-
extracted text
-
COMMUNITY COMPOSITION AND
INFLUENCE OF FOREST STRUCTURE ON
BIRDS IN THE EVERGREEN STATE COLLEGE FOREST RESERVE
by
Jora Rehm-Lorber
A Thesis
Submitted in partial fulfillment
of the requirements for the degree
Master of Environmental Studies
The Evergreen State College
March 2009
i
© 2009 by Jora Rehm-Lorber. All rights reserved.
ii
This Thesis for the Master of Environmental Studies Degree
by
Jora Rehm-Lorber
has been approved for
The Evergreen State College
by
________________________
Alison Styring, PhD
Member of the Faculty
________________________
Dylan Fischer, PhD
Member of the Faculty
________________________
Joan C. Hagar, PhD
Research Wildlife Biologist, USGS
________________________
Date
iii
ABSTRACT
Community Composition and Influence of Forest Structure on Birds in The Evergreen
State College Forest Reserve
Jora Rehm-Lorber
The temperate rainforests of the Pacific Northwest support the highest abundances of
birds of any coniferous forest system in North America. Birds are indicators of
ecological health and provide a number of ecosystem services such as pollinating plants,
dispersing seeds and controlling insect and rodent populations. Many birds in the Pacific
Northwest are experiencing dramatic declines, especially within lowland temperate
rainforests which are under development pressure and may face ecological changes with
a warming climate. The objective of this thesis is to lay the foundation for
avian science endeavors at the Evergreen State College (TESC). In this study
I describe baseline bird population findings and their relationships to forest structure
and vegetation attributes measured in 44 permanent forest plots. TESC bird abundance
was estimated at 11.86 birds/ha representing 55 different species. Using community
ordination methods, significant differences were found in avian community structure
among forest types. Comparison of regression models suggested deciduous overstory was
the best predictor of overall bird abundance. Indicator species analysis revealed species
specific examples in relation to forest type. Snag decay stage diversity was negatively
related to avian diversity, but was not affected by attributes of DWD. Sapling biomass
had a positive relationship with avian diversity, but not abundance. While these findings
are supported with data from only one breeding season, long-term data collection will
help to test and evaluate the best predictors of bird abundance and diversity in this
ecosystem. Also described are the necessary field protocols, tools and research
considerations for the newly created Evergreen Avian Monitoring Program (EAMP) to
continue long-term monitoring efforts.
iv
Table of Contents
Page
Title Page……………………………………………………………………………….....i
Copyright Page……………………………………………………………........................ii
Approval Page……………………………………………………………………………iii
Abstract…………………………………………………………………………………...iv
Table of Contents………………………………………………………….........................v
List of Tables…………………………………………………………………………….vii
List of Figures…………………………………………………………………………...viii
Acknowledgements……………………………………………………….........................ix
Chapter One- Avian Community Composition and Abundance Estimates at the
Evergreen State College
Abstract……………………………………………………………………........................1
1.1 Introduction
1.11 The Evergreen Ecological Observation Network……………........................2
1.12 Overview on monitoring bird populations……...……………........................4
1.2 Methods
1.21 Study site………………………………………………………………….....8
1.22 Data collection……………………………………………….........................9
1.23 Statistical analysis……………………………………………......................11
1.3 Results
1.31 Abundance estimates……………………………………………………….12
1.32 Habitat densities and community composition…………..............................14
1.33 Priority species……………………………………………………………...14
1.34 Nesting species……………………………………………….......................15
1.4 Discussion……………………………………………………………………………16
1.5 Literature cited…..…………………………………………………………………...19
Tables…………………………………………………………………………………….22
Figures……………………………………………………………………………………24
Chapter Two- Influence of Forest Structure on Birds in a Lowland Puget Sound
Rainforest
Abstract…………………………………………………………………………………..31
2.1 Introduction…………………………………………………………………………..32
2.2 Methods
2.21 Study site……………………………………………………………………38
2.22 Avian data collection……………………………………………………….40
2.23 Forest structure and vegetation data collection……………………………..40
2.231 Habitat typing……………………………………………………...41
2.232 Trees……………………………………………….........................41
2.233 Snags………………………………………………........................42
2.234 Downed woody debris………………………………………….....43
2.235 Saplings……………………………………………………………43
2.236 Understory vegetation transects…………………………………...43
2.24 Data analysis………………………………………………………………..44
2.3 Results
2.31 Trees……..………………………………………………………………….45
2.32 Snags………………………...……….……………………………………..47
v
Table of Contents (Continued)
________Page
2.33 Downed Woody Debris……………………………………………………..48
2.34 Understory…………………………………………………………………..48
2.35 Indicator Species Analysis………………………………………………….48
2.4 Discussion…………………………………………………………………………....49
2.5 Literature cited……………………………………………………………………….56
Tables…………………………………………………………………………………….61
Figures……………………………………………………………………………………66
Chapter Three- Protocols and Research Recommendations for the Evergreen Avian
Monitoring Program
Abstract…………………………………………………………………………………..78
3.1 Introduction
3.11 Monitoring………………………………………………………………….79
3.12 Research and monitoring needs in Pacific Northwest forests........................81
3.2 Focal Species: demographic monitoring
3.21 Productivity and Survivorship……………………………………………...83
3.22 Nest searching ……………………………………………………………...85
3.3 Protocols
3.31 Point counts…………………………………………………………………86
3.32 Nest searching and monitoring……………………………………………..90
3.33 Mist netting and banding…………………………………………………...95
3.4 Monitoring nocturnal species………………………………………………………...96
3.5 Monitoring birds of the nearshore environment……………………………………..97
3.6 Recommendations for future projects……………………………………………….99
3.7 Education and outreach
3.71 Importance and need………………………………………………………101
3.72 Interpretation………………………………………………………………102
3.73 Field based classes and university collaboration…………………….........103
3.74 Banding workshops and community classes………………….…………...103
3.75 The Puget Sound Bird Observatory……………………………………….104
3.8 Conclusions………..…………………………………………………......................105
3.9 Literature cited…..………………………………………………………………….106
Tables…………………………………………………………………………………...110
Figures……………………………………………………………………......................111
Appendices
A. Avian species list (common and scientific names)…………………………………..x
B. Plant species list (common and scientific names)…………………………………..xii
C. Blank data sheets and examples for program use…………………………………..xiv
D. Supplemental tables and figures…………………………………………………….xv
vi
List of Tables
Chapter One
Table 1. 2008 breeding bird survey species specific density table.
Table 2. Breeding species confirmed by behavioral and direct nest observations.
Chapter Two
Table 1. Variables used in habitat community analyses to identify factors driving avian
abundance, richness and diversity in the Evergreen State College forest reserve.
Table 2. Pearson and Kendall correlations with NMS ordination axes (n=44).
Table 3. Significant indicator species of habitat type in the Evergreen State College forest
reserve.
Table 4. Habitat relationships derived from linear regression analysis for avian
community indices in the Evergreen State College forest reserve.
Table 5. Model comparison results for a) abundance, b) bird species richness, c) bird
diversity, for each habitat model examined in the Evergreen State College forest reserve.
Chapter Three
Table 1. List of confirmed breeding species.
vii
List of Figures
Chapter One
Figure 1. Aerial photo of TESC campus and surrounding lands.
Figure 2. Map of EEON permanent plot locations and forest types.
Figure 3a. 1939 Orthophoto of the western half of future TESC property showing
extensive land clearing.
Figure 3b. Western half of TESC property as it looks today with large scale forest
regeneration.
Figure 4. Density estimates for all species pooled for each habitat type (confidence
intervals of 95%).
Figure 5a. Density estimates (with 95% confidence intervals) for 9 bird species detected
in mixed conifer/deciduous forests of The Evergreen State College, Washington.
Figure 5b. Density estimates (with 95% confidence intervals) for 9 bird species detected
in mixed conifer/deciduous forests of Mount Rainier National Park, Washington.
Figure 6. Species area curve for forest birds sampled in EEON forest plots during the
2008 breeding bird survey.
Chapter Two
Figure 1. Map of EEON permanent forest plots showing the diversity of nine different
forest types throughout the reserve.
Figure 2a. 1939 Orthophoto of the western half of future TESC property showing
extensive land clearing.
Figure 2b. Western half of TESC property as it looks today with large scale forest
regeneration.
Figure 3. NMS Ordination of avian community composition for 4 different forest stand
types, Douglas-fir (
), mixed conifer ({), mixed conifer/mixed hardwood (),
hardwood () in The Evergreen State College forest reserve
Figure 4. Comparison of avian a) abundance b) species richness and c) diversity in four
different forest types of the Evergreen State College forest reserve.
Figure 5. Graphical representations of NMS ordinations of species specific composition
for 4 different forest stand types, Douglas-fir (
), mixed conifer ({), mixed
conifer/mixed hardwood (), hardwood () in the Evergreen State College forest
reserve, Olympia, WA.
Figure 6. Significant habitat relationships for community indices A) avian diversity B)
avian species richness C) avian abundance (log).
Chapter Three
Figure 1a. Arial photograph of the Evergreen State College’s organic farm showing
surrounding forested lands (Google Earth image, 2008).
Figure 1b. Evergreen State College’s organic farm banding site with experimental net
locations.
viii
Acknowledgements
This work would not have been possible without the collaboration and assistance of many
students and faculty of the Evergreen State College. I thank the undergraduate programs
of Introduction to Environmental Studies: Land (2006/07), Temperate Rainforests (2007),
and Field Ecology (2008). Various members of the Evergreen Ecosystem Ecology (E3)
lab assisted in data collection in the summer of 2008 and discussed chapter drafts. I am
indebted to the patience, humor and hard work of the 2008 EEON lab manager, Justin
Kirsch. Thesis advisors Alison Styring, Dylan Fischer and Joan Hagar gave their
invaluable input and expertise in the many disciplines this thesis covers. What intended
to be a 2 year masters program turned into nearly 3 and I am grateful to my family and
friends for their support, encouragement and patience. Funding for the Evergreen
Ecological Observation Network was provided by The Evergreen State College.
ix
Chapter One
Avian Community Composition and Abundance Estimates at the Evergreen State College
Abstract
The temperate rain forests of the Pacific Northwest support the highest abundances of
birds of any coniferous forest system in North America, however many of these bird
species are experiencing dramatic declines. The objective of this thesis is to lay the
foundation for avian science endeavors at the Evergreen State College (TESC) stemming
from the creation of the Evergreen Ecological Observation Network (EEON) which
monitors the campus’s temperate rainforest ecosystem. This chapter details landbird
conservation programs and baseline population findings for the Evergreen Avian
Monitoring Program (EAMP). In the spring and summer of 2008 I completed the first
comprehensive bird community study at the college. I estimated overall and species
specific densities for twelve common forest breeders, and compiled a complete yearround species list and a list of confirmed breeding species from observations throughout
the 2008 breeding season. These data serve as baseline information for the monitoring
program and as an educational tool for local ornithology at the college and for the
surrounding community. Overall, TESC bird density was estimated at 11.86 birds/ha for
55 detected species. Density estimates, diversity indices and species area curves will
guide the work of future avian monitoring and student research. Over the long term, this
scientific data will inform the college’s forest management objectives as the student body
continues to grow and land use decisions influence the forest reserve system.
1
1. 1 Introduction
1.11 The Evergreen Ecological Observation Network
The 1,033-acre second-growth rainforest surrounding the developed campus of
the Evergreen State College (TESC) is one of the largest in the south Puget Sound (Hall
et al. 1976). Other large forested areas with lowland temperate rainforests (i.e. Capitol
State Forest) are present but consist of patchwork landscapes and are managed for timber
and other resources by private and state entities (Franklin et al. 2002). Since the
acquisition of land by the college in 1968, the forest has been left largely unmanaged
aside from routine trail and road maintenance. A recently revised campus master plan
addresses land use principles for the college and emphasizes the role of its natural areas
to be preserved for recreation, cultural and educational development (Zimmer Gunsul
Frasca Architects 2008). The forested area acquired by the college is often referred to as
a “reserve” because it represents a forested island in an increasingly urbanized area
(Figure 1). Native habitats of coastal and lowland areas are heavily encroached upon by
urbanization and as the Puget Sound experiences rapid population growth, protected
mature and intact forests within the basin will provide vital wildlife habitat (Rich et al.
2004). Several large development companies have purchased existing areas of forested
land adjacent to college property in the last ten years and completed planned residential
communities (Figure 1). This suburban development has further reduced surrounding
forest habitat and increased the biological and cultural value of TESC’s undeveloped
areas.
The college’s forest reserve has existed as a unique field site in some capacity
since the 1970’s, providing students and faculty with a living laboratory right outside the
door. Other Pacific Northwest field sites are outside of the Puget Sound basin, often far
2
inland and at elevations well above sea level. Additionally, many lowland temperate
forests in the Pacific Northwest are young coniferous forests, managed in 50 year
rotations (Altman and Hagar 2007), making an unmanaged 80 year old lowland forest
increasingly rare and worthy of study.
Many students and faculty with a variety of educational and professional
backgrounds see the forest as an excellent place to incorporate environmental and
sustainable awareness into their curriculum. The interdisciplinary nature inherent to the
college’s philosophy has allowed for students to spend time in the forest through the
mediums of spiritual and cultural expression, art, humanities and science. Aside from the
positive affect the Evergreen forest has had on components of core level and introductory
interdisciplinary courses, field based work in the biological sciences on campus is
enhanced and continues to flourish as a result of the campus’s forest system (Greenberg
and Hartley 1998, Kennedy and Quinn 2001, Kazakova et al. 2007, International Canopy
Network 2008, Zimmer Gunsul Frasca Architects 2008).
Faculty in the biological sciences at the college saw the potential to create a long
term research network that would allow the college and collaborators to track temporal
changes occurring within the forest ecosystem during a time of climatic change. Through
a grant from the college, a team of faculty created the Evergreen Ecological Observation
Network (EEON) in 2005. Since its inception, the network has supported the work of
dozens of independent student projects, as well as faculty research and has served as a
learning tool for a variety of programs. Establishment of the network has been a
collaborative process involving many TESC students of all educational levels and several
dedicated individuals. After the establishment of 52 gridded study plots, a subset of 10
was intensively studied by undergraduates whom collected data on forest structure and
3
vegetation. Over the last two years student research projects involving aspects of forest
ecology have occurred in these plots while concurrent field work collected baseline data
on 37 additional plots. Today EEON consists of 47 working plots with complete forest
structure data on live trees, snags, downed-woody debris (DWD) and understory
vegetation.
The incorporation of other scientific disciplines into EEON was an important
component emphasized during the planning stages. The network was designed with the
college’s tradition of interdisciplinary study in mind and intends to support and facilitate
research and monitoring from a variety of disciplines. This collaboration will assist in
our understanding of ecological communities and processes, and expose students to many
disciplines outside their field of study. The objective of my study is to bridge the
disciplines of forest ecology and wildlife science in order to describe the status and
distribution of birds living in the forest reserve in relation to their habitat. In this chapter
I provide the first abundance estimates of TESC’s forest birds and describe the structure
and composition of the bird community. Birds were chosen as study subjects here
because they are easily observable, well studied and charismatic. Additionally, many
Pacific Northwest forest bird species are experiencing dramatic declines (Marzluff and
Sallabanks 1998, Donovan et al. 2002, Rich et al. 2004). This study will provide baseline
data for the creation of an avian monitoring program linked to EEON; a viable and
logical progression in developing the network’s breadth and scope.
1.12 Overview on monitoring bird populations
Many bird species serve as indicators of habitat quality, with changes in their
populations linked to changes in ecological health (Marzluff and Sallabanks 1998, Zack
2002, Rich et al. 2004). Understanding bird population dynamics is critical to
4
conservation efforts (Thomas 1996). Conservation of forest birds is important because
these species provide many ecological services ranging from controlling insect and
rodent populations to pollinating plants and dispersing seeds (Gill 2006). Birds that
primarily breed in forests face threats to reproduction success and increases in mortality,
both of which are now widely accepted in the scientific world to be linked to habitat loss
(Marzluff and Sallabanks 1998, Donovan et al. 2002, Norris and Pain 2002, Plummer
2002, Ruth et al. 2003). Migratory birds are particularly susceptible to habitat loss
because they require a diverse and geographically large range of habitats at different
stages in their life (Robbins et al. 1989, Donovan et al. 2002). For example, a
Neotropical migrant breeder of the temperate rainforest may require a multi-layered
forest canopy for breeding, dense riparian zones during migration, and dry deciduous
woodland for wintering, all of which are experiencing their own habitat degradation.
The temperate rainforests of the Pacific Northwest support the highest abundances
of birds of any coniferous forest system in North America (Altman 1999). Although
these forests support a large number of birds, populations are changing due to many
species experiencing dramatic declines (Sharp 1996, Plummer 2002, Sauer et al. 2006).
Another 1.4 million people are expected in Puget Sound by the year 2020 and with them
additional urban development (Lombard 2006). With increasing habitat loss throughout
the Puget Sound basin, there is an urgent need for monitoring programs to track changes
in wildlife populations and other ecological changes through time (Lombard 2006). The
creation of the Evergreen Avian Monitoring Program (EAMP) as part of EEON will
represent one of the only long-term monitoring efforts for landbirds of lowland temperate
rainforests outside of the national park system (Siegel et al. 2004, Wilkerson et al. 2005).
The protected lands owned by TESC coupled with the stability of ongoing scientific
5
research at the college provide an excellent opportunity to establish a multifaceted
monitoring program combining the essential components utilized by other well known
monitoring programs around the country and standardized to achieve comparisons among
locations and projects (Ralph et al. 1995).
One of the largest and most comprehensive of these monitoring programs is the
North American Breeding Bird Survey (BBS). Currently the BBS comprises over 4,100
routes across North America and provides estimates of population trends for 420 bird
species (Sauer et al. 2008). Despite their widespread coverage, BBS routes occur only
along roads and collect only relative abundance data to generate population trends.
Estimates of population size or absolute abundance estimates are not possible with BBS
population indices data. In general, these population indices are seldom comparable
among species and monitoring programs. Thomas (1996) suggests BBS methods can be
applied in areas with more intensive studies underway, in an attempt to quantify observer
differences and estimate variations in detectability. Using methods comparable to the
BBS, EAMP will eventually allow for a more detailed description of bird populations in
the south Puget Sound area with absolute abundance estimates related to many habitat
characteristics.
Another major player in the development of landbird conservation plans for North
America was the creation of Partners in Flight (PIF) in 1990. The partnership, which
represents private, non-profit and public organizations aims to a) help at -risk species
before they become imperiled, stemming from the view that conservation implementation
is most effective before populations reach crisis levels, b) keep common birds common
by monitoring populations, and c) achieve bird conservation objectives by advocating for
“combining, coordinating and increasing” voluntary resources (Rosenberg 2004).
6
In 2004, landbird conservation priorities were synthesized in the Partners in Flight North
American Landbird Conservation Plan (Rich et al. 2004). State reports written by PIF
regional coordinators outline each state’s priority species, their population objectives and
numerical targets, and divide population estimates for each species into Bird
Conservation Regions (BCR’s) and primary breeding habitats.
Individual species assessments are based on the PIF North American Species
Assessment Database which utilizes BBS data. Because BBS methodologies are designed
to cover large areas with limited resources, BBS trends for many species are lacking or
have low precision (Rich et al. 2004). A very limited amount of alternative data exists to
supplement BBS routes and therefore many species are lacking population trend (PT)
scores (a score of 1-5 from large population increases to large population declines).
The future work of EAMP may assist updates to PIF state documents, including
the implementation of new population trend scores. EAMP monitoring protocols and
research plans will aim to augment the power of the BBS and work to test hypotheses
about causes of population change in priority species. In the longer term, EAMP may
provide comparative data on the mechanisms influencing landbird responses to
conservation implementation. Priority species and population objectives for birds
detected during EAMP 2008 surveys are described in appendix D.
Once established, EAMP can mirror the work of other nationally recognized
organizations with large scale and long term monitoring efforts currently underway, with
a focus on birds of the Puget Sound lowland rainforests. By following the standard
protocols of other monitoring programs, EAMP can engage in data sharing to link bird
populations here on campus to a larger continental or global context. Increasingly, avian
biologists and conservationists are emphasizing the imperativeness of collaborative
7
science when attempting to understand declines in bird populations. Data sharing
networks such as the Avian Knowledge Network (ANK) were created to allow scientists
and citizens alike to access and provide information about bird populations (Avian
Knowledge Network 2008). Eventually EAMP will have its own AKN node with public
access available to other monitoring programs, organizations, students and citizens.
The objective of this thesis is to lay the foundation for avian science endeavors at
the Evergreen State College (TESC) stemming from the creation of the Evergreen
Ecological Observation Network (EEON). In this chapter I present the first
comprehensive data on breeding bird species of the TESC forest reserve and generate
reliable density estimates for the most common breeding species using distance
methodologies. To provide a context and stimulate further discussion and research, I
present these results in relation to conservation objectives and compare these first year
estimates to other density data from Pacific Northwest rainforests in Western
Washington.
