Citation
Estimating population sizes of the gray jay (perisoreus canadensis) using geographic information systems (GIS) and habitat preferences

Material Information

Title:
Estimating population sizes of the gray jay (perisoreus canadensis) using geographic information systems (GIS) and habitat preferences
Creator:
Berg, Jennifer Marie
Publication Date:
Language:
English
Physical Description:
x, 69 leaves : ; 28 cm

Subjects

Subjects / Keywords:
Gray jay ( lcsh )
Bird populations -- Estimates -- Geographic information systems ( lcsh )
Bird populations -- Estimates ( fast )
Gray jay ( fast )
Genre:
Geographic information systems. ( fast )
bibliography ( marcgt )
theses ( marcgt )
non-fiction ( marcgt )
Geographic information systems ( fast )

Notes

Bibliography:
Includes bibliographical references (leaves 66-69).
General Note:
Department of Geography and Environmental Sciences
Statement of Responsibility:
by Jennifer Marie Berg.

Record Information

Source Institution:
|University of Colorado Denver
Holding Location:
Auraria Library
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
71752695 ( OCLC )
ocm71752695
Classification:
LD1193.L547 2006m B47 ( lcc )

Full Text
ESTIMATING POPULATION SIZES OF THE GRAY JAY (PERISOREUS
CANADENSIS) USING GEOGRAPHIC INFORMATION SYSTEMS (GIS)
AND HABITAT PREFERENCES
by
Jennifer Marie Berg
B.S. University of Minnesota, 2000
A thesis submitted to the University of Colorado at Denver
and Health Sciences Center
in partial fulfillment of the requirements for the degree of
Master of Science
Environmental Sciences
2006


This thesis for the Master of Science
degree by
Jennifer Marie Berg
has been approved
by
Diana F. Tomback


Berg, Jennifer Marie (M.S., Environmental Sciences)
Estimating Population Sizes of the Gray Jay (Perisoreus canadensis) Using
Geographic Information Systems (GIS) and Habitat Preferences
Thesis directed by Professor Diana F. Tomback
ABSTRACT
We are devising a Geographic Information Systems (GIS) method to estimate wildlife
population sizes. When the species geographic range, habitat preference, territory or
home range size, and social group number per territory are known, a reasonable
estimate of population size is possible. We developed and tested this method using
two long-term study populations of the gray jay (Perisoreus canadensis): the
Algonquin Park population, Ontario, Canada, studied by Dan Strickland since the late
1960s, and the Fraser Experimental Forest (FEF) population, Colorado, studied by
Tom Nicholls from 1982-2005. The latter study lacks habitat preference and territory
size information. Without these data, our initial FEF population estimate substituted
habitat and territory information from Algonquin Park. Using color-banded gray
jays, we conducted field observations and ground-truthing in the FEF to determine
habitat preference and territory size. We learned that gray jay habitat use extends into
lodgepole pine (Pinus contorta) forest. Using this new information, we updated the
GIS based population estimate and obtained results close to the actual population


sizes based on banding data. These results suggest that GIS-based estimates can be
accurate and potentially useful in monitoring species of concern, whose geographic
range, habitat preference, territory size and group number per territory are known.
This abstract accurately represents the content of the candidates thesis. I recommend
its publication.
Signed


DEDICATION
I dedicate this thesis to Nathan Lansing and my family for their support throughout all
of my academic endeavors.


ACKNOWLEDGEMENT
My thanks to my advisor, Dr. Diana Tomback, for her support and assistance though
this research project. I also wish to thank Kristin Grompone, Mario Perez, and Dan
Maddox for their assistance with field observations. My thanks to Dr. John Wyckoff
and Dr. Michael Greene, for serving on my thesis committee. I also wish to thank
The Graduate Council of the Downtown Campus of the University of Colorado at
Denver and Health Sciences Center (UCDHSC) for a 2005 Council Award for
Graduate Student Research (CAGSR) grant. Special thanks to Tom Nicholls and Dan
Strickland for their generous assistance in the field and throughout this research.


CONTENTS
Figures.............................................................ix
Tables..............................................................x
CHAPTER
1. INTRODUCTION AND PURPOSE....................................1
2. BACKGROUND REVIEW...........................................8
Life History.............................................8
Distribution: Rangewide and Colorado....................10
Habitat.................................................13
Territories.............................................15
GIS in Avian Studies....................................17
3. SITE DESCRIPTION......................................... 19
Algonquin Park, Ontario, Canada.........................19
Fraser Experimental Forest, Colorado, United States.....22
4. MATERIALS AND METHODS......................................25
GIS Methodology.........................................25
Fraser Experimental Forest, CO..........................35
5. RESULTS.................................................. 38
Initial GIS Based Population Estimate...................38
Field Data from the Fraser Experimental Forest..........41
Corrected GIS Based Population Estimate.................50
Vll


6. DISCUSSION AND CONCLUSION.....................59
REFERENCES............................................66
vm


FIGURES
Figure
2.1 Gray Jay Breeding Bird Survey Map........................................11
2.2 Gray Jay Occurrence in Colorado..........................................12
2.3 Gray Jay Vegetation Occurrences..........................................14
3.1 Algonquin Park, Ontario, Canada..............................:...........21
3.2 Colorado Headwaters Basin................................................23
4.1 Flow-Chart Showing Steps Involved in Method..............................28
4.2 Clipped FEF Boundary with Banding Sites Labeled..........................29
4.3 Banding Station Area at FEF..............................................34
5.1 Primary Habitat for the Colorado Headwaters Basin Before Ground-Truthing.39
5.2 Primary Habitat in the FEF Before Ground-Truthing........................40
5.3 Suet Cache in Lodgepole Pine at the FEF..................................48
5.4 Engelmann Spruce/Fir Mix in the Fraser Experimental Forest...............52
5.5 Lodgepole/Spruce/Fir Mix in the Fraser Experimental Forest...............53
5.6 Lodgepole Pine in the Fraser Experimental Forest.........................54
5.7 All Lodgepole Pine Dominated Areas of the Fraser Experimental Forest.....55
5.8. Corrected Primary Habitat for the Colorado Headwaters Basin.............56
5.9 Corrected Primary Habitat in the Fraser Experimental Forest..............57
IX


TABLES
Table
4.1 FEF Intemest Distances................................................32
5.1 All Gray Jays Trapped and Banded in September, 2005................41-43
5.2. Gray Jays Trapped and Color-Banded in September, 2005................46
5.3 Activity Log for September 23rd, 2005 Observation..................45-46
5.4 Fraser Experimental Forest Vegetation Species Composition............51
5.5 Population Estimated for the FEF and Colorado Headwaters Basin.......58
x


CHAPTER 1
INTRODUCTION AND PURPOSE
The purpose of this research is to combine geographic information systems
(GIS) techniques with the information available on distribution and territory sizes to
create a method to predict the carrying capacities for a wildlife species. This method
will provide rough estimates of population size from estimates of the habitats
carrying capacity. This study focuses on the gray jay (Perisoreus canadensis) as a
test case. I propose a geospatial technique for estimating carrying capacity of gray
jays in a portion of the Rocky Mountain Front Range of Colorado. This models
validation is based on ground truthing work on a long-term banded gray jay
population in the Fraser Experimental Forest, Colorado. A second long-term banded
gray jay population in Algonquin Park, Ontario, Canada, was used for comparison to
the Colorado population.
Human activities are increasingly affecting wildlife populations and are the
cause of many of the problems affecting biodiversity in the Rocky Mountains (Baron,
2002). Such habitat alteration includes decades of mining; timber harvest; land
development; introduction of exotic species and pathogens; disturbances caused by
recreational activities; fire suppression, and pollution such as pesticide use; acid mine
drainage; cattle grazing, and transportation (Baron, 2002).
1


Odell and Knight (2001) show that bird population sizes are affected by
exurban development in the Rocky Mountains. In their study, avian densities were
altered up to 180 m from homes on the perimeter of exurban developments.
Furthermore, habitat fragmentation influences the abundance, movements, and
persistence of many species (Villard et al., 1999). This is the case particularly in
Colorado. According to the U.S. Census Bureau, the population of Colorado
increased 8.4% between 2000 and 2005,1.75 times higher than the national average
(U.S. Census Bureau, 2006a). The population of the Mountain West region increased
11.7% between 2000 and 2005,2.3 times higher than the national average (U.S.
Census Bureau, 2006b)
With increasing threats to wildlife and more pressure on the U.S. Fish and
Wildlife Service to protect species, there is a need for new, cost-efficient conservation
tools. Currently, 398 animal species are listed as threatened or endangered in the U.S.
and 14 species are proposed for listing (U.S. Fish and Wildlife Service, 2006). There
are 138 animal species waiting as candidates for listing and many more species on
individual and organization petitions for listing (U.S. Fish and Wildlife Service,
2006). In Colorado, 31 animal species are listed as either threatened or endangered
(federal and state) (CO Division of Wildlife, 2006).
In this study, a method was developed using GIS to estimate the population
size of a select bird species based on carrying capacity. Carrying capacity is defined
as the maximum population an environment can sustain without causing damage to
2


the environment, such as overhrowsing (Bolen and Robinson, 1999). It is the
maximum population of a species that an ecosystem can sustain (Molles, 2002).
Carrying capacity is measured in terms of biomass or number of animals per species
per unit area. At carrying capacity, population size is considered constant and
population growth is zero (Molles, 2002). The idea behind carrying capacity is that a
given ecosystem can only support a certain number of individuals of a particular
species. I estimated the number of gray jays that a defined area can support. Because
carrying capacity is the population size at which growth stops, the geospatial model
estimate I developed may be considered equivalent to the maximum population of
gray jays that could occur in these areas.
This study proposes to estimate gray jay population size by designing and
testing a series of steps, which may apply to other species. The steps include: 1)
development of a species-specific procedure utilizing geographic information systems
(GIS) and vegetation data layers to estimate population sizes based on carrying
capacity, 2) conducting field observations and ground-truthing the assumed habitat
preference and territory sizes with a population of banded birds, 3) adjusting the
method to obtain a more accurate population count, and 4) evaluation of the
usefulness of the resulting information. I hypothesize that the population estimate
from this approach for gray jays, in a defined area of the Rocky Mountain Front
Range in Colorado, will be close to the actual population size in the area.
3


