Page 1

DYNAMICS AND VIABILITY OF A COUGAR POPULATION

IN THE PACIFIC NORTHWEST

By

CATHERINE MARIE SARAH LAMBERT

A thesis submitted in partial fulfillment of

the requirements for the degree of

MASTER OF SCIENCES IN NATURAL RESOURCE SCIENCES

WASHINGTON STATE UNIVERSITY

Department of Natural Resource Sciences

DECEMBER 2003

ii

To the Faculty of Washington State University:

The members of the Committee appointed to examine the thesis of

CATHERINE MARIE SARAH LAMBERT find it satisfactory and recommend that

it be accepted.

_______________________________________

Chair

_______________________________________

_______________________________________

iii

ACKNOWLEDGMENTS

This thesis is the result of a generous contribution from many agencies and

people. Research and fieldwork were funded by Bombardier Inc, Bonneville Power

Administration Fish and Wildlife Program, British Columbia Ministry of

Environment, Columbia Basin Fish and Wildlife Compensation Program, Idaho

Department of Fish and Game, U.S. Fish and Wildlife Service, U.S. Forest Service,

Washington Department of Fish and Wildlife, and Washington State University. Le

Fonds de recherche sur la nature et les technologies (Gouvernement du Québec,

Canada) provided me financial and in-kind support to attend Washington State

University.

I would like to express a heartfelt thanks to Dr. Robert Wielgus, my thesis

advisor as well as a mentor in many aspects of my life. I also greatly benefited from

the contact with Dr. Brian Dennis and Dr. Lisa Shipley, who both helped me to

become a better scientist and communicator. I would like to thank Hugh Robinson,

Donald Katnik, Ross Clarke, and Jon Almack for sharing their data and knowledge

with me, Hilary Cruickshank for her assistance in the field, and Tim Layser for his

generous hospitality. I also truly appreciated Dave Parker for making aerial

telemetry so enjoyable. Par-dessus tout, je remercie ma famille, pour l’inspiration, le

courage et l’amour qu’ils m’ont insufflé malgré les distances.

iv

DYNAMICS AND VIABILITY OF A COUGAR POPULATION

IN THE PACIFIC NORTHWEST

Abstract

By Catherine Marie Sarah Lambert, M.S.

Washington State University

December 2003

Chair: Robert W. Wielgus

Cougar (Puma concolor) populations are believed to be at high density and

increasing throughout western North America, especially in the Pacific Northwest, as

evidenced by increasing cougars/humans encounters. Harvest rates have increased

as a result. To test this hypothesis, I determined the density, fecundity, survival, and

growth rate of a cougar population in northeastern Washington, northwestern Idaho,

and southern British Columbia. From 1998 to 2003, 52 cougars were captured,

radio-collared, and monitored. I recorded fecundity through den site investigation

and snow tracking, and mortality by weekly telemetry. Survival rates were estimated

for kittens (0-1 yr), yearlings (1-2 yr), and adult (2-12 yr) males and females.

Average overall density was 1.09 cougars/100km

2

or 0.46 adults/100km

2

. I

estimated the litter size at 2.53 kittens, the birth interval at 18 months, the proportion

v

of reproductively successful females at 0.75, and the age of first reproduction at 30

months, for a maternity rate of 0.63 male or female kitten/year/adult female.

Average survival rate for all radio-collared cougars was 0.59, 0.77 for adult females,

0.44 for adult males, 0.37 for yearlings, and 0.57 for kittens. Hunting accounted for

92% of the mortalities of radio-collared cougars. Age- and sex-specific survival and

fecundity were entered into a stochastic two-sex matrix model. I used computer

simulations to determine the stochastic growth rate of the population and to assess its

viability over 25 years. The annual stochastic growth rate of this population was l =

0.80 (95%CI = 0.11). Starting with a total initial abundance of 357, the median

times to fall below a demographic collapse (N = 30 adults) and extirpation (N = 0)

were 8.5 and 25.9 years. My findings suggest that, contrary to popular belief,

cougars in the Pacific Northwest are currently at low to moderate densities and are

declining. Alternative hypotheses may account for the increased conflicts between

cougars and humans in this area.

vi

TABLE OF CONTENT

ACKNOWLEDGEMENTS…………………………………………………………iii

ABSTRACT………………………………………………………………………....iv

TABLE OF CONTENTS..………………………………………………………..…vi

LIST OF TABLES………………………………………………………………....viii

LIST OF FIGURES……………………………………………………………..…..ix

INTRODUCTION………………………………………………………………...….1

STUDY AREA………………………………………………………………...……..4

METHODS…………………………………………………………………...………6

Capture and monitoring………………………………………………………6

Density and abundance……………………………………………………….7

Fecundity……………………………………………………………………..8

Survival……………………………………………………………………….9

Population growth rate and short-term viability……………………….……10

RESULTS……………………………………………………………………….…..14

Density and abundance………………………………………………….…..14

Fecundity……………………………………………………………………14

Survival……………………………………………………………………...16

Population growth rate and short-term viability………...…………………..17

vii

DISCUSSION……………………………………………………………………….18

Density………………………………..……………………………………..18

Demographic rates…………………………………………………………..19

Population growth rate and viability…………………………………….…..21

Management implications…………..…………………………………..…...22

LITERATURE CITED.……………………………………………………………..33

viii

LIST OF TABLES

Table 1. Average density estimates for cougars in the study area from 1998 to 2003,

as defined by the annual composite adult female home ranges (AF HR) and the

abundance N of each demographic class, estimated with the number of captured

and monitored animals (Capt.) and the stable age distribution (SAD)..............24

Table 2. Annual survival rates for cougars in the Pacific Northwest (mean (sample

size)) for kittens (0-1yr), yearlings (1-2yr), and adult (2-12yr) males and

females, from 1998 to 2003.............................................................................25

ix

LIST OF FIGURES

Figure 1. Study area in the Pacific Northwest. The hatched polygons represent the

95% adaptive kernel composite home range of all cougars monitored from 1998

to 2003, and the triangles (N = 52) represent capture locations........................26

Figure 2. Monthly distribution of the cougar mortalities in the Pacific Northwest,

from 1998 to 2003. High mortality (HM) season was from 13 November to 20

May, while the low mortality (LM) season was from 21 May to 12 November.

