/09/2008
6/11/2008
6/06/2008
2nd Survey
7/01/2008
6/26/2008
6/27/2008
6/30/2008
6/26/2008
6/26/2008
7/01/2008
6/27/2008
7/1/2008
7/02/2008
7/02/2008
6/30/2008
6/26/2008
6/26/2008
6/30/2008
6/30/2008
7/01/2008
Survey Protocol
Survey methods differed in 2008 in that call playbacks were conducted at each
survey point immediately upon arrival. This differed from 2007 surveys where call
playbacks were conducted only if a standard point count survey failed to detect any
individuals. This change was designed to more efficiently determine the presence of an
individual and allow for the greater number of points covered within each transect. We
used a digital audio recording of a series of Pileated Woodpecker calls and drumming
broadcast over a Radioshack® “Power Horn” blaster at full volume. Based on several
field tests, our call playbacks could be detected from between 150 and 250 meters,
depending on field conditions (e.g. slope, tree density), by our observers. The call
playback survey was three minutes long and consisted of three 30 second call playbacks
each followed by a 30 second listening period. The direction the blaster was directed was
rotated 120 degrees from the previous broadcast position for each subsequent playback. If
at any point during the survey a Pileated Woodpecker was detected we ceased the
playback, recorded the type of detection (drumming, visual, or call), compass bearing,
and distance from the observer, and moved on to the next survey location. Any
incidental Pileated Woodpecker observations that occurred between points were also
recorded but not included in any analysis. All transects were surveyed twice during the
breeding season with the exception of three that we were unable to survey a second time
due to wildfire related travel restrictions in late June through mid July (Table 2). Surveys
for each transect were conducted at least two weeks apart.
164
Ch.3 Pileated Woodpecker
PRBO Avian Monitoring in the Lassen and Plumas National Forests - 2008
Figure 2. The location of Pileated Woodpecker survey transects in 2008 in the Lassen National
Forest. Four letter transect codes are defined in Table 2.
165
Ch.3 Pileated Woodpecker
PRBO Avian Monitoring in the Lassen and Plumas National Forests - 2008
Results
Surveys
Pileated Woodpecker were detected on 19 of the 20 transects that were surveyed
in 2008 with 17 transects having detections at more than one point (Table 3). Jellico,
within the HCRD was the only transect that did not have any detections. However, we
could only survey this site once in the season due to the wildfire related travel
restrictions. They were detected at 124 out of the 400 survey points (30%). They were
detected within 100 meters of the observer on 7 of the 20 transects (35%).
Table 3. Pileated Woodpecker call back survey transects where the species was detected within the
Lassen National Forest in 2008.
Transect Name
Aspen Flat
Bogard Buttes
Box Canyon
Bunchgrass Valley
Crater Mountain Road
Elam Creek
Grey’s Flat
Humboldt Road
Jennie Creek
Jellico
Little Davis Creek
Little Grizzly
Logan Mountain Road
Mineral Summit
North Battle Creek
Pratville
Stump Ranch
Suicide Cabin
Tamarack Swale
Upper Yellow Creek
Transect Code
ASFL
BOBU
BOCA
BUVA
CRMR
ELCR
GRFL
HURO
JECR
JELL
LDCR
LIGR
LMRO
MISU
NBCR
PRAT
STRA
SUIC
TASW
UYCR
Ranger District
Eagle Lake
Eagle Lake
Hat Creek
Hat Creek
Eagle Lake
Almanor
Eagle Lake
Almanor
Almanor
Hat Creek
Hat Creek
Almanor
Eagle Lake
Almanor
Hat Creek
Almanor
Almanor
Hat Creek
Hat Creek
Almanor
Detected
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
Detected < 100
meters from
survey point
X
X
X
X
X
X
X
We compared detection rates for the 2008 MIS surveys, 2007 MIS surveys, and
2007-2008 Plumas-Lassen passive point-count surveys (Table 4). The detection rates
(detections per point per visit) in 2008 were slightly lower than for the 2007 MIS surveys,
though these difference were not statistically significant (p>0.10). Both MIS surveys (that
incorporated callbacks) had significantly higher detection rates than the Plumas-Lassen
point counts for all detections and <50m. Detections at <100 m were equal for the 2008
MIS and Plumas-Lassen point counts and higher for the 2007 MIS survey.
166
Ch.3 Pileated Woodpecker
PRBO Avian Monitoring in the Lassen and Plumas National Forests - 2008
Table 4. Pileated woodpecker detections per point per visit by year for three separate survey efforts
in the Northern Sierra Nevada. Detections are shown for three detection distance bins with standard
error.
Distance to detection
2008 MIS
2007 MIS
All detections
0.21± 0.03
0.23 ± 0.045
2007-2008
Plumas-Lassen
0.12 ± 0.02
<100 m
0.03 ± 0.01
0.05 ± 0.021
0.03 ± 0.008
<50 m
0.02 ± 0.01
0.03 ± 0.017
0.005 ± 0.004
Model Performance
The predictive performance of our six models ranged from fair to excellent (0.7<
AUC <1.0) when tested against a 25% subset of the data. Our original full
presence/absence model had a good performance AUC value at 0.841. However, when
tested against 2008 survey results (our independent data), this dropped to 0.621 AUC
(Table 5) indicating poor predictive performance. Both the presence/absence vegetation
only and climate only models had lower test scores than the full model, thus rejecting the
hypothesis that fewer environmental variables would improve the model. The full
presence only model (excluding absence data) had a test score of 0.928 indicating
excellent performance. This was the best performing model overall. Its test value
dropped to 0.763 AUC, indicating fair model performance against the independent data
set. Although some loss in performance against independent data is expected, the full
presence only model was the only one that did not have poor test results against the
independent data.
Table 5. Test AUC values for models built with presence/absence data as well as presence only data.
Each model was tested with a 25% subset of the data used to build the model and also with
independent data.
Presence/Absence Model
Presence Only Model
25% subset test
Independent data
25% subset test
Independent data
0.841
0.621
0.928
0.763
Habitat Variables influencing Pileated Woodpecker Occurrence
The Maxent output includes a summary of how each environmental variable
affects a given model. We analyzed this output across all of the models and assessed the
relative contributions of each variable to the modeled occurrence probabilities of Pileated
Woodpeckers. The relative contribution each variable made to the Maxent algorithms
varied between the different models. The most consistent effects was for the percent of
the area comprised of moderate to dense tree cover (>40%) which had a strong positive
relationship with predicted suitability but become negative at high percentages (>80%).
167
Ch.3 Pileated Woodpecker
PRBO Avian Monitoring in the Lassen and Plumas National Forests - 2008
For the presence/absence models it provided the greatest contribution for the
vegetation only model (35.1%) and the greatest contribution of any vegetation variable to
the full model (12.1%). For the presence only models, the amount of habitat in the large
tree size (>24.0” DBH) and the moderate to dense tree cover categories were the second
and fourth highest contributors (22.1% and 14.1%) respectively. The effect of the
amount of habitat in large tree size class was positive until it reached approximately 60%
and then the effect weakened. The percent of Sierran mixed conifer and percent of white
fir also had positive relationships with suitability, while percent red fir had a mostly
negative relationship (Figure 3). These patterns held true when each variable was looked
at in isolation from other the variables.
168
Ch.3 Pileated Woodpecker
PRBO Avian Monitoring in the Lassen and Plumas National Forests - 2008
Figure 3. Model response curves for selected habitat variables for the presence only model, the best
performing model. These curves show how model prediction values changes (y-axis) for each
variable, keeping all other variables at their average sampled value.
% dense tree cover (>60% closure)
% large tree size ( >7.27m crown diameter)
% Sierran mixed conifer (within 1km radius)
% white fir (within 1km radius
% red fir (within 1km radius)
169
Ch.3 Pileated Woodpecker
PRBO Avian Monitoring in the Lassen and Plumas National Forests - 2008
Discussion
The final spatially explicit model predicting suitable Pileated Woodpecker habitat
performs well and is an improvement over a similar model developed using just point
count data prior to the 2007 surveys. The receiver operating characteristic from the final
models show that the final models predicted areas are much more likely to contain
Pileated Woodpecker compared to previous models.
Although 2008 MIS surveys that employed this model had slightly lower
detection rates than the 2007 MIS surveys with that used the older model, the detection
rates are not directly comparable and, in fact, we would expect the 2007 surveys to have
higher detection rates fore several reasons. We selected transect starting locations within
areas of high suitability for both years, but we did not constrain the rest of the survey
points to these same high-suitability areas. Transects in 2008 consisted of 20 points
compared to 6 points in 2007 and therefore covered much more ground. As our model
shows (Figure 1), suitable habitat is not evenly spaced throughout the forest; but instead
is clumped. Thus, since the surveys in 2008 covered approximately 10 km and the 2007
surveys only covered 2.5 it is clear that more of our survey locations were likely to occur
in areas of lower suitability than in 2007. Additionally, by covering more ground in 2008
the number of stations where we detected the same individual woodpecker would be less
than in 2007. In 2007 within our 2.5km transect it was possible, based on published
home range sizes, that a single woodpecker could span the entire transect. This
experience of “dragging” a territorial bird from point to point with playbacks is one of the
reasons we decided to use road based surveys in 2008 as they allowed us to cover much
more ground and be confident that birds detected at the beginning of the transect were
indeed different than those detected in the middle or end (>5km apart).
We found a significant increase in Pileated Woodpecker detections when
comparing both 2007 and 2008 MIS surveys to Plumas-Lassen point count surveys at
unlimited distances and within 50m of observers. With our study design it is not possible
to determine how much of the increase in detection rates is due to playbacks and how
much is due to the use of the models informing transect starting locations. The PlumasLassen point count locations surveyed in 2007 and 2008 are in areas that have much
higher predicted suitability for Pileated Woodpecker than the Lassen National Forest. As
the majority of sites in the Plumas-Lassen are in mixed conifer and white fir dominated
habitats on the Mt. Hough Ranger District which appears to be more suitable habitat than
some of the east side pine and higher elevation red fir forest which is a major component
of the Lassen (Figure 1). Thus we believe without playbacks and the use of the model to
select areas to survey the detection rates on the Lassen would be much lower than those
for the Plumas-Lassen study.
Placing surveys along roads allowed for far greater coverage of potential Pileated
habitat than would have been possible with normal off road foot based surveys. We were
able to cover almost twice as many total survey points in fewer days compared to the
2007 off road foot based surveys. We suggest any Pileated Woodpecker survey protocol
should employ playbacks and, assuming a broad network of roads occurring in moderate
to high density is present, road based surveys should be considered. When available the
use of existing survey data to develop habitat suitability models may also greatly increase
detection rates for this species, especially if every survey locations are only placed within
high and moderate suitability categories (not just starting points).
170
Ch.3 Pileated Woodpecker
PRBO Avian Monitoring in the Lassen and Plumas National Forests - 2008
Based on the model output and results from analysis of local vegetation features
in 2007 (Burnett et al. 2008) several clear patterns emerge with respect to the habitat
components that are important for this species in the Northern Sierra Nevada. Our results
suggest this species is more likely to occur in areas with high canopy closure and large
trees. These results are also consistent with findings from Oregon. Nelson (1988), found
this species density was greater in forests over 80 years old, and old growth stands with
>60% canopy closure were important for nesting and roosting, while mid-seral forests
(>40 years) were preferred for foraging (Bull et al. 1992, Mellen et al. 1992).
Testing distribution models in “real-world” applications is an important
component of evaluating a model’s suitability for its intended use (Vaughan and Ormerod
2005). Our results indicate that testing with non-independent data can provide optimistic
performance assessments for independent data. More importantly, results show that a
presence/absence model may be limited in its usefulness when applied outside of the
environment used to build the model. Although our presence/absence model tested well
with a subset of the original model-building data, it subsequently performed poorly when
tested against independent data. By comparison, the presence model resulted in an
excellent AUC score when tested with a subset of the data and a fairly good score was
achieved when it was tested against independent data. Even though playback surveys
clearly reduced our errors of omission, other factors can lead to faulty absence
information being used in the models thus reducing their performance compared to
presence only models.