1.2 Methods
1.21 Study site
The Evergreen State College lies southwest of Olympia, WA, Thurston County
(approx. 47°04’N, 122°58’W). The area receives an average of 130 cm of precipitation
per year, with nearly half the days in a year receiving substantial rainfall (Barrier and
Froyalde 1999). Average annual temperatures range from 3.9 to 15.6 degrees Celsius.
The campus is relatively flat with near sea level elevation (highest point is 74 meters).
The forest reserve is representative of a coastal temperate rainforest in the Western
Hemlock Zone, characterized by high productivity and complex forest structure (Franklin
and Dyress 1973). The forest itself is a mosaic of dominant stands of Douglas fir
8
(Pseudotsuga menziesii), codominant mixed hardwood and conifer stands, with red alder
(Alnus rubus), big-leaf maple (Acer macrophyllum) and scattered cottonwood (Populus
trichocarpa), nearly pure stands of red alder in early succession areas and wetter conifer
areas consisting of western red cedar (Thuja plicata) and western hemlock (Tsuga
heterophylla) associations (Figure 2). Douglas-fir forests, which dominate western
Oregon and Washington, support the highest bird densities of any coniferous forest
systems in North America (Wiens 1975). They also have been some of the most
intensively managed forests in the world, and are home to several high profile
endangered species such as the Northern Spotted Owl and Marbled Murrelet (Rich et al.
2004).
The topography of campus is generally flat with the highest elevation at 74 meters
above sea level. The landscape is characterized by gentle slopes bisected by five steepsloped, short (mostly ephemeral) streams with headwaters within, or just outside of the
reserve, and steep bluffs along the waterfront of Eld Inlet. My work took advantage of a
network of 10-meter radius permanent plots established in 2006 using a systematic
random grid with plots 250 meters apart across the reserve
(http://academic.evergreen.edu/projects/EEON).
1.22 Data collection
I conducted five minute variable circular plot (VCP) point counts from 23 April to
22 June 2008 at 47 permanent plots. I visited each plot once in the early season and once
in the late season to equally sample early nesting resident species and late nesting
migratory species. I recorded the horizontal distance to the nearest meter for each bird
detected with the aide of laser rangefinders and flagging. I recorded all birds seen and
heard, excluding birds that flew overhead and did not appear to be utilizing the habitat.
9
I conducted a pilot study from 17 March to 22 April 2008 using several VCP
analysis methods to assess forest songbird abundances (Reynolds et al. 1980, Ralph et al.
1995, Bibby et al. 2000). The pilot study offered an opportunity to practice point count
surveys, initially locate plots to save time during actual surveys, and identify the best
available protocol based on field work effort balanced with statistical power. I conducted
multiple surveys at each of the 47 permanent plots using a) the methods of surveys
conducted in June of 2006 which utilized 5 continuous 1 minute counts to increase the
number of sample units, and b) method of one 5 minute count divided into the first 3
minutes and the last 2 to allow for comparisons of data to the Breeding Bird Survey
(BBS). In the first 3 minutes all birds heard and seen are recorded and in the last 2
minutes only new individuals not previously heard or seen are recorded. The 2008 pilot
data was entered into Distance 5.0 (Thomas et al. 2005) to draw statistical comparisons of
avian abundance estimates for each of the above methods (see 1.24 statistical analysis).
Interestingly, combined three and two minute counts yielded slightly lower AIC values
and better model fits with truncation to 150 meters. Given the increased workload of
conducting one minute counts and the skill required to record all bird heard and seen
within a very short time period, five minute counts were selected for this study. In the
future other students may wish to try different methods in collaboration with five minute
counts, such as double sampling (Bart and Earnst 2002, Collins 2007) and double
observer methodologies (Forcey et al. 2006, Kissling and Garton 2006) .
Each 5 minute survey was separated into 3 and 2 minute periods, beginning
within 30 minutes of local sunrise and concluding 3 hours after sunrise (Ralph et al.
1995). Separating observations into the first three minutes and last two minutes improves
comparability with BBS routes which utilize three minute counts. All other bird species
10
not associated with the habitat were listed separately. Environmental data including
cloud cover, temperature, wind, precipitation and noise level (scale of 0-3) were also
recorded. Surveys were suspended due to high winds (>10mph) or precipitation which
penetrated the forest canopy. These environmental variables influence an observer’s
ability to detect birds and can also be tracked over the long-term to reveal potential
causes influencing bird populations. The level of road or construction noise can influence
detection rates because detections in dense forested habitats are generally greater than
90% aural (Ralph et al. 1995).
1.23 Statistical analysis
I estimated avian abundances using program DISTANCE (Thomas et al. 2005)1.
The program uses maximum likelihood to calculate a detection function based on
distance from the observer and uses this function to estimate densities per hectare
(Buckland et al. 2001). I used the half normal cosine model to estimate densities from 94
point counts at 47 plots. I used a combination of three criteria to select a final model. The
Akaike information criterion (AIC) is a method for determining model fit, where lower
AIC values yield better fits. I also used goodness of fit and Kolmogorov-Smirnov tests
which determine if the modeled and real datasets differ significantly. For each model, I
stratified by species and habitat to estimate species specific and pooled avian densities in
1
Among the wide array of methods for monitoring avian populations two prominent methodological approaches are widely used.
The first is the use of population indices generated by fixed radius point counts in which all birds are recorded that fall within a certain
radius (usual 50 meters in forested habitats) around the observer regardless of the bird’s actual distance (Hutto et al. 1986). The
statistical analyses of relative abundance data has historically yielded significantly different population trends for the same species
(Link et al. 1994, Sauer et al. 1994) and there is currently no consensus on what statistical methodologies best model actual population
trends (Thomas 1996). Distance sampling was developed in response to the widespread use of population indices. In distance
sampling the observer records the distance to each bird detected with the central premise that birds are harder to detect the further they
are from the observer. Actual observations are then modeled to account for individuals present but not detected during a survey
(Thompson 2002). Distance sampling has become the standard in much ornithological research and increasingly, monitoring projects
(Ellingson and Lukacs 2003, Siegel et al. 2004, Wilkerson et al. 2005). The debate on whether to use indices or distance methods
continues with many alternative approaches proposed in recent years to account for observer and measurement biases and address the
issues of conflicting results in population trends and management strategies (Nichols et al. 2000, Bart and Earst 2002, Bart et al. 2004,
Forcey et al. 2006, Kissling and Garton 2006, Collins 2007). In this study I use distance methods as a logical starting point, with data
collection allowing for transformation to index methods. This method also allows for the incorporation of other methodologies (i.e.
double-observer and double-sampling) to be tested by EAMP in the future.
11
order to extrapolate habitat differences. Forest types were ascertained through a student
research project which utilized aerial photography and GIS technology (Greenberg and
Hartley 1998). Greenberg and Hartley’s (1998) original work consisted of nine distinct
forest types (Figure 2). I simplified the forest typing classification system used in 1998
to help elucidate any possible differences in habitat selection for avian species detected in
these habitats. I separated plots into either a) Douglas-fir, b) maple, c) alder, d) mixed
conifer, e) mixed conifer/deciduous, f) mixed deciduous. To explore avian community
composition I calculated species richness, species evenness and Shannon’s and
Simpson’s diversity indices using program PCORD (McCune and Mefford 1999).
1.3 Results
1.31 Abundance estimates
During surveys I detected 2013 individual birds of 55 species (Appendix A).
Densities were estimated for 12 species with over 60 detections (Table 1) based on
recommendations by (Buckland et al. 2001). In order to ascertain reasonable estimates of
abundance with good model fits, distance sampling protocols recommend a relatively
high number of detections per a species. This means only common and easily detectable
species are considered here.
Five of the 12 species with density estimates are Neotropical migrants, whom
come to our forest each summer to breed and return each fall to various locations
throughout the Caribbean and central America. These species include Pacific-slope
Flycatcher (Empidonax difficilis), Swainson's Thrush (Catharus ustulatus), Blackthroated Gray Warbler (Dendroica nigrescens), Wilson's Warbler (Wilsonia pusilla), and
Western Tanager (Piranga ludoviciana) (Table 1). Two other species with density
estimates were American Robin (Turdus migratorius) and Purple Finch (Carpodacus
12
purpureus) which exhibit food driven migration patterns, moving to different areas of
Washington throughout the year (Wootton 1996, Sallabanks and James 1999). The
remaining five, Chestnut-backed Chickadee (Poecile rufescens), Red-breasted Nuthatch
(Sitta Canadensis), Winter Wren (Troglodytes troglodytes), Spotted Towhee (Pipilo
maculates), and Song Sparrow (Melospiza melodia) are resident species that spend their
entire lives in a local area. Resident species which maintain small territories and are
easily observable make good candidates for year round and over-wintering monitoring
efforts (see chapter three).
I qualitatively compared TESC abundance estimates to those of the same 12
species in mixed conifer/deciduous forests of Mount Rainier National Park (MRNP).
These data were collected with the same point count protocol during the breeding season
of 2003-2004 (Wilkerson et al. 2005). The extensive breeding bird counts which occur in
MRNP each year occur in all representative habitats, at all elevations throughout the
park. Many habitat types overlapped those found in TESC forests, however mixed
conifer/deciduous habitat provided the best overall representation of elevation and
topography of sites within MRNP. Abundance estimates were quite similar between the
two locations with TESC forests supporting slightly higher densities of all species except
Pacific-slope Flycatcher (Empidonax difficilis) and Chestnut-backed Chickadee (Poecile
rufescens) (Figure 5). MRNP had dramatically fewer points with detections and fewer
non-flyover detections overall for each species than those at TESC which influenced
statistical confidence in density estimates within the mixed conifer/deciduous habitat
(Table 1). Overall abundance in mixed conifer/deciduous forests at MRNP (all species
pooled) was estimated at 9.19 birds/ha (n=25). Overall TESC bird abundance estimates
(all species pooled) for 2008 were 11.86 birds/ha (n=47).
13
1.32 Habitat densities and community composition
Each 10-meter radius EEON plot represents a random sample of the surrounding
forest type since all plot locations were established randomly (see methods). In general,
the plot itself is representative of the forest habitat around the plot in which birds were
detected. Although our forests represent the broad habitat of Douglas-fir and mixed
conifer/hardwood, microhabitats exist in each plot, influencing bird abundances and even
detection probabilities (Ralph et al. 1995). Densities range from 10.64 (ind./ha) in mixed
hardwood plots to 13.37 (ind./ha) in pure alder plots (Figure 6). Density estimates for
hardwood dominated habitats are 12.15 (ind./ha) and 11.55 (ind./ha) for conifer
dominated habitats.
Measurements of diversity including species richness, species evenness and
Shannon’s and Simpson’s indices revealed substantial differences among plots but little
pattern relating to forest type (Appendix D). Mean species richness was 16.5 species per
plot. In general, plots with the highest species richness were either riparian, seasonally
wet, or had dominant or pure deciduous overstories. Appendix D provides an overview
of community composition among plots and areas of the TESC forest reserve. There
appears to be no substantial difference among the south, west, north and east portions of
the reserve. Chapter two explores patterns in bird-habitat relationships in closer detail.
1.33 Priority species
Four of the twelve species with density estimates are listed under the PIF North
American Landbird Conservation Plan as Tier IIA (see Appendix D) with high regional
concern within the Southern Pacific Rainforest Bird Conservation Region (Rosenberg
2004). Based upon BBS data Pacific-slope Flycatcher (Empidonax difficilis), Chestnutbacked Chickadee (Poecile rufescens), Black-throated Gray Warbler (Dendroica
14
nigrescens) and Purple Finch (Carpodacus purpureus) are experiencing declines in the
core of their ranges and require conservation action to reverse or stabilize trends (Altman
1999, Rich et al. 2004). These are species with a combination of high area importance
and declining (or unknown) population trends (Appendix D).
Aside from these four species that will require diligent monitoring and
conservation action in the near future, there are 13 other PIF priority species that I
detected of during the 2008 breeding bird survey (Appendix D). These species will likely
have reliable density estimates in 2009 and we can begin to estimate our own localized
trends within the next five years. Priority species with known population objectives and
with reliable abundance estimates at TESC will make for important focal species work on
campus (Appendix D). This information will be useful for planning conservation actions.
Species in need of additional information on habitat use, reproduction or behavior to
order to achieve conservation goals are described in greater detail under research project
recommendations in chapter three.
1.34 Nesting species
Although detection of a singing bird is a common indirect method for confirming
a breeding species, obtaining any direct breeding evidence is advisable. Many songbird
species sing on migration grounds or are detected by calls or movement. While
conducting breeding bird surveys, or while moving between survey stations, an observer
is likely to witness breeding activity in a variety of forms. I kept a detailed log of all
confirmed breeders where breeding behavior other than or in addition to singing was
observed. Occasionally nests were located and nest cards following Cornell Lab of
Ornithology protocols were created (Cornell Lab of Ornithology 2006). More often
behavior indicating a bird with a nest was observed such as an adult carrying food or
15
nesting material, or an adult feeding fledglings or juvenile birds. Locating nests in this
fashion is not a quantitative process without a systematic approach, but qualitative and
behavioral observation as well as detailed note keeping can aid in the development of
nest searching objectives and activities in the future. I confirmed 28 breeding species,
nine of which are cavity nesting species requiring snags or decaying portions of live trees
for nesting (Table 3).
1.4 Discussion
Our forest reserve hosts a variety of migrant bird species which rely on forested
habitats for successful breeding. The forest ecosystem hosts significant populations of
Neotropical migrants, including, Pacific-slope Flycatcher (Empidonax difficilis),
Swainson’s Thrush (Catharus ustulatus), Black-throated Gray Warbler (Dendroica
nigrescens), Wilson’s Warbler (Wilsonia pusilla) and Western Tanager (Piranga
ludoviciana). Several of these species are experiencing regional or continental declines
(Rich et al. 2004, Rosenberg 2004) and should be monitored for any population changes
through breeding surveys, nest searching and mist netting efforts.
The densities of 7 species of birds I estimated were slightly higher than those
recorded for the same species in other unmanaged forests of Washington by the same
sampling methods (Siegel et al. 2004, Wilkerson et al. 2005) but it will take several more
years to confirm this trend because confidence intervals for both locations overlapped. If
this is in fact the case, it is unknown if factors or a combination of factors, such as
proximity to salt water, elevation or a suburban interface are influencing these densities.
First year, baseline data on bird populations always present limitations for inference.
Less than 25 % of species detected during the 2008 survey generated reliable density
estimates.
16
Despite this small number of species specific density estimates, generation of
species area curves revealed an adequate sample size for the study area (Figure 6).
Buskirk and McDonald (1995) found multiple counts per point over the course of one
breeding season to yield improved coverage in forested habitats, and probably influenced
the results in this study. If time permits, 3 visits to each point during the breeding season
is ideal (Buskirk and McDonald 1995, Ralph et al. 1995).
Detections vary greatly among species, depending on degree of vocalization and
life history strategy and time of breeding (Buskirk and McDonald 1995), as well as
among observers (Buckland et al. 2001). Species with large territory sizes such as (Bandtailed Pigeon (Patagioenas fasciata) and Pileated Woodpecker (Dryocopus pileatus) are
difficult to estimate abundances for because they are not likely to be recorded frequently.
These species will either require several seasons of point counts to generate reliable
density estimates, or a larger survey area encompassing multiple home ranges.
Additionally, a greater amount of survey time is needed to confirm the presence or
absence of some species (Ralph et al. 1993, Buskirk and McDonald 1995). In the future,
with the addition of banding and nest searching monitoring tools to the monitoring
program, the number of species EAMP can include in analyses of population data will
increase.
The use of VCP point counts require more training and attention to
standardization than other methods of bird surveys such as fixed radius point counts, area
searches, or spot-mapping (Reynolds et al. 1980, Ralph et al. 1993). Abilities to
accurately estimate distances can vary greatly among observers and concern regarding
observer bias and data accuracy has been voiced in the literature (Hutto and Young
2002;2003). Over 90 percent of detections during the 2008 survey at TESC were aural,
17
indicating a need for training materials and bird song identification proficiency to become
a major component of the breeding bird survey. With skilled observers training
newcomers from year to year, the continued sustainability of EAMP and the replication
and comparability of results to other monitoring programs and regions should remain the
top priority. During the planning stages, it is likely experimental and pilot work will
continue to shape the direction and protocols of the monitoring program.
Although much troubleshooting and pilot work will need to continue over the next
few years, the completion of the first season of EAMP and the collection of initial count
data is a major step forward in the initial phases of the program. Long-term population
monitoring is essential to detect biologically significant changes to a population (Hutto
and Young 2002). Reliable population trends and density estimates for a number of
species will allow faculty and students involved in EAMP to develop meaningful and
testable hypotheses for future research. Scientific information about avian populations
and habitats on campus should be effectively disseminated to not only managers and
planners of the college, but applied to regional policy and management decisions
influencing temperate rainforest habitats.
18
Literature Cited
Adair, C. unpublished. Campus Orthophoto. The Evergreen State College, Olympia, WA.
Altman, B. 1999. Conservation strategy for landbirds in coniferous forests of western
Oregon and Washington. Oregon-Washington Partners in Flight.
Altman, B., and J. C. Hagar. 2007. Rainforest Birds: A Land Manager’s Guide to
Breeding Bird Habitat in Young Conifer Forests in the Pacific Northwest. U.S.
Geological Survey.
Avian Knowledge Network. 2008. Avian Knowledge Network: An online database of
bird distribution and abundance. Ithaca, New York. Accessed: [August 12, 2008]
Barrier, J., and R. Froyalde. 1999. Thurston County Profile. Thurston County Regional
Planning Council.
Bart, J., and S. Earnst. 2002. Double sampling to estimate density and population trends
in birds. Auk 119:36-45.
Bibby, C. J., N. D. Burgess, D. A. Hill, S. Mustoe, and S. Lambton. 2000. Bird Census
Techniques. 2nd edition. Academic Press, San Diego, California.
Buckland, S. T., D. R. Anderson, K. P. Burnham, J. L. Laake, D. L. Borchers, and L.
Thomas. 2001. Introduction to Distance Sampling: Estimating Abundance of
Biological Populations. Oxford University Press, Oxford, UK.
Buskirk, W., and J. L. McDonald. 1995. Comparison of point count sampling regimes for
monitoring forest birds. United States Forest Service General Technical Report.
Collins, B. T. 2007. Guidelines for using double sampling in avian population
monitoring. Auk 124:1373-1387.
Cornell Lab of Ornithology. 2006. A Nest Egg for Nest Watching. in Birdscope.
Donovan, T. M., C. J. Beardmore, D. N. Bonter, J. D. Brawn, R. J. Cooper, J. A.
Fitzgerald, R. Ford, S. A. Gauthreaux, T. Luke George, W. C. Hunter, T. E.
Martin, J. Price, K. V. Rosenberg, P. D. Vickery, and T. B. Wigley. 2002. Priority
research needs for the conservation of neotropical migrant landbirds. Journal of
Field Ornithology 73:329-339.
Forcey, G. M., J. T. Anderson, F. K. Ammer, and R. C. Whitmore. 2006. Comparison of
two double-observer point-count approaches for estimating breeding bird
abundance. Journal of Wildlife Management 70:1674-1681.
Franklin, J. F., and C. T. Dyress. 1973. Natural vegetation of Oregon and Washington.
Oregon State University Press, Corvallis, Oregon.
Franklin, J. F., T. A. Spies, R. V. Pelt, A. B. Carey, D. A. Thornburgh, D. R. Berg, D. B.
Lindenmayer, M. E. Harmon, W. S. Keeton, D. C. Shaw, K. Bible, and J. Chen.
2002. Disturbance and structural development of natural forest ecosystems with
silvicultural implications, using Douglas-fir forests as an example. Forest Ecology
and Management 155:39-423.
Gill, F. B. 2006. Ornithology. 3rd edition. W.H. Freeman, New York.
Greenberg, K., and A. Hartley. unpublished. Forest canopy cover typing. The Evergreen
State College, Olympia, WA.
Hall, D. J., D. H. Lockwood, and C. I. Lomax. 1976. Campus inventory and land use
planning. The Evergreen State College Press.
Hutto, R. L., and J. S. Young. 2002. Regional landbird monitoring: perspectives from the
Northern Rocky Mountains. Wildlife Society Bulletin 30:738-750.
19
_____. 2003. On the design of monitoring programs and the use of population indices: a
reply to Ellingson and Lukacs. Wildlife Society Bulletin 31:903-910.
International Canopy Network. 2008. The Forest Canopy Lab at the Evergreen State
College. Olympia, WA. Accessed: [November 10, 2008]
Kazakova, A., J. Kirsch, and D. Fischer. unpublished. First year estimate in variability of
productivity and diversity in the lowland Puget Sound permanent plot network.
The Evergreen State College, Olympia, WA.
Kennedy, P. G., and T. Quinn. 2001. Understory plant establishment on old-growth
stumps and the forest floor in western Washington. Forest Ecology and
Management 154:193-200.
Kissling, M. L., and E. O. Garton. 2006. Estimating detection probability and density
from point-count surveys: a combination of distance and double-observer
sampling. Auk 123:735-752.
Lombard, J. 2006. Saving Puget Sound: A conservation strategy for the 21st century.
American Fisheries Society, Bethesda, Maryland.
Marzluff, J. M., and R. Sallabanks, editors. 1998. Avian conservation : research and
management. Island Press, Washington, D.C.