Seasons were defined by the median date of the deaths and were used to meet

the assumption of constant mortality within each time interval required by the

Heisey-Fuller survival analysis. ......................................................................27

Figure 3. Structure of the Leslie matrix used to model population growth for female

and male cougar kittens (KitF, KitM), yearlings (YeaF, YeaM), and adult (xF,

xM, where x is the age class from 3 to 12) in the Pacific Northwest, from 1998

to 2003. In the first two rows, fecundity rates (F

x

) represented the probability

that a female of age x will give birth to a male (m

xm

S

(x-1)f

) or female (m

xf

S

(x-1)f

)

x

kitten. The survival rates for kittens (S

kf

, S

km

), yearlings (S

yf

, S

ym

), adult males

(S

xm

) and adult females (S

xf

) were the elements of the second-subdiagonal.......28

Figure 4. Adult and total cougar density in the Pacific Northwest, from 1998 to

2003, calculated with the 95% composite adaptive kernel home range method

for adult females, and projected with both capture data and stable age

distribution (SAD) vector to calculate adult male and yearling densities. Kitten

density was calculated with the SAD...............................................................29

Figure 5. Annual survival rates of adult females and deterministic growth rates l

1

(dominant eigenvalue of annual matrix model), between 1998 and 2003, for

cougars in the Pacific Northwest.....................................................................30

Figure 6. Simulated trajectory of the cougar population in the Pacific Northwest,

based on survival and fecundity rates from 1998 to 2003. The squares represent

the annual average abundances, the vertical lines are standard deviations, and

the empty circles are maximum and minimum values obtained in 5,000

simulations......................................................................................................31

xi

Figure 7. Time to fall from 357 cougars to demographic collapse (N = 79,

equivalent to 30 adults) and extirpation (N = 0), based on vital rates estimated

for cougars in the Pacific Northwest from 1998 to 2003, and simulated 5,000

times over 25 years. ........................................................................................32

xii

Dedication

I dedicate this thesis to the Cougars themselves

–fascinating, powerful, vulnerable animals–

May they be inhabit our forests and mountains for the centuries to come…

Go confidently in the direction of your dreams. Live the life you have imagined.

-Henry David Thoreau (1817 - 1862)

1

INTRODUCTION

Cougar (Puma concolor) populations are believed to be increasing

throughout western North America (Riley and Malecki 2001), as evidenced by the

increased frequency of reported encounters with humans (Beier 1991, Chester 2003)

and number of cougar harvested (Dawn et al. 2003). This is especially true in

Washington state (WA), where there is a current controversy about harvest

regulations (Martorello 2003). Although bounties for predators were abolished in

1960, the number of cougars hunted in the state during the past century has increased

steadily (Spencer et al. 2001). In 1996, hunting with hounds became illegal after

voter initiative 655 was passed, but the hunting season’s length and bag limit were

both increased to compensate. The years following initiative 655 were marked by a

constant increase in the number of complaints related to interactions between

humans and cougars –annual registered complaints rose from 247 in 1995 (the year

before the ban) to 495 in 1996, and 927 in 1998 (Washington Department of Fish

and Wildlife 1999). Since then, complaints have averaged about one thousand a

year. Because of increased complaints, the Engrossed Substitute Senate Bill (ESSB)

5001 was passed in 2001, allowing the use of hounds for safety removals. The 2001

Washington game status report (Martorello 2001) suggested that cougars were stable

to increasing statewide and were moderately exploited, so hunting regulations have

2

continued to become more permissive. To further satisfy local concerns, emergency

cougar hunts were adopted in March 2003 in Stevens, Ferry and Okanagan Counties,

in northeastern WA. Recent reports from neighboring Idaho (ID) (Nadeau 2003) and

British Columbia (BC) (Austin 2003) showed that hunting effort has also been

constantly increasing from 1998 to 2002, with harvest rates as much as 700% higher

than 20 years ago.

Many hypotheses were proposed to account for the increase number of

cougar complaints in WA. The most popular belief is that this indicates a very high

cougar density and increasing cougar population. Alternatively, more frequent

encounters and complaints may be caused by encroachment of expanding suburbs

into cougar habitat (Dickson and Beier 2002, Spencer et al. 2001). Hunting itself

could also be indirectly responsible for the increased frequency in cougar/human

conflicts. Carnivore populations under heavy hunting pressure may have a very

young age structure, as seen in African lions (Panthera leo) (Smuts 1978) and

wolves (Canis lupus) (Jedrezejewska et al. 1996). Removing older animals can

increase the density of immigrating/settling juveniles or sub-adults from adjacent

areas (Wielgus et al. 2001), which are often responsible for most of the attacks on

humans and livestock (Beier 1991). Finally, low levels of stakeholder acceptance of

cougars can amplify the perception that cougars are increasing (Riley and Decker

3

2000, Riley and Malecki 2001), and the large number of complaints could be a local

backlash to initiative 655.

My primary objective was to test the hypothesis that cougars are at very high

density and increasing, by determining the density and growth rate of the population

in northeastern WA, northwestern ID and southern BC from 1998 to 2003. I

examined the vital rates of cougars in two overlapping areas. Between 1998 and

2001, 32 individuals were radio-collared and monitored in the Selkirk Mountains in

northern ID (Panhandle area), southeastern BC (Kootenay region), and northeastern

WA (PendOreille county) (Katnik 2002). Seven of these 32 cougars were monitored

throughout the length of the study. Twenty-one other cougars were radio-collared

and monitored in the Colville National Forest in northeastern WA (Ferry and Stevens

counties) between 2002 and 2003.

4

STUDY AREA

The study area is located in the Northern Rocky Mountain US ecosystem

province (Bailey 1995) and the Southern Interior Mountains ecoprovince of British

Columbia (Demarchi 1996) (Fig.1). The minimum convex polygon of all cougar

locations in this study encompassed approximately 32,800 km

2

(48°21¢N-49°35¢N,

115°57¢W-119°04¢W), but the composite 95% contour kernel estimate for all

animals was only 5,600 km

2

. The area is characterized by rugged terrain with

numerous ridges (1,500-2,000 m) interspersed by valleys (500 m). Average winter

temperature is 0°C and the average summer temperature is 22°C. Precipitation falls

mostly in the winter and averages 510 to 1,020 mm per year.

A mixed evergreen-deciduous forest dominates the landscape. In lower

elevations, western red-cedar (Thuja plicata) and western hemlock (Tsuga

heterophylla) are the dominant tree species in mature forests, with black cottonwood

(Populus balsamifera trichocarpa) the climax in moister areas. Open mixed stands

of douglas-fir (Pseudotsuga menziesii) and ponderosa pine (Pinus ponderosa) are

common on more xeric south-facing slopes (Ketcheson et al. 1991). In higher

elevations, white spruce (Picea glauca) dominates the climax forest, with subalpine

5

fir (Abies lasiocarpa) composing the understory and lodgepole pine (Pinus contorta)

common following fire (Coupe et al. 1991).

