Individual Behavioral Phenotypes of the Cliff Chipmunk (Tamias Dorsalis): Effects on Female Reproductive Success and Juvenile Habitat Selection

Individual Behavioral Phenotypes of the Cliff Chipmunk (Tamias dorsalis): Effects on Female Reproductive Success and Juvenile Habitat Selection

Item Type text; Electronic Thesis

Authors Kilanowski, Allyssa LeAnn

Publisher The University of Arizona.

Rights Copyright © is held by the author. Digital access to this material is made possible by the University Libraries, University of Arizona. Further transmission, reproduction or presentation (such as public display or performance) of protected items is prohibited except with permission of the author.

Download date 27/09/2021 19:39:05

Link to Item http://hdl.handle.net/10150/582370 INDIVIDUAL BEHAVIORAL PHENOTYPES OF THE CLIFF CHIPMUNK (TAMIAS DORSALIS): EFFECTS ON FEMALE REPRODUCTIVE SUCCESS AND JUVENILE HABITAT SELECTION

Allyssa LeAnn Kilanowski

______

A Thesis Submitted to the Faculty of the

SCHOOL OF NATURAL RESOURCES AND THE ENVIRONMENT

In Partial Fulfillment of the Requirements For the Degree of

MASTERS OF SCIENCE WITH A MAJOR IN NATURAL RESOURCES

In the Graduate College

THE UNIVERSITY OF ARIZONA

2015

STATEMENT BY AUTHOR

This thesis has been submitted in partial fulfillment of requirements for an advanced degree at the University of Arizona and is deposited in the University Library to be made available to borrowers under rules of the Library.

Brief quotations from this thesis are allowable without special permission, provided that accurate acknowledgment of the source is made. Requests for permission for extended quotation from or reproduction of this manuscript in whole or in part may be granted by the head of the major department or the Dean of the Graduate College when in his or her judgment the proposed use of the material is in the interests of scholarship. In all other instances, however, permission must be obtained from the author.

SIGNED: _Allyssa LeAnn Kilanowski______

APPROVAL BY THESIS DIRECTOR

This thesis has been approved on the date shown below:

______8/7/15______John L. Koprowski Date Professor of Wildlife and Fisheries Science

ACKNOWLEDGMENTS

I wish to thank John Koprowski for his mentoring and guidance throughout this project. Without feedback from John and the Koprowski lab, this project would not have been successful. I sincerely appreciate all the time my fellow graduates spent reviewing my grant proposals and manuscripts.

I also wish to thank my family for their unwavering support and faith in my abilities throughout this project. I also want to thank the Tucson tennis community for providing an emotional support network and allowing me to use the courts as a place to work through graduate school trials.

This research would be impossible without funding from the University of

Arizona, T & E, Inc, the American Society of Mammalogists, the C. P. Patrick Reid

Scholarship, and the Arrington Memorial scholarship.

TABLE OF CONTENTS

LIST OF TABLES 5

LIST OF FIGURES 6

ABSTRACT 8

CHAPTER 1: INTRODUCTION 9

CHAPTER 2: PRESENT STUDY 17

APPENDIX A: Behavioral phenotypes, female reproductive success, and juvenile dispersal: Insights from a semi-fossorial, social rodent 19 ABSTRACT 19 INTRODUCTION 20 METHODS 23 Study Area and Species 23 Individual Behavioral Phenotypes 24 Repeatability of Personality 25 Factors Influencing Personality 26 Maternal Reproductive Success 26 Natal Habitat Preference Induction and Personality 27 RESULTS 29 Individual Behavioral Phenotypes 30 Repeatability of Personality 31 Factors Influencing Personality 31 Maternal Reproductive Success 32 Natal Habitat Preference Induction and Personality 32 DISCUSSION 33 Reversed Sexual Dimorphism 33 Behavioral Phenotypes and Repeatability 34 Factors Affecting Personality 35 Maternal Reproductive Success 36 Natal Habitat Preference Induction and Personality 36 TABLES AND FIGURES 39 REFERENCES 47

LIST OF TABLES

TABLE 1. PCA loadings for behaviors in two behavior tests (Open Field and Mirror

Image Stimulation) in cliff chipmunks, Mt. Graham, AZ, USA. Only the first two components were used in analyses. Units are proportion of time spent on each activity during the 7-min tests. Bold values indicate behaviors that were interpreted to contribute to a component. 43

TABLE 2. Estimated coefficients and confidence intervals of generalized linear models for the two principal components of the mirror image stimulation behavioral test

(sociality and image engagement) for cliff chipmunks (Tamias dorsalis) on Mt. Graham,

Arizona, USA in 2014. Bold values indicate significance (p < 0.05) and italicized values indicate potential biological significance (p < 0.1). 45

TABLE 3. Estimated coefficients and confidence intervals of linear models for the litter size and female home range models of cliff chipmunks (Tamias dorsalis) on Mt. Graham,

Arizona, USA in 2014. Two models are a 50% and 95% kernel density home range.

Bold values indicate significance (p < 0.05) and italicized values indicate potential biological significance (p < 0.1). 46

LIST OF FIGURES

FIGURE 1. Principal components analyses of behavioral phenotypes observed in a population of cliff chipmunks (Tamias dorsalis) on Mt. Graham, Arizona, USA based on two tests, the Open Field (OF, left) and Mirror Image Stimulation (MIS, right) tests.

Circles indicate a group of individuals with similar behavioral responses. Sex and age included to illustrate differences. OF PC 1 represents activity (loadings are low [i.e. unmoving] to high [i.e. locomotion], left to right); PC 2 represents vigilance (loadings are not alert to alert [i.e. unmoving with head up and aware of surroundings], bottom to top.

The MIS PC 1 represents sociality (loadings are low [i.e. wary of image] to high [i.e. unconcerned with image], left to right); PC 2 represents image engagement (loadings are unengaged [i.e. positioned away from image] to highly engaged [i.e. touching mirror], bottom to top). 39

FIGURE 2. Boxplot displaying the mean and range (1st and 3rd quartiles, including outliers) of novel vigilance, the first principal component of behavioral response clusters for the Open Field (left), and sociality, the first principal component of behavioral response clusters for the Mirror Image Stimulation (right) behavior tests performed on a population of cliff chipmunks (Tamias dorsalis) on Mt. Graham, Arizona, USA in 2014.

Letters above boxplots represent Tukey’s HSD post-hoc test. 40

FIGURE 3. Difference between natal and settled nest site characteristics for a population of cliff chipmunks (Tamias dorsalis) on Mt. Graham, Arizona, USA. 41

FIGURE 4. Mean of the observed individual differences between natal and settled nest site habitat characteristics for a population of juvenile cliff chipmunks (Tamias dorsalis) on Mt. Graham, Arizona, USA. The zero line indicates no difference between natal and 7 settled sites. Bars above the line occur when the natal nest site characteristic is greater than the settled site characteristic and points below the line occur when the natal site characteristic is less than the settled site characteristic. 42 8

ABSTRACT Differences among individual responses to behavioral stimuli have been observed throughout a variety of taxa and these individual differences can affect female reproductive success and juvenile settlement decisions. In this study, we examined the effect of reversed sexual dimorphism on behavior phenotype and the effect of behavior on maternal reproductive success and juvenile dispersal of a fossorial rodent (Tamias dorsalis) in southeastern Arizona. We found that multiple behavioral phenotypes existed within this population and female litter size was not affected by behavioral type. We also found that natal habitat preference induction (NHPI) does occur at the population level, but only weakly occurs for the individual. We also found no effect of personality on site selection. Our results indicate that sex and mass may explain differences in behavioral phenotypes; however, individual behavioral differences are weakly related to female reproductive success and settlement decisions during juvenile dispersal. 9

CHAPTER 1: INTRODUCTION

The movement of organisms away from their birthplace, termed natal dispersal

(Ronce 2007), primarily occurs because juveniles cannot remain with their parents as adults; however, natal dispersal also reduces local density of conspecifics (Byrom and

Krebs 1999), increases gene flow (Bohonak 1999), and affects individual fitness, both positively (Van Vuren and Armitage 1994) and negatively (Dingemanse et al. 2004).

Three stages of dispersal are recognized: 1) emigration, 2) exploration and travel, and 3) settlement (Bowler and Benton 2005, Ronce 2007). The first stage of dispersal, emigration, is well studied, however, the second and third stages require more research

(Bowler and Benton 2005, Ronce 2007, Mabry and Stamps 2008) as knowledge of these stages is critical to understanding the consequences of dispersal on individual fitness and population viability (Clobert et al. 2001). Crucial decisions made by individuals during dispersal are influenced by individual personality (Dingemanse et al. 2003, Sih et al.

2012), therefore it is important to consider the effect of individual behavioral variation on juvenile dispersal.

Personality

Differences among individual responses to behavioral stimuli have been observed throughout a variety of taxa (i.e. mammals [Martin & Réale 2007, Boon et al. 2007,

2008], fish [Dingemanse et al. 2007], reptiles [Cote and Colbert, 2007], invertebrates

[Sinn et al. 2006, Briffa and Greenaway 2011], birds [Dingemanse et al. 2003]). These behavioral differences, referred to as individual behavioral phenotypes (hereafter personality) are defined as a repeatable set of observable behaviors that are the result of an individual’s genotype interacting with its environment (Clark & Ehlinger, 1987; Sih et 10 al., 2012). Personality, in the form of traits such as aggressiveness or activity level, affects individual behaviors such as parenting style (Budeav et al. 1999) and dispersal

(Dingemanse et al. 2003), which in turn affects mating success (Godin and Dugatkin

1996), reproductive success (Réale et al. 2000), survival (Réale and Festa-Bianchet 2003,

Dingemanse et al. 2004), and ultimately, fitness (Dingemanse and Réale 2005, Réale et al. 2007).