Several factors may have contributed to the poorer performance of the model that
included absence data. First, we tested the models against passive point count data with
no playback surveys which we know results in relatively high errors of omission.
Additionally, we suspect this resident species that breeds in cavities begins breeding
earlier in spring than many other birds and thus detectability may be greater earlier in the
spring when most sites are inaccessible due to snow cover. Even with playbacks, due to
the size of their home ranges, Pileated Woodpeckers may not always be within hearing
range of call playbacks or human ears even though the observer is within an occupied
area. There may also be individual variation of the vigor with which this species
responds to playbacks. Indeed, only 7% of our 2008 MIS survey points had a
consistently positive detection from one survey to the next. This is not surprising given
this species shyness and large home range size. Additionally, population dynamics may
result in areas of suitable habitat being unoccupied by this species, resulting in unreliable
absence information (Hirzel et al. 2001). Such a scenario can cause a survey point to be
correctly classified as an absence point while incorrectly informing the model that the
point falls within unsuitable habitat. Long-term surveys may help to reduce these errors,
but we recommend that for this species, models using only presence data are likely to
outperform those that include absence data.
We might also be able to improve model performance by including more
accurate, specific, and local habitat variables. As our results from 2007 showed, habitat
features such as the number and size of snags, amount of downed woody debris, canopy
height, and basal area are important predictors for this species (Burnett et al. 2007).
Unfortunately, more detailed spatially explicit habitat data is not readily available for the
entire Lassen National Forest. While emerging remote-sensing techniques such as
LIDAR (Lefsky et al. 2002) can capture more specific habitat data, this type of
171
Ch.3 Pileated Woodpecker
PRBO Avian Monitoring in the Lassen and Plumas National Forests - 2008
information cannot always be practically captured and included in large scale species
distribution models. At the landscape level, examination of how this species responds to
resources at varying scales may be another way to help refine our models and our
understanding of this species and its needs.
Conclusions
Our results from 2007 and 2008 suggest the Pileated Woodpecker is more
abundant in the Lassen National forest than we suspected based on previous point count
results. Our habitat models though they could benefit from additional information
provide a spatially explicit tool for land mangers to determine the suitability of habitat
across the entire Lassen National Forest for this species. With results from local habitat
analysis in 2007 and model outputs in 2008 this effort has resulted in significantly more
information to help guide management for this species on the Lassen.
Based on our results and two years of experience implementing Pileated
Woodpecker monitoring in the field we believe the most appropriate approach for this
species should employ active playbacks and should seriously consider a road based
survey that employs vehicles to move quickly between distant survey points.
Literature Cited
Araújo, M.B., Pearson R.G., Thuiller§ W., and Erhard M. 2005. Validation of speciesclimate impact models under climate change. Global Change Biology 11:1504-1513.
Brotons, L., Thuiller, W., Araujo, M.B., and Hirzel, A.H. 2004 Presence-absence versus
presence-only modeling methods for predicting bird habitat suitability. Ecography
27:437-448
Bull, E.L., Holthausen, R.S., and Henjum, M.G. 1992. Roost trees used by Pileated
Woodpeckers in northeastern Oregon. Journal of Wildlife Management 56:786-793.
Bull, E.L. and Holthausen R.S. 1993. Habitat use and management of Pileated
Woodpeckers in northeastern Oregon. J. Wildl. Manage. 57: 335–345.
Bull, E.L., and Jackson J.A. 1995. Pileated Woodpecker (Dryocopus pileatus), The Birds
of North America Online (A. Poole, Ed.). Ithaca: Cornell Lab of Ornithology; Retrieved
from the Birds of North America Online:
http://bna.birds.cornell.edu.bnaproxy.birds.cornell.edu/bna/species/148
Burnett, R.D., Jongsomjit, D., Herzog, M., Stralberg, D., Ellis, T., and Humple, D. 2007.
Avian Monitoring in the Lassen and Plumas National Forests: 2007 Annual Report. A
PRBO report to the USFS.
Fielding, A.H., and Bell, J.F. 2007. A review of methods for the assessment of prediction
errors in conservation presence/absence models. Environmental Conservation 24:38-49.
172
Ch.3 Pileated Woodpecker
PRBO Avian Monitoring in the Lassen and Plumas National Forests - 2008
Hirzel, A.H., Helfer, V., and Metral, F. 2001. Assessing habitat-suitability models with a
virtual species. Ecological Modelling 145: 111-121.
Hijmans, R. J. et al. WORLDCLIM Bioclimatic Variables
(http://www.worldclim.org/bioclim.htm) (2008).
Lefskly, M.A., Cohen, W.B., Paker, G.G., and Harding, D.J. 2002 Lidar Remote Sensing
for Ecosystem Studies. BioScience 52:19-30.
LRMP 1992. Lassen National Forest Land Resource Management Plan. USDA Forest Service.
Lassen National Forest, Susanville, CA. Available at:
http://www.fs.fed.us/r5/lassen/projects/forest_plan/lrmp/lassen_lrmp_parta.pdf
Mayer, K.E. and Laudenslayer Jr., W.F. 1988. A Guide to Wildlife Habitats of California.
State of California, Resources Agency, Department of Fish and Game
Sacramento, CA. 166 pp.
Mellen, T.K., Meslow E.C. and Mannan R.W. 1992. Summertime home range and
habitat use of Pileated Woodpeckers in western Oregon. J. Wildl. Manage. 56: 96–103.
Nelson, S.K. 1988 Habitat use and densities of cavity nesting birds in the Oregon coast
ranges. Master’s Thesis. Oregon State University, Corvallis.
Pearce, J., and Ferrier, S. 2000. Evaluating the predictive performance of habitat models
developed using logistic regression. Ecological Modelling 133:225-245.
Phillips, S. J., R. P. Anderson, and R. E. Schapire. 2006. Maximum entropy modeling of
species geographic distributions. Ecological Applications 190:231-259.
Swets, J. 1988. Measuring the accuracy of diagnostic systems. Science 240:1285-1293.
Vaughan, I.P., and Ormerod S.J. 2005. The continuing challenge of testing species
distribution models. Journal of Applied Ecology 42:720-730
173
PRBO Avian Monitoring in the Lassen and Plumas National Forests - 2008
Chapter 4. Resident and Neotropical Migratory Bird Monitoring in
Mountain Meadows on the Almanor Ranger District
Ryan D. Burnett
PRBO Conservation Science
174
Chapter 4. Mountain Meadows
PRBO Avian Monitoring in the Lassen and Plumas National Forests - 2008
Background and Introduction
Mountain meadows are among the most important habitats for birds in California
(Siegel and DeSante 1999, Burnett and Humple 2003, Burnett et al. 2005); they support
several rare and declining species and are utilized at some point during the year by almost
every bird species that breeds in or migrates through the Sierra Nevada. Meadows also
perform a vital role as watershed wetlands that store and purify drinking water for
millions of Californians. And yet, most of these meadows are in a degraded state and
their value as wetlands and as critical habitat for birds and other wildlife has been
dramatically reduced.
Mountain meadows have mostly been lost or heavily degraded by human
activities over the past century (SNEP 1996, Siegel and DeSante 1999). The meadows
that remain are owned by a diverse set of interests including private industry, state and
federal agencies, and private landowners. Most mountain meadows in the Northern
Sierra, including the largest meadows, are privately owned.
The Lassen area supports populations of many declining and threatened riparian
meadow bird species, including Sandhill Crane, Swainson’s Thrush, Yellow Warbler, and
Willow Flycatcher. The area supports breeding populations of 12 of the 17 California
Partners in Flight Riparian Focal Species, 10 of which breed in meadows within the
Almanor Ranger District (Humple and Burnett 2004, RHJV 2004). With its large
diversity and abundance of meadow bird species, including the largest population of
Willow Flycatcher in the Sierra Nevada region (Humple and Burnett 2004), the Lassen
region is a conservation hotspot for meadow birds.
Meadow conservation and management in the Lassen region and throughout the
Sierra Nevada will require a collaborative effort between different land management
agencies and private landowners. Many of the largest meadows in the area are not owned
or managed by the Forest Service. Sites such as Battle Creek Meadow, Childs Meadow,
Humbug Valley, West Shore Lake Almanor, and Warner Valley are non-Forest Service
private holdings within the Almanor Ranger District (ARD). The majority of the
breeding bird species, especially Neotropical migrants, in the ARD use these and other
meadows during some portion of their annual cycle. In order to manage for breeding bird
populations, especially meadow dependent species such as Willow Flycatcher and
Sandhill Crane, the ARD must work collaboratively with the other meadow landowners
in the area in order to ensure for the long-term viability of these and other bird species.
In this chapter we summarize point count data from meadow monitoring on the
Almanor Ranger District from 2004 – 2008. We use a suite of meadow focal species to
compare abundance and richness metrics between meadows and change over time.
Methods
Site Selection
Several considerations went into selecting meadow sites we sampled. Following
an inventory of 16 meadows in the ARD area between 2000 and 2001 we selected a
subset of those sites to continue long-term meadow monitoring within. We were
interested in surveying sites that supported a riparian deciduous shrub (willows/alders)
bird community and especially those sites that had recently undergone management
175
Chapter 4. Mountain Meadows
PRBO Avian Monitoring in the Lassen and Plumas National Forests - 2008
changes (e.g. active restoration and/or removal of grazing). With these two
considerations in mind we attempted to choose sites that represented a range of elevations
and habitat conditions. With this strategy, we believe the sites selected are not
representative of the meadow conditions in the ARD area but represent some of the
higher quality riparian meadow bird habitat in the area.
Point Count Censuses
Point count data allow us to measure secondary population parameters such as
relative abundance of individual bird species and species richness. This method is useful
for making comparisons of bird communities across time, locations, habitats, and landuse treatments.
Standardized five-minute multiple distance band point count censuses (Buckland
et al. 1993, Ralph et al. 1995) were conducted at each of 88 stations along eight transects
in 2005 within the greater ARD area (Table 1). Each of the eight transects are located in
riparian meadow habitat (Figure 1). Point count stations were a minimum of 50 meters
from meadow edges where feasible; if the riparian corridor was less than 100 meters
wide, points were placed equidistant from each edge. At each site points were placed at
200 to 250 meter intervals and were configured in a manner that maximized spatial
coverage of the site.
Table 1. ARD area meadow and riparian point count transects surveyed by PRBO in 2008.
Transect
Carter Meadow
Fanani Meadow
Gurnsey Creek
Humbug Valley
Robber’s Creek
Soldier Meadow
West Shore Lake Almanor
Yellow Creek Riparian
Code
CAME
FAME
GUCR
HUVA
ROCR
SOME
WSLA
YCRI
# of
points
7
8
10
17
14
7
13
12
Year
established
2004
2003
1997
2003
2004
2001
2004
2001
2008
1st Visit
June 10
June 2
June 5
June 3
June 6
June 2
May 31
May 29
2008
2nd Visit
NS
June 16
June 24
June 18
June 20
June 16
June 14
June 17
All birds detected at each station during the five-minute survey were recorded.
Detections were placed within one of six categories based on the initial detection distance
from observer: less than 10 meters, 10-20 meters, 20-30 meters, 30-50 meters, 50-100
meters, and greater than 100 meters. Birds flying over the study area but not observed
using the habitat were recorded separately, and excluded from all analyses. The method
of initial detection (song, visual or call) for each individual was also recorded. Counts
began around local sunrise and were completed within four hours. Each transect was
visited twice each year between late May and the end of June, except Carter Meadow
which was only visited once in 2008 as a result of access restrictions due to wildfire
(Table 1). All surveys were conducted by the author who has been conducting point
counts in the Sierra Nevada since 2000. An electronic range finder was used to assist with
distance estimation at each point count station.
