Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Elevation and the avian phenotype : field and experimental studies of breeding dark-eyed juncos Bears, Heather 2007

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Notice for Google Chrome users:
If you are having trouble viewing or searching the PDF with Google Chrome, please download it here instead.

Item Metadata

Download

Media
831-ubc_2007-317372.pdf [ 8.93MB ]
Metadata
JSON: 831-1.0074960.json
JSON-LD: 831-1.0074960-ld.json
RDF/XML (Pretty): 831-1.0074960-rdf.xml
RDF/JSON: 831-1.0074960-rdf.json
Turtle: 831-1.0074960-turtle.txt
N-Triples: 831-1.0074960-rdf-ntriples.txt
Original Record: 831-1.0074960-source.json
Full Text
831-1.0074960-fulltext.txt
Citation
831-1.0074960.ris

Full Text

E L E V A T I O N AND T H E AVIAN P H E N O T Y P E : FIELD AND E X P E R I M E N T A L STUDIES OF BREEDING D A R K - E Y E D JUNCOS by HEATHER BEARS B.Sc. (Hons), Environmental Biology, University of Alberta, 1999 M.Sc. Zoology, University of British Columbia, 2003 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY in THE F A C U L T Y OF G R A D U A T E STUDIES (Forestry) THE UNIVERSITY OF BRITISH C O L U M B I A August, 2007 © Heather Bears, 2007 ABSTRACT Forty percent of the terrestrial planet is mountainous, yet little is known about how breeding elevation affects avian phenotypes. I studied dark-eyed juncos breeding at the extremes of their elevation range in Jasper, A B ( 'Low ' ; 1000 m and 'H igh ' ; 2000 m a.s.l.) from 2000-2005.1 compared reproductive and morphological traits in free-living birds between elevations to establish patterns of change with breeding elevation. I subsequently investigated mechanisms underlying those patterns by collecting hatchlings and adults of both sexes from each elevation and raising them in a common lab environment. The common lab experiment allowed me to determine the amount of variation due to phenotypic plasticity in response to local conditions. Measurements in the field included indicators of reproductive stage, seasonal reproduction, philopatry/survival, age ratios, morphometries, and local weather. Measurements in the lab included indicators of reproductive development over time (after birds were stimulated to breed with an increasing photoperiod), and morphometries (after birds replaced feathers in captivity). I addressed the following questions: (1) How do life-histories vary with breeding elevation?, (2) What environmental factors correlate with reproductive timing between elevations?, (3) Are reproductive differences between elevations due to local genetic adaptation or phenotypic plasticity?, and (4) Does morphology vary with elevation, and are differences due to local genetic adaptation or phenotypic plasticity? High-elevation birds in the field became reproductively capable >6 weeks later than low-elevation birds and produced half the number of broods and offspring per season. High-elevation habitats were not occupied by younger, smaller (i.e., less competitive) birds, and mark-recapture analysis suggested that high-elevation birds live longer. High-elevation males and females had longer tails and wings, respectively, than low-elevation birds. Both populations initiated more nests as rain and insect abundance increased. However, high-elevation birds initiated nests synchronously when the snow melted, while low-elevation birds initiated nests as growing degree days increased. In captive birds, the timing of breeding readiness was reversed in the lab relative to the field in both sexes and ages: i.e., high-elevation birds were able to breed earlier than low-elevation birds. Birds from both elevations increased morphological trait sizes in the lab relative to the field, suggesting they were released from growth constraints. However, after birds grew in the lab, relative differences between elevation groups remained (in the same direction as in the field), or were exacerbated. The reproductive and morphological responses observed in the lab relative to the field suggest that environmental constraints and countergradient forces can interact in complex, unexpected ways to shape avian phenotypes among elevations. i i i TABLE OF CONTENTS Abstract i i Table of Contents iv List of Tables vi List of Figures v i i List of Appendices x i List of Symbols x i i Acknowledgements x i i i Dedication xiv Co-Authorship Statement • xv 1 Chapter 1: General Introduction 1 1.1 Phenotypic Variation • 1 1.2 Phenotypic Variation with Elevation 3 1.3 Thesis Objectives 7 1.4 Study System 7 1.5 Thesis Overview 9 1.6 Literature Cited 13 2 Breeding Elevation as a Determinant of the Life-History Strategy of a Songbird (Junco hyemalis) 25 2.1 Introduction 25 2.2 Methods 27 2.2.1 Species 27 2.2.2 Study sites 28 2.2.3 Field procedures 28 2.2.4 Reproductive timing and seasonal reproductive success 29 2.2.5 Variation in reproductive development and termination 30 2.2.6 Apparent survival analysis 31 2.2.7 Data analysis 32 2.3 Results 33 2.3.1 Variation in egg-laying 33 2.3.2 Variation in reproductive development/termination 33 2.3.3 Variation in seasonal reproductive success 34 2.3.4 Variation in competitive status 35 2.3.5 Variation in apparent survival ....35 2.4 Discussion 36 2.5 Literature Cited 43 3 Weather and Timing of Breeding Among Elevations- Is Brood Initiation Correlated with The Same or Different Weather Variables Among Elevations? 61 3.1 Introduction 61 3.2 Methods 63 3.2.1 Study species and study sites 63 3.2.2 Environmental variables 63 3.2.3 Breeding phenology data 64 3.2.4 Statistical analyses 64 iv 3.3 Results 66 3.4 Discussion 67 3.5 References 76 4 Comparative Reproductive Phenology of Dark-eyed Juncos Breeding at Two Elavations: A Common Aviary Experiment 81 4.1 Introduction 81 4.2 Methods 84 4.2.1 Captive birds and experimental design 84 4.2.2 Sampling Techniques 85 4.2.3 Statistical Analyses 87 4.3 Results 88 4.3.1 Cloacal protuberance size 88 4.3.2 Functional gonads 88 4.3.3 Song frequencies 89 4.3.4 Ovary scores , 90 4.3.5 Brood patch 90 4.4 Discussion 91 4.5 References 105 5 Comparative Morphology of Dark-eyed Juncos Breeding at Two Elevations: A Common Aviary Experiment 110 5.1 Introduction 110 5.2 Methods 113 5.2.1 Field methods 113 5.2.2 Captive birds and experimental design 114 5.2.3 Data analysis 116 5.3 Results 117 5.3.1 Morphological differences among elevations 117 5.3.2 Common aviary experiment 117 5.4 Discussion 119 5.5 References 129 6 General Discussion 134 6.1 References 145 Appendices 150 Appendix I 150 Appendix II 151 Appendix III 152-153 Appendix I V 154 v LIST OF T A B L E S Table 1.1 Weather variables and insect biomass (food) summarized by year and month for high- and low-elevation sites in Jasper National Park, Alberta 15-16 Table 2.1 Mode l rankings and criteria from the computer program M A R K (White and Burnham 1999) for the predominant factors tested to explain dark-eyed junco (Junco hyemalis) monthly survival (9) and re-sighting probabilities (p) from 2000-2005 in high- and low- elevation sites in Jasper National Park. Akaike's Information Criterion (AIC), difference between A I C values between models (AAICc) , and A I C weights are included to show the level of support. Mode l selection results supported the {cp(elevation + /) p(elevation)} model as the most parsimonious with 0.999 weight, which describes survival probability as time-specific (p values, with recapture probability allowed to vary with elevation. The period symbol (.) means that the parameter was constant in the model 53 Table 3.1 Pearson correlation coefficients of four potential supplementary cues measured in high- and low-elevation habitats in Jasper National Park, Alberta, Canada. Variables are: S N O W = snow depth, IN= insect biomass, R N = cumulative rainfall (mm), GDD=growing degree days. Coefficients significant at a = 0.05 are marked with *, and those significant at 0.01 are marked with **. — means the correlation could not be measured 72 Table 3.2 Best-fit regression models for each variable examined (X) relative to cumulative percentage of broods (Y) initiated in high- and low-elevation habitats. For all parameters estimated, the asymptotic standard error was less than 3% of the estimated parameter value 73 Table 5.1 Pearson cross-correlation coefficients of eight morphological traits of wild-caught and captive dark-eyed juncos (Junco hyemalis) from Jasper National Park, Alberta, Canada, from 2000, 2001, 2004 and 2005. N = 298. Traits are: T A R -Tarsus length, W I N G = wing length, T A I L = tail length, W E I G H T = average v i weight, F A T = average fat score, B D = beak depth, B W = beak width, and B L = beak length. Bolded coefficients are significant at a = 0.05 125 Table 5.2 Morphological traits of wild-caught after-second year ( A S Y ) and second-year (SY) dark-eyed juncos (Junco by emails) at high- and low-elevations in Jasper National Park, Alberta, Canada, from 2000, 2001, 2004 and 2005. Traits are expressed as means, with sample sizes and standard deviations in brackets (N, SD). Means in bold are significantly different between elevations, at a = 0.05, based on linear mixed models. See Appendix III for details 126 v i i LIST OF FIGURES Fig. 1.1 The dark-eyed junco (Junco hyemalis), a songbird in the sparrow family emberizidae, identified by a dark hood that contrasts against a lighter body (darker in males than in females), and by white tail feathers that appear as white stripes when the tail feathers are fanned in courtship or territorial display 13 Fig. 1.2 Study site locations in Jasper National Park, Canada. Triangles are high-elevation sites and diamonds are low-elevation sites; those shaded black are field study sites, and those shaded white are field and collection sites 14 Fig. 2.1 Percentages of clutches initiated (date of first egg) by date in high- versus low-elevation study sites in 2000 [n(high)=67, n(low)=83], 2001[n(high)=81, n(low)=82], and 2004[n(high)=60, n(low)=58,]. Dates corresponding to 10 t h and 90th percentiles (initiation and cessation dates, respectively) are indicated by beginning and end of horizontal boxplot rectangles. Median initiation dates are indicated by the central lines within boxplots. The number of days over which broods were initiated at each elevation is indicated above boxplots 54-55 Fig. 2.2 Mean cloacal widths (±1SE) of male dark-eyed juncos from high- and low-elevation habitat throughout the 2004 breeding season. High- and low-elevation groups differed significantly from one another within A p r i l , M a y and July sample intervals, but not June 56 Fig. 2.3 Mean cloacal widths (±1SE) within high- and low- elevation male dark-eyed juncos, adjusted by life-history stage in 2000, 2001, 2004 and 2005. A n asterisk (*) indicates a significant difference between groups 57 Fig. 2.4 Mol t scores calculated from the right wing of high and low elevation males in 2004 (2 birds sampled from each of 8 sites within Jasper at each elevation at each v i i i sample interval). A n asterisk (*) indicates a significant difference between groups. 58 Fig. 2.5 Brood patch scores of female dark-eyed juncos from high- and low-elevation sites in Jasper National Park (including all years; n (high, low) = 30, 34). Brood patches were scored as: 1= N o brood patch formation, 2 = Partial feather loss/brood patch formation, 3 = fully developed brood patch, 4 = Brood patch that is re-feathering. The higher the mean brood patch score, the further along the average female is in her reproductive cycle 59 Fig. 2.6 Apparent monthly survival estimated from the top ranking model, {(p(elevation + t) /((elevation)}, where survival (9) is described as a function of breeding elevation and time, and recapture probability (p) is described as a function of breeding elevation only. This model had an A I C c weight of 0.999, and thus was strongly supported by these data. A t every time interval examined, survival estimates were significantly higher in high-elevation habitat 60 Fig. 3.1 Actual (observed) and predicted data (from best fit linear and non-linear regression models) showing the relationship of each weather variable (x) to the cumulative percentage of broods initiated (y) by dark-eyed juncos in low- (left column) and high- (right column) elevation habitats 74 Fig. 3.2 Graphical representation of the different relationship between cumulative percentage brood initiation ( C U M . % B R ) , snow depth, and growing degree day (GDD) , in high- and low-elevation birds. High-elevation birds all breed at a relatively low growing degree day values, and display a sharp peak in brood initiation when the snowpack melts to a value close to zero. Low-elevation birds, on the other hand do not experience snow cover during the breeding season, and display a strong non-linear correlation with growing degree day 75 ix Fig. 4 . 1 Schematic of four possible outcomes of how timing of reproductive development in dark-eyed juncos caught at different elevations may change when birds are brought into common aviary conditions and stimulated to breed by increasing daylength. Dotted and solid lines represent birds from high- and low-elevations, respectively, and text in italics denotes possible mechanisms (which may co-occur) for observed patterns 98 Fig. 4 . 2 Photoperiod prescribed to high- and low-elevation dark-eyed juncos (Junco hyemalis) in captivity in order to stimulate the temporal reproductive cycle, from 02 Jan to 20 Jun, 2005. Dark-eyed juncos were brought from short-days (8 hours of daylight) to long days (15 hours of daylight, mimicking summer solstice) in captivity over a period of 140 days (increase of 3.85 min /day). Birds were kept at the peak level o f 15 hours of daylight for an additional 40 days 99 Fig. 4 . 3 Changes in mean testis width (±1SE) over time in male dark-eyed juncos collected from high- and low-elevation habitats in Jasper National Park and subjected to identical increases in photoperiod in a common lab environment to stimulate reproductive onset 100 Fig. 4 . 4 Dates of onset of functional gonads in free-living (field) and experimental (laboratory) male birds collected as hatchlings and raised to breed for their first time in captivity ( ' S Y ' age) or as adult birds that had already bred for at least one season prior to collection and induction of breeding in the laboratory ( ' A S Y ' age). The absolute difference between onset and the order in which birds from high-and low-elevations became capable of breeding are shown. Field A S Y birds are only included for comparing the absolute differences in the days between populations becoming capable of breeding; actual initiation dates between the field (naturally stimulated) and the lab (artificially stimulated earlier in the season) are irrelevant since lab birds were artificially induced earlier in the season than either population breed naturally in the field 101 x Fig. 4.5 Changes in the mean number of songs (±1SE) per minute recorded from male dark-eyed juncos collected from high- and low- elevation habitat in Jasper National Park and subjected to an increasing photoperiod in a common lab environment to stimulate reproductive onset 102 Fig. 4.6 Average ovary scores for photoperiodically-induced female dark-eyed juncos collected at the end of the previous breeding season as 9-11 d hatchlings (SY) or adults ( A S Y ) from high- and low- elevation habitat 103 Fig. 4.7 Mean brood patch scores by date in lab-reared female juncos collected as 9-11 day hatchlings (SY at time of measurements; top) or as adults ( A S Y ; bottom) from high- and low-elevation habitats in Jasper National Park 104 Fig. 5.1 Schematic of four possible outcomes of how morphological traits of birds caught at different elevations may change when birds are brought into aviary under conditions of unlimited food supply. Solid and dotted lines represent birds from high- and low-elevations, respectively, and text in italics denotes possible mechanisms for observed patterns 126 Fig. 5.2 Mean values (± 1 SE) associated with wing and tail lengths of male and female dark-eyed juncos collected from low- and high-elevation sites in Jasper National Park, Canada. S Y refers to second-year, and A S Y refers to after second-year age classes 127 Fig. 5.3 Mean values (± 1 SE) associated with size and energy storage of male and female dark-eyed juncos collected from high- and low-elevation sites in Jasper National Park, Canada. S Y refers to second-year, and A S Y refers to after second-year age classes 128 xi LIST O F SYMBOLS A I C = Akaike's Information Criterion A S Y = after second year aged bird S Y = second year aged bird Eta = partial eta squared value (measure of effect size) R 2 coeff icient of variation cp (phi) = estimate of survival p (recapture probability) = probability of recapture of banded bird X 2 = Chi-squared statistic A C N K O W L E D G E M E N T S A PhD is not possible without the assistance of many valuable contributors along the way. First, I thank Dr. James N . M . Smith and Dr. Kathy Martin who were supervisors for this project, dealing with all the details along the way. Field work was made productive, relatively safe, and often hilarious by Graeme Brown, Anna Bendzsak, Melissa Bandura, Sarah Lord, and Karolyn Keir . Committee members Dr. Darren Irwin, Dr. Patricia Schulte, and Dr. Peter Arcese provided valuable suggestions and were always approachable. Members of the Martin lab provided intellectual feedback at many stages of my thesis progression. Lab technician Marty Mossop provided logistical support and a consistent, patient disposition that was much appreciated. Dr. Mark C. Drever provided a great deal of editorial and analytical aid. Dr. Barry Smith aided me with learning program M A R K , and Dr. Gary White aided with M A R K models. Personal funding was provided by N S E R C P G S - B , and the Cordula and Gunter Paetzold Graduate Fellowship. Research funding was provided by N S E R C funds to Dr. James N . M . Smith, N S E R C and Environment Canada funds to Dr. K . Martin, and an Alberta Conservation Association grant and Mountain Equipment Coop equipment loan to H . Bears. Many people outside academic venues also deserve thanks. Thanks to my friends and my boyfriend, Dave Campbell. Y o u are the most phenomenal people that I know. Thanks for not trading your inherent kindness to get by in a world where, too often, it is discouraged. Y o u have provided me with the emotional support I very much needed to get through those days where I was hit by repeated blows of extreme bad luck (you know the ones). Thanks to my family who was there from the start, to my Buddhist Society, Scopes Monkey Trial (the band) for many relieving jam sessions while writing, and to Mrs . Sandy Margetts for sparking my interest in biology. Finally, I thank Dr. James N . M . Smith for taking a chance on me and teaching me to believe in my own ideas. DEDICATION Tor Jamie This thesis is dedicated to Jamie Smith. If he had been here today, the writing in these pages would have been far more succinct. I hope he would have liked it anyhow. xiv CO-AUTHORSHIP S T A T E M E N T Analysis within chapters 1 and 4 of this thesis contain analytical contributions by Dr. Gary White and Dr. Mark C. Drever; as such they wi l l appear as co-authors for the papers to which they contributed. Specifically, the M A R K © survival analysis presented in chapter 2 was modified by Dr. Gary White (Colorado State University, Ft. Collins, Co.), designer of the program. Dr. Mark C. Drever conducted the mixed model statistical analysis for chapter 4, and wrote the methods section for the statistical methods used for this chapter. A l l other components of this thesis, including identification and design of the research program, the methods used, data collection, data analysis, and manuscript writing and presentation were done independently with editorial assistance of my supervisor and committee members. xv 1 I N T R O D U C T I O N 1.1 Phenotypic Var ia t ion One of the main goals of evolutionary ecologists is to identify patterns of phenotypic variation in animals over landscape-level gradients, and then to identify mechanisms that explain those patterns (Bergmann 1847; Al len 1877; Lack 1968; Zammuto and Mi l l a r 1985a,b; Perfito etal. 2004). The study of phenotypic variation along environmental clines provides insight into how traits respond to various environmental factors, and thus suggests traits that may be adaptive within certain contexts. Heritable trait variation within a species is required for natural selection, while phenotypic plasticity provides individuals with flexibility to adjust their phenotypes to environmental variation experienced within their lifetimes (Darwin 1859; Endler 1977; V i a and Lande 1985; Stearns 1989). The propensity for plasticity itself can also be heritable, and is subject to natural selection (Pigliucci 1998). Both forms of variation can lead to differentiation of phenotypes within a species, and may ultimately create reproductive barriers between populations that lead to species divergence and diversification (Irwin et al. 2001; Pfening et al. 2006). For instance, i f a correlation exists between a song type and a genetically encoded trait that is adaptive within a particular habitat, females may chose males with habitat-appropriate song types, leading to further divergence of populations (Wells and Henry 1992; Sedden 2005; Grant and Grant 1996; Slabekoorn and Smith 2002). If a phenotypically plastic trait such reproductive timing varies among habitats, populations may become temporally isolated from breeding with one another such that genetic drift or selection creates reproductive incompatibilities between populations even i f breeding synchrony is re-established (Quinn et al. 2000; Moore et al. 2005). Therefore, the study of intraspecific variation not only serves to produce hypotheses about traits that are likely adaptive in various habitats, it can also provide a glimpse into mechanisms that may precede and facilitate the speciation process (Foster et al. 1998). 1 One approach that evolutionary ecologists have used to determine possible causes of underlying phenotypic differences among habitats is to establish correlations between phenotypic variation and individual abiotic or biotic factors such as weather variables, resource availability, predation risk, competition, and breeding season duration (Krementz and Handford 1984, Martin 1995, Badyaev 1997, Conway and Martin 2000). A second approach has been to focus on landscape features or environmental clines that promote phenotypic or genetic variation in species in predictable ways (Lack 1968; Silverin et al. 1993; Galarowicz and Wahl 2003; Perfito et al. 2004). Well-known examples of the latter approach can be seen in Bergmann's and Al len 's rules, where species that live and breed further north tend to have smaller surface area to volume ratios and smaller and fewer protruding body parts than those breeding further south (Bergmann 1847, translated in James 1970; Al len 1877). While the latter approach does not tease apart influences that individual abiotic and biotic variables have on phenotypic variation, it has proven extremely useful for making predictions in natural environments, where abiotic and biotic variables shift concomitantly in predictable ways along certain clines. Knowing how phenotypic variation correlates with landscape features can be extremely useful in predicting how phenotypes w i l l be altered across a species' range, and how adaptable species w i l l be to changing environmental and habitat conditions. Comparative studies of organisms living along environmental gradients or between the extremes of their breeding range have proved useful for establishing how phenotypes respond to landscape level features and environmental clines (Allen 1877; Silverin et al. 1993; Moore et al. 2002, 2005). Some studies have compared multiple species and looked for shared characters among species residing in similar habitats in order to infer traits that are adaptive for particular habitat conditions (Badyaev 1997; Landmann and Winding 1995a,b). However, interspecific comparisons require accurate knowledge of phylogenetic relationships for the species examined, as differences among species could reflect disparate evolutionary histories of the lineages 2 examined, while similarities could reflect phylogenetic inertia, which is the persistence of benign traits from a shared ancestor to its descendants (Fjeldsa 1992). One solution to confounding influences of phylogeny is to use comparative methods that map traits onto phylogenetic data, which entails sampling across a broad array of species. However, because a large number of species must be sampled for this method to be effective, it restricts analysis to life-history traits that are easily measured, such as clutch size or egg mass. Other demographic rates are difficult to compare among studies using multiple species because estimates are sensitive to variation in capture methodology, spatial scale of study plots, and analytical techniques used. Therefore, intraspecific comparisons are still a preferred method for gaining insight into how environmental clines affect integrated phenotypes of populations that presumably have a similar evolutionary history (Robinson et al. 1996; Perfito et al. 2004). 1.2 Phenotypic Variation with Elevation Elevation gradients are important landscape features that influence phenotypic variation in plants and animals around the world, as nearly 40% of the terrestrial planet is mountainous ( U N E P 2002). Therefore, knowing how phenotypes are influenced by elevation as well as latitude would give ecologists considerable predictive power at the landscape level about how populations may change across their range. With increasing elevation, several physical variables change concomitantly over a short linear distance: precipitation, wind, snow cover, and solar radiation increase while temperature, barometric pressure, and growing season length decrease. High-elevation habitats experience regular hailstorms, wind, snow deposition, and freezing temperatures, often extending through the summer months (Martin 2001; Bears et al. 2003; Table 1.1). Biotic communities also change considerably with elevation, with high-elevation habitats having lower plant productivity, and generally hosting a lower faunal diversity (Kikkawa and Williams 1971; Martin 2001). Other costs and benefits (e.g., relative predation risk, interspecific competition, parasite exposure, etc.) may also shift considerably with elevation (Badyaev 1997; Martin 2001; Bears 2004). Possibly as a result, areas with steep environmental gradients and/or high topographic complexity are thought to be evolutionary hotspots (Lei et al. 2003; Roberts et al. 2007). Relatively few intraspecific, well-controlled study systems have been established that allow for the examination of the effects of breeding elevation on the integrated phenotypes of avifauna; yet birds are the most diverse vertebrate group in mountain areas, with over 70 species using high- elevation sites in western Canada (Wilson and Martin 2005). Birds cope with breeding at different elevations in a number of ways (Martin and Wiebe 2004). Many species of birds are elevational specialists, residing within relatively narrow elevation bands encompassing conditions within their range of tolerance (Scheid and Hashim 1997; Ghalambor et al. 2006). Other species are altitudinal migrants, utilizing different elevations throughout the breeding season in order to avoid or minimize exposure to physiologically challenging conditions (Hahn et al. 2004). Some other species are driven by intraspecific competition, with inferior competitors occupying less productive, high-elevation habitats (Pearson and Rohwer 1998; Rohwer 2004; Martin and Wiebe 2004). Finally, some species occupy wide elevation gradients with individuals displaying philopatry to specific elevations, allowing them to experience different costs and benefits for their lifetime, depending on breeding elevation (Bears et al. 2003; Perfito et al. 2004). Comparative studies of mammals (Dunmire 1960; Zammuto and Mi l l a r 1985a,b), and reptiles (Grant and Dunham 1990; Mathies and Andrews 1995) have utilized such elevationally widespread species in order to elucidate how phenotypic traits can differ among elevations in such systems. However, only a handful of studies on birds have compared single, species among elevations. Examples of elevationally expansive bird species include the dark-eyed junco (Junco hyemalis), chipping sparrow (Spizellapasserina), white-crowned sparrow (Zonotrichia leucophrys), and American robin (Turdus americanus). Most comparative studies of birds among elevations to date have been interspecific, and thus confounded by phylogeny and limited to few easily measured traits (Bjorklund 1994; Landmann and Winding 1995a; Badyaev 1997; Pearson and Rohwer 1998; Blackburn and Ruggiero 2001; Sandercock et al. 2005a,b), or have been intraspecific but confounded by latitude or complicated by the fact that each elevation compared was nested within different eco-regions (Widmer 1999; Perfito et al. 2004), or have examined males only (Perfito et al. 2004). Thus, no well-controlled, intraspecific, comparative studies have been conducted to explore how elevation affects the integrated avian phenotype, similar to those that have been conducted in botanical systems (e.g. Angert and Schemske 2005). Despite existing studies not being well-controlled or focusing on very few traits with breeding elevation, they are useful for predicting the direction that phenotypes may be expected to change with breeding elevation and are worth reviewing. Landmann and Winding (1995a,b) found that songbirds above the treeline in Nepal were larger than their low-elevation counterparts. A positive relationship between body weight and elevation was also present in endemic Andean birds (Blackburn and Ruggiero 2001). On the other hand, Widmer (1999) found no relationship between body weight or tarsus length with breeding elevation in the garden warbler (Sylvia borin), over a 1000 m elevation range in central Europe. In the Himalayas, high-elevation songbirds have longer, pointed wings and square-ended or shallow-forked tails, which were interpreted as adaptive features for flight stability in strong winds (Landmann and Winding 1993, 1995a, b). Beak size (Price 1991) and intraspecific variation in plumage (Graves 1985) can also vary with elevation. High-elevation birds may also exhibit behavioural differences, such as placing nests within more benign microclimates (Verbeek 1967; Medin 1987; Bohn and Landmann 1995; Landmann and Winding 1995b; Martin 2001; Bears 2002), and they may adapt physiologically or behaviorally to environmental conditions (Clemens 1988; Carey 1980; Scheid and Hashim 1997; Bears et al. 2003). 5 Birds at higher elevations tend to restrict their breeding activities to shorter periods due to a limited period of favorable conditions for breeding in high-elevation sites, which results in the production of fewer offspring per season (Krementz and Handford 1984; Badyaev 1997). Very little is understood about the mechanisms controlling different reproductive schedules in conspecifics breeding among elevations. Normally, the annual change in daylength in the spring is used by birds as the main proximate cue to initiate the physiological, morphological, and behavioral changes that must occur in advance of breeding (Famer and Lewis 1971; Dawson et al. 2001). In birds that respond to photoperiod, the daylength increases in the spring until a 'critical daylength' ( C D L ) is reached (Hamner 1966). However, birds experience identical photoperiods between elevations at the same latitude, yet they may have very different breeding schedules. Hence, birds either use supplementary cues such as weather variables in order to adjust their breeding schedules among elevations, or they respond to different critical day lengths among elevations, as they do among latitudes (Silverin et al. 1993). Differentiating between these hypotheses is important, as the former implies that birds w i l l be able to adjust their breeding schedules more rapidly in response to temporal aspects of climate changes such as earlier spring warming, while the latter hypothesis suggests that birds may be relatively fixed as to when they can breed. A single species of bird may breed across a wide elevation gradient for a number of reasons. First, less competitive, subordinate individuals may be forced to breed within the less desirable/less productive elevations due to intraspecific competitive exclusion from the more desirable elevations. In such a situation, a disproportionately higher number of less competitive classes (e.g., younger, smaller) would be expected to breed at the less desirable elevation. To date, there has been little support for this hypothesis in songbirds (Kollinsky and Landmann 1996; Widmer 1999, although see Rohwer 2004). Alternatively, a wide elevational range could be maintained by species experiencing different costs and benefits at various locations across 6 their elevation range, and individuals may adhere to different lifetime reproductive strategies in order to maximize the benefits of their local habitats. For instance, birds generally have fewer offspring per season at higher elevations (Krementz and Handford 1984; Badyaev 1997), yet may be compensated by alternate advantages, such as a lower exposure to parasites and diseases (Stabler et al. 1974; Braun et al. 1993), the production of offspring that survive better (Leary et al. 1999; Oddie 2000), or increased survival of adults or fledglings (Sandercock et al. 2005 a). 