The gray jay belongs to the order Passeriformes, family Corvidae. This
family includes crows, ravens, magpies, jays, and nutcrackers. The gray jay was
chosen as a test case for this population estimate because its habitat and range are
well known, it is a year-round resident of montane and subalpine forests, and it is a
habitat specialist.
The gray jay is not at risk in Colorado, which enables the use of habitat
information in a straightforward fashion to estimate carrying capacities. Although the
gray jay is not currently in danger, this status may change with growing human
population, increasing development, and more forest destruction. Habitat loss may
cause species decline; the effects of logging, road building, and recreation may be
especially harmful (Baron, 2002).
Other concerns that could impact gray jays include global climate change,
West Nile virus (Flavivirus), and the increase in conifer beetle infestations. The
corvids are a group of birds that has been particularly hard hit by the West Nile virus
(WNV) outbreak (Whitney, 2004). Cafffey et al. (2005) supports the possibility that
corvids are extremely vulnerable to the West Nile virus. Under laboratory conditions,
American crows (Corvus brachyrhynchos) exposed to West Nile virus were observed
to have a mortality approaching 100%. Increased corvid deaths caused by WNV in
areas throughout the United States signals an epidemic of the virus. Cafffey et al.
(2005) demonstrate how West Nile virus can drastically reduce populations of
American crows. The study documented a loss of 72% of study population members,
4


including 82% of juveniles. The American crow and other members of the Corvidae
family, such as the raven (Corvus corax), blue jay (Cyanocitta cristata), and gray jay
get sick very quickly from West Nile virus and may die in large numbers in affected
areas (Whitney, 2004). Between 1999 and 2002, more than 57,000 dead crows were
reported to authorities in the U.S. (Caffrey et al., 2005).
Gray jay habitat is affected by all major conifer beetle types, including
mountain pine beetle (Dendroctonus ponderosae), Douglas fir beetle (Dendroctonus
pseudotsugae), and spruce beetle (Dendroctonus rufipennis). One type of beetle in
particular, the mountain pine beetle, may cause great habitat losses in lodgepole pine,
because it kills healthy trees during outbreaks which may spread over thousands of
square kilometers of forest and cause catastrophic levels of tree mortality (Safranyik
et al., 2004). An intersection in all of these events will result in major losses of forest
habitat for the gray jay as well as other species. Because there are many different
ways human-caused disturbance changes Rocky Mountain forest ecosystems, and
these disturbances rarely occur alone, management solutions will need to address
many threats simultaneously (Baron, 2002).
The procedure being proposed evaluates population size on an ecosystem
level, and can factor in many threats to populations. Conservation needs related to
potential habitat loss can be addressed by monitoring the changes in gray jay carrying
capacity with the methods devised, taking habitat quality into consideration. With the
proposed method, it is possible to evaluate potential species-specific habitat on a
5


wider scale, and implement management to conserve habitat before a population
undergoes a detrimental decline due to loss of habitat. To learn more about the status
and future of these birds, it will be increasingly important to estimate carrying
capacities and changes in carrying capacities. The range and territory size are known
for many avian species based on observations, surveys (including the Breeding Bird
Survey), and literature. Except for a few species of well studied birds, namely
raptors, waterfowl and threatened or endangered species, there are no current
estimates of population size. Birds are often thought to be indicators of ecosystem
health due to their sensitivity to changes in the environment, so a method sensitive to
habitat losses from habitat destruction or alteration would lead to better management
(Maurer, 1993).
Most surveys of animal populations are based on counts of individuals
observed during a sampling period, which are used as indices to population size.
According to Link, Barker, and Sauer (1994) variability in these indices not only
reflects variability in population sizes among sites, but also variability due to the
inexactness of the counts. These surveys are merely a snapshot of the species
population size at a certain point in time. Repeated counts at survey sites can be used
to document this additional source of variability and, in some applications, to mitigate
its effects. The method developed in this study is useful over large geographical
areas as well as a longer time frame, and thus, may reduce such inherent problems
associated with other methods.
6


This method uses previously published information on range, territory size,
and habitat preference to estimate population numbers; therefore data are not needed
on age structure, sex ratios, birth rates, death rates, immigration, emigration, and the
environmental variables that influence these factors. This GIS method takes the
known species range for the gray jay and overlaps it with the gray jays preferred
habitat in order to separate out the preferred habitat from the entire range. The
territory sizes are added into the method to determine the carrying capacity. The
range, preferred habitat types for gray jays and territory size within each habitat type
are the three factors which are incorporated into the final maps used to estimate the
carrying capacity. Thus, population sizes may be estimated without the research
involved in collecting several years of actual population data for the species. This
method is appropriate for any avian or other wildlife species that exists at or near
carrying capacity and whose geographic range, territory size, and social system is
known.
7


CHAPTER 2
BACKGROUND REVIEW
The purpose of this literature review is to summarize studies of the range,
habitat, territory sizes, and life history of the gray jay, as well as current uses of GIS
in avian studies including population and habitat analysis. The GIS method proposed
here is based on availability of basic information on gray jays, including the gray jay
range, preferred habitat types and territory size within each habitat type. It is not
possible to estimate the population size of a species of concern, using this method,
without basic life history information.
Life History
Gray jays are able to survive in climatically hostile high latitude or high
elevation biomes by storing food within a territory which is used throughout the
winter by mated pairs (Strickland and Ouellet, 1993). Food storage is a corvid trait
which is the result of the evolution of behavioral and physiological traits. Gray jays
have a special adaptation for food storage: They produce copious amounts of sticky
saliva from enlarged mandibular salivary glands (one on each side of the base of the
bill) and use it to fasten food items in trees behind flakes of bark, under tufts of
lichen, in coniferous foliage or tree forks (Dow, 1963; Strickland and Ouellet, 1993).
8


Gray jays use spatial memory of cache sites to relocate food items (Bunch and
Tomback, 1986). Gray jays feed primarily on arthropods such as grasshoppers,
caterpillars, bees, wasps and beetles. However, they are generalist feeders and are not
limited in the choice of food. They have been observed feeding on seeds of conifers,
fruits, vegetation, fungi, small birds, bird eggs, small mammals, and animal carcasses
(becoming scavengers if presented with that opportunity) (Strickland and Ouellet,
1993; Ouellet, 1970).
Nicholls (2003) has been conducting long-term monitoring (1982-2005) of
birds associated with lodgepole pine dwarf mistletoe (Arceuthobium americanum),
and now focuses on gray jay longevity, site fidelity, and communication on the 93
km2 Fraser Experimental Forest, CO. Nicholls (2003) showed that gray jays are long-
lived and have a high degree of site fidelity. The oldest recorded gray jay was banded
in 1985 and was 17 years old when last captured in 2002.
Ha (1990) provided information on long-term patterns of gray jay social
organization. He estimated population size by using Jolly-Seber estimators. This is a
common mark/recapture technique in which a subset of the population is marked and
allowed back into the wild. At a later date (after enough mixing in the population
has occurred to make the marked populations distribution random), a survey is
performed and the ratio of marked to unmarked is calculated. This ratio is used as a
correction factor to arrive at a population estimate. Ha (1990) determined that gray
9


jays exhibit a monogamous and territorial mating system and are limited in dispersal
movements.
Nicholls (2003) studied the effects of WNV on the gray jays in the Fraser
Experimental Forest in Colorado, taking blood samples and testing for the antibody
that neutralizes the virus. Only two gray jays tested positive for this antibody.
However, highly susceptible species such as the corvids often die before antibodies
can develop (Nicholls, 2003). Nicholls (2003) also found there was a 37% decrease
in the number of gray jays netted in 2003 compared to 2002; and there was an overall
33% decline in the capture rate over the last 4 years, from 1999-2002, compared to
2003.
Distribution: Rangewide and Colorado
The gray jay is a year-round resident of North Americas boreal and subalpine
coniferous forests (Strickland and Ouellet, 1993). The U.S. Geological Survey
(USGS) (Fig. 2.1) Breeding Bird Survey shows the gray jay range, which extends
from north central Alaska to northern Labrador and Newfoundland, and south to
northern New York and New England in the East (USGS, 2003).
10


Figure 2.1. Gray Jay Breeding Bird Survey Map in the Contiguous United
States U.S.G.S.
In the West, the range extends through the Rocky Mountains to northern New
Mexico, central Arizona, and northern California.
In the Rocky Mountains, gray jays tend to be found at high altitudes, from
about 2,438 or 2,743 meters to timberline in Colorado and higher in the Southwest.
These elevations correspond to those of Engelmann spruce (Strickland and Ouellet,
1993). In times of extreme food scarcity there may be irruptions beyond the normal
range and occasional altitudinal movements in mountainous areas. The range remains
the same year-round, except for infrequent movements of some individuals to lower
latitudes in the east and lower altitudes in the west (Strickland and Ouellet, 1993).
11


According to the Colorado Division of Wildlife Natural Diversity Information
Source (NDIS) (2004), in Colorado gray jays are common residents in higher
mountains and occasionally are seen in the western valleys (five records) and eastern
plains near foothills (four records). Most lower-elevation sighting records occur in
fall or winter. This species wanders to low elevations far less frequently than other
montane corvids. Figure 2.2 shows the gray jay county occurrence in Colorado
(Colorado Division of Wildlife, 2004).
Figure 2.2 Gray Jay Occurrence in Colorado Natural Diversity
Information Source Colorado DOW
H Known to Occur
12


Habitat
According to Strickland and Ouellet (1993), the gray jay habitat consists
primarily of coniferous and mixed-coniferous-deciduous forests, with spruce (Picea
spp.) usually present. The Sierra Nevada is a striking example of the gray jays
apparent dependency on spruce as a habitat component. Although the gray jay range
in the Rocky Mountains stretches as far south as Arizona and New Mexico, in the
west, the range stops north of the Sierra Nevada. The Sierra Nevada coniferous
forests appear as if they could be gray jay habitat, but spruce is absent (Strickland and
Ouellet, 1993). Strickland and Ouellet (1993) state this patchy occurrence of spruce
is also the case in extreme northern California, where gray jay occurrence seems to
coincide with the isolated pockets of spruce. In Oregon, the gray jay is absent from
spruceless areas and occurs again in areas containing spruce.
Strickland (2005a) postulates that spruce is important to the gray jay due to its
effect on food storage, which enables gray jays to remain on their territory throughout
the winter. Strickland postulates that natural chemicals in the spruce penetrate stored
food and possibly act as food preservatives. This is important, because gray jays
begin storing their food in the summer, and it has to be preserved during the time
until temperatures drop. Spruce may be a vital component of gray jay habitat. The
presence of spruce may relate to the high variability that Strickland (2005a) has
observed between territory sizes and distance between territories.
13


A study was conducted by the University of Montana, Division of Biological
Sciences in association with the USFS Northern Region Landbird Monitoring
Program, on the relationship of species to habitat for just over 100 bird species that
were detected on point counts. The study depicts probabilities of occurrence on 100-
m-radius, 10-minute point counts across a series of major cover types for each of the
bird species. This study determined that gray jays were most often sighted in
relatively undisturbed conifer forests (Fig. 2.3).
Figure 2.3 Gray Jay Vegetation Occurrences University of Montana
CEDAR- HEMLOC K B^^BWB^B
SPRUCE-FIR -^BiMB^^M
LODGEPOLE jMBW
MIXED-CONIFER j
DOUGLAS-FIR
PONDEROSAPINE \ | j
GROUP SELECT
SHELTER WOOD Bi^B
SEED TREE ^^BB I
CLEARCUT BBBBBBBBB
POST-FIRE
SAGEBRUSH
GRASSLAND
AGRICULTURAL
MARSH. WETLAND
RIPARIAN SHRUB
C 0 TTONW D/AS PEN
RESIDENTAL
0 2 4 6 8 10 12
% Occurrence w/n 100 m
Although gray jays occur in harvested as well as uncut conifer forest types,
they are uniformly less abundant in the cut forest types, which suggests that they may
need older, relatively closed-canopy forests in order to maintain viable populations
14