176
Chapter 4. Mountain Meadows
PRBO Avian Monitoring in the Lassen and Plumas National Forests - 2008
Figure 1. PRBO meadow point count sites in the ARD area of Lassen National Forest, 2005.
177
Chapter 4. Mountain Meadows
PRBO Avian Monitoring in the Lassen and Plumas National Forests - 2008
Meadow Area Search and Territory Delineation
At each meadow an area search was conducted to note the presence and number of
territories for three of the rarest meadow breeders in the Sierra Nevada: Sandhill Crane, Willow
Flycatcher, and Swainson’s Thrush. Care was taken to delineate territories during point counting
and area searches, in transit between count stations, and then after completion of the point
counts. No more than one hour was spent searching after completion of counts at any one site.
Point Count Vegetation Assessment
Vegetation at each point count station was assessed using a relevé method, based on the
concepts summarized in Ralph et al. (1993). A 50-meter radius plot centered on each census
station was used. General habitat characteristics of the site were recorded (canopy shrub, and
herbaceous cover, riparian width, etc.) and the cover, abundance, and height of each vegetation
stratum (tree, shrub, herb, and ground) were estimated. Within each stratum, the species
composition was determined and each species’ relative cover recorded, as a percentage of total
cover for that stratum.
Statistical Analysis
Point count analysis was restricted to a subset of the species encountered. We excluded
species that do not breed in the study area as well as those species that are not adequately
sampled using the point count method (e.g., shorebirds, waterfowl, raptors, and swallows). For a
number of the analyses we used a suite of meadow focal species (Table 2).
Table 2. Avian focal species (listed in taxonomic order) for meadow monitoring in the ARD and their
conservation status. California Partners in Flight Riparian Focal species are noted in bold (RHJV 2004).
Species
Conservation Status
Sandhill Crane
State Threatened
Red-breasted Sapsucker
Declining in the Sierra1; NTMB
State Endangered, USFS Sensitive, NTMB
Willow Flycatcher
NTMB
Warbling Vireo
USFS Priority Land Bird Species, NTMB
Swainson’s Thrush
NTMB
Black-headed Grosbeak
State Species of Special Concern, NTMB
Yellow Warbler
MacGillivray's Warbler
NTMB
Significant Decline in Sierra1, NTMB
Wilson's Warbler
None
Song Sparrow
Lincoln's Sparrow
NTMB
1
= from Sauer et al. 2008. NTMB = Neotropical Migratory Bird
Species richness
The species richness index I used is the sum of species detected per point per year and
focal richness is the total number of focal species detected per point per year. Thus the species
richness (or focal richness) for a meadow is the total number of species detected at each point in
a year averaged across all the points in the transect. Presenting the mean species richness, as is
done herein, allows for comparisons between transects or habitats consisting of different
numbers of point count stations.
178
Chapter 4. Mountain Meadows
PRBO Avian Monitoring in the Lassen and Plumas National Forests - 2008
Indices of Abundance
I define the index of total bird abundance as the mean number of individuals detected per
station per visit. This number is obtained by dividing the total number of detections within 50
meters by the number of stations and the number of visits. The same method was employed for
creating abundance indices for focal species combined and individual species.
Trends
We used liner regression using Stata statistical software (Stata Corp 2005), to determine
trends in species richness and bird abundance indices. We assumed statistical significance at the
p<0.05 level.
Results
Song Sparrow was the most abundant meadow bird focal species detected from 2004 –
2008 in ARD meadows with and index of abundance of 1.17, followed by Yellow Warbler at
1.04 (Figure 2). Willow Flycatcher, a forest service sensitive species, had an index of abundance
of 0.08 while Black-headed Grosbeak was the least abundant focal species at 0.01.
Figure 2. The mean abundance (+/- standard error) of nine meadow focal species per point count visit from
2004 – 2008 all sites combined in wet riparian meadows in the Almanor Ranger District.
Focal Species Abundance
Black-headed Grosbeak
Red-breasted Sapsucker
Willow Flycatcher
Lincoln's Sparrow
Wilson's Warbler
Warbling Vireo
MacGillivray's Warbler
Yellow Warbler
Song Sparrow
0.00
0.20
0.40
0.60
0.80
1.00
1.20
1.40
Detections/Point Count Visit
Meadow Comparison
Warbling Vireo was the only species that was detected at each of the eight meadows;
Swainson’s Thrush was the least ubiquitous occurring at only at Humbug Valley. We are only
aware of three locations in the greater Almanor Ranger District where this species breeds,
179
Chapter 4. Mountain Meadows
PRBO Avian Monitoring in the Lassen and Plumas National Forests - 2008
Warner Valley, Ruffa Aspen, and Humbug Valley. Since 2004, Lincoln’s Sparrow were
detected only at two sites, Carter Meadow – where they are very abundant – and Robber’s Creek.
This species appears limited in the Almanor area to sites that are over 5000 feet. Sandhill Crane
were detected at West Shore Lake Almanor, Robber’s Creek, and Humbug Valley. Yellow
Warbler was detected at six sites and each of the remaining focal species was detected at exactly
seven sites (Figure 3).
Figure 3. Index of abundance (detections per point per visit within 50m of observers) for six meadow focal
species at eight sites in the Almanor Range District in 2008. Note that Carter Meadow was only surveyed
once in 2008 all other sites were visited twice.
Focal Species Abundance
Detec tions/Point Count Visit
3.00
Carter Meadow
Fanani Meadow
2.50
Gurnsey Creek
2.00
Humbug Valley
1.50
Robber’s Creek
Soldier Meadow
1.00
West Shore Lake
Almanor
Yellow Creek
Riparian
0.50
0.00
Yellow
Warbler
Song
Sparrow
MacGillivray's
Warbler
Warbling
Vireo
Wilson's
Warbler
Lincoln's
Sparrow
In 2008, Yellow Warbler was most abundant at West Shore Lake Almanor where the
index of abundance was 2.43 per point, followed by Humbug Valley at 1.41. Yellow Warbler
was not detected at Carter Meadow or Soldier Meadow in 2008. Song Sparrow was most
abundant in 2008 at Humbug Valley with an abundance index of 1.71 followed by Gurnsey
Creek at 1.25 and West Shore Lake Almanor at 1.19. Song Sparrow was detected at every site
except Carter Meadow. MacGillivray’s Warbler was most abundant at Fanani Meadow in 2008
with an abundance index of 0.94, followed by Gurnsey Creek at 0.75. Warbling Vireo was most
abundant at Carter Meadow with an abundance index of 0.86 followed by Fanani Meadow at
0.75. Wilson’s Warbler was most abundant at Carter Meadow with and index of 0.72 followed
by Gurnsey Creek with 0.65 and Yellow Creek Riparian with 0.50. Lincoln’s Sparrow was most
abundant at Carter Meadow with an index of 1.43 followed by Robber’s Creek with 0.29, the
only other point count site where this species was detected.
Temporal Patterns
Between 2004 and 2008 there was an increasing trend of 1.1% for species richness in
ARD meadows and a 0.1% increasing trend for meadow focal species richness (Figure 4). Total
bird abundance was increasing at 0.1% as was focal species abundance (Figure 5). None of these
trends was statistically significant. Combined focal species abundance and richness were
180
Chapter 4. Mountain Meadows
PRBO Avian Monitoring in the Lassen and Plumas National Forests - 2008
relatively consistent year to year. Focal species richness ranged from 2.74 in 2004 to a high of
3.06 in 2007. Focal species abundance ranged from 3.37 in 2008 to 3.80 in 2007.
Figure 4. Mean avian species richness and focal species richness across all Almanor Ranger District meadows
surveyed from 2004 through 2008 with standard error and predicted trend line. Neither trend was
statistically significant.
Species Richness
9.00
# of Species/Point Count Station
8.00
7.00
6.00
5.00
4.00
3.00
2.00
Total Species Richness
1.00
Focal Richness
0.00
2004
2005
2006
2007
2008
Figure 5. Mean total bird abundance and total focal species abundance across all Almanor Ranger District
meadows surveyed from 2004 – 2008 with standard error and predicted trend line. Neither trend was
statistically significant.
Avian Abundance
8.00
Individuals/Point Count Visit
7.00
6.00
5.00
4.00
3.00
2.00
Total Bird Abundance
Focal Species Abundance
1.00
0.00
2004
2005
2006
2007
2008
Yellow Creek Riparian had the highest focal species richness each year with a peak of
4.25 in 2007 (Figure 6). Focal richness at Gurnsey Creek, Fanani Meadow, and Carter Meadow
were also relatively high. Focal richness was lowest each year at Soldier Meadow with a peak
181
Chapter 4. Mountain Meadows
PRBO Avian Monitoring in the Lassen and Plumas National Forests - 2008
there of 2.0 in 2007. It was also relatively low each year at West Shore Lake Almanor, Humbug
Valley, and Robber’s Creek. However, with a marked drop in richness at Fanani Meadow in
2008, both Robber’s Creek and West Shore Lake Almanor were higher than this site this year.
Figure 6. Mean meadow bird focal species richness at each of eight meadow sites in the Almanor Ranger
District from 2004 to 2008.
Focal Species Richness
# of Focal Species/Point Count Station
4.50
Carter Meadow
4.00
Fanani Meadow
3.50
Gurnsey Creek
3.00
Humbug Valley
2.50
2.00
Robber’s Creek
1.50
Soldier Meadow
1.00
West Shore
Lake Alm a nor
0.50
Yellow Creek
Riparian
0.00
2004
2005
2006
2007
2008
Focal species abundance showed a different pattern than focal species richness (Figure
7). Focal abundance was highest at West Shore Lake Almanor in 2004 and 2005, Gurnsey Creek
in 2006, Carter Meadow in 2007, and Yellow Creek Riparian in 2008. However, as with focal
richness, focal abundance was markedly the lowest each year at Soldier Meadow. Humbug
Valley and West Shore Lake Almanor which were among the least rich in terms of focal species
had among the highest total abundance of focal species.
182
Chapter 4. Mountain Meadows
PRBO Avian Monitoring in the Lassen and Plumas National Forests - 2008
Figure 7. Mean meadow focal bird species combined abundance at each of eight meadow sites in the Almanor
Ranger District from 2004 to 2008.
Focal Species Abundance
Focal Species Detections/Point Count Visit
5.00
Carter Meadow
4.50
Fanani Meadow
4.00
3.50
Gurnsey Creek
3.00
Humbug Valley
2.50
Robber’s Creek
2.00
Soldier Meadow
1.50
1.00
West Shore
Lake Almanor
0.50
Yellow Creek
Riparian
0.00
2004
2005
2006
2007
2008
Rare Meadow Breeder Inventory
Willow Flycatcher’s were detected at five of the eight meadows surveyed between 2004
and 2008 (Table 3), though they were only detected at West Shore Lake Almanor, Humbug
Valley, and Robber’s Creek in each of the five years. One bird was detected singing in Soldier
Meadow in 2007 and its mate was subsequently found building a nest. In 2005 we documented
three males that had established territories at Fanani Meadow and at least two territories were
still occupied as of 2007; however none were detected here in 2008. The Willow Flycatcher
populations on the West Shore Lake Almanor site appears to be the most stable with 12-14
territories documented in our survey area each year. The area north of the causeway has only
been surveyed twice since 2003 and both times six additional territories were counted in this
area. The vast majority of the Willow Flycatcher territories at this site are on PG&E property
with only 1 to 2 on Forest Service land near the southern terminus of First Avenue in Chester.