1.3 Thesis Objectives The current review o f existing literature suggests that 4 dominant questions are at the forefront of comparative research on birds breeding among elevations (and analogous clines) that need to be addressed using appropriate, controlled study systems. The objectives of this thesis are to address the following general questions using a model songbird species breeding across a steep elevation gradient in a single ecoregion within Jasper National Park, A B , Canada: (1) How do life-history strategies vary with breeding elevation? (2) What cues correlate with reproductive timing between elevations? (3) Are reproductive differences observed with elevation primarily the result i of local genetic adaptation or phenotypic plasticity to local conditions?, and (4) Does morphology vary with elevation, and are morphological differences observed primarily the result of local genetic adaptation or phenotypic plasticity to local conditions? B y examining reproductive and morphological traits both in situ and in the context of a common garden experiment, I hoped to gain better insight into mechanisms that shape traits, potential trade-offs, and constraints that influence trait expression with elevation. 1.4 Study System I performed field investigations combined with an experimental rearing experiment to compare various interrelated phenotypic traits in a sparrow, the dark-eyed junco (Junco hyemalis; Nolan et al. 2002), breeding within high- (2000 m asl) and low- (1000 m asl) elevation habitats in the Canadian Rocky Mountains. The dark-eyed junco (Fig 1.1) is an emberizid commonly used in 7 studies of life-history and morphological variation along environmental gradients (Miles and Ricklefs 1984; Nolan and Ketterson 1990; Chandler and Mulv ih i l l 1990). The dark-eyed junco also has one of the widest elevational distributions in North America, and is able to breed from sea level to the subalpine-alpine treeline, making it an excellent candidate for exploring phenotypic variation in avifauna with elevation (Nolan et al. 2002). Jasper National Park was chosen as a study location because the habitat is unfragmented and juncos breed continuously along the elevational range in that area (Bears 2002); therefore allopatric genetic differention due to habitat fragmentation preventing gene flow is unlikely. Further, the abiotic cline at this latitude and elevation is particularly steep (Bears et al. 2003). The weather variables that changed along the specific elevation gradient used for my thesis were also shown to be comparable to the changes occurring along similar elevational gradients in other mountain areas in the interior of western Canada (Bears et al. 2003). In addition, I established this study system during my M S c thesis and thus baseline data existed prior to the initiation of my PhD research that was useful for establishing hypotheses (Bears 2002; Bears et al. 2003; Bears 2004). Within Jasper National Park, the dark-eyed junco breeds over a large portion of its potential elevation range, from the montane valley (1, 000 m asl) to the subalpine treeline (1,900-2,100 m asl). Juncos that breed in Jasper National Park likely migrate to the north-western United States or to south-western Canada to spend the winter months, but precise migration routes are unknown (Nolan et al. 2002). Study sites were located at the lowest elevations that juncos breed within the park in the montane valley (1000 - 1020m a.s.l.; 7 sites, separated by >10 km) and at the highest elevation that juncos breed in the park, at the sub-alpine/alpine treeline (1950-2100m a.s.l.; 8 sites, separated by >18km; Fig. 1.2). I focused on dark-eyed juncos breeding within the extremes of this elevation gradient only, as the elevational extremes were the most strongly contrasting and thus likely to favor different phenotypes. A l l high-8 elevation sites had slight southeast facing aspects; site characteristics are described elsewhere (Bears 2002; Bears et al. 2003; Bears 2004). Abiotic weather variables differed considerably between the elevations in my study area, and I include a general summary of weather data for my study system that is relevant to my whole thesis in Table 1.1. Abiotic differences observed between elevations in my study area were typical of those seen between similar elevations in other localities in interior western Canada (Bears et al. 2003). Compared to low-elevation sites, high-elevation sites in the summer had mean temperatures that were 4-5 °C colder, snowmelt was delayed by up to 3 months, and more precipitation received as snow and hail during April-August (Table 1.1). Compared to low-elevation study sites, high-elevation study sites had a shorter growing season (pers. obs.) and insects, a major food resource of dark-eyed juncos, emerged later in the summer (Table 1.1). Photoperiod was the same between elevations, as discussed in chapter 4 (Appendix II). 1.5 Thesis Overview In chapter 2,1 demonstrated some dramatic temporal shifts in the organization of life-history events in dark-eyed juncos breeding between elevations, and showed that these shifts are coincident with differences in the timing of the seasonal maturation of reproductive features between populations at different elevations. I tested the hypothesis that less competitive individuals are forced to breed within the less productive extreme of the elevation range against the alternative hypothesis that high- and low-elevation populations are locally adapted and that a shift in the life-history strategies of populations, from a high-reproduction to a high-survival life-history strategy, occurs with increasing breeding elevation. I found little evidence for the competitive exclusion hypothesis, and instead observed differences in life-history strategies between elevations, with a higher estimated survival, shorter reproductive season, and lower seasonal reproductive success in high- compared to low-elevation birds. Causes of the differences in reproductive phenology and effort were explored further in subsequent chapters 9 In chapter 3,1 examined the relationship between weather variables measured at each elevation, and the mean cumulative percentage of broods initiated in high- and low-elevation birds per season, which is a measure of population investment into reproduction. I examined ways in which weather variables were correlated with brood initiation, and determined that highl-and low-elevation populations initiated broods in relation to different weather conditions. However, weather variables may not have been causally linked to brood initiation, as birds may have also adapted to respond to different critical day lengths among elevations. I f populations did respond to different critical daylengths, correlations between photoperiod and weather variables would in chapter 3 would be correlative rather than causal. Therefore, in chapter 4 I determined whether the different reproductive schedules between high-and low-elevation birds could be due to differences in their responses to increasing photoperiod. In chapter 4 , 1 tested the hypothesis that differences in the temporal organization of reproductive events seen in chapters 2 and 3 were due to high- and low- elevation populations being locally adapted to respond to different critical daylengths to initiate reproductive onset in the spring, rather than having plasticity in responses to differences in weather between elevations. In order to differentiate between these two hypotheses, I raised dark-eyed juncos collected from high- and low-elevation study sites in a common aviary environment where they experienced identical supplementary cues. I increased the photoperiod artificially to simulate the advance of spring, and determined i f the differences in reproductive timing remained (due to different reactions to increasing daylength alone) or were eliminated ( i f supplementary cues that differ between elevations were causing differences between high- and low- field populations). Since the differences observed in the field were greatly reduced in the laboratory, I concluded that dark-eyed juncos were adjusting their breeding schedules in response to supplementary cues. However, since elevational differences remained, I suggested that local genetic adaptation or perinatal/maternal effects may also contribute to the variation. 10 In chapter 5,1 explored morphological differences between juncos breeding in high- and low-elevation habitats, and whether differences in morphological traits between populations were the result of local genetic adaptation or phenotypic plasticity. In order to differentiate between these two possibilities, I raised dark-eyed juncos collected from high- and low-elevation sites in a common aviary where they experienced identical supplementary cues for a full molt cycle. I measured traits in lab-raised birds that differed between elevations in the field. I tested the hypothesis that traits that differed in the field would be more similar in lab-raised birds since I expected that differences in wi ld birds would be due to phenotypic plasticity in response to differing environments. I tested this hypothesis against the alternative hypothesis that morphological differences would remain i f they were caused by genetic (or persistent perinatal/maternal effects) that were elevation-specific. I found that differences observed in the field remained or were exacerbated in common laboratory conditions, suggesting that local genetic adaptation and persistent perinatal/maternal effects are more important than phenotypic plasticity in creating morphological differences that exist between elevations. Finally, in chapter 6,1 reviewed my main findings, and discussed how they contribute to our overall understanding of shifts in phenotypes with elevation, and the broader theoretical implications of my results. Investigating how species adapt to shifting conditions with elevation, and the forces that underlie such variation, is fundamental for understanding and conserving mountain species. In my thesis, I enhanced our general understanding of phenotypic variation of a single bird species with breeding elevation. A s we understand better how species adapt naturally to elevation gradients, we w i l l be able to better predict how they w i l l react to future climate and habitat change (Luckman 1998; Storch 2000; Martin 2001). Finally, I determined whether and how uninterrupted elevation gradients promote reproductive and morphological variation in avifauna, as clines that promote high intraspecific variation (especially heritable) over a short linear distance are of conservation value. A s it is impossible to study all species 11 spanning wide elevation gradients, I used this model system to demonstrate shifts that can occur with elevation as a first step in establishing empirical evidence for the building of a theoretical framework. Work on other species in the future w i l l establish how far-reaching and general the patterns and mechanisms uncovered here are. 12 Fig. 1.1 The dark-eyed junco (Junco hyemalis), a songbird in the sparrow family emberizidae, identified by a dark hood that contrasts against a lighter body (darker in males than in females), and by white tail feathers that appear as white stripes when the tail feathers are fanned in courtship or territorial display. 13 Fig. 1.2 Study site locations in Jasper National Park, Canada. Triangles are high-elevation sites and diamonds are low-elevation sites; those shaded black are field study sites, and those shaded white are field and collection sites. 14 Table 1.1 Weather variables and insect biomass^, summarized by year and month at high- and low-elevation sites in Jasper, A B . Weather Mean Temp Minimum Temp Maximum Temp Rainfall Variable (°C) [range] (°C) [range] (°C) [range] (mm) [range] Elevation 2000 April High Low High Low High Low High Low -2.2[-l5.5-4.5] 3.0[-9.6-9.4] -7.9[-19.4-1.3] -4.4[-16.7-3.0] 3.4[-l1.9-12.3] 10.3[-6.7-21.3] 0(0-0] 0.3(0-2.8] May 3.4[12.8-12.4] 6.8[3.6-11.9] -1.3[-7.3-7.4] 0.2[-4.8-7.0] 8.1[1.7-17.4] 13.5[6.8-18.7] 0.6(0.4.7] 0.8(0-3.6] June 6.4[1.5-11.3] 11.5[7.0-15.3] 2.0[-1.6-5.9] 4.7[-l.4-9.0] 10.8[3.5-16.6] 18.8[10.4-25.6] 2.6(0-26.6] 2.2(0-16.6] July 14.6[6.7-21.5] 15.0[10.8-18.1] 5.0[1.5-9.4] 7.1 [1.8-11.3] 14.6[6.7-21.5] 22.9[13.9-30.4] 2.1(0-15.7] 1.1(0.8.2] Aug •.Average 8.6[2.8-14.8] 13.6[9.3-22.9] 3.6[-0.9-8.8] 5.3[-0.3-ll.l] 13.6[5.5-21.1] 21.3[13.2-30.2] 1.6(0-11.1] 0.9(0-8.2] 2001 April -1.9[-7.1-6.5] 3.4[-l.1-11.6] -6.9[-13.6-2.0] -3.1[-11.5-5.1] 3.l[-3.1-12.8] 10.0[2.3-20.8] 0.3(0-6.1] 4.2(0-5.4] May 3.4[-2.8-12.4] 8.7[2.4-15.6] -1.4[-7.3-7.4] 1.2[-6.8-7.1] 8.1[1.7-17.4] 16.1 [8.9-26.7] 0.6(0-4.7] 0.5(0-4.6] June 6.3[2-13.1] 12[8.4-18.3] 1.7[-2.0-8.4] 5.1[-0.30-10.7] 10.9[4.5-17.8] 18.8(13.4-26.3] 2.7(0-15] 2.2(0-10] July 9.6[3.7-16.7] 14.9[10.3-19.1] S.l[0.4-ll.l] 7.8[3.4-l1.7] 13.9[6.9-22.5] 21.9[13.8-21.7] 5.5(0-26.9] 3.5(0-27.2] Aug p . A v e r a g e i « 10.8[5.5-17.7] 14.9[11.1-19.3] 5.7[1.9-11.2] 5.6[1-11] mM4wsmi 16.0[8.4-24.2] 24.2[16.7-32.1] 1.3(0-17.3] M M 0.3(0-6.2] m m m 2004 April 0.7[-8.4-7.2] 6.3[-3.0-15.3] -4.9[-13.9-1.7] -2.2[-9.0-6.5] 6.3[-4.8-13.9] 14.4[-1.0-24.0] -0.4[-5.8-3.8] 0(0-0] May 2.0[-7.7-8.7] 7.2[-1.8-13.8] -2.4[-10.8-3.8] 0.4[-9.0-7.5] 6.4[-5.9-15.8] 14.0[1.0-25.5] 0(0-0] 2.4(0-12.5] June 8.7[2.1-14.3] 13.7[8.5-18.5] 3.5[-1.2-9.3] 5.8[-1.0-10.5] 13.9[4.9-21.5] 21.6[ 11-28.5] 0(0-0] 2.7(0-17.5] July 11.4[2.4-17.7] 16.6[8.3-22.0] 6.4[0.4-l1.8] 9.2[3.0-15.0] 16.3[3.7-24.5] 23.9[9.5-32.5] 3.3(0-33.2] 2.6(0-13.4] Aug i Average ' 10.5[1.9-18.8] 6 7N 14.8[5.8-21.3] 11 7 , ....  .. . .  5.9[0.2-12.2] 1 7 ~ 7.5[-0.5-15.5] 41 15.1[3.5-25.4] 116 .. .. , . ....... , . i 22[7-33] 19 2" . * . .. i .......... J . 3.1(0-15.8] 12 i., .  .. ,t J. i J . . . . 2.3(0-12.8] 2005 April -0.5[-7.8-9.3] 4.3[-l.6-11.1] -6.5[-12.9-2.4] -4.2[-10.0-3.6] 5.2[-2.6-16.2] 12.8[4.4-24.3] 0(0-0] 0.3(0-2.4] May 5.5[-2.0-14] 10.2[4.3-16.7] 0.4[-8.4-6.9] 1.3[-5.2-8.6] 10.5[4.4-21.7] 18.9(11.6-29.8] 1.1(0-9] 0.8(0-7.6] June 6.8[3.3-13.1] 12.5[9.1-17.1] 2.8[-0.7-6.4] 6.2[-1.9-9.8] 10.8[5.5-19.7] 18.8(12.4-28.2] 3.2(0-20.1] 2.2(0-12.2] July 8.7[3.8-15.3] 13.8[9.8-19.0] 4.3[0.7-11.4] 6.7[0.5-12.8] 13.0[6.8-20.0] 20.9(16.5-27.7] 2.5(0-12.5] 1.8(0-10.6] Aug 8.8[2.8-15.4] 12.8[8.1-17.9] 4.0[5.1-21.7] 4.4[-2.6-10.10] 13.6[5.1-21.7] 21.0(11.4-28.1] 3.0(0-24] 2.1(0-14.4] J Temperature and precipitation data are summarized as daily means taken from hourly measurements at weather stations within 5-10 km of our study sites at equivalent elevations, and from Hobo Pro © data loggers (temperature only) at each study site (3 per site). 15 Table 1.1 (con't) Weather variables and insect biomass¥ summarized by year and month at high- and low- elevation sites in Jasper, A B . Weather Mean Snow Depth (cm) Mean Snowfall (cm) Mean Insect Variable [range] [range] Biomass |± 1SE] (mg) Elevation High Low High Low High Low 2000 April 107.9[94-126] 0.4[0-3] 1.9[0-2.1] 0[0-0] May 78.4[55.0-98.0] 0[0-0] 0.8[0-9.2] 0.04[0-0.8] June 9.3[0-54] 0[0-0] 0[0-0] 0[0-0] July 0[0-0] 0[0-0] 0[0-0] 0[0-0] Aug 0[0-0] 0[0-0] 0.4[0-10] 0[0-0] April 101.9[87-111] 0.07[0-2] 2.2[0-6.2] 0.2[0-2.5] 10.8±0.8 18.0±2.9 May 69.2[19-92] 0.6[0-19] 0.7[0-9.2] 0.1[0-2] 20.1±1.4 30.8±0.8 June 2.3[0-ll] 0[0-0] 0.9[0-10] 0[0-0] 31.5±0.8 23.3±2.9 July 0.6[0-19] 0[0-0] 0[0-0] 0[0-0] 28.6±1.0 20.4±1.0 Aug 0[0-0] 0[0-0] 0[0-0] 0[0-0] 19.3±1.4 13.6±1.0 Average" " 34 8 01 08 ~ *01 22 1 ^ ^ ' 1 \ , 2 1 2 ""I"? k j 2004 April 64.1 [0-82] 0[0-0] 1.1[0-13] 0[0-0] May 36.6[1-61] 0.7[0-8.0] 0.6[0-7] 0.8[0-7.5] June 0[0-0] 0[0-0] 0[0-0] 0[0-0] July 0[0-0] 0[0-0] 0[0-0] 0[0-0] Aug k 0[0-0] 0[0-0] 0[0-0] 0[0-0] April May June July Aug .Average y.-.y 102.9[69-131] 26.6[0-69] 0[0-0] 0[0-0] 0[0-0] 0[0-0] 0[0-0] n/a n/a 0[0-0] 0[0-0] n/a 0[0-0] 0[0-0] 0[0-0] 0[0-0] 0[0-0] 0[0-0] 0[0-0] 0[0-0] ¥ In 2004, a Tanglefoot© insect trap (Tanglefoot Company, Grand Rapids, Ml) was placed at each high- and each low- elevation site and replaced bi-weekly. Insects were cleaned, dried, and weighed to measure temporal changes in mean insect biomass (mg), as insects are the main food source for hatchlings. Zeros mean that measurements were taken but no snow was recorded. N/A means that no measurements were taken. 16 1.6 R E F E R E N C E S Al len , J .A. 1877. The influence of physical conditions in the genesis of species. Radical Review 1: 108-140 Angert, A . L . , and Schemske, D . W . 2005. The evolution of a species' distributions: Reciprocal transplants across the elevation ranges of Mimulus cardinalis and M. Lewisii. Evolution 59 1671-1684. Badyaev, A . V . 1997. Avian life history variation along altitudinal gradients: A n example with cardueline finches. - Oecologia 111: 365-374. Bears, H . 2002. How and why a small songbird, Junco hyemalis, breeds over a steep elevation gradient. M S c . thesis, University of British Columbia, Vanvouver. Bears, H . , Smith, J . N . M . , and Wingfield, J .C. 2003. Adrenocortical sensitivity to stress in Dark-eyed Juncos [Junco hyemalis oregonus) breeding in low and high elevation habitat. - Ecoscience 10: 127-133. Bears, H . 2004. Parasite prevalence in Dark-eyed Juncos, Junco hyemalis, breeding at different elevations. - Can. Field Nat. 118: 235- 238. Bergmann, C. 1847. Ueber die Verhaltnisse der Warmeokonomie der Thiere zu ihrer Grosse. -- Gottinger Studien 3: 595-708. Bjorklund, M . 1994. Allometric relationships in three species of finches (Aves Fringillidae). -J. Zool. 233: 657-668. Blackburn, T . M . , and Ruggiero, A . 2001. Latitude, elevation and body weight variation in Andean passerine birds. - Global Ecol Biogeogr. 10: 245-259. Bohn C , and Landmann, A . 1995. Nest-site selection and nest construction in the Water pipit (Anthus spinolettd). - J . Fiir Ornithol 136:1-16. Braun, C .E . , Martin, K . , and Robb, L . A . 1993. White-tailed ptarmigan (Lagopus leucurus) In The birds of North America, No. 68. The Academy of Natural Sciences; Washington, D . C , U S A (Poole A . , and G i l l F. , Eds.). Pp 1-24. Carey, C. 1980. Adaptation of the avian egg to high elevation. - Am. Zool. 20: 449-459. Chandler, R . C . and Mulv ih i l l , R.S. 1990. Wing-shape variation and differential timing of migration in Dark-Eyed Juncos. - Condor 92: 54-61. Clemens, D.T. 1988. Ventilation and oxygen consumption in rosy finches and house finches at sea level and high elevation. - J. Comp. Physiol. 158: 57-66. Conway, C. J., and Martin, T .E. 2000. Evolution of passerine incubation behavior: influence of food, temperature, and nest predation. - Evolution 54: 670-685. Darwin, C R . 1859. The Origin of Species. V o l . X I . The Harvard Classics. New York: P.F. Collier & Son, 1909-14; Bartleby.com, 1859 (republished in 2001). www.bartleby.com/11/. Dawson, A . 2003. A comparison of the annual cycles in testicular size and moult in captive European starlings Sturnus vidgaris during their first and second years. - J. Avian Biol. 34: 119-123. Dunmire, W . W . 1960. A n altitudinal survey of reproduction in Peromyscus maniculatus. -Ecology 41: 174-182. Endler, J .A. 1977. Geographic variation, speciation and clines. Princeton University Press; Princeton, New Jersey, U S A . Farner, D.S. , and Lewis, R . A . 1971. Photoperiodism and reproductive cycles in birds. -Photophysiol. 6: 325-370. Fjeldsa, J. 1992. Biogeographic patterns and evolution of the avifauna of relict high-elevation woodlands of the Andes. - Steenstrupia 18: 9-62. Foster, S.A., Scott, R.J . , and Cresko, W . A . 1998. Nested biological variation and speciation. -Proc. Roy. Soc. B 353: 207-218. Galarowicz, T .L . , and Wahl , D . H . 2003. Differences in growth, consumption, and metabolism among walleyes from different latitudes. - TransAm. Fish. Soc. 132: 425-437. Ghalambor, C . K . , Huey, R . B . , Martin, P.R., Tewksbury, J.J., and Wang, G . 2006. Are mountain passes higher in the tropics? Janzen's hypothesis revisited.- Int. Comp. Biol. 46: 5-17 Grant, B . W . , and Dunham, A . E . 1990. Elevational covariation in environmental constraints and life-histories of the desert lizard Sceloporus merriami. - Ecology 71: 1765-1776. Grant, B .R. , and Grant, P.R. 1996. Cultural inheritance of song and its role in the evolution of Darwin's finches. Evolution 5 0 : 2471-2487. Grasso, M . J . , Savalli, U . M . , and Mumme, R. 1996. Status signalling in dark-eyed juncos: perceived status of other birds affects dominance interactions. - Condor 98:636-639. Graves, G.R. 1985. Elevational correlates of speciation and intraspecific geographic variation in plumage in Andean forest birds. - Auk 102: 556-576. Hahn, T.P., Pereyra, M . E . , Sharbaugh, S . M . and Bentley, G . E . 2004. Physiological responses to photoperiod in three cardueline finch species. - Gen. Comp Endocrin. 137: 99-108. Hamann, H.J . , Schmidt, K . H . , and Simonis, S. 1989. Elevation and size of eggs and clutch in the Great tit Parus major. - J. Fiir Ornithol. 130: 69-74. Hamner, W . H . 1966. Circadian control of photoperiodism in the House Finch demonstrated by interrupted-night experiments. - Nature 203: 1400-1401 Irwin, D .E . , Bensch, S. and Price, T .D. 2001. Speciation in a ring. - Nature 409: 333-337. James, F. C. 1970. Geographic size variation in birds and its relationship to climate. - Ecology 51: 365-390. Kikkawa, J., and Williams, W.T. 1971. Altitudinal distributions of land birds in New Guinea. - Search 2: 64-69. Koll insky, C , and Landmann, A . 1996. Altitudinal distribution of male black redstarts: are there age-dependent patterns? - Bird Study 43: 103-107. Krementz, D . G . , and Handford. P. 1984. Does avian clutch size increase with elevation? - Oikos 43: 256-259. Lack, D. 1968. Ecological adaptations for breeding in birds. Methuen, London. Landmann, A . , and Winding, N . 1993. Niche segregation in high-elevation Himalayan chats (Aves, Turdidae): does morphology match ecology.- Oecologia 95: 506-519. Landmann A . , and Winding N . 1995a. Gui ld organization and morphology of high-elevation granivorous and insectivorous birds: convergent evolution in an extreme environment.-Oikos 73: 237-250. Landmann, A . , and Winding, N . 1995b. Gui ld organization and morphology of high-elevation granivorous and insectivorous birds: convergent evolution in an extreme environment.-O/ATOJ 73: 237-250. Leary, J., Sullivan, K . A . , and Hillgarth, N . 1999. Relationships among dominance, foraging proficiency, and condition in juvenile dark-eyed juncos. - Auk 116: 1136-1141. L e i , F . M . , Qu, Y . H . , Tang, Q.Q., and A n , S.C. 2003. Priorities for the conservation of avian biodiversity in China based on the distribution patterns of endemic bird genera. - Biodiv. Cons. 12: 2487-2501. Luckman, B . H . 1998. Landscape and climate change in the central Canadian Rockies during the 20th Century. - Can. Geogr. 42: 319-336. Martin, K . , Holt, R.F. , and Thomas, D . W . 1993. Getting by on high: ecological energetics of Arctic and alpine grouse: Life in the cold III. In ecological, physiological, and molecular mechanisms. Westview Press; Boulder, Colorado, U S A (Carey, C , Florant, G . L . , Wunder, B . A . , and Horowitz, B . , Eds.). Pp 33-41. Martin, T. E . 1995. Avian life history evolution in relation to nest sites, nest predation, and food.- Ecolog. Monogr. 65: 101-127. Martin, K . 2001. Wildlife communities in alpine and subalpine habitats. In Wildlife-habitat relationships in Oregon and Washington. Oregon State University Press; Corvallis, U S A (Johnson, D . H . , and O ' N e i l , T .A . , Managing Directors). Pp 239-260. Mathies, T.E. , and Andrews, R . M . 1995. Thermal and reproductive biology of high and low elevation species populations of the lizard Sceloporus scalaris: Implications for the evolution of viviparity. - Oecologia 104:101-111. Medin, D . E . 1987. Breeding birds of an alpine habitat in the southern Snake Range, Nevada.-Western Birds 18: 163-168. Miles , D . B . and Ricklefs, R . E . 1984. The correlation between ecology and morphology in deciduous forest birds. - Ecology 65: 1629-1640. Moore, I.T., Perfito, N . , Wada, H . , Sperry, T.S., and Wingfield, J.C. 2002. Latitudinal variation in plasma testosterone levels in birds of the genus Zonotrichia. - Gen. Comp. Endocrinol. 129: 13-19. Moore, I.T., Bonier, F. , and Wingfield, J.C. 2005. Reproductive asynchrony and population divergence between two tropical bird populations. - Behav. Ecol. 16: 755-762. Nolan, V . and Ketterson, E .D . 1990. Timing of autumn migration and its relation to winter distribution in dark-eyed juncos. - Ecology 71: 1267-1278. Nolan, V . , Jr., Ketterson, E. D . , Cristol, D . A . . , Rogers, C M , Clotfelter, E . D. , Titus, R. C , Schoech, S.J. and Snajdr, E . . 2002. Dark-eyed Junco (Junco hyemalis). In The Birds of North America, no. 716. Academy of Natural Sciences, Philadelphia, and American Ornithologists' Union, Washington (Poole, A.and Gi l l s , F. , Eds). Oddie, K . 2000. Size matters: competition between male and female great tit offspring. J. Anim. Ecol. 69: 903-912. Pearson, S.F., and Rohwer, S. 1998. Influence of breeding phenology and clutch size on hybridization between hermit and Townsend's warblers. - Auk 115: 739-745. Perfito, N . , Tramontin, A . D , Meddle, S., Sharp, P, Af ik , D. , Gee, J., Ishii, S., Kikuchi , M . , and Wingfield, J.C. 2004. Reproductive development according to elevation in a seasonally breeding male songbird. - Oecologia 140:201-210. Pfennig, D .W. , Rice, A . M . and Martin, R . A . 2006. Ecological opportunity and phenotypic plasticity interact to promote character displacement and species coexistence. - Ecology 87: 769-779. Pigliucci, M . , and Schlichting, C D . 1998. A Reaction Norm Perspective. Sinauer Associates Inc, Sunderland, Massachusetts. Pp 387. Price, T. 1991. Morphology and ecology of breeding warblers along and altitudinal gradient in Kashmir India. J. Anim. Ecol. 60: 643-664. Quinn, T.P., Unwin, M . J . , and Kinnison, M . T . 2000. Evolution of temporal isolation in the wi ld : Genetic divergence in timing of migration and breeding by introduced chinook salmon populations. - Evolution 54: 1372-1385. Roberts, J .L. , Brown, J .L. , Schulte, R., Arizabal, W. , Summers, K . 2007. Rapid diversification of colouration among populations of a poison frog isolated on sky peninsulas in the central cordilleras of Peru. - J. Biogeogr. 34: 417-426. Robinson, B . W . , Wilson, D.S. , and Shea, G.O. 1996. Trade-offs of ecological specialization: A n intraspecific comparison of pumpkinseed sunfish phenotypes. - Ecology 11: 170-178. Rohwer, S. 2004. Using age ratios to infer survival and despotic breeding dispersal in hybridizing warblers. - Ecology 85: 423-431. Sandercock, B . K . , Martin, K . , and Hannon, S.J. 2005a. Life history strategies in extreme environments: comparative demography of arctic and alpine ptarmigan. - Ecology 86: 2176-2186. Sandercock, B . K . , Martin, K . , and Hannon, S.J. 2005b. Demographic consequences of age-structure in extreme environments: population models for arctic and alpine ptarmigan. - Oecologia 146: 13-24. Scheid, P., and Hashim, S. 1997. Birds over Mount Everest: extreme hypoxia tolerance. J. Physiol. 499.P, 9S Seddon, N . 2005. Ecological adaptation and species recognition drives vocal evolution in neotropical suboscine birds.- Evolution 5 9 : 200-215. Silverin, B . , Massa, R. , and Stokkan, K . A . 1993. Photoperiodic adaptation to breeding at different latitudes in great tits. - Gen. Comp. Endocrinol. 9 0 : 14-22. Slabbekoorn, H . , and Smith, T .B . 2002. Bi rd song, ecology and speciation. - Philos. T. Roy. Soc. 5 . 3 5 7 : 4 9 3 - 5 0 3 . Stabler, R . M . , Kitzmillar, N . J . , and Braun, C E . 1974. Hematozoa from Colorado birds, I V . Galliformes - J. Parasitol. 6 0 : 536-537. Stearns, S . C 1989. The evolutionary significance of phenotypic plasticity. - BioSci. 3 9 : 436-445. Storch, I. 2000. Grouse: General survey and conservation action plan 2000-2004. WPA/Bi rd l i fe /SSC Grouse Specialist Group and the World Pheasant Association. I U C N , Gland, Switzerland, and Cambridge, Reading, U K . U N E P World Conservation Monitoring Centre, Mountain Watch. 2002. Pages 14-15 in Environmental Change and Sustainable Development in Mountains. 80 pages, [http://www.unep-wcmc.org/]. Verbeek, N . A . M . 1967. Breeding biology and ecology of the horned lark in alpine tundra. -Wilson. Bull. 7 9 : 208-217. V i a , S., and Lande, R. 1985. Genotype-environment interaction and the evolution of phenotypic plasticity. - Bioscience 3 9 : 436-445. Wells, M . M . , and Henry, C S . 1992. The role of courtship songs in reproductive isolation among populations of green lacewings of the genus Chrysoperla (Neuroptera, Chrysopidae). -Evolution 4 6 : 31-42. Widmer, M . 1999. Altitudinal variation of migratory traits of the garden warbler Sylvia borin. PhD Thesis, University of Zurich, Zurich, Switzerland. Wilson S., and Martin K . 2005. Songbird use of high-elevation habitat during the fall post-breeding and migratory periods. - Ecoscience 12: 561-568. Zammuto, R . M . , and Mil la r , J.S. 1985a. Environmental predictability, variability, and Spermophilus columbianus life history over an elevational gradient. - Ecology 66: 1784-1789. Zammuto, R . M . , and Mil lar , J.S. 1985b. A consideration of bet-hedging in Spermophilus columbianus. - J. Mammal. 66: 652-660. 24 2 BREEDING ELEVATION AS A DETERMINANT OF THE LIFE-HISTORY STRATEGY OF A SONGBIRD (Junco hyemalis) 2.1 INTRODUCTION Species that breed along steep elevation gradients in the temperate zone experience very different conditions over relatively compressed spatial scales. With increasing elevation, temperature and growing season length decrease, hail and snow storms become more frequent, snow cover persists for longer periods of time, plant productivity is lower, and there is a delay in insect and berry emergence (Kikkawa and Williams 1971; Korner 1999; Hegelbach 2001; Bears et al. 2003; Diereg et al. 2006). Species respond in a number of ways to these elevational clines. Some species restrict breeding to narrow elevational bands that encompass conditions to which they are adapted (Ghalambor et al. 2006). Other species migrate between elevations throughout the breeding season to minimize exposure to challenging conditions (Inouye et al. 2000; Hahn et al. 2004), while others occur over a wide elevation range because inferior competitors are excluded from preferred elevations (Pearson and Rohwer 1998; Rohwer 2004; Martin and Wiebe 2004). Finally, some species breed over wide elevation ranges by adopting elevation-specific life-history strategies to maximize their lifetime reproductive success (Dunmire 1960; Zammuto and Mi l l a r 1985a,b), involving developing functional adaptations and making optimal trade-offs (e.g. Descamps et al. 2006) to deal with different constraints among elevations. This last group of species is ideal for comparative studies that aim to determine life-history traits that are selected for or constrained among elevations in a set region, as confounding influences of different phylogenetic histories or geographic zones are avoided. Despite birds being the most diverse vertebrate group in mountainous areas, no studies have determined how life-history strategies shift in populations of songbirds among elevations (Martin A version of this chapter is in preparation for submission to the journal Ecology with H. Bears, K. Martin, and G. C. White as authors. 