White-tailed deer (Odocoileus virginianus) were the most abundant ungulate,

but mule deer (Odocoileus hemionus, see Robinson et al. 2002), elk (Cervus

elaphus), moose (Alces alces) and mountain caribou (Rangifer tanrandus caribou,

see Katnik 2002) were also present. Common predator species, beside cougars,

included coyotes (Canis latrans), black bears (Ursus americanus), bobcats (Lynx

rufus), and, to a lesser extent, grizzly bears (Ursus arctos), lynx (Lynx canadensis),

and wolves (Canis lupus).

Cougar harvest was permitted throughout the study area but regulations

varied amongst jurisdictions. In WA, the hunting season lasted from 1 August to 15

March from 1998 to 2003. Hounds were prohibited in 1998, but their use was

reintroduced in 2000 to meet public safety needs. In ID, the hunting season lasted

from 15 September to 31 March for the 1998-99 and 1999-2000 seasons, and from

30 August to 31 March for the seasons 2000-2001 to 2002-2003. Hounds were

allowed during most of the season, but starting in 1999 they were prohibited between

September and December. In BC, cougar hunting in the southeastern Kootenay

region lasted from 10 September to 28 February, and lasted until 31 March for the

southwestern Kootenays. Hounds were allowed during the entire season.

6

METHODS

Capture and monitoring

The study area was intensively searched for cougar tracks during the snow-

covered period (November to April) each year from 1998 to 2003. Hounds were

released on fresh tracks and used to tree cougars (Hornocker 1970). Efforts were

made to capture and collar all cougars larger than about 30kg (yearlings and adults).

Treed cougars that were not too high for safe capture (those £5 m from the ground)

were immobilized by injecting 0.4 ml/10 kg of body mass of ketamine hydrochloride

(200 mg/ml) and xylazine hydrochloride (20 mg/ml), with a projectile dart in the

hindquarter (Ross and Jalkotzy 1992, Spreadbury et al. 1996). Cougars were fitted

with mortality-sensing VHF radio-collars, and administrated 0.2 mg/kg of body mass

of yohimbine hydrochloride, a chemical antagonist of the immobilizing drugs (Seal

et al. 1987). Cougars were classified either as kitten (0-1 yr), yearling (1-2 yr) or

adult (2+ yrs), based on their general appearance and tooth wear (Ashman and Greer

1976). Released cougars were located about once a week by aerial or ground

telemetry, from 14 December 1998 to 20 May 2003.

7

Density and abundance

I estimated minimum mean annual density (adults/100km

2

and

cougars/100km

2

) by counting all captured animals in the annual 95% adaptive kernel

composite home range of the adult females (Wielgus et al. 1994), using the animal

movement extension for ArcView‚ (ESRI, Redlands, California, U.S.A.) developed

by Hooge and Eichenlaub (1997). Densities of adult females were derived from how

many were monitored annually, divided by the area of their annual 95% adaptive

kernel composite home range. I believe that most adult females were captured in

these relatively small (1,800-5,600 km

2

, see Table 1) annual composite home ranges.

Densities of adult males and yearlings were estimated using two different techniques.

The first method counted how many animals were captured or monitored each year,

divided by the adult female composite home range area. The second method, based

on survival and fecundity rates, used the stable age distribution (dominant right

eigenvector, Morris and Doak 2002:222) to determine the proportion of adult males

and yearlings present each year, according to adult female density. These

proportions were then multiplied by the number of adult females in the annual

composite home range. In both cases, kitten density was assessed using the stable

age distribution method. Average cougar abundance for the entire study area was

obtained by extrapolating the average density estimate over the 32,800km

2

minimum

convex polygon of all studied animals.

8

Fecundity

The maternity rates (m

x

) were the average number of kittens of each sex (m

xm

,

m

xf

) produced during 1 year by any mother of age x (Case 2000). Because cougars

can reproduce all year long (birth-flow population), frequent checks of potentially

reproductive females were conducted. The reproductive status at capture was

determined by presence of milk in nipples, color and size of the nipples (to detect

previous maternities), and sightings of kittens or yearlings. Subsequent changes in

the reproductive status were detected by weekly telemetry locations. A decrease in a

female’s movements was investigated as a possible birth event (Beier et al. 1995,

Hemker et al. 1986). I determined litter size by investigating the dens or by track

counts (Banhurst and Lindzey 1989, Kerley et al. 2002, Ross and Jalkotzy 1992) for

kittens from 0-3 months. I calculated annual maternity rates by dividing the average

litter size by the average successful interbirth interval, and multiplying by the

proportion of kitten of each sex and the proportion of reproductively successful adult

females (i.e., pregnant or involved in raising a litter). The latter parameter is

sometimes ignored in calculating maternity rates, but failing to consider the

proportion of unsuccessfully reproducing females would bias kitten production

upwards by including only successful females during short study periods (Wielgus et

al. 2001). Fecundity rates (F

x

) of my post-breeding model were defined as F

x

=

m

x+1

s

xf

(Case 2000), where s

xf

are the annual survival rates of females of age x. I

9

assumed F

x

had the same standard deviations as s

xf

(see below), because sample sizes

were too small to estimate the standard deviations for fecundity.

Survival

I determined the survival of kittens by comparing direct observations or track

counts on snow (Kerley et al. 2002, Ross and Jalkotzy 1992) at different time

intervals. Because I could not monitor kittens daily, I evaluated their survival rate as

the proportion of survivors after the first year of life (365 days).

To estimate survival of radio-collared animals, I used the Mayfield method

(1961) modified by Heisey and Fuller (1985), because of its high precision when

vital rates are constant within time intervals (Samuel and Fuller 1996). I estimated

average annual survival rates by using the geometric mean of all years,

†

S

t

t=1

Y

’

Ê

Ë

Á

¯

˜

1/Y

,

where t represents each year of the study, Y the total number of years, and S

t

the

annual survival rates. I computed the 95% confidence interval (CI) with Akçakaya’s

unbiased weighed variance (2002),

†

S

t

(1- S

t

t=1

Y

Â

)

N

t

t=1

Y

Â

, where N

t

is the annual number of

cougars monitored in each class. This variance estimate minimizes sampling error

and accounts for demographic stochasticity (Akçakaya 2002).

10

To meet the required assumption of constant mortality of the Mayfield

method, I analyzed the statistical distribution of the deaths over a year (365 days) to

determine the time intervals when survival probabilities were constant. This yielded

two seasons: the high mortality (HM) season (13 November to 20 May) and the low

mortality (LM) season (21 May to 12 November) (Fig. 2). Most of the deaths (22 of

23) during the HM season were hunting-related, whereas the only death that occurred

during the LM season was caused by starvation. Intervals for each period were

chosen based on the median date of the deaths for each period. I tested for

differences in survival between the sex and age classes using a two-tailed t-test.