Settlement

The third phase of dispersal, settlement, occurs when an individual makes the decision to remain in a location (Bowler and Benton 2005, Ronce 2007). Some factors that influence the decision to settle include the habitat type (Stamps and Swaisgood 2007,

Selonen et al. 2007), the distance of the settlement location from the natal area (Loew

1999, Selonen and Hanski 2010), and the availability of food (Bryrom and Krebs 1999).

One explanation for understanding the cues used by individuals to select a settlement location is the natal habitat preference induction (NHPI) hypothesis that states that animals will choose to settle in an area that is similar to their natal area (Davis and

Stamps 2004, Stamps et al. 2009). NHPI is a common term for this concept. Other descriptions of NHPI include habitat imprinting (Mannan et al. 2006), Hopkin’s host selection principle (Barron 2001), preference induction (Dethier 1982), and natal habitat- biased dispersal (Selonen et al. 2007). NHPI may be adaptive, as habitat that produces individuals old enough to disperse is most likely habitat of reasonable quality, thus individuals use cues from the natal area to help reduce search time and increase fitness

(Stamps and Swaisgood 2007). NHPI is still controversial as it has been observed in some taxa (Peromyscus boylii, Mabry 2008; Pteromys volans, Selonen et al. 2007; 11

Drosphilia melanogaster, Stamps and Blozis 2006), but not in others (Accipiter cooperii,

Mannan et al. 2006). The effect of natal habitat preference is complicated by the effects of individual personality. Individuals with a more bold and exploratory personality may be more likely to accept novel cues than a shy and less active individual (Stamps and

Swaisgood 2007) as bold individuals may have a wider threshold of acceptable cues

(Weins 1985).

Maternal Reproductive Fitness

Female behavioral phenotype can affect reproductive success and survival; more aggressive and bold individuals demonstrate a slightly enhanced success (Smith and

Blumstein 2007). In great tits (Parus major), aggressive individuals are better able to defend the nest against predators, (Hollander et al. 2008) and bold females have increased foraging success (Biro and Stamps 2008) leading to increased reproductive success.

Conservation Implications

Climate change and anthropogenic factors are rapidly altering the environment.

Efforts to manage threatened or endangered species will benefit from an understanding of dispersal behaviors. A high proportion of released animals leave the reintroduction area to settle in a new location, one that is not always desired by managers (Watters and

Meehan 2007). This creates two problems, the first is that released animals that leave the release area are at increased risk of mortality and second, released animals that remain may represent a single type of personality thereby reducing the variability and ultimately the viability of the introduced population (Davis and Stamps 2004). Information on how individual personalities influence dispersal decisions can mitigate these negative outcomes from reintroductions. 12

Current Research

In this thesis, I examine the effect of personality on maternal reproductive success and juvenile dispersal of a semi-fossorial small rodent species (cliff chipmunk, Tamias dorsalis) in southeastern Arizona, USA. Small mammals are ideal to study personality and the relationship to maternal reproductive success and juvenile dispersal because of their abundance and existence of personalities (Martin and Réale, 2007; Boon et al.,

2008; Selonen and Hanski, 2010). In red squirrels (Tamiasciurus hudsonicus), female personality did not influence litter size, but was related to growth rate and offspring survival (Boon et al., 2008). In grey-sided voles (Clethrionomys rufocanus), juvenile dispersal was influenced by social behavior, with dispersing females displaying asocial behaviors such as avoiding other individuals and having larger home ranges (Ims, 1990).

REFERENCES

Barron, A. B. 2001. The Life and Death of Hopkins’ Host-Selection Principle. Journal of

Insect Behavior 14:725–737.

Biro, P. A. and J. A. Stamps. 2008. Are animal personality traits linked to life-history

productivity? Trends in Ecology and Evolution. 23:361–368.

Bohonak, A. J. 1999. Dispersal, gene flow, and population structure. The Quarterly

Review of Biology 74:21–45.

Boon, A. K., D. Réale, and S. Boutin. 2007. The interaction between personality,

offspring fitness and food abundance in North American red squirrels. Ecology

letters 10:1094–104. 13

Boon, A. K., D. Réale, and S. Boutin. 2008. Personality, habitat use, and their

consequences for survival in North American red squirrels Tamiasciurus

hudsonicus. Oikos 117:1321–1328.

Bowler, D. E., and T. G. Benton. 2005. Causes and consequences of animal dispersal

strategies: relating individual behaviour to spatial dynamics. Biological Reviews

80:205–225.

Briffa, M., and J. Greenaway. 2011. High in situ repeatability of behaviour indicates

animal personality in the beadlet anemone Actinia equina (Cnidaria). PloS. ONE.

6:e21963.

Budaev, S., D. Zworykin, and A. Mochek. 1999. Individual differences in parental care

and behaviour profile in the convict cichlid: a correlation study. Animal behaviour

58:195–202.

Byrom, A. E., and C. J. Krebs. 1999. Natal dispersal of juvenile arctic ground squirrels in

the boreal forest. Canadian Journal of Zoology 77:1048–1059.

Clark, A., and T. Ehlinger. 1987. Pattern and adaptation in individual behavioral

differences. Perspectives in Ethology 7:1–47.

Clobert, J., E. Danchin, A.A. Dhondt, and J. D. N. 2001. Dispersal. Oxford University

Press.

Cote, J. and J. Clobert. 2007. Social personalities influence natal dispersal in a lizard.

Proceedings of the Royal Society of Biological Sciences 274:383–390.

Davis, J. M., and J. A Stamps. 2004. The effect of natal experience on habitat

preferences. Trends in ecology & evolution 19:411–6. 14

Dethier, V. G. 1982. Mechanism of host-plant recognition. Entomologia Experimentalis

et Applicata 31:49-56.

Dingemanse, N. J., C. Both, P. J. Drent, and J. M. Tinbergen. 2004. Fitness consequences

of avian personalities in a fluctuating environment. Proceedings of the Royal

Society: Biological Sciences 271:847–52.

Dingemanse, N. J., C. Both, A. J. van Noordwijk, A. L. Rutten, and P. J. Drent. 2003.

Natal dispersal and personalities in great tits (Parus major). Proceedings of the

Royal Society Biological Sciences 270:741–7.

Dingemanse, N., and D. Réale. 2005. Natural selection and animal personality. Behaviour

142:1159–1184.

Dingemanse, N. J., J. Wright, A. J. N. Kazem, D. K. Thomas, R. Hickling, and N.

Dawnay. 2007. Behavioural syndromes differ predictably between 12 populations of

three-spined stickleback. Journal of Animal Ecology. 76:1128–1138.

Godin, J.-G. J., and L. A. Dugatkin. 1996. Female mate preference for bold males in the

guppy, Poecilia reticulata. Proceedings of the National Academy of Science

93:10262–10267.

Hollander, F. A., T. Van Overveld, I. Tokk and E. Matthysen. 2008. Personality and nest

defence in the great tit (Parus major). Ethology. 114:405–412.

Ims, R. A. 1990. Determinants of natal dispersal and space use in grey-sided voles

Clethrionomy rufocanus: a combined field and laboratory experiment. Oikos

57:106–113.

Loew, S. 1999. Sex-biased dispersal in eastern chipmunks, Tamias striatus. Evolutionary

Ecology 13:557–577. 15

Mabry, K. E., and J. A. Stamps. 2008. Dispersing brush mice prefer habitat like home.

Proceedings of the Royal Society Biological Sciences 275:543–8.

Mannan, R. W., R. N. Mannan, C. A. Schmidt, W. A. Estes-Zumpf, and C. W. Boal.

2006. Influence of natal experience on nest site selection by urban nesting Cooper’s

hawks. The Journal of Wildlife Management 71:64–68.

Martin, J. G. a., and D. Réale. 2007. Temperament, risk assessment and habituation to

novelty in eastern chipmunks, Tamias striatus. Animal Behaviour 75:309–318.

Réale, D., and M. Festa-Bianchet. 2003. Predator-induced natural selection on

temperament in bighorn ewes. Animal Behaviour 65:463–470.

Réale, D., B. Gallant, M. Leblanc, and M. Festa-Bianchet. 2000. Consistency of

temperament in bighorn ewes and correlates with behaviour and life history. Animal

behaviour 60:589–597.

Réale, D., S. M. Reader, D. Sol, P. T. McDougall, and N. J. Dingemanse. 2007.

Integrating animal temperament within ecology and evolution. Biological Reviews of

the Cambridge Philosophical Society 82:291–318.

Ronce, O. 2007. How does it feel to be like a rolling stone? Ten questions about dispersal

evolution. Annual Review of Ecology, Evolution, and Systematics 38:231–253.

Selonen, V., and I. K. Hanski. 2010. Movements of dispersing flying squirrels in relation

to siblings and parents. Behavioral Ecology and Sociobiology 64:1019–1027.

Selonen, V., I. K. Hanski, and A. Desrochers. 2007. Natal habitat-biased dispersal in the

Siberian flying squirrel. Proceedings of the Royal Society Biological Sciences

274:2063–8. 16

Sih, A., J. Cote, M. Evans, S. Fogarty, and J. Pruitt. 2012. Ecological implications of

behavioural syndromes. Ecology letters 15:278–89.