McCune, B., and M. J. Mefford. 1999. PC-ORD Multivariate analysis of ecological data
version 4. MjM Software Design, Gleneden Beach OR.
Norris, K., and D. J. Pain, editors. 2002. Conserving bird biodiversity: general principles
and their application. Cambridge University Press, New York, NY
Plummer, T. 2002. Reversing Avian Population Declines in Northwest Forests. in
Partners In Flight Oregon/Washington Chapter Newsletter.
Ralph, C. J., S. Droege, and J. R. Sauer. 1995. Managing and monitoring birds using
point counts: standards and applications.
Ralph, C. J., G. R. Geupel, P. Pyle, T. E. Martin, and D. F. DeSante. 1993. Handbook of
field methods for monitoring landbirds.
Reynolds, R. T., J. M. Scott, and R. A. Nussbaum. 1980. A variable circular-plot method
for estimating bird numbers. Condor 82:309-313.
Rich, T. D., C. Beardmore, H. H. Berlanga, P. Blancher, M. Bradstreet, G. Butcher, D.
Demarest, E. Dunn, C. Hunter, E. Inigoelias, J. Kennedy, A. Martell, A. Panjabi,
D. Pashley, K. Rosenberg, C. Rustay, S. Wendt, and T. Will. 2004. Partners in
flight North American landbird conservation plan. Partners in Flight.
Robbins, C. S., J. R. Sauer, R. S. Greenberg, and S. Droege. 1989. Population declines in
North American birds that migrate to the neotropics. Proc. Natl. Acad. Sci.
86:7658-7662.
Rosenberg, K. V. 2004. Partners in flight continental priorities and objectives defined at
the state and bird conservation region levels: Washington. Partners in Flight.
Ruth, J. M., D. R. Petit, J. R. Sauer, M. D. Samuel, F. A. Johnson, M. D. Fornwall, C. E.
Korschgen, and J. P. Bennett. 2003. Science for avian conservation: priorities for
the new millennium. Auk 120:204-211.
Sallabanks, R., and F. C. James. 1999. American Robin (Turdus migratorius), in Poole,
A., and Gill, F., eds., The Birds of North America, No. 462. The Birds of North
America Online
Sauer, J. R., J. E. Hines, and J. Fallon. 2006. The North American breeding bird survey,
results and analysis 1966-2006. USGS Patuxent Wildlife Research Center,.
20
_____. 2008. The North American breeding bird survey, results and analysis 1966-2006.
USGS Patuxent Wildlife Research Center.
Sharp, B. E. 1996. Avian population trends in the Pacific Northwest. Bird Populations
3:26-45.
Siegel, R. B., R. L. Wilkerson, and S. Hall. 2004. Landbird Inventory for Olympic
National Park (2002 - 2003) Final Report. Report Cooperative Agreement
H9471011196.
Thomas, L. 1996. Monitoring long-term population change: why are there so many
analysis methods? Ecology 77:49-58.
Thomas, L., J. L. Laake, S. Strindberg, F. F. C. Marques, S. T. Buckland, D. L. Borchers,
D. R. Anderson, K. P. Burnham, S. L. Hedley, J. H. Pollard, J. R. B. Bishop, and
T. A. Marques. 2005. DISTANCE. Research Unit for Wildlife Population
Assessment, , University of St. Andrews, United Kingdom.
Wiens, J. A. 1975. Avian communities, energetics, and functions in coniferous forest
habitats. USDA For. Serv. Gen. Tech. Rept. WO-1.
Wilkerson, R. L., R. B. Siegel, and J. Schaberl. 2005. Landbird Inventory for Mount
Rainier National Park (2003-2004) Final Report. Report Cooperative Agreement
H9471011196.
Wootton, J. T. 1996. Purple Finch (Carpodacus purpureus). The Birds of North America
Online in Poole, A., and Gill, F., eds., The Birds of North America, No. 208.
Zack, S. 2002. The oak woodland bird conservation plan: a strategy for protecting and
managing oak woodland habitats and associated birds in California (S. Zack, lead
author). in California Partners in Flight, editor. Version 2.0. Point Reyes Bird
Observatory Stinson Beach, CA.
Zimmer Gunsul Frasca Architects. 2008. Campus Master Plan. The Evergreen State
College.
21
Tables
22
23
Figures
Figure 1. Aerial view of TESC campus outlined in yellow. Residential
development has reduced forest cover surrounding the campus, as exemplified
in the area northeast of the forest reserve.
24
Figure 2. Map of EEON permanent forest plots and forest types
25
Figure 3a.1939 Orthophoto of the western half of future TESC property showing
extensive land clearing (scanned and prepared by C. Adair)
26
Figure 3b. Western half of TESC property as it looks today with large scale forest
regeneration and an average forest age of 80 years.
27
Figure 4. Density estimates for all species pooled for each habitat type (confidence intervals of 95%).
The average bird density in hardwood habitats was 12.15 birds/ha, while the average in conifer forests was 11.55 birds/ha.
28
Figure 5a. Density estimates (with 95% confidence intervals) for 9 bird species detected
in mixed conifer/deciduous forests of The Evergreen State College, Washington.
Figure 5b. Density estimates (with 95% confidence intervals) for 9 bird species detected
in mixed conifer/deciduous forests of Mount Rainier National Park, Washington. Bird
species acronyms are defined in Appendix A.
29
Figure 6. Species area curve for forest birds sampled in EEON forest plots during the
2008 breeding bird survey. Dashed lines indicate 95% confidence bands. The number of
species detected (54) begins to level off around 30 subplots.
30
Chapter Two
Influence of Forest Structure on Birds in a Lowland Puget Sound Rainforest
Abstract
Birds are indicators of ecosystem health and provide a number of ecosystem services
such as pollinating plants, dispersing seeds and controlling insect and rodent populations.
In this study I explore the relationship between a bird community in a lowland, secondgrowth temperate rainforest and the forest structure and vegetation attributes measured in
44 permanent forest plots. I hypothesized that 1) the bird community would differ
significantly among habitat types and 2) several structural attributes of the forest would
be important predictors of bird abundance and diversity. I predicted positive avian
community responses to 1) abundance of deciduous trees 2) increases in the decay stage
and class diversity of course woody debris (snags and downed woody debris (DWD) and
3) increases in understory plant species richness, cover and sapling biomass. Community
ordination revealed significant differences in avian community structure among habitat
types and comparison of regression models suggested deciduous overstory was the best
predictor of total bird abundance. Sapling biomass had a positive relationship to avian
diversity, but not abundance. Indicator species analysis revealed species specific
examples in relation to habitat type. While these findings are supported with data from
only one breeding season, long-term data collection will help to test and evaluate the best
predictors of bird abundance and diversity in Pacific Northwest lowland temperate
rainforests.
31
2.1 Introduction
Structural attributes of forest stands are recognized as important to understanding
and managing forest ecosystems. Structure is the attribute most often manipulated in
management, is a readily measured surrogate for functions that are otherwise difficult to
measure directly, and has direct values for products or ecosystem services (Franklin et al.
2002). In the Pacific Northwest, there is a widely recognized need to both retain existing
coniferous old-growth forest and allow young and mature forests to develop structural
attributes of old-growth (Ruggerio et al. 1991, Altman 1999). Due to their structural
complexity, the western hemlock forests of Oregon and Washington that constitute the
majority of coniferous forests, are a high priority for regional avian conservation plans
(Altman 1999, Rich et al. 2004, Rosenberg 2004). The coniferous forests of TESC have
been unmanaged for over 40 years and are transitioning from young to mature with stand
ages around 80 years old (Spies and Franklin 1991). The younger forests preserved by
TESC will provide recruitment into old-growth status over the next several decades,
contributing to wildlife habitat and regional bird conservation efforts.
Mature coniferous forests support many avian species who rely on their multilayered structural complexity for successful breeding (Rosenberg 2004). Dozens of
studies have examined the response of breeding birds to forest structure at both the stand
and landscape levels, often with conflicting results (MacArthur and MacArthur 1961,
Thomas et al. 1979, Manuwal 1991, Ralph et al. 1991, McGarigal and McComb 1995,
Willson and Comet 1996, Sallabanks et al. 2006). Despite these mixed results, the
importance of some habitat characteristics for breeding birds in coniferous forests has
been documented through multiple scientific studies. They include decay stage and size
32
of snags (Thomas et al. 1979, Cline et al. 1980, Ganey and Vojta 2004), canopy
heterogeneity and diversity in foliage height (MacArthur and MacArthur 1961, Beedy
1982), forest floor complexity (Maser et al. 1979, Hansen et al. 1995, Bull et al. 1997)
including understory vegetation (Hagar et al. 2007), and tree species richness, density and
canopy cover (James and Wamer 1982, Verner and Larson 1989). All of these forest
attributes, however, may exhibit high variability within maturing second growth forests
(Rosenberg 2004).
The presence of large live trees have important implications for avian
conservation efforts in coniferous forests (Rich et al. 2004). Large trees provide nesting
and foraging resources for many landbird species. For example, the presence of deeply
fissured bark increases the surface area for bark foraging birds such as the Brown Creeper
(Certhia americana) (Weikel and Hayes 1999). Layering of vegetation within the midstory of mature coniferous forests provide vertical gradients which enhance complexity
and density of the overall canopy cover (MacArthur and MacArthur 1961), likely
influencing avian diversity, abundance and community composition (Hansen et al. 1995,
Hagar et al. 1996, Willson and Comet 1996). Multiple layering tends to reduce the
amount of understory vegetation due to light limitations, but allows for an extensive
organic debris layer to form over the forest floor, especially with the presence of
deciduous tree species (Ralph et al. 1991). The presence of deciduous trees contributes to
litter layer depth in coniferous dominated Pacific Northwest forests (Ruggerio et al.
1991). Forest floor associated species then utilize this litter layer (Davis 1957, Bull et al.
1997). Many forest birds forage on and nest in deciduous trees in coniferous forests
(Beedy 1982, James and Wamer 1982, Gumtow-Farrior 1991, Ruggerio et al. 1991,
33
Hansen et al. 1995, Donovan et al. 2002, Norris and Pain 2002, Hagar et al. 2007). An
increase in avian diversity in hardwood patches may be driven by mechanisms such as
richness in fruits and foliage-dwelling insects and higher densities of cavities per tree in
some hardwood species (Gumtow-Farrior 1991). In Puget Sound lowland forests,
deciduous trees can contribute significantly to canopy cover during the late spring and
summer months (Table 3). Dense stands of conifer trees impede development of an
understory, which reduces overall biodiversity and may limit some species (Altman and
Hagar 2007) while providing habitat for others needing access to an open forest floor
(Hansen et al. 1995).
Hardwood verses conifer dominance may play an especially important role in
avian responses to forest structure. In Pacific Northwest second-growth forests
hardwoods often dominate young stands, and there is evidence of further recruitment of
hardwood from sapling counts (Appendix D). Species such as Pacific-slope Flycatcher
(Empidonax difficilis), Black-throated Gray Warbler (Dendroica nigrescens) and
Wilson’s Warbler (Wilsonia pusilla) have been found to exhibit associations with
hardwood mosaics within coniferous landscapes (Ruggerio et al. 1991, McGarigal and
McComb 1995, Hagar et al. 1996).
Canopy gaps may also play a role in shaping forest bird communities. In
unmanaged forests, openings in the canopy due to mortality from root rot and other
natural events provide further opportunities for development of a deciduous understory.
Canopy gaps in themselves may have important implications for forest bird distribution,
essentially creating edge effects (Harris 1988). Birds near edges suffer from increased
nest predation and cowbird parasitism (Marzluff and Sallabanks 1998, Manuwal and
34
Manuwal 2002), however avian diversity may increase in canopy gaps, as proximity to
preferred habitat increases and as species that favor more open habitat move in
(Sallabanks et al. 2000). Naturally occurring canopy gaps within the forest reserve as
well as the proximity of agricultural land and roads may have an important influence on
avian community composition.
Many species of birds in temperate forests use or are dependent upon dead
standing trees (snags) for breeding or foraging (Cline et al. 1980, Spies et al. 1988, Bull
et al. 1997). In the past few decades the importance of snags to wildlife, especially
primary and secondary cavity nesters, has gained increasing attention (Thomas et al.
1979). Cavity nesters and snag foragers play a critical role in forest ecosystems by eating
insects and controlling pest outbreaks (Bull et al. 1997). The quality and size of snags are
considered to be a primary factor in maintaining healthy populations of cavity-dependent
species (Cline et al. 1980, Hallet et al. 2001, Ganey and Vojta 2004, Spiering and Knight
2005, Smith et al. 2008). Snag size and density in second-growth forests vary greatly due
to past forestry practices. In unmanaged forests, even of a relatively young age, there is a
regular supply of dying and dead trees due to natural processes of decay. The recruitment
of large snag size (>100cm) however, is dependent on the death and decay of large trees
(Spies et al. 1988). Conservation efforts for sang dependent and associated species are
underway, including snag management practices which allow recruitment to occur
through the retention of large live trees and the creation of artificial snags (Hallet et al.
2001).
The majority of snag research has focused on understanding and managing for
snag density, but several recent studies indicate that snag quality (decay stage, diameter
35
and height) may be a better predictor of snag-dependent bird abundance (Bull et al. 1997,
Spiering and Knight 2005, Smith et al. 2008). One possible indicator of snag quality may
be the representation of many different stages of decay, referred to hereafter as snag stage
diversity. Birds that use snags do so for different reasons and select for different direct or
indirect food and nesting resources that snags of different decay stages provide. For
example, following the concept of ecological niches (MacArthur 1958, Wiens 1992), one
might expect to find an increase in snag dependent species if more decay stages are
available.
Other course woody debris (CWD) such as logs, root wads and branches also play
an important role for Pacific Northwest wildlife, including forest birds (Spies et al. 1988,
Bull et al. 1997, Rosenberg 2004). Species that forage and nest on or near the ground and
within the understory are associated with complex vegetative structure and habitat
attributes which provide food, cover, perching locations, and nesting sites (Maser et al.
1979). In managed forests, forest floor components such as downed logs, stumps, and
root wads are often reduced thus impacting associated forest birds (Altman 1999,
Rosenberg 2004). Ground foraging species such as thrushes (Turdidae) (Mack and Yong
2000) and sparrows (Emberizidae) rely on decaying organic debris generated from CWD
because of the abundance of arthropods found there (Davis 1957). Manuwal (1991)
predicted forest fragmentation and simplification of forest structures such as DWD
through management practices, to result in the decline of forest floor associated species.
Finally, plant species composition certainly can have a strong effect on avian
communities. Multilayered versus single-layered forest canopies clearly can affect avian
diversity and abundance (see above). However, understory diversity may also affect
36
avian communities. Understory plants in Pacific Northwest coniferous forests contribute
to the majority of the forest’s vegetative diversity (Spies and Franklin 1991). A dense
understory in conifer dominated forests provide an abundant food resource of flowers,
seeds, fruits and insects for birds (Willson and Comet 1996). Some studies examining the
role of understory vegetation in avian responses to forest structure show that higher plant
species richness provides for a higher number of avian species because of unique
foraging and nesting needs (Altman and Hagar 2007). Other work in Pacific Northwest
coniferous forests found strong nesting associations of some species to particular plants
(Leonard 1998, Leu 2000). Hagar et al (2007) found species specific foraging
associations to specific plants that hosted distinct arthropod communities. The
understory component of forest structure as been largely neglected in forest management
practices and may be a significant driver for some species in the forest bird community.
The purpose of this study is to examine baseline data of a Puget Sound temperate
rainforest avian community in relation to forest structure by asking the question: What
are the best predictors of avian abundance and diversity? To achieve this objective, I
analyzed the response of forest birds to four primary measures of forest structure: the
abundance, size and quality (decay stages) of snags, the number and quality (decay
classes) of downed-woody debris (DWD), overstory species composition (forest type)
and canopy cover, understory vegetation cover and species richness, and sapling richness
and biomass (Table 1). First, I tested whether forest type (Douglas-fir, mixed conifer,
mixed conifer/deciduous or deciduous) and the degree of a deciduous overstory
influenced the avian community and whether the distance to the nearest canopy gap (>.20
ha) was positively related to avian abundance or diversity. I predicted hardwood
37
dominated stands to have the highest diversity of birds and increases in a deciduous
overstory to positively influence avian abundance and diversity. I predicted avian
abundance and diversity to also increase with decreasing canopy gap distance. Second, I
examined snag characteristic data with 13 species of snag-dependent birds detected
during point counts during the 2008 breeding season (Appendix D). I predicted snag
quality and snag-dependent bird abundance and diversity to be positively related.
Particularly, snag decay stage diversity should be a stronger driver of this bird
community than snag density alone. Congruently, the same attributes for snags should
also apply to DWD. Specifically, I predicted DWD abundance, decay class and decay
class diversity to be positively related to forest floor associated birds (Appendix D).
Third, I tested whether understory cover, species richness, and sapling richness and
biomass influenced avian species associated with the understory. I predicted understory
cover and understory plant species richness to be positively related to avian abundance
and diversity. Because saplings contribute to a large amount of overall understory
biomass and provide additional foraging area for insectivores (see above), I also
predicted sapling biomass to increase avian abundance and diversity. Finally, I discuss
the results in relation to the findings of other bird-habitat studies in similar ecosystems.
2.2 Methods
2.21 Study site
My study took place on The Evergreen State College (TESC) forest reserve located
three miles northwest of Olympia, Washington (approx. 47°04’N, 122°58’W) in a
network of permanent plots hereafter referred to as the Evergreen Ecological Observation
Network (EEON). The EEON permanent plot network consists of 44 10-meter radius
38
permanent plots created in 2006 in a 250-meter grid across the reserve
(http://academic.evergreen.edu/projects/EEON - accessed 3-31-09). Acquired in 1968
and representing the largest land area of any college in Washington State, the forest
reserve comprises over 80% of the entire campus and approximately 314 hectares
(Zimmer Gunsul Frasca Architects 2008) . The land was previously owned by the
Washington Department of Natural Resources (DNR) and was last clear cut between
1937 and 1939 (Figure 2a, 2b). The reserve also includes 3,300 feet of waterfront on Eld
Inlet of the Puget Sound adding to its ecological diversity. The forest is primarily
Douglas-fir (Pseudotsuga menziesii (Mirb.) Franco) dominated with substantial stands of
hardwood (Alnus rubra Bong. and Acer macrophyllum Pursh), western red cedar (Thuja
plicata ex D. Don) and western hemlock (Tsuga heterophylla (Raf.) Sarg.) mosaics.
Hardwoods make up an average of 38% of all trees within permanent plots at this study
site (Table 2; data from EEON). Along coastal plots, mature grand fir (Abies grandis
(Douglas ex D. Don) Lindl.) are present, and throughout the reserve pacific madrone
(Arbulus menziesii Pursh), Scouler’s willow (Salix scouleriana Barratt ex Hook.) and
black cottonwood (Populus balsamifera L. ssp. Trichocarpa (Torr. & A. Gray ex Hook.)
Brayshaw) are also present. Much of the understory vegetation with adequate light is
dominated by dense thickets of sword fern (Polystichum munitum (Kaulf.) C. Presl), salal
(Gaultheria shallon Pursh) and Oregon grape (Mahonia nervosa (Pursh) Nutt.). The
topography of campus is generally flat with the highest elevation at 74 meters above sea
level. The landscape is characterized by gentle slopes bisected by five steep-sloped, short
(mostly ephemeral) streams with headwaters within, or just outside of the reserve, and
steep bluffs along the waterfront of Eld Inlet. My work took advantage of a new network
39
of 44 10-meter radius permanent plots created in 2006 in a 250 meter grid across the
reserve (http://academic.evergreen.edu/projects/EEON).
2.22 Avian data collection
I conducted five minute variable circular plot (VCP) point counts from 23 April to
22 June 2008. I recorded all birds seen and heard, excluding birds that flew overhead and
did not appear to be utilizing the habitat, in 47 established forest plots. I visited each plot
once in the early season and once in the late season to equally sample early nesting
resident species and late nesting migratory species. I recorded the horizontal distance to
the nearest meter for each bird detected with the aide of laser rangefinders and flagging.
Each survey was separated into 3 and 2 minute periods, beginning within 30 minutes of
local sunrise and concluding 3 hours after sunrise (Ralph et al. 1995). Separating
observations into the first three minutes and last two minutes improves comparability
with Breeding Bird Survey (BBS) routes which utilize three minute counts. In this study,
data was analyzed from the full five minute count. All other avian species not associated
with the habitat including flyovers, were listed separately. Environmental data including
cloud cover, temperature, wind, precipitation and noise level (scale of 0-3) were also
recorded. Surveys were suspended due to high winds (>10mph) or precipitation which
penetrated the forest canopy. Environmental variables recorded for each survey can be
tracked over time to reveal potential causes influencing avian populations. The level of
road or construction noise can influence detection rates because detections in dense
forested habitats are generally greater than 90% aural (Ralph et al. 1995).