(University of Montana, 1994). In Colorado, the gray jay primarily occurs in spruce-
fir forests, but can also be found in limber pine and lodgepole pine forests, as well as
krummholz treeline forest (Colorado Division of Wildlife, 2004). However, these
studies all report observed gray jay occurrences in these other forest types, and not
actual habitat use as an all-purpose territory.
Territories
Gray jays nest during late February and March in cold and snowy conditions,
with little fresh food supply and temperatures as low as -30 C during egg incubation
(Strickland and Ouellet, 1993; Ha, 1990). Once fledged, the young (usually one or
two) stay in the territory until the next breeding season (Ha, 1990). Gray jays do not
breed cooperatively, which means that the young do not stay to help at the nest
(Strickland and Ouellet, 1993). As stated by Strickland and Ouellet (1993), breeding
pairs occupy and defend permanent, all-purpose territories. Territories are normally
acquired by a nonbreeder moving from one territory to another and filling a breeding
vacancy (Strickland and Ouellet, 1993). These all-purpose territories vary in size
with location (Strickland and Ouellet, 1993): 0.69 km2 in Reserve de la Verendrye,
Quebec; 1.46 km2 in Algonquin Park, Ontario; 0.65 km2 in Manitoba; and 0.41 km2 in
the Yukon. The territories can be compared to similar contiguous hexagons with
nests at centers. According to Ha (1990), dispersal movements of gray jays appear to
15


be limited. Ha (1990) noted that no band recoveries were made at an appreciable
distance from the site at which the gray jay was banded.
The group size occupying these all-purpose territories is roughly two birds,
although 1 or 2 nonbreeders will often accompany the gray jay breeding pair outside
the breeding season (Ha, 1990). Young bom in the spring stay on the natal territory
until early June. The dominant brood member expels its sibling, and one or both may
remain on the natal territory (one may move into an adopted territory) until the next
breeding season, when they are expelled by the adults (Strickland, 1990; Strickland
and Ouellet, 1993). According to Strickland and Ouellet (1993), at the onset of
breeding on twelve territories in the Yukon there was a mean of 3.3 individuals per
territory; most Alaska territories had summer groups of at least 3 and often 4 jays;
and fall populations observed in Quebec and Ontario consisted of territories with 2, 3,
or 4 birds.
Ha (1990) studied the group sizes in open pine forests and pine-spruce-fir
forests in the Colorado State Forest throughout the year and determined that the mean
group size varied from 2.3 adults in January through March, 2.2 adults in April
through June, 2.9 in July through September (2.3 adults and 0.6 juveniles), and 2.3 in
October through December (2.1 adults and 0.2 juveniles). The group generally
consists of a mated pair occasionally joined by a related or adopted juvenile or an
unrelated adult jay. Ha (1990), concludes that gray jays exhibit a monogamous and
territorial mating system.
16


GIS in Avian Studies
Geospatial technologies have been used successfully in a number of avian
studies. For example, Weih et al. (1996) applied GIS and remote sensing imagery to
avian research, in a case study in the Ouachita National Forest, Arkansas. The
objectives of this study were to characterize avian abundance, species richness, and
diversity, and develop and validate methods for predicting breeding season avian
community composition (presence/absence and abundance). Weih et al. (1996) used
GIS and remote sensing in the planning and design and their study. This research
involved the use of 524 bird census points which were selected using GIS analysis.
The GIS layers used to characterize the area were hydrology, soil types, road and
stream locations, elevations, slope, and aspect. Landsat Thematic Mapper imagery
was used to identify vegetation spectral types. Future plans for this research are to
use the GIS layers to measure landscape properties surrounding each bird census plot.
In another avian study based on GIS, Kelsey and Collins (2000) used
estimates of known territory sizes in several different habitats occupied by the island
scrub-jay (Aphelocoma insularis) on Santa Cruz Island, California, to develop an
overall estimate of the population size of this endemic species. They suggest that
their baseline population estimate will be useful for monitoring the influence of future
island management efforts on the island scrub-jay.
Kelsey and Collins (2000) used estimates of territory size in several different
habitats that are occupied by the island scrub-jay and the total area of these habitats
17


on the island to develop an empirical estimate of the population size. Two study sites
were set up on the island. The main site had a population of marked birds that had
been the subject of demographic and breeding biology studies since 1974. On the
second site, Kelsey and Collins (2000) established a new marked population by
capturing breeding and non-breeding birds and banding them for individual
identification. This second population was established in order to include island
scrub-jay use of pine forest habitat in the analysis. Size varied among habitat types
for the territories mapped, with an overall mean territory size of 0.014 km2.
The territories were mapped using field observations and were represented as
points on a map. These point maps were transferred to geographically referenced
digital versions of the same photographs used in the field using the GIS program,
ArcView. From these digital point maps of territory use, the investigators were able
to calculate territory size and the total area of each vegetation type within each
territory. They calculated the size of the total breeding population on the island by
multiplying the average territory size for each habitat by the total area of that habitat
available on Santa Cruz Island.
GIS has been used in various other avian studies not related to habitat or
population studies. Other applications of GIS to avian studies include the
development of a cost-effective raptor protection plan for electrical utilities (Landon
and Harness, 2002) and mapping the occurrence of West Nile virus (USGS, 2005).
18


CHAPTER 3
SITE DESCRIPTION
Two study sites were used in this study, both sites representing long-term
studies of gray jays. The Fraser Experimental Forest (FEF), Colorado, was the site
chosen for the population estimate. However, the long-term gray jay study which
occurred at the FEF did not include information on habitat use and territory size,
which is necessary for this population estimate procedure. A second site, Algonquin
Park, Ontario, provided the necessary, missing information.
Algonquin Park. Ontario. Canada
Algonquin Provincial Park, established in 1893, is a 7725 km2 protected area
in south-central Ontario (The Friends of Algonquin Park, 2006b). Algonquin Park is
located between Ontario's Georgian Bay and the Ottawa River and lies in a transition
zone between deciduous forests typical of areas to the south of the Park, and
coniferous forests, typical of areas to the north.
The result is that both forest types are found within Park boundaries (The
Friends of Algonquin Park, 2006a and 2006b). The deciduous forest (hardwood -
forest) is the dominant habitat type in Algonquin Park covering the western two-
thirds of the park, comprised mainly of sugar maple (Acer saccharwri). Other
19


deciduous tree species in Algonquin Park are American beech (Fagus grandifolia),
yellow birch (Betula alleghaniensis), Eastern hemlock (Tsuga canadensis), and white
pine (Pinus strobus) (The Friends of Algonquin Park, 2006b). The east side of
Algonquin Park is characterized by lower elevations and warmer temperatures,
suitable conditions for coniferous forest. The white pine and the red pine (Pinus
resinosa) are the dominant tree species on the Park's east side (The Friends of
Algonquin Park, 2006b). In the northern portion of Algonquin Park spruce bogs are
common (The Friends of Algonquin Park, 2006b).
Dan Strickland began studying the gray jays in Algonquin Park over 30 years
ago. His studies began as biology and life history studies, and now focus on the
decline of gray jays in the park (2005b). Beginning in the 1970s, individual
territories in the Algonquin gray jay study area unexpectedly became vacant.
Occasionally, an affected territory may be re-occupied, but this is always just
temporary, and the slow, otherwise steady trend continues towards fewer occupied
areas. Presently the number of active territories is only 41% (N = 44) of that
observed 30 years ago.
Strickland (2005b) studies the birds along highway 60 in Algonquin Park (Fig.
3.1). The study population in the park is not bounded. This is a linear study area and
territories are observed on either side of the highway. The highway crosses all three
of the main ecosystems: black spruce (Picea mariana)/ lowland peat bogs, hardwood,
and upland coniferous forest.
20


Figure 3.1 Algonquin Park, Ontario, Canada
The gray jays preferred habitat is black spruce/lowland peat bog. According
to Strickland (2005b), 25-30 years ago gray jays were found everywhere; this is not
the case anymore. Strickland (2005b) has observed a 100% reduction in gray jays in
the hardwood habitat (hardwood forest with ribbons of coniferous habitat along
lakeshores, creeks and the edges of beaver ponds). The decline began earliest in these
territories and they are now totally devoid of gray jays. There is also about a 60%
decline in the upland coniferous habitat of white spruce (Picea glauca), white pine,
balsam fir (Abies balsamea), trembling aspen (Populus tremuloides) and white birch
(Betula papyrifera). Strickland has noted about a 30% decline in the black
21


spruce/lowland peat bog habitat, which he notes to be the gray jays preferred habitat
in Algonquin Park.
Strickland (2005b) follows about 20 nesting pairs or territories each year. The
nestlings are color-banded before fledging, and detailed observations are made on
ecology, nesting behavior, social interactions, and dispersal for nestlings and the
parents. Due to the decline in occupied territories, the study area changes in location
year-to-year to make sure at least 20 pairs are observed. Strickland (2005b) has
determined that in Algonquin Park, Ontario, gray jays appear to be dependent on
spruce. He has also determined that they occupy all purpose territories that are
approximately 1.46 km .
Strickland (2005c) estimates gray jay density by measuring distances between
nests on adjacent territories and estimating territory size with the formula: territory
size = 3/2 tan 30 d2, where d is the distance between centers of adjacent similar
hexagons. The density of gray jays in the area is calculated by multiplying density of
breeding territories by the mean number of birds on each territory. Dan Strickland
calculates density twice a year for gray jays in Algonquin Park, once in the fall
(around October 15th) and again at the start of the breeding season (around March 1st).
Fraser Experimental Forest. Colorado. United States
The Fraser Experimental Forest (FEF) was established in 1937 near Fraser,
Colorado in the central Rocky Mountains (USDA Forest Service RMRS, 2006).
22