Population growth rate and short-term viability

I constructed a post-breeding, two-sex, age-classified, stochastic matrix

model to estimate annual stochastic population growth (l) and its 95%CI. I assumed

that (1) environmental conditions in successive years were uncorrelated; (2)

fecundity and survival rates were uncorrelated, but annual survival rates of each sex-

and age-class were correlated; (3) all cougars were equally at risk within the study

area (no spatial structure); (4) no catastrophes or bonanzas occurred; (5) cougars

were senescent at 12 years old (Beier 1996); and (6) male density did not affect

female fecundity, with the exception that a male density of 0 would cause a 0

fecundity rate. Different survival rates of adult males and females during this study

11

justified the use of a two-sex model (Caswell 2001). My two-sex structured model

was a Leslie matrix composed of 24 classes (Fig. 3), where males and females of the

same age succeeded each other from row to row and column to column (Caswell

2001), and was filled with the average vital rates for the period 1998-2003. The

dominant eigenvalue of this matrix corresponds to the finite rate of increase

†

l

1

of the

population and the dominant right eigenvector to the stable age distribution (Caswell

2001). I assessed the relative influence of each parameter r on the population growth

rate by determining and comparing their elasticity,

†

E

r

, defined as

†

E

r

=

r

l

1

∂l

1

∂

r

(Morris and Doak 2002), where

†

∂l

1

is the change in the growth rate caused by a 10%

increase,

†

∂r, in each parameter. As part of a descriptive analysis of population

growth from 1998 to 2003, I examined annual survival rates in 5 distinct matrices

and calculated annual dominant eigenvalues, despite having few animals represented

in each sex and age category during some years. As I didn’t have sufficient data for

kitten survival during some years, I used the average kitten survival rate for all

annual matrices.

My model took the form n(t) = L* n(t-1), where n is the vector of population

abundance, t is the time in years, and L is a Leslie matrix (Fig. 3). At time 0, n was

the vector of initial abundance, defined by the average density and stable age

distribution extrapolated to the entire 32,800 km

2

study area.

12

Environmental stochasticity was incorporated into my model by randomly

generating the F

x

and S

x

elements of L at every time step from normal distributions

with standard deviations given in a matrix V. Because there was no biological

reason why the variances of the average survival rates of each sex- and age-class

should be different within a year, and because annual sample sizes were low for

some classes, the standard deviation of annual survival rates for all radio-collared

animals was used to define environmental stochasticity in V (Caswell 2001).

Demographic stochasticity was included by drawing the number of survivors

in each sex-age class from a binomial distribution (based on the abundance N

i

(t) of

each i

th

class and on the survival rates s

i

from L), and the number of kittens born each

year from a Poisson distribution, using the random number generator of RAMAS 4.0

(Applied Biomathematics, Setauket, New York, USA). Age at first reproduction was

set at 1.5 years (see results).

To obtain the stochastic growth rate l, I projected the population 30 times

over 1,000 years, or until it fell to N = 0 animals. For every simulation, I computed

the geometric mean of [N(t+1)/N(t)] over all pairs of adjacent years to obtain a first

estimate of l (Morris and Doak 2002). I selected the arithmetic mean and 95% CI

of all simulations as the best estimate of the population growth rate.

13

I assessed population viability by examining the population trajectories of

5,000 simulations of my model projected over 25 years to determine the median time

to reach the demographic collapse (N = 79) and extirpation (N = 0). I set

demographic collapse at a total abundance of 79 cougars over the 32,800 km

2

area,

which corresponds to an abundance of 30 adults, because populations below 30

adults are more vulnerable to Allee effects and extreme demographic stochasticity

effects (Wielgus 2002). All simulations were performed with the software RAMAS

4.0.

14

RESULTS

Density and abundance

Fifty-two cougars were captured and monitored from 14 December 1998 to

20 May 2003, for a total of 19,337 cougar-radio-days and 2,737 radio-locations.

Twenty were males (8 adults and 12 yearlings), and 32 were females (27 adults and 5

yearlings). Annual density estimates and trends for both methods (the annual count

of monitored cougars in each class and the stable age distribution multiplied by the

adult female density) were very similar (Fig. 4). Average annual density appeared to

decline from a high of 1.46 cougars/100km

2

in 1998-99 to 0.85 cougars/100km

2

in

2002-03, with an average annual decline of 12±16% (mean±95% CI) (Table 1). The

average density for all years was 0.46±0.24 adults/100km

2

, and 1.09±0.54

cougars/100km

2

. Total abundance in the 32,800km

2

study area was estimated at 357

cougars, based on 1.09 cougars/100km

2

. To be conservative, I used 357 as the

starting point of every simulation, despite the apparent declining trend in density

observed during the latter years of the study.

Fecundity

Based on observations of 15 litters, I estimated mean litter size at 2.53±1.01.

I sexed 17 kittens and found a M:F ratio of 1:1.13, not different from equality (z =

15

0.25, P = 0.81), so I assumed each sex represented 50% of the litter. I obtained only

one successful interbirth interval (18 months), but used this value in my model

because it was consistent with results from studies with larger sample sizes (17.4

months, Logan and Sweanor 2001; 17.3 months, Lopez-Gonzalez 1999; 19.7

months, Ross and Jalkotzy 1992). I estimated the age of production of the first

successful litter to be 30 months, based on 29.1 months for Logan and Sweanor

(2001), 30 months for Ross and Jalkotzy (1992), and 29-30 months for Spencer et al.

(2001). Furthermore, I estimated that 75% of all adult females were reproductively

successful (pregnant or with offspring) in any year, which could be an

overestimation, considering the 41%-64% range reported by Toweill et al. (1984),

Ross and Jalkotzy (1992), Robinette et al. (1961), and Toweill et al. (1988). I

suspected that some births occurred without being noticed, so I remained more

conservative at 75%. Based on these parameters, I estimated the maternity rate to be

0.63 kittens of each sex per year (

†

2.53⋅ 0.5⋅ 0.75

1.5

) for females older than 2, and 0.32 for

females between 1 and 2 (half the adult rate because females start to reproduce at 1.5

years old).

16

Survival

Hunting accounted for 22 of the 24 deaths of radio-collared animals, and

indirectly caused the death of 5 of 21 dependent kittens by death of the mother. Nine

of the 52 monitored cougars were censored after losing their collars or after they

moved out of the study area. Four additional individuals were excluded from all

survival analyses because 1 died as a result of capture procedures and 3 others lost

their radio-collar or could not be found in the study area after only one day. Overall,

only 21 of the 52 radio-collared cougars lived or were monitored for longer than one

year after capture. The average annual survival rate for all radio-collared cougars

from 1998 to 2003 was 0.59±0.20, but major variations occurred between age and

sex classes, and from one year to another (Table 2). Average survival rate was

higher for adult females (0.77) than for adult males (0.33) (t =7.80, df = 34, P <

0.01) or yearlings (0.34) (t = 10.33, df = 43, P < 0.01). Survival did not differ

between yearling males (0.37) and females (0.32) (t = 0.35, df = 15, P = 0.73), so I

pooled them. Yearling survival (0.34) was lower than kitten survival (0.57) (t =

4.92, df = 36, P < 0.01), so I kept the two estimates distinct in my model. In total, I

recorded only 2 natural mortalities for radio-collared cougars: 1 adult female was

found with a broken neck and 1 adult male starved. The standard deviation of the

annual rates of all radio-collared cougars was 0.17, which was used in the matrix V

to incorporate environmental stochasticity in my model.