2.23 Forest structure and vegetation data collection
Forest structure and vegetation variables were collected by students and faculty of the
40
Evergreen State College over the course of two years, between the summer of 2006 and
the summer of 2008 (Kirsch et al. in prep, Fischer et al. unpublished). Intensive
structural data was first collected for 21 plots located in the south end of the forest
reserve. Forest structure variables and methods were determined by the EEON network
during its inception in 2006. The forest structure data used in this study and their
methods are described below and are stored in an online database (Evergreen Ecological
Observation Network 2008). A complete list of all plant codes, common and scientific
names are provided in Appendix B. I conducted bird surveys in 47 plots of the network,
but used habitat and vegetation data for 44 plots (plots K7, K6 and J5 are omitted from
habitat analysis because forest structure data was not available).
2.231 Habitat typing
Forest types were ascertained through a student research project which utilized
aerial photography and GIS technology (Greenberg and Hartley unpublished). Greenberg
and Hartley’s (unpublished) original work consisted of nine distinct forest types (Figure
3). I simplified their forest typing classification system to help elucidate any possible
differences in habitat selection for avian species detected in these habitats. I did this by
ground-truthing each plot and its surrounding area, comparing it visually in the field to
the 1998 data, and then matching the plot with either a) Douglas-fir, b) mixed conifer, c)
mixed conifer/hardwood, or d) hardwood, based on visual estimates of percent cover.
2.232 Trees
Each live tree within the plot was measured for tree species, diameter at breast
height (DBH) (cm), height (m), and height to live crown (m). Tree and live crown height
were determined using a Sunto® clinometer (Vantaa, Finland) and laser rangefinder
41
(U.S. Department of Agriculture Forest Service 2005). I took all canopy cover
measurements using a spherical densitometer after the completion of each bird survey. I
used the average of the two surveys to determine the canopy cover for each plot during
the survey period (April-June). Tree biomass was determined using allometric biomass
estimation equations from the database software package BIOPAK (Means et al. 1994)
similar to (Kirsch et al. in prep) (see http://academic.evergreen.edu/projects/EEON/). I
calculated the distance to the nearest canopy gap (>0.20ha) in ArcMap GIS
(Environmental Systems Research Institute 1992-2005) using recently obtained Light
Detection and Ranging (LiDAR) data. Briefly, a raster calculator was used to develop a
map of tree height based on subtraction of high hits and bare earth hits from recently
obtained LiDAR data (Watershed Sciences Inc. 2008, Stewart 2009)
2.233 Snags
All snags occurring within the plot were measured for tree species, height (m),
DBH (cm) and decay stage. Heights were measured using the same methods as live trees
except when a reliable estimate (<3 meters) was possible. Snag decay was evaluated on a
scale of 1-9 from the classification system developed by the USDA Forest Service
(Thomas et al. 1979). The stages are identified as; dying tree with green remaining (1)
decline (2-browning of needles), death (3-loss of needles, but fine branching still
evident), loose bark (4 loss of fine branching, cracks in bark), bark lost (5-tew branches
remain), broken (6 top of tree lost), decomposed (7 advanced decay, additional breakage
of the trunk). Down material (8-most of trunk is on the ground), and stump (9). As many
relic old growth stumps remain, stumps were noted as cut if such a determination was
possible.
42
2.234 Downed Woody Debris
Downed woody debris (DWD) are classified from snags when they lean at or
below a 45 degree angle. They include dead downed and dead trees, shrub boles, and
tree limbs. Each piece of DWD over 10cm DBH (known as course woody debris) was
measured for volume using three diameter measurements (one at each end and one in the
center) and total DWD length (Harmon and Sexton 1996). Mass of DWD was determined
using volume multiplied by estimated decay-stage-specific density form Harmon and
Sexton (1996). Decay class was assigned on a scale of 1-5 from the classification system
described in Maser et al (1979). In this study, only the numbers of course DWD present
in the plot were used in analyses and fine downed woody debris were not measured or
counted. DWD biomass was not included in this study.
2.235 Saplings
Sapling counts were completed for each plot during the summer of 2008.
Saplings were defined as woody species with a diameter less than 5cm. Saplings were
separated into trees and woody shrubs and included as trees one meter or taller and
woody shrubs two meters and taller. Trees were defined as those which were
representative of the dominant overstory. This allows for saplings that have the greatest
chance of survival to maturity to be included in the sapling count so presence and
diversity of saplings may be tracked over time. The numbers of each species of sapling
was recorded for plot and DBH measurements were taken. Sapling biomass was then
calculated using allometric biomass estimation equations from the database software
package BIOPAK (Means et al. 1994).
2.236 Understory vegetation transects
43
We determined understory cover and diversity using point intercept transects in
each cardinal direction from the center post to plot edge (Brower et al. 1998). The
numbers of hits every 10cm along the 100cm transect were recorded for each species or
as bare ground. We identified all vascular plants except grasses to species (Hitchcock
and Cronquist 1973). If trees and shrubs were encountered along transects they were
counted only if they did not meet the height requirements for saplings. Total percent
cover, species richness and diversity indices were calculated for each plot.
2.24 Data Analysis
I generated community indices for all avian species detected, including relative
abundance, species richness and diversity (McCune and Mefford 1999, McCune and
Grace 2002). For this general community analysis, I used 21 habitat variables for live
trees, snags, DWD, saplings and understory vegetation (Table 1). For community
analyses I used an ordination technique known as non-metric multidimensional scaling
(NMDS) with a Bray-Curtis distance measure to examine the avian community among
multiple habitat variables. The equivalent to an ANOVA procedure, differences in the
avian community and forest type were tested using a multi-response permutation
procedure (MRPP) in PCORD (McCune and Mefford 1999, Gleneden Beach, OR).
Analysis of Variance (ANOVA) with post hoc comparisons (Tukey’s Honest Significant
Difference (HSD)) was used to further test for differences among forest type for avian
abundance, species richness and diversity. I used bi-plot vectors in PCORD to
distinguish any strong correlations potentially driving the avian community. Pearson’s
correlation coefficients were used to quantify correlations of the community similarity
matrix with habitat variables along axis 1 and 2. I then used post-hoc regression analysis
44
to test if these correlations were significant. Finally, I preformed an indicator species
analysis in the same program with graphical depictions of species ordinations in order to
distinguish each species stand type indicator.
Using community indices as dependent variables and structural attributes of the
forest as predictors, I used simple linear regression in JMP-statistics 7.0 (academic
version, SAS Institute Inc. Cary, NC) to examine support for hypotheses regarding
habitat relationships. Because so many predictor variables were examined (see Table 1)
results should be viewed with caution since spurious results are more likely when higher
numbers of predictor variables are examined in multiple linear regressions. I used
Akaike’s Information Criterion (AIC) values (Burnham and Anderson 2002) output from
JMP regression analysis to begin an exploratory analysis of which variables explained the
most variation in the avian communities of each habitat type (i.e. overstory, snags, DWD,
understory). Because my analysis was not parsimonious in its choice of factors to
analyze, these results should be interpreted with caution. For example, AICc (rather than
AIC) should have been used in these analyses due to low sample sizes and analysis of
multiple models (see Burnham and Anderson 2002). Additionally, I did not include an
intercept model for AIC comparisons, and so my analyses assume that models including
predictor variables were more important than intercept only models. Because avian
abundance did not met assumptions of normality, I log transformed all relative avian
abundance data. All regressions and modeling was completed using an alpha of 0.05
(type I error probability).
2.3 Results
2.31 Trees
45
I tested whether forest type (Douglas-fir, mixed conifer, mixed conifer/deciduous
or deciduous) and the degree of a deciduous overstory influenced the avian community. I
predicted hardwood dominated stands to have the highest diversity1 of birds and increases
in a deciduous overstory to positively influence avian abundance and diversity. The
MRPP procedure with NMS visualization revealed avian community structure to differ
significantly among forest types (MRPP A=0.03, P=0.0008; Figure 3). The MRPP Astatistic shows the effect size of forest type on the avian community. Here a 0.03 shows
forest type having a moderate effect on avian community structure (McCune and Grace
2002). Bi-plot vectors along two axes suggested the strongest factor structuring the avian
community to be the degree of a deciduous overstory (Pearson’s r=0.51, r²=0.26; Table 2;
Figure 3). The degree of understory cover was also a strong driver for the avian
community (Pearson’s r=0.43, r²=0.19; Table 2; Figure 3) although these two parameters
were weakly autocorrelated (Appendix D). Biplot vectors provide a visual representation
of community similarity along 2 axes, however when correlations are restricted to one
axis there is increased stress in multidimensional space and therefore stronger
relationships (r²=0.31, r²=0.19 respectively, Table 2).
Analysis of variance analyses on avian abundance also showed a significant
difference among forest types (F=4.44, P=0.009, Figure 4). Post hoc tests suggested this
difference could be attributed particularly hardwood stands. Avian species richness and
diversity did not differ significantly among forest type (F=2.27, P=0.09, F=1.13, P=0.35
respectively) although they were generally higher in hardwood stands (Figure 4).
1
Hereafter I refer to Shannon’s diversity index McCune, B., and J. B. Grace. 2002. Analysis of Ecological
Communities. MjM Software Design, Gleneden Beach, OR. as simply “diversity”.
46
Linear regressions revealed a number of significant habitat relationships for avian
abundance, species richness and diversity (Table 4). Supporting ordination findings, the
only significant variable predicting avian abundance was percent deciduous overstory
(R²=0.31, P<0.001, Tables 2 and 4, respectively). The degree of a deciduous overstory
was only weakly related to avian species richness (R²=0.07, P=0.04). Deciduous
overstory was not significantly related to avian diversity (P=0.23). Although I predicted
avian abundance and diversity to increase with decreasing canopy gap distance, I found
no significant relationship between avian abundance or diversity and canopy gap distance
(P=0.39, P=0.10, respectively). Overstory and tree habitat modeling showed deciduous
cover to be the single best predictor of avian community structure (AIC=154.95, Table
5).
2.32 Snags
I predicted snag quality and snag-dependent bird abundance and diversity to be
positively related. Particularly, snag decay stage diversity should be a stronger driver of
this bird community than snag density alone. The density of snags were only weakly
related to avian abundance (R²=0.10, P=0.02), species richness (R²=0.14, P=0.007) and
diversity (R²=0.07, P=0.04) and snag decay stage diversity were significantly related to
avian richness (R²=0.12, P=0.01) but not diversity (P=0.05) or abundance (P=0.06).
Contrary to my predictions, these attributes of snags had negative relationships to these
avian community indices (Figure 6). Snag decay diversity was the best snag habitat
model to show an influence on avian species richness (AIC=12.37). The best models for
other community indices were snag decay stage for avian diversity (AIC=-104.53) and
snag size (DBH) for avian abundance (AIC=202.31).
47
2.33 Downed Woody Debris
Congruently, I predicted the same attributes for snags should also apply to DWD.
Specifically, I predicted DWD abundance, decay class and decay class diversity to be
positively related to forest floor associated birds. I found no significant relationships
between avian community indices and any DWD variable (Table 4). For DWD, the best
models were decay diversity for avian abundance (AIC=-186.25) and species richness
(AIC=11.58). For avian diversity, AIC used in model selection were too close for an
appropriate model to be chosen (Table 5).
2.34 Understory
For understory forest structure variables, the only significant variable related to
avian community structure was sapling biomass (Figure 6). Avian diversity was
significantly related to sapling biomass (R²=0.16, P=0.004, Figure 6) but understory
cover was not (P=0.05). Avian species richness was related to sapling biomass as well
(R²=0.11, P=0.02, Figure 6). The best models for understory structure and avian diversity
used sapling biomass as a predictor variable (AIC=-170.16). The best model for avian
abundance was sapling species richness (AIC=-213.60) and the best model for avian
species richness was understory species richness (AIC=38.54).
2.35 Indicator Species Analysis
Indicator species analysis revealed Cassin’s Vireo (Vireo cassinii), Warbling
Vireo (Vireo gilvus), Black-capped Chickadee (Poecile atricapillus), Chestnut-backed
Chickadee (Poecile rufescens), Swainson’s Thrush (Catharus ustulatus), and Song
Sparrow (Melospiza melodia) to be associated with a particularly forest type (Table 3).
Graphical representation of each species matrix and correlation values revealed which
48
stand type drove these indicators (Figure 5). Both species of Vireo (r=0.173, r=0.197
respectively), Black-capped Chickadees (r=0.0628) and Song Sparrows (r=0.625) were
indicators of hardwood stands. Chestnut-back chickadees were indicators of coniferous
dominated stands (r=-0.581). Swainson’s Thrushes were only weakly correlated to
conifer stand types (r=0.146).
2.4 Discussion
Forest type is an important factor influencing avian community structure at this
study site. Specifically, a deciduous tree component appears to positively influence both
avian abundance and species richness. Avian diversity was not significantly affected by
deciduous tree abundance, probably because species evenness is lower in deciduous
stands (Appendix D). These results corroborate with the findings of other bird-habitat
studies in conifer dominated forests (James and Wamer 1982, Willson and Comet 1996).
Deciduous trees represent a small portion of tree biomass (Kazakova et al. 2007) at this
study site, and appear to be a limiting foraging and nesting resource for many avian
species. Contrary to other studies, overstory canopy cover and the height of live crowns
(all tree species) did not affect the avian community (James and Wamer 1982). The
variation in live crowns and overlapping canopies at different heights that were
unmeasured in this study may have confounded these results (MacArthur and MacArthur
1961). Supporting findings in other forested ecosystems, distance to canopy gaps did not
play a significant role in shaping avian community structure (Sallabanks et al. 2000 and
references therein). In all analyses, my findings suggest deciduous cover to influence the
avian community, supported by varying levels of significance.
49
Contrary to my predictions, the diversity of snag decay stages does not appear to
provide additional bird habitat for a wider array of species in this study area. I observed
fewer species in plots with many different stages of decay. These results could mean
snag-dependent birds are selecting for only a few specific decay stages of snags that offer
the best nesting and foraging resources. Brown Creepers (Certhia Americana) and Redbreasted Nuthatches (Sitta Canadensis) utilize snags and dying trees with remaining bark
harboring insects and places to cache food (Hendricks 1995, Ghalambor and Martin
1999, Weikel and Hayes 1999) . Primary cavity nesters such as woodpeckers and
chickadees may prefer snags with more decay, allowing easier excavation (Ganey and
Vojta 2004). Since woodpeckers were detected less often than other birds, additional
years of sampling, with additional woodpecker detections, many provide very different
snag decay stage diversity results. Despite these initial results being contrary to my
predictions, this study is the first I know of to utilize plot level snag data to quantify snag
decay stage diversity.
My snag density results are contrary to the findings of several other snag studies
(Bull et al. 1997, Spiering and Knight 2005, Smith et al. 2008) and should be viewed with
caution. It is possible that each permanent plot where snags were sampled was too small
(20m diameter) to adequately represent the surrounding landscape of snags. It is also
likely that 2 5-minute counts over the entire breeding season did not allow for an
adequate sampling of less common snag-dependent bird species (i.e. woodpeckers).
Here, all snags regardless of size were used in analyses, however future analyses might
truncate a minimum snag diameter to include in density estimates. Additionally, several
seasons of surveys will be necessary to understand how snag availability in this study
50
area influences the abundance, richness and diversity of birds. Further research should
quantify avian use of cavities found in order to determine the role of snag decay stage and
snag size on cavities.
Contrary to other studies (Ruggiero et al. 1991, Bull et al. 1997), forest floor
complexity did not significantly influence the diversity and abundance of the forest floor
avian community. My findings do suggest avian diversity to be slightly higher in plots
with greater understory cover, although the results were non significant (Table 4). Many
species may be selecting for nest sites and places to feed dependent juveniles that offer
substantial cover over structures such as logs and stumps where arthropods are abundant
and a nest site entrance can be well concealed (Willson 1974). In this sense, cover may
be more important during the breeding season than the presence of logs of stumps if they
are not adequately concealed, but this is highly dependent on the individual bird species.
In this lowland rainforest study site, saplings were comprised of 16 deciduous
species and 7 evergreen species, further contributing to a multi-layered and diverse forest
(Appendix D). Sapling biomass was the strongest understory driver of avian species
richness and diversity. An increase in biomass increases surface area for foraging and
may positively influence the numbers of species able to forage there. Hagar (2007) found
higher arthropod prey abundance on deciduous than evergreen trees and shrubs in the
understory of Douglas-fir forests in Western Oregon. Given the high degree of a
deciduous component in the understory of this study site, a greater number of
insectivorous species may be foraging there. The extent and nature of avian use in the
understory is worthy of future investigation.
51
Indicator species results for stand type may be attributed to some species being
found in only one forest type and their sample size. For example, both species of Vireo
were found only in pure hardwood stands but occurred in low numbers decreasing their
correlation coefficient (Figure 5). Swainson’s Thrushes were only weakly correlated to
Douglas-fir and mixed conifer stand types (r=0.146) but were one of the more common
species sampled. Swainson’s Thrushes favor more open forest floor environments for
foraging on arthropods in leaf and needle litter (Mack and Yong 2000). Douglas-fir and
mixed conifer forests impede light reaching the forest floor and inhibit understory growth
(Ruggiero et al. 1991).
These indicator species results support the findings of previous studies where
Black-capped and Chestnut-backed Chickadees are sympatric and have shown resource
partitioning (Sturman 1968). In a lowland rainforest of the Pacific Northwest, Sturman
(1968) found Black-cap’s to be associated with hardwood/deciduous habitat whereas
Chestnut-back’s were associated with coniferous habitat. An additional study supported
these findings by quantifying 3.5 times as many records of Black-capped Chickadees in
deciduous trees as in conifers, and more than five times as many records of Chestnutbacked Chickadees in conifers as in deciduous trees (Smith 1967).
Song Sparrows were a somewhat surprising indicator species given their high
densities in nearly every forest type of the reserve (chapter 1). Their documented
increases in densities near water and riparian areas (Arcese et al. 2002) may be driving
their indicator status in hardwood plots. Song Sparrows may occur in higher densities in
hardwood plots because of increased food availability; however this should be tested with
further study. Given that my indicator species analysis results are somewhat difficult to
52
interpret, it is possible a larger study area with more distinct stand types is needed to fully
understand indicator species analysis results and determine if it is a useful tool for birds
in these habitats.
The objective of this study was to examine habitat relationships in an avian
community in an unmanaged lowland temperate rainforest using data from the first year
of a long-term monitoring effort. As with many multivariate datasets there are many
factors I was unable to examine in this study that may have influenced the results. Given
this data represents the first of its kind at our study site, I was unable to consider
community composition changes from year to year. I did not examine the effects of
proximity to water or moisture gradients, which has been found to influence avian
communities in other regions (Smith 1977, Anthony et al. 1996), species-specific
interactions, or the influence of foliage height diversity (MacArthur and MacArthur 1961,
Verner and Larson 1989). No other influences above the stand level were examined.
Many other factors possibly influenced avian community composition at this study site
and should be addressed in future research.
Avian-habitat regressions presented here explained less than 20% of the variation
in avian community indices. It should be seriously considered that other factors were
equally significant and their contribution may have dramatically influenced R² values,
their negative and positive relationships, and ordination outputs. These factors may be
environmental, or related to differences in individual species ecology.
Comparisons of AIC values to help select the most appropriate habitat model are
useful for understanding what predicts avian distribution and composition. Here I
compared AIC values only for each habitat type (i.e. either snag, DWD, tree or
53
understory). It would also be useful to compare across all habitat types where all species
detected are pooled. These model comparisons will require a larger sample size over
more than one breeding season. Other types of model selection methods (Burnham and
Anderson 2002) may be valuable in assessing bird-habitat relationships at this study site.
Burnham and Anderson describe the importance of parsimonious selection of each
predictor variable prior to analysis. Using AICc in model selection analysis may have
provided more clearer habitat models as the sample size was low relative to the number
of models in this study. In the future an intercept model for each AIC (or AICc)
comparison should be used to address assumptions that predictor variable models are
more important than their intercept counterpart (see Burnham and Anderson 2002 for
further explication of AIC habitat modeling). Because I did not use AICc or intercept
models these initial avian-habitat relationships should be viewed with extreme caution.
One avenue to exploring additional variables and to assist in the collection of
vegetation data is the use of Light Detection and Ranging (LiDAR) data. LiDAR
provides fine grained 3-D data on the physical structure of any terrestrial environment
(Vierling et al. 2008). The complexity of temperate rainforests provide an excellent use
for this new technology. Although several studies have discussed the theoretical
implications for LiDAR (Davenport et al. 2000, Hyde et al. 2005) to influence our
understanding of avian habitat preferences, very few studies have examined wildlife data
in relation to LiDAR habitat data. Vierling et al (2008) believe LiDAR has the potential
to dramatically reduce if not replace labor intensive and time consuming field
measurements. This study site now possesses 2008 LiDAR data for a large area
including the forest reserve and surrounding areas. The implementation of this analytical
54
tool could provide new insights into how the avian community responds to forest
structure. LiDAR data also can expand analyses to larger landscape scales.
To understand patterns of avian community composition in relation to TESC
forest structure it will be imperative for avian data collection to continue and possibly for
habitat variables to be amended as patterns emerge. With 1-3 years of additional breeding
bird surveys, following the same sampling methodologies, a clearer picture of the best
predictors of bird distribution in the forest reserve will emerge. Much of the forest
structure data now available can be used for several more years before substantial
structural changes occur. An exception would be a natural disturbance dramatically
contributing to DWD recruitment and other structural attributes of the forest. These longterm data on forest structure and the avian community will provide the college with new
insights into the ecology of its forest reserve and elucidate bird-habitat relationships.