The FEF is located in the southeastern portion of the Colorado Headwaters Basin.
Figure 3.2 shows the location of the Colorado Headwaters Basin in Colorado.
Figure 3.2 Colorado Headwaters Basin shown in green.
Colorado Division of Wildlife, 2004
The FEF spans alpine to lower subalpine ecosystems. The elevation within
the FEF varies from 2680 3900 m, and about one-third of the FEF occurs above
timberline at 3350 m. Overall, the climate is cool and humid with long, cold winters
and short, cool summers. Average annual temperature at forest headquarters (2745
m) is 0.5 C, and frost can occur any month of the year. At forest headquarters, mean
monthly temperature for January is -10 C and for July, 12.7 C. Annual
23


precipitation at forest headquarters averages 584 mm (range 432-711 mm), and
average annual precipitation over the entire FEF is 737 mm. Nearly two-thirds of the
precipitation falls as snow from October to May (USDA Forest Service RMRS,
2006).
The FEF includes subalpine forests and alpine tundra typical of the central
Rocky Mountains. In the forested areas below timberline, Engelmann spruce (Picea
engelmannii) and subalpine fir (.Abies lasiocarpa) are predominant trees at higher
elevations, or on north facing slopes, and along streams; lodgepole pine (Pinus
contorta) is the predominant tree at lower elevations and on drier upper slopes.
Tom Nicholls (University of Minnesota) has conducted a long-term (1982-
2005) study of the FEF gray jays, which began as a study to monitor birds associated
with lodgepole pine dwarf mistletoe. The study now focuses on gray jay biology,
ecology, health, longevity, site fidelity, and as of 2005, habitat use. Birds have been
trapped annually in the late summer froml982 2005 (with the exception of 2003)
using cell traps and/or 40 foot mist nets placed at the same trapping locations on the
Fraser Experimental Forest.
24


CHAPTER 4
MATERIALS AND METHODS
GIS Methodology
Geographic information systems (GIS) data were obtained from the Colorado
Division of Wildlifes Natural Diversity Information Source (NDIS). NDIS data are
provided in UTM Zonel3, North American Datum of 1927 (NAD27) projection and
were reprojected to UTM Zone 13, North American Datum of 1983 (NAD83)
projection. Metadata are found within the datasets and contain the information
decoding which vegetation type is represented at locations on the map. Metadata are
information about spatial data which describe the content, source, lineage, methods,
developer, coordinate system, extent, structure, spatial accuracy, attributes, and
responsible organization for the spatial data (Bolstad, 2002). The vegetation maps are
provided by the Colorado Vegetation Classification Project (CVCP). The CVCP is
an interagency, cooperative effort by the Colorado Division of Wildlife
(CDOW), Bureau of Land Management (BLM), and U.S. Forest Service (USFS) to
provide a landscape-level vegetation dataset for the state of Colorado.
The methods used by the CDOW, BLM, and USFS to develop these maps are
as follows. The vegetation maps are raster grid files tiled by basin, or watershed.
Raster data models define the image as a regular set of cells in a grid pattern.
25


Typically these cells are square and evenly spaced in the x and y direction and have a
cell dimension, defining the size of the cell (Bolstad, 2002). The cell dimension
specifies the length and width of the cell in surface units. In this case, the cell
dimension is specified as a square 25 meters on each side (625 m2). Individual basins
were delineated from aerial photographs taken during spring and fall of 1993 1997
by Landsat Thematic Mapper (TM) imagery and processed using an unsupervised
classification procedure. Field gathered GPS data were used to label and group the
classes into the final classification map. Landsat TM imagery was used to determine
vegetation coverage. The data are processed at a resolution of 25 meters.
A description of the specific vegetation types and/or tree species making up a
particular cover type category within a particular basin is given in the metadata. The
final classification scheme lists cover types in hierarchical form. The CVCP has
based this classification on the classification system devised by Anderson et al.
(1976) titled A Land Use and Land Cover Classification System for Use with Remote
Sensor Data.
The land cover maps were downloaded by basin (for the Colorado Headwaters
Basin), as well as the entire state of Colorado, in the form of interchange files (.e00
extension) from the website (http://ndis.nrel.colostate.edu/ftp/cvcp/index.html) and
processed using ArcGIS 9.0 through a remote desktop connection to the FAST
computer lab at the University of Colorado at Denver and Health Sciences Center.
Using conversion tools in ArcToolbox, I imported the files and changed them to
26


coverage files. I then converted the coverage files to shapefiles and added them into
ArcGIS 9.0. In order to determine the extent of the gray jays primary habitat in the
area, I assigned new attributes to the attributes table. Within the attribute table, the
vegetation cover type category attributes are originally assigned a unique number. I
reassigned them nominally and categorized the data by the nominal vegetation
attribute to distinguish the different land cover types (0 or NoData, for vegetation
types that are not primary gray jay habitat and 1, for vegetation types that are primary
gray jay habitat). Each cell from the land cover file is assigned a specific vegetation
type. Using known gray jay distribution (rangewide and in Colorado) and habitat
preferences I determined the cells in which the gray jays can be found.
I also created an additional attribute to determine the number of birds that are
present in the all-purpose territories. Attribute data are used to record the non-spatial
characteristics of an entity. They are a list of characteristics that help describe and
define the features represented in a GIS, in this case the number of birds present in
each all-purpose territory and the size of the territory. I used the field calculator
function on ArcGIS 9.0 to estimate carrying capacities. I added columns, or fields,
for each of the four month divisions (Jan-March, April-June, July-September,
October-December) in which the estimated group size is known. For the analysis, I
used 2.9 as the group size, which is the group size during September when banding
occurred, based on the research done by Ha (1990) on changes in group sizes of gray
jays in the Colorado State Forest throughout the year. I added a column for the area,
27


which is determined by multiplying the total number of cells in primary habitat by the
size of the cell, 625 m2. I added a final column for the number of territories. The
number of territories was determined by dividing the total area by the average
territory size. To determine the carrying capacity for each time of year, I multiplied
the average group size (respectively, 2.3,2.2,2.9, and 2.3) by the number of .
territories. In ArcGIS 9.0,1 was able to select the attribute table for the primary
vegetation layer to view these results. The following flow-chart (Fig. 4.1) illustrates
the steps involved in this method.
Figure 4.1 Flow-chart showing steps involved in method
This entire technique was carried out three times for the FEF. For the purpose
of this GIS analysis, the area described as the FEF is a clip of the vegetation raster
data bound by the points X-min: 419885 m, Y-min: 4412626 m, X-max: 428320 m,
28


and Y-max: 4421205 m. This area is smaller than the actual FEF boundary due to the
irregular FEF boundary which can not be clipped exactly in raster data. The area was
clipped to exclude portions of the FEF far outside of the banding sites. The clipped
FEF area is 72.2 km2, whereas the actual FEF area is 93.2 km2. The clipped FEF area
is shown with clear designations of the banding locations in Figure 4.2.
Figure 4.2 Clipped FEF Boundary with Banding Sites Labeled.
Note: banding stations shown in red are stations where color-banding
took place; stations shown in black are stations where gray jays were
banded only with aluminum band.
29


The following descriptions of cover type were taken from the Colorado
Vegetation Classification Project metadata. According to the metadata, when several
different vegetation types occur in a cover type, species are generally listed in order
of greatest occurrence. Co-dominance is defined as two species with each having
>25% and <75% crown cover over the same site.
The first of the three calculation runs was completed before ground-truthing in
the FEF and was based on the assumption that gray jays require habitat comprised
predominantly of spruce. The cover types whose attributes were reassigned a value
of 1, designating them as primary habitat, were as follows: Engelmann spruce/fir mix
defined as coniferous forest dominated by Engelmann spruce and sub-alpine fir, with
Douglas fir and lodgepole pine; spruce/fir regeneration defined as harvested
Engelmann spruce/sub-alpine fir sites, in regeneration; spruce/lodgepole pine mix
defined as coniferous forest codominated by Engelmann spruce and lodgepole pine;
spruce/fir/aspen mix defined as mixed deciduous/coniferous forest codominated by
Engelmann spruce/sub-alpine fir and trembling aspen, associated conifers include
white fir and Douglas fir. The first run also used the 1.46 km2 average territory size
from Algonquin Park, Ontario because territory size specific to the FEF was not yet
known.
The second run was completed after field observations in the FEF, which
determined the gray jays actively use lodgepole pine forest as primary habitat. This
run included the original habitat types (spruce dominated) and also included the
30


lodgepole pine dominated forests. The additional lodgepole pine cover types added
were: lodgepole pine defined as coniferous forest dominated by lodgepole pine,
associated tree species include sub-alpine fir, Douglas fir, Engelmann spruce, and
aspen; lodgepole/spruce/fir mix defined as coniferous forest codominated by
lodgepole pine, Engelmann spruce, and subalpine fir; lodgepole pine/aspen mix,
defined as mixed deciduous/coniferous forest codominated by lodgepole pine and
trembling aspen. The territory size of 1.46 m2 was kept constant again in this second
run to compare to the actual population size in order to determine its accuracy as
territory size in the FEF.
In the third run, the habitat types were the same as in the second, and the
territory size was adjusted to the specific territory size calculated for the FEF of 0.28
km I estimated this mean territory size for the FEF by using a mean intemest
distance. I estimated the mean distance between adjacent banding sites where two or
three gray jays were banded (Table 4.1).
31


Table 4.1 FEF Internest Distances,
Adjacent Banding Sites Internest Distance (km)
MTN6-MTN7 0.278
MTN7-T0WER 0.477
TOWER-MTN6 0.703
AVERAGE 0.486

Adjacent Banding Internest
Sites Distance (km)
MTN5-MTN6 0.439
MTN6-MTN7 0.278
MTN7-MTN5 0.666
AVERAGE 0.461

Adjacent Banding Internest
Sites Distance (km)
MTN4-MTN5 0.258
MTN5-MTN6 0.439
MTN6-MTN4 0.492
AVERAGE 0.396

Adjacent Banding Internest
Sites Distance (km)
MTN3-MTN4 0.617
MTN4-MTN5 0.268
MTN5-MTN3 0.652
AVERAGE 0.512

Adjacent Banding Internest
Sites Distance (km)
KING-GOS 0.89
GOS-CLEAR 0.488
CLEAR-KING 0.694
AVERAGE 0.691

Adjacent Banding Internest
Sites Distance (km)
KEJ-GOS 0.59
GOS-CLEAR 0.488
CLEAR-KEJ 0.987
AVERAGE 0.688
Adjacent Banding Sites Internest Distance (km)
FIRE 1-MUD 1 0.392
MUD 1-CLEAR 0.672
CLEAR-FIREl 0.744
AVERAGE 0.603

Adjacent Banding Internest
Sites Distance (km)
FIRE 1-MUD 1 0.392
MUD1-FIRE2 0.605
FIRE2-FIRE1 0.287
AVERAGE 0.428

Adjacent Banding Internest
Sites Distance (km)
BYR-OPEN 0.774
OPEN-ELKVIEW 0.552
ELKVIEW-BYR 0.567
AVERAGE 0.631

Adjacent Banding Internest
Sites Distance (km)
BYR-DITCH 0.967
DITCH-ELKVIEW 0.457
ELKVIEW-BYR 0.552
AVERAGE 0.659

Adjacent Banding Internest
Sites Distance (km)
DITCH-ELKVIEW 0.457
OPEN-ELKVIEW 0.552
OPEN-DITCH 0.896
AVERAGE 0.635