Both West Shore Lake Almanor and the Humbug Valley-Yellow Creek Riparian area had
two pairs of Sandhill Crane in 2008 (Table 3). No colts were observed at either site in 2008 but
we did document a nest and subsequent adults with a colt above the causeway at Lake Almanor
in 2007. We have observed a pair of Sandhill Crane in Swain Meadows at the terminus of the
Robber’s Creek transect on two occasions since 2005 though we did not see them here in 2008
and have not documented them breeding here in any year.
In 2007, a single Swainson’s Thrush was detected along Miller Creek adjacent to
Humbug Valley and in 2008 a singing male in the same location as 2007 as well as one within
the valley were detected. A single bird was detected south of Ruffa Ranch on Forest Service
183
Chapter 4. Mountain Meadows
PRBO Avian Monitoring in the Lassen and Plumas National Forests - 2008
property in an aspen/mountain alder stand was detected in 2008, the same location where one
was detected in 2005. Warner Valley is the only other location I am aware of that this species
breeds at in the Lassen National Forest.
Table 3. Total number of rare riparian bird species territories detected from area searches at each meadow
site, 2008. Transect codes are presented in Table 1. An area search of meadow habitat around the Ruffa
Aspen site was conducted in 2008 to include in the rare species inventory but data from that point count
transect is now included in our aspen dataset (see Chapter 2).
Site
ASPN CAME GUCR FAME
Species
Willow Flycatcher
3
0
0
0
Sandhill Crane
0
0
0
0
Swainson's Thrush
1
0
0
0
* Includes Miller Creek area east of the valley proper.
HUVA*
ROCR
SOME
WSLA
YCRI
8
2
2
1
0
0
0
0
0
13
2
0
0
0
0
Discussion
Wet Meadows with extensive riparian deciduous vegetation support rich and abundant
bird populations, they are used extensively following the breeding season by the majority of
upland breeding species, and they are the preferred habitat of several species of conservation
interest. Since wet meadows represent less than 1% of National Forest land in the Sierra Nevada,
and have been heavily degraded over the past century, meadow restoration and conservation
should be among the highest priorities of land mangers in the Sierra Nevada.
Our meadow monitoring in the ARD has been restricted to areas with riparian deciduous
vegetation. While many of these sites have been altered – or undergoing passive restoration
following cessation of grazing in the last 15 years - they should be viewed as among the best
riparian meadow bird sites in the greater Almanor Ranger District area. Outside of these eight
sites there are very few meadows in the ARD area that support large and diverse populations of
riparian meadow dependent bird species. Nearly all of the largest meadows are privately owned
and have been severely degraded by a century of excessive livestock grazing as well as other
mismanagement (roads, culverts, and dams). Sites such as Battle Creek Meadows, Childs
Meadows, Swain Meadow, and Mountain Meadows (including upper Goodrich Creek) are nearly
devoid of riparian vegetation. Though we have not conducted formal bird monitoring surveys at
these sites it is unlikely they support riparian meadow dependent bird populations at all similar to
the eight sites we have surveyed.
The ARD area meadows support higher bird abundance than any other habitat type in the
Lassen region we have surveyed. Only aspen habitat (see Chapter 2) has slightly higher species
richness. Meadows in the greater ARD area are among the most important for meadow birds in
the Sierra Nevada. Yellow Warbler, a California Bird Species of special concern, reaches its
greatest reported density in the state here (RHJV 2004, Heath 2008). The area also harbors more
Willow Flycatcher than any other similarly sized area of the Sierra Nevada and a breeding
population of the state threatened Greater Sandhill Crane. With a wealth of mountain meadows
and many in a degraded state, the Lassen area should be considered an ideal location to focus
restoration actions to benefit these and other meadow dependent bird species.
The populations of meadow-dependent focal bird species appear to be stable across the
sites we monitored in the ARD area since 2004. While this may be considered a positive sign, as
many of these species have shown declines in the Sierra Nevada, it also suggests that many of
184
Chapter 4. Mountain Meadows
PRBO Avian Monitoring in the Lassen and Plumas National Forests - 2008
the sites that have been recovering following grazing removal (and some active restoration) are
not rapidly increasing their capacity for meadow-dependent birds. It appears that many of these
sites could benefit from some additional restoration actions. For many of the sites, removal of
encroaching conifers (Robber’s Creek, Gurnsey Creek, Soldier Meadow) and/or planting of
willows could greatly increase the value of the habitat here (Soldier Meadow, Swain Meadow
below Robber’s Creek). Both Humbug Valley and Yellow Creek have sections of stream
channel that have been isolated from their floodplains and may benefit from more significant
restoration actions that restore a wet meadow condition. An increase in riparian deciduous
vegetation (e.g. Salix, Populus, and Alnus spp.) at many of these sites would greatly enhance
their value to meadow birds.
A priority for meadow conservation in the ARD and surrounding areas should be
protecting and enhancing the largest wet meadows, especially for Sandhill Crane and Willow
Flycatcher. However, our results also show that species such as Lincoln’s Sparrow, Wilson’s
Warbler, and Warbling Vireo are much more abundant in the smaller and higher elevation
meadows, such as Carter. Several other higher elevation meadow sites Robber’s Creek, Hay
Meadow, and Spenser Meadow (where we have conducted post-breeding banding), also support
breeding Lincoln’s Sparrow. Unlike Carter, each of these meadows also have breeding Song
Sparrows co-occurring with Lincoln’s Sparrow. Other researchers have suggested Song Sparrow
expanded its elevation range upwards in the last 75 years in the Sierra Nevada (Siegel and
DeSante 1999, Moritz 2007). It is not clear how this possible invasion of higher elevation
meadows by Song Sparrows will impact the more diminutive Lincoln’s Sparrow but higher
elevation sites should be considered a unique and valuable resource for certain meadow
dependent birds. Thus, we recommend managing the larger meadow complexes at lower
elevations (3500 – 5000 feet) for species such as Sandhill Crane and Willow Flycatcher (Childs
Meadow, Battle Creek Meadow, Deer Creek Meadow, Humbug Valley, West Shore Lake
Almanor) while also protecting and, where necessary, enhancing higher elevation sites to support
species such as Lincoln’s Sparrow and Wilson’s Warbler.
Conclusions
With the loss and degradation of riparian meadow habitat and it disproportionate
importance to birds, restoration and prudent management of meadows in the Lassen region
should be among the highest priorities of land mangers here. Increasing the function and
resiliency of wet willow-filled meadows should be a high priority.
The ARD meadows support dense populations of riparian and meadow species, including
the largest population of Willow Flycatcher in the Sierra Nevada. It is necessary to protect both
lower elevation meadows and higher elevation meadows in order to sustain populations of all
meadow-dependent birds, as very few meadows are good for all meadow species. Meadow
restoration in the Lassen region will require partnerships between the U.S. Forest Service, local
government agencies (e.g. Plumas Corp.), watershed groups, and non-profit organizations (e.g.
TNC and PRBO). Working together these groups have the potential to dramatically increase the
value of meadow habitats for birds in this region.
185
Chapter 4. Mountain Meadows
PRBO Avian Monitoring in the Lassen and Plumas National Forests - 2008
Literature Cited
Buckland, S. T., D. R. Anderson, K. P. Burnham, and J. L. Laake. 1993. Distance sampling:
estimating abundance of biological populations. Chapman and Hall, London.
Burnett, R. D., and D. L. Humple. 2003. Songbird monitoring in the Lassen National Forest:
Results from the 2002 field season with summaries of 6 years of data (1997-2002). PRBO report
to the U.S. Forest Service.
Burnett, R.D., D.L. Humple, T. Gardali, and M. Rogner. 2005. Avian Monitoring in the Lassen
National Forest. 2004 Annual Report. PRBO report to the U.S. Forest Service.
Heath, S.K. 2008. Yellow Warbler (Dendroica petechia) In Shuford, W.D. & T. Gardali (eds.),
California Bird Species of Special Concern. Studies of Western Birds 1. Western Field
Ornithologists, Camarillo, CA and California Department of Fish and Game, Sacramento.
Moritz, C. 2007. A Re-survey of the Historic Grinnell-Storer Vertebrate Transect in Yosemite
National Park, California. A U.C. Berkeley Museum of Vertebrate Zoology report to the
National Park Service.
(available at: http://mvz.berkeley.edu/Grinnell/pdf/2007_Yosemite_report.pdf)
RHJV (Riparian Habitat Joint Venture). 2004. Version 2.0. The riparian bird
conservation plan: a strategy for reversing the decline of riparian associated birds in
California. California Partners in Flight. http://www.prbo.org/calpif/pdfs/riparian.v2.pdf.
Siegel, R.B. and D.F. DeSante. 1999. Version 1.0 The draft avian conservation plan for
the Sierra Nevada Bioregion: conservation priorities and strategies for safeguarding Sierra bird
populations.
SNEP (Sierra Nevada Ecosystem Project) 1996. Sierra Nevada Ecosystems. Volume 1,
chapter 1. Regents of the University of California.
http://ceres.ca.gov/snep/pubs/web/PDF/v1_ch01.pdf
Ralph, C. J., G. R. Geupel, P. Pyle, T. E. Martin, & D. F. DeSante. 1993. Field Methods
for Monitoring Landbirds. USDA Forest Service Publication, PSW-GTR 144, Albany,
CA.
Ralph, C.J., Droege, S., Sauer, J.R., 1995. Managing and monitoring birds using point counts:
standards and applications. In: C. J. Ralph, J. R. Sauer and S. Droege (Eds.), Monitoring bird
populations by point counts. USDA Forest Service, General Technical Report PSW-GTR 149,
161-169.
186
Chapter 4. Mountain Meadows
PRBO Avian Monitoring in the Lassen and Plumas National Forests - 2008
Shuford, W.D., Gardali, T. (Eds.), 2008. California Bird Species of Special Concern. Studies of
Western Birds No. 1. Western Field Ornithologists, Camarillo, CA and California Department of
Fish and Game, Sacramento.
Stata Corp. 2005. Intercooled Stata 8.2 for Windows. Stata Corp. LP College Station,
TX.
187
13 January 2009
California Spotted Owl Module: 2008 Annual Report
Principal Investigator:
John J. Keane
Sierra Nevada Research Center
Pacific Southwest Research Station
U.S. Forest Service
2121 2nd Street, Suite A-101
Davis, CA 95616
530-759-1704; jkeane@fs.fed.us
Research Team:
Claire V. Gallagher, Ross A. Gerrard, Gretchen Jehle, Paula A. Shaklee
Sierra Nevada Research Center
Pacific Southwest Research Station
U.S. Forest Service
1731 Research Park Drive
Davis, CA 95618
530-759-1700
Introduction
Knowledge regarding the effects of fuels and vegetation management on California
spotted owls (Strix occidentalis occidentalis; CSOs) and their habitat is a primary
information need for addressing conservation and management objectives in Sierra
Nevada forests. The specific research objectives of the California spotted owl module as
identified and described in the Plumas-Lassen Study (PLS) Plan are:
1) What are the associations among landscape fuels treatments and CSO density,
distribution, population trends and habitat suitability at the landscape-scale?
2) What are the associations among landscape fuels treatments and CSO reproduction,
survival, and habitat fitness potential at the core area/home range scales?
3) What are the associations among landscape fuels treatments and CSO habitat use and
home range configuration at the core area/home range scale?
4) What is the population trend of CSO in the northern Sierra Nevada and which factors
account for variation in population trend?
188
13 January 2009
5) Are barred owls increasing in the northern Sierra Nevada, what factors are associated
with their distribution and abundance, and are they associated with reduced CSO territory
occupancy?
6) Does West Nile Virus affect the survival, distribution and abundance of California
spotted owls in the study area?
7) What are the effects of wildfire on California spotted owls and their habitat?