25 2001). Intraspecific studies have instead focused on elevational variation of a few life-history traits such as clutch size (Krementz and Handford 1984; Hamann et al. 1989), breeding phenology (males only-Perfito et al. 2004), seasonal productivity (Purcell 2006), body condition (Widmer and Biebach 2001), physiological adaptation (Bears etal. 2003), and age specific patterns of elevational occupancy (Kollinsky and Landman 1996). Interspecific studies have also examined elevational differences in breeding times and productivities (Badyaev 1997; Sandercock et al. 1995a,b), offspring quality (Badyaev and Ghalambor 2001), and survival (Sandercock et al. 2005a,b). These studies provide data that can be combined to make predictions about life-history shifts with increasing elevation. Birds at higher elevations typically breed for shorter periods and produce fewer broods of an equivalent or lower clutch size than birds at lower elevations (Krementz and Handford 1984; Hamann et al. 1989; Badyaev 1997). The mechanisms orchestrating this shift are unclear, however. Normally an increase in photoperiod in the spring is the main cue used by birds to time the physiological, morphological, and behavioral changes that must occur to initiate breeding (Rowan 1925; Hamner 1966; Famer and Lewis 1971; Dawson et al. 2001). Eventually, birds become unresponsive to long days, reproduction ceases, gonads regress, and molting begins ('photorefractoriness': Farner et al. 1983; Nicholls et al. 1988). However, since day length is the same among elevations at the same latitude, birds at higher elevations may modify their reproductive schedules in response to supplementary cues such as weather variables and food supply (Wingfield et al. 1992; Perfito et al. 2004), which differ among elevations (Table 1.1), or may modify the daylength to which they respond, relative to lower elevation birds. Birds at higher elevations that have lower seasonal productivities may also survive longer (Sandercock et al. 2005a,b), produce higher quality offspring (Badyaev and Ghalambor 2001), and have fewer parasite infections (Stabler etal. 1974; Bears 2004). 26 In order to evaluate shifts in overall life-history strategies, however, one needs to study multiple traits in a single, controlled system allowing the observation of how they co-vary. Indeed, Sandercock et al. (2005a,b) showed the benefits of analyzing many potentially interrelated life-history traits, as they were able to suggest that birds switched from a high-reproductive to a high-survivorship life history strategy with increasing breeding elevation. However, these authors compared different grouse species among elevations located at very different latitudes. Therefore, it was unclear whether these trends were due to different phylogenetic histories, confounding regional/latitudinal differences, or a true effect of breeding elevation on the avian phenotype. In this chapter, I compared multiple interrelated life-history variables in a single songbird species, the dark-eyed junco (Junco hyemalis), breeding at the high- (2000m; sub-alpine) and low-(1000m; montane) elevation extremes of its range in Jasper National Park, Alberta. I compared shifts in reproductive schedules, timing of growth and recrudescence of gonads, seasonal reproduction, indicators of competitive status for this species (arrival time, age, and size), and local survival of dark-eyed juncos between elevations over 4 years. I tested the following main hypotheses: i) Reproductive Restriction Hypothesis - Dark-eyed junco males and females in high- compared to low-elevation habitat should have a shorter breeding period, corresponding to differences in the timing of gonad development, ii) Reproductive Reduction Hypothesis - High-elevation birds should produce fewer offspring per season as a result of a shorter breeding period that restricts the number of broods that can be produced, iii) Competitive Exclusion/Suboptimal Habitat Hypothesis -Competitive exclusion forces less competitive birds to occupy high-elevation habitats, and iv) Trade-off Hypothesis - Benefits such as increased survival may compensate birds breeding at higher elevations that experience lower seasonal reproduction. 2.2 M E T H O D S 2.2.1 Study species - Dark-eyed juncos breed from sea-level to the sub-alpine treeline in North America. Adult dark-eyed juncos feed on insects, berries, and seeds, while hatchlings and 27 young fledglings are fed insects and berries only. Dark-eyed juncos in my study region nest on the ground (Nolan et al. 2002), and juncos at high-elevation sites in Jasper National Park can nest up to 70 cm below ground (H. Bears, pers. obs.). Dark-eyed juncos produce 1-3 clutches of 2-5 eggs, which are incubated by females only (Nolan et al. 2002). Males and females both feed and defend young, but males are more involved in territorial defense. Adult and juvenile dark-eyed juncos show consistent philopatry in many regions (Nolan et al. 2002). There is a well-established competitive hierarchy in dark-eyed juncos, where younger, smaller, and later arriving birds compete less successfully for preferred resources and breeding areas (Cristol et al. 1990; Grasso et al. 1996). Additional information on this species is provided in chapter 1. 2.2.2 Field Site Description - Dark-eyed juncos were captured and monitored at eight 50 -70 ha sites in Jasper National Park ( 5 2 ° 5 3 \ 118°3'; F ig . 1.1), Alberta, from 01 Apr i l to 20 August, 2000, 01 M a y to 20 August, 2001, 12 A pr i l to 30 August, 2004, and OUune to July 30, 2005. Sites are described in chapter 1. 2.2.3 Field Procedures - Juncos were captured in mist nets using taped recordings of male juncos. A l l birds were marked with U.S . Fish and Wildlife Service aluminum bands, and unique color band combinations. In order to monitor changes in reproductive stages through time, I caught ca. an equal number of new (unhanded) juncos from each site during each sampling bout, during which 2 sites were visited per day for 4 days; this sampling protocol allowed me to look at the frequencies of birds in various reproductive stages throughout the season at each elevation with an equal representation of birds from all sites within each elevation group. Sample sizes vary with the measure taken, and the number of years over which a trait was married; sample sizes are indicated for each trait in the results section. Activities at the nest, morphological features associated with reproduction, and other behavioral clues (e.g., delivery of insects to and from an area with a nest) were observed in order to place birds in the following breeding status categories: unpaired, pre-laying, laying, incubating, tending to nestlings, tending to fledglings, post-reproductive (male gonads 28 are recrudescing, female brood patch is re-feathering, or body molt it occurring near end of the breeding season), or unknown. For each bird, I took standard morphological measures (analysed in chapter 4), and juncos were aged as hatching year ( H Y ) , second-year (SY) , after second-year age ( A S Y ) , or unknown, according to Pyle (1997). Territories of captured birds were marked and mapped using a GPS unit. 2 . 2 . 4 Reproductive Timing and Seasonal Reproductive Success - In 2000, surveys were conducted every 1-3 days from 01-15 A p r i l to determine arrival times in high- and low-elevation sites. Additional surveys were conducted to monitor reproductive schedules and the presence of previously banded birds in study sites. Banded birds were visited weekly to monitor the temporal schedules of reproductive activities within high- and low-elevation sites. I relied heavily on behavioural observations of reproductive stage of banded birds. Determining reproductive schedules using traditional nest-monitoring methods was not possible for this system, as a large proportion of high-elevation nests were subterranean and inaccessible. Weekly visits to territories of banded birds allowed me to determine the onset of breeding, hatchling and fledging dates, initiation of second and subsequent nesting attempts, and cessation of breeding. When young dark-eyed juncos of known pairs were just about to leave the nest, or had just left the nest, parents were aggressive, using vocalizations and conspicuous behavioral displays that indicated the presence of hatchlings. Recent fledglings, defined as young that had left the nest but were confined to walking along the ground and could not yet fly, were easily found, aged and counted during this stage. Both parents made frequent feeding visits to each recent fledgling, which were slow moving, unable to fly, and located within ca. 5-10 m of the nest site. Recent fledglings at this stage were aged as 10-12 days (Nolan et al. 2002). I then determined the date of clutch initiation by back-dating, where I began with the estimated age of the offspring (10-12 days), and subtracted the incubation period and clutch size (equivalent to number o f egg-laying days; 1 day per egg) for the appropriate elevation (Nolan et al. 2002). I found a small number of nests containing eggs or just prior to egg-laying at both elevations, however, and 29 followed them to record clutch size and incubation period. Eggs and hatchlings were not handled in order to minimize human disturbance. From surveys I was able to construct accurate egg-laying schedules. A t high- and low-elevation sites, respectively, the mean incubation periods used in back-dating equations were 12±1 days (n=4) and 12±1 days (n=10), and the mean clutch sizes used were 3.8±0.8 (n=9) and 3.5±0.5 (n=15). Each estimate made using this method wi l l have a maximum error of ca. ± 4-5 days. The number of recent fledglings per brood at each elevation was determined before fledglings became capable of flight. I repeated counts of fledglings during surveys when fledglings were 25-30 days of age and capable of flight but still in close family groups at both elevations. Known families were observed for 15-30 minutes until I had viewed and identified every fledgling in the family. While there was little observer error associated with counting hatchlings that have just left the nest and are not very mobile and on the ground, there was more potential error associated with fledgling counts; thus, I relied more on the former measure in my interpretations. I did not calculate life-history schedules in 2005, as the field season did not span the entire breeding period. 2.2.5 Variation in Reproductive Development and Termination - In 2004,1 measured morphological indicators of reproductive stage in order to compare when high- and low- elevation populations initiated and terminated breeding capacities (i.e., grew or recrudesced reproductive features). When male songbirds become reproductively capable of breeding in the spring, the distal ends of the ductus deferens (seminal glomera) fill with new spermatozoa resulting in a swelling of the cloaca referred to as the cloacal protuberance (CP), which is the external male genitalia (King 1981). C P dimensions vary in size over the course of the breeding season relative to male reproductive condition, and thus can be used as a measure of reproductive capacity (Lombardo 2001). For each male caught, I measured the cloacal length, width, and volume (King 1981, Lake 1981; Kempenaers et al. 1999), all of which rendered comparable results; thus only C P widths are presented, as in Deviche et al. (2000). A C P of ca. one half of the maximum size is considered 30 functional for reproduction (Gwinner 1986). Since C P size is not a reliable indicator of reproductive stage in females, I examined the brood patch area as an indicator of reproductive stage in females, as brood patches develop in response to increasing oestrogen and prolactin action that enable reproductive readiness (Baily 1952). I scored females from 1-4, where 1= N o brood patch formation, 2 = Partial feather loss/brood patch formation, 3 = Fully developed brood patch, 4 = Formerly active brood patch that is re-feathering. Therefore, the higher the brood patch score, the further along the female bird was in her reproductive cycle. Termination of seasonal reproductive activities in males was determined using primary molt score in addition to decreasing cloacal protuberance size. Molt ing of primaries indicates that a bird has completed reproductive activities, as molting is a significant physiological burden that is normally only feasible when reproductive duties are completed (Lindstrom et al. 1993, Schieltz and Murphy 1995). Thus, I examined males throughout the 2004 season at both elevations to compare their schedules of pre-basic feather molt, which starts with the loss and replacement of proximal primary wing feathers, and is completed when all feathers are replaced for full adult breeding plumage (Pyle 1997). Mol t was scored using the Ginn and Melvi l le (1983) method, whereby 0 is given to old feathers, 1- 4 to missing feathers and feathers growing in, and 5 to a completely grown, new feather. Mol t scores were calculated as the mean score of all flight feathers (3 tertials; 10 primaries, 6 secondaries) on the right wing. A s repeatedly catching the same individuals was not feasible for a time course analysis of reproductive indicators, new, unhanded birds were captured at each time interval. 2.2.6 Apparent Surv iva l Analysis - Surveys were conducted on each of the 8 study sites within Jasper National Park boundaries each month during the field season for each year. Surveys began at sunrise, and my assistant and I walked transects stopping at every corner of a 100m x 100m grid network using a GPS tracking device. When a junco was observed, its banding status was recorded. If I was uncertain of the identity of the bird, an assistant captured the bird in a mist net to 31 determine the band number. Due to time restrictions, less search time per transect per plot were used in some months; however, equivalent search times were always used between elevations. Therefore, the relative estimates generated between low- and high-elevation habitats (the comparison of interest here) are more meaningful than any absolute measures of survival generated from mark recapture data. I prepared my banded bird data for 153 low- and 140 high-elevation adult males for analyses of survival rates using the program M A R K (White and Burnham 1999) with the Cormack-Jolly-Seber model (Cormack 1964, Jolly 1965, Seber 1965). Data from S Y and A S Y males were analyzed for 15 occasions with unequal time intervals. I considered 17 models incorporating elevation as an attribute group and time effects in apparent survival (cp) and re-sighting (p) probabilities (Table 2.1), with apparent survival standardized to 1-month intervals. Interval lengths were 1, 1, 1,9, 1, 1, 1, 20, 1, 1, 1, 1, 10, and 1 month(s). Information-theoretic procedures using Akaike Information Criterion (AICc) (Burnham and Anderson 2002) were used for model selection. Goodness-of-fit of the global model {(p(elevation*r)/?(elevation*0} was evaluated with the procedures in program R E L E A S E (Burnham et al. 1987). I note that the focus on field work in 2004 was collection hatchlings for the common garden experiment, and so I did not have adequate time to band hatchlings within monitoring sites to estimate returns from 2004-2005; thus survival of first year birds could not be estimated. 2.2.7 Data Analysis and Software - Data are presented as means ± 1 standard error (n, number of birds). I pooled data across study sites by elevation to increase power, which was justified as preliminary data analysis showed that there were no anomalous site means with respect to other sites within the same elevation category, and because I used an approximately balanced design in site sampling. Preliminary nested A N O V A results with site nested by elevation also showed a non-significant nested term, enabling the removal of site level variance from the analysis. I tested data using two-way analyses of variance ( A N O V A ) followed by Tukey's post hoc tests. F-values are presented as F(df(effect)/df(error)). For female brood patch scores, I used Friedman's A N O V A on 32 ranks, followed by Holm-Sidak post hoc tests (Sokal and Rohlph 1995), because samples sizes were small, and data were non-normally distributed, with unequal variances. Tests were performed in SPSS 11 (Statistical Package for the Social Sciences), except for the survival modeling, which were performed using M A R K (White and Burnham 1999). A l l tests were two-tailed and alpha (a) was set to 0.05 and Bonferroni-adjusted (Holm 1979) as indicated. Partial Eta-squared values (eta2), which represent the proportion of variance of the dependent variable explained by an independent variable, are included with parametric analyses as a standard post hoc measure of effect size. Strengths of eta2 values are interpreted according to Cohen (1988) where 0.01=small effect, 0.06=moderate effect, and 0.14=large effect. Power, or ability of my data to detect a difference when one exists in nature, was > 0.90 unless otherwise indicated. 2.3 RESULTS 2.3.1 Variation in Egg-laying Schedules - The date by which the first 10 % of birds had initiated egg-laying was ca. 25-45 days later in high- compared to low- elevation habitats in al l years (Fig. 2.1). Birds in high-elevation habitats also ended their breeding seasons earlier than those in low elevation habitats, as 90% of all broods were completed 15-20 days earlier. The breeding season lasted only 38- 43 days at high elevations compared to 95-99 days at low-elevations. Median brood initiation occurred at similar times between elevations. 2.3.2 Variation in Reproductive Development and Termination- Males: Two- way A N O V A s conducted on C P width in males revealed a large interaction effect between elevation and month [F(4, 146)=2.22, p=0.001, eta2=0.72, Fig . 2.2]. Differences in CP widths between capture intervals, as determined by a Tukey's post hoc test, revealed differences in C P between elevations at the first, second, and fourth capture intervals. Compared to low-elevation males, high-elevation males had smaller cloacal protuberances early in the breeding season until late May . Cloacal protuberance declined again by 01-July. Low-elevation males, on the other hand, developed functional gonads prior to the beginning of the monitoring period. Mean C P sizes of S Y birds were 33 slightly lower than those of A S Y birds at every sample day at both elevations, although this difference was not significant within either elevation [F ( l , 146)=1.00, p>0.32]. CPs in juncos did not differ between elevations when they were categorized by life-history stages, with the exception that males with fledglings had smaller cloacas in high-versus low-elevation sites [F(4, 228)=6.32, p=0.02; Fig . 2.3]. When I compared males tending to fledglings of their last brood at low-elevations, versus males tending fledglings of their only (hence last) brood at high- elevations, this difference disappeared (p>0.37). Mol t scores did not differ between low- and high-elevation males as the breeding season progressed, but variation was high, and high-elevation males achieved significantly higher molt scores by the final sampling period on 20-Aug [F(4,76)=3.60, p=0.01; Fig . 2.4]. Females: Elevation had a significant impact on timing of brood patch development (Friedman's test, X 2=54.79, df = 2, p= 0.001, n=64). Females in low-elevation habitat had more developed brood patches until 15-June compared to high-elevation females (Holm Sidak post hoc, p<0.05, F ig . 2.5). However, there was no difference in the time that females began re-feathering their brood patches in both elevations (Holm Sidak post hoc, p>0.05). Therefore, in both males and females, the significant differences in physiological indicators of temporal breeding readiness and termination generally corresponded to differences in egg-laying schedules. 2.3.3 Variation in Seasonal Reproductive Success. - High-elevation females produced fewer broods on average than at low-elevations; high-elevation pairs produced only 0-1 broods (Mean = 0.75 ± 0.2, n = 32) while low-elevation pairs produced 0-3 broods (Mean = 1.55 ± 0.5, n = 48) per season (t-test, t=3.42, df=78, p<0.01). The mean number of young per brood, as counted shortly after hatchlings left the nest (not yet capable of flight), did not differ between elevations [High (n = 40): 2.7 ± 0.2; low (n = 45): 2.9 ± 0.2]. The average number of fledglings at age 25 days observed with parents at high- and low- elevations were also similar [High (n=20): 2.5 ± 0.3; L o w (n=20): 2.3 ± 0.5; t-test, t=2.4, n=38, p>0.05]. The average high-elevation junco produced 2.0 34 successful hatchlings, 1.9 of which survived to 25 days of age and the average low-elevation junco produced 4.5 successful hatchlings, 3.6 of which survive to 25 days of age. 2.3.4 Variation in Competitive Status of Populations - The proportions of S Y : A S Y male birds that could be aged reliably in high-elevation (36:54) and low-elevation (43:53) habitats did not differ (Fisher's Exact Test, p=0.30). The numbers of S Y : A S Y female birds that could be aged reliably in high-elevation (7:8) versus low-elevation (8:8) habitats were also not different (Fisher's Exact Test, p=0.71). According to early spring surveys conducted in 2000, dark-eyed juncos arrived at breeding sites at similar times, between 02 and 07 Apr i l . Birds remained in flocks during the initial arrival stage before establishing territories, which were often identical for marked birds from year to year within a site. 2.3.5 Variation in Apparent Survival - Goodness-of-fit tests from R E L E A S E suggested no lack of fit o f the global model (x2 =28.0, df=39, p=0.90). Therefore, no correction for over-dispersion was applied to these data. Mode l selection results evaluated using Aikake 's Information Criterion (AIC) supported the {(p(elevation + /) p(elevation)} model with 0.999 weight (Table 2.1). This model describes survival probability as time-specific cp values and with recapture probability allowed to vary with elevation. The estimated difference on the logit scale between the high- to low-elevation populations was 1.42 (SE = 0.37, 95% CI 2.14 to 0.69), demonstrating significantly higher survival rate estimates for males in high-elevation habitats at every time point examined relative to low-elevation estimates (Fig. 2.6; Appendix I). Hatchlings and females could not be modeled using the Cormack-Jolly-Seber approach due to sparseness of data; however, I made some qualitative observations about natal and female philopatry. In 2001,1 re-captured or re-sighted 5/20 (25%) low-elevation and 2/10 (20%) high-elevation birds that were banded as hatchlings in 2000, providing qualitative evidence for natal philopatry. For females, I re-captured 7/28 (25%) and 5/27 (19%) birds in high- and low-elevation sites, respectively. Most females were re-caught/sighted one season after 35 being banded; however, one high-elevation females was caught >3 years after being banded (banded in 2000, caught in 2004). 2.4 D I S C U S S I O N I demonstrated four main results in the present study. First, data supported the reproductive restriction hypothesis, as dark-eyed juncos in high-elevation habitat compressed the length of their reproductive season relative to birds at low-elevations. This compressed breeding schedule at high-elevations corresponded to a delay in growth, and earlier termination of reproductive structures in males and females. Second, data supported the reproductive reduction hypothesis, as high-elevation birds produced fewer broods and hence fewer offspring per season compared to low-elevation birds. Third, I failed to find support for the competitive exclusion hypothesis, as high-elevation populations were not composed of a higher proportion of younger, less-competitive classes of birds. Fourth, I found support for the trade-off hypothesis, as apparent survival in high-elevation males was higher than in low-elevation males. Taken together, I showed a strong influence of elevation on the life-history strategies o f populations similar to those seen in interspecific comparisons. M y first finding was that high-elevation juncos had a compressed egg-laying period relative to low elevation juncos. It is common for non-tropical avian species to limit reproductive activity to the time of year when reproductive attempts are more likely to succeed (Perrins 1970; Wingfield 1983). Reproductive success declines during the breeding season for many bird species (Perrins 1970), and breeding time is at least partially heritable (van Noordwijk et al. 1981; Price et al. 1988; Sheldon et al. 2003), meaning that directional selection should lead to evolutionary changes to breed as early as possible. The fact that directional selection does not push breeding dates to be earlier in most cases, has become the classical example for evolutionary stasis (Visser et al. 1998, 2006; Meri la et al. 2001), suggesting that the selection for an early breeding date is up against a resource limitation at the time of egg-production (Perrins 1970; Giennap et al. 2006). Egg production and rearing of offspring is costly in terms of energy, nutrients, and future fitness demands a high 36 proportion of a bird's daily energy budget (Monaghan et al. 1998; Stevenson and Bryant 2000). Therefore, dark-eyed juncos from high-elevations may be energetically constrained to breed for only ca. 40-50% as long in dark-eyed juncos from low-elevations. Constraints on high-elevation breeding times could be due to multiple proximate factors. Most variables that change with increasing breeding elevation are known to negatively affect the seasonal productivity of birds. Colder temperatures are associated with a later reproductive onset in avifauna, resulting in shorter breeding seasons, fewer broods, and a decrease in seasonal reproductive success (e.g. Stevenson and Bryan 2000; Naef-Daenzer 2004; Torti and Dunn 2005; Weggler 2006). Persistent snow cover in spring can delay reproduction in ground nesting birds, resulting in fewer clutches (Inouye et al. 2000; Hendricks 2003). Severe storm events in the spring can also delay egg-laying and induce reproductive failure (Wiebe and Martin 1995; Coulter and Bryan 1995; Martin and Wiebe 2004). Finally, delays in food emergence correlates with delayed egg-laying and lower seasonal productivity in great and blue tits (Parus major and P. caeruleus; Svensson and Nilsson 1995; Thomas et al. 2001) and Eurasian dippers (Cinclus cinclus; Hegelbach 2001). Since all o f these variables shift collectively with increasing elevation (Table 1.1), it is perhaps not surprising that a negative correlation between seasonal reproductive success and breeding elevation has been documented in this study, and in others (Krementz and Handford 1984; Hamann et al. 1989; Badyaev 1997; Perfito et al. 2004). The correlative relationships between weather and insect biomass and brood initiation between elevations are examined in more detail in chapter 3. High-elevation juncos also terminated breeding earlier than low-elevation juncos. N o second broods were observed at high-elevation sites. The first clutch was likely initiated late enough that high-elevation birds attempting a second brood would have had insufficient time to raise the brood before their pre-basic molt began. Feather replacement presents a significant physiological and energetic burden that birds do not normally initiate while involved in reproductive duties (Lindstrom 37 et al. 1993, Schieltz and Murphy 1995). There is also a constraint on the rate at which new feather mass can be accumulated, and i f birds molted while tending a late second brood, smaller feathers of poorer quality might result (Dawson et al. 2000). Hence, high-elevation juncos may gain more from terminating reproduction directly following their first brood and investing energy into feeding fledglings and molting. This finding contrasts with results of Perfito et al. (2004), who found that male song sparrows (Melospiza melodia morphna) breeding at coastal sites at 270 m asl began molting earlier than conspecifics breeding at 1210 m asl. Perfito et aVs (2004) study examined an elevational difference that was less extreme and at a lower elevation compared to my study, andthus their birds may not have encountered the same late season energetic restrictions. Further, only males were examined by Perfito et al. (2004). Paralleling the different egg-laying schedules between elevations, males and females differed in the phenology of growth and termination of reproductive features between elevations. I showed that, in high- compared to low-elevation habitats, males enlarged their cloacal protuberances later and began cloacal recrudescence and molt earlier. In addition, when I examined male cloacal protuberance size by breeding stage, I found that high-elevation males recrudesced their cloacas when they were still attending their first brood of fledglings, as they no longer required active testes for a second brood. Likewise, I found that a higher proportion of females formed full brood patches later in the breeding season and had re-feathered brood patches more by the last sampling interval in high- compared to low-elevation habitats. These data show that high- and low-elevation birds do not become physiologically capable of reproduction at the same time, and high-elevation birds wait for favorable conditions to begin breeding. Instead, birds confined reproductive development to appropriate but different breeding intervals, which can conserve energy (Thomas et al. 2001). The temporal differences in timing of reproductive maturation and termination of structures associated with breeding between elevations raises interesting questions about the cues used to initiate the physiological changes that precede clutch initiation. Increasing daylength is normally the 38 primary cue that temperate songbirds use to initiate development and maturation of reproductive structures (Wingfield and Kenagy 1991; B a l l 1993; Wingfield and Farner 1993). Later in the season, birds become refractory to the stimulatory effects of long days, their gonads regress, and feather replacement (molt) begins (Nicholls et al. 1988). Since daylength is the same between elevations, juncos may rely on supplementary cues such as temperature, snow cover, and nutritional status in order to alter their reproductive schedules, or populations have evolved to respond to different critical day lengths among elevations, as they can with latitude (Silverin et al. 1993; Ftahn 1998). I provide some support for the hypothesis that populations respond to different critical daylengths between elevations in chapter 4, but show a complicated role of elevation-specific plasticity to the environment in reproductive timing as well . . M y second finding was that, due to the shorter breeding period in high-elevation habitats, high-elevation juncos produced fewer than half the number of offspring per season as their low-elevation conspecifics. The discrepancy in the number of offspring per female per season between elevations was mainly due to high-elevation birds producing less than half of the broods as low-elevation birds. Other studies on songbirds have also documented a reduction in the number of broods per season with increasing breeding elevation (Krementz and Handford 1984; Hamann et al. 1989; Badyaev 1997). High-elevation birds did not compensate for having fewer broods by producing more chicks per brood, but instead produced broods with similar numbers as low-elevation birds. Other intra- and inter-specific studies on clutch size and fledgling production in songbirds have also demonstrated that clutch size remained constant (Hamann et al. 1989) or decreased (Krementz and Handford 1984; Badyaev 1997; Fargalla 2004; but see Purcell 2006) with increasing elevation. M y third finding was that, although high-elevation birds had a lower seasonal reproductive success, high-elevation populations were not composed of inferior competitors. Smaller, younger, and later arriving dark-eyed juncos are known to be inferior competitors (Cristol et al. 1990; Grasso et al. 1996). A t high-elevation sites, dark-eyed juncos were not smaller (see chapter 5), and populations did not contain a higher proportion of younger birds. High- and low-elevation juncos also arrived at similar times in Apr i l . Kol l insky and Landmann (1996) also found no support for the hypothesis that younger male black redstarts were forced to settle at higher elevations, and instead observed a similar age structures and pattern of philopatry between elevations. Similarly, Widmer and Biebach (2001) found no difference in the sizes and age ratios of garden warblers at the extremes of a 1000 m elevation gradient in Europe. However, male hermit and Townsend's warblers arrived later at high-elevation breeding sites after spring migration, and yearlings were displaced into higher-elevation habitat (Pearson and Rohwer 1998; Rohwer 2004). M y fourth finding was that adult males in high-elevation habitats had higher estimated survival rates. The differences revealed in survival models, however, could also be due to higher emigration rates from low-elevation sites, as emigration and mortality are treated identically in CJS models. I observed a similar trend of higher female return rates to high-elevation habitats. Therefore, i f elevation influences survival high-elevation birds may trade lower current reproductive effort for longer lives and may be compensated for their seasonal reproductive deficit on a lifetime basis i f they are able to breed for more seasons. A n inverse correlation between present and future reproduction with survival has been shown in other vertebrates (Reznick 1985; Stjernman et al. 2004; Morton et al. 2004; Sandercock et al. 2005a; Parejo and Danchin 2006; Descamps et al. 2006; Ruf et al. 2006), and studies have suggested mechanisms for how this trade-off might occur (Dawson et al. 2000; Stjernman et al. 2004; Wiersma et al. 2004). One hypothesis to explain a trade-off between reproduction and survival in birds is that the high workload during reproduction compromises resistance to parasites and that the resulting increase in parasitaemia lessens the prospects of future survival (Stjernman et al. 2004). I previously found lower incidences of blood parasites in high- compared to low-elevation adults and fledglings in this system (Bears 2004). Another hypothesis is that shunting energy into reproduction for an extended period during the 40 breeding season leaves adults and fledglings at a lower nutritional state, which decreases their chances of overwinter survival (Badyaev 1997; Badyaev and Ghalambor 2001). Previously, I showed that high-elevation fledglings had greater fat reserves prior to fall migration (Bears 2002), perhaps due to greater synchronicity between breeding and a peak in food supplies (chapter 3), less competition for food resources, or a greater parental effort per offspring (Badyaev and Ghalambor 2001). Fledglings with larger fat reserves often survive better in the fall (Leary et al. 1999; Oddie 2000). I also previously showed that high-elevation males and females experienced lower levels of physiological stress in response to noxious stimuli relative to low-elevation birds (Bears et al. 2003). Predation of adults could also be greater at low-elevations, such that low-elevation birds must '•V produce as many young as possible early in life, as their chances of being depredated within or between seasons is substantially higher, while high-elevation birds can afford to wait. Predation of fledglings over 25-30 days of age may also be higher in low-elevation habitat, although in the yellow-eyed junco (Junco phaenotus), predation is far less important once birds are capable of flight (Sullivan 1989). Future studies that compare predation rates or juvenile survival between elevations would be helpful in evaluating these possibilities. In conclusion, I showed that a number of inter-related life-history traits change with increasing elevation in a ground nesting songbird. High-elevation juncos had a compressed egg-laying schedule, as they developed reproductive features later in the season, and recrudesced those structures earlier. A s a result of the compressed reproductive period, high-elevation birds had fewer broods and offspring per season, compared to low-elevation birds. High-elevation populations were not composed of inferior birds forced to breed at upper elevations. Instead, a shift from high-reproductive to high-survivorship life-history strategy may be occurring with elevation. M y results appear to fit the life-history shift expected with breeding elevation as suggested in the interspecific comparison of Sandercock et al. (2005a,b). In my next chapter, I explore how weather variables correlate with breeding phenology in high- and low-elevation birds. In later chapters, I also 41 determine whether differences uncovered in the present chapter are primarily due to local genetic adaptation or phenotypic plasticity. 