17

Population growth rate and short-term viability

The deterministic finite rate of increase l

1

of the average projection matrix

was 0.87. The stable age distribution was 30% adult females, 8% adult males, 25%

yearlings (1:1 sex ratio), and 37% kittens (1:1 sex ratio). Overall, individuals 4 years

and older represented only 22% of the population. Annual deterministic growth rates

followed the pattern of adult female survival rates (Fig. 5), starting with l

1

= 1.24 in

1998 and then progressively declining to 1.04, 0.92, 0.59, and 0.87 in 2002-03.

Elasticity analysis confirmed that adult female survival affects growth rate the most,

with

†

E

f

= 0.69, whereas the elasticity value was 0 for adult males, 0.17 for yearling

survival, 0.17 for kitten survival, and 0.19 for maternity rate (elasticities do not sum

to 1 because both maternity and female survival rates were involved in fecundity

rates).

The matrix projections yielded a stochastic population growth rate l of

0.80±0.11 (Fig. 6). Median time to demographic collapse (N = 79) was 8.5 years

and median time to extirpation (N = 0) was 25.9 years (Fig. 7).

18

DISCUSSION

Density

Average density estimates (0.46 adults and 1.09 cougars/100km

2

) were in the

lower range of densities reported by other cougar researchers (from 0.44 to 13.03

cougars/100km

2

, Smallwood 1997). My estimates were comparable to densities

observed in 2 other populations in Utah (0.37 adults/100km

2

, Lindzey et al. 1994;

0.96 cougars/100km

2

, Lopez-Gonzalez 1999), but considerably lower than those

observed in 5 other populations, including Alberta (2.7 to 4.2 cougars/100km

2

, Ross

and Jakoltzy 1992), British Columbia (3.5 to 3.7 cougars/100km

2

, Spreadbury et al.

1996), Idaho (2.9 adults/100km

2

, Hornocker 1970), and New Mexico (1.72 to 3.91

cougars/100km

2

, Logan and Sweanor 2001). Even the starting density, at 1.47

cougars/100km

2

in 1998-99, was considerably lower than most other populations,

suggesting that the common perception that cougars were at very high density was

incorrect.

Although intense trapping efforts were repeated each winter in an attempt to

capture and radio-collar all cougars in the study area, my estimates should be

considered minimum because some uncaptured individuals undoubtedly existed in

the annual 95% contour kernel home range of the adult females. The progressive

increase in the composite home range area may also have contributed to the observed

19

decline in density (Smallwood 1997). However, uncaptured animals in the 95 %

composite female home range would likely be balanced by captured animals

venturing outside the composite range. For example, most males spent considerable

periods of time outside the adult female composite home range. Furthermore, my

estimates included all animals captured every year, even cougars that were

monitored for only few days. Other studies of cougar populations probably suffered

the same problem of uncaptured animals and lack of population closure, so my

density estimates should be directly comparable to others in the literature.

Demographic rates

Litter size (2.53) was consistent with those reported elsewhere (2.4, Hemker

et al. 1986; 2.4, Lindzey et al. 1994; 3, Logan and Sweanor 2001; 2.2, Ross and

Jalkotzy 1992). As discussed above (see results), insufficient data made it necessary

to estimate other parameters involved in maternity rates (interbirth interval, age at

first litter, proportion of successful reproducing females) from literature. Because

the elasticity value of maternity rate was relatively low (0.19), uncertainties in

fecundity values had a negligible influence on population projection results.

Survival rates indicated this cougar population was very heavily hunted (92%

deaths of radio-collared cougars) compared to other populations (5%, Logan and

Sweanor 2001; 15%, Logan et al. 1986; 35%, Spencer et al. 2001). The overall

20

survival (0.58) was similar to that encountered in Arizona (0.62, Cunningham et al.

2001). Survival rate of adult females (0.77) was similar to those of other cougar

populations (0.82, Logan and Sweanor 2001; 0.77, Lopez-Gonzalez 1999; 0.77 for

all adults, Spencer et al. 2001), and higher than the 0.67 reported by Cunningham et

al. (2001). Adult male survival (0.33) was much lower than in other areas (0.58,

Cunningham et al. 2001; 0.91, Logan and Sweanor 2001; 0.62, Lopez-Gonzalez

1999). Similarly, kitten and yearling survival rates were lower than most rates

reported in literature. Kitten survival (0.57) was below 0.72 (Hemker et al. 1986)

and 0.76 (Robinette 1961) previously reported, but similar to rates encountered New

Mexico (0.59 to 0.63, Logan and Sweanor 2001) and Utah (0.42 for cubs until

dispersal, Lopez-Gonzalez 1999). Yearling survival rate (0.34) was lower than

previously reported (0.56 and 0.88 for subadult males and females, Logan and

Sweanor 2001; 0.60 and 0.80 for juvenile males and females, Spencer et al. 2001).

The elasticity of adult female survival outweighed that of any other

parameter, highlighting the large influence of adult female survival on population

growth, and minimizing the effect of uncertainties for other parameters that had

smaller sample sizes and larger standard deviations.

21

Population growth rate and viability

The observed declining trend in annual density (12±16%) corresponded to

the decline indicated by the stochastic (20±11%) and deterministic (13%) growth

rates. Because the stochastic growth rate incorporates environmental and

demographic variability, it is believed to be more realistic than the deterministic

growth rate (Caswell 2001, Morris and Doak 2002). The difference between the 2

values in my model (∆l= 0.07) is probably related to the large magnitude of

environmental variation in vital rates (SD = 0.17). Regardless of which method was

used, both indicated a steeply declining population.

My results are limited by the uncertainties in some of the model parameters

and density estimates. However, the very high elasticity value for adult female

survival (0.69) and large sample size (N = 28) for this parameter indicated that the

population would still be declining even if the other parameters were doubled to

unrealistic values. Additionally, if my density estimates had been higher, the

simulated population trajectories would have reached extirpation after a longer time

delay, but the growth rate would not have been affected. A limitation for population

projections is the underlying model assumption that all means and standard

deviations will remain constant in the future. Emigration from neighboring

populations could significantly increase the viability of this population (Beier 1993,

22

Sweanor et al. 2000), or a very sparse cougar density could trigger a lower hunting

effort.