Understanding how birds and other organisms respond to different forest attributes will
inform future land use planning and management of this unique and ecologically valuable
tract of land.
55
2.5 Literature Cited
Adair, C. unpublished. Campus Orthophoto. The Evergreen State College, Olympia, WA.
Altman, B. 1999. Conservation strategy for landbirds in coniferous forests of western
Oregon and Washington. Oregon-Washington Partners in Flight.
Altman, B., and J. C. Hagar. 2007. Rainforest Birds: A Land Manager’s Guide to
Breeding Bird Habitat in Young Conifer Forests in the Pacific Northwest. U.S.
Geological Survey.
Anthony, R. G., G. A. Green, E. D. Forsman, and S. K. Nelson. 1996. Avian abundance
in riparian zones of three forest types in the Cascade mountains, Oregon. Wilson
Bulletin 108:280-291.
Arcese, P., M. K. Sogge, A. B. Marr, and M. A. Patten. 2002. Song Sparrow (Melospiza
melodia). Poole, A., and Gill, F., eds., The Birds of North America, No. 704.
Beedy, E. C. 1982. Bird communities and forest structure in the Sierra Nevada of
California. Condor 83:97-105.
Brower, J. E., J. H. Zar, and C. N. Von Ende. 1998. Field and Laboratory Methods for
General Ecology. 4th edition. Wm. C. Brown Publishers, Dubuque, IA.
Bull, E. L., C. G. Parks, and T. R. Torgersen. 1997. Trees and logs important to wildlife
in the interior Columbia river basin. USDA Forest Service General Technical
Report PNW-GTR-391, Pacific Northwest Research Station
Burnham, K. P., and D. R. Anderson. 2002. Model selection and multimodel inference: a
practical information-theoretic approach. 2nd edition. Springer-Verlag, New
York.
Cline, S. P., A. B. Berg, and H. M. Wight. 1980. Snag characteristics and dynamics in
Douglas-fir forests, western Oregon. Journal of Wildlife Management 44:773786.
Davenport, I. J., R. B. Bradbury, G. Q. A. Anderson, G. R. F. Hayman, J. R. Krebs, D. C.
Mason, J. D. Wilson, and N. J. Veck 2000. Improving bird population models
using airborne remote sensing. Int J Remote Sens 21:2705-2717.
Davis, J. 1957. Comparative foraging behavior of the Spotted and Brown towhees. Auk
74:129-166.
Donovan, T. M., C. J. Beardmore, D. N. Bonter, J. D. Brawn, R. J. Cooper, J. A.
Fitzgerald, R. Ford, S. A. Gauthreaux, T. Luke George, W. C. Hunter, T. E.
Martin, J. Price, K. V. Rosenberg, P. D. Vickery, and T. B. Wigley. 2002. Priority
research needs for the conservation of neotropical migrant landbirds. Journal of
Field Ornithology 73:329-339.
Environmental Systems Research Institute. 1992-2005. Version 9.1. ArcMap [GIS
software] Redlands, CA.
Evergreen Ecological Observation Network. 2008. EEON Database v.1.0.0.8 [web
application]. Olympia, WA. Accessed: [August 12, 2008]
Fischer, D., J. Kirsch, and A. Kazakova. unpublished. The Evergreen Ecological
Observation Network. Olympia, WA.
Franklin, J. F., T. A. Spies, R. V. Pelt, A. B. Carey, D. A. Thornburgh, D. R. Berg, D. B.
Lindenmayer, M. E. Harmon, W. S. Keeton, D. C. Shaw, K. Bible, and J. Chen.
2002. Disturbance and structural development of natural forest ecosystems with
56
silvicultural implications, using Douglas-fir forests as an example. Forest Ecology
and Management 155:39-423.
Ganey, J. L., and S. C. Vojta. 2004. Characteristics of snags containing excavated
cavities in northern Arizonia mixed-conifer and pondersoa pine forests. Forest
Ecology and Management 199:323-332.
Ghalambor, C. K., and T. E. Martin. 1999. Red-breasted Nuthatch (Sitta canadensis).
Poole, A., and Gill, F., eds., The Birds of North America, No. 459.
Greenberg, K., and A. Hartley. unpublished. Forest canopy cover typing. The Evergreen
State College, Olympia, WA.
Gumtow-Farrior, D. A. 1991. Cavity resources in Oregon white oak and Douglas-fir
stands in the mid-Willamette Valley, Oregon., Oregon State Univ, Corvallis.
Hagar, J. C., K. M. Dugger, and E. E. Starkey. 2007. Arthropod prey of Wilson's
Warblers in the understory of Douglas-fir forests. Wilson Journal of Ornithology
119:533-546.
Hagar, J. C., W. C. McComb, and W. H. Emmingham. 1996. Bird communities in
commercially thinned and unthinned Douglas-fir stands of western Oregon.
Wildl. Soc. Bull. 24:353-366.
Hallet, J. G., T. Lopez, M. A. O'Connell, and M. A. Borysewicz. 2001. Decay dynamics
and avian use of artifically created snags. Northwest Science 75:378-386.
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.
Harmon, M. E., and J. Sexton. 1996. Guidelines for measurements of woody detritus in
forest ecosystems. US Long Term Ecol. Res. Network Off., Univ. of Wash,
Seattle, WA, USA.
Harris, L. D. 1988. Edge effects and conservation of biotic diversity. Conservation
Biology 2:330-332.
Hendricks, P. 1995. Ground-caching and covering of food by a Red-breasted Nuthatch.
Journal of Field Ornithology 66:370-372.
Hitchcock, C. L., and A. Cronquist. 1973. Vascular Plants of the Pacific Northwest.
University of Washington Press, Seattle, WA.
Hyde, P., R. Dubayah, B. Peterson, J. B. Blairb, M. Hoftona, C. Hunsakerc, R. Knoxb,
and W. Walkerd. 2005. Mapping forest structure for wildlife habitat anaylsis
using waveform lidar: validation of montane ecosystems. Remote Sens Environ
96:427-437.
James, F. C., and N. O. Wamer. 1982. Relationships between temperate forest bird
communities and vegeation structure. Ecology 63:159-171.
Kazakova, A., J. Kirsch, and D. Fischer. unpublished. First year estimate in variability of
productivity and diversity in the lowland Puget Sound permanent plot network.
The Evergreen State College, Olympia, WA.
Kirsch, J., A. Kazakova, and D. Fischer. in prep. Carbon Storage and Productivity
Estimates in a Second-Growth Forest. Forest Science.
Leonard, J. P. 1998. Nesting and foraging ecology of Band-tailed Pigeons in western
Oregon Oregon State University, Corvallis, OR.
Leu, M. 2000. Breeding territory settlement patterns and mate choice in a monochromatic
tyrannid flycatcher. Univeristy of Washington, Seattle, WA.
57
MacArthur, R. H. 1958. Population ecology of some warblers of Northeastern coniferous
forests. Ecology 39:599-619.
MacArthur, R. H., and J. W. MacArthur. 1961. On bird species diversity. Ecology
42:594-598.
Mack, D. E., and W. Yong. 2000. Swainson’s thrush (Catharus ustulatus). Poole, A., and
Gill, F., eds., The Birds of North America, No. 540.
Manuwal, D. A. 1991. Spring bird communities in the southern Washington Cascade
Range. USDA For. Serv. Rep. 285: 160–174, .
Manuwal, D. A., and N. J. Manuwal. 2002. Effects of habitat fragmentation on birds in
coastal forests of the the Pacific Northwest. Studies in Avian Biology 25:103-112.
Marzluff, J. M., and R. Sallabanks, editors. 1998. Avian conservation : research and
management. Island Press, Washington, D.C.
Maser, C., R. G. Anderson, G. Ralph, K. Cromack, J. T. Williams, and R. E. Martin.
1979. Dead and down woody material. In: Thomas, J.W. (Ed.), Wildlife Habitats
in Managed Forests: the Blue Mountains of Oregon and Washington. USDA
Agriculture Handbook No. 553.
McCune, B., and J. B. Grace. 2002. Analysis of Ecological Communities. MjM Software
Design, Gleneden Beach, OR.
McCune, B., and M. J. Mefford. 1999. PC-ORD Multivariate analysis of ecological data
version 4. MjM Software Design, Gleneden Beach OR.
McGarigal, K., and W. C. McComb. 1995. Relationships between landscape structure
and breeding birds in the Oregon Coast Range. Ecol. Monogr. 65:235-260.
Means, J. E., H. A. Hansen, G. J. Koerper, P. B. Paul B. Alaback, and M. W. Klopsch.
1994. Software for computing plant biomass--BIOPAK users guide. Portland,
OR: U.S. Department of Agriculture, Forest Service, Pacific Northwest Research
Station.
Norris, K., and D. J. Pain, editors. 2002. Conserving bird biodiversity: general principles
and their application. Cambridge University Press, New York, NY
Ralph, C. J., S. Droege, and J. R. Sauer. 1995. Managing and monitoring birds using
point counts: standards and applications.
Ralph, C. J., P. W. Paton, and C. A. Taylor. 1991. Habitat association patterns of
breeding birds and small mammals in Douglas-fir and hardwood stands in
northwestern California and southwestern Oregon. Pages 379-394 in L. F.
Ruggiero, K. B. Aubry, A. B. Carey, andM. H. Huff, editors. Wildlife and
vegetation of unmanaged Douglas-fir forests. USDA For. Serv. Gen. Tech.
Rep,Portland, OR.
Rich, T. D., C. Beardmore, H. H. Berlanga, P. Blancher, M. Bradstreet, G. Butcher, D.
Demarest, E. Dunn, C. Hunter, E. Inigoelias, J. Kennedy, A. Martell, A. Panjabi,
D. Pashley, K. Rosenberg, C. Rustay, S. Wendt, and T. Will. 2004. Partners in
flight North American landbird conservation plan. Partners in Flight.
Rosenberg, K. V. 2004. Partners in flight continental priorities and objectives defined at
the state and bird conservation region levels: Washington. Partners in Flight.
Ruggerio, L. F., K. B. Aubry, A. B. Carey, and M. H. Huff. 1991. Wildlife and vegetation
of unmanaged Douglas-fir forests.
Ruggiero, L. F., L. L. C. Jones, and K. B. Aubry. 1991. Plant and animal habitat
associations in Douglas-fir forests of the Pacific Northwest: an overview. in L. F.
58
Ruggiero, K. B. Aubry, A. B. Carey, andM. H. Huff, editors. Wildlife and
vegetation of unmanaged Douglas-fir forests. USDA For. Serv. Gen. Tech. Rep.
Sallabanks, R., J. B. Haufler, and C. A. Mehl. 2006. Influence of forest vegetation
structure on avian commmunity composition in west-central Idaho. Wildlife
Society Bulletin 34:1079-1093.
Sallabanks, R., J. R. Walters, and J. A. Collazo. 2000. Breeding bird abundance in
bottomland hardwood forests: habitat, edge, and patch size effects. Condor
102:748-758.
Smith, C. Y., I. G. Warkentin, and M. T. Moroni. 2008. Snag avaliability for cavity
nesters across a chronosquence of post-harvest landscapes in western
Newfoundland. Forest Ecology and Management 256:641-647.
Smith, K. G. 1977. Distribution of summer birds along a forest moisture gradient in an
Ozark watershed. Ecology 58:810-819.
Smith, S. M. 1967. An ecological study of winter flocks of Black-capped and Chestnutbacked Chickadees. Wilson Bulletin 79:200-207.
Spiering, D. J., and R. L. Knight. 2005. Snag density and use by cavity nesting birds in
managed stands of the Black Hills National Forest. Forest Ecology and
Management 214:40-52.
Spies, T. A., and J. F. Franklin. 1991. The structure of natural young, mature, and oldgrowth Douglas-fir forests in Oregon and Washington. in L. F. E. Ruggiero,
editor. Wildlife and Vegetation of unmanaged Douglas-fir forests. USDA For.
Serv. Gen. Tech. Rep. ,Portland, OR.
Spies, T. A., J. F. Franklin, and T. B. Thomas. 1988. Course woody debris in Douglas-fir
forests of western Oregon and Washington. Ecology 69:1689-1702.
Stewart, G. 2009. Personal Communication. D. Fischer, January 20, 2009, The Evergreen
State College, Olympia, WA.
Sturman, W. 1968. Description and Analysis of Breeding Habitats of the Chickadees,
Parus Atricapillus and P. Rufescens. Ecology 49:418-431.
Thomas, J. W., R. G. Anderson, C. Maser, and E. L. Bull. 1979. Snags. In: Thomas, J.W.
(Tech. Ed.), Wildlife Habitats in Managed Forests: the Blue Mountains of Oregon
and Washington. USDA Agriculture Handbook No. 553.
U.S. Department of Agriculture Forest Service. 2005. Forest Inventory and Analysis
national core field guide. U.S. Department of Agriculture Forest Service.
Verner, J., and T. A. Larson. 1989. Richness of breeding bird species in mixed-conifer
forests of the Sierra Nevada, California. Auk 106:447-463.
Vierling, K. T., L. A. Vierling, W. A. Gould, S. Martinuzzi, and R. M. Clawges. 2008.
Lidar: shedding new light on habitat characterization and modeling. Front Ecol
Environ 6:90-98.
Watershed Sciences Inc. 2008. LiDAR Remote Sensing: Olympia, Washington.
Watershed Sciences.
Weikel, J. M., and J. P. Hayes. 1999. The foraging ecology of cavity-nesting birds in
young forests of the northern Coast Range of Oregon. Condor 101:58-66.
Wiens, J. A. 1992. The ecology of bird communities. Cambridge University Press,
Cambridge, UK.
Willson, M. F. 1974. Avian community orgainzation and habitat structure. Ecology
55:1017-1029.
59
Willson, M. F., and T. A. Comet. 1996. Bird communities of northern forests: patterns of
diversity and abundance. Condor 98:337-349.
Zimmer Gunsul Frasca Architects. 2008. Campus Master Plan. The Evergreen State
College.
60
Tables
Table 1. Variables used in habitat community analyses to identify factors driving avian abundance, richness and diversity in the Evergreen State College forest reserve, Olympia, WA.
Structure type
Response Variables
CODE
UNITS
DESCRIPTION
Avian species abundance
BDIV
H'
Shannon-Weaver Index of diversity
Avian species richness
BSR
species/plot
bird species richness
Avian species diversity
BAB
log(no./plot)
relative bird abundance log transformed
Habitat/stand type
STTYPE
categorical
1=Douglas-fir, 2=mixed conifer, 3=mixed conifer/hardwood, 4=hardwood
Canopy gap distance (m)
CPYGAP
m
Distance to nearest canopy gap
# of trees
TREES
no./plot
number of trees in plot
Tree biomass (kg)
TRBIO
kg
total tree biomass in plot
Average tree height (m)
CRHT
m
average height of live crown
Average crown height (m)
TRHT
m
average tree height
Deciduous cover (%)
DECDCOV
%
percent deciduous trees
Overstory canopy cover (%)
CPYCOV
%
average overstory canopy cover
Tree species richness
TREERICH
no./plot
number of overstory tree species in plot
# of snags
SNAGS
no./plot
number of standing dead trees
snag DBH (cm)
SGDBH
cm
average snag diameter at breast height
snag height (m)
SGHT
m
average height of standing dead trees
Snag decay diversity
SGDECDIV
categorical
average snag decay stage
average snag decay stage
SGDECSTG
categorical
number of snag decay stages represented in plot
# of DWD
DWD
no./plot
number of DWD counted in plot
DWD decay diversity
DWDDECDIV
categorical
number of DWD decay classes represented in plot
average DWD decay stage
DWDDECCL
categorical
average DWD decay class
Understory species richness
VEGRICH
no./plot
number of plant species, excluding trees and saplings in each plot
Understory cover (%)
VEGCOV
%
combined cover of all plant species excluding trees and saplings in each plot.
Sapling biomass (kg)
SAPRICH
no./plot
number of sapling tree species in plot
Sapling species richness
SAPBIO
kg
total sapling biomass in plot
Predictor Variables
LIVE TREES
SNAGS
DWD
UNDERSTORY
61
62
63
64
65
Figures
Figure 1. Map of EEON permanent forest plots showing nine different forest types across
the reserve (Greenberg and Hartley, unpublished data). Forest types in this study were
simplified into 1) Douglas-fir 2) mixed conifer 3) mixed conifer/hardwood 4) hardwood.
66
Figure 2a.1939 Orthophoto of the western half of future TESC property showing
extensive land clearing (scanned and prepared by C. Adair)
67
Figure 2b. Western half of TESC property as it looks today with large scale forest
regeneration (http://academic.evergreen.edu/projects/EEON/, accessed April 11,
2009)
68
A=0.03
P=0.0008
Figure 3. NMS Ordination of avian community composition for 4 different forest
stand types, Douglas-fir (), mixed conifer (), mixed conifer/mixed hardwood
(), hardwood () in The Evergreen State College forest reserve, MRPP
A=0.03, P=0.0008. Bi-plot vectors show correlations of the matrix with percent
deciduous cover (Pearson’s r=0.261) and percent understory cover (Pearson’s
r=0.185).
69
A
F=4.44
P=0.009
B
F=2.27
P=0.090
C
F=1.13
P=0.350
70
71
72
Figure 5. Graphical representations of NMS ordinations of species specific composition for 4
different forest stand types, Douglas-fir (), mixed conifer (), mixed conifer/mixed hardwood
(), hardwood () in the Evergreen State College forest reserve, Olympia, WA. Greater
abundances of each species are represented by larger symbols on the ordination graph. Axis
graphs show degree of correlation between each species’ abundance and community similarity
along each axis. Both species of Vireo, Black-capped chickadees and Song Sparrows appear to
be indicators of hardwood stands. Chestnut-backed chickadees appear to be indicators of
coniferous dominated stands.
73
A) Avian diversity
R²=0.06,
P=0.05
R²=0.07,
P=0.04
R²=0.16, P=0.004
74
B) Avian species richness
R²=0.07, P=0.04
R²=0.14, P=0.007
75
R²=0.12, P=0.01
R²=0.11, P=0.02
76
C) Log avian abundance
R²=0.10, P=0.02
Figure 6. Significant habitat relationships based on regression models for community indices
A) Avian diversity B) Avian species richness C) Log avian abundance. Snag density and snag
decay stage diversity were negatively related to all community indices. Sapling biomass and
percent deciduous cover explained the greatest variations in the bird community (R²=0.15,
R²=0.11, R²=0.32, respectively).
77
Chapter Three
Protocols and Research Recommendations for the Evergreen Avian Monitoring
Program
In order to understand recent global declines in birds and to guide conservation efforts,
long-term monitoring programs are necessary in conjunction with in depth species
specific research. Increasingly, avian population monitoring is guiding the recovery of
declining species and assisting initiatives to keep healthy population numbers steady.
Many birds in the Pacific Northwest are also experiencing these declines, especially
within lowland temperate rainforests which are under development pressure and may face
ecological changes with a warming climate. The Evergreen State College’s forest reserve
offers an ideal setting in which to monitor populations of temperate rainforest birds,
contributing to other long-term monitoring and conservation efforts. This chapter
provides the necessary field protocols, tools and research considerations for EAMP to
become a successful and effective component of the Evergreen curriculum and the larger
south Puget Sound community. The protocol should be considered a working document
with future additions and revisions made by students and faculty as the first years of
monitoring are carried out and continuing contributions to the field of avian conservation
are made.
78
3.1 Introduction
3.11 Monitoring
Long-term monitoring is widely used across the globe to understand population
dynamics of birds (Marzluff and Sallabanks 1998, Ruth et al. 2003, Rich et al. 2004).
Monitoring methodologies are based upon the objective they wish to achieve. Four
common purposes and goals for monitoring are to: 1) show trends in populations over
time at the landscape level (i.e the North American Breeding Bird Survey); 2) determine
the influence of specific management actions; 3) conservation-oriented studies on
reproductive success and survivorship; 4) determining the quality and quantity of habitat.
Long-term population data that show trends over time are designed to identify significant
or non-significant declines or increases in species within and among continental regions
(Sauer et al. 2008). Usually spearheaded or run by governmental and academic groups,
specific management actions or habitat conditions are assessed in relation to bird
community responses (Ruggiero et al. 1991, Hansen et al. 1995, Young and Hutto 2002).
Conservation minded non-profits such as Institute for Bird Populations (IBP) and PRBO
Conservation Science conduct long-term monitoring of reproductive success and
survivorship, usually through bird banding and nest searching methods.
Many successful monitoring programs link three types of data together to achieve
an understanding of the bird community in a given area (Ralph et al. 1993). Successful
monitoring programs bring together data on a) population size or trends of various
species occupying habitat; b) the demographic parameters of these populations; and c)
detailed habitat variables linked to the population and demographic data (Ralph et al.
1993). The newly created Evergreen Avian Monitoring Program (EAMP) will collect
79
data on population trends of campus birds, reproductive success, survivorship and the
habitat variables influencing this data. The objective of EAMP is to become a long-term
monitoring program grounded in proven methods of data collection that can be replicated
and shared with the greater scientific community, helping to identify declining species
and understand the causes.