42 2.5 R E F E R E N C E S Badyaev, A . V . 1997. Avian life history variation along altitudinal gradients: an example with cardueline finches.- Oecologia 111 365-374. Badyaev, A . V . , and Ghalambor, C K . 2001. Evolution of life histories along elevational gradients: trade-off between parental care and fecundity. Ecology 82: 2948-2960. Bai ly , R . B . 1952. The incubation patch of passerine birds. - Condor 54: 212-256. Ba l l , G.F. 1993. The neural integration of environmental information by seasonally breeding birds. -Am. Zool. 33: 185-199. Bears, H . 2002. How and why a small songbird breeds over a steep elevation gradient. M . S c . Thesis, University of British Columbia. Bears, H . , Smith, J . N . M . , and Wingfield, J .C. 2003. Adrenocortical sensitivity to stress in Dark-eyed Juncos (Junco hyemalis oregonus) breeding in low and high elevation habitat. - Ecoscience 10: 127-133. Bears, H . 2004. Parasite prevalence in dark-eyed juncos (Junco hyemalis) breeding at different elevations. - Can. Field Nat. 118: 235-238. Burnham, K . P., Anderson, D. R., White, G . C , Brownie, C , and Pollock, P . H . 1987. Design and analysis methods for fish survival experiments based on release-recapture. American Fisheries Society Monograph N o . 5. Bethesda, Maryland, U S A . Burnham, K . P . , and Anderson, D.R. 2002. Mode l selection and multimodel inference: a practical information-theoretic approach. Springer-Verlag, New York. Cohen, J. 1988. Statistical power analysis for the behavioural sciences. Hillside, N J , McGraw H i l l . Cormack, R. M . 1964. Estimates of survival from the sighting of marked animals. - Biometrika 51:429-438. 43 Coulter, M . C , and Bryan, A . L . 1995 Factors affecting reproductive success of wood storks (Mycteria americana) in east-central Georgia. - Auk 112: 237-243 Cristol, D . A . , Nolan, V(Jr)., and Ketterson, E . D . 1990. Effect of prior residence on dominance status of dark-eyed juncos, Junco hyemalis. - Anim. Behav. 40: 580-586. Dawson, A . , S. A . Hinsley, P. N . Ferns, R. H . C. Bonser, and Eccleston, L . 2000. Rate of moult affects feather quality: A mechanism linking current reproductive effort to future survival. -Proc. Roy. Soc.B 267:2093-2098. Descamps, S., Boutin, S, Bertrauux, D , and J . M . Gaillard, J . M . 2006. Best squirrels trade long life for an early reproduction. - Proc. Roy. Soc. B 273: 2369-2374. Deviche, P., Wingfield, J . C , and Sharp, P.J. 2000. Year class differences in the reproductive system, plasma prolactin and corticosterone concentrations and onset of pre-basic molt in male dark-eyed juncos (Junco hyemalis) during the breeding period. Gen. Comp. Endocrinol. 118: 425-435. Diereg, D . A . , Adam, N .R . , Mackey, B . E . , Dahlquist, G . H . , and Coffelt, T . A . 2006. Temperature and elevation effects on plant growth, development, and seed production of two Lesquerella species. - Ind. Crop. Prod. 24: 17-25 Dunmire W . W . 1960. A n altitudinal survey of reproduction in Peromyscus maniculatus. -Ecology 41: 174-182 Enstrom, D . A . , Ketterson, E .D . , and Nolan, V.(Jr), 1997. Testosterone and mate choice in the dark-eyed junco. - Anim. Behav. 54:1135-1146 Fargallo, J .A. 2004. Latitudinal trends of reproduction in the blue tit Parus caeruleus. -Ardeola 51: 177-190. Farner, D.S., and Lewis, R . A . 1971. Photoperiodism and reproductive cycles in birds. -Photophysiol. 6: 325-370. Farner, D . S., Donhan, R. S., Matt, K . S . , Mattocks, P.W., Moore, M . C , and Wingfield, J .C. 1983. The nature of photorefractoriness. In Av ian endocrinology: environmental and ecological perspectives (Mikami, S., Homma K . , and Wada, M . , Eds). Pp 149-166. Tokyo: Japanese Scientific Society Press. Ghalambor, C . K . , Huey, R . B . , Martin, P.R., Tewksbury, J.J., and Wang, G . 2006. Are Mountain passes higher in the tropics? Janzen's hypothesis revisited. - Int. Comp. Biol. 46: 5-17 Gienapp, P., Postma, E . , and Visser, M . E . 2006. Why breeding time has not responded to selection for earlier breeding in a songbird population. - Evolution 60: 2381-2388. Grasso, M . J . , Savalli, U . M . , and Mumme, R. 1996. Status signalling in Dark-eyed Juncos: perceived status of other birds affects dominance interactions. - Condor 98: 636-639. Ginn, H . B . , and Melvi l le , D.S. 1983, reprint in 2000. Moult in Birds: British Trust for Ornithology, Tring, United Kingdom. Gwinner, E . 1986. Circannual Rhythms. Springer-Verlag, Heidelberg. 125 pp. Hahn, T.P. 1998. Reproductive seasonality in an opportunistic breeder, the Red Crossbill, Loxia curvirostra. - Ecology 79: 2365-2375. Hahn, T.P., Pereyra, M . E . , Sharbaugh, S . M . and Bentley, G .E . 2004. Physiological responses to photoperiod in three cardueline finch species. - Gen. Comp Endocrin. 137: 99-108. Hamann H . J., Schmidt, K . H . , and Simonis, S. 1989. Elevation and size of eggs and clutch in the great tit Parus major.-J. Fiir Ornithol 130: 69-74. Hamner, W. H . 1966. Circadian control of photoperiodism in the House Finch demonstrated by interrupted-night experiments. - Nature 203: 1400-1401 Harper, E .D. , and Skinner, N . D . 1998. Clinical nutrition of small psittacines and passerines. - Semin. Av. Exot. Med 7: 116-127. Hegelbach, J. 2001. Water temperature and phytophenology indicate the earlier onset of oviposition in Eurasian Dipper (Cinclus cinclus) from the Swiss Lowlands. - J. Fiir Ornithol. 142: 284-294. Hendricks, P. 2003. Spring snow conditions, laying date, and clutch size in an alpine population of American Pipits. - J . Field Ornithol. 74:423-429. Holm, S. 1979. A simple sequentially rejective multiple test procedure. - Scand. J.Stat. 6: 65-70. Inouye, D . W . , Barr, W . A . , Armitage, K . B . , and Inouye, B . D . 2000. Climate change is affecting altitudinal migrants and hibernating species. - PNAS'97: 1630-1633. Jolly, G . M . 1965. Explicit estimates from capture-recapture data with both death and immigration-stochastic model. - Biometrika 52:225-247. Kempenaers, B . , Congdon, B . , Boag, P., and Robertson, R.J . 1999, Extrapair paternity and egg hatchability in Tree Swallows: Evidence for the genetic compatibility hypothesis?-Behav. Ecol. 10: 304-311. Kikkawa, J., and Williams, W.T. 1971. Altitudinal distribution of land birds in New Guinea. Search 2: 64-65. K ing , A . S. 1981. Cloaca: Pages 107-148 in Form and Function in Birds, vol . 2 (King, A . S., and McLel land, J., Eds.). Academic Press, New York. Lindstrom, A . , G . H . Koll insky, C , and Landmann, A . 1996. Altitudinal distribution of male Black Redstarts: Are there age-dependent patterns? Bird Study 43: 103-107. Korner, C . 1999. Alpine Plant Life: Functional Plant Ecology of High Mountain Ecosystems. Springer-Verlag, Berl in and Heidelberg, Germany. Krementz, D . G . , and Handford, P. 1984. Does avian clutch size increase with elevation? -Oikos 43: 256-259. Lake, P. E . 1981, Male genital organs: Pages 1-61 in Form and Function in Birds, vol. 2 (King, A . S . , and McLel land, J., Eds.). Academic Press, New York. Lapp, S., Byrne, J., Townshend, I., and Kienzle, S. 2005. Climate warming impacts on snowpack accumulation in an alpine watershed. - Int. J. Climatol. 25: 521-536. Leary J, Sullivan K . A . , and Hillgarth N . 1999. Relationships among dominance, foraging proficiency, and condition in juvenile dark-eyed juncos. -Auk 116: 1136-1141. Lindstrom, A . , Visser, G . H . , and Daan, S. 1993, The energetic cost of feather synthesis is proportional to basal metabolic rate. - Physiol. Zool.. 66 490-510. Lombardo, M . P . 2001. Individual and seasonal variation in external genitalia of male Tree Swallows. - Auk 118: 789-795. Martin, K . 2001. Wildlife communities in alpine and subalpine habitats. In Wildlife-habitat relationships in Oregon and Washington. Oregon State University Press; Corvallis, U S A (Johnson, D . H . , and O ' N e i l , T .A . , Managing Directors). Pp 239-260. Martin, K . and Wiebe, K . L . 2004. Coping mechanisms o f alpine and arctic breeding birds: Extreme weather and limitations to reproductive resilience. - Int. Comp. Biol. 44: 177-185. Merl ia , J., Sheldon, B . C . , and Kruuk, L . E . B . 2001. Explaining stasis: microevolutionary studies in natural populations. - Genetica 112: 199-222. Monaghan P., Nager R . G . , and Houston, D . C . 1998. The price of eggs: increased investment in egg production reduces the offspring rearing capacity of parents. - Proc. Roy. Soc. B 265: 1731-1735. Morton, M . L . , Pereyra, M . E . , Crandall, J.D., MacDougall-Shackleton, E . A . , and Hahn, T.P. 2004. Reproductive effort and return rates in the Mountain White-crowned Sparrow. Condor 106: 131-138. Naef-Daenzer L , Nager, R . G . , Keller, L . F . , and Naef-Daenzer, B . 2004. Are hatching delays a cost or a benefit for Great Tit Pants major parents? - ARDEA 92: 229-237. Nicholls, T.J., Goldsmith, A . R . , and Dawson, A . 1988. Photo- refractoriness in birds and comparison with mammals. - Physiol. Rev. 68: 133-176. Nolan, V . , Jr., Ketterson, E . D . Cristol, D . A . , Rogers, C M . , Clotfelter, R . C J i t u s , R . C , Schoech, S.J., and Snajdr, E . 2002. Dark-eyed Junco (Junco hyemalis): In The Birds of North America, No . 716 (Poole, A . , and Gil ls , F., Eds.). Academy of Natural Sciences, Philadelphia, and American Ornithologists Union, Washington. Oddie K . 2000. Size matters: competition between male and female great tit offspring. J. Anim. Ecol. 69: 903-912. Parejo, D. , and Danchin, E . 2006. Brood size manipulation affects frequency of second clutches in the blue tit. - Behav. Ecol. Sociobiol. 60: 184-194. Pearson, S. F . , and Rohwer, S. 1998. The influence of breeding phenology and clutch size on hybridization between Hermit and Townsend's warblers. - Auk 115: 739-745. Perfito, N . , Tramontin, A . D , Meddle, S., Sharp, P, Af ik , D . , Gee, J., Ishii, S., Kikuchi , M . , and Wingfield, J.C. 2004. Reproductive development according to elevation in a seasonally breeding male songbird. - Oecologia 140:201-210. Perrins, C M . 1970. The timing of birds' breeding seasons. - Ibis 112:242-255. Price, T, Kirkpatrick, M , and Arnold, S.J. 1988. Directional selection and the evolution of breeding date in birds. Science 240: 798-799. Purcell, K . L . 2006. Abundance and productivity of warbling vireos across an elevational gradient in the Sierra Nevada. - Condor 108: 315-325. Pyle, P. 1997. Identification guide to North American birds - part 1. Slate Creek Press, P O Box 219, Bolinas, C A 94924-0219, U S A . 732 pp. I S B N 0-9618940-2-4. Reznick, D . 1985. Costs of reproduction: an evaluation of the empirical evidence. - Oikos 44: 257-267. Rohwer, S. 2004. Using age ratios to infer survival and despotic breeding dispersal in hybridizing warblers. - Ecology 85: 423-431. Rowan, W . 1925. Relation of light to bird migration and developmental changes. - Nature 115: 494-495. Ruf, T., Fietz, J., Schlund, W. , and Bieber, C. 2006. High survival in poor years: Life history tactics adapted to mast seeding in the edible dormouse. - Ecology 87: 372-381. Sandercock, B . K . , Martin, K . , and Hannon, S.J. 2005a. Life history strategies in extreme environments: comparative demography of arctic and alpine ptarmigan. - Ecology 86: 2176-2186. Sandercock, B . K . , Martin, K . , and Hannon, S.J. 2005b. Demographic consequences of age-structure in extreme environments: population models for arctic and alpine ptarmigan. - Oecologia 146: 13-24. Schieltz, P. C , and Murphy, M . E . 1995. Diurnal variation in oxygen consumption by molting and nonmolting sparrows. - Comp. Biochem. Physiol. A. 112: 265-272. Seber, G . A . F. 1965. A note on multiple-recapture census. - Biometrika 52: 249-259. Sheldon B . C . , Kruuk L . E . B . , and Meri la , J. 2003. Natural selection and inheritance of breeding time and clutch size in the collared flycatcher. - Evolution 57: 406-420. Silverin, B . , Massa, R. and Stokkan, K . A . 1993. Photoperiodic adaptation to breeding at different latitudes in Great tits. - Gen. Comp. Endocrinol. 90: 14-22. -Biol. Reprod. 71: 979-986. Stjernman, M . , Raberg, L . , and Nilsson, J .A. 2004. Survival costs of reproduction in the blue tit (Parus caeruleus): a role for blood parasites? - Proc. Roy. Soc. B 271: 2387-2394. Sokal, R. R., and Rohlf. F.J . 1995. Pages 321-342 in Biometry: the principles and practice o f statistics in biological research. 3rd edition. W . H . Freeman and Co.: New York. 887 pp. Stabler, R . M . , Kitzmiller, N . J . , and Braun, C E . 1974. Hematozoa from Colorado birds. VI.Galliformes. J. Parasitol. 60: 536-537. Stenseth, N . C , and Mysterud, A . 2002. Climate, changing phenology, and other life-49 history traits: Non-linearity and match-mismatch to the environment. - PNAS 99: 13379-13381. Stevenson, I.R., and Bryant, D . M . 2000. Avian phenology- climate change and constraints on breeding. - Nature 406: 366-367. Sullivan, K . A . 1989. Predation and starvation: age specific mortality in juvenile juncos (Junco phaenotus). - J. Anim. Ecol. 58: 275-286. Svensson, E , and Nilsson, J .A. 1995. Food- supply, Territory quality, and Reproductive Timing in the Blue Tit (Parus caeruleus). - Ecology 76: 1804-1812. Thompson, D . B . A . , and Whitfield, D.P. 1993. Research on mountain birds and their habitats: research progress report. - Scottish Birds 17: 1-8. Thomas, D .W. , Blondel, J., Perret, P., Lambrechts, M . M . , and Speakman, J.R. 2001. Energetic and Fitness Costs of Mismatching Resource Supply and Demand in Seasonally Breeding Birds. - Science 291: 2598-2600. Torti, V . W . , and Dunn, P.O. 2005. Variable effects of climate change on six species of North American birds. - Oecologia 145: 486-495 Lindstrom, A . , Visser, G . H . , and Daan, S. 1993. The energetic cost of feather synthesis is proportional to basal metabolic rate. - Physiol. Zool. 66 490-510. V a n Noordwijk, A . J . , van Balen, J .H. , and Scharloo, W . 1981. Genetic- variation in the timing of reproduction in the great tit. - Oecologia 49: 158-166 Visser, M . E . 1998. Warmer Springs lead to mistimed reproduction in great tits (Parus major). -Proc. Roy. Soc.B 265: 1867-1970. Visser, M . E . , Holleman, L . J . M . , and Gienapp, P. 2006. Shifts in caterpillar biomass phenology due to climate change and its impact on the breeding biology of an insectivorous bird. -Oecologica 147: 164-172. Weggler, M . 2006. Constraints on, and determinants of, the annual number of breeding attempts in the multi-brooded Black Redstart Phoenicurus ochruros. - Ibis 148: 273-284. White, G .C . , and Burnham, K . P . 1999. Program M A R K : survival estimation from populations of marked animals. - Bird Study 46: 120-139. Widmer, M . 1999. Altitudinal variation of migratory traits of the garden warbler Sylvia borin. PhD Thesis, University of Zurich, Zurich, Switzerland. Widmer, M . , and Biebach, H . 2001. Changes in body condition from spring migration to reproduction in the garden warbler Sylvia borin: a comparison of lowland and a mountain population. - Ardea 89:57-68. Wiebe, K . L . and Martin, K . 1995. Physiological and environmental effects on laying rates in ptarmigan. - Condor 97: 708-717. Wiersma, P., Sleman, C , Speakman, J.R., and Verhulst, S. 2004. Birds sacrifice oxidative protection for reproduction. - Proc. Roy. Soc.B 271:S360-S363. Wingfield, J .C. 1983. Environmental and endocrine control of reproduction: an ecological approach. In: Avian Endocrinology: Environmental and Ecological Perspectives (Mikami, S.I, Homma, K . , and Wada, M ) . Pp. 265-288. Berlin, Springer-Verlag. Wingfield J . C , and Kenagy, G.J. 1991. Natural regulation of reproductive cycles: Vertebrate endocrinology: fundamentals and biomedical implications (Schreibman, M . , and Jones, R . E . Eds). Pp 181-241. New York, Academic. Wingfield, J . C , Hahn, T.P., Levin , R., and Honey, P. 1992. Environmental predictability and control of gonadal cycles in birds.- J. Exp. Zool. 261: 214- 231. Wingfield, J .C. and Farner, D.S. 1993. Endocrinology of reproduction in wi ld species. in Avian Biology, V o l . 9 (Farner, D. S., K ing , J.R., and Parks, K . C , Eds.). Pp. 163-327. Academic Press, San Diego, California. Wingfield, J . C , Hahn, T.P., Maney, D . L . Schoech, S.J., Wada, M . , and Morton, M . L . 51 2003. Effects of temperature on photoperiodically induced reproductive development, circulating plasma leuteinizing hormone and thyroid hormones, body weight, fat deposition and molt in mountain white-crowned sparrows, Zonotrichia leucophrys oriantha. - Gen. Comp. Endocrinol. 131: 143-158. Zammuto, R . M . , and Mil la r , J.S. 1985a. Environmental predictability, variability, and Spermophilus colnmbianus life history over an elevational gradient. - Ecology 66: 1784-1789. Zammuto, R . M . , and J.S. Mil lar . 1985b. A consideration of bet-hedging in Spermophilus columhianus. - J. Mammol. 66: 652-660. 52 Table 2 .1 Mode l rankings and criteria from the computer program M A R K (White and Burnham 1999) for the predominant factors tested to explain males dark-eyed junco (Junco hyemalis) monthly survival ((p) and re-sighting probabilities (p) from 2000-2005 in high- and low- elevation sites in Jasper National Park. Akaike's Information Criterion (AIC), difference between A I C values between models (AAICc) , and A I C weights are included to show the level of support. Mode l selection results supported the {(p(elevation + t) /^(elevation)} model as the most parsimonious with 0.999 weight, which describes survival probability as time-specific cp values, with recapture probability allowed to vary with elevation. The period symbol (.) means that the parameter was constant in the model. Model A I C c A A I C c A I C c Model Num. Deviance Weights Likelihood Par. {(p(elevation +1) p(elevation)} 658.79 0 0.99 1 17 622.94 M 0 p(elevation)} 672.46 13.66 0.0011 0.0011 16 638.81 {^(elevation*/) p(.)} 676.35 17.56 0.00015 0.0002 28 615.26 {(p(elevation*r) /?(elevation)} 678.02 19.22 0.000070 0.0001 29 614.55 M 0 / > ( • ) } 680.70 21.91 0.000020 0.00 15 649.26 {^(elevation*?) p(t)} 694.59 35.79 0.00 0.00 41 601.33 M 0 / ? ( 0 } 696.31 37.52 0.00 0.00 28 635.22 {(p(elevation) p(t)} 696.87 38.08 0.00 0.00 16 663.23 {(p(/) ^(elevation*/)} 699.87 41.08 0.00 0.00 41 606.62 {(p(elevation*?) /?(elevation*?)} 703.28 44.49 0.00 0.00 50 586.11 {(p(elevation) /?(elevation)} 707.60 48.81 0.00 0.00 4 699.48 {(p(elevation) p(.)} 711.35 52.56 0.00 0.00 3 705.28 M . ) p(elevation)} 711.63 52.84 0.00 0.00 3 705.56 M O M O I 715.56 56.76 0.00 0.00 15 684.11 {(p(elevation) ^(elevation*?)} 717.15 58.36 0.00 0.00 30 651.28 M . ) ^(elevation*/)} 720.12 61.33 0.00 0.00 29 656.65 W O K ) } 730.17 71.38 0.00 0.00 2 726.14 53 H = n ~ H i g h • T T I - Low c TO 6) UJ TJ 0 W 0 x : o O (a) 2000 38 days 99 days DATE Fig. 2 .1 Percentages of clutches initiated (date of first egg) by date in high- versus low-elevation study sites in 2000 [n(high)=67, n(low)=83], 2001 [n(high)=81, n(low)=82], and 2004[n(high)=60, 54 n(low)=58,]. Dates corresponding to 10 t h and 90th percentiles (initiation and cessation dates, respectively) are indicated by beginning and end of horizontal boxplot rectangles. Median initiation dates are indicated by the central lines within boxplots. The number of days over which broods wen initiated at each elevation is indicated above boxplots. 55 1 0 9 • r ~ i r~ 1 N = 9 8 30 28 21 18 11 10 7 4 A p r i l M a y J u n e J u l y A u g u s t Month Fig. 2.2 Mean cloacal widths (±1SE) of male dark-eyed juncos from high- and low- elevation habitat throughout the 2004 breeding season. High- and low-elevation groups differed significantly from one another within A p r i l , M a y and July sample intervals, but not June. Sample sizes indicated on x-axis are the number of individual birds. 56 E £ 1 0 8 -C 5 6 CO o CD o i H i g h • L o w N = 1 0 2 0 9 1 5 N e s t Bu i ld ing/ Incubat ion 3 0 4 5 F e e d i n g ha tch l i ngs 2 9 3 5 F e e d i n g f l edg l ings 1 5 2 0 P o s t -r e p r o d u c t i v e P r e - l a y i n g Life-History Stage Fig . 2.3 Mean cloacal widths (±1SE) within high- and low- elevation male dark-eyed juncos, adjusted by life-history stage in 2000, 2001, 2004 and 2005. A n asterisk (*) indicates a significant difference. 57 High Low O o CO 3 4 2 H 7 ^ N 3 3 01-Jun 8 8 8 8 13- 01-Jul Aug Sample Date 8 8 10-Aug 8 8 20-Aug F i g . 2.4 Mean molt scores (±1SE) calculated from the right wing of high and low elevation males in 2004 (2 birds sampled from each of 8 sites within Jasper at each elevation at each sample interval). A n asterisk (*) indicates a significant difference between groups. 58 M—Low Apr May May Jun Jun Jul 01 - Jul 15- Aug Aug 15-20 01-15 15-20 01-15 15-30 15 30 01-15 15-30 Date Fig. 2.5 Mean brood patch scores (±1SE) of female dark-eyed juncos from high- and low-elevation sites in Jasper National Park (including al l years; n (high, low) = 30, 34). Brood patches were scored as: 1= N o brood patch formation, 2 = Partial feather loss/brood patch formation, 3 = fully developed brood patch, 4 = Brood patch that is re-feathering. The higher the mean brood patch score, the further along the average female is in her reproductive cycle. 59 1 " 0.9 -CO 0.8 -> • f c 0.7 -ZJ CO 0.6 -0.5 -c CD 0.4 -CD Q. 0.3 -CL 0.2 -< 0.1 -0 --0 - A - .A. - - High • Low 10 12 14 16 2 4 6 8 Interval Fig. 2.6 Apparent monthly survival estimated from the top ranking model, {<p(elevation + t) />(elevation)}, where survival (cp) is described as a function of breeding elevation and time, and recapture probability (p) is described as a function of breeding elevation only. This model had an A I C c weight of 0.999, and thus was strongly supported by these data. A t every time interval examined, survival estimates were significantly higher in high-elevation habitat. 60 3 WEATHER, FOOD ABUNDANCE, AND BREEDING AMONG ELEVATIONS - IS BROOD INITIATION CORRELATED WITH THE SAME OR DIFFERENT SUPPLEMENTARY CUES AMONG ELEVATIONS? f 3.1 INTRODUCTION The global average temperature has increased by approximately 0.6 °C over the past 100 years, and it is projected to continue to rise at a rapid rate due to human-induced increases in carbon dioxide levels (Houghton et al. 2001). Evidence for an effect of climate change on timing of breeding in North American birds breeding in the temperate zone is accumulating from long term data sets on multiple species around the globe. For instance, in response to increasing spring temperatures over the past few decades, Mexican jays (Aphelocoma ultramarina) and tree swallows (Tachycineta bicolor) have advanced their breeding dates by 9-10 days (Brown et al. 1999; Dunn and Winkler 1999). The trend towards earlier breeding in temperate-zone birds is expected to continue under conditions of increased seasonal duration of the growing season and more favorable spring climate (Kallander and Karlsson 1993; Visser 1998; Visser et al. 2006). Thus, further advances in the timing of breeding are expected as temperatures continue to rise. The advancement of breeding dates are subsequently expected to influence clutch size (Moller 2002), the rate of recruitment of young birds to populations, and population growth (Saether et al. 2000; Sillet et al. 2000), but spring storms and increased stochasticity may limit the ability of populations to increase their seasonal reproductive success (Martin and Wiebe 2004). Although 40 percent of the terrestrial planet is considered mountainous ( U N E P 2002), no studies have examined how responses of animals to climate change scenarios w i l l differ among elevations. Long-term data used to examine the advance in avian breeding dates are from low-elevation systems, and it is unclear whether birds breeding at higher-elevations should be expected t This paper is in preparation for submission to the journal Climate Change. H. Bears will be the solo author unless significant data, analytical or written contributions are added prior to submission 61 to react similarly. Differences in the responses of populations among elevations might be expected i f timing of reproductive efforts correlate with different weather variables at different elevations. Likewise, differences in responses of populations among elevations might be expected i f birds correlate reproductive efforts with similar weather variables, but climate change affects elevation zones differently (Lapp et al. 2005). Reproductive schedules of dark-eyed juncos breeding within high-(2000 m) and low-(1000 m) elevation habitats in Jasper National Park, Alberta, differ greatly (Fig. 2.1). Dark-eyed juncos breeding in high-elevation sites initiate their first brood 25-45 days later and ended their breeding seasons 15-20 days earlier than those at low-elevation sites. Similar to my study, a shorter seasonal reproductive period and corresponding lower reproductive success is often documented in birds breeding at increasing elevations (Krementz and Handford 1984; Hamann et al. 1989; Badyaev 1997; Perfito et al. 2004). Since photoperiod, the primary cue for reproductive development (Rowan 1925), is identical among elevations at the same latitude, differences in reproductive schedules may be related to differences in weather conditions (Table 1.1). Weather conditions and food supply can modify the effects of photoperiod and fine-tune reproductive timing in birds (Wingfield and Famer 1993; Wingfield et al. 2003). Colder temperatures can delay reproductive onset in birds, resulting in shorter breeding seasons, and fewer broods {e.g. Stevenson and Bryan 2000; Naef-Daenzer 2004; Torti and Dunn 2005; Weggler 2006). Snow cover in spring can delay reproductive onset in ground nesting species (Inouye et al. 2000; Hendricks 2003; Martin and Wiebe 2004). Spring storms can also delay egg-laying and induce reproductive failure (Wiebe and Martin 1995; Coulter and Bryan 1995). Finally, dates of egg-laying correlate with food emergence times in great tits and blue tits [Parus major and P. caeruleus; Svensson and Nilsson 1995; Thomas et al. 2001) and Eurasian dippers (Cinclus cinclus; Hegelbach 2001). The goal of the present chapter is to explore how weather and food availability correlate with brood initiation in the dark-eyed junco, breeding at high- and low-elevation sites in Jasper. The dark-62 eyed junco is a ground-nesting emberizid which has frequently been used in studies of life-history variation along environmental gradients (Nolan and Ketterson 1990; Bears 2002; Bears et al. 2003; Bears 2004), and Jasper National Park was an ideal study location because the abiotic cline at this latitude and elevation is particularly steep (Bears et al. 2003). Compared to low-elevation habitats in this region, high-elevation habitats are consistently colder, have a snowpack that does not melt until June (whereas low elevation sites do not have snow on the ground during the breeding months), have a delayed date of insect emergence, and receive rain at different times (Table 1.1). I ask two questions in this chapter that are necessary to forecast how climate change wi l l impact populations among elevations: (1) Which weather variables correlate with the cumulative percentage of broods initiated (i.e., timing of clutch initiation) in high- and low-elevation birds?, and (2) Do high- and low- elevation birds initiate broods in correlation with similar weather variables? 3.2 METHODS 3.2.1 Study Species and Study Site.- Dark-eyed juncos breed from sea-level to the sub-alpine treeline in North America. Adult dark-eyed juncos feed on insects, berries, and seeds, while hatchlings and young fledgling are mainly fed insects. Dark-eyed juncos in my study region nest on the ground (Nolan et al. 2002), and juncos at high-elevation sites in Jasper National Park often nest below ground in burrows or tunnels (Bears 2002). Dark-eyed juncos produce 1-3 clutches of 2-5 eggs, which are incubated by females only (Nolan et al. 2002). Males and females both feed and defend young, and males are more involved in territorial defense. Adult and juvenile dark-eyed juncos show philopatry in many regions (Nolan et al. 2002). Dark-eyed juncos were captured and monitored at eight 50 -70 ha study sites in Jasper National Park ( 5 2 ° 5 3 \ 118°3'; F ig . 1.1), Alberta, from 01 Apr i l to 20 August, 2000, 01 M a y to 20 August, 2001, 12 A p r i l to 30 August, 2004, and June 01 to July 30, 2005. Study species and sites are described in more detail in chapter 1. 3.2.2 Environmental variables.- To describe environmental conditions experienced by birds in high- and low-elevations of potential importance to their reproductive schedules, I analysed daily 63 minimum (°C), maximum (°C), and mean (°C) temperatures, wet precipitation (mm), snowfall (cm), and the depth of the snowpack at sites (cm) for the relevant months (Apri l - August) for 2000, 2001, and 2004. Insect biomass (mg) was collected in 2001 only, as described in Table 1.1. 3.2.3 Breeding Phenology Data.- Brood initiation data from chapter 2 (Fig. 2.1) are used as the response variable in analyses in this chapter. Field methods for detennining brood initiation dates are presented in detail in chapter 2. 