Management implications

My results reject the hypothesis that cougars are at very high densities and

are increasing in northeastern WA, southern BC, and northwestern ID. Cougar

densities were not higher than reported in most other areas, and the population

appears to be declining. The increased level of cougar complaints following 1996

doesn’t seem to be due to excessively high densities and growth rates, but could be

related to (1) very young age structure of the population caused by heavy hunting;

(2) increased human intrusion into cougar habitat; or (3) low level of social

acceptance of cougars by local stakeholders.

The conflicts surrounding cougar management in my study area are not

uncommon in other regions of North America. Other populations that are believed

to be increasing, as suggested by more frequent cougar encounters, could be

declining as well. Conducting similar demographic analyses in different geographic

areas where complaints are increasing could shed light on this situation. Educational

programs, such as the one undertaken in western WA (CAT project, Koehler and

23

Nelson 2003), might help alleviate the tension presently surrounding cougar

management throughout western North America.

24

Table 1. Average density estimates for cougars in the study area from 1998 to 2003,

as defined by the annual composite adult female home ranges (AF HR) and the

abundance N of each demographic class, estimated with the number of captured and

monitored animals (Capt.) and the stable age distribution (SAD).

Adult

females (N)

Adult

males (N)

Yearlings

(N)

Kittens

(N)

95% composite

adaptive kernel

AF HR (km

2

)

Capt.

Capt. SAD Capt. SAD

SAD

Adults

/100km

2

Cougars

/100km

2

1998-99

1,826

9

4 2.3

2 7.3

11.1

0.67

1.47

1999-00

3,308

13

2 3.3

8 10.6

16.1

0.47

1.21

2000-01

3,582

12

3 3.1

7 9.8

14.9

0.42

1.05

2001-02

4,744

14

3 3.6

5 11.4

17.3

0.36

0.86

2002-03

5,641

17

3 4.4

5 13.8

21.1

0.37

0.85

Average

3,820

13

3 3.3

5.4 10.6

16.1

0.46

1.09

25

Table 2. Annual survival rates for cougars in the Pacific Northwest (mean (sample

size)) for kittens (0-1yr), yearlings (1-2yr), and adult (2-12yr) males and females,

from 1998 to 2003.

1998-99 1999-00 2000-01 2001-02 2002-03 Mean±95%CI

Kittens

NA

0.25 (4)

1 (3)

0.75 (8)

0.43 (7)

0.57±0.21 (21)

Yearlings

1 (2)

0.53 (7)

0.24 (7)

0.17 (5)

0.41 (5)

0.34±0.35 (17)

Adult males

0.21 (3)

0.50 (2)

0.27 (3)

0.26 (3)

1.00 (3)

0.33±0.47 (8)

Adult females

1 (9)

0.90 (12) 0.91 (11) 0.52 (13)

0.76 (16)

0.77±0.20 (28)

Total radio-collared

0.69 (14) 0.76 (21) 0.65 (20) 0.35 (22)

0.77 (23)

0.59±0.20 (52)

26

Figure 1. Study area in the Pacific Northwest. The hatched polygons represent the

95% adaptive kernel composite home range of all cougars monitored from 1998 to

2003, and the triangles (N = 52) represent capture locations.

27

0

1

2

3

4

5

6

7

8

9

10

Nov

Dec

Jan

Feb

Mar

Apr

May

Jun

Jul

Aug

Sep

Oct

Death frequency

LM

HM

Figure 2. Monthly distribution of the cougar mortalities in the Pacific Northwest,

from 1998 to 2003. High mortality (HM) season was from 13 November to 20 May,

while the low mortality (LM) season was from 21 May to 12 November. Seasons

were defined by the median date of the deaths and were used to meet the assumption

of constant mortality within each time interval required by the Heisey-Fuller survival

analysis.

28

KitF

KitM

YeaF

YeaM

3F

3M

… 11F

11M

F12

M12

KitF

m

yf

S

kf

0

m

3f

S

yf

0

m

4f

S

3f

0

… m

12f

S

11f

0

m

13f

s

12f

0

KitM

m

ym

S

kf

0

m

3m

S

yf

0

m

4m

S

3f

0

… m

12m

S

11f

0

M

13m

s

12f

0

YeaF

S

kf

0

0

0

0

0

… 0

0

0

0

YeaM

0

S

km

0

0

0

0

… 0

0

0

0

3F

0

0

S

yf

0

0

0

… 0

0

0

0

3M

0

0

0

S

ym

0

0

… 0

0

0

0

…

…

…

…

…

…

…

… …

…

…

…

11F

0

0

0

0

0

0

… 0

0

0

0

11M

0

0

0

0

0

0

… 0

0

0

0

12F

0

0

0

0

0

0

… S

11f

0

0

0

12M

0

0

0

0

0

0

… 0

S

11m

0

0

Figure 3. Structure of the Leslie matrix used to model population growth for female

and male cougar kittens (KitF, KitM), yearlings (YeaF, YeaM), and adult (xF, xM,

where x is the age class from 3 to 12) in the Pacific Northwest, from 1998 to 2003.

In the first two rows, fecundity rates (F

x

) represented the probability that a female of

age x will give birth to a male (m

xm

S

(x-1)f

) or female (m

xf

S

(x-1)f

) kitten. The survival

rates for kittens (S

kf

, S

km

), yearlings (S

yf

, S

ym

), adult males (S

xm

) and adult females (S

xf

)

were the elements of the second-subdiagonal.

29

0.00

0.20

0.40

0.60

0.80

1.00

1.20

1.40

1.60

1.80

1998-99

1999-00

2000-01

2001-02

2002-03

Abundance/100km^2

All cougars -SAD

All cougars -Capture data

Adults only -SAD

Adults only -Capture data

Figure 4. Adult and total cougar density in the Pacific Northwest, from 1998 to

2003, calculated with the 95% composite adaptive kernel home range method for

adult females, and projected with both capture data and stable age distribution (SAD)

vector to calculate adult male and yearling densities. Kitten density was calculated

with the SAD.

30

0

0,2

0,4

0,6

0,8

1

1,2

1,4

1998-99

1999-00

2000-01

2001-02

2002-03

Deterministic growth rate

Adult female survival

Figure 5. Annual survival rates of adult females and deterministic growth rates l

1

(dominant eigenvalue of annual matrix model), between 1998 and 2003, for cougars

in the Pacific Northwest.

31

0

100

200

300

400

500

600

700

800

900

1000

0

2

4

6

8

10

12

14

16

18

20

22

24

Year

Cougar abundance

Figure 6. Simulated trajectory of the cougar population in the Pacific Northwest,

based on survival and fecundity rates from 1998 to 2003. The squares represent the

annual average abundances, the vertical lines are standard deviations, and the empty

circles are maximum and minimum values obtained in 5,000 simulations.