The banding sites where two or three gray jays were banded were chosen
because it is assumed those banding sites represent a territory. Banding sites where
more than three gray jays were banded are likely located between several territories.
The banding stations where two or three gray jays were banded are: SAGE 2, MUD
1, FIRE 1, FIRE 2, CLEARCUT, KEJ, GOSHAWK, KING, MTN 3, MTN 4, MTN
5, MTN 6, MTN 7, TOWER, BYR, OPEN, DITCH, ELKVIEW, DEADHORSE, and
POND 1. The distance between banding stations is assumed to be the equivalent of
an intemest distance, which is based on the assumption that territories may be likened
to similar contiguous hexagons with nests at centers (Strickland and Ouellet, 1993).
For the FEF the average intemest distance (d) is 0.56 km. Using the formula from
Strickland and Ouellet (1993), to calculate territory size (3/2 tan 30 d2), where d
equals the intemest distance, I calculated a minimum territory size of 0.28 km2 for the
FEF.
In the third run I defined an area closely surrounding the banding stations to
estimate population size. I did this to compare my population estimate with the actual
number of birds that were banded in this area. This will test the accuracy of the
method and the calculated FEF specific territory size, before I extrapolate the ground-
truthed information to the rest of the FEF and the Colorado Headwaters Basin. In
ArcGIS 9.0,1 used the draw tools to add oval shapes around the banding sites, with
the edges of the ovals approximately 0.14 km out from the banding sites it
encompassed. Based on the territory size calculation of 0.28 km2, 0.14 km is roughly
33


the distance from the center of a territory to its edge. I added the areas inside these
y
ovals and determined a total banding area of 8.36 km (Figure 4.3).
Figure 43 Banding Station Area at FEF.
Black dots represent banding stations.
Using the same GIS technique as above, I made four additional maps
depicting and calculating the habitat type composition for the FEF. These four maps
show the areas where the vegetation is coniferous forest dominated by Engelmann
spruce and sub-alpine fir; coniferous forest codominated by lodgepole pine,
34


Engelmann spruce, and subalpine pine with lodgepole pine having the greatest
occurrence; coniferous forest dominated by lodgepole pine; the combination of the
last two types shows all areas dominated by lodgepole pine.
Fraser Experimental Forest. CO
Ground-truthing is necessary to support or correct the habitat use and territory
assumptions made in the first GIS method run. In my ground-truthing at the Fraser
Experimental Forest, I attempted to determine: 1) the habitat use by the gray jay in
this area as opposed to the habitat use in Algonquin Park; 2) the extent of Engelmann
spruce in the study area and 3) an estimate of the territory size in the FEF as opposed
to the known territory size in Algonquin Park. I conducted this study at the Fraser
Experimental Forest, CO, where gray jay population numbers are known based on the
long-term gray jay study (1982-2003) (Nichols, 2003).
In 2005, gray jays were trapped and banded in the Fraser Experimental Forest
from September 2nd through 7th by myself, Dr. Tom Nicholls, Dr. Kurt Reed, Leann
Egeland, and assisted by Mary Lou Nicholls. When possible, trapped gray jays were
identified to age and sex. They were also weighed and measured for wing and tail
length. Gray jays were baited to traps and nets using suet, and all birds were released
after receiving an official U.S. Fish and Wildlife Service Bird Banding Laboratory
uniquely numbered metal band. A subset of gray jays was also color-banded for
specific use in this habitat study. The birds were banded with combinations of red,
35


yellow, green, blue, and white bands. When two bands were placed on the same leg,
they were noted in the log as X/X R. For example, R/A R would read red band
over aluminum band on right leg and W L Y/A R would read white band on left
leg and yellow band over aluminum band on right leg. We attempted, where
possible, to place the aluminum band on the right leg and under the color band when
a color band was also placed on the right leg.
On September 22nd through 23rd, observations of the color banded gray jays in
Fraser Experimental Forest were collected by myself, Dr. Diana Tomback and two
students from Dr. Tombacks lab, Kristin Grompone and Mario Perez. September
22nd was spent familiarizing ourselves with the site and developing an observation
plan for the 23rd. Observations began on September 23rd at approximately 9:00 am
and continued until 5:00 pm with an hour break at 12:30 pm. A total of 35 person-
field hours were spent during this observation period.
Attempts were made at several of the banding stations to lure the gray jays
into the site by playing a tape recording of their calls and placing suet at conspicuous
locations. The tape recording was obtained from Tom Nicholls and was a collection
of gray jay vocalizations he recorded in August 1993, while banding gray jays at the
FEF. Suet balls, approximately 3-4 cm in diameter, were either tied to low hanging
branches with white string or were placed on top of short snags. This was successful
at two of the locations, SAGE 4 and GOSHAWK. Observations at SAGE 4 took
36


place for 74 minutes, at two separate times. Observations at the GOSHAWK banding
site took place for a full 80 minutes from 2:40 pm to 4:00 pm.
The goal of these observations was to determine the habitat use of gray jays in
the Fraser Experimental Forest. The observation team consisted of four people
divided into two teams, so that different groups of gray jays could be followed
simultaneously. We noted which species of tree was used for caches and the diameter
at breast height (DBH) of trees used for caching. At one of the locations, we were
able to determine the maximum distance suet was cached from the food source. We
also attempted to determine if the gray jays were seen at the same location they were
banded at (to support site/territory fidelity).
Additional field observations were attempted on February 11th, 2006. The
observation team consisted of myself, Dr. Diana Tomback, and three students from
Dr. Tombacks lab: Kristin Grompone, Mario Perez, and Dan Maddox. The team
arrived at approximately 8:00 am in the town of Fraser. The temperature was -34C
when the team arrived in Fraser. We waited until approximately 9:30 am and
conducted observations until 1:30 pm. During that time, there were no gray jays nor
other birds observed. The temperature during that time reached a maximum of
approximately -11C.
Based on the ground-truthing at the Fraser Experimental Forest, I adjusted the
estimates of population size from on-site determination of habitat preference and the
newly estimated average territory size specific to the Fraser Experimental Forest.
37


CHAPTER 5
RESULTS
Initial GIS Based Population Estimate
I ran an initial GIS based population estimate for the Fraser Experimental
Forest based on habitat preference and territory size information obtained from Dan
Stricklands work in Algonquin Park, Ontario. This was done to test the methods and
the assumptions that the habitat preferences and territory sizes noted in the literature
were consistent with the FEF. Prior to our field observations in the FEF, it had not
been noted in the literature that gray jays use lodgepole pine forest as part of their
habitat. The initial estimate assumed gray jay territories occur only in areas
dominated by Engelmann spruce/fir mix. The initial estimate also assumed an
average territory size of 1.46 km The primary habitat for the Colorado Headwaters
Basin is shown in Figure 5.1 and the primary habitat for the FEF is shown in Figure
5.2. The FEF is seen as a cluster of banding locations in the lower right portion of the
map.
38


Figure 5.1 Primary Habitat for the Colorado Headwaters Basin Before Ground-Truthing
39


Figure 5.2 Primary Habitat in the FEF Before Ground-Truthing
Primary Habitat
*
J
ii
Sj&ki *
j'hpep' V| *ELK
JDR5E if.9, ELK1
id 1 ____
GAGE
SAGE4
SAGE3
SAGE2,

SAQ^1 MUD1
/
FIRE2
KEJ
+<* ESP / FIRE1 GOSHAWK
t CURNAE* *
i i r
iiV -
t^ERI # > IJL^AQU/klNG
ELKVIEW^PEH>^T,^f
jj^EE^J^?2. .!* ^N1 -'V
Â¥w ; jct ;.M^5 %
a# /* .'Wju ..-j n
rNi

0 0.5 1
Kilometers
The first calculation estimates 701.17 km2 as the area of primary habitat for
the Colorado Headwaters Basin. This results in an estimated 480.25 gray jay
territories and 1,392 gray jays in the Colorado Headwaters Basin. The area of
primary habitat in the FEF is estimated as 14.29 km containing 9.79 territories and
an estimated 28 gray jays.
40


Field Data from the Fraser Experimental Forest
The field data collection for the Fraser Experimental Forest began in
September, 2005 with trapping and banding of the gray jays. During the 2005
banding effort, a total of 88 gray jays were trapped and banded. Of the 88 total gray
jays, 38 were previously banded gray jays. The summary of all gray jays trapped and
banded in 2005 is shown in Table 5.1.
Table 5.1 All gray jays trapped and banded in September, 2005.
Birds aged as hatch year (HY) and after hatch year (AHY)
new/retrap designated by an N (unhanded birds trapped in 2005), R (banded birds
trapped in 2005), N/R (birds that were trapped twice in 2005, and were new
captures during their first capture of 2005), and R/R (birds that were trapped
twice in 2005, and were recaptures during their first capture of 2005).
Age New/Retrap Band Number Code HY/AHY Date Trapped Location Trappe<
5 R 9822-51927 318 AHY 9/2/2005 BYR
1 N 9822-52061 318 AHY 9/2/2005 BYR
6 R 1453-70390 319 AHY . 9/5/2005 CLEARCUT
3 R 9822-52050 319 AHY 9/5/2005 CLEARCUT
5 R 9822-51965 318 AHY 9/2/2005 DEADHORSE
0 N 9822-52059 318 HY 9/2/2005 DEADHORSE
1 N 9822-52060 318 AHY 9/2/2005 DEADHORSE
11 R 1003-46190 318 AHY 9/2/2005 DITCH
0 N 9822-52067 318 HY 9/2/2005 DITCH
0 N/R 9822-52067 318 HY 9/3/2005 DITCH
6 R 1453-70361 318 AHY 9/3/2005 ELK VIEW
0 N 9822-52069 318 HY 9/3/2005 ELK VIEW
1 N 9822-52070 318 AHY 9/3/2005 ELK VIEW
1 N 9822-52064 318 AHY 9/2/2005 ELK1
7 R 1453-70330 318 AHY 9/2/2005 ESP
5 R 9822-51926 318 AHY 9/2/2005 ESP
1 N 9822-52062 318 AHY 9/2/2005 ESP
1 N 9822-52063 319 AHY 9/2/2005 ESP
1 N 9822-52068 318 AHY 9/2/2005 ESP
2 R 9822-52040 319 AHY 9/5/2005 FOOL
0 N. 9822-52087 319 HY 9/5/2005 FOOL
1 N 9822-52088 319 AHY 9/5/2005 FOOL
1 N 9822-52089 319 AHY 9/5/2005 FOOL
0 N 9822-52090 319 HY 9/5/2005 FOOL
6 R 1003-46252 319 AHY 9/5/2005 FR1
4 R 9822-51961 319 AHY 9/5/2005 FR1
1 N 9822-52085 319 AHY 9/4/2005 FR2
41