3.2.4 Statistical Analyses. - Brood initiation data (number of clutches initiated by date) were converted to cumulative % brood initiation (cumulative % of total broods in the population initiated by date) at each elevation for each year of the study, as this measure allows one to view the cumulative, relative investment of the population over the season into breeding activities and to examine that investment relative to ambient conditions. Weather variables were also converted to cumulative variables. Temperature data was converted to growing degree days (GDD) , using the equation: [1] G D D = (Max Temp+Min Temp)/2 - Baseline Baseline values (value above which some activity, such as plant budding occurs) typically vary from 5-15 °G (Myers et al. 2004). To aid comparison with other studies, I used 10°C (Myers et al. 2004). The baseline w i l l not affect correlative conclusions, model formulations, or comparisons between elevations, which are the goals of the present paper. I examined the relationships between each weather variable and the cumulative percentage brood initiation data graphically using the curve estimation tool in SPSS (11). Snowfall data showed no relationship with cumulative % brood initiation as most of the snowfall occurred outside the brood initiation period. Thus, this variable was not analyzed further. Insect biomass (measured in 2001 only) was positively and linearly correlated with cumulative % brood initiation at both elevations. I fitted a regression line using the linear equation: [2] Cumulative % Brood Initiation = B(insectbiomass)+C 64 Where B is the slope and C is the intercept. A non-linear growth response was observed between rainfall and cumulative % brood initiation at both elevations, and between growing degree days and cumulative % brood initiation in low-elevation habitats. I fitted a non-linear regression models to these data using the equations: [3] Cumulative % Brood Initiation = C - A*exp(-B(cum. rainfall)) [4] Cumulative % Brood Initiation = C - A*exp(-B(growing degree days)) Where A , B and C are parameters estimated separately for each regression: A n exponential decay relationship between cumulative % brood initiation and snow depth in high-elevation habitats was observed. I fitted a non-linear regression model to these data using the equation: [5] Cumulative % Brood Initiation = B*exp(-snowdepth) Prior to parameter estimation, data residuals (e) were iteratively examined for autocorrelations between residuals at increasing distances from one another (i.e. between eo, and eo-i: eo-2, eo-3, etc). A s no significant autocorrelations between residuals were encountered (first-order r=0.55, p>0.08), autoregressive models were not necessary and standard linear and non-linear regression models were justified. To develop best-fit cumulative % brood initiation models relative to each of the variables, linear and non-linear regression functions in SPSS 11 were used to estimate parameters A , B , and C in each equation. (A, B , and C were estimated for each equation and do not represent the same parameters between equations). The residual sum of squares and corrected total sum of squares from the nonlinear regression analyses were used to calculate the coefficients of determination (R ) between the predicted model and the observed field data (Schabenberger and Pierce 2002). A s variables w i l l inevitably be correlated with one another, and thus the power values of the effect of each variable such as temperature, precipitation, snow depth, and insect biomass w i l l necessarily be inflated and should never be interpreted as strength of causality, and only as strength 65 of correlation when viewed as isolated variables. The 2-tailed a-value for evaluating significance was = 0.05. 3.3 R E S U L T S Partial correlation coefficients between weather variables at each elevation are summarized in Table 3.1, displaying inter-correlations between weather variables. Best-fit models for each weather variable in relation to cumulative % brood initiation are summarized in Table 3.2, along with their corresponding R 2 and p-value. In low-elevation habitat, the weather variables that were significantly correlated with the cumulative % of broods initiated were: growing degree days, insect biomass, and rainfall. Snow depth was not related to brood initiation in low-elevation birds as snow ground cover was absent during the breeding season in low-elevation habitats. In high-elevation habitat, the weather variables correlated with the percentage of broods initiated were: insect biomass, snow depth, and rainfall. Best-fit models for each variable and the cumulative percentage broods initiated are shown superimposed onto observed data in F ig 3.1. Weather variables that were elevation-specific in their relationship with the cumulative percentages of broods initiated were snow depth and growing degree days. Low-elevation birds increased the percentage of broods initiated with increasing growing degree days with a strong coefficient of variation (0.92). In contrast, high-elevation birds initiated almost all broods at the same, low growing degree day value (ca. 0, Fig. 3.1) and exhibited a spike in the cumulative percentage of broods initiated when the ground became mostly snow free. This relationship was the strongest distinguishing feature between responses of high- and low-elevation birds to weather variables and is depicted graphically in Fig . 3.2. Finally, when the models created from all years were tested on each year individually, R 2 values varied by only 2-3%, suggesting the models were robust for making predictions within the range of values examined. The R 2 values for rainfall were the most variable between years, ranging from 0.65-0.75 for low-elevation sites, and from 0.69-0.78 for high-elevation sites between years. 66 3.4 D I S C U S S I O N In this chapter, I demonstrated that (1) two supplemental cues for breeding (insect abundance, rainfall) were significantly correlated with the cumulative percentage of broods initiated by both high- and low-elevation birds populations, and (2) two supplemental cues for breeding (growing degree days for low-elevation birds and snow depth for high-elevation birds) were highly significant in predicting the cumulative percentage brood initiation, but were entirely elevation-specific. These findings suggest that changes in insect emergence dates and rainfall patterns w i l l affect populations similarly between elevations. However, changes in temperature w i l l likely have a profound effect on the productivity of low-elevation birds, while changes in snow depth and date of snow melt w i l l have a profound effect on the productivity of high-elevation birds. In other words, climate change may affect the productivity of high- and low-elevation birds in disparate ways. M y first finding was that the cumulative percentage of broods initiated was positively correlated with insect abundance and cumulative rainfall. I cannot deduce cause and effect from these analyses, nor can I treat the relationship between cumulative rainfall, insect abundance, and cumulative percentage of broods initiated independently, as rainfall and insect abundance co-vary more than other variables. However, initiating breeding attempts during periods where insect abundance is increasing appears to be universally important for insectivorous and omnivorous birds. Egg-development and incubation of eggs by females, as well as territorial defense of nest sites by males are costly duties, and a sufficient abundance of readily available calories must be available for birds from the onset of reproduction. A single egg can weigh up to 10% of the female body weight and is costly in terms of energy, nutrients, and future fitness. Nest construction and rearing of offspring also demand a high proportion of the daily energy budget for a breeding female (Monaghan et al. 1998; Stevenson and Bryant 2000). Producing hatchlings in periods with poor weather or insufficient food resources can lead to abnormal energetic expenses and/or death of parents and offspring (Thompson and Whitfield 1993; Thomas er al. 2001; Stenseth and Mysterud 67 2002), and so there is strong selective pressure for birds to breed later in response to constraints than to breed earlier despite them. Other authors have shown that dates of insect emergence correlate strongly with dates of egg-laying and seasonal productivity in great tits and blue tits (Parus major and P. caeruleus; Svensson and Nilsson 1995; Thomas et al. 2001) and in Eurasian dippers (Cinclus cinclus; Hegelbach 2001). I found that the relationship between insect biomass and cumulative percentage of broods initiated was stronger in high- relative to low-elevation birds. This is likely due to the highly synchronous laying period at high-elevations (chapter 2) that correlates with the delayed peak in insect abundance. It is interesting to note that high-elevation birds also began initiating broods when insects were more plentiful, which is likely the result of a synchronous eruption of insects in high-elevation sites, which birds correlate hatchling emergence with. L o w -elevation birds, on the other hand, breed throughout the season, with a smaller amplitude increase in broods initiated during the earlier, low-elevation peak in insect biomass; however, there is less of a peak in insect availability in low-elevation compared to high-elevation sites that birds can correlate reproductive effort with. I suspect that the strong correlation between cumulative rainfall and cumulative percentage brood initiation was not linked directly, but rather that rainfall affects birds through an effect on their food. Hematocrit values from birds in 2001 suggested that birds were not dehydrated in my study sites (Bears unpubl. data). M y second finding was that the cumulative percentage of broods initiated by high- and low-elevation birds showed strong correlations with different weather variables. Low-elevation birds showed a strong, positive, non-linear correlation with growing degree days. Therefore, for low-elevation birds, increases in spring temperatures predicted by climate change models can be expected to lead to a higher percentage of broods initiated earlier. Temperatures are often reported to affect reproductive onset in avifauna, whereby warmer temperatures result in longer breeding seasons, more broods, and an increase in seasonal reproductive success (e.g. Stevenson and Bryan 2000; Naef-Daenzer 2004; Torti and Dunn 2005; Weggler 2006). However, i f insect emergence is 68 unable to advance in parallel, low-elevation birds may become out of synch with food resources, and a mismatched resource supply and demand may result, leading to poor reproductive success or failure (e.g., Visser 1998; Visser et al. 2006; Stenseth and Mysterud 2002). High-elevation birds did not increase brood initiation in response to increasing growing degree days. In fact, high-elevation birds initiated most of their broods at a relatively low growing degree range, and instead had a strong non-linear negative relationship between snow depth and cumulative percentage of broods initiated {i.e. most nests were initiated in high-elevation habitats as the ground became snow free). Persistent snow cover in spring has been speculated to be a primary factor delaying reproductive onset in other bird species (Inouye et al. 2000), and snow pack is thought to be important in driving the annual variation in mean laying date in the American pipit (Anthus rubescens; Hendricks 2003), which also requires snow-free ground to begin nesting. Climate change, therefore, may impact the reproductive phenology of low-elevation birds via temperature increases, and high-elevation birds through alterations to spring snowpack. Luckmann (1998) summarized available weather data from the central Rocky Mountains, and showed some strong effects of climate change in this region. Mean annual temperatures have increased by 1.4 °C over the last 100 years, and the greatest increases have been in winter temperatures. A tree-ring-based temperature reconstruction indicated that summer and spring temperatures in the last half of the 20th century were higher than any period over the last 900 years. Glaciers have also lost ca. 25% of their volume in the last 100 years, and are smaller now than they have been at any time in the last 3000 years. Thus, in the Rocky Mountains, significant climate change has occurred during the late 20th century that is ongoing and exceptional in the context of the last 1000 to 3000 years. A s growing degree days reach higher values earlier in the season, advancements in brood initiation in low-elevation birds are expected. In high-elevation environments in the Rocky Mountains, on the other hand, climate change models predict an increase in winter precipitation, and high-elevation snowpack (Lapp et al. 2005). However, warmer temperatures often 69 increase the rate of snowmelt, and lengthen the midsummer period (Harte et al. 1995; Inouye et al. 2000). Therefore, the productivity of high-elevation birds may be affected in either direction. When increases in spring temperatures are unable to melt a larger snow pack remaining from the preceding winter, reproduction is expected to be delayed and productivity negatively impacted. However, once spring temperatures have increased sufficiently to melt the snow more rapidly despite a potentially greater volume in some years, reproduction is expected to occur earlier and productivity should be positively impacted. Again, the capacity for insect emergence to track such climatic changes in parallel with breeding dates w i l l also determine the ultimate success of a shifted breeding schedule. In this chapter I showed the necessity for considering impacts of climate change on phenology of vertebrate reproduction with considerations of breeding elevation. Reproductive investment was not simply correlated with the same weather variables in high- and low-elevation populations. Instead, breeding efforts at either elevation were correlated differently with some variables. Even though variables measured were inter-correlated, growing degree days and snow depth were entirely elevation-specific in their predictive values, and as such are highly useful variables to use for elevation-specific predictions about cumulative percentage of broods initiated. I f the results of my study are similar for other ground nesting bird species, it may be important to consider high- and low-elevation populations separately when predicting the impacts of climate change on avian phenology and productivity. However, I am still unable to say whether these variables are causally related or simply auto-correlated with breeding efforts. A n alternative hypothesis to the assumption that brood initiation in populations adjusted in response to weather variables is that birds have adapted to respond to the increases in spring photoperiod differently between elevations, with populations responding to different "critical daylengths" (Hamner 1966; Wingfield et al. 1992; Silverin et al. 1993). Differing critical daylengths between elevations could create spurious correlations with weather that are correlated with photoperiod. If this latter mechanism were occurring, it would explain why some populations are 70 unable to move their breeding dates in response to changing weather patterns (Giennap et al. 2006), and explain why I saw little annual variation in phenology despite larger scale differences in weather. In chapter 4, therefore, I conduct a common aviary experiment to test whether populations from each elevation respond to different critical daylengths in order to initiate reproduction. 71 Table 3 . 1 Pearson correlation coefficients of four potential supplementary cues measured in high-and low-elevation habitats in Jasper National Park, Alberta, Canada. Variables are: S N O W = snow depth, IN= insect biomass, R N = cumulative rainfall (mm), GDD=growing degree days. Coefficients significant at a = 0.05 are marked with *, and those significant at 0.01 are marked with **. — means the correlation could not be measured H i g h - Elevation Low-Elevat ion S N O W I N R N G D D S N O W I N R N G D D S N O W 1 -0.91* -0.65** -0.35** 1 I N 1 0.90* -- 1 0.95* 0.95* R N 1 0.73** 1 0.89** G D D 1 1 72 Table 3.2 Best-fit regression models for each variable examined (X) relative to cumulative percentage of broods (Y) initiated in high- and low-elevation habitats. For all parameters estimated, the asymptotic standard error was less than 3% of the estimated parameter value. Variable(X) High- elevation Low-elevation (R 2 andp) (R 2 and p) Insect Biomass Y = 0.87(X)- 109.42 Y = 0.89(X)+ 13.82 (R 2 = 0.93;p=0.01) (R 2 = 0.77; p=0.01) Cum. Rainfall Y = 113.77-112.96*exp(-0.008(X)) Y = 113.33-107.64*exp(-0.009(X)) (R 2 = 0.87; p=0.001) (R 2 = 0.78;p=0.001) Growing Y=(97.48)-(78.68)*exp(-0.03(X)) Degree Days N o Relationship (R 2=0.92;p<0.0001) Snow Depth Y=73.3*exp(-X) (R 2=0.77; pO.OOOl) N o Relationship 73 Low High 120 100 -80 -6 0 -4 0 -2 0 0 40 ^ Observed Predicted 7 7 ^ 60 80 100 120 0 I n s e c t B i o m a s s ( m g ) 150 2 0 0 • • 400 0 50 C u m . R a i n f a l l ( m m ) 100 150 200 250 4 » — 0 50 100 S n o w D e p t h ( c m ) * • I T 150 0 50 100 150 200 250 300 G r o w i n g d e g r e e d a y s Fig. 3.1 Actual (observed; dark diamonds) and predicted data (lighter squares; best fit regression calculated by parameterization for each data set; linear for insect data and non-linear for other variables). Graphs show the relationship of each weather variable (x) to the cumulative percentage of broods initiated (y) by dark-eyed juncos in low- (left column) and high- (right column) elevation habitats. 74 F i g . 3.2 Graphical representation o f the different relationship between cumulative percentage brood initiation, snow depth (cm), and growing degree day, in high- and low-elevation birds. High-elevation birds all breed at a relatively low growing degree day values, and display a sharp peak in brood initiation when the snowpack melts to a value close to zero. Low-elevation birds, on the other hand do not experience snow cover during the breeding season, and display a strong non-linear correlation with growing degree days. 75 3 . 5 R E F E R E N C E S Badyaev, A . V . 1997. Avian life history variation along altitudinal gradients: an example with cardueline finches.- Oecologia 111 365-374. Bears, H . 2002. How and why a small songbird, Junco hyemalis, breeds over a steep elevation gradient. M S c . thesis, University of British Columbia. Bears, H . , Smith, J . N . M . , and Wingfield, J .C. 2003. Adrenocortical sensitivity to stress in Dark-eyed Juncos (Junco hyemalis oregonus) breeding in low and high elevation habitat. - Ecoscience 10: 127-133. Bears, H . 2004. Parasite prevalence in Dark-eyed Juncos, Junco hyemalis, breeding at different elevations. - Can. Field Nat. 118: 235- 238. Brown, J .L. , Shou-Hsien, L . , and Bhagabati, N . 1999. Long-term trend toward earlier breeding in an American bird: A response to global warming? - Ecology 9 6 : 5565-5569. Coulter, M . C , and Bryan, A . L . 1995 Factors affecting reproductive success of wood storks (Mycteria americand) in east-central Georgia. - Auk 112: 237-243. Dunn, P.O., and Winkler, D . W . 1999. Climate change has affected the breeding date o f tree swallow throughout North America. - Proc. Biol. Sci. 2 6 6 : 2487-2490. Gienapp, P., Postma, E. , and Visser, M . E . 2006. Why breeding time has not responded to selection for earlier breeding in a songbird population. - Evolution 6 0 : 2381-2388. Hamann H . J., Schmidt, K . H . , and Simonis, S. 1989. Elevation and size of eggs and clutch in the great tit Parus major. - J. Fiir Ornithol 1 3 0 : 69-74. Hamner, W . M . 1966. Photperiodic control of the annual testicular cycle in the House Finch demonstrated by interrupted-night experiments. - Nature 2 0 3 : 1400-1401. Harte, J. and Shaw, M . R . 1995. Shifting dominance within a montane vegetation community: results of a climate warming experiment. - Science 2 6 7 : 876-880. Hegelbach, J. 2001. Water temperature and phytophenology indicate the earlier onset of oviposition in Eurasian Dipper (Cinclus cinclus) from the Swiss Lowlands. - J. Fiir Ornithol. 142: 284-294. Hendricks, P. 2003. Spring snow conditions, laying date, and clutch size in an alpine population of American Pipits. - J. Field Ornithol. 74:423-429. Houghton, J. 2001. The science of global warming. - Interdisc. Sci. Rev. 26: 247-257. Inouye, D .W. , Barr, W . A . , Armitage, K . B . , and Inouye, B . D . 2000. Climate change is affecting altitudinal migrants and hibernating species. - PNAS 97: 1630-1633. Kallander, H . , and Karlsson, J. 1993. Supplemental food and laying date in the European Starling. - Condor 95: 1031-1034. Krementz, D . G . , and Handford, P. 1984. Does avian clutch size increase with elevation? -Oikos 43: 256-259. Luckmann, B . H . 1998. Landscape and climate change in the central Canadian Rockies during the 20th century. - Can. Geogr. 42: 318-336. Moller , A . P . 2002. North Atlantic Oscillation ( N A O ) effects of climate on the relative importance of first and second clutches in a migratory passerine bird. - J. Anim. Ecol. 71 : 201-2002 Monaghan P., Nager R . G . , and Houston, D . C . 1998. The price of eggs: increased investment in egg production reduces the offspring rearing capacity of parents. - Proc. Roy. Soc. B 265: 1731-1735. Myers, M . W . , Curran, W.S. , VanGessel, M . J . , Calvin, D .D. , Mortensen, D . A . , Majek, B . A . , Karsten, D .D . , and Roth, G .W. 2004. Predicting weed emergence for eight annual species in the northeastern united states.- Weed Sci. 52: 913-919. Naef-Daenzer L , Nager, R . G . , Keller, L . F . , and Naef-Daenzer, B . 2004. Are hatching delays a cost or a benefit for Great Tit Parus major parents? - ARDEA 92: 229-237. 77 Nolan, V . and Ketterson, E . D . 1990. Timing of autumn migration and its relation to winter distribution in dark-eyed juncos. - Ecology 71: 1267-1278. Nolan, V . , Jr., Ketterson, E . D . Cristol, D . A . , Rogers, C M . , Clotfelter, R.C.,Titus, R . C . , Schoech, S.J., and Snajdr, E . 2002. Dark-eyed Junco (Junco hyemalis): In The Birds of North America, No. 716 (Poole, A . , and Gi l l s , F., Eds.). Academy of Natural Sciences, Philadelphia, and American Ornithologists Union, Washington. Perfito, N . , Tramontin, A . D , Meddle, S., Sharp, P, Af ik , D . , Gee, J., Ishii, S., Kikuchi , M . , and Wingfield, J.C. 2004. Reproductive development according to elevation in a seasonally breeding male songbird. - Oecologia 1 4 0 : 2 0 1 - 2 1 0 . Rowan, W. 1925. Relation of light to bird migration and developmental changes. - Nature 1 1 5 : 494-495. Saether, B . E . 2000. Population dynamical consequences of climate change for a small temperate songbird population. - Science 287: 854-2000. Schabenberger, O. and Pierce, F.J. 2002. Contemporary Statistical Models for the Plant and Soil Sciences. Boca Raton, F L : C R C . Pp. 185-213. Sillet, T.S. 2000. Impacts of a global climate cycle on population dynamics of a migratory songbird. - Science 288 : 2040 2000 Silverin, B . , Massa, R., and Stokkan, K . A . 1993. Photoperiodic adaptation to breeding at different latitudes in great tits. - Gen. Comp. Endocrin. 9 0 : 14-22. Stenseth, N . C . , and Mysterud, A . 2002. Climate, changing phenology, and other life-history traits: Non-linearity and match-mismatch to the environment. - PNAS 99: 13379-13381. Stevenson, I.R., and Bryant, D . M . 2000. Avian phenology- climate change and constraints on breeding. - Nature 4 0 6 : 366-367. Svensson, E , and Nilsson, J .A. 1995. Food- supply, Territory quality, and Reproductive Timing in the Blue Tit (Parus Caeruleus). - Ecology 76: 1804-1812. Thompson, D . B . A . , and Whitfield, D.P. 1993. Research on mountain birds and their habitats: research progress report. - Scottish Birds 17: 1-8. Thomas, D .W. , Blondel, J., Perret, P., Lambrechts, M . M . , and Speakman, J.R. 2001. Energetic and Fitness Costs of Mismatching Resource Supply and Demand in Seasonally Breeding Birds. - Science 291: 2598-2600. Torti, V . W . , and Dunn, P.O. 2005. Variable effects of climate change on six species of North American birds. - Oecologia 145: 486-495 U N E P World Conservation Monitoring Centre, Mountain Watch. 2002. Pages 14-15 in Environmental Change and Sustainable Development in Mountains. 80 pages. [http://www.unep-wcmc.org/]. Visser, M . E . 1998. Warmer Springs lead to mistimed reproduction in great tits (Parus major). -Proc. Roy. Soc.B 265: 1867-1970. Visser, M . E . , Holleman, L . J . M . , and Gienapp, P. 2006. Shifts in caterpillar biomass phenology due to climate change and its impact on the breeding biology of an insectivorous bird. -Oecologia Ul: 164-172. Wilson, S., and Arcese, P. 2003. E l Nino drives timing of breeding but not population growth in the song sparrow (Melospiza melodia). P N A S 100: 11139-11142. Weggler, M . 2006. Constraints on, and determinants of, the annual number of breeding attempts in the multi-brooded Black Redstart Phoenicurus ochruros. - IBIS 148: 273-284. Wiebe, K . L . and Martin, K . 1995. Physiological and environmental effects on laying rates in ptarmigan. - Condor 97: 708-717. Wingfield, J . C , Hahn, T.P., Levin, R., and Honey, P. 1992. Environmental predictability and control of gonadal cycles in birds.- J. Exp. Zool. 261: 214- 231. Wingfield, J .C. and Farner, D.S. 1993. Endocrinology of reproduction in wi ld species. Pages 163-327, in Avian Biology, V o l . 9 (Farner, D . S., K ing , J.R. and Parks, K . C . , Eds.). Academic Press, San Diego, California. Wingfield, J . C , Hahn, T.P., Maney, D . L . Schoech, S.J., Wada, M . , and Morton, M . L . 2003. Effects of temperature on photoperiodically induced reproductive development, circulating plasma leuteinizing hormone and thyroid hormones, body weight, fat depositio and molt in mountain white-crowned sparrows, Zonotrichia leucophrys oriantha. - Gen. Comp. Endocrinol. 131: 143-158. 80 4 C O M P A R A T I V E R E P R O D U C T I V E P H E N O L O G Y O F D A R K - E Y E D J U N C O S B R E E D I N G A T T W O E L E V A T I O N S : A C O M M O N A V I A R Y E X P E R I M E N T * 4.1 I N T R O D U C T I O N In birds breeding in the temperate zone, seasonal reproduction tends to occur when factors like food supply and ambient conditions facilitate survival of parents and offspring (Thomson 1950; Lack 1968). Accurately timed reproduction is a life-history trait that is subject to strong natural selection, as the feasible period for breeding is often short in the annual cycle (Murton and Westwood 1977), and mistimed reproduction can be energetically costly and may lead to the death of parents and/or offspring (Thomas et al. 2001). The annual change in daylength is an accurate signal for the transition between winter and spring, so it is often used as the primary proximate cue to initiate the physiological, morphological, and behavioral changes that must occur in advance of breeding (Farner and Lewis 1971; Dawson et al. 2001). Daylength increases in the spring until a 'critical daylength' ( C D L ) is reached (Hamner 1966), at which point the reproductive axes of photoperiodic birds are stimulated, and a cascade of endocrine events occur that leads to the development of gonads and the activation of reproductive behaviors. Eventually, birds become unresponsive to the stimulatory effects of long days, reproduction ceases, gonads regress (gonadal recrudescence), and molting begins, a stage referred to as photorefractoriness (Farner et al. 1983; Nicholls et al. 1988). The critical daylength required for the initiation of reproductive development can differ among species and populations within a species depending on local conditions, especially between latitudes where photoperiods shift considerably (Silverin el al. 1993; Lambrechts et al. 1996). In addition, supplementary cues such as temperature, food supply, weather conditions, or behavioral stimuli can act to modify the effects of photoperiod and to fine-tune reproductive timing (Wingfield * This chapter is in preparation for submission to the Journal of Animal Ecology. H. Bears will be the solo author unless significant additional data, analytical, or written contributions are added prior to submission 81 et al. 1992; Perfito et al. 2004). Differences in the adjustment of reproduction to different environmental conditions can either be the result of genetic differences in how birds respond to thresholds (e.g., birds adapt to respond to different critical daylengths or temperatures), or the result of phenotypic plasticity (e.g., birds respond directly to supplementary cues such as temperature to adjust the temporal progression of reproduction). Phenotypic plasticity is also often based on reaction norms that are themselves subject to natural selection (Pigliucci and Schlichting 1998). Consequently, identifying the effects of genetic and environmental sources of phenotypic variation is important for understanding population differences in reproductive timing, and for predicting how climate changes w i l l affect avian reproductive schedules and productivity between localities. A comparative study of annual reproductive cycles of free-living populations of dark-eyed juncos (Junco hyemalis) inhabiting sub-alpine (2000 m 'High ' elevation) or montane (1000 m ' L o w ' elevation) habitat revealed substantial differences in the temporal organization of reproductive activities with elevation, despite both populations experiencing identical photoperiods (Appendix II). Differences in egg-laying schedules between elevations were also tied to differences in the timing of development of reproductive structures (chapter 2). Birds of both sexes breeding within high-elevation habitat developed their gonads (or brood patches in the case of females) > 4 weeks later in the spring than their low-elevation conspecifics, and began egg-laying ca. 4-6 weeks later. Since both populations regressed gonads at about the same time in mid-summer, high-elevation birds terminated reproductive attempts after the completion of their initial attempt and progressed through molt more rapidly. Mark-recapture analyses suggested that both high- and low-elevation populations are philopatric, and therefore populations may be influenced by natural selection for locally adaptive traits, although relative rates of emigration are unknown (chapter 2). High-elevation populations experience extremely different abiotic and biotic conditions during the breeding season compared to low-elevation populations, although the photoperiod changes they experience are identical. Conditions that differ at the high-elevation habitat include 82 persistent snow cover, frequent early season storms, colder temperatures, different temporal schedules of rainfall, and a later peak in insect biomass abundance (Table 1.1). The compressed breeding season of high-elevation birds may either be the result of evolutionary changes that have evolved during the process of adapting to breed successfully within high-elevation habitat (e.g. high-elevation populations adapting to respond to longer critical daylengths than low-elevation populations). Alternatively, the compressed breeding season may be the result of phenotypic plasticity of individuals exposed to the different environmental conditions within low- and high-elevation habitats. The goal of this chapter was to differentiate between genetic vs. environmental causes for the different temporal reproductive schedules between elevations and to elucidate how and to what degree they contribute to the observed differences in the timing of reproduction. I performed a rearing experiment to infer likely causes of differences in timing of reproductive features observed between elevations. Birds from 2 elevations were reared under conditions of unlimited food supply, termed a 'common aviary' (analogous to common garden experiments in botany), and timing of the development of reproductive traits were compared between field and aviary environments. With the exception of one other study on adult male song sparrows, Melospiza melodia morphna (Perfito et al. 2004), such experiments have not been performed on songbirds to help determine the causes of phenotypic differences associated with elevational gradients. Reproductive traits of birds caught at different elevations may change in a number of ways when brought into the aviary (Fig. 4.1a-d). First, differences in reproductive timing development observed in the field may remain in the lab, implying genetic differences, or differences that were set early in development and prior to collection (i.e. early perinatal or maternal effects) that cause each population to responds to a different critical daylength (Fig. 4A) . Second, differences in reproductive timing may disappear in the lab, implying that birds to respond to secondary environmental cues to time reproductive development in the wild, and are not genetically different (Fig. 4B). Third, differences in reproductive timing are exacerbated in the lab, implying 83 genetic or persistent early developmental differences combined with different environmental constraints between elevations (Fig. 4C). Finally, the timing of reproductive development could be inverted in the lab, implying that strong countergradient effects, which are either genetic or perinatal/maternal and influence the bird's phenotype early in development, are present in high-elevation populations in order to counteract the relatively strong environmental constraints. Thus, when environmental influences are removed in the lab, I might see an inversion (Fig. 4D). 