Sinn, D. L., L. A. Apiolaza, and N. A. Moltschaniwskyj. 2006. Heritability and fitness-

related consequences of squid personality traits. Journal of Evolutionary Biology

19:1437–1447.

Smith, B. R., and D. T. Blumstein. 2007. Fitness consequences of personality: a meta-

analysis. Behavioral Ecology. 19:448–455.

Stamps, J. A., and S. A. Blozis. 2006. Effects of natal experience on habitat selection

when individuals make choices in groups: a multilevel analysis. Animal Behaviour

71:663–672.

Stamps, J. A., V. V. Krishnan, and N. H. Willits. 2009. How different types of natal

experience affect habitat preference. The American Naturalist 174:623–30.

Stamps, J. A., and R. R. Swaisgood. 2007. Someplace like home: Experience, habitat

selection and conservation biology. Applied Animal Behaviour Science 102:392–

409.

Van Vuren, D., and K. Armitage. 1994. Survival of dispersing and philopatric marmots :

what is the cost of dispersal? Oikos 69:179–181.

Watters, J. V., and C. L. Meehan. 2007. Different strokes: Can managing behavioral

types increase post-release success? Applied Animal Behaviour Science 102:364–

379.

Weins, J. A. 1985. Habitat selection in variable environments: Shrub-steppe birds.

Pages 227-251 in Habitat Selection in Birds. Ed. M. L. Cody. Academic Press, Inc:

Orlando, FL. 17

CHAPTER 2: PRESENT STUDY

The methods, results, and conclusions of this study are presented in the paper appended to this thesis. The following is a summary of the major findings in this document. A detailed description of this research is presented in Appendix A and is formatted for submission to Behaviour.

Our study was conducted in 200 ha of mixed-conifer forest above 3,000 m on Mt.

Graham, an isolated 3,267-m peak in the Pinaleño Mountains, Graham County, Arizona,

USA. Mt. Graham is a habitat mosaic due to insect outbreaks and fire. We studied a population of semi-fossorial rodents, the cliff chipmunk (Tamias dorsalis). From June –

September 2014, we captured adults and juveniles, marked them with ear tags and colored washers, recorded reproductive status and mass, and radio-collared 6 females and

11 juveniles. We performed two behavior tests at the capture site; a 7-min open field

(OF) and 7-min mirror image stimulation (MIS) behavior test that measured an individual’s response to a novel environment (exploration and activity, OF) and to a conspecific

(aggression and sociality, MIS).

To determine female reproductive success, we used radio telemetry to track home range of females for 1 month, the time during which juveniles are still nursing. We also obtained counts of the number of juveniles in each female’s litter. To determine if habitat characteristics were different between natal and settled nest sites at the population level, we measured 6 habitat characteristics at both natal and settled nest sites of juveniles. We compared values of all habitat characteristics to see if juveniles settled in nest sites similar to natal sites. We also used personality score to examine any influence of behavioral phenotype on differences between natal and settled nest sites. 18

We identified 4 groups of behavioral responses to the OF test and 3 groups of responses to the MIS test. An individual’s activity, vigilance, sociality, or image engagement scores did not affect female litter size or 50% kernel home ranges (Table 3).

Vigilance was related to female 95% kernel home ranges (highly vigilant individuals had larger home ranges). At the population level, juveniles settled in nest sites that did not differ from natal nest sites for any of the six habitat characteristics. At the individual level, juveniles settled in nests that did differ from natal nests. Personality did not explain differences between natal and settled nest site characteristics.

Although we found no effect of personality on litter size in chipmunks, other mammalian species had an increase in reproductive success for bold and aggressive individuals. Other mechanisms such as food abundance, female body size, or stress hormones are important determinants of litter size in small mammals. NPHI has been documented across taxa and our result that juveniles prefer to settle in habitat similar to their natal habitat agrees with previous research. Settling in habitat similar to the natal habitat may increase survival, reproductive success, and juvenile survival. Our results indicate that sex and mass may explain differences in behavioral phenotypes; however, individual behavioral differences are weakly related to female reproductive success and settlement decisions during juvenile dispersal. 19

APPENDIX A: Behavioral phenotypes, female reproductive success, and juvenile dispersal: Insights from a semi-fossorial, social rodent

Short Title: Behavioral phenotypes in chipmunks

Allyssa L. Kilanowski1 & John L. Koprowski1

1School of Natural Resources and the Environment, University of Arizona, Tucson, AZ

85721-0001, USA [email protected] (ALK) [email protected] (JLK)

Abstract

Differences among individual responses to behavioral stimuli occur throughout a variety of taxa and these individual differences can affect female reproductive success and juvenile settlement decisions. We examined the effect of body size and life-stage on behavior phenotype and the effect of behavior on maternal reproductive success and juvenile dispersal of a fossorial rodent (Tamias dorsalis) in southeastern Arizona. We found that multiple behavioral phenotypes existed and female litter size was not affected by behavioral type. Furthermore, natal habitat preference induction (NHPI) does occur at the population level, and weakly occurs at the individual level. There was no influence of personality on site selection. Our results indicate that sex and mass may explain differences in behavioral phenotypes; however, individual behavioral differences are 20 weakly related to female reproductive success and settlement decisions during juvenile dispersal.

Key Words: personality, small mammal, home range, litter size, natal habitat preference induction, Tamias dorsalis, Arizona

Introduction

Differences among individual responses to behavioral stimuli occur in a variety of taxa (i.e. mammals [Martin & Réale, 2007; Boon et al., 2008], fish [Dingemanse et al.,

2007], reptiles [Cote & Colbert, 2007], invertebrates [Sinn et al., 2006; Briffa &

Greenaway, 2011], birds [Dingemanse et al., 2003]). These behavioral differences, referred to as individual behavioral phenotypes (hereafter personality) are defined as a repeatable set of observable behaviors (Clark & Ehlinger, 1987; Sih et al., 2012) and can vary across life-stage (Bekoff, 1977). Multiple behavioral phenotypes often exist within a population, likely a result of a relationship with individual fitness trade-offs such as survival or reproduction (Smith & Blumstein, 2007). While much work has been done to elucidate the effect of personality on individual fitness (Boon et al., 2008; Cote et al.,

2008; Hollander et al., 2008), variance of the results could indicate that we are just beginning to understand the effect of personality (Smith & Blumstein, 2007; Biro &

Stamps, 2008; Dammhahn, 2012).

Intrinsic individual factors, such as body mass or sex, can affect an individual’s personality (Brown et al., 2007). Behavioral differences between males and females are 21 frequently observed (Smith & Blumstein, 2007), and body condition and life-history strategies are two common explanations for personality differences (Ralls, 1977; Brown et al., 2007; Biro & Stamps, 2008; Réale et al., 2009; Dammhahn, 2012). Most studies of personality examine systems where females are smaller than males (Oddie, 2000;

Dingemanse et al., 2003; Dingemanse et al., 2007; Kitano et al., 2007). Species with male biased size dimorphism have males that are more aggressive than females

(Huntingford & Ruiz-Gomez, 2009) and more exploratory in a novel environment

(Brown et al., 2007; King et al., 2013). Species with reversed sexual dimorphism

(females larger than males) have females that are more exploratory in a novel environment or more aggressive than males (Chapple, 2005; While et al., 2009).

However, reversed sexual dimorphism primarily occurs in birds (Dingemanse et al.,

2003; Dingemanse et al., 2007) and fishes (Brown et al., 2007; Piyapong et al., 2009;

King et al., 2013), and less commonly in mammalian species (but see Ralls 1976).

Although body size is correlated with personality (Sinn et al., 2006), another key component of personality is the effect of personality on reproductive success and juvenile dispersal.

Female behavioral phenotype can affect reproductive success and survival; more aggressive and bold individuals demonstrate slightly enhanced success (Smith &

Blumstein, 2007). In great tits (Parus major), aggressive individuals are better able to defend the nest against predators, (Hollander et al., 2008) and bold females have increased foraging success (Biro & Stamps, 2008) leading to increased reproductive success. 22

The effect of personality on juvenile dispersal, specifically settlement decisions, is largely unknown, but may be disproportionate compared to environmental cues (Waters

& Meehan, 2007). To be sure, personality is likely to interact with other drivers to determine habitat selection. The natal habitat preference induction (NHPI) hypothesis states that animals will choose to settle in an area that is similar to their natal area (Davis and Stamps 2004, Stamps et al. 2009). NHPI may be adaptive, as habitat that produces individuals old enough to disperse is most likely habitat of reasonable quality, thus individuals use cues from the natal area to help reduce search time and increase fitness

(Stamps and Swaisgood 2007). NHPI has been studied in a variety of taxa including mice

(Peromyscus boylii, Mabry & Stamps, 2008), squirrels (Pteromys volans, Selonen et al.,

2007), birds (Accipiter cooperii, Mannan et al., 2006), and flies (Drosophila melanogaster, Stamps & Blozis, 2006); however, the interactive effects of differences among individual responses to behavioral stimuli on NHPI is less clear (Waters &

Meehan, 2006; Stamps & Swaisgood, 2007).