The sampling methodologies of EAMP are designed to complement the work of
other programs and to provide comparisons within the Pacific Northwest region and
beyond to other applicable regions. The 418 hectares that EAMP covers comprises a
limited amount of habitats but will provide monitoring data for an area currently
underrepresented in monitoring efforts (Battaglia 2000). Inferences can provide a
localized assessment of bird populations as well as comparability to other regional
projects. Initial results from EAMP work will identify patterns that need additional
monitoring strategies and the development of hypothesis-driven research. Over time
EAMP can assist other large and small scale programs in explaining bird population
dynamics at local, regional and continental levels.
The methods of monitoring for EAMP follow the approaches of other well
established programs across North America in habitats similar to our own (Martin and
Geupel 1993, PRBO Conservation Science 2004, Siegel et al. 2004, Wilkerson et al.
2005). Because this is a long-term endeavor the population results for a wider variety of
species, demographic data and habitat relationships will not be available for several more
years. Typically results from avian research and monitoring programs are published after
anywhere from 3-20 years of data collection. An important goal of EAMP is to
collaborate with outside organizations and partners. One avenue for collaboration and
80
data sharing is the Avian Knowledge Network (ANK), a relatively new online database
which links hundreds of short and long-term avian datasets from around the continent
(Avian Knowledge Network 2008).
3.12 Research and monitoring needs in Pacific Northwest forests
Research reviewed by Partner in Flight (PIF) of Washington indicates most
reproductive studies at the community and landscape levels to be located in the Cascade
Mountains and inland areas of Washington and Oregon (Rosenberg 2004). Few
population, reproduction and behavioral bird studies cover the habitat of the Puget Sound
basin’s remaining temperate rainforests (Battaglia 2000). Furthermore, many bird/habitat
studies occur on federal lands due to their size and diversity of habitats (Hansen et al.
1995, Anthony et al. 1996, Hagar et al. 1996, Weikel and Hayes 1999, Nott et al. 2005).
PIF describes a need for data on reproductive success of focal species that can provide
information on where source and sink habitats are occurring (Altman 1999). A wider
array of western coniferous forest habitat needs to be included in monitoring efforts, at
varying elevations and locations throughout the northern pacific rainforest region
(Rosenberg 2004). Monitoring studies also require a broader range of research
objectives, including everything from species-specific reproductive information to
population changes at community and landscape scales (Rich et al. 2004, Rosenberg
2004).
In the past several decades ornithologists have begun to study the relationships
between breeding, stopover and wintering grounds that influence bird conservation
(Robbins et al. 1989). Documented declines in some species have not always been
explained from breeding studies which has emphasized the importance of over-wintering
81
and migratory stop-over studies (Plummer 2002, Rich et al. 2004). Current long-term,
over-wintering studies are being conducted in areas of Latin America targeted at
Neotropical migrants (Ralph et al. 2005, DeSante et al. In Press.) as well as the northern
temperate zones for species which breed in northern boreal forest and tundra (Humple et
al. 2001, PRBO Conservation Science 2004, Samuels et al. 2005) . The forests of
Evergreen host 6 common species which spend the winter here and are candidates for
over-wintering survival studies; Ruby-crowned Kinglet (Regulus calendula), Varied
Thrush (Ixoreus naevius), Hermit Thrush (Catharus guttatus), Golden-crowned Sparrow
(Zonotrichia atricapilla), Fox Sparrow (Passerella iliaca), and Lincoln’s Sparrow
(Melospiza lincolnii) (Wahl et al. 2005). Over-wintering studies could be carried out in
conjunction with the already established Olympia area Christmas Bird Count (CBC)
(National Audubon Society 2002). Taking CBC forest passerine data and applying it to
productivity and survivorship data collected on campus via banding efforts will help
elucidate the population dynamics and behavior of our wintering songbirds (particular
sparrows which occur in high numbers and are easily caught in mist-nets). PRBO
Conservation Science has operated a color-banded focal species project for over 20 years
and tracks territories and survivorship throughout the year (PRBO Conservation Science
2004). Observers resight color-banded birds by routine area searches around the study
site. Such monitoring has helped determine the success of restoration efforts in degraded
riparian habitat while at the same time providing long term data on populations of
resident and over-wintering species (Samuels et al. 2005). A pilot project of color
banding resident and over-wintering species should take place in the future to assess the
82
feasibility and resources required to carry out a color banding study on focal species
(Ralph et al. 1993).
3.2 Focal Species: demographic monitoring
3.21 Productivity and Survivorship
Mist-netting involves capturing birds in mesh nets, banding them with a unique
number sequence, and taking measurements and data on species, age, sex, breeding
status, molt and physical condition (Pyle 1997). Mist-netting data provide indices of
productivity through the assessment of adult breeding condition and analysis of juvenile
to adult capture ratios (DeSante et al. 2008). In addition to USGS bands placed on each
bird caught, in depth studies of focal species that are common and easily captured in
mist-nets can contribute greatly to general knowledge of a species (Ralph et al. 1993).
Life history studies now often involve the placement of a unique color-band sequence (3
colors plus the USGS required band) placed on the legs of individuals (Koronkiewicz et
al. 2005). In addition to recapturing the bird in the mist-net throughout its life, colorbanded individuals can be resighted unobtrusively with the aide of observers. Many
successful color-banding projects augment survivorship data by resighting color-banded
birds that are never again caught in a mist-net (Martin and Geupel 1993, PRBO
Conservation Science 2004). Concurrently, monitoring can track individual territories
and pairs throughout the breeding season. Resident species are excellent candidates for
intensive life history studies because recapturing and/or resighting an individual is more
likely (Martin and Geupel 1993, DeSante et al. 2008).
Examining conditions and habitats on wintering and migration grounds may help
to reveal the causes for declines in many species (Robbins et al. 1989, Marzluff and
83
Sallabanks 1998). In addition to resident species, TESC has several species that return
each winter before moving further north or into the mountains to breed (Ruby-crowned
Kinglet (Regulus calendula), Varied Thrush (Ixoreus naevius), etc.). These species
should not be overlooked in monitoring efforts (Rich et al. 2004) conducted by EAMP.
Similarly mist-netting efforts can occur during fall and spring migration which can track
arrival and departure dates of Neotropical breeding species over time as well as the extent
and distribution of other species moving through the area (Root et al. 2003, Sparks et al.
2005). Migration net-netting will provide average arrival and departure dates to
document avian responses to climate change (Root et al. 2003) and may be more feasible
for the involvement of regular school year programs at TESC in the early stages of
EAMP.
Banding methods assess the population health of only a small proportion of bird
species in an area, usually about 10 species (Ralph et al. 1993). Permanent net locations
with fixed flag poles can provide an opportunity to capture mid and upper canopy species
with the use of stackable nets (Humphery et al. 1968). Otherwise there is the severe
limitation of only sampling ground and understory associated species (Bonter et al.
2008). This is an addition that should be employed after mist-netting stations are
operating comfortably as it is labor intensive and costly to implement (personal
observation).
Information on survivorship and productivity indices are valuable but the data
loses habitat specificity throughout the breeding season as adults birds disperse to feed
young (Martin and Geupel 1993). For this reason mist-netting should be complimented
with other monitoring methods, such as intensive point count surveys conducted at least
84
twice during the breeding season, to account for birds that may have dispersed (Ralph et
al. 1993). Surveys are conducted around the mist-netting station and should be
completed at about the mid point of the 10 day interval between banding days (PRBO
Conservation Science 2004). The use of intensive and extensive point counts are
described below under protocols.
3.32 Nest searching
Nest searching techniques involve the establishment of intensive plots where an
individual works to locate and track the development and outcome of each nest (Martin
and Geupel 1993). Upon competition of the nesting attempt, the vegetation around the
nest is measured along with that of a control area in representative habitat away from the
nest (PRBO Conservation Science 2004). Nest searching is usually done between May
and August and each nest is visited at least once every four days to check the status of the
nest and record observations. Nest searching is labor and training intensive and applies to
fewer species than mist-netting but provides a direct assessment of reproductive success
and can elucidate rates and causes of predation and parasitism as well as basic breeding
and nesting behavior (Martin and Geupel 1993).
Due to the high canopy and dense undergrowth found throughout the TESC forest
reserve, directly locating nests of focal species will present difficulties. Resident species
generally nest near or on the ground with excellent concealment (Ehrlich et al. 1988,
Wahl et al. 2005). Observational approaches that indirectly confirm nesting activity and
reproductive success may be a viable alternative for species that are particularly difficult
to nest search for (Vickery et al. 1992). Such approaches use breeding behaviors linked to
stages in the reproductive cycle to determine a pair’s nesting status. Reproductive
85
behaviors can be transformed into reproductive indices and used as measures of fitness
(Mayfield 1961). Such tactics are used in a variety of situations where a species may be
rare or endangered and nest disturbances during breeding may influence a population
(Ralph et al. 1993)
Another indirect reproductive monitoring tool is documenting song types to
indicate mated status of singing males. Although this approach does not confirm success
or failure it indicates pairing or certain species in particular habitats, distinguishing paired
species from non-paired ones. This may be important in certain habitats of TESC reserve
that are connected to particular species such as cottonwood and willow stands in riparian
areas of campus and warbling vireos (Vireo gilvus) (personal observation).
Alternatives to nest searching and monitoring should be considered only after an
attempt to start a nest searching program over one season has been made (Ralph et al.
1993). After one or two pilot seasons of nest monitoring it will become clearer whether
direct reproductive monitoring is a viable option for EAMP. If so few nests are located
as to not support existing data or resources are limited for locating nests, further research
on alternatives should be made along with recommendations for protocol changes. The
nest monitoring protocols are included here in order to expedite a pilot project to assess
its viability as a monitoring tool.
3.3 Protocols
3.31 Point counts
The continued collection of breeding bird survey data, as I have done in 2008 will
be a vital component to the success of EAMP. Such surveys are known as extensive
point counts because they cover an entire study area (Ralph et al. 1993). Extensive
86
surveys remain the primary method for monitoring population changes in forest birds
(PRBO Conservation Science 2004). Each year a minimum of one survey should be
completed in each plot of EEON (Buskirk and McDonald 1995, Siegel et al. 2001).
Yearly point counts will enable EAMP to assess changes in populations in fixed locations
and habitats (permanent plots within the network), the differences in population changes
and species composition between habitats, and the how population trends change over
time. In conjunction with mist netting and banding, intensive point counts located
around the nets at each station should follow the same protocols described below except
there should be a minimum of two surveys spread out over the banding season. Current
monitoring programs do intensive point counts at stations early, mid and late in the
breeding season) (Ralph et al. 1993, Siegel et al. 2001).
In order to calculate absolute abundances of birds each year, EAMP protocols use
a variable circular plot (VCP) point count method where the distance is recorded from the
observer to each bird heard or seen (Reynolds et al. 1980). Although the detection radius
is theoretically infinite, studies have shown that 99% of birds are detected within 125
meters of the observer (Reynolds et al. 1980). In temperate North America the survey
period should run during the height of the breeding season when detection rates are most
stable (Ralph et al. 1993, Siegel et al. 2001). In Western Washington at low elevations
this is generally the beginning of May into the beginning of July (Barrier and Froyalde
1999).
Although there are varying point count protocols EAMP follows the
methodologies of long term monitoring programs such as Point Reyes Bird Observatory
(PRBO) and Institute for Bird Populations (IBP). Counts should begin approximately 15
87
minutes after local sunrise and should be completed before 10am, usually within 3-4
hours (Ralph et al. 1993, Ralph et al. 1995) . Counts should not be conducted if weather
conditions possibly could reduce detection of birds. Winds above 10 mph, continuous
rain or extreme temperatures warrant cancelling a count. At each point record the general
weather conditions, with temperature, cloud cover and wind speed, plot name, starting
time (24 hour clock) and noise level (road or construction are common on campus). A
sample data sheet derived from PRBO Conservation Science (2004) for VCP counts is
provided in Appendix C.
At each point, approach the point quietly, minimizing disturbance. If disturbance
is unavoidable, wait 2-5 minutes quietly for activity to resume to normal. If a bird was
flushed within 10 meters of the point when you arrived, include it in the count. Set your
watch to five minutes and begin the count once you are ready to record. If noise or
another disturbance interrupts the count, cross out the survey, note the disturbance, wait
until it passes and begin the survey again. All surveys should last five minutes with
detections divided into the first three and the last two minutes of the survey (Ralph et al.
1995). Record every species detected regardless of distance, with the appropriate four
letter AOU code (American Ornithologists' Union 2008). For unknown species enter
XXXX. If the group of bird is known substitute the last two letters for that group. For
example, enter XXFL for an unknown flycatcher. Unidentified birds should be followed
and identified if time permits. If no birds are detected at a point make a note on the data
sheet. For each detection also record the distance to the bird when it was first detected to
the nearest meter and the type of identification used as S-song, C-call or V-visual (PRBO
Conservation Science 2004). Other less common types are D for drumming woodpecker
88
and H for humming hummingbird. If a bird is in flight or high in a tree when it is
detected record the distance the bird would be if it were on the ground directly below it
(Buckland et al. 2001). Birds flying over but not using the habitat are not recorded but
should be noted in the field journal to assist with long-term natural history observations
and an updated species list for the college. Record any breeding activity observed as
follows (PRBO Conservation Science 2004):
CO- copulation
TD- territorial display
DD- distraction display
FC- food carry
FL- fledgling(s) observed
FS- fecal sac carry
MC- material carry
NF- nest found
PA- pair
A hardcopy datasheet should exist for every point count conducted, photocopied
and stored in the lab. Ideally all data should be entered the same day as the survey to
reduce errors and make necessary corrections while the field day is fresh in your mind.
When a datasheet is entered the date and your initials should be written on the bottom of
the hardcopy. If time allows datasheets should be proofed with two people before the end
of the season and the date and initials noted again on the sheet (PRBO Conservation
Science 2004).
Adequate training in bird identification by observation, song and call is
imperative to the success of any monitoring program (Ralph et al. 1995, Wilkerson et al.
2005). Each surveyor should learn the songs and calls of all western forest bird species
89
with the aid of audio devices and computer software (around 90 species) (Wahl et al.
2005). Learning how to use rangefinders effectively and estimate distances requires
experience and practice. Observers should begin practicing by estimating the distance to
known objects such as flagging and checking afterwards with the rangefinders (Siegel et
al. 2004, Wilkerson et al. 2005). Afterwards the surveyor should practice counts in the
early breeding season when resident species have begun to sing. The point counter(s) for
each year should help the program’s continuation by advertising, recruiting and training a
point counter(s) for the following breeding season.
3.32 Nest searching and monitoring
Nest monitoring is a helpful component to any avian monitoring program because
it provides data on nesting success or failure, trends in recruitment and natural history
information that may be lacking or incomplete for many species (Martin and Geupel
1993). For most programs nests are located for color-banded individuals. One or both of
the parents will be identifiable and can be tracked through the breeding season. If a nest
is successful the nestlings can be color-banded just before they fledge in order to continue
identifying individual pairs in coming breeding seasons (PRBO Conservation Science
2004). A reasonable start to a nest monitoring program at TESC would be to locate the
nests of color-banded individuals around the organic farm banding station (figure 1).
Nest searching and monitoring is a particularly invasive methodology with potential
impacts that should be minimized at all times (Martin and Geupel 1993). PRBO has put
together a list of nest searching rules and have been amended here for our study area
(PRBO Conservation Science 2004):
1) Distress calling by adults should never continue for more than 5 minutes. If the nest cannot be
located return on a different day.
90
2) Do not approach a nest or attempt to locate a nest you know is close by if a corvid is present (jays,
ravens or crows).
3) When checking or locating a nest never leave a dead end trail to the nest. “Fake” check other
bushes and trees and make other trails in the area.
4) Minimize impact to the vegetation around the nest
5) Do not take others to see the nest. Only the person monitoring the nest should know where it is
located.
6) Use a pen or stick to check the status of a nest so as not to leave your scent around the nest.
7) Move in and out of the nest area quickly when checking a nest, complete datasheets when away
from the nest.
8) Never use flagging or other visual aids to mark a nest location. If a general area must be flagged
flag at least 50 meters away from the nest and write the cardinal direction to the nest, date and nest
number on the flagging. Remove all flagging after the nest is no longer in use.
9) After finding a nest, GPS its approximant location, memorize the location and write down a
detailed description of how to reach it again.
The life history traits of many North American species are poorly understood,
stemming mostly from the misconception that nests are too difficult to find (Martin and
Geupel 1993). When learning to nest search it is helpful to read existing natural history
and nest strategies of the focal species. The Birds of North America series offers detailed
accounts of every avian species breeding in North America (Poole 2005). A good
understanding of behavioral cues seen around nest sites, which comes from practice, is
also an important factor to successfully locating nests (Martin and Geupel 1993). Most
people can begin to locate nests on their own with a few days of practice and watching
others locate nests. For example, during the 2008 breeding season while conducting
91
point counts, I located the nests or saw material or food carries for 27 different avian
species (Table 1).
For resident species in the lowland Puget Sound such as sparrows and wrens, nest
construction may begin as early as late March (Wahl et al. 2005). With some exceptions
such as wrens and woodpeckers the female of most species builds the nest and incubates
the eggs (Ehrlich et al. 1988). For this reason locating and following the activities of the
female is the most effective method for locating a nest (Martin and Geupel 1993). A
mated female will exhibit specific behavioral towards her male partner including rapid
movement around the male with no harassment, and food or copulation begging. Every
effort should be made to locate a nest during construction to obtain complete nesting data
(Mayfield 1975) and minimize disturbance (PRBO Conservation Science 2004). During
nest construction females will carry material in the beak, often with long direct flights to
the or near the nest location (Martin and Geupel 1993). Material is not often seen with
the naked eye and a good pair of 7 or 8x42 binoculars are essential to seeing fine material
such as spider web, lichen or flower seed (personal observation). Locating the source of
nesting material and watching birds come and go is recommended. Once a material carry
is seen follow the bird visually. If the bird disappears into nearby vegetation wait for the
bird to return for more material. Several trips to and from the same location may help
confirm the general location of a nest. If the observer is to close to the nest a bird will
often sit above or nearby the nest nervously until the observer leaves (PRBO
Conservation Science 2004). If this occurs or a bird drops nest material, move away
quickly and relocate. Once the area of the nest site has been identified return later to
92
locate the exact location when the female is not present. Checking for the nest right away
while the female is watching can cause abandonment (Martin and Geupel 1993).
During egg laying, females generally lay one egg a day and may only visit the nest at
that time (Martin and Geupel 1993). Visits and incubation will become more frequent as
more eggs are laid. To locate nests during egg laying use behavioral cues. Both birds
will often look at the nest when they arrive. Females staying in one area without feeding
can indicate a nest nearby. Copulation will occur for each egg laid and generally occurs
above, at or very near the nest. The completion of egg laying and the beginning of
incubation is easily identified by increased singing by the male and difficult detection of
the female. If a female is located and moving and foraging very quickly for short periods
of time she will probably return to the nest soon. Most passerines incubate for 20-30
minutes and feed for 6-10 minutes (Gill 2006). Following a foraging female for 30
minutes may indicate she is not incubating. Incubating females will be generally
conspicuous when foraging but more cautious as they near the nest (Martin and Geupel
1993).
The nestling stage is the easiest time to locate nests (Ralph et al. 1993) but provides
the least amount of data (Mayfield 1961). Both parents will feed young and remove fecal
sacs from the nest, increasing trips to and from the nest site. Parents will be particularly
vocal with frequent distress calling if you are near the nest (PRBO Conservation Science
2004). Locating nests in this stage from far away should be done to eliminate distress
calling or interruptions in feeding.
Once a nest is located a nest card should be filled out (Appendix C) and the nest
checked every 2-4 days depending on the nesting stage (PRBO Conservation Science
93
2004). The nest card shown in the appendix of this chapter or its equivalent should be
used for data comparison to other nest monitoring projects. Earlier in the nesting period
nest should be checked less frequently, while later when nestlings are near fledgling age
the nest should be checked more often to get a reliable or exact date of nest success or
failure (Mayfield 1961). Ideally, nest cards should include the date of the first egg laid,
the clutch completion date, hatch date, and fledge or fail date. Determining the nest
outcome is one of the most important aspects of nest monitoring (Mayfield 1961;1975).
The date determination of at least one of these major nesting events allows you to
determine the nest age by counting backwards using existing natural history species
accounts. Knowing the nest age provides statistical estimates of the probability of nest
survivorship for each species (Bart and Robson 1982). Clearly describe each visit on the
nest card, the date, parent or young behavior and the status of the nest. Prior to the
breeding season a cheat sheet outlining the life history attributes for each species can be
helpful in the field (PRBO Conservation Science 2004). This will help observers
determine nesting events such as hatching date (Martin and Geupel 1993). Metadata on
recording nest data and an example of a nest card can be found in Appendix C.