Table 5.1 Cont.
Age New/Retrap Band Number Code HY/AHY Date Trapped Location Tra
1 N 9822-52086 319 AHY 9/4/2005 FR2
1 N 0962-22105 319 AHY 9/6/2005 GAGE
0 N 0962-22106 319 HY 9/6/2005 GAGE
5 R 9822-51973 319 AHY 9/6/2005 GAGE
1 N 9822-52099 319 AHY 9/6/2005 GAGE
0 N 9822-52100 319 HY 9/6/2005 GAGE
1 N 9822-52091 319 AHY 9/5/2005 GOS
1 N 9822-52092 319 AHY 9/5/2005 GOS
5 R 9822-51925 318 AHY 9/3/2005 HAZ
1 N 9822-52071 318 AHY 9/3/2005 HAZ
1 N 9822-52072 318 AHY 9/3/2005 HAZ
1 . N 9822-52073 318 AHY 9/3/2005 HAZ
1 N 9822-52074 318 AHY 9/3/2005 HAZ
0 N 0962-22107 319 HY 9/6/2005 HQ
7 R/R 1453-70330 319 AHY 9/7/2005 HQ
3 R 9822-52058 318 AHY 9/2/2005 HQ
3 R/R 9822-52058 319 AHY 9/6/2005 HQ
1 R 9822-52063 319 AHY 9/7/2005 HQ
1 N 9822-52066 318 AHY 9/2/2005 HQ
4 R 9822-52005 319 AHY 9/5/2005 KEJ
0 N 9822-52095 319 HY 9/5/2005 KEJ
8 R 1003-46217 319 AHY 9/5/2005 KING
1 N 9822-52094 319 AHY 9/5/2005 KING
7 R 1453-70337 318 AHY 9/2/2005 LOG
0 N 9822-52081 318 HY 9/4/2005 MT2
3 R 9822-52041 318 AHY 9/4/2005 MT3
1 N 9822-52077 318 AHY 9/4/2005 MT3
6 R 1003-46253 318 AHY 9/4/2005 MT4
6 R 1453-70399 318 AHY 9/4/2005 MT4
5 R 9822-51964 318 AHY 9/4/2005 MT4
8 R 1003-46214 318 AHY 9/3/2005 MT5
7 R 1453-70345 318 AHY 9/4/2005 MT5
1 N 9822-52075 318 AHY 9/3/2005 MT5
9 R 1003-46242 318 AHY 9/4/2005 MT6
0 N 9822-52079 318 HY 9/4/2005 MT6
4 R 9822-51984 318 AHY 9/4/2005 MT7
1 N 9822-52076 318 AHY 9/3/2005 MT7
1 N 9822-52080 318 AHY 9/4/2005 MT7
6 R 1453-70396 319 AHY 9/5/2005 MUD1
5 R 9822-51991 319 AHY 9/5/2005 MUD1
1 N 9822-52093 319 AHY 9/5/2005 MUD1
13 R 1003-46116 319 AHY ' 9/5/2005 MUD2
5 R 9822-51925 318 AHY 9/2/2005 OPEN
0 - N 9822-52065 318 HY 9/2/2005 OPN
3 R 9822-52054 318 AHY 9/2/2005 POND
3 R 9822-52055 318 AHY 9/2/2005 POND
7 R 1453-70382 318 AHY 9/4/2005 ROCK
1 N 9822-52078 318 AHY 9/4/2005 ROCK
42


Table 5.1 Cont.
A.ge New/Retrap Band Number Code HY/AHY Date Trapped Location Trap;
1 N 9822-52083 318 AHY 9/4/2005 ROCK
1 N 9822-52084 318 AHY 9/4/2005 ROCK
11 R 1003-46183 319 AHY 9/6/2005 SAGE2
3 R 9822-52047 319 AHY 9/6/2005 SAGE2
0 N 0962-22108 319 HY 9/6/2005 SAGE4
6 R 9822-51948 319 AHY 9/5/2005 SAGE4
6 R 9822-51951 319 AHY 9/6/2005 SAGE4
1 N 9822-52096 319 AHY 9/5/2005 SAGE4
1 N 9822-52097 319 AHY 9/5/2005 SAGE4
1 N 9822-52098 319 AHY 9/5/2005 SAGE4
1 N 0962-22113 300 AHY 9/7/2005 SAGE5
1 N 0962-22110 301 AHY 9/7/2005 ST.CAMP
1 N 0962-22111 301 AHY 9/7/2005 ST.CAMP
1 N 0962-22112 300 AHY 9/8/2005 ST.CAMP
8 R 1003-46229 318 AHY 9/2/2005 STREAM
5 R 9822-51988 318 AHY 9/2/2005 TOWER
1 N 9822-52082 318 AHY 9/4/2005 TOWER
Of the 88 total gray jays that were trapped and banded in 2005, 39 of those
were also color banded (designated by the 319 banding code in Table 1). Each color-
banded gray jay was given a unique color combination for individual identification.
The color combination, location trapped and banded, aluminum band number, age
(reported as hatch year, HY, or after hatch year, AHY), and the first year the gray jay
was trapped and banded (if a retrap) are listed in Table 5.2.
43


Table 5.2 Gray Jays Trapped and Color-Banded in September, 2005.
Band Location Trapped Band Number HY/AHY 1st Year Banded
Combination
R/A-R FIRE2 -105.87724, 39.91018 9822-52085 AHY
Y/A-R FIRE2 -105.87724, 39.91019 9822-52086 AHY
G/A-R FOOL -105.87001, 39.89795 9822-52087 HY
B/A-R FOOL -105.87001, 39.89796 9822-52088 AHY
W/A-R FOOL -105.87001, 39.89797 9822-52040 AHY 2003
R-L&A-R FOOL -105.87001, 39.89798 9822-52089 AHY
Y-L&A-R FIRE1 -105.87401, 39.90633 9822-51961 AHY 2001
G-L&A-R FIRE1 -105.87401,39.90634 1003-46252 AHY 1998
B-L&A-R FOOL -105.87001, 39.89798 9822-52090 HY
W-L&A-R GOS -105.86051, 39.90475 9822-52091 AHY
R-L&A-R GOS -105.86051,39.90476 9822-52092 AHY
R/Y-L&A-R MUD2 -105.86230,39.90876 1003-46116 AHY 1993
G L & G/A R MUD1 -105.87048,39.90876 9822-51991 AHY
B/G L & A R MUD1 -105.87048,39.90877 1453-70396 AHY 1999
W L W/A R MUD1 -105.87048,39.90878 9822-52093 AHY
R L & Y/A R KING -105.86504, 39.89749 9822-52094 AHY
Y/G L & A R KING -105.86504, 39.89750 1003-46217 AHY 1997
G L & B/A R KEJ -105.85699,39.90931 9822-52005 AHY 2002
B-L&W/A-R KEJ -105.85699, 39.90932 9822-52095 HY
W L & R/A R CLEARCUT -105.86604, 39.90368 9822-52050 AHY 2003
R/G L & A R CLEARCUT -105.86604, 39.90369 1453-70390 AHY 1999
Y-L&B/A-R SAGE4 -105.86461, 39.92639 9822-52096 AHY
G L & W/A R SAGE4 -105.86461,39.92640 9822-52097 AHY
B L & R/A R SAGE4 -105.86461, 39.92641 9822-51948 AHY 2000
W L & Y/A R SAGE4 -105.86461, 39.92642 9822-52098 AHY
R L & B/A R HQ -105.88367, 39.90527 9822-52058 AHY 2003
Y L & W/A R GAGE -105.87836, 39.91018 9822-52099 AHY
G/A-L&R-R GAGE -105.87836, 39.91019 9822-52100 HY
B L & Y/A R GAGE -105.87836, 39.91020 0962-22105 AHY
W L & G/A R GAGE -105.87836, 39.91021 0962-22106 HY
R-L&W/A-R HQ -105.88367, 39.90527 0962-22107 HY
Y- L & R/A R SAGE4 -105.86461, 39.92642 0962-22108 HY
G L & Y/A R SAGE4 -105.86461, 39.92643 9822-51951 AHY 2000
B/G-L&A-R SAGE2 -105.87426, 39.91660 1003-46183 AHY 1995
W-LB/A-R SAGE2 -105.87426, 39.91661 9822-52047 AHY 2003
Y L & G/A R GAGE -105.87836, 39.91018 9822-51973 AHY 2001
G L & B/A R HQ -105.88367,39.90527 9822-52063 AHY
B L & W/A R ST. LOUIS CAMPGROUND 0962-22110 AHY
W-L&R/A-R ST. LOUIS CAMPGROUND 0962-22111 AHY
44


During the field observations made almost two weeks after banding in the
FEF, on September 23rd, 2005, the observation team conducted successful
observations at two of the locations, SAGE 4 and GOSHAWK. Table 5.3 shows the
activity log of the birds at the two locations.
Table 5.3 Activity Log for September 23rd, 2005 Observation
PICO stands for lodgepole pine (Pinus contorta); POTR stands for quaking
aspen {Populus tremuloides)
Location Bird ID Activity Tree Species Code DBH (cm) Tree Notes
caching PICO
caching POTR
caching PICO 24.1
caching PICO 27
B-LR/A-R unsure POTR 9
caching PICO 32.3
caching POTR 9.5
caching POTR 10
SAGE 4 caching PICO 29.2
caching PICO 28.7
caching PICO 22.3
caching PICO 34.1
caching PICO 29.6
Y-LB/A-R caching PICO 23.8
caching PICO 21.8
caching PICO 8.8
caching PICO 19.8
caching POTR
45


Table 5.3 cont.
W-LA-R caching caching PICO PICO 34.7 42.7
caching PICO 27.8
caching PICO 31.3
A-R caching PICO 47.6
caching PICO 23
caching PICO 28.2
caching PICO 19.4
caching PICO 32.5
caching PICO 27.7
R-LA-R caching PICO 8.8
caching PICO 24.6
caching PICO 40.7
caching PICO 30.6
A-R caching, foraging caching, foraging PICO PICO 26.5 32.5 dead dead (pine beetle)
GOSHAWK Unbanded caching, foraging PICO 26.7
caching, foraging PICO 40.8 dead
caching, foraging PICO 25.5
caching, foraging PICO 6.7
caching, foraging PICO 20.9
W-R* caching, foraging caching, foraging PICO PICO 39.3 17 dead
caching, foraging PICO 28.4
caching, foraging PICO
caching, foraging PICO
caching, foraging PICO
*may have lost aluminum band on right leg


At SAGE 4, two gray jays were observed: B-L R/A-R and Y-L B/A-R. These
two birds were both color banded at this same location, SAGE 4, on September 5th,
2005. Observations at SAGE4 occurred at two different times. Observations began
at approximately 9:30 am and continued until 10:00 am. At 10:00 am, 30 minutes
into this observation, a Stellers jay (Cyanocitta stelleri) entered the site and began
following the pair of gray jays as they cached. About 23 minutes later (53 minutes
into the first observation) the pair of gray jays left the area. Altogether B-L R/A-R
was observed making 5 caches in lodgepole pine and 2 caches in aspen. In the first
53 minutes of observation, Y-L B/A-R was observed making 26 caches in lodgepole
pine. We returned 30 minutes later, at approximately 10:50 am, and attracted the
gray jays back to the site. In the next 21 minutes of observation, 7 additional caches
were observed. Of the 33 total observed lodgepole pine caches, 8 were confirmed
visually by finding the cache in the tree (Fig. 5.3).
47