Small mammals are ideal to study personality and the relationship to maternal reproductive success and juvenile dispersal because of their abundance and existence of personalities (Martin & Réale, 2007; Boon et al., 2008; Selonen & Hanski, 2010). In red squirrels (Tamiasciurus hudsonicus), female personality does not influence litter size, but does have an effect on juvenile growth rate and offspring survival (Boon et al., 2008). In grey-sided voles (Clethrionomys rufocanus), juvenile dispersal is weakly influenced by social behavior, with dispersing females displaying asocial behaviors such as avoiding other individuals and having larger home ranges (Ims, 1990). 23

In this study, we asked the following questions for a semi-fossorial rodent population with reversed sexual dimorphism:

1) Do personalities exist?

2) Is personality consistent across time?

3) Does reproductive status, age, body mass, or sex correlate with personality?

4) Are ecological consequences of personality evident in maternal reproductive

success and juvenile settlement decisions?

Methods

Study Area and Species

Our study was conducted in 200 ha of mixed-conifer forest >3,000-m on Mt.

Graham, an isolated 3,267-m peak in the Pinaleño Mountains, Graham County, Arizona,

USA (2° 42′ 5.87″ N, 109° 52′ 18.87″ W). The dominant trees are Douglas-fir

(Pseudotsuga menziesii), southwestern white pine (Pinus strobiformis), and corkbark fir

(Abies lasiocarpa var. arizonica), mixed with Engelmann spruce (Picea engelmanii), aspen (Populus tremuloides) and ponderosa pine (Pinus ponderosa, Sanderson &

Koprowski, 2009). Mt. Graham is a habitat mosaic due to insect outbreaks and fire

(Wood et al., 2007).

We studied a population of semi-fossorial rodents, the cliff chipmunk (Tamias dorsalis), a granivorous rodent that ranges from Mexico to Utah, with populations as far east as New Mexico and west into Nevada (Hart, 1992). Although this species is considered semi-fossorial, the Mt. Graham population has been documented to climb trees to forage (often stealing food that was cached by the Mt. Graham red squirrel 24

[Tamiasciurus hudsonicus grahamensis, Edelman et al., 2005]), nesting in tree cavities, and using tree cavities for maternal nests (Hoffmeister, 1956; personal observation). This population exhibits reversed sexual dimorphism, with females being larger than males

(Levenson, 1990; Hart, 1992).

Individual Behavioral Phenotypes

From June – September 2014, we captured (3 x 3.5 x 9”, Large Folding

Galvanized Sherman Trap Sherman Trap, H.B. Sherman Traps, Tallahassee, FL) adults and juveniles, affixed ear tags (Monel #1, National Band & Tag, Newport, KY) and colored washers, and recorded age (adult or juvenile, i.e. young of the year), sex, reproductive status (nursing, scrotal, or non-reproductive) and mass. We performed two behavior tests at the capture site; a 7-min open field (OF) and 7-min mirror image

stimulation (MIS) behavior test (Boon et al., 2007; Martin & Réale, 2007) that measured an individual’s response to a novel environment (exploration and activity, OF; Martin &

Réale, 2007) and to a conspecific (aggression and sociality, MIS; Boon et al., 2007).

Individuals were placed in a behavioral arena (60 x 60 x 50 cm extruded polycarbonate box) containing a false floor with four blind holes and a mirror (45 x 45 cm) attached to one side of the arena and equipped with a removable cover. The arena lid contained a small hole for insertion of a USB digital web camera (LifeCam VX 6000, Microsoft,

Redmond, WA) to record animal behaviors. Animals were released into the arena, at which point the OF test began. Initially, the mirror was covered and animal activity was recorded for 7-min (OF test), after which the mirror cover was removed (MIS test) and the animal remained in the arena for another 7-min. We used JWatcher software (Blumstein 25 et al., 2011) to review behavior videos and quantify the total amount of time each individual spends on activities (Table 1). We used a principal components analysis (R

Development Core Team, 2010, R package “stats”, using singular value decomposition of our scaled and centered behavior data matrix) to reduce dimensionality of behaviors and to create two synthetic variables per behavioral test. Each synthetic variable permitted description of an individual’s personality compared to other animals in the sample population. Finally, we used K-means cluster analysis on all behavioral data summarized by the principal components analysis (Sokal & Rohlf, 1995; R package “stats”, R

Development Core Team, 2010) to determine if groups of personality responses to our behavioral tests exist. We used a one-way ANOVA (R package “stats”, R Development

Core Team, 2010) to determine if clusters are unequal along the first principal component and a post-hoc Tukey’s HSD (R package “stats”, R Development Core Team, 2010) test to determine which clusters are statistically different (Sokal & Rohlf, 1995).

Repeatability of Personality

Repeatability occurs when an individual exhibits the same behavioral response to a stimulus upon each occasion that the animal experiences the stimulus (Bell et al., 2009).

A strongly repeatable behavioral response can indicate that individuals are exhibiting personalities. For a subset of individuals recaptured at a later date, we performed a second behavioral test under the same procedure to determine if individual responses to the behavioral tests are consistent over time. Animals were always tested at site of capture. We performed a principal components analysis as previously described on a dataset containing the first and second behavior recordings that only included individuals 26 with a repeat behavior test. We performed a K-means cluster analysis (described above) on the subset of behavioral data and recorded the number of individuals that remained in the same behavioral cluster for the first and second behavioral tests. All repeatability analyses were conducted in R (R Development Core Team, 2010). We used R (R package

“stats”) to quantify the absolute value between behavioral test scores for each individual.

As the first principal component accounts for the most variation, we only quantified the difference between scores for the first principal component of each behavioral test.

Factors Influencing Personality

We used R to perform a stepwise regression (R package “stats” using both forward and backward elimination, entry or elimination of a variable into a model required a lower modified model AIC than the previous unaltered model) of a generalized linear model to determine if these factors influence personality scores. Stepwise model selection was performed separately for each personality principal component (4 total model sets) and all models were limited to 3-way or less interactions. Due to a lack of data for juvenile males (2 males), we did not include any interactions of sex and age in any model set. We also exclude the interaction of age and reproductive status, as all juveniles are non-reproductive; and we exclude interactions of mass and age because juveniles are always smaller than adults. One response variable, sociality (first principal component for the Mirror Image Stimulation test), was highly skewed and we performed a log transformation of the response variable before model selection.

Maternal Reproductive Success 27

In June-August, 2014, we used the methods described above to capture 6 nursing females using the protocols described above. We collared (SOM 2070, Wildlife Materials

Inc., Murphysboro, IL) nursing females before performing the initial behavior test. After juveniles emerged (2 July -2 August), we used radio telemetry to determine home range of females for 1 month when juveniles were still nursing. We obtained 16-30 points per female (25.67 points ± 5.61). We calculated 50% and 95% kernel density home ranges using ArcView (ESRI, Redlands, CA). We used R to perform a stepwise regression (same method as above) to determine if OF or MIS behavior scores affected home range size

(either 50% or 95%).

When juveniles were just beginning to emerge from the nest, we observed maternal nests on multiple days (mean: 2.8 days, range: 1-6) at sunrise to count juveniles during emergence and obtain a count of litter size. We used R to perform a stepwise regression (R package “stats” using both forward and backward elimination, entry or elimination of a variable into a model required a lower modified model AIC than the previous unaltered model) to determine if OF or MIS behavior scores affected litter size.

Natal Habitat Preference Induction and Personality

In June-August, 2014, we used the protocols described previously to capture 9 juveniles. We collared (SOM 2070, Wildlife Materials, Murphysboro, IL) and ear-tagged juveniles before performing the behavior test. We used radio telemetry to determine movements of juveniles for 1-3 months, during which juveniles were dispersing from the natal nest. We located night nests for each juvenile twice per month and for each nest we recorded the nest location, nest type (i.e. cavity, ground), snag class (if nest was in a snag, 28 classification system adapted from Smith and Mannan, 1994) and tree species (if nest was in a tree or snag, i.e. Pseudotsuga menziesii or Abies lasiocarpa var. arizonica).

To determine nest site characteristics, we used ArcMap (ESRI, Redlands, CA), to create a 12 m buffer (the perceptual range of a juvenile) around each nest used by juveniles, including the natal nest. We used the methods described in Mech and Zollner

(2002) to calculate perceptual range, the distance at which juveniles can perceive suitable environmental characteristics or cues, of juveniles. Our perceptual range calculations assumed a horizon height of 1.5 m, critical angular diversion of 63 and 69, and a body mass of 49 g. We used two angular divergence values to calculate two perceptual range

(lower: 11.85 m; upper: 13.08 m) distances, and we chose a perceptual range in the middle of our calculated results (12 m). We used zonal statistics to extract mean values for six habitat characteristics (average tree height, total basal area, live basal area, percent canopy cover, total woody volume, and deadfall volume) around each juvenile nest within a 12 m radius. We used remote sensing LiDAR (Light Detection and Ranging) to collect habitat characteristic data. We included natal and settled nests (the last known nest for each juvenile). Due to difficulties with maintaining collars, all juveniles have unknown fates. Therefore, we consider the last known nest location as a settled nests.

Our method is not optimal for studying NHPI as the last known nest may not be a settlement site; however, given the difficulties of tracking dispersing individuals, we consider the last known nest site as the best estimate of an individual’s settlement site selection. To fully understand if a juvenile cliff chipmunk settled, we need to track juveniles until they hibernate. Adult chipmunks remain within the same area across fall, 29 winter, and spring (Kilanowski, personal observation). Therefore, tracking juveniles until hibernation would indicate a settlement location.