4.2 METHODS 4.2.1 Captive Birds and Experimental Design- In 2004, dark-eyed juncos were collected and fitted with individually identifying number bands as early fledglings (ca. 9-11 days of age, just after leaving the nest but before they could fly) from different nests in sites just outside Jasper National Park boundaries (see Chapter 2, Fig . 1.2). I collected only one bird per nest in territories at least 500 m apart. Adult ( A S Y ) juncos were also collected to enable a comparison of birds experiencing their first breeding season in captivity versus those that had a prior breeding season in their elevation of origin, which may have worked to alter their reproductive schedules to match their breeding elevation. High-elevation juncos that were collected all hatched between 21 June and 07 July, and low-elevation juncos that were collected all hatched between 20 May and 15 July. Birds were held in an outdoor aviary (6 m x 1.5 m x 2 m) at the Palisades Research Centre (1020 m asl) in Jasper until young fledglings reached independence by 30 days, at which point birds were transported to the University of British Columbia Animal Care facility in Vancouver and placed in a roofed, % walled semi-outdoor aviary (10 m x 15 in x 10 m). Birds flew freely and were given trees, perches, roosting shelters, and heat lamps from 01 Sept to 31 Dec, 2004. Eight food and water dishes (1 in x 1 m) were provided to eliminate competition for resources. Juncos were fed a veterinarian recommended diet ad libitum, which consisted of fresh berries, soft Science-diet© cat food, cooked egg, suet, Haitz © vitamin/mineral enriched seed mix, and soft-bodied crickets. Prime © amino acid 84 powder was added to water daily. Charcoal, oyster shells, and grit were sprinkled on the aviary floor as digestive aids. On 01 January, 2005 when birds had finished molting, they were caged individually (Gina© cages; 33.02 cm x 45.72 cm x 58.42 cm) and transferred to a photoperiod (full spectrum natural light; 8L: 16D) and temperature-controlled (19 °C) room at U B C ; their diets remained the same as in the aviary. This temperature was chosen because it would be experienced during the reproductive period at both elevations in Jasper. B y the time the experiment was initiated in January, the birds that had originally been collected at 9-11 days old had matured to full grown "second year" birds in captivity, and hence I refer to them as the S Y age group hereafter. The adult birds are classified as after second year ( A S Y ) age class. The initial photoperiod represented short day lengths that would be experienced during winter solstice. A l l individually-caged birds were housed together in a single chamber such that all aspects of the "environment", including the social environment, were comparable for every bird. Males and females could see and hear one another, but could not make physical contact. Beginning on 02 Jan, 2005, the photoperiod was increased at a rate of 3.86 min per day, similar to the rate in nature during the spring, until a photoperiod of 17L:7D was reached 140 days later (15 May) , which approximates the summer solstice day length in Jasper. After this point, the day length was kept constant for an additional 40 days to allow birds to become desensitized to long days, which leads to eventual gonad recrudescence and molt. The photoperiod regime prescribed is shown in Figure 4.2. Procedures were done with appropriate permits ( U B C Animal Care # A04-0018; Jasper National Park #518, 038, 008, 005; Canadian Wildlife Service CWS04-A001; Environment Canada #2000/067, WSA-6/00) . 4.2.2 Sampling techniques- For males, the width of the cloaca was measured weekly, as this measure closely approximates the increase in testis volume for this species (Deviche et. al. 2000), but does not require surgery, minimizing disturbance to the birds. Further, this was the reproductive measure taken from free-living birds in the field (chapter 2). I also recorded the number of songs per 85 minute in males every 2 weeks after the lights came on in the morning as a behavioural indicator of reproductive readiness (Dloniak and Deviche 2001). For song monitoring, I entered the bird housing unit and sat in the middle of the room such that I could view all caged birds. I allowed birds to adjust to my presence for 5 minutes prior to starting song counts, which continued for 15 minutes uninterrupted where I recorded how often each bird sang. Every three to five weeks, female birds were laparotomized under light Isoflurane anaesthesia to assess gonadal size (Wingfield and Farner 1976) with the aid of a veterinarian. During laparoscopics, female ovaries were scored on a scale of 1 - 5 as in MacDougall-Shackleton et al. (2001), where l=smooth, with no visible follicular development, 2=slightly granular appearance, 3=initial development of follicular hierarchy, 4= obvious follicles with evident hierarchy, 5=large yolky follicles. Incisions were treated with antiseptic and sealed with Histoacryl. Ovarian measurements were limited by the availability of a veterinarian skilled in songbird laparoscopies, and thus no surgeries could be done after 01 M a y when veterinarian aid was unavailable. Female brood patch development was also measured every 2-weeks for the duration of the experiment, as it is an indicator of oestradiol and prolactin action that signals reproductive readiness (Baily 1952; Enstrom et al. 1997). Brood patches were scored weekly from 1-4, where 1= No brood patch formation, 2 = Partial feather loss/brood patch formation, 3 = Ful ly developed brood patch, 4 = Formerly active brood patch that is re-feathering. Recovery from surgery was rapid and incisions healed quickly. Weight (g) and intrafurcular fat score (0-5; see chapter 2) were also measured weekly to ensure animals were maintaining their body weight, and are analysed in chapter 5. Molt scores on the right wing of males and females (see chapter 2) were also assessed at the time of surgeries as an indicator of reproductive termination. To compare the timing of reproductive capacities between high- and low-elevation birds, I used threshold values of gonads. In males, the onset and the end of 'functional' gonads were defined as the dates at which testicular widths reached Vi of their maximum value. This threshold value was selected on the assumption that testes in the cloaca start producing sperm at ca. half-maximum 86 volume (Gwinner 1986). These dates were calculated by interpolation of individual fitted curves for each experimental bird. In females, ovary scores could only be compared at equivalent time periods, and a precise onset of mature gonads could not be defined or determined due to less frequent sampling. For a comparison of reproductive timing between free-living and experimental juncos, 1 used data for the timing of reproductive activity (chapter 2). The experiment began with 16 high-and .15 low- males, and 15 high- and 15 low-elevation females, but with mortality these numbers were reduced to 14 high-elevation males (8 " S Y " males raised from hatchlings and 6 " A S Y " males collected as adults) and 14 low-elevation males (8 " S Y " and 6 " A S Y " ) , and 1.0- high-elevation females (6 " S Y " and 4 " A S Y " ) and 1 M o w elevation females (6 " S Y " and 5" A S Y ) by the end of the experimental period. Birds that died during the study period were not included in analyses because I used repeated measures to track caged individuals over time. 4.2.3 Statistical Analyses- A l l statistical analyses were performed with SPSS 11 (SPSS Inc., Chicago, IL) with reference to Sokal and Rohl f (1995). Changes in gonad size, male song frequency or follicle score over time were assessed using repeated-measures analysis o f variance ( r m A N O V A ) with time as a within-subject variable and elevation of origin (low- or high-) and age group ( H Y and A S Y ) as between-subject factors. As these data violated sphericity assumptions as determined by examining Mauchly's W statistic, I used the Huynh-Feldt epsilon correction to assess values for any effect tested. If the overall between-subject effects were significant in r m A N O V A s , Tukey's b post hoc comparisons were done. Data are presented as means ±1SE unless otherwise indicated. Eta-squared values (eta^), which represent the proportion of variance of the dependent variable explained by an independent variable, are included as a standard measure of effect size. Strengths of eta2 values are interpreted according to Cohen (1988) where 0.01=small effect, 0.06=moderate effect, and 0.14=large effect. F-values are presented as F(df(effect)/df(error)) with Huynh-Feldt epsilon corrected df values. Tests were two-tailed and the significance level set at a=0.05. 87 4.3 R E S U L T S 4.3.1 Cloacal Protuberance Size - Within-Groups Contrasts (Where Groups = "Low SY", "Low ASY", "High SY", "High ASY"): The photoperiod increase in the laboratory had a large effect on cloacal size in males over time, which began to increase in late February/early March until mid-Apri l , stabilized at a maximum size from mid-Apri l to the end of May, and then began to decline again in all groups [F(5, 100)= 1464, p<0.001, eta2=0.99, p o w e i - 1.00]. There was a strong two-way interaction between time and elevation of origin [F(5, 100)= 8.35, p<0.0001, eta2=.29, power= 0.99], with high-elevation groups increasing their cloacal protuberances earlier than low-elevation groups, when average differences at each time period up to functional size (50% of maximum cloacal size) are considered. There was also an interaction between time and age class, with older birds growing larger cloacal protuberances earlier than younger birds [F(5, 100)= 2.85, p=0.02, eta2=0.13, power= 0.82]. There was also a moderate 3-way interaction between time, elevation of origin, and age [F(5, 100)=3.07, p=0.01, eta2=0.13. powei-0.86] where S Y males at low-elevations had smaller cloacas relative to all other groups (Fig. 4.3). Between-Groups Contrasts: There was a large effect of elevation of origin on cloacal protuberance size, with high-elevation birds having larger cloacal protuberances compared to low-elevation birds at multiple comparable time intervals [F(.l,20)= 14.13, p=0.()01, eta2=0.41, power=0.95]. There was also a large affect of age on cloacal protuberance size in this experiment, with older birds having larger cloacal protuberances at many set time intervals than younger birds [F(l,20)=5.23, p=0.03, eta2=0.21, power=0.59]. Males of all groups began the regression of their gonads at similar times (Fig 4.3); however I was unable to monitor the entire gonadal regression period prior to termination of the experiment on 10 June, when field studies began. 4.3.2 Funct ional Gonads- The average dates (±1SE) by which functional gonads had developed (defined as the mean date at which 50% of maximum size reached) in males collected as adults and as hatchlings from high-elevation habitats were 18 Mar(±2d) and 19-Mar(±2d), respectively. The average dates (±1SE) by which functional gonads had developed in males collected as adults and as hatchlings from low-elevation habitats were 21 Mar(±4d) and 28 Mar(±4d), respectively (Fig. 4.4). In general, birds that originated from high-elevation habitats developed functional gonads slightly earlier than those from low-elevation habitats. High-elevation males shifted their gonad development time from >30 days later than low-elevation birds to 4 or 8 days earlier in the lab, which is a 34 to 38 day shift in timing between high- and low-elevation populations. In the same measurements taken from males living in the field (Chapter 2), cloacas reached a functional size ca. > 4 weeks (30 days) later in male juncos living within high- compared to low-elevation habitats (Fig. 4.4). In other words, the large difference in the timing of functional gonad development observed between high- and low-elevation males in the field was greatly reduced in the lab; however low-elevation males developed first in the field, while high-elevation males developed first in the lab. Males of S Y age originating from low-elevations had the latest development of functional gonads. 4.3.3 Song Frequency- Within- Subjects Contrasts: The photoperiod increase in the laboratory over time had a large effect on song frequency in males, with increases in song-frequencies from late-February, which began to decrease again in mid-May [F(6.6, 131.7)—731.63, pO.0001 , eta2=0.97, power= 0.99]. There was a strong two-way interaction between time and elevation of origin, with high-elevation birds generally increasing their song-frequencies earlier than low-elevation birds [F(6.6, 131.7)=9.84, p<0.0001, eta2=0.33, power=0.99]. There was a large two-way interaction between time and age, with older birds increasing their song frequency earlier than younger birds [F(6.6, 131.7)=6.76, p<0.000.1, eta2=0.25, power= 0.99]. There was also a strong three-way interaction between time, elevation of origin, and age, again with the young low-elevation group being the slowest to increase song frequency to the photoperiod regime [F(6.6, 131.7)=3.45, p=0.002, eta =0.15, power=0.95] (Fig.4.5). Between-Subjects Contrasts: There was a large effect of elevation of origin on song frequency, with high- elevation birds singing more frequently than low-89 elevation birds at some set time intervals [F(l,20)=28.73, p=0.001, eta2=0.59, powei-0.99]. There was also a large effect of age, with older low-elevation birds singing more at some set time intervals than younger low-elevation birds [F(l,20)=8.20, p=0.001, eta2=0.29, power=0.78]. A l l groups of males decreased the frequency of their morning chorus at similar times among groups (Fig. 4.5). 4.3.4 Ovary Score.- Within-subjects contrasts'. The photoperiod increase in the laboratory had a large effect on ovary development, with some degree of ovary development observed by 01 March, which increased at subsequent sample points in all groups thereafter [F(3.2,51.5)=79.19, p<0.0001, eta"=0.82, power=l .00]. A moderate two-way interaction between time and age was also observed, with older high-elevation female birds developing their ovaries earlier than younger female birds [F(3.2, 51.5)=2.75, p=0.05, eta2=0.15, power = 0.65] (Fig. 4.6). Between-Subjects Contrasts: There was a large effect of age on ovary development, with older females having more developed ovaries than younger females at one or more comparable sampling intervals 2 [F(l,16)=8.36, p=0.01, etaz=0.34, power= 0.77], Elevation of origin did not have a significant influence on ovary development (p>0.30), but the power for this tests was very low (power=0.18), and so I may have been unable to detect this effect size as statistically different with the sample size investigated. Use of a larger sample size may have rendered a significant result. No significant interactions between age and elevation were detected (p>0.90), but the power for this test was again low (power=0.10). Females laid eggs on some occasions (3 observations), as they were found on the bottom of feeding trays. 4.3.5 Brood Patch . - Within- Subjects Contrasts: When brood patch scores were averaged for each group at each time-interval, the photoperiod increase in the laboratory over time had a large effect on brood patch development, causing advancement of brood patch score by early to late March in all groups [F(10, 161)=427.6.1, p O . O O l , eta :=0.97, power=0.99]. There was a large effect of elevation of origin [ F ( l , 16)=30.42, p=0.001, eta2=0.66, power=0.99] on brood patch development, with high-elevation females reaching higher brood patch scores earlier than low-90 elevation females. There was a large effect of age on brood patch score, with older females from both habitats developing brood patches earlier than younger females [F(l,16)=4.90, p=0.04, eta^=0.24, power=0.55]. There was also a large two-way interaction between time and elevation of origin [F(10, 161)=6.82, p=0.019, eta2=0.30, power=0.68] on brood patch development. No other two- or three-way interactions were observed (all p>0.10). Betiveen-Subjects Contrasts: There was ; large effect of elevation of origin on brood patch formation, with high-elevation females having more advanced brood patches than low-elevation females at multiple set time intervals [F(l,16)=30.42, pO .001 , eta2=0.66, power=0.99; Fig. 4.7]. There was also a large effect of age on brood patch development, with older females having more advanced brood patches at multiple comparable sample intervals compared to low-elevation females [F(l,16)=4.90, p=0.04, eta2=0.04, powei-0.55; Fig . 4.7]. There was no significant interaction between age and elevation (p=0.33). 4.4 D I S C U S S I O N In this chapter, I examined the underlying basis for the strong variation in reproductive timing and seasonal development of reproductive structures between dark-eyed juncos breeding within high- and low-elevation habitats shown in chapter 2. The substantial differences in reproductive timing between free-living high- and low-elevation male and female dark-eyed juncos (chapter 2) were greatly decreased in captive first and second year breeding males and females that were exposed to identical environmental conditions in the laboratory and exposed to a controlled treatment of increasing photoperiod. Experimental males and females originating from high- and low-elevation habitats were more similar in their timing of maturation and reproductive development in the lab than in the field (Fig. 4.D). Both sexes altered the timing of reproductive maturation in the common environment (more similar between populations and inverted), and the order in which populations matured were reversed between the lab and the field in both sexes (i.e., low-elevation birds matured earlier in the field, while high- elevation birds matured earlier in the lab). The reversal that occurred in the lab meant that high-elevation birds developed reproductive structures 34-38 days 91 earlier than they did relative to low-elevation birds in the field. This shift in high-elevation timing was dramatic and beyond the range of normal, inter-annual variation seen in Fig. 2.1.1 also showed that birds collected as adults, presumed to have already experienced a breeding season in the wild, matured slightly earlier than birds caught as hatchlings. These results suggest that the pronounced differences in the onset of reproduction between high- and low-elevation dark-eyed juncos revealed in chapter 2 were mainly the result of phenotypic plasticity to different environmental conditions. However, the reversal in order of maturation between elevational groups suggests the potential for countergradient variation in this system, whereby genetic selection (or early maternal/perinatal effects) counteracts or diminishes the effects of the environment on the phenotype of an animal (Levins 1969; Conover andSchultz 1995; Mel l i sh et al 2000). Finally, because high-elevation birds became reproductively active so much earlier relative to low-elevation birds in the lab compared to the field, this group may have evolved a genetic predisposition for a higher level of phenotypic plasticity than the low-elevation group. A high level of phenotypic plasticity can evolve in species that live in highly unpredictable environments, such as high-elevations (Levins 1968; V i a et al. 1995; Schlichting and Pigliucci 1998). In the Rocky Mountains of Alberta, the dark-eyed junco expressed very different reproductive schedules and seasonal reproductive success rates between the extremes of its elevational distribution (chapter 2). In their natural habitats, high-elevation males and females become capable of breeding and initiating egg-laying 3-4 weeks later in the season compared to low-elevation juncos. One hypothesis for this pattern was that male and female dark-eyed juncos had undergone local genetic adaptation such that high- and low-elevation populations responded to different critical day lengths: between elevations. Another hypothesis was that the strong cline in abiotic and biotic variables with elevation (chapter 2) was responsible for the temporal differences in reproductive development in the field. I found direct evidence for an effect of the environment on the timing of seasonal changes required for reproduction in this species. When birds experienced I 92 environmental differences, as they did breeding within high- and low-elevations in the field, temporal initiation of reproductive development differed by over 30 days between populations. However, when birds were exposed to identical conditions in the lab, the temporal differences were reduced to only 4-8 days and the order in which populations developed was reversed. This suggests that the environmental conditions in the habitats of origins accounted for a large proportion of the temporal differences in reproductive timing that I observed between free-living populations, since the experimental conditions eliminated these environmental differences leading to populations breeding at more similar times. Daylength is considered to be the primary environmental cue for timing seasonal reproduction in temperate songbirds (Rowan 1925; Murton and Westwood 1977; Gwinner 1986), while supplementary cues, such as temperature, rainfall, and food availability aid in fine-tuning reproduction once daylength reaches a critical level (e.g. Johnson 1954; Hahn 1995, 1998; Widmer and Biebach 2001). High- and low-elevation birds in my study system experienced identical photoperiods, and yet differed in reproductive timing by over 30 days. I was able to decrease this temporal difference in reproductive timing by simply removing environmental differences experienced by either population suggesting a primary role for supplementary cues in driving temporal differences among elevations. This study, therefore, is a powerful demonstration of the degree to which supplementary cues can influence reproductive timing, altering reproductive schedules by one month or more out of a possible 3-4 breeding months, despite equivalent photoperiod exposure. In a less intensive elevation gradient study of song sparrows (Melospiza melodia morphna) from coastal (0-1.0 m) and montane (280-1220 m) habitats, Perfito et al. (2004) also reported larger temporal differences in the development of reproductive features in birds measured in the wi ld versus birds held in the lab and allowed to respond to natural increases in daylength. However, Perfito et al. (2004) investigated only A S Y males, included a smaller sample size, and examined an elevational difference that was lower in elevation and less extreme (270-1210 93 ra). M y study was conducted over a steeper, higher elevation gradient and demonstrated a strong influence of supplementary cues in the reproductive development of both males and females. As opposed to Perfito et al. (2004), I also included birds collected at a much younger age when they had just left the nest and were unable to fly (9-11 d), when their development was less influenced by their elevation of origin prior to collection. Data on hand-reared male and female dark-eyed juncos also suggested that genetic selection and environmental effects may oppose one another in determining the timing of reproduction in this species. Although both high- and low-elevation males developed their cloacal protuberances at more similar times in the lab compared to the field, indicating a strong environmental effect, I also observed a surprising reversal in which group developed first. High-elevation males grew functional gonads and began to sing slightly earlier, an indicator of reproductive readiness (Dloniak and Deviche 2001), compared to low-elevation males in the lab. This pattern was the opposite from what was observed in the field where low-elevation males developed gonads and began to sing earlier than high-elevation males. A similar trend was observed in females, although this trend was not significant. Sti l l , high-elevation females achieved a more advanced ovarian stage and had more developed brood patches earlier than low-elevation females in the lab. In free-living populations, however, low-elevation females developed their brood patches earlier than high-elevation females. These data suggest that populations differ in their innate responsiveness to increasing daylength, with high-elevation birds able to respond slightly faster than low-elevation birds. Although the present results argue in favor of a genetic basis for the differences found in the lab between high- and low-elevation juncos, the present experiment was performed on birds that had spent at least their first 9-11 days of life in their natural habitats. Hence perinatal effects cannot be completely excluded (Mousseau and Fox 1998). I also cannot mie out the possibility that maternal effects caused the observed differences in reproductive timing between high- and low-elevation experimental birds. However, i f perinatal/maternal effects were involved, they acted differently on high- and low-elevation birds because differences in reproductive timing and responses to lab conditions were elevation-specific. Therefore, it appears that high-elevation birds of both age classes have the innate potential for higher annual fitness as they are reproductively ready earlier than low-elevation birds when released from environmental constraints. High-elevation birds may be hard-wired to mobilize resources into reproduction as soon as possible (at an early critical daylength), or they may put on more mass by eating more, which is not likely from personal observations of levels of food in feeding trays, or by being more metabolically efficient with calories such that energy stores can be shunted into reproductive expenses (see chapter 5). However, an excess of resources are not present in high-elevation habitats early in the season (Table 1.1), and so this phenomenon may only be observed when constraints are decreased. Simply worded, i f the environment slows a bird down, selection or perinatal/maternal effects may be required, and selected, to speed the bird up, and the balance of these opposing forces is what is observed in nature. Thus, when I removed the environmental influences in the lab, I was able to see an underlying countergradient force. High-elevation populations may need countergradient mechanisms or they would be unable to finish breeding before late season deterioration of weather. In other words, i f high-elevation birds had to wait for the snow to melt and for the growing degrees to reach an equivalent value of those at which low-elevation birds initiate most of their broods (chapter 3), they might simply inn out of time. An interesting subsequent study would be to conduct a common garden experiment at colder temperatures mimicking those experienced at high-elevations. In such an experiment, I would predict that high-elevation birds would develop reproductive features earlier than low-elevation birds, however, I would expect this pattern to be much larger in magnitude (i.e., high-elevation birds breeding much earlier than low-elevation birds in the lab) similar to the patterns seen in frogs collected from high- and low- elevations and raised in the lab at temperatures mimicking high-elevation conditions (Berven et al 1979). Finally, high-elevation birds may have evolved a greater capacity for phenotypic plasticity in response to changing conditions, as they live and breed in 95 variable, unpredictable habitats (Martin and Weibe 2004; Bears et al. 2003), which can select for high phenotypic plasticity (Levins 1968; V i a etal. 1995; Schlichting and Pigliucci 1998). These latter two ideas are not mutually exclusive, however, as countergradient selection in the context of this experiment would make a population appear more plastic to the conditions examined. Determining an overall reaction norm pattern for each population (i.e.. the relative capacities of each population for adaptive plasticity to the entire range of variables encountered by the species as a whole) would require observations of high-and low-elevation populations in many different environmental contexts, which is normally logistically unfeasible for birds. In males and females from both elevations, I also observed a tendency for birds collected as adults to develop reproductive traits earlier in response to increasing daylength than those collected as 9-11 day old hatchlings from a comparable elevation. There is considerable evidence from other studies that various aspects of the breeding performance of birds improve with age (Curio 1983; Saether 1990), due in large part to older birds having an earlier laying date. Dawson (2003) demonstrated that testicular maturation was advanced by 3-4 weeks in captive A S Y European starlings Stumus vulgaris compared to S Y birds. Further, starlings breeding for their second year regressed their testes 2 weeks later than first year breeders. Soekman et al. (2004) also demonstrated that experience with photo-stimulation during the first breeding year may prime the hypothalamo-pituitary-gonadal axis to respond to photic cues more rapidly or robustly in subsequent years in European starlings. M y results agree with these studies that show older birds that have experienced prior breeding seasons may be able to initiate clutches earlier. Taken together, my data suggest that environmental variables act to strongly delay the timing of reproduction in juncos breeding within high-elevation habitats, while selection pressures and/or maternal effects oppose these environmental effects. These results corroborate the results of chapter 3 and suggest that snowpack, and possibly limited food early in the season, likely constrains high-elevation birds to breed later. In nature, the environmental effects (constraints of snow cover, food) 96 appear to win out over genetic and/or maternal effects, and thus high-elevation males and females delay reproduction and undergo earlier reproductive termination. Environmental conditions within high-elevation sites in my system may promote this pattern through constraints on high elevation birds, as metabolic costs early and late in the breeding season in high-elevation habitats are likely to be greater, due to colder temperatures, prolonged snow cover, frequent storms, and a delayed peak in the availability of insects as food (Table 1.1). Experimental studies in which photosensitive birds were exposed to the same long day lengths but different ambient temperatures provide some evidence for a direct constraining effect of lower temperatures on the reproductive system (Silverin and Viebke 1994; Wingfield et al. 1996, 1997, 2003). Other studies have shown that the temporal peaks in food resources can constrain birds breeding in periods of low food abundance (Visser et al. 2006). However, in the laboratory where environmental effects were controlled for and an unlimited supply of nutritious food and water was supplied, I was able to observe an underlying propensity in high-elevation birds to develop slightly faster than low-elevation birds to the increasing photoperiod. In conclusion, the present study suggests that the profound temporal differences in development of reproductive features in dark-eyed juncos in high- vs low-elevation habitats can mainly be attributed to phenotypic plasticity and countergradient genetic differences or maternal effects, as illustrated in Fig. 4D. Countergradient forces may play a role in counteracting the degree to which environmental variables are able to compress the reproductive period in high-elevation birds, yet these effects are masked in the wild. High-elevation populations may also have an inherently greater capacity for plasticity than low-elevation populations. In the next chapter I w i l l test for differences in morphological features in dark-eyed juncos in high- and low-elevation habitats, and examine underlying causes of variation in morphological features to see i f morphological differences observed in the field are also decreased and reversed in the lab as they were with reproductive traits. 