Figure 53 Suet Cache in Lodgepole Pine at the FEF
During the 80 minute observation at GOSHAWK, at least six different gray
jays were observed. There were several unknown birds at the site, and the
observation team was unable to positively identify the birds based on bands. Four
banded birds, including three color banded birds, were observed at the site, W-L A-R,
A-R, R-L A-R, and W-R. It is possible the bird positively identified as W-R lost one
or more of its bands. The most plausible scenario is that it was banded as W/A-R and
it lost its aluminum band. W/A-R was banded at FOOL, which is approximately 1.12
km from GOSHAWK (determined in GIS analysis using distance measuring tool).
A-R is likely a bird banded in a previous year at or near GOSHAWK that was not
48


captured in 2005 and color banded. W-L A-R and R-L A-R were both banded at
GOSHAWK on September 5th, 2005.
During the observations at GOSHAWK, gray jays foraged on small branches
and on the ground. A territorial dispute occurred at this site, marked by a feeding
frenzy. The feeding frenzy involved five or more gray jays very aggressively taking
suet and caching. During this time, R-L A-R was observed acting aggressively,
following other gray jays to their cache, attempting to disrupt caching, and at one
point stealing the cache from an unknown gray jay. An attempt was made to follow
R-L A-R with the stolen cache, but we were unable to determine where the stolen
cache was taken. Displacement on the suet occurred frequently throughout this
feeding frenzy. An unknown gray jay was also observed with a large black berry in
its bill.
All 27 observed caches made during this 80 minute observation period
occurred in lodgepole pine. We were able to obtain DBH measurements for 24 of
these lodgepole pines, recorded in Table 5.3. The average DBH of the lodgepole
pines used for caching was 27.4 cm. At this site we documented the two maximum
distances the caches were from the suet source. The two maximum distances were
41.7 m and 42.5 m from the suet source, indicating a preference for larger trees.
At the Fraser Experimental Forest gray jay nests sites have not been located
and mapped as they have been in Algonquin Park. Because this important piece of
information was missing, I assumed that the birds are trapped and banded at a
49


location that is in, or is very close to, their actual territory. The long term banding
study by Tom Nicholls, allows the analysis of site gray jay site fidelity. Through my
analysis of data supplied by Tom Nichols, strong site fidelity by gray jays is
supported. Between 1982 and 2005, birds that were recaptured, were never
recaptured at more than 5 different locations, and 76.6 percent of that time they were
recaptured each time at the same location. Several of the gray jays have been trapped
and banded at the same location for many years, including band number 1043-44698,
trapped and banded 12 times at GOSHAWK, band number 1003-46210, trapped and
banded 11 times at FOOL, and band number 1163-20900, trapped arid banded 10
times, also at FOOL.
Corrected GIS Based Population Estimates
The revised GIS based population estimates for the Colorado Headwaters
Basin and the Fraser Experimental Forest is based on habitat preference and territory
size information obtained from the above mentioned field work from the FEF. After
initial analysis of the raster vegetation data for the FEF, it became clear that this area
is dominated by lodgepole pine and not Engelmann spruce. Table 5.4 shows the
species composition of the FEF based on the raster vegetation data.
50


Table 5.4 Fraser Experimental Forest Vegetation Composition
Cell Count % of total FEF area Community Description
43 0.0 Grass Dominated
79 0.1 Grass/Forb Mix
365 0.3 Sagebrush Community
166 0.1 Shrub/Grass/Form Mix
366 0.3 Sagebrush/Grass Mix
469 0.4 Sagebrush/Mesic Mtn. Shrub Mix
3290 2.8 Aspen
22406 19.4 Engelmann Spruce/Fir Mix
37839 32.7 Lodgepole Pine
263 0.2 Spruce/Fir Regeneration
47 0.0 Spruce/Lodgepole Pine Mix
39508 34.2 Lodgepole/Spruce/Fir Mix
472 0.4 Fir/Lodgepole Pine Mix
160 0.1 Spruce/Fir/Aspen Mix
5205 4.5 Lodgepole Pine/Aspen Mix
86 0.1 Rock
1497 1.3 Talus Slopes & Rock Outcrops
1529 1.3 Alpine Grass/Forb Mix
521 0.5 Subalpine Grass/Forb Mix
229 0.2 Shrub Riparian
1017 0.9 Willow
34 0.0 Herbaceous Riparian
Figures 5.4, 5.5, 5.6, and 5.7, show the vegetation composition for the Fraser
Experimental Forest. Figure 5.4 shows the locations of Engelmann spruce/sub-alpine
fir mix, dominated by Engelmann spruce. Many of the banding stations are located
outside of this habitat type. Figure 5.5 shows the locations of lodgepole
pine/Engelmann spruce/sub-alpine fir mix that is dominated by lodgepole pine.
Figure 5.6 shows the locations of lodgepole pine. Figure 5.7 shows the vegetation in
Figures 5.5 and 5.6 combined to show all lodgepole pine dominated areas.
51


Figure 5.4 Engelmann Spruce/Fir Mix in the Fraser Experimental Forest
52


Figure 5.5 Lodgepole/Spruce/Fir Mix in the Fraser Experimental Forest


Figure 5.6 Lodgepole Pine in the Fraser Experimental Forest
54


Figure 5.7 All Lodgepole Pine Dominated Areas of the Fraser Experimental Forest
55


Based on the field observations, it was determined that the gray jays in the
FEF use lodgepole pine forest, as shown by their foraging and caching in lodgepole
pine dominated areas. The corrected primary habitat, including Engelmann
spruce/subalpine fir mix as well as all lodgepole pine habitat types, for the Colorado
Headwaters Basin is shown in Figure 5.8, and the corrected primary habitat for the
FEF is shown in Figure 5.9.
Figure 5.8 Corrected Primary Habitat for the Colorado Headwaters Basin.
56


Figure 5.9 Corrected Primary Habitat in the Fraser Experimental Forest
The revised area of primary habitat for the Colorado Headwaters Basin is
2982.3 km2. Under the assumption that the territory size is the same for the gray jays
in the FEF as in Algonquin Park, Ontario, there are 2,042 territories. The estimated
number of gray jays in the Colorado Headwaters Basin is 5,923. The revised area of
y
primary habitat for the FEF is 65.85 km which includes 45.1 territories (again,
assuming a territory size of 1.46 km2). There are an estimated 130 gray jays in the
FEF.
57


To determine if the assumption is correct that the gray jays in the Fraser
Experimental Forest defend the same size territory as the gray jays of Algonquin
Park, it is necessary to determine a population estimate for the area closely clipped to
the banding sites (refer back to Fig. 4.2) using the 1.46 km2 territory size. If the
territory size in the FEF is the same, the population estimate should be close to 88, the
number of birds that were trapped and banded in this area in September, 2005.
The total area near the banding area is 8.36 km2, which equals 5.7 territories at
1.46 km2 per territory. The estimated number of gray jays in the FEF is 16. This is
well below the 88 actual gray jays trapped and banded in September 2005. Thus, it is
likely the territory size in the FEF is smaller than the territory size of Algonquin Park.
Table 5.5 shows the final revised population estimates and the values involved in the
calculations.
Table 5.5 Population Estimates for the FEF and Colorado Headwaters Basin
Banding Area FEFb CO Headwaters Basin
A. Count of Cells Containing Primary Habitat 105,366 4,771,718
B. Area of Primary Habitat (A x 0.625km2) 8.36a 65.85 2982.32
C. Mean Intemest Distance (km) 0.56 0.56 0.56c
D. Territory Size (3/2 tan 30 C2) 0.28 0.28 0.28c
E. Number Territories (B/D) 30 235 10,651
F. Number of Gray Jays per Territory in Fall 2.9 2.9 2.9
Total Number of Gray Jays (E x F) 86 682 30,888
Actual Number of Gray Jays (from banding) 88 NA NA
a area measured in GIS by boundary
b FEF refers to the clipped FEF boundary
c values extrapolated from FEF data
58


CHAPTER 6
DISCUSSION AND CONCLUSION
The goal of this study was to develop an estimated carrying capacity for the
gray jay, in selected geographic areas within its range and preferred habitat, using a
Geographic Information System (GIS) method. An initial population estimate was
made for the Fraser Experimental Forest, Colorado. The method devised to make this
population estimate involved using information about the species (in this case, the
gray jay) range, habitat preference, territory size, and animal group number per
territory with a GIS model. Because two of these critical pieces of information were
missing for the FEF, it was necessary to fill in this information from a previous study.
Dan Strickland has been studying gray jay life history in Algonquin Park, Ontario for
close to 40 years. His studies provided information on gray jay habitat preference
and territory size.
These two missing pieces of information for the FEF were substituted with
information from Dan Stricklands Algonquin Park study. The newly devised GIS
method was applied and an initial population estimate was made for the FEF and the
Colorado Headwaters Basin, which encompasses the FEF. Engelmann spruce (the
species of spruce found in Colorado) was assumed to be an essential vegetation type.
Dan Stricklands work in Algonquin Park suggests that the gray jays there are
59


dependent on black spruce. The gray jays in the Fraser Experimental Forest are
known to inhabit lodgepole pine as part of their territories, but until this study, it had
not been previously documented that they used lodgepole pine.
The second missing piece, territory size, has been studied in several other
parts of the gray jay range. The average estimated territory size from Dan
Stricklands work in Algonquin Park is 1.46 km2. This is the largest known gray jay
territory size and using this value extrapolated to the FEF would result in a
conservative population estimate (likely to be lower than the actual population).
These two missing pieces of information from the FEF were input into the devised
GIS method and produced an estimate of 1392 gray jays in the Colorado Headwaters
Basin and 28 gray jays in the FEF.
There are limitations to this initial method. After the initial estimates were
made, I began analysis of the GIS data and confirmed that Algonquin Park has an
entirely different vegetation composition than that of the Fraser Experimental Forest.
While the FEF does contain Engelmann spruce, it is dominated by lodgepole pine
habitat types. It can not be assumed that the gray jays in the FEF use the same habitat
as the gray jays in Algonquin Park or that the territory sizes are similar to those in
Algonquin Park.
A limitation of the FEF territory size is that the nest locations are not known.
In Algonquin Park, Dan Strickland likens territories to contiguous hexagons with
nests at centers. Dan Strickland determines nest locations and bases the territory size
60