32

0

0.2

0.4

0.6

0.8

1

1.2

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25

Year

Cumulative probability

Extirpation

Demographic collapse

Figure 7. Time to fall from 357 cougars to demographic collapse (N = 79,

equivalent to 30 adults) and extirpation (N = 0), based on vital rates estimated for

cougars in the Pacific Northwest from 1998 to 2003, and simulated 5,000 times over

25 years.

33

LITERATURE CITED

A

KÇAKAYA

, H.R. 2002. Estimating the variance of survival rates and fecundities.

Animal Conservation 5:333-336.

A

SHMAN

, D.L.,

AND

K. G

REER

. 1976. Age techniques. Pages 199-204 in: G.C.

Christensen and R.J. Fisher, editors. Transactions of the Mountain Lion

Workshop, U.S. Fish and Wildlife Service, Portland, Oregon, USA.

A

USTIN

, M. 2003. British Columbia mountain lion status report [abstract]. In press.

Proceedings of the Seventh Mountain Lion Workshop; 2003 May 15-17,

Jackson, Wyoming, USA.

B

ANHURST

, D.,

AND

F.G. L

INDZEY

. 1989. Detecting female mountain lions with

kittens. Northwest Science 63:35-37.

B

AILEY

, R.G. 1995. Descriptions of the Ecoregions of the United States. 2nd ed.

(rev. and expanded). Miscellaneous publication 1391 (rev.). U.S. Department

of Agriculture, Forest Service, Washington DC, USA.

B

EIER

, P. 1991. Cougar attacks on humans in the United States and Canada.

Wildlife Society Bulletin 19:403-412.

_____. 1993. Determining minimum habitat areas and habitat corridors for cougars.

Conservation Biology 7:94-108.

_____, D. C

HOATE

,

AND

R.H. B

ARRETT

. 1995. Movement patterns of mountain

lions during different behaviors. Journal of Mammalogy 76:1056-1070.

_____. 1996. Metapopulation models, tenacious tracking, and cougar conservation.

Pages 293-323 in: D.R. McCullough, editor. Metapopulation and wildlife

conservation, Island Press, Washington, D.C., USA.

B

ROOK

, B.W., J.J. O’G

RADY

, A.P. C

HAPMAN

, M.A. B

URGMAN

, H.R. A

KÇAKAYA

,

AND

R. F

RANKHAM

. 2000. Predictive accuracy of population viability analysis

in conservation biology. Nature 404:385-387.

34

C

ASE

, T.J., editor. 2000. An illustrated guide to theoretical ecology. Oxford

University Press, New York, New York, USA.

C

ASWELL

, H. 1997. Matrix methods for population analysis. Pages 19-58 in: S.

Tuljapurkar and H.Caswell, editors. Structured-population models in marine,

terrestrial, and freshwater systems. Population and community biology series

18, Chapman & Hall, New York, New York, USA.

_____. 2001. Matrix population models: Construction, analysis, and interpretation.

Second edition. Sinauer Associates, Sunderland, Massachusetts, USA.

C

HESTER

, T. 2003. Mountain lion attacks on people in the U.S. and Canada

[online]. Available from: http://tchester.org/sgm/lists/lion_attacks.html.

Accessed 6 October 2003.

C

OUPE

, R., A.C. S

TEWART

,

AND

B.M. W

IKEEM

. 1991. Englemann spruce-subalpine

fir zone. Pages 223-236 in D. Meidinger and J. Poplar, editors. Ecosystems of

British Columbia. Special report series 6, B.C. Ministry of Forests, Research

Branch, Victoria, British Columbia, Canada.

C

UNNINGHAM

, S.C., W.B. B

ALLARD

,

AND

H.A. H

EATHER

. 2001. Age structure,

survival, and mortality of mountain lions in southeastern Arizona.

Southwestern Naturalist 46:76-80.

D

AWN

, D., M. K

UTILEK

, R. H

OPKINS

, S. A

NAND

,

AND

S. T

ORRES

. 2003.

Management of cougars (Puma concolor) in the western United States. In

press. Proceedings of the Seventh Mountain Lion Workshop; 2003 May 15-17,

Jackson, Wyoming, USA.

D

EMARCHI

, D.A. 1996. An introduction to the ecoregions of British Columbia.

Ministry of Environment, Lands and Parks, Wildlife Branch, Victoria, British

Columbia, Canada.

D

ICKSON

, B.

AND

P. B

EIER

. 2002. Home-range and habitat selection by adult

cougars in Southern California. Journal of Wildlife Management 66:1235-

1245.

J

EDRZEJEWSKA

, B., W. J

EDRZEJEWSKI

, A.N. B

UNEVICH

, L. M

ILKOWSKI

,

AND

H.

O

KARMA

. 1996. Population dynamics of Wolves Canis lupus in Bialowieza

35

Primeval Forest (Poland and Belarus) in relation to hunting by humans, 1847-

1993. Mammal Review 26:103-126.

H

EISEY

, D.M.

AND

T.K. F

ULLER

. 1985. Evaluation of survival and cause-specific

mortality rates using telemetry data. Journal of Wildlife Management 49:668-

674.

H

EMKER

, T.P., F.G. L

INDZEY

, B.B. A

CKERMAN

,

AND

A.J. B

UTTON

. 1986. Survival

of cougar cubs in a non-hunted population. Pages 327-332 in: S.D. Miller and

D.D. Everett, editors. Cats of the world. National Wildlife Federation,

Washington, D.C., USA.

H

OOGE

, P.N.,

AND

B. E

ICHENLAUB

. 1997. Animal movement extension to Arcview,

version 1.1. Alaska Biological Science Center, U.S. Geological Survey,

Anchorage, Alaska, USA.

H

ORNOCKER

M.G. 1970. An analysis of mountain lion predation upon mule deer and

elk in the Idaho primitive area. Wildlife Monograph 21.

K

ATNIK

, D.D. 2002. Predation and habitat ecology of mountain lions (Puma

concolor) in the southern Selkirk mountains. Dissertation, Washington State

University, Pullman, Washington, USA.

K

ETCHESON

, M.V., T.F. B

RAUMANDLE

, D. M

EIDINGER

, G. U

TZIG

, D.A. D

EMARCHI

,

AND

B.M W

IKEEM

. 1991. Interior cedar-hemlock zone. Pages 167-181 in D.

Meidinger and J. Poplar, editors. Ecosystems of British Columbia. Special

report series 6, B.C. Ministry of Forests, Research Branch, Victoria, B.C.