To determine if any of our six habitat characteristics were different between natal and settled nest sites at the population level, we performed a permutation test (Sokal and

Rohlf 1995, R package “exactRankTests”). To compare categorical variables (tree nest species, nest type, and snag class) among natal and settled sites, we performed a Pearson

Chi-Square test (Sokal and Rohlf 1995). We used a paired permutation test to examine the difference between natal and settled nests among individuals for all six habitat characteristics (Sokal and Rohlf 1995, R package “exactRankTests”).

To determine if personality explains differences between natal and settled sites, we used a two-way MANOVA and included settlement (0 = final known nest; 1 = natal nest), personality (cluster obtained from the Open Field test), and the interaction of settlement and personality as factors. All calculations were conducted in R (R package

“stats”).

All behavioral tests and animal handling methods were approved by the

University of Arizona’s Institutional Animal Care and Use Committee (Protocol #08-

024). Field methods were conducted under permits from the United State Department of

Agriculture Forest Service (SAF2035-01) and Arizona Game and Fish Department

(SP654189). All results presented in this paper are formatted as mean ± SD, unless otherwise noted.

Results 30

We captured and performed behavioral tests on 37 adults (20 females; 17 males) and 11 juveniles (9 females; 2 males). Adults that received a behavior test exhibited sexual dimorphism (11.75%: mean female mass = 75.20 g ± 5.96; mean male mass =

67.29 g ± 5.23; t = 4.30, df = 35, p < 0.0001). Juveniles were not sexually dimorphic

(mean female mass = 49.33 g ± 7.58; mean male mass = 49.00 g ± 5.66; t = 0.07 df = 2, p

= 0.95).

Individual Behavioral Phenotypes

We identified 4 groups of behavioral responses to the OF test and 3 groups of responses to the MIS test (Figure 1). We retained two components from each behavioral test, which in combination accounted for 56% (OF test) and 54% (MIS test) of the observed variance (Table 1). The first principal component of the OF test (hereafter activity) was characterized by active and exploratory behaviors (i.e. immobile, locomotion, wall-climb, and sniffing) and the second principal component (hereafter vigilance) was characterized by vigilance behaviors (i.e. animal immobile, either alert or not alert, Table 1). The first principal component of the MIS test (hereafter sociality) was characterized by social behaviors (i.e. animal alert, immobile, and wary of image, or grooming) and the second principal component (hereafter image engagement) was characterized by social contact behaviors (i.e. animal touching mirror or tail waving,

Table 1). Differences existed among the 4 OF personality groups (ANOVA, F3, 44 = 59.08, p <0.001, Figure 2) and the 3 MIS personality groups (ANOVA, F2, 45 = 150.03, p <0.001,

Figure 2). Activity/exploration differed between vigilant explorers and non-vigilant non- explorers (PC scores; vigilant explorers: 1.34 ± 0.80, p <0.001; non-vigilant non- 31 explorers: -0.64 ± 1.10, p <0.001), vigilant explorers and vigilant non-explorers (PC scores; vigilant non-explorers: -1.27 ± 0.43, p <0.001), vigilant explorers and non- vigilant explorers (PC scores; non-vigilant explorers: 5.26 ± 1.96, p <0.001), non-vigilant non-explorers and non-vigilant explorers, and vigilant non-explorers and non-vigilant explorers. Vigilant non-explorers and non-vigilant non-explorers did not differ (p =

0.22). Sociality differed between amicable engagers and asocial avoiders (PC scores; amicable engagers: 3.76 ± 1.41, p <0.001; asocial avoiders: -1.11 ± 0.71, p <0.001) and asocial avoiders and amicable avoiders (PC scores; amicable avoiders: 4.93 ± 1.37, p

<0.001), but amicable engagers and amicable avoiders were not clearly distinct (p =

0.11).

Repeatability of Personality

Of the original 48 animals given behavior tests, 20 adults (12 females, 8 males) and 7 juveniles (5 females, 2 males) received a repeat behavior test. In the OF test, 17 of

27 animals changed behavioral cluster and in the MIS test, 6 of 27 animals changed behavioral cluster. For the OF test, individual behavior scores along the activity/exploration axis changed by 1.78 ± 2.03 principal component units (17.75% of the total axis range) and for the MIS test, individual behavior scores along the sociality axis changed by 2.05 ± 3.09 principal component units (19.23% of the total axis range).

Factors Influencing Personality

Activity/exploration was influenced by sex (females more vigilant than males,

GLM, t = -1.959, p = 0.057, CI: 0, 1.006), and negatively affected by mass (heavy animals 32 were less active, GLM, t = -1.323, p = 0.193, CI: 0.912, 1. 018). The effect of body mass on activity varied by sex, with females exhibiting a positive relationship and males exhibiting a negative relationship (GLM, t = 2.031, p = 0.048, CI: 1.005, 1.349). Vigilance was affected by age (juveniles less vigilant than adults, GLM, t = -3.134, p = 0.003, CI:

0.108, 0.599) and sex (males less vigilant than females, GLM, t = -2.024, p = 0.049, CI:

0.224, 0.976).

Sociality was affected by reproductive status (nursing females less social than non-reproductive individuals; Table 2). Image engagement was influenced by sex (males less engaged than females) and reproductive status (scrotal males more engaged than non-reproductive individuals; Table 2).

Maternal Reproductive Success

Nursing females exhibited home range sizes (50% kernel density: 0.26 ± 0.18 ha,

95% kernel density: 1.50 ± 0.99 ha) as reported for other populations (Hart 1992). An individual’s activity, vigilance, sociality, or image engagement scores did not affect female litter size or 50% kernel home ranges (Table 3). Vigilance affected female 95% kernel home ranges (highly vigilant individuals had larger home ranges; Table 3).

Natal Habitat Preference Induction and Personality

Comparing means for natal and settlement sites suggested that juveniles settled in nest sites that were similar to natal nest sites for all of the six habitat characteristics

(permutation test; percent canopy cover, p = 0.49; total basal area, p = 0.40; live basal area, p = 0.91; average tree height, p = 0.37; deadfall volume, p = 0.34; total woody 33 volume, p = 0.11; Figure 3). Also at the population–wide assessment, categorical habitat characteristics were not different among settled and natal nest sites (tree species: χ2 = 6.2, df = 4, p = 0.18; snag class: χ2 = 8.67, df = 4, p = 0.07; nest type: χ2 = 4.44, df = 3, p =

0.22).

For individual juveniles, settlement nests differed from natal nests in percent canopy cover (permutation test; p = 0.06), average tree height (permutation test; p =

0.06), total woody volume (permutation test; p = 0.06), total basal area (permutation test; p = 0.03), live basal area (permutation test; p = 0. 03), and volume of deadfall

(permutation test; p = 0.02, Figure 4). Personality did not explain differences between natal and settled nest site characteristics (MANOVA, F3 = 0.67, p = 0.79).

Discussion

Reversed Sexual Dimorphism

In many studies of personality, females are smaller than males (Oddie, 2000;

Dingemanse et al., 2003; Dingemanse et al., 2007; Kitano et al., 2007), but reversed sexual dimorphism occurs in some birds (Dingemanse et al., 2003; Dingemanse et al.,

2007) and fishes (Brown et al., 2007; Piyapong et al., 2009; King et al., 2013), and less commonly in mammalian species (but see Ralls, 1976). Similar to other T. dorsalis populations, our Mt. Graham population exhibited reversed sexual dimorphism (Hart,

1992); however, our adult body mass was much lower than that reported by Levenson

(1990). In Utah, juveniles exhibit sexual dimorphism from, June – August often well after emergence (Hart 1992). We weighed our juveniles immediately after nest 34 emergence and sexual dimorphism does not occur this early in development (Kilanowski, unpublished data).

Behavioral Phenotypes and Repeatability

Other small mammal species have discernable behavioral phenotypes (Benus &

Rondigs, 1996; Rauw et al., 2000; Benus, 2001; Cavigelli & McClintock, 2003;

Daniewski & Jezierski, 2003; Krackow, 2003; Martin & Réale, 2007; Boon et al., 2008).

We found that behavioral phenotypes did exist in cliff chipmunks. Cliff chipmunks did not display aggression toward the mirror image; instead, they avoided, ignored, or interacted with the image. This population nests socially (Kilanowski et al., unpublished data), and in some social species, individuals display reduced conspecific aggression.

However, our non-aggressive response from chipmunks is different from chickadees

(Peocile atricapillus, a social flocking individual) that displayed aggression to a mirror image at foraging sites (Sumitani, 2014). The accuracy of the mirror test can vary by species. In three species of cichlid (Neolamprologus pulcher, Telmatochromis vittatus, and Lepidiolamprologus eleongatus), each with different levels of sociality, the mirror image test only predicts intraspecific aggression for one species (Balzarini et al., 2014).

The mirror image test may have been unable to detect aggressive behavior in our social species; however, the mirror test did show reasonable repeatability in response.

Therefore, the mirror test for cliff chipmunks may measure sociality or tolerance, rather than aggression because aggressive interactions between chipmunks were not observed in the field (Kilanowski, personal observation). A meta-analysis of repeatability across 35 multiple taxa found that sociality/aggression are usually more repeatable than exploratory behaviors (Bell et al., 2009), corroborating our result.