97 A Maintenance Differences between elevations, maintained in aviary -Persistent genetic differences -Early perinatal/maternal effects x CD E "TO CD N CO T J CO C o si TD CD i— CO CO CD CO r -B Convergence C Divergence Inversion Differences between elevations, convergence in aviary -Environmental differences Differences between elevations, divergence in aviary -Differing constraints -Persistent genetic differences -Perinatal/maternal effects Differences between elevations, Inversion in aviary -Differing constraints -Persistent countergradient genetic differences -Countergradient perinatal/maternal effects FIELD AVIARY F i g . 4.1 Schematic of four possible theoretical outcomes of how timing of reproductive development in dark-eyed juncos caught at different elevations may change when birds are brought into common aviary conditions and stimulated to breed by increasing daylength. Dotted and solid lines represent birds from high- and low-elevations, respectively, and text in italics denotes possible mechanisms (which may co-occur) that could explain each of the 4 patterns depicted. 98 L O o L O o L O o * - T- CN CN co L O O L O O L O • > - • » - CN CN L O O L O O L O O * - CN CN CO L O O L O O L O T- CN CN L O O L O O L O O T- CN CN CO L O O January Feb March April May Jun Day of Experiment F i g . 4.2 Photoperiod prescribed to high- and low-elevation dark-eyed juncos (Junco hyemalis) in captivity in order to stimulate the temporal reproductive cycle, from 02 Jan to 20 Jun, 2005. Dark-eyed juncos were brought from short-days (8 hours of daylight) to long days (15 hours of daylight, mimicking summer solstice) in captivity over a period of 140 days (increase of 3.85 min /day). Birds were kept at the peak level of 15 hours of daylight for an additional 40 days. 99 E E .c T3 Q. O 12 10 8 6 4 0 • A — H i g h S Y • A — H i g h A S Y -o— Low S Y Low A S Y ^ ^ ^ <eP ^ . eP ^ ^ ^  ^  ^  ^  ^  ^  ^  ^VVV^ ^  ^  ^  if ° f N * f r p f *v Date of Experiment N J <r v * " \ N " \ * ^ Fig. 4 . 3 Changes in mean cloacal protuberance width (±1SE) over time in male dark-eyed juncos collected from high- and low-elevation habitats in Jasper National Park and subjected to identical increases in photoperiod in a common lab environment to stimulate reproductive onset. 100 Field Adult Laboratory SY Year Laboratory ASY Year > 30 days, Low First A H i g h • L o w 1 8 days, High First r. 4 days, High First A i c ra co CN CO ci co o CO Q. <: CO CL < I CN CO CO i n CM c 3 00 Date of Experiment F i g . 4.4 Dates of onset of functional gonads in free-living (field) male birds, compared with experimental (laboratory) male birds collected as hatchlings and raised to breed for their first time in captivity ( ' S Y ' age) or as adult birds that had experienced at least one breeding season in the wi ld prior to collection and induction of breeding in the laboratory ( ' A S Y ' age). The absolute difference between onset and the order in which birds from high- and low-elevations became capable of breeding are shown. Field A S Y birds are only included for comparing the absolute differences in the days between populations becoming capable of breeding; actual initiation dates between the field (naturally stimulated) and the lab (artificially stimulated earlier in the season) are irrelevant since lab birds were artificially induced earlier in the season than either population breed naturally in the field. 101 2.5 - - f t - - H i g h S Y Date of Experiment Fig . 4.5 Changes in the mean number of songs (±1SE) per minute recorded from male dark-eyed juncos collected from high- and low-elevation habitat in Jasper National Park and subjected to an increasing photoperiod in a common lab environment to stimulate reproductive onset. 102 4.5 1-Jan 1-Feb 1-Mar 1-Apr 1-May Date of Experiment/Surgery F i g . 4 . 6 Average ovary scores for photoperiodically-induced female dark-eyed juncos collected at the end of the previous breeding season as 9-11 d hatchlings (SY) or adults ( A S Y ) from high-and low- elevation habitat. 103 4.5 25-Jan 15-Feb 10-Mar 30-Mar 20-Apr 15-May 5-Jun i 5-Feb 25-Feb 20-Mar 10-Apr 5-May 25-May 15-Jun Date F i g . 4 . 7 Mean brood patch scores by date in lab-reared female juncos collected as 9-11 day hatchlings (SY at time of measurements; top) or as adults ( A S Y ; bottom) from high- and low-elevation habitats in Jasper National Park. 104 4.5 REFERENCES Baily, R. E. 1952. The incubation patch, of passerine birds. - Condor 54:121-36. Cohen, J. 1988. Statistical power analysis for the behavioural sciences. Hillside, N J , M c G r a w H i l l . Conover, D.O. , and Schultz, E.T. 1995. Phenotypic similarity and the evolutionary significance of countergradient variation. - TREE 10: 248-252. Curio, E . 1983. Why do young birds reproduce less well? Ibis 125: 400-404. Dawson, A . , K ing , V . M . , Bentley, G . E . and Ba l l , G.F. 2001. Photoperiodic control of seasonality in birds. - J. Biol. Rhythm 16: 366-381. Dawson, A . 2003. A comparison of the annual cycles in testicular size and moult in captive European starlings Sturnus vulgaris during their first and second years. - J. Avian Biol. 34:119-123. Dloniak, S .M. , and Deviche, P. 2001. Effects of testosterone and photoperiodic condition on song production and vocal control region volumes in adult male dark-eyed juncos (Junco hyemalis). - Horm. Behav. 39: 95-1-5. Enstrom, D . A . , Ketterson, E .D. , and Nolan, V . 1997. Testosterone and mate choice in the dark-eyed junco. Anim. Behav. 54: 1135- 1146. Farner, D.S. , and Lewis, R . A . 1971. Photoperiodism and reproductive cycles in birds. -Photophysiol. 6: 325-370. Farner, D . S., Donhan, R. S., Matt, K . S . , Mattocks, P.W., Moore, M . C , and Wingfield, J .C. 1983. The nature of photorefractoriness. In. Avian endocrinology: environmental and ecological perspectives (Mikami, S. Homma, K . , and Wada, M . , Eds). Pp 149-166. Tokyo: Japanese Scientific Society Press. Gwinner, E . 1986. Circannual Rhythms. Springer-Verlag, Heidelberg. 125 pp. Hahn, T.P. 1995. Integration of photoperiodic and food cuesto time changes in reproductive physiology by an opportunistic breeder, the red crossbill, Loxia curvirostra (Aves: Carduelinae). - J. Exp. Zool. 212: 213-226. Hahn, T.P. 1998. Reproductive seasonality in an opportunistic breeder, the red crossbill, Loxia curvirostra. - Ecology 79: 2365-2375. Hamner, W. H . 1966. Circadian control of photoperiodism in the House Finch demonstrated by interrupted-night experiments. - Nature 203: 1400-1401 Johnston, R.F. 1954. Variation in breeding season and clutch size in song sparrows of the Pacific coast. - Condor 56: 268-273. Lapp, S., Byrne, J., Townshend, I., and Kienzle, S. 2005. Climate warming impacts on snowpack accumulation in an alpine watershed. - Int. J. Climatol. 25: 521-536. Lack, D . 1968. Ecological Adaptations for Breeding in birds.- Methuen, London. Lambrechts, M . M . , Perret, P., and Blondel, J. 1996. Adaptive differences in the timing of egg laying between different populations of birds result from variation in photoresponsiveness. - Proc. Roy. Soc. B 263: 19-22. Levins, R. 1968. Evolution in Changing Environments. Princeton University Press, Princeton, N J . Levins, R. 1969. Thermal acclimation and heat resistance in Drosophila species. - Am. Nat. 103: 483-499 MacDougall-Shackleton, S.A., MacDougall-Shackleton, E . A . , and Hahn, T.P. 2001. Physiological and behavioural responses of female mountain white-crowned sparrows to natal- and foreign-dialect songs. - Can. J. Zool. 79: 325-333. Mousseau T . A . , and Fox, C W . 1998. The adaptive significance of maternal effects. - TREE 13: 403-407. Murton, R . K . , and Westwood, N . J . 1977. Avian breeding cycles. Oxford University Press, Oxford. Nicholls, T.J., Follet, B . K . , Goldsmith, A . R . , and Pearson, H . 1988. Possible homologies between photorefractoriness in sheep and birds - the effect of thyroidectomy on the length of the ewes breeding season. - Reprod. Nutr. Dev. 28: 375-385. Perfito, N . , Tramontin, A . D . , Meddle, S., Sharp, P., Af ik , D. , Gee, J., Ishii, S., Kikuchi , M . , and Wingfield, J.C. 2004. Reproductive development according to elevation in a seasonally breeding male songbird. - Oecologia 140: 201-210. Pigliucci, M . , and Schlichting, C D . 1998. A Reaction Norm Perspective. Sinauer Associates Inc, Sunderland, Massachusetts. Pp 387. Rowan, W . 1925. Relation of light to bird migration and developmental changes. - Nature 115: 494-495. Sa;ther, B . E . 1990. Age-specific variation in reproductive performance. - Curr. Ornithol. 7: 251-283. Schlichting, C D . and Pigliucci, M . 1998. Phenotypic Evolution: a Reaction Norm Perspective. Sinauer Press, Sunderland, M A . Silverin, B . , Massa, R., and Stokkan, K . A . 1993. Photoperiodic adaptation to breeding at different latitudes in great tits. - Gen. Comp. Endocrin. 90: 14-22. Silverin, B . , and Viebke, P .A . 1994. L o w temperatures affect the photoperiodically induced L H and testicular cycles differently in closely related species of tits (Parus spp.). -Horm. Behav. 28: 199-206 Sockman, K . W . , Williams, T.D. , Dawson, A . , and Ba l l , G.F. 2004. Prior experience with photostimulation enhances photo-induced reproductive development in female European starlings: A possible basis for the age-related increase in avian reproductive performance. -Biol. Reprod. 71: 979-986. Sokal, R. R., and Rohlf. F.J . 1995. Biometry: the principles and practice of statistics in biological research. 3rd edition. W . H . Freeman and Co.: New York. Pages 321-342. Thomson, A . L . 1950. Factors determining the breeding seasons of birds: an introductory review. - Ibis 92: 173-184 Thomas, D .A . , Blondel, J., Ferret, P., Lambrechts, M . M . , and Speakman, J.R. 2001. Energetic and fitness costs of mistiming resource supply and demand in seasonally breeding birds. Science 291: 2598-2600. V i a , S., Gomulkiewicz, R., De Jong, G . , Scheiner, S . M . , Schlichting, C D . and Van Tienderen, P . H . 1995. Adaptive phenotypic plasticity: consensus and controversy.- Trends Ecol. Evol. 10: 212-217. Visser, M . E . , Holleman L . J . M . , and Gienapp, P. 2006. Shifts in caterpillar biomass phenology due to climate change and its impact on the breeding biology of an insectivorous bird.-Oecologia 147: 164-172. Weggler, M . 2006. Constraints on, and determinants of, the annual number of breeding attempts in the multi-brooded Black Redstart Phoenicurus ochruros. - Ibis 148: 273-284. Widmer, M . , and Biebach, H . 2001. Changes in body condition from spring migration to reproduction in the garden warbler Sylvia borin: a comparison of lowland and a mountain population.- Ardea 89: 57-68. Wingfield, J . C , and Farner, D.S. 1976. Avian endocrinology - field investigations and methods. - Condor 78: 570-573. Wingfield J . C , Hahn T.P., Wada M . , Astheimer L . B . , and Schoech, S. 1996. Interrelationship of day length and temperature on the control of gonadal development, body weight, and fat score in white-crowned sparrows, Zonotrichia leucophrys gambelii.- Gen.Comp. Endocrin. 101: 242-255. Wingfield J . C , Hahn T.P., Wada M . , and Schoech, S.J. 1997. Effects of day length and 108 temperature on gonadal development, body mass, and fat depots in white-crowned sparrows, Zonotrichia leucophryspugetensis.- Gen. Comp. Endocrin. 107: 44-62 Wingfield J . C , Hahn T.P., Maney D . L . , Schoech S.J., Wada M . , and Morton, M . L . 2003. Effects of temperature on photoperiodically induced reproductive development, circulating plasma luteinizing hormone and thyroid hormones, body mass, fat deposition and molt in mountain white-crowned sparrows, Zonotrichia leucophrys oriantha.- Gen. Comp. Endocrinol. 131: 143-158. Wingfield, J . C , Hahn, T.P., Lev in , R. , and Honey, P. 1992. Environmental predictability and control of gonadal cycles in birds.- J. Exp. Zool. 261: 214- 231. 109 5 C O M P A R A T I V E M O R P H O L O G Y O F D A R K - E Y E D JUNCOS B R E E D I N G A T T W O E L E V A T I O N S : A C O M M O N A V I A R Y E X P E R I M E N T 8 5 . 1 INTRODUCTION The morphologies of species often vary along environmental gradients (James 1983, 1991; Price 1991). Early studies assumed that morphological variation reflected genetic variation that was maintained by natural selection for locally adaptive traits (Mayr 1963). Subsequent studies have shown that morphological differences can result from phenotypic plasticity, where environmental variables directly influence the phenotype of an animal within its lifetime to create non-heritable trait variation (Stearns 1989; Zink 1982). Studies have also demonstrated that phenotypic plasticity may itself be subject to natural selection (Pigliucci 2005). Despite the recognition that many causes and constraints shape the morphologies of species, assumptions about adaptive traits still linger from the application of Mayr ' s (1963) logic that phenotypic variation along clines probably results from adaptive selection, and can be used to determine the direction of selection on traits in various habitats (e.g. Graves 1985; Price 1991; Badyaev 1997; Blackburn and Ruggiero 2001; Erickson et al. 2004). Experiments are increasingly being used to test these assumptions, with some surprising results. F o r example, amphibians develop more slowly at higher elevations; hence it was assumed that slow development was selected for at high-elevations or that development was constrained in high-elevation environments and that selection was not occurring (reviewed in Morrison and Hero 2003). However, when green frogs, Rana clamitans, were collected from high- and low-elevation habitats and placed in a common environment, high-elevation frogs grew faster than low-elevation frogs (Berven et al. 1979). Thus, genotypes encoding for faster development were being selected in high-elevation frogs in order to counteract the developmentally retarding § A vers ion of this chapter h a s b e e n a c c e p t e d for publication a n d is in press . Bears , H., Drever, M . C , a n d Martin, K. (2007). Comparat ive morphology of the dark-eyed junco breeding at two elevat ions: a c o m m o n aviary experiment. Journal of Av ian Biology, M S 4191 . 110 impact of colder temperature on development. In their natural environment, temperature effects won out over genetic effects and so the direction of selection could not be observed. When environmental effects were equalized, however, the direction of genetic selection was exposed. Multiple authors have cited patterns of morphological variation in songbirds along elevation gradients. High-elevation songbirds in Nepal were larger in size, and had longer wings than their low-elevation counterparts (Landmann and Winding 1995). A positive relationship between weight and elevation was also reported in other bird species (Blackburn and Ruggiero 2001; Laiolo and Rolando 2001; Soobramoney et al. 2005). Tai l feathers in chats and finches also change shape with increasing elevation (Landmann and Winding 1993, 1-995). Beak size (e.g., Price 1991), fat (e.g., Martin et al. 1993), and plumage (e.g., Graves 1985) have also been shown to vary with elevation. Differences in morphologies uncovered between high- and low-elevation birds may either be the result of micro-evolutionary changes that have evolved during the process of adapting to breed successfully within local habitats, or of phenotypic plasticity of individuals exposed to the different environmental conditions within low- and high-elevation habitats. High-elevation birds experience extremely different abiotic and biotic conditions during the breeding season compared to low- elevation birds. Conditions that differ at high-elevations include persistent snow cover, frequent snow and hail storms, colder temperatures, more rainfall and snowfall, and a later peak in insect biomass abundance (Bears et al. 2003), all of which may limit the growth of morphological traits through constraint, or provide selective pressure that has led to shifts in mean values of morphological traits. The goal of the present study was to test for morphological differences in free-living high- and low-elevation birds and then to differentiate between genetic versus environmental causes for the differences observed. I performed a field investigation combined with an experimental rearing experiment to compare morphological traits in a sparrow, the dark-eyed junco (Junco hyemalis ; Nolan et al. 2002), breeding within low- (1000 m asl) and high-111 elevation (2000 m asl) habitats in the Canadian Rocky Mountains. The dark-eyed junco is an emberizid commonly used in studies of morphological and life-history variation along environmental gradients (Miles and Ricklefs 1984; Nolan and Ketterson 1990; Chandler and M u l v i h i l l 1990; Bears 2002; Bears et al. 2003; Bears 2004). Jasper National Park was chosen as a study site because the habitat is unfragmented and juncos breed continuously along the elevational range, and therefore allopatric genetic differentiation due to habitat fragmentation preventing gene flow is unlikely. Further, the abiotic cline at this latitude and elevation is particularly steep (Bears et al. 2003). I used field measurements to describe differences in eight morphological traits in dark-eyed juncos breeding within low- and high-elevation habitats. Following the field investigation, I performed a rearing experiment to infer likely causes of morphological differences observed between elevations. Birds were caught at different elevations and reared under conditions of unlimited food supply, termed a 'common aviary' (analogous to common garden experiments in botany), and morphological traits were compared between field and aviary environments. With the exception of a few studies (Altshuler et al. 2004; Rasner et al. 2004; Y e h and Price 2004), such experiments have rarely been performed on songbirds to help determine the causes of morphological differences associated with elevational gradients. Assuming constraints in growth are eased when birds are reared in the aviary, morphological traits of birds caught at different elevations may change in a number of ways when brought into the aviary (Fig. 5.1). First, as a null response I could expect to see no difference in morphological trait values between elevations in the field, and this lack of a difference is maintained when populations are reared in a common aviary (Fig. 5.1 A ) . This result would imply that the populations are genetically the same, and that their morphological traits are not affected by phenotypic plasticity in response to conditions that differ between elevations in the field. Second, differences in morphological traits observed in dark- eyed juncos between elevations in the field might be subsequently maintained in the aviary (Fig. 5. IB). This result 112 would imply that there were persistent genetic differences between the populations or that the early perinatal environment or maternal effects prior to capture set the morphological growth rate of traits according to elevation. Third, differences in morphological traits might be observed between elevations in the field, but trait values converge in the common aviary environment where conditions are identical (Fig. 5.1C). This scenario implies that differences in the field were due to phenotypic plasticity in response to abiotic and biotic differences that exist between low-and high-elevation habitats. Fourth, differences in morphological traits between elevations in the field might be small or not present, but trait values diverge between elevational populations when they are raised in a common environment (Fig. 5. ID). This result could imply that there are genetic or perinatal/maternal differences in the ability of either population to grow the traits measured, but that these differences oppose the influence of the environment. For example, high-elevation populations may be genetically selected, or hardwired through the early perinatal environment, to allocate more resources into feather growth, yet I cannot observe this underlying propensity manifested in differing feather lengths because high-elevation environments limit feather growth more than low-elevation environments. When birds are released from constraints in a common environment, I am able to observe differences in feather lengths. These 4 scenarios assume field conditions impose constraints on growth that are then relaxed in the aviary. However, all figures could also be drawn with horizontal lines, implying that there were no constraints on growth of morphological traits in the field as compared to the laboratory. 5.2 METHODS 5.2.1 Field methods- I studied dark-eyed juncos in Jasper National Park, Alberta, Canada. Juncos breed from A p r i l to mid-August in low-elevation sites, and from June to late-July at high-elevation sites (2000 - 2100 m asl; High). A total of 273 dark-eyed juncos were measured at 7 low- and 8 high- elevation sites (Fig. 1.1). Site characteristics are described in detail in Bears et al. (2003). Juncos arrive at high- and low-elevation sites at the same time 113 annually, exhibit breeding and natal site fidelity in this region (chapter 2). Juncos were attracted by playing recorded songs of male juncos, and captured in mist nets, a technique that captured many more males than females. The ages of juncos were categorized as hatchling year ( H Y ; hatched the summer that I first caught them), second-year (SY) , or after second year ( A S Y , hatched the summer prior to the summer in which they were caught or earlier; Pyle et al. 2001). The following morphological measurements were taken: tarsus length (to 0.01 mm), unflattened wing chord length (to 0.1 mm), body weight (to O.lg), tail length (i.e. rectrix; to 0.1 mm), beak width, depth and length (to 0.01 mm), and fat score (0-5 scale) according to Pyle et al. (2001). Birds of an unknown age or with worn feathers were not used, and feather lengths did not change over the capture period (Pearson's r = -0.10, p = 0.43). U.S . Fish and Wildlife Service numbered aluminum bands and color bands were affixed to each bird to identify individuals. 5.2.2 Captive Birds and Experimental Design- In July, 2004,1 caught 43 young (9-11 days; $\ 18 L o w , 7 High; $ : 9 L o w , 9 High) and 27 adult (S: 8 Low, 6 High; $ : 6 L o w , 6 High) juncos for a common aviary experiment**. One 9-11 day old (hatchling year; H Y ) young bird was collected per brood from nesting attempts at least 500 m apart. Adult (after second year; A S Y ) juncos were also collected on the same sites. Comparing these two different age groups allowed us to see i f living at an elevation for a longer period of time affects the ability of birds to change their phenotype in the common aviary, or i f they maintain the capacity for plasticity into adulthood. High-elevation juncos collected hatched between 21 June and 07 July, and low-elevation juncos between 20 May and 15 July. Birds were held in an outdoor aviary (6 m x 1.5 m x 2 m) at the Palisades Research Centre (1020 m asl) in Jasper until they reached independence by 30 days, at which point birds were transported to the University of British Columbia Animal Care facility in Vancouver, British Columbia, Canada, and placed in a roofed, 3A walled semi-A common garden approach was chosen since a reciprocal transplant experiment was not possible due to logistical constraints of differences in the timing of active nests, high risk of mortality during egg transport, and subterranean placement of high-elevation nests. 114 outdoor aviary (10 m x 15 m x 10 m). Birds flew freely and were given trees, perches, roosting shelters, and heat lamps. Eight food and water dishes (1 m x 1 m) were provided to eliminate competition for resources. Juncos were fed a veterinarian recommended diet ad libitum, which consisted of fresh berries, soft Science-diet© cat food, cooked egg, suet, Hartz © vitamin/mineral enriched seed mix, and soft-bodied crickets. Prime © amino acid powder was added to water daily. Charcoal, oyster shells, and grit were sprinkled on the aviary floor as digestive aids. Juncos undergo a full molt from the end of the breeding season to December (Pyle et al. 2001), and thus all birds replaced their feathers ca. 4.5 month after being in captivity. In January 2005 birds had finished molting and were at their maximum adult feather lengths (i.e. there was no further growth of feathers after this point, and feather length cannot increase again until molt the following year). B y the time the experiment was initiated, the birds that had originally been collected as young, 9-11 day old hatchlings had matured to full grown "second year" birds in captivity, and hence are referred to as the S Y age group in analyses. The adult birds caught as adults are classified as after second year ( A S Y ) age class in analyses. Birds were caged individually (Gina© cages; 33.02 cm x 45.72 cm x 58.42 cm) and transferred to a photoperiod- and temperature-controlled (19 °C) chamber at U B C in January 2005; their diets remained the same as in the aviary. The temperature and photoperiods were chosen to represent conditions experienced at both elevations in Jasper during the breeding season, such as to fall within the range of tolerance of both populations. In captive birds, I only measured morphological traits that could change after collection. For example, tarsus length and beak dimensions would have been close to fixation and irreversible by the time birds were collected, but wing and tail lengths could change as all birds underwent a complete replacement of wing and tail feathers in captivity. Weight and fat scores could also change in captivity over the course of a year. I took measurements every 2 months until June (i.e. January, March, and June) to account for slight changes over time, which were not seen (measurements were nearly identical 115 between measures). The averages from all measurements from each individual were used in analyses. Procedures were done with appropriate permits (chapter 4). 5.2.3 Data analyses- Analysis of morphological data is often done with multivariate methods (Manly 2005). However, morphological traits of juncos examined here were not strongly inter-correlated (Table 5.1), and thus would not benefit from data reduction techniques such as Principal Components Analysis (PCA) . In addition, I needed to analyse traits that became fixed during development (tarsus length, and beak measures) separately from traits that could vary annually or seasonally (wing chord length, tail length, mean weight, and mean fat score) and could thus be expected to vary under field and aviary conditions. I therefore used a series of linear mixed effects models (Pinheiro and Bates 2000) to assess how each morphological trait varied as a function of elevation ('high' or ' low') , rearing condition ( 'field' or 'aviary'), and their interaction. A s juncos can vary in size and weight depending on age and sex (Nolan et al. 2002), I considered the age and sex groups separately and fit different series of models depending whether the trait became fixed at development or could vary temporally. For wing chord length, tail length, mean weight, and mean fat score, I first fit a model that had site of origin as a random effect, and elevation, rearing, and their interaction as fixed effects as my test for divergence or convergence of a morphological trait in the aviary (Fig. 5.1, panels C and D). I f there was no significant interaction effect, I fit the main effects model with elevation and rearing as fixed effects, and based my inferences on this reduced model. Including site as a random effect allowed us to account for potential site-level differences and to deal with the non-independence of birds that were captured on the same site. For tarsus length and the three beak measures, I fit one model with elevation as a fixed effect and site of origin as a random effect. A l l models were fit with R (R Development Core Team 2006) using restricted maximum likelihood with the lme4 package. Random effects and residuals from all regressions were normally distributed, and residuals were homoscedastic. 116 5.3 RESULTS 5.3.1 Morphological differences among elevations- Morphological traits of free-living, wild-caught juncos tended to have similar mean values between high- and low-elevations, with two main exceptions that had large effect sizes (Table 5.2). First, high-elevation females of both age classes had longer wing lengths than low-elevation females by ca. 2 mm (Appendix III). Second, high-elevation males of both age classes had longer tail lengths than low-elevation males by 2-3 mm (Table 5.2, Appendix III). High-elevation females of both age classes also had longer tail lengths than low-elevation females, but this difference was not significant (Table 5.2, Appendix III). In addition, small differences in tarsus length, beak length, and mean fat score were observed between elevations, but these differences were not consistent among sexes or age classes. Tarsus lengths of high-elevation S Y females and A S Y males were shorter by 0.2-0.3 mm than tarsus lengths of low-elevation birds. Beak lengths of high-elevation S Y females were shorter by 0.5 mm than beak lengths of low-elevation S Y females. A S Y females at high-elevations tended to have smaller mean fat scores than A S Y females at low-elevations. 5.3.2 Common Aviary Experiment- The four traits measured in the aviary (wing chord length, tail length, mean weight, and mean fat score) tended to increase in magnitude relative to measures taken from birds in the field, although the details of the responses varied by trait (Figs. 5.2 and 5.3). For wing chord length, there was no evidence of an interaction between elevation and rearing effects, and the main effects model indicated a significant rearing effect, such that wing lengths of juncos raised in the aviary were 1.0-2.2 mm longer than wing lengths of free-living juncos. This difference in wing length was significant for all age/gender groups, except for A S Y males (Appendix III), where the low-elevation juncos raised in the aviary had similar wing lengths to wild-caught birds. Tai l lengths of low-elevation juncos tended to be similar for aviary-raised juncos relative to wild-caught birds, whereas tail lengths of high-elevation juncos tended to be longer or similar for aviary- raised juncos relative to wild-caught birds. These 117 differences, however, were not statistically significant, and I found no evidence of an interaction between elevation and rearing. The main effects model for al l 4 age/gender groups indicated only weak evidence for an effect of rearing on tail length. For mean weight, there was no evidence of an interaction effect for females of both age classes (Appendix III), and weight of females in general showed little difference between elevations and rearing environments (Fig. 5.3). In contrast, the interaction effect between elevation and rearing was significant for males of both age classes (Appendix III), indicating that the differences in mean weight between aviary- and wild-reared juncos depended on the elevation of origin. High-elevation males raised in the aviary had significantly greater mean weights than wild-caught males, whereas low-elevation males did not. Mean fat score of aviary-raised juncos tended to be greater than wild-caught birds, although this difference was only significant for A S Y females and S Y males. There was no evidence of an interaction between rearing and elevation for any of the 4 age/gender groups, indicating that mean fat scores of birds from both elevations of origins differed in the same way between the field and aviary. In summary, the eight morphological traits I measured in free-living birds had similar values between high- and low-elevations, with two consistent exceptions with large effect sizes. Females at high-elevation sites had longer wings than females from low-elevation sites, and males from high-elevation sites had longer tails than males from low-elevation sites. When juncos were reared in the aviary and provided with unrestricted access to food, the four morphological traits I measured showed a tendency to be greater in magnitude. Juncos from high- and low-elevations had similar responses to the aviary environment, with one exception: males from high-elevation sites had greater weight gain relative to wild-caught juncos from high-elevation sites than males from low-elevation sites relative to wild-caught males from low-elevation sites. 118 5 . 