When the nest becomes inactive, measurements of the nest and the surrounding
vegetation should be taken. This allows for nest success or failure to be linked to nest
site selection and habitat variables (PRBO Conservation Science 2004). Generally
protocols for vegetation measurements involve measuring the nest itself, the plant the
nest is in and the vegetation within a 1.5 meter radius around the nest. Data on the cover
and height of all shrub, forb and tree species should be collected. Vegetation protocols
should be developed during the planning phases of nest monitoring after several field
94
trials to determine what data is most time efficient to collect while providing the most
amount of information (Martin and Geupel 1993). The PRBO terrestrial monitoring
handbook provides a detailed example of vegetation protocol (PRBO Conservation
Science 2004).
Most nest monitoring programs utilize a statistical analysis known as the Mayfield
Method to estimate nesting success for focal species (Mayfield 1961). The number of
days that a nest has eggs or nestlings is used to calculate daily mortality rates and
generate nesting success models for species (Mayfield 1961;1975, Hensler and Nichols
1981). Various data entry and analysis software exist using the Mayfield Method and
should be researched during the planning stages of this monitoring component (see the
California Avian Data Center at: http://data.prbo.org/cadc2/index.php?page=songbirdnest-observations).
3.33 Mist netting and banding
Minimum protocol suggests an operation of 8-12 nets open at least once every ten
days between May and August (Ralph et al. 1993), however a pilot project for
demographic work may be necessary to determine the best start date. Ralph et al. (1993)
recommend May 1- August 28 for most of temperate North America, however local
weather and climate play a critical role. A sampling interval of ten days allows for at
least one make up weekend in the case of inclement weather, divides each month of the
season into equal proportions, and provides a direct comparison to other locations (PRBO
Conservation Science 2004). Nets should not be opened in extreme moisture (usually
morning fog), rain, wind or excessive heat or cold. Nets should be closed if any of these
situations develop or in the event of bird predation in the net or a predator intently
95
watching a net (DeSante et al. 2008). Pilot banding under classes at TESC have occurred
at the campus organic farm (Figure 1a) and a tentative net locations have been mapped
(Figure 1b). This area is at the interface of a variety of habitats, has net lanes cleared and
reliably catches birds. So far banding has occurred over various seasons.
There should be a minimum of two experienced banders operating the nets at all
times, with initial opening 15 minutes after local sunrise (Ralph et al. 1993). Net checks
should be done every 30-40 minutes in ideal weather conditions, and move often if
weather is cold or hot (DeSante et al. 2008). Each bird caught should be placed in an
individual cotton bag and brought immediately back to the lab for banding once the net
checks are complete. Bags should be washed regularly to reduce the spread of avian
diseases (PRBO Conservation Science 2004). Nets should be closed 5-6 hours after you
opened the first net depending on how many net hours we wish to obtain (Ralph et al.
1993). Always record how long each net was open in the net hours log found in the
banding book. Also record total net hours in the banding journal located at the back of
the banding book. The journal entry each day should include a description of activities
and participants, time of net opening and closing, and a banding summary include
species, sexes and ago groups (PRBO Conservation Science 2004). Protocols for
collecting bird data during banding is outlined in the Identification Guide to North
American Birds: Part I (Pyle 1997). Blank and example datasheets with metadata can be
found in the Appendix C.
3.4 Monitoring Nocturnal species
Point count protocols adequately sample diurnal landbirds and are especially
useful in forested habitats (Ralph et al. 1995, Bibby et al. 2000). Nocturnal landbirds
96
such as owls however, require separate protocols which address life history, behavior and
reliable detection rates (Johnsgard 2002). Owl surveys are generally conducted via
playback of common calls given by the species surveyed for (Forsman 1983). A survey
route is predetermined in a given area with fixed call locations established randomly with
equal distance apart (Takats et al. 2001). Currently these points are not set up but should
occur outside EEON plots in order to minimize site impact. Here I describe methods
used under the Guidelines for Nocturnal Owl Monitoring in North America (Takats et al.
2001). Surveys begin within 30 minutes after local sunset and continue until completed.
Surveys are conducted for 10 minutes at each station with calls given for 2 minutes
followed by 2 minutes of listening for a response. Each survey begins with 2 minutes of
silent listening before playback to allow data to be compared across the continent
regardless of playback protocols used (Takats et al. 2001). If you hear an owl note down
the species, which minute it was heard in (first, second or both) and estimate the distance
and direction to the bird. If Great Horned Owls (Bubo virginianus) are heard at any time
at a station, the survey at that station is discontinued. Great Horned’s are known to
displace and sometimes predate other owl species (Houston et al. 1998). All nocturnal
species that respond are recorded. Focal species to survey for are to be determined but
will probably include Barred Owl (Strix varia), Western Screech Owl (Megascops
kennicottii), and Great Horned Owl (Bubo virginianus). Protocols may vary for each
species based on proven detection methods (Takats et al. 2001). The collection of data on
nocturnal species will depend on student and faculty interest and should be considered
with other monitoring priorities in mind.
3.5 Monitoring birds of the nearshore environment
97
The Evergreen State College campus comprises 3,300 feet of shoreline on Eld
Inlet, Puget Sound (Zimmer Gunsul Frasca Architects 2008). The nearshore environment
is an important component to the campus ecology and supports many avian species
(Appendix A). A variety of seabirds make the Puget Sound their home in the nonbreeding season and can easily be observed from the campus beach. Seabirds are top
predators feeding on other marine organisms (shellfish and fish) and can aid in
predictions of commercial fishery stocks (Cairns 1992). The Puget Sound Assessment
and Monitoring Program (PSAMP) indicates nearly all seabird species wintering in Puget
Sound are declined since 1979 (Seattle Audubon Society 2008). However, due to
differences in past seabird survey protocols, discrepancies in population trend data have
occurred. Seattle Audubon is developing partnerships for monitoring in South Puget
Sound where population trends in seabirds are currently unknown (Seattle Audubon
Society 2008).
The Puget Sound Seabird Survey occurs once a month from October to April.
Surveys are 15-30 minutes in duration during a four hour window with the high tide (2
hours before high tide to 2 hours after high tide). All birds on the water within 300 meters
of the shoreline are recorded with the aid of binoculars and spotting scopes. The
directional bearing from the observer to the bird is recorded along with the distance from
the horizon to the bird (in millimeters with a ruler). These distances allow individual
seabirds present but not detected to be included in abundance estimates (Buckland et al.
2001) since diving birds under water will often be missed (Seattle Audubon Society
2008). Large rafts (flocks) of seabirds are counted with a clicker counter and the distance
and bearing are taken from the middle of the raft. If monitoring the avian community of
98
the shoreline and Eld Inlet follows PSSS protocols, EAMP can provide standardized data
to Seattle Audubon and other seabird conservation organizations.
3.6 Recommendations for future projects
In addition to the monitoring work and potential studies I have presented above,
there are many opportunities for individual student research projects through the structure
of quarter long undergraduate projects linked to 16 credit programs, individual learning
contacts, and Master of Environmental Studies (MES) theses. One benefit of an existing
large ecological dataset and the establishment of a permanent plot network is that
students may pursue their own interests from a large range of possible projects. Some
examples of individual and group projects over varying lengths of time are described
below with potential research questions.
Distribution and abundance of cavity dependent birds
TESC supports over a dozen cavity dependent species (Appendix D). Snags of
varying sizes and decay classes are found in nearly every plot on campus (Appendix D).
Recent wildlife management studies with snags suggest that snag quality (possessing
characteristics necessary for cavity excavation) may be more important than snag
abundance which has been the typical habitat management strategy (Cline et al. 1980,
Bull et al. 1997, Smith et al. 2008). Cavity presence in snags and its influence on the
cavity nesting species has not been examined in Puget Sound lowland rainforests. In
hardwood forests of eastern North America, cavity presence in snags is positively
correlated to DBH and decay class (Smith et al. 2008). Does an increase in snag and
cavity availability increase abundance of snag dependent birds? What are the best
predictors of cavity presence in snags?
99
Avian use in experimental cottonwood plots
Small pockets of mature cottonwoods exist in several areas of campus (personal
observation, appendix B). These stands are sometimes near a permanent plot, but no
cottonwoods are found in any of plots themselves. Deciduous trees are limited on
campus and may influence the abundance of some songbirds (Hansen et al. 1995, Altman
and Hagar 2007). Many studies involving cottonwoods and birds occur in riparian
habitats, especially arid environments of the southwest (Johnson et al 1977).
Cottonwoods on campus occasionally occur outside of riparian habitat because of the
extremely wet environment. A nursery of young cottonwoods are ready for planting and
could assist in an experimental study of avian use in cottonwoods on campus. Does the
presence of cottonwood in lowland temperature rainforests influence the avian
community? What is the role of cottonwood in patterns of avian diversity, abundance
and nesting success?
Comparative breeding bird studies to low elevation Capitol State Forest stands
The 90,000 acre Capitol State Forest (CSF) has been owned and managed by the
Department of Natural Resources since 1957 (Felt 1975). The close proximity and
similar ecology to the TESC forest reserve provides the opportunity to conduct
comparative bird studies on forest structure and avian habitat use. This work would
constitute one of the few comparative avian studies to occur in low elevation rainforests
of the Puget Sound. For example, a one quarter project with the addition of point count
data from CSF could determine how 400 hectares of managed land at CSF compares to
the same amount of unmanaged land at Evergreen.
100
Incorporation data into the Avian Knowledge Network (AKN) database
AKN holds data from over 400,000 locations, mostly in forested environments
(Avian Knowledge Network 2008), and allows a student to work with large online
datasets and hypothesis generation. A student would gain skills in generating viable
hypotheses working with raw avian datasets and engaging in their statistical analysis.
There is a need for a student to manage the existing data from EAMP and coordinate with
AKN staff to upload our own data into the network.
3.7 Education and Outreach
3.71 Importance and need
In order for EAMP and bird conservation in general to be successful the message
must be communicated to a broad audience outside of the academic and scientific
community (Ruth et al. 2003, PRBO Conservation Science 2004). The general public,
policy makers, landowners, children of all ages are all excellent targets for outreach.
TESC has a long legacy of community involvement through events such as Super
Saturday, work study internship arrangements with local businesses and non-profits, and
a local and independent radio station. With such a relationship already established it is
both productive and feasible for EAMP to develop education and outreach projects along
with its scientific ones. The use of newspaper articles, radio/TV features, workshops, bird
festivals, demonstration sites, and school activities have proven to be very effective for
many bird observatories and long-term programs (Zack 2002, PRBO Conservation
Science 2004). Education and outreach can be facilitated through TESC’s existing
relationships with community organizations such as Black Hills Audubon Society
(BHAS) and The Nature Conservancy of Washington, and future partners such as the
101
Puget Sound Bird Observatory (PSBO).
3.72 Interpretation
Interpretation within the forested parts of campus is an excellent way to
disseminate key concepts of the ecology of our forest, including the bird life. Displays,
and information kiosks located along heavily used sections of the TESC tail system are
gaining increasing interest. Because birds are easily observed on campus, often from
trails, wildlife information kiosks should highlight birds using different components of
the habitat, especially those that are critical to reproduction or survival (Zack 2002). For
conservation efforts, kiosks should emphasize the causes of declining bird populations
and the current research projects underway to reverse them. Integrating the public into the
benefits of the TESC forest will encourage its respect and appreciation and generate
support for academic work conducted there. The general public that utilizes trails and
accesses the TESC forest may be interested in learning of the work going on there.
Participation in college sponsored events is an effective way to advertise the
college and promote issues in conservation. The college already hosts many community
events that draw large public crowds including the annual Science Fair and Super
Saturday. International Migratory Bird Day or a Puget Sound bird festival would offer
unique opportunities for the public and students to learn about bird conservation and
TESC academic programs and research. Such events may also attract potential funders,
collaborators and existing bird organizations, potentially increasing the visibility of
EEON and EAMP. For scientists who wish to interact with the public, an emphasis on
education and outreach as a part of the Evergreen science curriculum is extremely
valuable. Teaching students how to effectively communicate with others outside a
102
specific discipline is a necessary skill to take out into the world upon graduation.
3.73 Field based classes and university collaboration
Our college is unique in its widespread use of field based classes in many
disciplines. In many cases getting to the field is as easy as stepping out the backdoor into
the forest. The proximity of the forest to campus sets Evergreen apart from nearly every
other college in the country (Hall et al. 1976). Field based classes and projects located on
campus eliminate the costs of transportation, lodging, and administrative resources. The
establishment of EEON emerged from the idea that faculty could facilitate ongoing
student work directly on campus. Educating faculty about EEON and its potential
academic opportunities will then allow faculty to engage students in additional field
based curriculum right outside their door. Those that engage in projects linking field data
collection and lab analysis will save time and resources with their field site located at
their institution. As knowledge of EEON grows, students in varying disciplines may be
interested in previously unexplored aspects of the forest. Students from other colleges
around the Puget Sound often travel great distances to study the ecology or other
scientific aspects of temperate rainforests. Students at University of Washington and
University of Puget Sound travel to sites such as Wind River and H.J. Andrews
Experimental Forests, hundreds of miles away from their institutions (HJ Andrews
Experimental Forest 2002, Pacific Northwest Research Station 2007). Although these
forests offer many attributes TESC cannot, our campus forest provides a viable additional
field location and a much closer alternative for many student projects.
3.74 Banding workshops and community classes
The establishment of a banding program as part of EAMP is well underway and
103
students enrolled in undergraduate programs, individual learning contracts and four credit
electives have already learned about the operation of banding stations and obtaining data
from live birds (i.e. Ornithology, Spring 2007, Avian Monitoring and Research Methods,
Spring 2008, Avian Natural History, Summer 2008). Outside of academia, banding
workshops are an increasingly popular means of educating the public about bird
conservation while simultaneously training volunteers to become skilled in bird banding
(Ralph et al. 2005, DeSante et al. 2008). TESC currently offers community and evening
weekend special student classes. Banding workshops and classes could be effectively
offered in this context along with banding components in more traditional TESC
programs. Such workshops and classes could be offered either directly through the
college or facilitated under the direction of a local non-profit such as BHAS or PSBO.
3.75 The Puget Sound Bird Observatory
Within the past few months a new regional non-profit as been created filling an
important gap in avian conservation efforts throughout the Puget Sound basin. Newly
created, the Puget Sound Bird Observatory (PSBO) has begun to establish objectives for
the future and help fund several current projects (Puget Sound Bird Observatory 2008).
The potential for collaboration between EAMP and PSBO in regards to education and
community outreach is compelling. PSBO provides the opportunity to bring more
expertise in field ornithology to campus projects and offer internships at other sites, and
Evergreen may provide PSBO with additional membership and funding support through
the academic sector. PSBO has begun its activities by hosting several banding
workshops and MAPS sites around the Tacoma area. EAMP’s own banding efforts may
be enhanced through collaborative efforts with PSBO in terms of equipment, volunteers
104
and project analysis.
3.8 Conclusions
The Evergreen Avian Monitoring Program is designed to assist in achieving
scientific understandings and remedies for the alarming declines of many avian species
and provide students with a venue for studies in Ornithology. The program will fill a
regional gap with the collection of long-term avian data in a lowland temperate rainforest,
and will bring together many different types of monitoring projects to answer a wider
array of ornithological questions. The field site of the program is unique because it is
physically located at a major public institution which offers the advantage of a variety of
expertise and resources centrally located, as well as better use of academic research
funds. These protocols on point counting, nest searching and bird banding can be used to
develop field manuals for the various projects, evoke discussion about moving various
projects forward, and inspire students and faculty to carry out ornithological research at
the college. Such actions will help EAMP achieve its objective of becoming a long-term
monitoring program grounded in TESC tradition and shared with the greater scientific
community.
105
3.9 Literature Cited
Altman, B. 1999. Conservation strategy for landbirds in coniferous forests of western
Oregon and Washington. Oregon-Washington Partners in Flight.
Altman, B., and J. C. Hagar. 2007. Rainforest Birds: A Land Manager’s Guide to
Breeding Bird Habitat in Young Conifer Forests in the Pacific Northwest. U.S.
Geological Survey.
American Ornithologists' Union. 2008. The A.O.U check-list of North American Birds,
seventh edition. Accessed: [October 15, 2008].
Anthony, R. G., G. A. Green, E. D. Forsman, and S. K. Nelson. 1996. Avian abundance
in riparian zones of three forest types in the Cascade mountains, Oregon. Wilson
Bulletin 108:280-291.
Avian Knowledge Network. 2008. Avian Knowledge Network: An online database of
bird distribution and abundance. Ithaca, New York. Accessed: [August 12, 2008]
Barrier, J., and R. Froyalde. 1999. Thurston County Profile. Thurston County Regional
Planning Council.
Bart, J., and D. S. Robson. 1982. Estimating survivorship when the subjects are visited
periodically. Ecology 63:1078-1090.
Battaglia, D. 2000. A directory of landbird monitoring and assessment projects in Oregon
and Washington. Boise, ID.
Bibby, C. J., N. D. Burgess, D. A. Hill, S. Mustoe, and S. Lambton. 2000. Bird Census
Techniques. 2nd edition. Academic Press, San Diego, California.
Bonter, D. N., E. W. Brooks, and T. M. Donovan. 2008. What are we missing with only
ground-level mist nets? Using elevated nets at a migration stopover site. Journal
of Field Ornithology 79:314-320.
Buckland, S. T., D. R. Anderson, K. P. Burnham, J. L. Laake, D. L. Borchers, and L.
Thomas. 2001. Introduction to Distance Sampling: Estimating Abundance of
Biological Populations. Oxford University Press, Oxford, UK.
Bull, E. L., C. G. Parks, and T. R. Torgersen. 1997. Trees and logs important to wildlife
in the interior Columbia river basin. USDA Forest Service General Technical
Report PNW-GTR-391, Pacific Northwest Research Station
Buskirk, W., and J. L. McDonald. 1995. Comparison of point count sampling regimes for
monitoring forest birds. United States Forest Service General Technical Report.
Cairns, D. K. 1992. Bridging the gap between ornithology and fisheries biology: use of
seabird data in stock assessment models Condor 94:811-824.
Cline, S. P., A. B. Berg, and H. M. Wight. 1980. Snag characteristics and dynamics in
Douglas-fir forests, western Oregon. Journal of Wildlife Management 44:773786.
DeSante, D. F., K. M. Burton, P. Velez, D. Froehlich, and D. R. Kaschube. 2008. MAPS
Manual: 2008 Protocol. Contribution No. 127 of The Institute for Bird
Populations.
DeSante, D. F., T. S. Sillett, R. B. Siegel, J. F. Saracco, C. A. Romo de Vivar Alvarez, S.
Morales, A. Cerezo, D. R. Kaschube, M. Grosselet, and B. Mila. In Press. MoSI
(Monitoreo de Sobrevivencia Invernal): Assessing habitat-specific overwintering
survival of Neotropical migratory landbirds. USDA Forest Service Gen. Tech.
Rep. PSW-GTR-191.
106
Ehrlich, P. R., D. S. Dobkin, and D. Wheye. 1988. The Birder's Handbook: A field guide
to the natural history of North American birds. Simon and Schuster, New York,
NY.
Felt, M. E. 1975. Capitol Forest: the forest that came back. Washington State Department
of Natural Resources, Olympia, WA, USA.
Forsman, E. D. 1983. Methods and materials for locating and studying spotted owls. US.
Department of Agriculture, Forest Service, Pacific Northwest Forest and Range
Experiment Station; Gen. Tech. Rep. PNW-162.
Gill, F. B. 2006. Ornithology. 3rd edition. W.H. Freeman, New York.
Hagar, J. C., W. C. McComb, and W. H. Emmingham. 1996. Bird communities in
commercially thinned and unthinned Douglas-fir stands of western Oregon.
Wildl. Soc. Bull. 24:353-366.
Hall, D. J., D. H. Lockwood, and C. I. Lomax. 1976. Campus inventory and land use
planning. The Evergreen State College Press.
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.
Hensler, G. L., and J. D. Nichols. 1981. The Mayfield method of estimating nesting
success: a model, estimators and simulation results. Wilson Bulletin 93:42-53.
HJ Andrews Experimental Forest. 2002. Long Term Ecological Research (LTER).
Accessed: [October 20, 2008].
Houston, C. S., D.G. Smith, and C. Rohner. 1998. Great Horned Owl (Bubo virginianus).
Poole, A., and Gill, F., eds., The Birds of North America, No. 372
. Accessed: [November 10, 2008]
Humphery, P. S., D. Bridge, and T. E. Lovejoy. 1968. A technique for mist-netting in the
forest canopy. Journal of Field Ornithology 39:43-50.
Humple, D. L., N. Nur, G. R. Geupel, and M. P. Lynes. 2001. Female-biased sex ratio in
a wintering population of Ruby-crowned Kinglets. Wilson Bulletin 113:419-424.
Johnsgard, P. A. 2002. North American owls: biology and natural history. Smithsonian
Institution Press, Washington, D.C.
Koronkiewicz, T., E. H. Paxton, and M. K. Sogge. 2005. A technique to produce
aluminum color bands for avian research Journal of Field Ornithology 76:94-97.
Martin, T. E., and G. R. Geupel. 1993. Nest-monitoring plots: methods for locating nests
and monitoring success. Journal of Field Ornithology 64:507-519.
Marzluff, J. M., and R. Sallabanks, editors. 1998. Avian conservation : research and
management. Island Press, Washington, D.C.