Factors Affecting Personality

Intrinsic individual factors affect an individual’s personality (Brown et al., 2007;

Huntingford & Ruiz-Gomez, 2009; While et al., 2009; King et al., 2013). In cliff chipmunks, sex, mass, and age were predictors of an individual’s activity and vigilance behavior in a novel environment. Our activity/exploration behavioral responses are very different from other taxa, especially the interaction between mass and sex. We found that large males are less active than small males, however this is the opposite of other mammalian (Dammhahn, 2012; Edwards et al., 2013), aquatic (Brown & Irving, 2014), and avian systems (Hollander et al., 2007). Our finding that vigilance in a novel environment is lower in males than females is similar to other small mammals

(Dammhahn, 2012), but different from large mammals (Gartner & Powell, 2012). Cliff chipmunk exploratory behavior in a novel environment is clearly affected by sex and mass.

In our chipmunk population, both sex and reproductive status were predictors of sociality and image engagement (a proxy for conspecific engagement). We observed that male chipmunks were less social than females, a common trend in mammals, fishes, birds, and insects (reviewed in Smith & Blumstein, 2007), however scrotal male chipmunks were more engaged with the mirror image than females and non-reproductive

(male and female) chipmunks. We performed behavior tests during mating season when males are competing for mates, which may explain the high male engagement with mirror 36 images. Despite the promiscuous mating system of cliff chipmunks (Hart, 1992), where males are usually aggressively competing for mates, chipmunk male engagement with the mirror image was not aggressive, but more amicable. Male yellow pine chipmunks display minimal aggression to other males during the mating season (Schulte-Hostedde &

Millar, 2002). Chipmunks in the Mt. Graham population nest communally which may explain why, even during mating season, males were non-aggressive in their interactions with a mirror image.

Maternal Reproductive Success

Although we found no effect of personality on litter size, other mammalian species had an increase in reproductive success for bold and aggressive individuals (Biro

& Stamps, 2008; Hollander et al., 2008; Réale et al., 2009). Other mechanisms such as food abundance (Jonsson et al., 2002), female body size (Rieger, 1996), or stress hormones (Pratt & Lisk, 1989) are important determinants of litter size in small mammals.

In small mammals, metabolism and personality may be linked (Biro & Stamps,

2010; Lantová et al., 2011). We found that vigilance in a novel environment is correlated with the size of a nursing female’s home range. The high energetic requirements of nursing females (Millar, 1979; Humphries & Boutin, 1999; Fletcher et al., 2012) require an increase in home range (Lindstedt et al., 1986; Tufto et al., 1996; Saïd et al., 2005), which in turn requires increased vigilance as females forage in unfamiliar areas.

Natal Habitat Preference Induction and Personality 37

NPHI occurs across taxa (Davis & Stamps, 2004; Mabry & Stamps, 2008;

Selonen et al., 2007; Stamps & Blozis, 2006) and our result that juveniles prefer to settle in habitat similar to their natal habitat agrees with previous research (reviewed in Davis

& Stamps, 2004). Settling in habitat similar to the natal habitat may increase survival, reproductive success, and juvenile survival (reviewed in Stamps & Davis, 2006). While examining food resources at natal and settled sites is beyond the scope of this paper, it is important to consider because juveniles may be conditioned to survive on specific food items (Davis & Stamps, 2004) and selecting a settlement habitat similar to the natal habitat increases the likelihood of a juvenile finding food resources that it has previously experienced.

When examining NHPI for each individual through the use of pairwise tests, we observed much more variability in responses, with some individuals demonstrating behavior consistent with NHPI and others settling in nest sites that differed greatly from the natal site. While NHPI is used by many species, other factors determine selection of settlement site (Mannan et al., 2006; Ousterhout et al, 2014). Juvenile cliff chipmunks may not use forest structural habitat cues, such as tree height or canopy cover, to assess nest site quality. Juveniles may use conspecific abundance (Armstrong, 1995; Dunham,

2000; Van Zant & Wooten, 2003; Betts et al., 2008; Le Gouar et al., 2008; Morris &

MacEachern, 2010), heterospecific cues (Mönkkönen et al., 1999; Forsman et al., 2008;

Hromada et al., 2008), vulnerability to predation (Tabeni et al. 2012), individual body condition (Stamps, 2006; Ousterhout et al., 2014), or food abundance (Morin et al., 2005;

Morris & MacEachern, 2010) to evaluate habitat quality. Our results indicate that mass and sex may explain differences in behavioral phenotypes, and individual behavioral 38 differences are weakly related to female reproductive success and settlement decisions during juvenile dispersal.

Acknowledgements

We wish to thank T&E, Inc, the American Society of Mammalogists, and the

University of Arizona for providing financial support for this project. We also thank R.

William Mannan, David Christianson, Hsiang Ling Chen, Kendell Bennett, Amanda

Veals, Maria Altemus, and Kirsten Fulgham for feedback on early drafts of this manuscript. Special thanks to Melissa Merrick for designing the behavioral arena and providing advice on methods throughout this project.

Figures

Figure 1. Principal components analyses of behavioral phenotypes observed in a population of cliff chipmunks (Tamias dorsalis) on Mt. Graham, Arizona, USA based on two tests, the Open Field (OF, left) and Mirror Image Stimulation (MIS, right) tests.

Figure 2. Boxplot displaying the mean and range (1st and 3rd quartiles, including outliers) of novel vigilance, the first principal component of behavioral response clusters for the

Open Field (left), and sociality, the first principal component of behavioral response clusters for the Mirror Image Stimulation (right) behavior tests performed on a population of cliff chipmunks (Tamias dorsalis) on Mt. Graham, Arizona, USA in 2014. Letters above boxplots represent Tukey’s HSD post-hoc test. 41

Figure 3. Difference between natal and settled nest site characteristics for a population of cliff chipmunks (Tamias dorsalis) on Mt. Graham, Arizona, USA. 42

Figure 4. Mean (± SE) of the observed individual differences between natal and settled nest site habitat characteristics for a population of juvenile cliff chipmunks (Tamias dorsalis) on Mt. Graham, Arizona, USA. The zero line indicates no difference between natal and settled sites. Bars above the line occur when the natal nest site characteristic is greater than the settled site characteristic and points below the line occur when the natal site characteristic is less than the settled site characteristic. 43

1 Table 1. PCA loadings for behaviors in two behavior tests (Open Field and Mirror Image

2 Stimulation) in cliff chipmunks, Mt. Graham, AZ, USA. Only the first two components

3 were used in analyses. Units are proportion of time spent on each activity during the 7

4 min tests. Bold values indicate behaviors that were interpreted to contribute to a

5 component.

Open1Field1Test Mirror1Image1Stimulation1Test Behavior Components Behavior Components 1 2 1 2 Activity/0 Image0 Vigilance Sociality Exploration Engagement

Alert !0.14 0.68 Alert0(back0of0arena) 0.01 !0.42 Climb0wall 0.33 0.13 Alert0(front0of0arena) E0.06 0.33 Dig 0.36 E0.11 Climb0Wall 0.31 0.30 Groom 0.10 E0.05 Dig 0.24 E0.18 HeadEdip 0.32 0.05 Groom 0.26 0.27 Immobile !0.19 !0.67 HeadEdip 0.28 !0.29 Jump 0.32 E0.15 Immobile0(back0of0arena) E0.08 E0.15 Locomotion 0.43 E0.02 Immobile0(front0of0arena) !0.13 0.06 Rear0and0sniff 0.39 0.05 Jump 0.30 0.14 Scan 0.25 0.14 Locomotion 0.35 E0.14 TailEwave 0.30 E0.14 Movement0Towards0Mirror 0.29 !0.27 Nose0to0Mirror0Touch 0.31 0.28 Scan 0.34 0.08 Rear0and0Sniff 0.31 0.20 TailEwave 0.26 !0.41

Standard0deviance 2.07 1.35 Standard0deviance 2.37 1.59 Proportion0of0Total0 Proportion0of0Total0 0.39 0.17 0.37 0.17 deviance deviance Cumulative0Proportion0 Cumulative0Proportion0of0 0.39 0.56 0.37 0.54 6 of0Variance Variance

7 Alert – An animal that is immobile but aware of its surroundings; eyes open and head up.

8 In the MIS test, animals were classified as in the front of arena (half closest to the mirror)

9 or back of arena (half farthest away from mirror).

10 Head-dip – Animal inserts nose/head into 1 of 4 holes in the false floor 44

11 Immobile – Animal is immobile and appears unaware of surroundings: eyes closed and

12 head down. In the MIS test, animals were classified as in the front of arena (half closest to

13 the mirror) or back of arena (half farthest away from mirror).

14 Rear and sniff – Animal is either on all fours or resting on hindquarters. In some cases,

15 animals sniffed the air while in these positions and this data was lumped into a single

16 category.

17 Scan – Animal body was unmoving, all four paws on ground; head moved side to side as

18 animal visually assessed surroundings.

19 Tail-wave – Animal body unmoving and head alert, tail waves side to side 45

20 Table 2. Estimated coefficients and confidence intervals of generalized linear models for

21 the two principal components of the mirror image stimulation behavioral test (sociality

22 and image engagement) for cliff chipmunks (Tamias dorsalis) on Mt. Graham, Arizona,

23 USA in 2014. Bold values indicate significance (p < 0.05) and italicized values indicate

24 potential biological significance (p < 0.1).