Current information on the distribution and density of CSOs across the HFQLG study
area is required to provide the data necessary to build predictive habitat models and
provide baseline population information against which we will assess post-treatment
changes in CSO populations and habitat. Continued monitoring on the Lassen
Demographic Study Area is critical for estimating CSO population trends and status. Our
focus in 2008 was to conduct landscape inventories of CSO distribution and abundance,
and continue banding to provide the required data and baseline information to meet the
objectives of Research Questions 1-4 identified above. Complete landscape inventory
surveys were conducted across 9 of 11 survey areas in 2008 (Figure 1). Surveys were not
conducted in 2 survey areas in 2006-2008 (SA-5, SA-7, Figure 1). Surveys were not
conducted in these 2 study areas in 2006-2008 because sufficient data for determining the
number and distribution of CSO sites for initial habitat modeling efforts was collected in
2004-2005. Details on survey methods are described in the study plan. Efforts were made
to monitor the pair and reproductive status of each owl, and to capture, uniquely colormark, and collect blood samples from each individual owl across the study area. Capture
and color-marking is necessary to estimate survival and population trend, and to assess
exposure to West Nile Virus (WNV)(Research Question #5). We also recorded all barred
and hybrid barred-spotted owls encountered in the study area and synthesized all existing
barred owl records for the northern Sierra Nevada to address Research Question #6.
Additionally, we conducted the second year of a radio-telemetry study on CSOs within
SA-4 in the Meadow Valley project area to document home range size and configuration,
and to assess habitat selection relative to the recently implemented treatments. In
response to a need for information on the association between CSOs and wildfire we
initiated an assessment of CSO distribution, abundance and habitat associations in the
Moonlight and Antelope Complex fire areas. These fires burned in 2007 and we initiated
surveys in 2008 to assess the immediate post-fire response of CSOs. We have added a
seventh research questions to reflect this new research direction.
Results
CSO Numbers, Reproductive Success, Density and Population Trends:
A total of 72 territorial CSO sites were documented across the core PLS study area in
2008 (Figure 2). This total consisted of 52 confirmed pairs, 9 unconfirmed pairs (i.e., one
189
13 January 2009
member of pair confirmed as territorial single plus single detection of opposite sex bird),
and 11 territorial single CSOs (single owl detected multiple times with no pair-mate
detected). Ten pairs successfully reproduced in 2008 (16.4% of confirmed/unconfirmed
pairs). Of these ten pairs, eight were located on the Lassen NF and 2 were located on the
Plumas NF. One of the Plumas pairs involved a spotted owl paired with a sparred owl
(spotted-barred hybrid) that produced one hybrid fledgling. The number of successful
nests on the Plumas was the lowest reproductive year we have recorded during the study.
A total of 17 fledged young were documented in 2008 (1.70 young per successful nest)
(Table 1). Across the recent five years of the study, CSO reproduction has been highest in
2004 and 2007 in terms of the percent of CSO pairs that successfully reproduced, and in
terms of the number of young fledged per successful nest. Approximately 50% of CSO
pairs successfully reproduced in 2004 and 2007 whereas the proportion of pairs
successfully reproducing ranged between 14%-18% in 2005, 2006 and 2008. The number
of young produced per successful nest was more similar across years, ranging between
1.47 -1.81. CSO reproduction is known to vary with spring weather: precipitation
patterns were more similar in 2004 and 2007, with total precipitation relatively low
during March-April of 2004 and 2007 as compared to 2005 and 2006 (Figure 3).
However, this pattern varied in 2008 as precipitation was low in March-April. Although
precipitation was low during these two months in 2008, CSO reproduction was low in
2008. The low reproduction may be associated with the heavy snowpack that persisted
into May-June as a result of heavy snowfall in January-February that accumulated across
much of the study area. Additionally, small mammal numbers appeared to be low in 2008
(D.Kelt, pers. comm.)
The Lassen Demographic Study Area (SA-1A, SA-11, SA-12, SA-13, SA-14, SA-15)
and Plumas NF Survey Areas (SA-2, SA-3, SA-4, SA-5, SA-7) were fully integrated in
2005 to define the overall Plumas-Lassen Study project area and provide consistent CSO
survey effort across the project area. (Figures 1 & 2). We estimated the crude density of
CSOs based on the number of territorial owls detected across 9 survey areas during 2008
surveys at the Survey Area spatial scales (Tables 2 and 3). The estimated crude density
across the overall study area in 2008 was 0.067 territorial owls/km2. Overall study area
crude densities are not directly comparable across years because different total areas were
surveyed in each year. However, crude density estimates within individual Survey Areas
indicate similar densities and number of territorial sites (pair sites plus territorial single
sites) between 2004-2008 for the survey areas on the Plumas NF (SA-2, SA-3, SA-4),
while numbers have declined somewhat on the Lassen survey areas (SA-1A, SA-11, SA12, SA-13, SA-14, SA-15) between 2005-2008 (Tables 2 and 3). Due to an active
wildfire (Cub Fire) and associated safety concerns we could not conduct surveys at two
CSO territories within SA-13 in 2008. Thus, the crude density estimate for SA-13 in 2008
is not directly comparable to previous years.
The most recent information on CSO population trends is included in the January 2006
meta-analysis, conducted to estimate CSO population trends and to assess population
status in response to a petition to list the CSO under the Endangered Species Act
(Blakesley et al. 2006). These data continue to provide the best estimates of CSO
population trends. Data collected between 1990-2005 from four CSO demographic
190
13 January 2009
studies across the Sierra Nevada and southern Cascades, including the Lassen
Demographic Study Area, were analyzed as part of the meta-analysis workshop. The
Lassen Demographic Study Area is contained within the overall PLS study area and
consists of survey areas SA-1A, SA-11, SA-12, SA-13, SA-14 and SA-15 in Figure 1.
Full details on meta-analysis methods and results are provided in Blakesley et al. (2006).
In synopsis, across the four study areas, results indicated that the Lassen Study CSO
population exhibited the strongest evidence for a population decline between 1990-2005.
Mean lambda for the Lassen Demographic Study was 0.973, with 95% confidence limits
ranging from 0.946-1.001 (Table 4).
Habitat Assessment – Nest/Roost Plot Scale
We documented a total of 103 CSO territorial sites between 2004-2006. We overlayed
the primary nest/roost locations for each of the 103 CSO sites with the CWHR vegetation
classes available within the VESTRA photo-interpreted vegetation map for the PLS to
examine nest/roost-site habitat association patterns. Approximately 53% of the nest sites
were located within CWHR 5M, 5D and 6 size classes (Table 5, Figure 4). An additional
37% of the sites were located within CWHR size class 4M and 4D polygons. CWHR size
class 4 is defined as stands with average tree sizes of 12-24 inch diameter-at-breastheight (dbh) trees. Of the 38 sites located in size class 4 polygons, 25 (66%) were in size
class 4 polygons with a large tree component (i.e., presence of >24 inch dbh trees).
Overall, about 90% of the sites were located within CWHR 4M, 4D, 5M, 5D, and 6 size
classes. The remaining 10 sites were located in more open, smaller-tree size polygons,
with nests or roosts located within remnant, scattered larger trees (Table 5, Figure 4).
While the distribution of nest site locations relative to broad vegetation classes provides
insight into patterns of nest-site habitat, we also conducted vegetation sampling at nest or
primary roost sites to describe vegetation structure and composition. Vegetation plot
sampling was conducted at 80 CSO territories across 2005-2007. Vegetation plots were
centered on CSO nest trees, or on a primary roost tree for sites where no nest has been
documented, and were measured using the national Forest and Inventory Assessment
(FIA) protocol. The FIA protocol is used nationally by the USDA Forest Service for
inventorying and monitoring vegetation. FIA sampling consists of measuring vegetation
structural and compositional variables within a 1-ha plot centered on a CSO nest or roost
tree. Only one plot was collected from each CSO territory, with the most frequently used
nest tree serving as the plot center location, or the most recent nest tree used at sites
where no nest tree was used more frequently than another. CSO nest sites were
characterized by mean total basal areas of 260.8 ft2/acre, 7.4 snags (>15 inch dbh)/acre,
and 10.7 trees (>30 inch dbh)/acre (Table 6). Under the FIA protocol, canopy cover is
modeled based on the tree inventory list. The modeled canopy cover for these plots
averaged 64.1%. Shrub cover averaged 7.7%. Fuel loads averaged 0.75 tons/acre for 1-hr
fuels, 4.0 tons/acre for 10-hr fuels and 4.44 tons/acre for 100-hr fuels (Table 6). Use of
the FIA sampling protocol will facilitate monitoring of vegetation and development of
CSO habitat models that can be used as adaptive management planning tools. Habitat
191
13 January 2009
models are currently being evaluated that can be used to assess projected changes in CSO
nesting habitat suitability under varying fuels and vegetation treatment scenarios.
Habitat Assessment – Core Area/Home Range Scale
Core area habitat associations around 102 CSO nest/roost sites was assessed by using a
Geographic Information System (GIS) and the VESTRA photo-interpreted vegetation
map to determine the vegetation patterns within a 500 acre (201 ha) circle centered on
each of the CSO territory sites. To compare the CSO sites with the general availability of
habitat across the study area we also assessed the same vegetation patterns around 130
points determined by placing a systematic grid across the study area. For this summary
we assessed vegetation using the USDA Forest Service Region 5 classification system.
Overall, CSO core areas averaged 75.7% suitable habitat (classes 3N, 3G, 4N, 4G)
whereas the grid points averaged 61.9% (Table 7, Figure 5). Approximately 32% of CSO
core areas was composed of large tree polygons (>24inch dbh, >=40% canopy cover)
compared to 19.6% of the grid points (Table 7, Figure 6).
Radio-Telemetry – Meadow Valley Project Area
Eight adult territorial CSOs were radio-tagged during April-June of 2007 within SA-4 in
the Meadow Valley Project Area. The sample included 3 males and 5 females. CSOs
were fitted with 12g backpack-mounted transmitters from Holohil Systems with projected
radio life expectancy of 1.5 years. We attempted to locate each radio-tagged CSO 5 times
over each 2-week sample period between April and September 2007. CSOs were tracked
from the ground using vehicles and hand-held H-antennas. Approximately 30 locations
were recorded for each individual between April and September. The eight birds tagged
in 2007 were followed at reduced effort during the 2007-2008 nonbreeding period to
determine wintering locations and post-breeding movements. Two mortalities were
recorded during the nonbreeding period. One male and one female died of apparent
natural causes.
An additional 2 territorial CSOs (one male, one female) were radio-tagged during AprilJune 2008 within SA-4 in the Meadow Valley Project Area. The six birds tagged in 2007
plus the 2 new birds tagged in 2008 were followed between April-September 2008 using
the same sampling protocol and frequency described above for the 2007 breeding period.
These data will be used to investigate CSO home ranges sizes and configurations, as well
as habitat selection within home ranges relative to available vegetation and fuels
treatments. These efforts to assess CSO use of the post-treated landscape in SA-4
(Meadow Valley) are severely hampered by the lack of post-treatment vegetation data
(see discussion below under Meadow Valley Project Area Case Study for further details).
The radio-telemetry data will also be used to look at the plot-scale vegetation structure
and composition at CSO activity locations determined via telemetry. Eighty-seven
192
13 January 2009
vegetation plots were measured to the standard FIA protocol between August-November
2008 at a sub-sample of CSO activity locations.
Meadow Valley Project Area Case Study:
The Meadow Valley Project Area (MVPA) is the first area within the PLS where the full
implementation of HFQLG treatments has occurred. Treatments were implemented on
the ground within this project area during 2001-2008, with primarily light-thinning and
underburning occurring in 2001-2005, and Defensible Fuel Profile Zones and Group
Selections implemented during 2005-2008. The MVPA corresponds closely with the
boundaries of SA-4 of the PLS.