Mayfield, H. F. 1961. Nesting success calculated from exposure. Wilson Bulletin 73:255261.
_____. 1975. Suggestions for calculating nest success. Wilson Bulletin 87:456-466.
National Audubon Society. 2002. The Christmas Bird Count historical results [Web
application]. Accessed: [September 29, 2008]
Nott, M. P., D. F. DeSante, P. Pyle, and N. Michel. 2005. Managing Landbird
Populations in Forests of the Pacific Northwest Region. . Publication No. 254 of
The Institute for Bird Populations.
Pacific Northwest Research Station. 2007. Wind River Experimental Forest. Accessed:
[November 1, 2008]
107
Plummer, T. 2002. Reversing Avian Population Declines in Northwest Forests. in
Partners In Flight Oregon/Washington Chapter Newsletter.
Poole, A. E. 2005. The Birds of North America Online [web application]. The Birds of
North America Online. Ithaca, NY. Accessed: [October 1, 2008].
PRBO Conservation Science. unpublished. PRBO's Terrestrial Ecology Division
Procedures Manual. PRBO Conservation Science, Stinson Beach, CA.
Puget Sound Bird Observatory. 2008. Puget Sound Bird Observatory. Seattle, WA, USA.
Accessed: [November 15, 2008]
Pyle, P. 1997. Indentification Guide to North American Birds, Part I. Slate Creek Press,
Bolinas, CA.
Ralph, C. J., S. Droege, and J. R. Sauer. 1995. Managing and monitoring birds using
point counts: standards and applications.
Ralph, C. J., G. R. Geupel, P. Pyle, T. E. Martin, and D. F. DeSante. 1993. Handbook of
field methods for monitoring landbirds.
Ralph, C. J., M.J. Widowson, R.I. Frey, P.A. Herrera, and B. P. O'Donnell. 2005. An
Overview of a Landbird Monitoring Program at Tortuguero, on the Caribbean
Coast of Costa Rica. Pages 831-838 in Proceedings of Bird Conservation
Implementation and Integration in the Americas: Proceedings of the Third
International Partners in Flight Conference. 2002 March 20-24; .831-838.
Reynolds, R. T., J. M. Scott, and R. A. Nussbaum. 1980. A variable circular-plot method
for estimating bird numbers. Condor 82:309-313.
Rich, T. D., C. Beardmore, H. H. Berlanga, P. Blancher, M. Bradstreet, G. Butcher, D.
Demarest, E. Dunn, C. Hunter, E. Inigoelias, J. Kennedy, A. Martell, A. Panjabi,
D. Pashley, K. Rosenberg, C. Rustay, S. Wendt, and T. Will. 2004. Partners in
flight North American landbird conservation plan. Partners in Flight.
Robbins, C. S., J. R. Sauer, R. S. Greenberg, and S. Droege. 1989. Population declines in
North American birds that migrate to the neotropics. Proc. Natl. Acad. Sci.
86:7658-7662.
Root, T. L., J. T. Price, K. R. Hall, S. H. Schneider, C. Rosenzweig, and J. A. P. . 2003.
Fingerprints of global warming on wild animals and plants. Nature 421:57-60.
Rosenberg, K. V. 2004. Partners in flight continental priorities and objectives defined at
the state and bird conservation region levels: Washington. Partners in Flight.
Ruggiero, L. F., L. L. C. Jones, and K. B. Aubry. 1991. Plant and animal habitat
associations in Douglas-fir forests of the Pacific Northwest: an overview. in L. F.
Ruggiero, K. B. Aubry, A. B. Carey, andM. H. Huff, editors. Wildlife and
vegetation of unmanaged Douglas-fir forests. USDA For. Serv. Gen. Tech. Rep.
Ruth, J. M., D. R. Petit, J. R. Sauer, M. D. Samuel, F. A. Johnson, M. D. Fornwall, C. E.
Korschgen, and J. P. Bennett. 2003. Science for avian conservation: priorities for
the new millennium. Auk 120:204-211.
Samuels, I. A., T. Gardali, D. L. Humple, and G. R. Geupel. 2005. Winter site fidelity
and body condition of three riparian songbird species following a fire Western
North American Naturalist:45-52.
Sauer, J. R., J. E. Hines, and J. Fallon. 2008. The North American breeding bird survey,
results and analysis 1966-2006. USGS Patuxent Wildlife Research Center.
Seattle Audubon Society. 2008. Puget Sound Seabird Survey: Survey Protocol. Seattle,
WA, USA. Accessed: [November 3, 2008]
108
Siegel, R. B., D. F. DeSante, and M. P. Nott. 2001. Using point counts to establish
conservation priorities: how many visits are optimal? Journal of Field Ornithology
72:228-235.
Siegel, R. B., R. L. Wilkerson, and S. Hall. 2004. Landbird Inventory for Olympic
National Park (2002 - 2003) Final Report. Report Cooperative Agreement
H9471011196.
Smith, C. Y., I. G. Warkentin, and M. T. Moroni. 2008. Snag availability for cavity
nesters across a chronosequence of post-harvest landscapes in western
Newfoundland Forest Ecology and Management 256:641-647.
Sparks, T. H., F. Bairlein, J. G. Bojarinova, O. Hüppop, E. A. Lehikoinen, K. Rainio, L.
V. Sokolov, and D. Walker. 2005. Examining the total arrival distribution of
migratory birds. Global Change Biology 11:22-30.
Takats, L. D., D.M. Francis, G.L. Holroyd, J.R. Duncan, K.M. Mazur, R. Cannings, W.
Harris, and D. W. Holt. 2001. Guidelines for Nocturnal Owl Monitoring in North
America Beaver Hill Bird Observatory and Bird Study.
Vickery, P. D., M. L. Hunter, and J. V. Wells. 1992. Use of a new reproductive index to
evaluate relationships between habitat quality and breeding success. Auk
109:697-705.
Wahl, T. R., B. Tweit, and S. Mlodinow. 2005. Birds of Washington: status and
distribution. Oregon State University Press, Corvallis, OR.
Weikel, J. M., and J. P. Hayes. 1999. The foraging ecology of cavity-nesting birds in
young forests of the northern Coast Range of Oregon. Condor 101:58-66.
Wilkerson, R. L., R. B. Siegel, and J. Schaberl. 2005. Landbird Inventory for Mount
Rainier National Park (2003-2004) Final Report. Report Cooperative Agreement
H9471011196.
Young, J. S., and R. L. Hutto. 2002. Use of a landbird monitoring database to explore
effects of partial-cut timber harvesting. Forest Science 48:373-378.
Zack, S. 2002. The oak woodland bird conservation plan: a strategy for protecting and
managing oak woodland habitats and associated birds in California (S. Zack, lead
author). in California Partners in Flight, editor. Version 2.0. Point Reyes Bird
Observatory Stinson Beach, CA.
Zimmer Gunsul Frasca Architects. 2008. Campus Master Plan. The Evergreen State
College.
109
Tables
Table 1. Confirmed breeding species. Astricks indicate possible focal species to be included in color banding and nest searching activities
Latin Name
Common Name
Haliaeetus leucocephalus
Bald Eagle
Accipiter cooperii
Cooper's Hawk
Patagioenas fasciata
Band-tailed Pigeon
Selasphorus rufus
Rufous Hummingbird
Sphyrapicus ruber
Red-breasted Sapsucker
Picoides pubescens
Downy Woodpecker
Picoides villosus
Hairy Woodpecker
Dryocopus pileatus
Pileated Woodpecker
Empidonax difficilis
Pacific-slope Flycatcher
Vireo gilvus
Warbling Vireo
Cyanocitta stelleri
Steller's Jay
Corvus corax
Common Raven
Poecile atricapillus
Black-capped Chickadee
Poecile rufescens
Chestnut-backed Chickadee
Sitta canadensis
Red-breasted Nuthatch
Certhia americana
Brown Creeper
Troglodytes troglodytes
*Winter Wren
Catharus ustulatus
*Swainson's Thrush
Turdus migratorius
*American Robin
Dendroica nigrescens
Black-throated Gray Warbler
Wilsonia pusilla
*Wilson's Warbler
Piranga ludoviciana
Western Tanager
Pipilo maculatus
*Spotted Towhee
Melospiza melodia
*Song Sparrow
Junco hyemalis
*Dark-eyed Junco
Carpodacus purpureus
Purple Finch
Carduelis pinus
Pine Siskin
110
Figures
Figure 1a. Arial photograph of the Evergreen State College’s organic farm showing
surrounding forested lands (Google Earth image, 2008). Inset area of the farm’s
banding site shown below.
111
Figure 1b. Evergreen State College’s organic farm banding site with experimental net
locations.
112
Appendix A. Avian species detected during the 2008 breeding season in the
Evergreen State College forest reserve, Olympia, WA. Asterisks indicate species
documented but not detected during point counts. Shorebird and waterbird species
are omitted.
Latin Name
Haliaeetus leucocephalus
Accipiter cooperii
Buteo jamaicensis
Patagioenas fasciata
Bubo virginianus*
Strix varia*
Selasphorus rufus
Megaceryle alcyon
Sphyrapicus ruber
Picoides pubescens
Picoides villosus
Colaptes auratus
Dryocopus pileatus
Contopus sordidulus
Empidonax hammondii
Empidonax difficilis
Vireo cassinii
Vireo huttoni
Vireo gilvus
Cyanocitta stelleri
Corvus brachyrhynchos
Corvus corax
Tachycineta bicolor
Tachycineta thalassina*
Petrochelidon pyrrhonota*
Stelgidopteryx serripennis*
Hirundo rustica*
Poecile atricapillus
Poecile rufescens
Psaltriparus minimus
Sitta canadensis
Certhia americana
Thryomanes bewickii
Troglodytes troglodytes
Regulus satrapa
Regulus calendula
Catharus ustulatus
Catharus guttatus
Turdus migratorius
Bombycilla cedrorum
Common Name
Bald Eagle
Cooper's Hawk
Red-tailed Hawk
Band-tailed Pigeon
Great Horned Owl
Barred Owl
Rufous Hummingbird
Belted Kingfisher
Red-breasted Sapsucker
Downy Woodpecker
Hairy Woodpecker
Northern Flicker
Pileated Woodpecker
Western Wood-Pewee
Hammond's Flycatcher
Pacific-slope Flycatcher
Cassin's Vireo
Hutton's Vireo
Warbling Vireo
Steller's Jay
American Crow
Common Raven
Tree Swallow
Violet-green Swallow
Cliff Swallow
Northern Rough-winged Swallow
Barn Swallow
Black-capped Chickadee
Chestnut-backed Chickadee
Bushtit
Red-breasted Nuthatch
Brown Creeper
Bewick's Wren
Winter Wren
Golden-crowned Kinglet
Ruby-crowned Kinglet
Swainson's Thrush
Hermit Thrush
American Robin
Cedar Waxwing
Acronym
BAEA
COHA
RTHA
BTPI
GHOW
BARO
RUHU
BEKI
RBSA
DOWO
HAWO
NOFL
PIWO
WEWP
HAFL
PSFL
CAVI
HUVI
WAVI
STJA
AMCR
CORA
TRSW
VGSW
CLSW
NRWS
BARS
BCCH
CBCH
BUSH
RBNU
BRCR
BEWR
WIWR
GCKI
RCKI
SWTH
HETH
AMRO
CEDW
x
Vermivora celata
Dendroica petechia
Dendroica coronata
Dendroica nigrescens
Dendroica townsendi
Wilsonia pusilla
Piranga ludoviciana
Pipilo maculatus
Melospiza melodia
Zonotrichia leucophrys
Junco hyemalis
Pheucticus melanocephalus
Agelaius phoeniceus
Molothrus ater
Carpodacus purpureus
Carpodacus mexicanus
Loxia curvirostra
Carduelis pinus
Carduelis tristis
Coccothraustes vespertinus
Orange-crowned Warbler
Yellow Warbler
Yellow-rumped Warbler
Black-throated Gray Warbler
Townsend's Warbler
Wilson's Warbler
Western Tanager
Spotted Towhee
Song Sparrow
White-crowned Sparrow
Oregon Junco
Black-headed Grosbeak
Red-winged Blackbird
Brown-headed Cowbird
Purple Finch
House Finch
Red Crossbill
Pine Siskin
American Goldfinch
Evening Grosbeak
OCWA
YWAR
YRWA
BTYW
TOWA
WIWA
WETA
SPTO
SOSP
WCSP
ORJU
BHGR
RWBL
BHCO
PUFI
HOFI
RECR
PISI
AMGO
EVGR
xi
Appendix B. List of trees, shrubs and herbaceous plants found in the Evergreen
State College forest reserve, Olympia, WA sampled in EEON permanent plots,
during the summer of 2008. Asterisks indicate a prominent species of the forest but
not found during sampling
Scientific name
Common Name
TREES
Abies grandis
Grand Fir
Acer macrophyllum
Big leaf maple
Alnus rubra
Red alder
Corylus cornuta
Beaked hazlenut
Cornus nuttallii
Pacific Dogwood
Ilex aquifolium
English holly
Picea sitchensis
Sitka Spruce
Pseudotsuga menziesii
Douglas-fir
Rhamnus purshiana
Cascara
Salix scouleriana
Scouler's willow
Thuja plicata
Western redcedar
Tsuga heterophylla
Western hemlock
Populus balsamifera trichocarpa* Black Cottonwood
SHRUBS
Gaultheria shallon
Salal
Mahonia nervosa
Dull Oregon-grape
Malus fusca
Oregon crabapple
Holodiscus discolor
Oceanspray
Lonicera involucrata
Black twinberry
Oemleria cerasiformis
Indian plum
Oplopanax horridus
Devil's club
Rosa gymnocarpa
Baldhip rose
Rubus parviflorus
Thimbleberry
Rubus spectabilis
Salmonberry
Rubus ursinus trailing
Blackberry
Sambucus racemosa
Red elderberry
Symphoricarpos albus
Common snowberry
Vaccinium ovatum
Evergreen huckleberry
Vaccinium parvifolium
Red huckleberry
HERBS
Adiantum pedatum
Maidenhair fern
Arabis furcata
Columbia Gorge rockcress
Asarum caudatum
Wildginger
Asplenium viride
Green Spleenwort
Athyrium filix-femina
Lady fern
Blechnum spicant
Deer fern
Bromus vulgaris
Columbia brome
Bromus sp.
Unknown grass
Carex deweyana
Dewey sedge
Acronym
ABGR
ACMA
ALRU
COCO
CONU4
ILAQ80
PISI
PSME
RHPU
SASC
THPL
TSHE
POBAT
GASH
MANE
MAFU
HODI
LOIN
OECE
OPHO
ROGY
RUPA?
RUSP?
RUUR?
SARA
SYAL
VAOV
VAPA
ADPE
ARFU
ASCA
ASVI
ATFI
BLSP
BRVU
Bromus sp.
CADE
xii
Circaea alpina
Claytonia sibirica
Corallorhiza sp.
Dactylis glomerata
Dicentra formosa
Dryopteris expansa
Epilibium angustafolium
Epilobium ciliatum
Galium aparine
Galium triflorum
Galium trifidum
Hedera helix
Hydrophyllum tenuipes
Lactuca muralis
Linnaea borealis
Lonicera ciliosa
Melica subulata
Maianthemum dilatatum
Myosotis laxa
Nemophila parviflora
Oenanthe sarmentosa
Phalaris arundinacea
Polypodium glycyrrhiza
Polystichum munitum
Pteridium aquilinum
Rubus discolor
Stellaria crispa
Tiarella trifoliata
Tolmiea menziesii
Trientalis latifolia
Trillium ovatum
Vancouveria hexandra
Veronica americana
Vicia sp.
Viola palustris
Viola sempervirens
Small enchanter's nightshade
Siberian miner's lettuce
Pacific Coralroot
Orchard grass
Pacific bleeding heart
Spiny woodfern
Fireweed
Fringed willowherb
Cleavers
Fragrant bedstraw
Small bedstraw
English ivy
Pacific waterleaf
Wall lettuce
Twinflower
Orange honeysuckle
Alaska oniongrass
False lily of the valley
Small flowered forget-me-not
Small flower nemophila
Pacific water-parsley
Reed canary grass
Licorice fern
Sword fern
Bracken fern
Himalayan blackberry
Curled starwort
Threeleaf foamflower
Piggy-back plant
Broadleaf starflower
Trillium
Inside-out flower
American brooklime
Vetch
Marsh violet
Evergreen violet
CIAL
CLSI
COME
DAGL
DIFO
DREX
EPAN2
EPCI
GAAP
GATR
GATR2
HEHE
HYTE
LAMU
LIBO3
LOCI
MESU
MADI
MYLA
NEPA
OESA
PHAR
POGL
POMU
PTAQ
RUDI2
STCR
TITR
TOME
TRLA
TROV
VAHE
VEAM
Vicia sp.
VIPA
VISE
xiii
Appendix C. Blank datasheets for the point count, banding and nesting searching
components of the Evergreen Avian Monitoring Program. Includes the following:
1) Point count field work checklist
2) Nest record cards
3) Nest record sheet
4) VCP form
5) Banding form
6) Net hours data sheet
7) Banding journal
xiv
mo/day
time
YEAR
SPECIES
NEST #
OBS
DATE FOUND
min. at nest
RESULTS
DISTRESS
after (1-4)
FLUSH (1-4)
CONSPIC
before (1-4)
(and age y)
DISTRESS
after (1-4)
FLUSH (1-4)
CONSPIC
before (1-4)
(and age y)
BHCO
CONTENTS
AGE OF
YOUNG
CONTENTS
B=building
E=eggs
Y=young
min. at nest
BHCO
CONTENTS
time
RESULTS
AGE OF
YOUNG
mo/day
OBS
DATE FOUND
CONTENTS
B=building
E=eggs
Y=young
YEAR
SPECIES
NEST #
xv
FINDING DATA (circle one)
RATE ABOVE: 0 1 2 3 4
RATE BELOW: 0 1 2 3 4
RATE APPROACH: 0 1 2 3 4
HUMAN PATH: 0 1 2 3 4
FIND DISTURBANCE: 0 1 2 3 4
FIND METHOD: F PB L SS NBC YB PY
TIMESPENT:
NUM PARENT VISITS:
SEARCH RADIUS:
NUM PREVIOUS TRY
MAP
NEST SITE DESCRIPTION:
FINDING DATA (circle one)
RATE ABOVE: 0 1 2 3 4
RATE BELOW: 0 1 2 3 4
RATE APPROACH: 0 1 2 3 4
HUMAN PATH: 0 1 2 3 4
FIND DISTURBANCE: 0 1 2 3 4
FIND METHOD: F PB L SS NBC YB PY
TIMESPENT:
NUM PARENT VISITS:
SEARCH RADIUS:
NUM PREVIOUS TRY
MAP
NEST SITE DESCRIPTION:
xvi
xvii
EAMP Variable Circular Plot (VCP) Point Count Data Form
State
Forest Type
First Name
Last Name
Point #
Time
Plot
Month
Address
Species
code
Pg. _____ of
_____
Day
Year
Telephone
Dist.
(m)
Estimated
how?
(R/WO/E)
Cue
(V/S/C)
Cluster
Size
Interval
((1,2,3 min)
Visit
Email
Behavioral
Observations
Behav. Obs: AF = aerial foraging, CO = copulation, MC = material carry, FC = food carry, NF = nest found, FL = fledglings, FS = fecal sac carry, DD = distraction display, PA = pair, DI = display
Weather Information: Please estimate temperature, cloud cover (% of sky covered by clouds), and approximate wind
speed.
jksadjlj
______˚ F or C (circle one) ______%
______ mph, knots, or kmph (circle one) ______ (Y/N)
Temperature
Cloud Cover Wind Speed
Raining
Canopy Cover (Y/N)
ENTERED__________ PROOFED __________
x
xi
xii
EVERGREEN AVIAN MONITORING PROGRAM
BANDING JOURNAL
DAY:
LOCATION:
initials _____
DATE:
exported weather data
!
WEATHER
WIND
INI
TIME
DIRECTION
(True)
TEMPERATURE:
CLOUDS
FORCE
(kmph)
HIGH:
PERSONNEL:
VISIBILITY
(km)
WEATHER
(codes)
BAROMETER
(inches)
TEMP
(Celsius)
% COVER
(tenths)
TYPE
(code)
RAIN (mm)
Current
LOW:
NUMBER OF VISITORS:
WEATHER SUMMARY:
ACTIVITIES:
xiii
Annual
July-June
EVERGREEN AVIAN MONITORING PROGRAM
Nets Open:
DAY:
DATE:
Nets Closed:
Total Net Hours:
Explanations and Times for Closed Nets:
BANDING SUMMARY
NEW CAPTURES
UNBANDED
SPECIES
HY
AHY
SY
ASY
U
RECAPTURES
HY
AHY
SY
ASY
U
SITE BIRD LIST (Seen, Heard or Banded)
xiv
Appendix D. Supplemental tables and figures
xv
x
xi
xii
xiii
xiv
Sapling counts in 44 EEON plots exemplifying the role of the deciduous community to
overall forest structure via recruitment.
Correlation between percent deciduous canopy cover and understory vegetation cover in
EEON permanent plots.
xv