Variables Coefficient SE t+value P+value 95%+CI+(lower) 95%+CI+(upper)

Sociality Age$(adult$as$reference) 1.386 0.833 71.663 0.135 0.049 1.281 Sex$(female$as$reference) 1.823 1.178 1.547 0.160 0.615 62.347 Reproductive$Status$(no$signs$as$reference) Nursing 71.618 0.833 71.942 0.088 0.039 1.015 Scrotal 71.929 1.361 71.418 0.195 0.010 2.091

Image+Engagement Sex$(female$as$reference) 70.986 0.586 71.684 0.099 0.118 1.175 Reproductive$Status$(no$signs$as$reference) Nursing 0.694 0.642 1.080 0.286 0.568 7.047 25 Scrotal 1.701 0.705 2.412 0.020 1.376 21.813 46

26 Table 3. Estimated coefficients and confidence intervals of linear models for the litter

27 size and female home range models of cliff chipmunks (Tamias dorsalis) on Mt. Graham,

28 Arizona, USA in 2014. Two models are a 50% and 95% kernel density home range.

29 Bold values indicate significance (p < 0.05) and italicized values indicate potential

30 biological significance (p < 0.1).

Variables Coefficient SE t+value P+value 95%+CI+(lower) 95%+CI+(upper)

95% Activity/Exploration 00.20 0.16 01.23 0.43 0.60 1.12 Vigilance 0.65 0.07 8.79 0.07 1.65 2.20 Sociality 0.16 0.03 5.37 0.12 1.11 1.25 ImageBEngagement 0.18 0.03 6.46 0.10 1.13 1.26

50% Activity/Exploration 00.71 0.33 02.16 0.28 0.26 0.94 Vigilance 00.22 0.15 01.43 0.39 0.60 1.08 Sociality 0.16 0.06 2.46 0.25 1.03 1.33 ImageBEngagement 00.04 0.06 00.78 0.58 0.85 1.07

Litter+Size 31 Intercept 2.33 0.42 5.53 0.00 4.51 23.56

32 47

33 References

34 Armstrong, D. P. (1995). Effects of familiarity on the outcome of translocations, II. A

35 test using New Zealand robins. – Biol. Cons. 71:281-288.

36 Balzarini, V. & Taborsky, M. (2014). Mirror, mirror on the wall: the predictive value of

37 mirror tests for measuring aggression in fish. – Behav. Ecol. Sociobiol. 68:871-878.

38 Bekoff, M. 1977. Mammalian dispersal and the ontogeny of individual behavioral

39 phenotypes. – Am. Nat. 111:715-732.

40 Bell, A. M., Hankison, S. J. & Laskowski, K. L. (2009). The repeatability of behaviour: a

41 meta-analysis. – Anim. Behav. 77:771-783.

42 Benus, R. F. (2001). Coping in female mice from lines bidirectionally selected for male

43 aggression. – Behav. 138:997-1008.

44 Benus, R. F. & Rondigs, M. (1996). Patterns of maternal effort in mouse lines

45 bidirectionally selected for aggression. – Anim. Behav. 51:67-75.

46 Betts, M. G., Hadley, A. S., Rodenhouse, N., & Nocera, J. J. (2008). Social information

47 trumps vegetation structure in breeding site selection by a migrant songbird. – Proc.

48 R. Soc. B. 275:2257-2263.

49 Biro, P. A. & Stamps, J. A. (2008). Are animal personality traits linked to life-history

50 productivity? – Trends. Ecol. Evol. 23:361-368.

51 Blumstein, D.T., Daniel, J.C. & Evans, S. C. (2011). JWatcher Video v 1.0. University of

52 California Los Angeles.

53 Boon, A. K., Réale, D. & Boutin, S. (2007). The interaction between personality,

54 offspring fitness and food abundance in North American red squirrels. – Ecol. Lett.

55 10:1094-1104. 48

56 Boon, A. K., Réale, D., & Boutin, S. (2008). Personality, habitat use, and their

57 consequences for survival in North American red squirrels Tamiasciurus

58 hudsonicus. – Oikos 117:1321-1328.

59 Briffa, M. & Greenaway, J. (2011). High in situ repeatability of behaviour indicates

60 animal personality in the beadlet anemone Actinia equina (Cnidaria). – PloS. ONE.

61 6:e21963.

62 Brown, C. & Irving, E. (2014). Individual personality traits influence group exploration

63 in a feral guppy population. – Behav. Ecol. 25:95-101.

64 Brown, C., Jones, F. & Braithwaite, V. A. (2007). Correlation between boldness and

65 body mass in natural populations of the poeciliid Brachyrhaphis episcopi. – J. Fish.

66 Biol. 71:1590-1601.

67 Cavigelli, S. A. & McClintock, M. K. (2003). Fear of novelty in infant rats predicts adult

68 corticosterone dynamics and an early death. – Proc. Natl. Acad. Sci. USA.

69 100:16131-16136.

70 Chapple, D. G. (2005). Life history and reproductive ecology of White’s skink, Egernia

71 whitii. – Aus. J. Zool. 53:353-360.

72 Clark, A. B. & Ehlinger, T. J. (1987). Pattern and adaptation in individual behavioral

73 differences. – Perspect. Ethol. 7:1-47.

74 Cote, J. & Clobert, J. (2007). Social personalities influence natal dispersal in a lizard. –

75 Proc. R. Soc. B. 274:383-390.

76 Cote, J., Dreiss, A. & Clobert, J. (2008). Social personality trait and fitness. – Proc. R.

77 Soc. B 275:2851-2858. 49

78 Daniewski, W. & Jezierski, T. (2003). Effectiveness of divergent selection for open-field

79 activity in rabbits and correlated response for body weight and fertility. – Behav.

80 Genet. 33:337-345.

81 Dammhahn, M. (2012). Are personality differences in a small iteroparous mammal

82 maintained by a life-history trade-off? – Proc. R. Soc. B. 279:2645-2651.

83 Davis, J. M., & Stamps, J. A. (2004). The effect of natal experience on habitat

84 preferences. – Trends. Ecol. Evol. 19:411-6.

85 Dingemanse, N. J., Both, C., van Noordwijk, A. J., Rutten, A. L. & Drent, P. J. (2003).

86 Natal dispersal and personalities in great tits (Parus major). – Proc. R. Soc. B.

87 270:741-747.

88 Dingemanse, N. J., Wright, J., Kazem, A. J. N., Thomas, D. K., Hickling, R. & Dawnay,

89 N. (2007). Behavioural syndromes differ predictably between 12 populations of

90 three-spined stickleback. – J. Anim. Ecol. 76:1128-1138.

91 Dunham, K. M. (2000). Dispersal pattern of mountain gazelles, Gazella gazella,

92 released in central Arabia. J. Arid. Environ. 44:247–258.

93 Edelman, A. J., Koprowski, J. L. & Edelman, J. L. (2005). Kleptoparasitic behavior and

94 species richness at Mt. Graham red squirrel middens. – USDA Forest. Serv. Proc.

95 RMRS-P. 36:395-398.

96 Edwards, A. M., Best, E. C., Blomberg, S. P., & Goldizen, A. W. (2013). Individual traits

97 influence vigilance in wild female eastern grey kangaroos. – Aus. J. Zool. 61:332-

98 341.

99 Fletcher, Q.E., Speakman, J.R., Boutin, S., McAdam, A.G., Woods, S.B. & Humphries,

100 M.M. (2012) Seasonal stage differences overwhelm environmental and individual 50

101 factors as determinants of energy expenditure in free-ranging red squirrels. – Func.

102 Ecol. 26:677-687.

103 Forsman, J.T., Hjernquist, M. B., Taipale, J. & Gusafsson, L. (2008). Competitor

104 density cues for habitat quality facilitating habitat selection and investment

105 decisions. – Ecology. 19:539-545.

106 Gartner, M. C. & Powell, D. (2012). Personality assessment in snow leopards (Uncia

107 uncia). – Zoo. Biol. 31:151-165.

108 Hart, E. B. (1992). Tamias dorsalis. – Mammal. Spec. 399:1-6

109 Hoffmeister, D. (1956). Mammals of the Graham (Pinaleno) Mountains, Arizona. – Am.

110 Mid. Nat. 55:257-288.

111 Hollander, F. A., Van Overveld, T., Tokk, I. & Matthysen, E. (2008). Personality and

112 nest defence in the great tit (Parus major). – Ethology. 114:405-412.

113 Hromada, M., Antczak, M., Valone, T. J., & Tryjanowski, P. (2008). Settlement

114 decisions and heterospecific social information use in shrikes. – PloS. One. 3:e3930.

115 Humphries, M. M., & Boutin, S. (1999). Mass-dependent reproduction or reproduction-

116 dependent mass? A comment on body mass and first-time reproduction in female

117 sciurids. – Can. J. Zool. 77:171-173.

118 Huntingford, F. A. & Ruiz-Gomez, M. L. (2009). Three-spined sticklebacks Gasterosteus

119 aculeatus as a model for exploring behavioural biology. – J. Fish. Biol. 75:1943-

120 1976.

121 Ims, R. A. (1990). Determinants of natal dispersal and space use in grey-sided voles

122 Clethrionomy rufocanus: a combined field and laboratory experiment. – Oikos.

123 57:106-113. 51

124 Jonsson, P., Hartikainen, T., Koskela, E. & Mappes, T. (2002). Determinants of

125 reproductive success in voles: Space use in relation to food and litter size

126 manipulation. – Evol. Ecol. 16:455-467.