We began monitoring CSOs SA-4 in 2003 and have annually monitored the distribution,
abundance and reproduction of CSOs within SA-4. Additionally, we have color-banded
all individuals within this area, with the exception of one male who could not be
captured. Full survey methods are described in detail in our study plan (available from
field project leaders) and are consistent with USDA Forest Service R5 survey methods.
Briefly, we conduct 3 nocturnal broadcast surveys during the breeding period (AprilAugust) across a network of survey points to detect CSOs. When a CSO is detected we
then conduct dusk status surveys to pinpoint roost and nest locations for each bird. Status
surveys are used to determine the social status of each bird (pair or single), nesting and
reproductive status (breeding, non-breeding, unknown), and to identify color-banded
individual birds.
In general, in years of higher CSO reproduction, such as occurred in 2004 and 2007, it is
easier to establish pair and reproductive status and to identify individual birds as they are
more vocal and exhibit stronger ties to their core areas. In years of lower reproduction,
such as occurred in 2005, 2006, and 2008, it is more difficult to determine the status of
birds as they tend to range more widely and are not as vocal and territorial, particularly
the females. Based on our cumulative survey results, we then use accepted, standardized
methods for estimating the overall number of territorial sites (confirmed pairs,
unconfirmed pairs and territorial singles) for each year. Confirmed pairs consist of a
reproductive pair of CSOs or, at non-reproductive sites, the detection of a male and
female on more than one occasion within 1/2-mile of each other across the breeding
period. Unconfirmed pairs consist of two sightings of one sex and one detection of the
opposite sex within 1/2-mile of each other across the breeding period. Territorial singles
are considered to be individuals that are detected on at least 2 occasions within a 1/2-mile
distance across the breeding period without a detection of the opposite sex. Birds
detected on only a single occasion across the breeding period are not considered to be
territorial.
Figure 7 illustrates the proposed treatment locations and the cumulative number and
distribution of CSO territorial sites across the six years between 2003-2008. The number
of territorial sites across SA-4 varied annually between 6-9 (Table 8, Fig. 8). Overall, the
193
13 January 2009
numbers of territorial sites was fairly similar with 7 sites documented between 20042006, an increase to 9 territorial sites during the high reproductive year that occurred in
2007, and then decreasing to 6 territorial sites in 2008.
Of note, one of the territorial sites occupied from 2004-2007 was not occupied in 2008,
thus reducing the number of territorial sites to 6 as compared to the base of 7 documented
between 2004-2006. The territory in question was located at Maple Flat in the extreme
Northwest corner of SA-4. This site was located at high elevation (approx. 1700m, 5600
ft.) relative to the other sites in the SA. Treatments occurred in this site during Fall 2007.
Both the male and female from this pair were out-fitted with radio-transmitters in Spring
2007. Both birds remained on their territory through the 2007 breeding period. In
November 2007 the male traveled approximately 29 km (18 mi.) to Coyote Gap,
dropping to about 1400m (4600 ft) elevation. He remained in this area through February
2008, when he was found dead. He had been scavenged and a cause of death could not be
determined. The female left the Maple Flat site in September 2007, moved north to near
Seneca where she remained until early November. She then was detected in Maple Flat
on 7 November 2007, after which she then migrated down west slope of the Sierra
Nevada, ultimately settling for the winter near Lake Oroville at approximately 480m
(1600 ft.) elevation. This winter location was a straight-line distance of 56 km (35 miles)
from the Maple Flat breeding site. The female remained at this winter location through
early-March 2008. She moved back upslope and was detected within 3.2 km (2 miles) of
the Maple Flat breeding site in mid-March 2008. She then began to wander and moved 24
km (15 miles) north to Wolf Creek near Greenville, eventually settling 14.5 km (9 miles)
from Maple Flat near Seneca in early June 2008. She remained in this area through
October 2008 when she was recaptured and the radio-transmitter was removed. No new
CSOs were detected or colonized the Maple Flat site in 2008.
Whether the treatments may have caused the Maple Flat site to be unoccupied in 2008
can not be determined with certainty. CSOs are known to exhibit breeding dispersal from
a territory following the death of a mate. The radio-tagged female returned to the Maple
Flat area in Spring 2008, apparently may not have found a male present and then
dispersed until eventually settling 14.5 km away in SA-2 for the summer. Whether or not
the Maple Flat site, or an alternate site in the near vicinity, will be re-colonized will
require additional years of monitoring. Of importance, 2008 was the lowest reproductive
year recorded on the Plumas NF, with only 2 nests documented across all of the Plumas
NF sites. Thus, the conditions leading to the low reproductive activity in 2008 may have
resulted in a low probability of recruitment for a new male colonizing the site in 2008.
For example, a higher number of territorial sites (9) were documented within SA-4 in the
higher breeding year of 2007 as compared to the 7 territorial sites documented in the low
reproductive years of 2005-2006. Also, higher CSO reproduction in 2007 may result in
increased number of recruits available to colonize sites in 2009-2010.
To date, we have not observed apparent nor dramatic changes in the numbers of
territorial CSO sites within SA-4 as an immediate acute response to treatments. These
initial findings should be tempered by the need to assess possible chronic, or longer-term,
responses by CSOs. Specifically, we recommend that monitoring be continued to assess:
194
13 January 2009
(1) long-term occupancy, abundance and distribution of CSOs across the project area to
document longer-term responses to address concerns that site fidelity in such a long-lived
species may obscure possible negative effects of habitat change over the short term; and
(2) to continue to monitor color-banded birds to assess longer-term associations between
CSO survival, reproduction, and recruitment related to changes in habitat. Each of the
pieces of above information is necessary to fully assess the potential acute and chronic
responses of CSOs to landscape treatments.
One significant impediment to fully assessing the associations between CSO response
and treatments in the MVPA is the lack of accurate spatial mapping of: (1) the specific
locations where treatments were actually implemented on the ground; (2) the specific
site-specific treatments that were implemented on a piece of ground; (3) when the
treatments were implemented on the ground (which year at minimum); and (4) the
resulting post-treatment vegetation structure and composition. Understanding the what,
where, when, and effects of treatments is the foundation on which subsequent adaptive
management assessments will be constructed. Generalized project-planning polygon
mapping showing where treatments may occur, and what specific treatments might be
implemented, are not specific and accurate enough for post-treatment adaptive
management assessment of treatment effects on CSOs and their habitat.
Accurate post-treatment vegetation maps are needed at resolutions appropriate for
differing management and research objectives. In the case of the CSO, we have not
observed dramatic short-term changes in CSO numbers across the broader MVPA in
response to treatments. However, we have documented some changes in the distribution
and occupancy of CSO territories where treatments have occurred within SA-4 that may
be related to treatments. We require accurate post-treatment vegetation information if we
are to fully explore the response of CSOs and their habitat to treatment effects within an
adaptive management framework. This is a critical information issue that requires
immediate attention.
Moonlight and Antelope Complex Fire Area Case Study
A primary source of uncertainty regarding the effects of fuels treatments is an assessment
of risk to CSOs and their habitat from treatments versus the risk from wildfire that occurs
across untreated landscapes. Our PLS work to date has focused on assessing CSO
distribution, abundance and habitat associations across the untreated overall project area
landscape and being in position to monitor effects as treatments are implemented within
specific project areas, as illustrated by the MVPA case study described above. In 2008 we
were fortunate to have the opportunity and funding support from the Plumas National
Forest to inventory CSO distribution, abundance, and status across the Moonlight and
Antelope Complex Fire Area (MACFA) that burned in 2007. These two fires burned
adjacent to each other in 2007 and both were primarily high severity fires. The MACFA
covers approximately 88,000 acres. We conducted CSO surveys across this entire
landscape and within a 1.6 km (1 mile) unburned buffer surrounding the MACFA during
the 2008 breeding period. We used our standardized survey protocol and conducted 3
195
13 January 2009
nocturnal surveys across the landscape with follow-up visits to attempt to located
nest/roost locations for birds detected on nocturnal surveys. These methods are described
in the section above and fully in protocol described in the study plan. We also contracted
to obtain pre- and post-fire vegetation maps to be able to assess changes to the vegetation
and CSO habitat.
The high-severity fires that burned in the MACFA resulted in significant changes to the
vegetation (Figures 9,10,11). The amount of suitable CSO habitat (CWHR classes 4M,
4D, 5M, 5D) within the 88,000 acre MACFA decreased from 70.1% of the landscape to
5.8% of the landscape following the fires. The largest increase in the post-fire landscape
occurred in the CWHR classes <= 2D which increased from 8.2% to 64.9%. The
remaining forested areas across the post-fire landscape were predominantly classified as
either 4P (18.5%) or 4S (7.9%) (Fig 9 & 10).
We are still in the process of synthesizing all of the pre-fire CSO survey information for
the MACFA as there is not a solid baseline of consistently collected survey information
prior to the fire such as exists for our core PLS project area. Nevertheless, this synthesis
may provide us with a reasonable estimate of the pre-fire distribution and abundance of
CSO sites across the MACFA. All or parts of at least 23 PACs were located within the
pre-fire MACFA. Given the lack of continuous annual CSO survey effort we are
uncertain what proportion of those PACs were occupied in 2007 prior to the fires.
During our 2008 surveys we documented a single confirmed pair of CSOs (non-breeding)
within the MACFA, with the female from this pair being the only female we detected
within the fire area (Fig 11). We had 10 single detections of male CSOs across the burned
area. In each of these ten cases we were not able to locate the birds at nests or roosts on
follow-up status surveys. Each of these ten locations occurred primarily in the middle of
the night when birds are out foraging and none of the detections occurred within 1/2-mile
of each other as required to classify these individuals as territorial birds under currently
accepted protocols. Within the unburned 1-mile buffer area surrounding the burned area
we documented 5 confirmed pairs, 1 unconfirmed pair, 1 territorial male single, and 6
single detections (4 males, 2 sex unknown). Thus, in the immediate unburned buffer area
we observed territorial sites whereas we only were able to document the single confirmed
territorial pair within the burned area.
In this first year of survey work in the MACFA we were able to document significant
changes to the vegetation and amounts and distribution of CSO habitat as a result of the
high-severity wildfires. Our initial year of CSO survey work suggests that the immediate
post-fire landscape may not support territorial CSO sites as evidenced by the single
confirmed pair of owls that we documented in 2008. The lack of additional confirmed
pairs, lack of female detections, timing of male detections, and our inability to pin down
nest/roost locations for any of the males detected within the burned area supports this
hypothesis. Alternatively, perhaps CSOs within the burned area were behaving
differently and were not as responsive to surveys in the burned landscape as compared to
unburned buffer and broader PLS project area. Additionally, 2008 was a very low
reproductive year on the Plumas and perhaps female CSOs were less vocal and less likely
196
13 January 2009
to be detected within the fire area as compared to the buffer and PLS project area.
However, both of these explanations do not seem probable as we do not suspect that there
would be significant differences in response rates between CSOs in the burned area
compared to the buffer area. Our 2009 surveys will shed light on these issues as we will
be able to re-survey the burned and unburned areas to document if CSOs are able to
persist on territories within the MACFA.
It is important to determine both the acute and chronic responses of CSOs and their
habitat to wildfire as it is unknown if CSOs can persist over both the short-term and longterm in these areas. Whether a landscape that has experienced wildfire can support CSOs
likely depends on the pre-fire habitat suitability and variable fire severity patterns both
within individual fires and across different fires. Largely low-mid severity fires may have
positive or neutral effects on CSOs and their habitat while high severity fires may result
in greater negative effects.
Banding, Blood Sampling, West Nile Virus Monitoring
Forty-seven owls were captured and banded in 2008. Blood samples were collected from
21 individuals that will be screened at the University of California, Davis for West Nile
Virus (WNV) antibodies. None of the 158 individual blood samples collected from 20042007 have tested positive for WNV antibodies. The 2008 samples have not been analyzed
to date.