K

ERLEY

, L.L., J.M G

OODRICH

, D.G. M

IQUELLE

, E.N. S

MIRNOV

, H.B. Q

UIGLEY

,

AND M.G. H

ORNOCKER

. 2002. Reproductive parameters of wild female

Amur (Siberian) tigers (Panthera tigris altaica). Journal of Mammalogy

84:288-298.

K

OEHLER

, G.

AND

E. N

ELSON

. 2003. Project CAT (cougars and teaching):

integrating science and community in development planning [abstract]. In

press. Proceedings of the Seventh Mountain Lion Workshop; 2003 May 15-17,

Jackson, Wyoming, USA.

36

L

INDZEY

, F.G., W.D. V

AN

S

ICKLE

, B.B. A

CKERMAN

, D. B

ARNHURST

, T.P. H

EMKER

,

AND

S.P. L

AING

. 1994. Cougar population dynamics in Southern Utah.

Journal of Wildlife Management 58:619-624.

L

OGAN

, K.A.

AND

L.L. S

WEANOR

. 2001. Desert puma: Evolutionary ecology and

conservation of an enduring carnivore. Island Press, Washington, D.C., USA.

_____, L.L. Irwin, and R. Skinner. 1986. Characteristic of a hunted mountain lion

population in Wyoming. Journal of Wildlife Management 50:648-654.

L

OPEZ

-G

ONZALEZ

, C.A. 1999. Implicaciones para la conservacion y el manajo de

pumas (Puma concolor) utilizando como modelo una poblacion sujeta a

caceria deportiva. Dissertation, Universidad Nacional Autonoma de Mexico,

Mexico.

M

ARTORELLO

, D.A. 2001. Cougar status and trend report. Game status and trend

report, Wildlife Program, Washington Department of Fish and Wildlife,

Olympia, Washington, U.S.A.: 159-161.

_____. 2003. Cougar harvest statistics pre- and post-initiative banning use of

hounds for hunts. 2003 annual meeting and workshops, The Washington

Chapter of the Wildlife Society; 2003 April 15-17, Port Townsend,

Washington, USA.

M

AYFIELD

, H. 1961. Nesting success calculated from exposure. Wilson Bulletin,

73:255-261.

M

ORRIS

, W.F. AND D.D. D

OAK

. 2002. Quantitative conservation biology; Theory

and practice of population viability analysis. Sinauer Associates, Sunderland,

Massachusetts, USA.

N

ADEAU

, S. 2003. Idaho mountain lion status report. In press. Proceedings of the

Seventh Mountain Lion Workshop; 2003 May 15-17, Jackson, Wyoming,

USA.

R

ILEY

, S.J.

AND

D.J. B

ECKER

. 2000. Wildlife stakeholder acceptance capacity for

cougars in Montana. Wildlife Society Bulletin 28:931-939.

37

____

AND

R.A. M

ALECKI

. 2001. A landscape analysis of cougar distribution and

abundance in Montana, USA. Environmental Management 28:317-323.

R

OBINETTE

, W.L., J.S. G

ASHWILER

,

AND

O.W. M

ORRIS

. 1961. Notes on cougar

productivity and life history. Journal of Mammalogy 42:204-217.

R

OBINSON

, H, S., R.B. W

IELGUS

,

AND

J.C. G

WILLIAM

. 2002. Cougar predation and

population growth of sympatric mule deer and white-tailed deer. Canadian

Journal of Zoology 80:556-568.

R

OSS

, P.I., AND M.G. J

ALKOTZY

. 1992. Characteristics of a hunted population of

cougars in Southwestern Alberta. Journal of Wildlife Management 56:417-

426.

S

AMUEL

, M.D. AND M.R. F

ULLER

. 1996. Wildlife radiotelemetry. Pages 370-418

in: T.A. Bookhout, editor. Research and management techniques for wildlife

and habitats. Fifth edition revised, Wildlife Society, Bethesda, Maryland,

U.S.A.

S

EAL

, U.S., D.L. A

RMSTRONG

, AND L.G. S

IMMONS

. 1987. Yohimbine

hydrochloride reversal of ketamine hydrochloride and xylazine hydrochloride

immobilization of Bengal tigers and effects on hematology and serum

chemistry. Journal of Wildlife Diseases 23:296-300.

S

MALLWOOD

, K.S. 1997. Interpreting puma (Puma concolor) population estimates

for theory and management. Environmental Conservation 24:283-289.

S

MUTS

, G.L. 1978. Effects of population reduction on the travels and reproduction

of lions in Kruger National Park. Carnivore 1:61-72.

S

PENCER

, R.D., P

IERCE

, D.J., S

CHIRATO

, G.A., D

IXON

, K.R.,

AND

C.B. R

ICHARDS

.

2001. Mountain lion home range, dispersal, mortality, and survival in the

Western Cascade Mountains of Washington. Final report. Washington

Department of Fish and Wildlife, Olympia, Washington, USA.

S

PREADBURY

, B.R., K. M

USIL

, J. M

USIL

, C. K

AISNER

,

AND

J. K

OVAK

. 1996. Cougar

population characteristics in southeastern British Columbia. Journal of

Wildlife Management 64:962-969.

38

S

WEANOR

, L.L., K.A. L

OGAN

,

AND

M.G. H

ORNOCKER

. 2000. Cougar dispersal

patterns, metapopulation dynamics and conservation. Conservation Biology

14:798-808.

T

OWEILL

, D.E., C. M

ASER

, L.D. B

RYANT

,

AND

M.L. J

OHNSON

. 1984. Size and

reproductive characteristics of western Oregon cougars. Pages 176-184 in: J.

Roberson and F. Lindzey, editors. Proceedings of the Second Mountain Lion

Workshop, Zion National Park, Utah, USA.

_____, _____, _____,

AND

_____. 1988. Reproductive characteristics of eastern

Oregon cougars. Northwest Science 62:147-150.

W

ASHINGTON

D

EPARTMENT OF

F

ISH AND

W

ILDLIFE

. 1999. Fact sheets: cougars

[online]. Available from: http://www.wa.gov/wdfw/factshts/cougar.htm.

Accessed 30 September 2003.

W

IELGUS

, R.B. 2002. Minimum viable population and reserve sizes for naturally

regulated grizzly bears in British Columbia. Biological Conservation 106:381-

388.

_____, F.L. B

UNNELL

, W.L. W

AKKINEN

,

AND

P.E. Z

AGER

. 1994. Population

dynamics of Selkirk mountain grizzly bears. Journal of Wildlife Management

58:266-272.

_____, F. S

ARRAZIN

, R. F

ERRIERE

,

AND

J.C

LOBERT

. 2001. Estimating effects of

adult male mortality on grizzly bear population growth and persistence using

matrix models. Biological conservation 98:293-303.