127 King, A. J., Fürtbauer, I., Mamuneas, D., James, C. & Manica, A. (2013). Sex-differences

128 and temporal consistency in stickleback fish boldness. – PLoS. ONE. 8:31-36.

129 Kitano, J., Mori, S. & Peichel, C. L. (2007). Sexual dimorphism in the external

130 morphology of the threespine stickleback (Gasterosteus aculeatus). – Copeia. 2:336-

131 349.

132 Krackow, S. (2003). Motivational and heritable determinants of dispersal latency in wild

133 male house mice (Mus musculus musculus). – Ethology. 109:671-689.

134 Lantová, P., Zub, K., Koskela, E., Šíchová, K. & Borowski, Z. (2011). Is there a linkage

135 between metabolism and personality in small mammals? The root vole (Microtus

136 oeconomus) example. – Physiol. Behav. 104:378-383.

137 Le Gouar, P., Robert, A., Choisy, J.-P., Henriquet, S., Lecuyer, P., Tessier, C., &

138 Sarrazin, F. (2008). Roles of survival and dispersal in reintroduction success of

139 Griffon vultures (Gyps fulvus). – Ecol. Appl. 18:859-972.

140 Levenson, H. (1990). Sexual size dimorphism in chipmunks. – J. Mammal. 71:161-170.

141 Lindstedt, S. L., Miller, B. J. & Buskirk, S. W. (1986). Home range, time, and body size

142 in mammals. – Ecol. 67:413-418

143 Mabry, K. E., & Stamps, J. A. (2008). Dispersing brush mice prefer habitat like home. –

144 Proc. R. Soc. B. 275:543-548. 52

145 Mannan, R. W., Mannan, R. N., Schmidt, C. A., Estes-Zumpf, W. A., & Boal, C. W.

146 (2006). Influence of natal experience on nest site selection by urban nesting

147 Cooper’s hawks. – J. Wild. Manag. 71:64-68.

148 Martin, J. G. A. & Réale, D. (2007). Temperament, risk assessment and habituation to

149 novelty in eastern chipmunks, Tamias striatus. – Anim. Behav. 75:309-318.

150 Mech, S. G., & Zollner, P. A. (2002). Using body size to predict perceptual range. –

151 Oikos. 98:47-52.

152 Millar, J. S. (1979). Energetics of lactation in Peromyscus maniculatus. – Can. J. Zool.

153 57: 1015-1019.

154 Mönkkönen, M., Härdling, R., Forsman, J. T., & Tuomi, J. (1999). Evolution of

155 heterospecific attraction: using other species as cues in habitat selection. – Evol.

156 Ecol. 13:91-104.

157 Morin P., Berteaux, D., & Klvana, I. (2005). Hierarchical habitat selection by North

158 American porcupines in southern boreal forest. – Can. J. Zool. 83:1333-1342.

159 Morris, D., & MacEachern, J. (2010). Active density-dependent habitat selection in a

160 controlled population of small mammals. – Ecol. 91:3131-3137.

161 Oddie, K. R. (2000). Size matters: competition between male and female great tit

162 offspring. – J. Anim. Ecol. 69:903-912.

163 Ousterhout, B. H., Luhring, T. M., & Semlitsch, R. D. (2014). No evidence of natal

164 habitat preference induction in juveniles with complex life histories. – Anim. Behav.

165 93:237-242. 53

166 Piyapong, C., Krause, J., Chapman, B. B., Ramnarine, I. W., Louca, V. & Croft, D. P.

167 (2010). Sex matters: A social context to boldness in guppies (Poecilia reticulata). –

168 Behav. Ecol. 21:3-8.

169 Pratt, N. C., & Lisk, R. D. (1989). Effects of social stress during early pregnancy on

170 litter size and sex ratio in the golden hamster (Mesocricetus auratus). – J. Reprod.

171 Fert. 87:763-769.

172 R Development Core Team. (2010). R: A language and environment for statistical

173 computing. R Foundation for Statistical Computing, Vienna, Austria.

174 Ralls, K. (1976). Mammals in which females are larger than males. – Q. Rev. Biol.

175 51:245-276.

176 Ralls, K. (1977). Sexual dimorphism in mammals: Avian models and unanswered

177 questions. – Am. Nat. 111:917-938.

178 Rauw, W. M., Lutting, P., Bakken, M., Schuurman, T., De Veer, C. J. M & Vangen, O.

179 (2000). Behavioral differences in non-reproductive adult females in a long-term

180 selection experiment for litter size in mice. – App. Anim. Behav. Sci. 66:249-262.

181 Réale, D., Martin, J., Coltman, D. W., Poissant, J. & Festa-Bianchet, M. (2009). Male

182 personality, life-history strategies and reproductive success in a promiscuous

183 mammal. – J. Evol. Biol. 22:1599-1607.

184 Rieger, J. F. (1996). Body size, litter size, timing of reproduction, and juvenile survival

185 in the Uinta ground squirrel, Spermophilus armatus. – Oecologia. 107:463-468.

186 Saïd, S., Gaillard, J.M., Duncan, P., Guillon, N., Servanty, S., Pellerin, M., Lefeuvre, K.,

187 Martin, C. & Van Laere, G. (2005). Ecological correlates of home-range size in 54

188 spring-summer for female roe deer (Capreolus capreolus) in a deciduous woodland.

189 – J. Zool. 267:301-308.

190 Sanderson, H. R. & Koprowski, J. L. Eds. (2009). The last refuge of the Mt. Graham red

191 squirrel: Ecology of endangerment. The University of Arizona Press: Tucson, AZ.

192 Schulte-Hostedde, A. I., & J. S. Millar. (2002). ‘Little chipmunk’ syndrome? Male body

193 size and dominance in captive yellow-pine chipmunks (Tamias amoenus). – Ethol.

194 108:127-137.

195 Selonen, V. & Hanski, I. K. (2010). Condition-dependent, phenotype-dependent and

196 genetic-dependent factors in the natal dispersal of a solitary rodent. – J. Anim. Ecol.

197 79:1093-1100.

198 Selonen, V., Hanski, I. K., & Desrochers, A. (2007). Natal habitat-biased dispersal in

199 the Siberian flying squirrel. – Proc. R. Soc. B. 274:2063-2068.

200 Sih, A., Cote, J., Evans, M., Fogarty, S. & Pruitt, J. (2012). Ecological implications of

201 behavioural syndromes. – Ecol. Lett.15:278-289.

202 Sinn, D. L., Apiolaza, L. A. & Moltschaniwskyj, N. A. (2006). Heritability and fitness-

203 related consequences of squid personality traits. – J. Evol. Biol. 19:1437-1447.

204 Smith, B. R. & Blumstein, D. T. (2007). Fitness consequences of personality: a meta-

205 analysis. – Behav. Ecol. 19:448-455.

206 Smith, A. A., & Mannan, R. W. (1994). Distinguishing characteristics of Mount Graham

207 red squirrel midden sites. – J. Wild. Manag. 58:437-445.

208 Sokal, R. R., & Rohlf, F. J. (1995). Biometry. Macmillan:Freeman, NY.

209 Stamps, J. A. (2006). The silver spoon effect and habitat selection by natal dispersers. –

210 Ecol. Lett. 9:1179-85. 55

211 Stamps, J. A., & Blozis, S. A. (2006). Effects of natal experience on habitat selection

212 when individuals make choices in groups: a multilevel analysis. – Anim. Behav.

213 71:663-672.

214 Stamps, J. A., & Davis, J. M. (2006). Adaptive effects of natal experience on habitat

215 selection by dispersers. – Anim. Behav. 72:1279-1289.

216 Stamps, J. A., & Swaisgood, R. R. (2007). Someplace like home: Experience, habitat

217 selection and conservation biology. – App. Anim. Behav. Sci. 102:392-409.

218 Stamps, J. A., Krishnan, V. V., & Willits, N. H. (2009). How different types of natal

219 experience affect habitat preference. – Am. Nat. 174:623-630.

220 Sumitani, M. (2014). Mirror effects on feeding black-capped chickadees (Poecile

221 atricapillus). – J. Young. Investig. 26:23-31.

222 Tabeni, S., Marcos, N., Rosi, M. I. & Bender, B. (2012). Vulnerability of small and

223 medium-sized prey mammals in relation to their habitat preferences, age classes and

224 locomotion types in the temperate Monte Desert, Argentina. – Mamm. Biol. 77:90-

225 96.

226 Tufto, J., Andersen, R. & Linnell, J. (1996) Habitat use and ecological correlates of home

227 range size in a small cervid: the roe deer. – J. Anim. Ecol. 65:715-724.

228 Van Zant, J. L., & Wooten, M. C. (2003). Translocation of Choctawatchee beach mice

229 (Peromyscus polionotus allophrys): hard lessons learned. – Biol. Cons. 112:405-

230 413.

231 Watters, J. V., & Meehan, C. L. (2007). Different strokes: Can managing behavioral

232 types increase post-release success? – App. Anim. Behav. Sci. 102:364-379. 56

233 While, G. M., Sinn, D. L. & Wapstra, E. (2009). Female aggression predicts mode of

234 paternity acquisition in a social lizard. – Proc. R. Soc. B. 276:2021-2029.

235 Wood, D. J. A., Drake, S., Rushton, S. P., Rautenkranz, D., Lurz, P. W. W. &

236 Koprowski, J. L. (2007). Fine-scale analysis of Mt. Graham red squirrel habitat

237 following disturbance. – J. Wildl. Manag. 71:2357-2364.