Barred and Sparred (Spotted x Barred hybrid) Distributional Records
We detected the presence of 3 barred owl and 4 sparred owls during 2008 surveys within
our intensive study area. Our synthesis and update of barred-sparred owl records through
2008 based on Forest Service and California Department of Fish and Game databases
indicates that there are a minimum of 38 individual records across the HFQLG Project
Area and a minimum total of 51 across the Sierra Nevada (Figure 7). This includes a
minimum total of 18 records that have been documented within our intensively surveyed
study area. The first barred owl in the region was reported in 1989. The first documented
breeding in the PLS survey area was in 2007, and repeated again in 2008. The pattern of
records suggests that barred/sparred owls have been increasing in the northern Sierra
Nevada between 1989-2008.
California Spotted Owl Diet
A single diet survey plot was established at a CSO nest or roost location at each CSO
territory on the Plumas National Forest during 2003-2007. Systematic searches for
pellets and prey remains were conducted in each plot during each year. A total of
approximately 3398 pellets have been collected during 2003-2007 (2003 = 606; 2004 =
807; 2005 = 838; 2006 = 516; 2007 = 552). To date 2846 pellets, covering the period
197
13 January 2009
2003-2006, have been sorted and all prey items identified to species, or taxonomic group
when species identification could not be ascertained. Scientific names of all species
recorded in pellets are listed in Appendix A. Mammals comprised the dominant
taxonomic group identified in the diet. Across years the four most frequently detected
species or taxa were the dusky-footed woodrat (recorded in 45% of pellets), northern
flying squirrel (recorded in 35% of pellets), Peromyscus species (recorded in 31% of
pellets) and other small mammals (recorded in 31% of pellets) (Fig. 13a, Table 9). In
terms of biomass contributions across years, dusky-footed woodrats contributed 45% of
the estimated biomass, followed by northern flying squirrels (20%) (Fig 13b, Table 10).
There appears to be some level of annual variation in diet (Figures 14a&b, Tables 9 &
10). For example the annual estimate of biomass contribution from woodrats was
estimated to range from 38-54% in any single year. The 552 pellets collected in 2007
have been sorted and identification of all prey species is near completion. Our objective
was to sample over several years to assess temporal variation in diets and possible
relationships to variation in CSO reproduction, and to sample widely over space in order
to investigate potential variation in CSO diets associated with elevation and vegetation
conditions. Once the 2007 samples are fully complete we will have enough information
to meet these objectives.
Summary 2003-2008
Our efforts from 2003-2007 have focused on collecting the initial data to address our
primary research objectives and provide the baseline data for monitoring HFQLG
implementation. In conjunction with the now fully integrated Lassen Demographic Study
we have collected landscape-scale information on the distribution and abundance of
CSOs across approximately 650,000 acres of land. Determining the accurate number and
distribution of CSO sites requires multiple years of survey and marking of individual
CSOs to delineate separate territories and identify individual birds that move among
multiple sites within and across years. These baseline data are fundamental for
developing empirically based habitat models for understanding CSO habitat associations
and developing adaptive management tools and models. The near completion of the
Meadow Valley area projects in 2007-2008 marked the first landscape series of HFQLG
treatments to be implemented within the study area, providing the first opportunity to
address treatment effects within a case study framework. Our baseline information on
CSO distribution and habitat associations, coupled with our 2007-2008 radio-telemetry
work, will allow us to assess associations between CSOs and vegetation changes. In
2008 we were now able to begin monitoring CSO distribution and abundance in the
Meadow valley project area, providing the first empirical data from a treated landscape.
Additionally we were able to expand our work to address the effects of wildfire and
CSOs and their habitat through our initial survey work in the Moonlight-Antelope
Complex fire area. In summary, we are working towards being able to broadly address
CSO management questions across a gradient of landscape conditions ranging across
untreated landscapes, landscapes treated to meet desired fuels/vegetation conditions, and
198
13 January 2009
landscapes that have experienced wildfire in order to address primary management
issues.
Dedicated monitoring of CSOs on the Lassen Demographic study continues to provide
critically valuable demographic and population trend information for determining the
status of CSOs. The declining population trend estimated through the meta-analysis of
the Lassen Demographic Study data and the apparent decline in numbers of CSOs
observed between 2005-2007 within the Lassen NF survey areas warrant close continued
monitoring of the status of CSOs within the study area, along with continued
management focus on providing high-quality CSO habitat during the planning and
implementation of HFQLG treatments. We lack similar long-term demographic data for
the Plumas NF study areas, but our baseline information on CSO distribution and
abundance suggests that numbers of territorial CSOs and sites have been similar across
2004-2008.
Our focused diet analyses have broadened and deepened our understanding of CSO diets
and sources of variation in CSO diets among pairs and across environmental gradients.
Monitoring of WNV exposure coupled with demographic monitoring has provided an
opportunity to assess if WNV may ultimately be a factor influencing CSO viability. To
date we have not had a positive detection for WNV within CSOs. Finally, through our
research into historical and current occurrence records, in conjunction with our field
surveys, we have been able to document the colonization of the northern Sierra Nevada
by barred owls. Our results indicate that barred owls are increasing in the northern Sierra
Nevada and may become an increasing risk factor to CSOs.
Current Research: 2009
In 2009 we will continue monitoring owl distribution, abundance, demography, and
population trend across the core PLS study area. We will also conduct our second year of
CSO surveys to document distribution, abundance and habitat associations within the
Moonlight and Antelope Complex fire area. We will augment our existing work with two
additions. First, we will conduct CSO surveys in Empire (Plumas NF) and Scotts John
Creek (Lassen NF) proposed project areas. This new work, coupled with our ongoing
work in the Meadow Valley (Plumas NF) and Creeks (Lassen NF) will provide the
baseline data for four of the first project areas that are scheduled for implementation and
position us to assess effects to CSOs and their habitat as these first projects are
implemented on the ground. Together this work will provide a more comprehensive base
of knowledge regarding CSO habitat associations and the effects of treatments and
wildfires.
In addition to continuing field surveys in 2009 designed to address our six research
questions, we have broadened our emphasis on the development of predictive habitat
relationship models as described in the module study plan. We have continued to work
closely with biologists on the Plumas and Lassen National Forests, and the R5 Regional
199
13 January 2009
Office, to identify and define the types of analyses and tools that would best address
management needs. Baseline information collected during this study forms the
foundation for this phase of the research. The combination of broad-scale landscape CSO
distribution data, in conjunction with detailed demographic information available from
the Lassen Demographic Study, will facilitate exploration and development of predictive
habitat models for use in an adaptive management framework and to directly monitor
implementation of the HFQLG project.
Literature Cited
Blakesley, J.A., M.E. Seamans, M.M. Connor, A.B. Franklin, G.C. White, R.J. Gutierrez,
J.E. Hines, J.D. Nichols, T.E. Munton, D.W.H. Shaw, J.J. Keane, G.N. Steger, B.R.
Noon, T.L. McDonald, S. Britting. 2006. Demography of the California Spotted Owl in
the Sierra Nevada: Report to the US Fish and Wildlife Service on the January 2006 MetaAnalysis. February 2006.
200
13 January 2009
Table 1. California spotted owl reproduction on the Plumas and Lassen National Forests
2004-2008.
Year
2004
2005
2006
2007
2008
Percent of confirmed/unconfirmed pairs
with successful nests
49.4%
17.7%
13.8%
55.4%
16.4%
Young fledged per
successful nest
1.68
1.47
1.50
1.81
1.70
201
13 January 2009
Table 2. Crude density of territorial California spotted owls across survey areas on the
Plumas and Lassen National Forests 2004-2008. Locations of survey areas are identified
in Figure 1.
Survey Area
SA-2
SA-3
SA-4
SA-5
SA-7
SA-1A
SA-1B**
SA-11
SA-12
SA-13
SA-14
SA-15
Total Study
Area
Crude Density of Territorial
Owls (#/km2)
2005*
2006* 2007*
2008*
Size
(km2)
182.4
214.4
238.2
260.2
210.3
190.4
130.3
179.4
215.8
152.9
318.7
196.8
2004*
0.126
0.075
0.059
0.069
0.071
NI***
NI
NI
NI
NI
NI
NI
0.143
0.093
0.050
0.069
0.062
0.042
0.023
0.045
0.097
0.105
0.053
0.086
0.115
0.115
0.132
0.089
0.103
0.098
0.046
0.071
0.046
NS**** NS****
NS****
NS
NS
NS
0.042
0.053
0.042
NS
NS
NS
0.033
0.033
0.045
0.070
0.074
0.070
0.085
0.065 0.050*****
0.044
0.035
0.047
0.036
0.056
0.081
2489.8
0.078
0.073
0.060
0.066
0.067
*Total Area surveyed each year: 2004 = 1,106 km2; 2005 = 2,490 km2; 2006 = 1,889 km2; 2007 =
1,889 km2; 2008 = 1,877 km2
**NI = not included. Project level area surveyed only in 2005. Included for comparative
purposes.
***Lassen Demographic Study Area – incorporated into the overall study in 2005.
****Survey areas not surveyed in 2006-2008.
*****This survey area was not completely surveyed during 2008 because of wildfire activity in
the area. Two CSO territories within the study area could not be surveyed.
202
13 January 2009
Table 3. Number of pairs (confirmed and unconfirmed) and territorial single California
spotted owls across the Plumas-Lassen Study survey areas on the Plumas and Lassen
National Forests, California, 2004-2008.
2004
Survey Pairs TS*
Area
SA-2
11
1
SA-3
7
2
SA-4
7
0
SA-5
8
2
2005
Pairs TS*
2006
Pairs TS*
2007
Pairs TS*
2008
Pairs TS*
1
1
3
--
10
11
8
NS**
1
0
1
--
12
9
5
NS**
12
10
5
9
2
0
2
0
10
9
4
NS**
**
SA-7
SA-1A
SA1B**
SA-11
SA-12
**
0
3
1
--
**
7
NI***
NI
1
---
6
4
3
1
0
0
NS
4
NS
-0
--
NS
5
NS
-0
--
NS
4
NS
-0
--
----
4
10
8
0
1
0
3
1
6
0
7
1
3
8
5
0
0
0
3
7
SA-13
NI
NI
NI
2
1
1
SA-14
SA-15
NI
NI
3****
*
---
8
8
1
1
7
3
0
1
5
4
1
3
7
8
1
0
*TS = Territorial Single.
**NI = not included. Project level area surveyed only in 2005. Included for comparative
purposes.
***Lassen Demographic Study Area – incorporated into the overall study in 2005.
****Survey areas not surveyed in 2006-2008.
***** This survey area was not completely surveyed during 2008 because of wildfire activity in
the area. Two CSO territories within the study area could not be surveyed.
203
13 January 2009
Table 4. Mean estimated population lambda (population change) for California spotted
owls on four study areas in the southern cascades and Sierra Nevada, 1990-2005
(Blakesley et al. 2006)
Study Area
Lambda
Standard Error
95% Confidence Interval
Lassen National
0.973
0.014
0.946-1.001
Forest
Sierra National
0.992
0.013
0.966-1.018
Forest
Sequoia-King
1.006
0.031
0.947-1.068
Canyon National
Park
Eldorado National
1.007
0.029
0.952-1.066
Forest
204
13 January 2009
Table 5. Distribution of California spotted owl nest/primary roost sites (n = 103) across
CWHR tree size classes within the Plumas-Lassen Study on the Plumas and Lassen
National Forests, 2004-2006.