on those nest locations. Since nest locations have not been identified for the FEF, we
are using the banding locations to roughly represent territory centers. This is a
limitation, although this limitation was tested by analyzing site fidelity. It could be
hypothesized that if a bird shows high site fidelity (defined as a high percent of
recaptures at the same location) that particular banding site is within that birds
territory (even if not located at the center). Because birds were recaptured 76.7
percent of the time at the same location, this hypothesis is supported. Thus, even
though the banding location may not be located at the center of a territory, it is likely
within the territory of birds trapped there.
The exceptions to this assumption are the banding locations where more than
2 or 3 gray jays (the number of gray jays occupying a territory in September) were
trapped. In 2005 this occurred at ESP, FOOL, GAGE, HAZ, HQ, ROCK, and
SAGE4. These banding locations are probably located between two adjacent
territories and attract birds from both. This is supported by the field observations at
GOSHAWK. Although only two birds were initially trapped and banded at
GOSHAWK, six birds occurred at the site during field observations. Two of the birds
were initially banded at the site, one was banded at a near-by site, and three could not
be identified due to lack of color bands.
Attracting six birds to this one area resulted in a territory dispute, marked by
aggressive behavior, attempts to block other birds from caching, and at one point a
gray jay stealing another gray jays cache. It is likely that these six birds are from
61


two (possibly three) adjacent territories bordering this one banding location. We
measured the two maximum distances caches were made from the suet source to be
0.042 and 0.043 km. Since the birds were not traveling far with the suet, it is
reasonable to assume that the site was close to two or more territories.
With more accurate information regarding habitat preference and territory size
for the gray jays of the Fraser Experimental Forest, I was able to determine a
corrected GIS based population estimate for the FEF and extrapolate this to the
Colorado Headwaters Basin. In the first correction I added the new habitat preference
information concerning lodgepole pine, but kept the territory size the same from the
Algonquin Park study, and calculated new estimates for the Colorado Headwaters
Basin and the FEF. I now determined there to be 5923 gray jays in the Colorado
Headwaters Basin and 130 gray jays in the FEF.
With accurate habitat information, I was able to test the territory size used
from Algonquin Park (1.46 km2) to see if this territory size would correctly estimate
the 88 gray jays that were trapped and banded within the banding area. Using the
Algonquin Park territory size, I estimated only 16 gray jays for the banding area, well
below the actual 88 that were banded there. It was clear the territory size for the FEF
is smaller than the territory size for Algonquin Park. Using mean distances between
adjacent banding stations, I calculated the FEF territory size to be considerable less,
at 0.28 km2.
62


Using a combination of the corrected habitat information and corrected
territory size for the Fraser Experimental Forest I was able to calculate a gray jay
population estimate for the Colorado Headwaters Basin of 30,888 gray jays and 682
gray jays in the FEF. Through the use of ground truthing, I was able to correct my
population estimate and determine a more accurate population estimate for the gray
jay.
Kelsey and Collins (2000) estimated 7,000-12,500 island scrub-jays on the
250 km Santa Cruz Island, which is equal to 28-50 island scrub-jays per 1 km My
estimate of 682 gray jays in the 72.2 km2 FEF is equal to 9 gray jays per 1 km2. After
comparison of my estimate to the estimated population size of island scrub-jays on
Santa Cruz Island, my estimate of gray jays in the study areas appears to be
reasonable.
While limitations of this study have been addressed, future research could
focus on the limitation of known territory centers. Additional work could attempt to
define actual nesting sites of the gray jays in the FEF. By determining and plotting
the nesting sites, a more accurate estimate of territory size is possible. Determination
of nesting sites, and documentation of the nesting pairs, also gives a more accurate
ground-method population estimate for comparison to the GIS based population
estimate than comparison with a banded population alone.
The effects of habitat alteration, including exurban development and the
consequences of fire, fire suppression, conifer beetle infestations, or global warming,
63


may then be used to determine effects on population sizes and to make predictions for
effective management. However, it is clear that the natural history of each species,
including home range or territory size, social system, habitat requirements, and
sensitivities, must be known and incorporated into the methodology. Ground-truthing
is an integral part of this approach, requiring a field component that involves at the
minimum bird counts, and optimally, a banded population to study.
The proposed methodology could potentially be very useful for conservation
efforts. This method can be used to estimate carrying capacity for any species whose
social systems, habitat preferences, and range are known. This method could be used
to estimate carrying capacities for many different species of birds and other wildlife,
as well as tracking changes in carrying capacities and habitat quality over time. This
approach could prove to be a powerful tool for use in avian studies. With the high
number of species waiting protection under the Endangered Species Act, there is an
urgent need for more efficient conservation tools. While species such as the gray jay
are not currently threatened nor at risk, given their numbers, they may be locally at
risk. West Nile vims may likely have one of the most ecologically devastating effects
of recent times (Caffrey et al., 2005; Whitney, 2004).
According to Kelsey and Collins (2000), in order to assess the viability of a
population, it is helpful to develop a population estimate that will determine if that
particular species is threatened and will establish a baseline with which future
estimates can be compared. This method incorporating GIS techniques with known
64


available information is applicable to all bird species and will hopefully be adopted
for monitoring programs.
65


REFERENCES
Baron, J. (Ed.). (2002). Rocky Mountain futures: An ecological perspective.
Washington D.C.: Island Press.
Bolen, E. G., & Robinson, W. L. (1999). Wildlife ecology and management (4th ed.).
New Jersey: Prentice-Hall, Inc.
Bolstad, P. (2002). GIS fundamentals: A first text on geographic information systems.
White Bear Lake, MN: Eider Press.
Bunch, K., & D. Tomback. (1986). Bolus recovery by gray jays: an experimental
analysis. Animal Behavior. 34. 754-762.
Caffrey, C., Smith, S., & Weston, T. (2005). West Nile virus devastates an American
crow population. The Condor. 107.128-132.
Colorado Division of Wildlife. (2004, December 17). Natural Diversity Information
Source. [Online] Available: http://ndis.nrel.colostate.edu/wildlifespx.asp7SpCode
=041247
Colorado Division of Wildlife. (2006, April 24). Threatened and Endangered List.
[Online] Available: http://wildlife.state.co.us/WildlifeSpecies/SpeciesOf
Concern/ ThreatenedEndangeredList/ListOfThreatenedAndEndangeredSpecies.
htm
Dow, D. D. (1963). The role of saliva in food storage by the gray jay. The Auk. A
Quarterly Journal of Ornithology. 82(21. 139-154.
The Friends of Algonquin Park. (2006a, March 1). Algonquin Provincial Park.
[Online] Available: http://www.algonquinpark.on.ca/nature.html
The Friends of Algonquin Park. (2006b, March 1). Introduction to Algonquin Park.
[Online] Available: http://www.sbaa.ca/about.asp?cn=275
Ha, J.C. & Lehner, P.N. (1990). Notes on gray jay demographics in Colorado.
Wilson Bull, 102(41 698-702.
66


Kelsey, R. & Collins, C. (2000). Estimated population size of the island scrub-jay
Aphelocoma insularis. Bird Conservation International. 10.137-148.
Landon, M. & Harness, R. (2002, November 12). Using GIS to develop a cost-
effective protection plan for electrical utilities. EDM International, Inc. [Online]
Available: http://www.edmlink.com
Link, W. A., Barker, R. J., Sauer, J. R., & Droege, S. (1994). Within-site variability in
surveys of wildlife populations. Ecology. 95.1097-1108.
Maurer, B.A. (1993). Biological diversity, ecological integrity, and neotropical
migrants: New perspectives for wildlife management. In: D. M. Finch & P. W.
Stangel (Eds.), Status and management of neotropical migratory birds. Gen. Tech
Rep. RM-229. (pp. 24-32). Fort Collins, CO: U.S. Department of Agriculture,
Forest Service, Rocky Forest and Range Experiment Station.
Molles, M. (20021. Ecology: concepts and applications. New York, NY: McGraw-
Hill.
Nicholls, T.H. (2003). Gray jay progress report, Fraser Experimental Forest, CO. U.S.
Forest Service, North Central Research Station. St. Paul, MN.
Odell, E. & Knight, R. (2001). Songbird and medium-sized mammal communities
associated with exurban development in Pitkin County, Colorado. Conservation
Biology. 15(41.1143-1150.
Ouellet H. (1970). Further observations on the food and predatory habits of the gray
jay. Canadian Journal of Zoology. 48(21. 327-330.
Safranyik, L., Shore, T., & Linton, D. (2004). Bark beetle (Coleoptera: Scolytidae)
diversity in spaced and unmanaged mature lodgepole pine (Pinaceae) in
southeastern British Columbia. Forest Ecology and Management. 200. 23-38.
University of Montana. (1994). Habitat relationship model, gray jay. [Online]
Available: http ://biology.dbs.umt.edu/dbs/habitat.htm.
U.S. Census Bureau. (2006a). National and State Population Estimates. Table 2:
Cumulative Estimates of Population Change for the United States and States, and
for Puerto Rico and State Rankings: April 1, 2000 to July 1, 2005. [Online]
Available: http://www.census.gov/popest/states/NST-pop-chg2004.html
67


U.S. Census Bureau. (2006b). National and State Population Estimates. Table 9:
Cumulative Estimates of Population Change for the United States, Regions, and
Divisions and their National Rankings: April 1,2000 to July 1,2005. [Online]
Available: http://www.census.gov/popest/states/NST-pop-chg.html
U.S. Department of Agriculture Forest Service RMRS. (2006, March 1). Fraser
Experimental Forest: About FEF. [Online] Available: http://www.fs.fed.us/rm/
ffaser/about/index.shtml.
U.S. Fish and Wildlife Service. (2006, April 10). General statistics for endangered
species [Online] Available: http://ecos.fws.gov/tess_public/SummaryStatistics.do
U.S. Geological Survey. (2003, December 17). Gray jay Breeding Bird Survey map.
[Online] Available: http://www.nenature.com/GrayJayBreedingMap.htm
U.S. Geological Survey. (2005, February 2). West Nile virus maps, bird. [Online]
Available: http://westnilemaps.usgs.gov/us_bird.html
Strickland, D. (2005a, January 10). Personal communication.
Strickland, D. (2005b, June 6). Personal communication.
Strickland, D. (2005c, June 29). Personal communication.
Strickland, D. (1990). Juvenile dispersal in gray jays: Dominant brood member expels
siblings from natal territory. Canadian Journal of Zoology, 69. 2935-2945.
Strickland, D. (1996). Trees of Algonquin Provincial Park. The Friends of Algonquin
Park. Whitney, ON.
Strickland, D. & Ouellet, H. (1993). Gray jay (Perisoreus Canadensis). The Birds of
North America: Life Histories for the 21st Century. The Academy of Natural
Sciences of Philadelphia.
Villard, M., Trzcinski, M., & Merriam, G. (1999). Fragmentation effects on forest
birds: Relative influence of woodland cover and configuration on landscape
occupancy. Conservation Biology. 13(41. 774-783.
68


Weih, R. C, Tappe P. A., Bennett C.W., & Thill, R. E. (1996). Applying GIS and
remote sensing to avian research: A case study on the Ouchita National Forest.
[Online] Available: http://gis.esri.com/library/userconf/proc96/TO300/PAP273
/P273.HTM
Whitney, H. (2004). Wildlife diseases factsheet: West Nile virus surveillance.
Government of Newfoundland and Labrador Agriculture Department of Natural
Resources.
69