CWHR
Size
Class*
Barren
3S
3M-LT
3D
4P
4M
4M-LT
4D
4D-LT
5M
5D
6
CWHR Size Class Description
Open, sparse tree coverage
6-12 inch dbh, ,20% CC
6-12 inch dbh, 40-60% CC, large trees recorded
6-12 inch dbh, >60% CC
12-24 inch dbh, 20-40% CC
12-24 inch dbh, 40-60% CC
12-24 inch dbh, 40-60% CC, large trees recorded
12-24 inch dbh, >60% CC
12-24 inch dbh, >60% CC, large trees recorded
>24 inch dbh, 40-60% CC
>24 inch dbh, >60% CC
>24 inch dbh, >60% CC, multi-layer canopy
Number
of Nests
Percent
1
1
1
4
3
3
12
10
13
25
9
21
1.0
1.0
1.0
3.9
2.9
2.9
11.7
9.7
12.6
24.3
8.7
20.1
*defined by average tree size (dbh = diameter at breast-height) and average percent canopy cover
(CC).
205
13 January 2009
Table 6. Nest-site (1 ha (2.47 acres)) habitat characteristics collected using the Forest
Inventory and Analysis sampling protocol at California spotted owl nest sites (n = 80) on
the Plumas and Lassen National Forests, California, 2005-2006.
Variable
Mean
SE
Total Basal Area (ft2/acre)
260.8
6.47
# Trees >= 30 inch dbh (#/acre)
10.7
0.58
Basal Area Trees >= 30 inch dbh (ft2/acre)
96.0
5.70
# Trees >= 24 inch dbh (#/acre)
19.9
0.90
Basal Area Trees >= 24 inch dbh (ft2/acre)
131.7
6.29
# Trees <12 inch dbh (#/acre)
383.5
26.36
2
Basal Area Trees , <12 inch dbh (ft /acre)
50.1
2.71
# Snags >=15 inch dbh (#/acre)
7.4
0.80
Mean Duff Depth (inches)
3.0
0.16
Duff (tons/acre)
67.4
3.64
Mean Litter Depth (inches)
2.3
0.18
Litter (tons/acre)
23.7
1.81
1 Hour Fuels (tons/acre)
0.75
0.03
10 Hour Fuels (tons/acre)
4.0
0.21
100 Hour Fuels (tons/acre)
4.4
0.28
Shrub Cover (%)
7.7
1.16
Canopy Cover (%)*
64.1
1.24
* estimated through Forest Vegetation Simulator modeling of plot-based tree lists.
206
13 January 2009
Table 7. Distribution of USDA Region 5 vegetation classes (Mean (SE)) within 500 acre
(201 ha) circles centered on California spotted owl (CSO) territories (n = 102) and
systematic grid (Grid) points (n = 130) within the Plumas-Lassen Study on the Plumas
and Lassen National Forests, 2004-2006.
R5 Size
Class*
Non-forest
Total Size 1
2P & 2S
2N
2G
3P&3S
3N
3G
4P&4S
4N
4G
Total 4N &
4G
Total
Suitable
habitat
R5 Size Class Description
CSO
Grid
Sum of non-forest land types
Sum of 1G,1N, 1P, 1S: <6 inch dbh,
all %CC classes
6-12 inch dbh, 10-39% CC
6-12 inch dbh, 40-69% CC
6-12-24 inch dbh, >=70% CC
12-24 inch dbh, >10-39% CC
12-24 inch dbh, 40-69% CC
12-24 inch dbh, >=70% CC
>24 inch dbh, >10-39% CC
>24 inch dbh, 40-69% CC
>24 inch dbh, >=70% CC
Sum of 4N & 4G: >24 inch dbh, >=
40% CC
Sum of classes 3N, 3G, 4N, 4G =
>12 inch dbh, >40% CC
4.4 (1.0)
1.7 (0.3)
8.4 (1.2)
1.6 (0.3)
3.4 (0.4)
3.8 (0.6)
1.6 (0.5)
9.2 (0.8)
37.2 (2.4)
6.2 (1.0)
1.0 (0.3)
25.8 (2.0)
6.5 (0.1)
32.4 (2.3)
4.1 (0.5)
4.4 (0.9)
0.5 (0.1)
16.1 (1.3)
38.5 (1.8)
3.8 (0.7)
2.1 (0.4)
17.3 (1.6)
2.4 (0.8)
19.6 (1.8)
75.7 (2.19)
61.9 (1.75)
*defined by average tree size (dbh = diameter at breast-height) and average percent canopy cover
(CC).
207
13 January 2009
Table 8. Annual number of California spotted owls documented during the breeding
period (April-August) in SA-4 (Meadow Valley Project Area) between 2004-2008 on the
Plumas National Forest, California..
Year
2003
2004
2005
2006
2007
2008
Confirmed
Pairs
7
7
4
3
8
5
Unconfirmed
Pairs
0
0
1
1
0
0
Territorial
Singles
1
0
2
3
1
1
Total Territorial
Sites
8
7
7
7
9
6
208
13 January 2009
Table 9. Frequency of occurrence of species or taxonomic groups in California spotted
owl pellets collected on the Plumas National Forest, California, 2003-2006.
2003
(n=607)
2004
(n=813)
2005
(n=886)
2006
(n=540)
All Years
(n=2846)
Dusky Footed
Woodrat
0.4745
0.3924
0.4537
0.5259
0.4547
Northern Flying
Squirrel
0.4135
0.3653
0.3115
0.3037
0.3472
Peromyscus ssp.
0.2669
0.3370
0.3251
0.2685
0.3053
Other small
Mammals
0.2306
03407
0.3002
0.3556
0.3074
Other Large
Mammals
0.0626
0.1095
0.0542
0.0370
0.0685
Birds
0.0923
0.1230
0.1479
0.1093
0.1216
Insects
0.1351
0.1796
0.2077
0.1167
0.1669
Table 10. Percent biomass contribution of species or taxonomic groups in California
spotted owl diets based on pellet analysis on the Plumas National Forest, California,
2003-2006.
2003
(n=607)
2004
(n=813)
2005
(n=886)
2006
(n=540)
All Years
(n=2846)
Dusky Footed
Woodrat
48.03
37.87
42.33
53.51
44.75
Northern Flying
Squirrel
24.10
21.12
17.79
14.34
19.53
Peromyscus ssp.
5.15
7.47
6.19
4.27
6.01
Other small
Mammals
6.55
11.32
11.64
11.43
9.85
Other Large
Mammals
7.54
11.31
11.11
5.36
7.83
Birds
7.87
10.02
15.41
8.87
11.07
Insects
0.75
0.89
1.29
1.02
0.96
209
13 January 2009
Figure 1. (A) Location of CSO Survey Areas surveyed in 2004-2008. (B) Example of
original survey plot consisting of multiple Cal-Planning watersheds. (C) Example of
Primary Sampling Units for surveying for CSOs. See text and study plan for further
details .
210
13 January 2009
Figure 2. Distribution of California spotted owl territories within CSO survey plots
across the Plumas and Lassen National Forests, 2008.
211
13 January 2009
Figure 3. Monthly precipitation totals for Quincy, California, during January-May, 20042008 (data from Western Regional Climate Center).
Monthly Precipitation (Quincy, CA)
14
Precipitation (inches)
12
10
2004
8
2005
2006
6
2007
2008
4
2
0
January
February
March
April
May
212
13 January 2009
Figure 4. Distribution of California spotted owl (n = 103) nest sites by California Wildlife
Habitat Relationship (CWHR) database vegetation classes on the Plumas and Lassen
national Forests, California, 2004-2007. Descriptions of the CWHR classes are provided
in Table 5 within the text of this document.
30
25
# nests
20
15
Series1
10
5
0
BAR
3S
3MLT
3D
4P
4M
4MLT
4D
4DLT
5M
5D
6
Size/Density type of nest's veg polygon
213
13 January 2009
Figure 5. Percent suitable habitat (>=12 inch dbh trees with >=40% canopy cover) within
500 acre (201 ha) circles centered on California spotted owl (CSO, n = 102) and
systematic grid points (Grid, n = 130) on the Plumas and Lassen National Forests,
California, 2004-2007.
35
30
Percent of Observations
25
20
CSO
GRID
15
10
5
0
0-20
21-30
31-40
41-50
51-60
61-70
71-80
81-90
91-100
Percent Suitable Habitat (3N, 3G, 4N, 4G)
214
13 January 2009
Figure 6. Percent large tree habitat (R5 classes 4N &4G: >=24 inch dbh trees with
>=40% canopy cover) within 500 acre (201 ha) circles centered on California spotted owl
(CSO, n = 102) and systematic grid points (Grid, n = 130) on the Plumas and Lassen
National Forests, California, 2004-2007. Descriptions of R5 classes are provided in Table
7 within the text of this document.
25
Percent of observations
20
15
CSO
GRID
10
5
0
0%
1-10%
11-20%
21-30%
31-40%
41-50%
51-60%
61-70%
71-80%
81-90%
215
13 January 2009
Figure 7. Distribution of proposed Meadow Valley Project Area forest management
treatments and cumulative distribution of California spotted owl territorial sites between
2003-2008 in Survey Area-4 of the Plumas-Lassen Study, Plumas National Forest,
California.
216
13 January 2009
Figure 8. Annual summary distribution of California spotted owl territorial sites between
2003-2008 across Survey Area-4 (Meadow Valley Project Area) of the Plumas-Lassen
Study, Plumas National Forest, California.
217
13 January 2009
Figure 9. Distribution of pre- and post-fire California Wildlife Habitat Relationship
vegetation classes within the Moonlight-Antelope Complex fire areas 2008 on the Plumas
and Lassen National Forests, California.
70.0
60.0
Percent
50.0
40.0
Pre-Fire (%)
Post-Fire (%)
30.0
20.0
10.0
0.0
<= 3D 3M 3P 3S 4D 4M 4P 4S 5D 5M 5P 5S
2D
CWHR Class
218
13 January 2009
Figure 10. Maps of (a) pre-fire and (b) post-fire California Wildlife Habitat Relationship
vegetation classes within the Moonlight-Antelope Complex fire areas 2008 on the Plumas
and Lassen National Forests, California.
(a)
(b)
219
13 January 2009
Figure 11. Distribution of California spotted owls detected in 2008 within the MoonlightAntelope Complex Fire Area and a 1.6 km buffer and wildfire burn severity classes on
the Plumas and Lassen National Forests, California.
220
13 January 2009
Figure 12. Distribution of Barred and Sparred (Spotted-Barred hybrids) Owls between
1989-2008 within the HFQLG Project area.
221
13 January 2009
Figure 13. Cumulative (a) frequency of occurrence in pellets and (b) percent biomass
contribution of species or taxonomic groups in California spotted owl diets based on
pellet analysis across 2003-2006 on the Plumas National Forest, California.
(a)
Frequency of Prey - All Years
0.5
0.4
0.3
All Years
0.2
Insects
Other Large
Mammals
Woodrat
0
Peromyscus
0.1
(b)
Percent Biomass- All Years
45
40
35
30
25
20
15
10
5
Insects
Birds
Other Large
Mammals
Other Small
Mammals
Peromyscus
Flying
Squirrel
Woodrat
0
222
13 January 2009
Figure 14. Annual (a) frequency of occurrence in pellets and (b) percent biomass
contribution of species or taxonomic groups in California spotted owl diets based on
pellet analysis across 2003-2006 on the Plumas National Forest, California.
(a)
Prey Frequency By Year
0.6
0.5
2003
0.4
2004
0.3
2005
2006
0.2
Insects
Birds
Other Large
Mammals
Other Small
Mammals
Flying
Squirrel
Woodrat
0
Peromyscus
0.1
(b)
Percent Biomass by Year
55
50
45
40
35
2003
30
2004
25
2005
20
2006
15
10
Insects
Birds
Other Large
Mammals
Other Small
Mammals
Flying
Squirrel
Woodrat
0
Peromyscus
5
223