4 DISCUSSION The first main finding of my study was that some morphological traits of free-living juncos differed between 1000 m and 2000 m asl in my study area in the Canadian Rocky Mountains. The traits that varied consistently among sex and age groups were tail lengths in males and wing chord lengths in females: high-elevation males of both age groups had longer tails (rectrices) than low-elevation males, and females at high-elevations had longer wings than at low-elevations. Rectrices aid in balance and steering (Maybury et al. 2001) and therefore strong selection pressure may exist for longer tails in high-elevation habitats for flight and perching in windy conditions with lower barometric pressure (Lindstrom et al. 1993; Landmann and Winding 1993; Maybury et al. 2001). Longer rectrices may be especially beneficial for males that participate in lengthy courtship and territorial displays on treetops, which are particularly windy positions in high-elevation habitats. Rectrices are also used in courtship and territorial displays in this species (Hi l l et al. 1999; Yeh and Price 2004), and sexual selection has been shown to enhance the amount of white on rectrices in males that defend sparser habitat and compete more strongly for females (Yeh and Price 2004). Sex ratios and sexual selection pressures between elevations are not yet established for this system. In females, I found a stronger effect of elevation on wing length. High-elevation females had longer wings than low-elevation females in the field, which may also serve to increase flight efficiency in windy, low-barometric pressure conditions at higher elevations (Lindstrom et al. 1993; Landmann and Winding 1993). Longer wings may also indicate that high-elevation females fly further to their wintering sites, as wing length is correlated with migration distance in this differentially migrating species (Chandler and M u l v i h i l l 1990). Some smaller magnitude, age- and gender-specific differences were also noted in other morphological traits in free-living dark-eyed juncos between elevations. Tarsi were slightly shorter in high-elevation males and females in my study system. Laiolo and Rolando (2001) 119 found that alpine chough (Pyrrhocorax graculus) had shorter tarsi and other extremities compared to its close relative, the lower elevation red-billed chough (Pyrrhocorax pyrrhocorax). Laiolo and Rolando (2001) suggested this was due to Al len 's rule, which states that animals adapted to colder climates have smaller/shorter protruding body parts to reduce heat loss through the surface area of extremities (Allen 1877). Failure to detect differences in tarsi in both age classes may have been due to insufficient power and a small effect size, and larger sample sizes may reveal similar differences in both age categories. Fat scores in females were also slightly higher in low-elevation environments in the A S Y age group, which may be due to the fact that low-elevation A S Y females have the most broods per season, and fat stores generally increase in advance of yolk-loading in the egg (Hau et al. 2004). A S Y females in low-elevation habitat may also have access to larger quantities of fatty foods or burn fat less frequently as they encounter fewer storm events that require the utilization of fat reserves. Finally, beak lengths were slightly longer in low-elevation S Y females in the field. Where habitat and seed size/type co-vary, selection can favour differing beak dimensions for foraging efficiency. For instance, with changing food and habitat types, Price (1991) found that beak measures varied along altitudinal gradients in Phylloscopus warblers (Price 1991). In a north-temperate emberizid, the song sparrow (Melospiza melodia), selection acted on beak length rather than beak depth or width (Schluter and Smith 1986). However, longer beaks have also been shown to benefit birds in preening to k i l l ectoparasites, which are common in young birds (Clayton et al. 2005). If external parasites are higher in low-elevation birds, similar to patterns seen in blood parasites (Bears 2004) , there may be strong selective pressure for longer maxillary overhangs on the culmens of young birds. Maxil lary overhangs are prone to breakage and wear over time, which may explain why beak lengths in older A S Y age groups no longer differed between elevations (Clayton et al. 2005) . 120 The second main finding in my study was that there was a shift in morphological trait measures when birds originating from high- and low-elevation habitats were raised in a common environment for a full molt cycle. In general, morphological traits measured in males and females from both elevations tended to increase in magnitude when birds were raised in the laboratory, demonstrating that these traits are phenotypically flexible and respond to differences in the environment. Since most trait sizes increased in the lab, the common environment likely decreased energetic constraints by supplying more resources, decreasing energetic demands, or both, which allowed birds to allocate more resources to growth. Feather replacement and growth demand a great deal of energy, and optimal feather lengths are not reached i f molt occurs when resources are not plentiful, o f poor quality, or when metabolic demands are high (Harper and Skinner 1998; Myhre and Steen 1979; Andreev 1999; Pearcy and Murphy 2001). The non-significant reduction in the lengths of rectrices in A S Y females here was most likely due to above average rectrix wear in one or more of a small sample size of individuals, but the difference between the lab and field were non-significant. For most traits, there was no interaction between elevation and rearing condition (field and aviary) and therefore populations from either elevation showed similar responses when brought into the aviary (Fig 5.IB). In other words, most traits displayed phenotypic plasticity, but there were also persistent genetic or perinatal/maternal differences underlying population responses that prevented traits from converging under aviary conditions, and high- and low-elevation populations maintained trait size differences. Rasner et al. (2004) also reported that wing and tail lengths in dark-eyed juncos can evolve rapidly, and may be longer in mountain areas: members of an isolated population of dark-eyed juncos (Junco hyemalis thurberi) residing within University of California San Diego campus have shorter wings and tails than nearby mountain populations, which are thought to have colonized the U C S D campus in 1980. A 121 common garden experiment indicated that the morphological differences likely had a genetic or perinatal basis (Rasner et al. 2004). I observed a slightly different pattern for weight in A S Y males in the common aviary experiment as compared to the shifts observed in wing, tail lengths, and fat scores. While free-living males did not differ in weight between elevations in the wild, high-elevation males gained significantly more weight than low-elevation males in the common lab environment, resulting in a large difference in weight between birds originating from high- and low-elevations in the lab. A similar pattern was observed in S Y males, but was not significant. Results for male weight, therefore, most closely match the divergence scenario in Fig. 5. ID. These results suggest that high-elevation males, especially in the older age category may have a greater innate capacity to put on weight compared to low-elevation males, either due to higher feeding rates, greater metabolic efficiency or other internal trade-offs. The greater weight gaining capacity in high-elevation males was likely masked in the field by stronger environmental constraints on weight in high-elevation environments. When the constraints were removed, however, as they were in the common environment, I was able to observe the underlying propensity for high-elevation males to put on more weight. The exacerbation of male weight in the laboratory could be an example of countergradient variation, whereby genetic selection works to counteract or diminish the constraining effects of the environment on the weight of high-elevation birds (Levins 1969; Conover and Schultz 1995; Mer i la et al. 2001). In the field, the environmental effects masked genetic- and/or perinatal or maternally-derived differences in weight gaining capacity, but when environmental conditions were equalized I was able to observe the underlying differences in weight gaining capacity. Although the present results suggest a potential genetic basis for the differences between low- and high-elevation juncos, my experiment was performed on birds that had spent their first week of life or longer in their natural habitats. Hence perinatal and early maternal effects cannot 122 be completely excluded, as these can influence the metabolic pace of other vertebrates, and species raised in areas with lower primary productivity may experience a lower metabolic pace for life (Mousseau and Fox 1998; Mueller and Diamond 2001; Wikelski et al. 2003) or may have a permanently altered morphology (Smith 1998). However, i f perinatal/maternal effects were involved, they acted differently on high- and low-elevation birds because differences in morphologies and responses to lab conditions were elevation-specific. In conclusion, the present study revealed a novel pattern for vertebrate homeotherms; morphological differences that were present between populations of dark-eyed juncos from different elevations persisted or were amplified when both populations were held in a common lab environment. The results of the field and aviary portions of the experiment show phenotypic plasticity in certain traits, but also suggest a role of genetic selection and/or perinatal/maternal effects that underlie how juncos interact with the environment to produce morphological phenotypes. The present study suggests that elevation gradients in this region may provide strong selective pressure for population differentiation. Additional studies that decrease the contributions of maternal and early developmental effects are important next steps. 123 Table 5.1. Pearson cross-correlation coefficients of eight morphological traits of wild-caught and captive dark-eyed juncos (Junco hyemalis) from Jasper National Park, Alberta, Canada, from 2000, 2001, 2004 and 2005. N = 298. Traits are: T A R = Tarsus length, W I N G = wing length, T A I L = tail length, W E I G H T = average weight, F A T = average fat score, B D = beak depth, B W = beak width, and B L = beak length. Bolded coefficients are significant at a = 0.05. T A R W I N G T A I L W E I G H T T A R W I N G 1 0.15 T A I L W E I G H T -0.16 0.17 0.43 0.14 F A T -0.05 -0.04 -0.01 0.10 B D -0.04 0.19 0.05 0.10 0.01 B W 0.11 0.17 0.02 0.05 -0.09 0.21 B L 0.00 -0.09 0.06 0.06 -0.09 0.27 0.13 1 0.01 F A T B D B W B L 124 Table 5.2. Morphological traits of wild-caught after-second year (ASY) and second-year (SY) dark-eyed juncos {Junco hyemalis) at high-and low-elevations in Jasper National Park, Alberta, Canada, from 2000, 2001, 2004 and 2005. Traits are expressed as means, with sample sizes and standard deviations in brackets (N, SD). Means in bold are significantly different between elevations, at a = 0.05, based on linear mixed models. See Appendix III for details. Age/Gender A S Y S Y A S Y S Y Females Females Males Males Elevation High Low High Low High Low High Low Wing Length 74.2 72.6 74.7 72.9 77.9 77.6 77.5 77.5 (mm) (11,3.2) (11, 5.0) (13,1.3) (12, 1.2) (63, 0.5) (53,0.1) (43, 0.2) (30,0.3) Tail Length 60.1 59.4 59.2 58.1 62.2 59.9 62.0 59.2 (mm) (11,1.4) (11,4.1) (13,0.6) (12, 1.2) (63, 0.7) (53,0.1) (43, 0.3) (30, 0.3) Weight 18.2 17.8 17.6 18.1 18.4 18.6 18.3 18.5 (g) (10,2.0) (9, 3.6) (10, 1.0) (9,1.4) (63, 0.7) (52, 0.1) (43, 0.3) (30, 0.4) Fat Score 1.9 2.6 2.2 1.9 1.8 1.9 1.8 1.7 (scale 0-5) (11,3.0 (11,3.6) (13,0.7) (12,1.6) (63, 0.3) (53,0.1) (43, 0.2) (30, 0.4) Tarsus Length 20.2 19.5 19.6 20.0 20.2 20.5 20.4 20.7 (mm) (11,2.2) (11,4.6) (13,0.9) (12, 0.8) (62, 0.8) (53, 0.2) (43, 0.3) (30, 0.3) Beak Depth 5.4 5.3 5.4 5.4 5.4 5.5 5.4 5.4 (mm) (11,2.7) (11,3.0) (13,1.1) (12, 1.1) (62, 0.8) (53, 0.2) (42, 0.3) (30, 0.4) Beak Width 5.6 5.6 5.6 5.7 5.8 5.9 5.8 5.9 (mm) (11,2.5) (11,4.4) (13,0.9) (12, 1.0) (63, 0.7) (53, 0.2) (43, 0.3) (27,0.4) Beak Length 7.9 8.0 7.7 8.2 8.0 7.9 8.0 7.9 (mm) (11,2.2) (11,2.7) (13,0.8) (12, 0.9) (62, 0.8) (53, 0.2) (43, 0.3) (30, 0.4) 125 A No differences between elevations, maintained in aviary Similar constraints FIELD AVIARY FIELD FIELD FIELD AVIARY AVIARY AVIARY Differences between elevations, maintained in aviary Persistent genetic differences Early perinatal effects Differences between elevations, convergence in aviary Environmental differences No differences between elevations, divergence in aviary Differing constraints Persistent genetic differences Perinatal effects Fig. 5.1 Schematic of four possible outcomes of how morphological traits of birds caught at different elevations may change when birds are brought into aviary under conditions of unlimited food supply. Solid and dotted lines represent birds from high- and low-elevations, respectively, and text in italics denotes possible mechanisms for observed patterns. 126 SY Females FIELD AVIARY ASY Females H L FIELD AVIARY SY Females H - L -31 FIELD AVIARY ASY Females E E CD c CD I CO c E E cn c CD I CD C E E CD c CD CO SY Males 82 80 78 76 74 72 82 80 78 76 74 72 66 64 62 60 58 56 E E^  - C A—> CD C CD I CD 66 64 62 60 58 56 323-= b L H . L - -=cr FIELD AVIARY ASY Males H - L-t l_ FIELD AVIARY SY Males . — H -jr FIELD AVIARY ASY Males H .L -FIELD AVIARY FIELD AVIARY Fig. 5.2 Mean values (± 1SE) associated with wing and tail lengths of male and female dark-eyed juncos collected from low- and high-elevation sites in Jasper National Park, Canada. S Y refers to second-year, and A S Y refers to after second-year age classes. 127 SY Females U) 21 CD C CO CD 19 18 17 H L FIELD AVIARY ASY Females 3 2 1 I) 2 0 £ 19 an 18 CD 17 4.0 o 3.5 o CO -4—* 3.0 CO LL 2.5 an 2.0 Me 1.5 2> 4.0 o 3.5 o CO 3.0 -+-» CO LL 2.5 H ^ L "I FIELD AVIARY SY Females ^ 1.5 FIELD AVIARY ASY Females FIELD AVIARY SY Males 3 21 2o CD C CO 0 19 18 17 H L FIELD AVIARY ASY Males 3 21 20 19 18 17 'CD H L FIELD AVIARY SY Males CD 4.0 8 3.5 <2 3.0 CO LL 2.5 CO CD 2.0 1.5 FIELD AVIARY 2 4.0 8 3.5 <2 3.0 CO LL 2.5 ASY Males 2.0 H 1.5 H L -33= I FIELD AVIARY Fig. 5.3 Mean values (± 1SE) associated with size and energy storage of male and female dark-eyed juncos collected from high- and low-elevation sites in Jasper National Park, Canada. S Y refers to second-year, and A S Y refers to after second-year age classes. 128 5.5 R E F E R E N C E S Allen , J .A. 1877. The influence of physical conditions in the genesis of species. Radical Review 1: 108-140 Altshuler, D . , Dudley, L . R . and McGuire , J .A. 2004. Resolution of a paradox: Hummingbird flight at high elevations does not come without a cost. - PNAS 101: 17731-17736. Andreev, A . V . 1999. Energetics and survival of birds in extreme environments. - Ostrich 1: 13-22. Badyaev, A . V . 1997. Avian life-history variation along altitudinal gradients: an example with cardueline finches. - Oecologia 111: 365- 374. Bears, H . 2002. How and why a small songbird, Junco hyemalis, breeds over a steep elevation gradient. M S c . thesis, University of British Columbia. Bears, H . 2004. Parasite prevalence in Dark-eyed Juncos, Junco hyemalis, breeding at different elevations. - Can. Field Nat. 118: 235- 238. Bears, H . , Smith, J . N . M . and Wingfield, J .C. 2003. Adrenocortical sensitivity of Dark-eyed Juncos (Junco hyemalis oregonus) breeding in low and high elevation habitat. -Ecoscience 10: 127-133. Berven, K . A . , G i l l , D . E . and Smith-Gills, S.J. 1979. Countergradient selection in the Green Frog, Rana clamitans. - Evolution 33: 519-623. Blackburn T . M . and Ruggiero, A . 2001. Latitude, elevation and body weight variation in Andean passerine birds. - Global Ecol. Biogeogr. 10: 245-259. Chandler, R . C . and Mulv ih i l l , R.S. 1990. Wing-shape variation and differential timing of migration in Dark-Eyed Juncos. - Condor 92: 54-61. 129 Clayton, D . H . , Moyer, B .R. , Bush, S.E., Jones, T .G . , Gardiner, D.W., Rhodes, B . B . and Goller, F. 2005. Adaptive significance of avian beak morphology for ectoparasite control. - Proc. Roy. Soc. Bill: 811-817. Conover, D .O. and Schultz, E.T. 1995. Phenotypic similarity and the evolutionary significance of countergradient variation. - TREE 10: 248-252. Erickson, V . J . , Mandel, N . L . and Sorenson, F .C . 2004. Landscape patterns of phenotypic variation and population structuring in a selfing grass, Elymus glaucus (blue wildrye). -Can. J. Bot. 82: 1776-1789. Graves, G.R. 1985. Elevational correlates of speciation and intraspecific geographic variation in plumage in Andean forest birds. - Auk 102: 556-576. Harper, E .D. , and Skinner, N . D . 1998. Clinical nutrition of small psittacines and passerines. -Semin. Avian Exot. Pet. 7: 116-127. Hau, M . , Wikelski , M . , Gwinner, H . and Gwinner, E . 2004 Timing of reproduction in a Darwin's finch: temporal opportunism under spatial constraints. - Oikos 106: 489-500. H i l l , J .A. , Enstrom, D .E . , Ketterson, E.D., .Nolan, V.J r and Ziegenfus, C . 1999. Mate choice based on static vs. dynamic secondary sexual traits in the Dark-eyed Junco. - Behav. Ecol. 10: 91-96. James, F .C . 1983. The environmental component o f geographic variation in the size and shape of birds: transplant experiments with blackbirds. - Science 221: 184-186. James, F. C. 1991. Complementary descriptive and experimental studies of clinal variation in birds.- Am. Zool. 31: 694-706. Laiolo, P. and Rolando, A . 2001. Ecogeographic correlates of morphometric variation in the Red- billed Chough Pyrrhocorax pyrrhocorax and the Alpine Chough Pyrrhocorax graculus. - Ibis 143: 602-612. Landmann, A . and Winding, N . 1993. Niche segregation in high-elevation Himalayan chats (Aves, Turdidae): Does morphology match ecology? - Oecologia 95: 506-519. Landmann, A . and Winding, N . 1995. Gui ld organization and morphology of high-elevation granivorous and insectivorous birds: convergent evolution in an extreme environment. -Oikos 7 3 : 237-250. Levins, R. 1969 Thermal acclimation and heat resistance in Drosophila species. - Am. Nat. 1 0 3 : 483-499. Lindstrom, A . , Visser, G . H . and Daan, S. 1993. The energetic cost of feather synthesis is proportional to basal metabolic rate. - Physiol. Zool. 66: 490-510. Manly , B .F . J . 2005. Multivariate Statistical Methods: A Primer. Third Edition. Chapman and H a l l / C R C Press, Boca Raton, Florida, U S A . Martin K . , Holt, R .F . and Thomas, D . W . 1993. Getting by on high: ecological energetics of Arctic and alpine grouse. In: Florant, G . L . , Wunder, B . A . and Horowitz, B.(eds), Life in the cold III: ecological, physiological, and molecular mechanisms, pp. 33-41. Westview Press; Boulder, Colorado, U S A . Maybury, W.J . , Rayner, J . M . V . and Couldrick, L . B . 2001. Lift generation by the avian tail. -Proc. Roy Soc. B 268: 1443-1448. Mayr, E . 1963. Animal species and evolution. Harvard University Press, Cambridge. Meri la , J., Kruuk, L . E . B . and Sheldon, B . C . . 2001. Cryptic evolution in a wi ld bird population. -Nature 4 1 2 : 76-79. Miles , D . B . and Ricklefs, R . E . 1984. The correlation between ecology and morphology in deciduous forest birds. - Ecology 65: 1629-1640. Morrison, C. and Hero J . - M . 2003. Geographic variation in life-history characteristics of amphibians: a review. - J. Anim. Ecol. 7 2 : 270-279. Mousseau, T. A . and Fox, C W . 1998. Maternal effects as adaptations. Oxford University Press, New York, U S A . 131 Mueller, P. and Diamond, J. 2001. Metabolic rate and environmental productivity: W e l l -provisioned animals evolved to run and idle fast. - PNAS 98: 12550-12554. Myhre, K . and Steen, J .B. 1979. Body temperature and aspects of behavioral temperature regulation in some neonate sub-arctic and arctic birds. - Ornis Scand. 10: 1-9 Nolan, V . and Ketterson, E . D . 1990. Timing of autumn migration and its relation to winter distribution in dark-eyed juncos. - Ecology 71: 1267-1278. Nolan, V . , Jr., Ketterson, E . D . , Cristol, D . A . . , Rogers, C M , Clotfelter, E . D. , Titus, R. C , Schoech, S.J. and Snajdr, E . . 2002. Dark-eyed Junco (Junco hyemalis). In: Poole, A.and Gi l l s , F. (eds), The Birds of North America, no. 716. Academy of Natural Sciences, Philadelphia, and American Ornithologists' Union, Washington. Pearcy, S.D. and Murphy, M . E . 2001. Essential amino-acid metabolism in molting and non-molting sparrows in different nutritional states. - Comp. Biochem. Physiol. 118: 1157-1163. Pigliucci , M . 2005. Evolution of phenotypic plasticity: where are we going now? - TREE 20: 481-486. Pinheiro, J.C. and Bates, D . B . 2000. Mixed-Effects Models in S and S - P L U S . Springer-Verlag, New York, New York, U S A . Price, T. 1991. Morphology and ecology of breeding warblers along an altitudinal gradient in Kashmir, India. - J. Anim. Ecol. 60: 643-664. Pyle P., Howell , S.N.G., Yunick, R.P. and DeSante, D .F . 2001. Identification guide to North American passerines. Slate Creek Press, Bolinas, C A . R Development Core Team. 2006. R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. U R L : http://www.R-proiect.org. 132 Rasner, C . A . , Yeh , P., Eggert, L . S . Hunt, K . E . Woodruff, D.S. and Price, T .D. 2004. Genetic and morphological evolution following a founder event in the dark-eyed junco, Junco hyemalis thurberi.- Mol. Ecol. 13: 671-681. Schluter, D . and Smith, J.N.M.1986. Natural-selection on beak and body size in the song sparrow. - Evolution 40: 221-231. Smith, J.I. 1998. Allometric influence on phenotypic variation in the Song Sparrow (Melospiza melodia ). Zool. J. Linnean Soc- 122:427-454 Soobramoney, S., Downs, C.T. , and Adams, N . J . 2005. Morphological variation in the Common Fiscal Lanius collaris along an altitudinal gradient in southern Africa. - Ostrich 76: 130-141. Stearns, S.C. 1989. The evolutionary significance of phenotypic plasticity. - BioScience 39: 436-44. Wikelski , M . , Spinney, L . Schelsky, W. , Scheuerlein, A . and Gwinner, E . 2003. Slow pace of life in tropical sedentary birds: a common-garden experiment on four stonechat populations from different latitudes. - Proc. Roy. Soc. B 270: 2383-2388. Yeh , P., and Price, T.P. 2004. Adaptive phenotypic plasticity and the successful colonization of a novel environment. - Am. Nat. 4: 532-542. Zink, R . M . 1982. Patterns of genetic and morphological variation among sparrows in the genera Zonotrichia, melospiza, Junco, and passerella. - Auk 99: 632-649. 133 6 G E N E R A L D I S C U S S I O N As described in the introduction of this thesis, one main goal in evolutionary ecology is to identify patterns of variation over landscape-level gradients that are useful for making predictions in natural ecosystems, and then to work to identify mechanisms that explain those patterns (Bergmann 1847 and Al l en 1877 reviewed in Ashton 2001; Lack 1968; Zammuto and Mi l l a r 1985a,b; Perfito et al. 2004). Studies that have established general patterns over environmental clines across landscapes (e.g. Lack 1968; Moore et al. 2002; Marcel l 2002) have indeed aided in making predictions at the landscape level and have helped in designing studies that have answered mechanistic questions about adaptive biology and processes that lead to phenotypic diversification (Berven et al. 1979; Lauren et al. 2003; Angilletta et al. 2004). M y dissertation was one of the first well-controlled comparative studies of intraspecific variation in birds among elevations. Such studies are needed to identify patterns that would aid in predicting directions of change along mountain gradients, and for examining underlying causes of variation. Other intraspecific comparisons have only focused on phenotypic variation in birds with elevation for a few traits in males, rather than the phenotypes of both sexes (e.g. Widmer 1999; Perfito et al. 2004). In my thesis, I sought to provide a comprehensive study of the variation of interrelated aspects of the avian phenotype in both males and females. In my system, which used the dark-eyed junco breeding at the extremes of its elevational range in the northern Rocky Mountains of Canada, I demonstrated shifts in a wide range of traits between 1000m and 2000m elevation sites. I have also revealed mechanisms underlying variation in temporal differences in reproductive schedules and morphologies. Although the phenotypic traits shown to vary with elevation in my dissertation are diverse, there are some common themes that emerge when they are considered holistically. Here, I discuss how the main findings of my thesis contribute to a better understanding of intraspecific variation in avifauna with elevation and other sub-134 disciplines of biology, including reproductive biology, evolutionary biology, and climate change research. Finally, I discuss some main themes that emerge when considering these results collectively. While more bird species that span wide elevational gradients w i l l need to be compared to establish the generality of the themes uncovered here, demonstrating what "can be" is a first step in establishing a body of scientific evidence to generate theory. In chapter 2,1 demonstrated a delayed onset of growth and earlier recrudescence of reproductive structures in high- relative to low-elevation birds that corresponded with a breeding season that was approximately half as long. A s a result, high-elevation birds produced fewer broods and young per season relative to low-elevation birds. Delayed reproductive schedules are frequently reported in birds breeding over a continuum of elevations (Verbeck 1967; Beason 1995; Krementz and Handford 1984; Badyaev 1997). Other studies have also found that fewer offspring per brood are produced at high-elevations (Krementz and Handford 1984; Badyaev 1997). The striking disparity in seasonal reproductive timing and success over this spatially compressed, 1000 m elevation gradient could have signaled that the high-elevation habitats were peripheral for this species, as has been shown for the high-elevation segment of the elevational range of Townsend's warblers (Dendroica townsendi; Rohwer 2004). However, based on current indicators of competitive class in juncos (age, size, and arrival times; Cristol et al. 1990; Grasso et al. 1996), I did not find a higher proportion of poor competitors in high-elevation habitat. Other studies have also failed to support the hypothesis that poor competitors are forced to breed at higher elevations (Kollinsky and Landmann 1996; Widmer 1999). The paradox of lower seasonal reproduction in a competitively equivalent population would be resolved i f high-elevation birds adhered to different life-history strategies resulting in similar lifetime reproductive success rates among elevations. Sandercock et al. (2005a,b) showed that, in congeneric grouse species breeding among elevations at different latitudes, species in the highest-elevation sites were the longest lived, but had the lowest seasonal reproduction. These 135 authors suggested that low current reproduction is traded for longer lives at higher elevations, but their results may have also been due to differences in evolutionary histories among species compared, or latitudinal differences. Morton et al. (2004) also observed a reproduction/survival trade-off in high-elevation habitat. These authors studied white-crowned sparrows (Zonotrichia leucophrys) breeding in an energetically demanding, high-elevation habitat and found that the offspring of smaller clutches had a higher over-winter survival rates. Badyaev and Ghalambor (2001) also found an inverse correlation between seasonal reproduction and offspring quality in cardueline finches along an elevation gradient due to an increased parental effort at high-elevations. The Morton et al. (2002) study, however, lacked a low-elevation group for comparison, and Badyaev and Ghalambor (2001) compared different species. In my thesis, I showed a trend towards higher survival rates in high-compared to low-elevation males, supporting the hypothesis that birds may be switching from a high-reproductive to a high-survivorship life-history strategy, with a similar trend of higher return rates in females. I interpreted this result with caution, as estimates could also signal differences in emigration rates of birds from high- vs. low-elevation sites. Other results from this study system, however, make the higher-survival interpretation more plausible. I previously showed that, compared to low-elevation birds, high-elevation adults and fledglings suffered from lower levels of blood parasitemia (Bears 2004), and lower levels of physiological stress in response to noxious stimuli (Bears et al. 2003). Fledglings from high-elevation habitats also had higher fat scores prior to spring migration (Bears 2002), which can increase over-winter survival (Leary et al. 1999). Presumably, these features could promote longer lives in high-elevation adults and fledglings alike. I was unable to determine whether low-elevation birds were required to produce more offspring to compensate for higher mortality rates of fledglings over 25 days of age within low-elevation habitat. Therefore, an important next study would focus efforts into banding and radio-tracking fledglings for comparing survival rates among elevations. 136 Given the striking discrepancy in seasonal reproduction between elevations uncovered in chapter 2,1 wished to explore causes of this difference in more detail. In chapter 3,1 sought to understand the relationship between different weather variables and investment in clutch initiation between elevations. Prior studies showed that climate warming is affecting the timing of migration and reproduction in multiple altitudinal migrants (Inouye et al. 2000; Hendricks 2003). However, prior studies had not compared the relationship between weather variables and reproductive investment among elevations, and whether birds track similar weather variables that differ seasonally among elevations in order to time their life-history events, or whether they organize their breeding schedules to match different weather variables and thresholds. In chapter 3,1 showed that the same weather variables were not correlated with reproductive investment in high- and low-elevation birds. Instead, I showed a positive, non-linear correlation between growing degree days and cumulative percentage of broods initiated in low-elevation birds, and a negative, non-linear correlation between date of snowmelt and a highly synchronized peak in brood initiation in high-elevation birds. Interestingly, the synchronized peak in high-elevation initiation events occurred at a lower growing degree day value than the value at which most low-elevation birds breed. Thus, a high proportion of high-elevation birds bred at a low growing degree day value with advanced snowmelt, while low-elevation birds did not deal with snow on the ground during the breeding season, and breeding initiation correlated strongly with increasing growing degree day values. The fact that high-elevation birds respond to snow depth and not temperature, and that low-elevation birds respond to temperature (and obviously not snow depth) is novel, and important in light of climate change, which might affect ground nesting birds differently among elevations. In this way, high-elevation habitats may be similar to high latitude habitats where the onset of breeding is relatively insensitive to temperature but strongly correlated to snow melt (Martin and Wiebe 2004; Hannon et al. 1978). 137 If weather variables are causally linked to reproductive phenology and not simply correlative, increasing spring temperatures should advance the breeding dates of low-elevation birds relatively more than at high-elevations. Other species of birds have also been shown to advance their breeding dates in response to increases in spring temperature (e.g. Inouye et al. 2000; Hegelbach 2001; Stenseth and Mysterud 2002; Naef-Daenzer et al. 2004). High-elevation birds, on the other hand, w i l l not likely breed earlier due to a direct influence of increasing temperature, as they are already able to breed at quite low growing degree day values. Instead, high-elevation birds may be impacted by changes in snow depth (e.g. Hendricks 2003). Increasing spring temperature may increase the rate of snowmelt (Harte et al. 1995; Inouye et al. 2000) and lead to earlier breeding in high-elevation birds. On the other hand, the early stages of climate warming are expected to increase winter precipitation, and thus a larger snowpack may need to be melted in the spring (Lapp et al. 2005), and unless increasing spring temperatures can make up for the larger snowpack in faster melting rates, high-elevation breeding may be delayed in some years. In addition, both high- and low-elevation birds showed a positive relationship between cumulative percent brood initiation and insect biomass and cumulative rainfall, and so the ability of insect emergence to shift in parallel with laying dates wi l l be important in the success of altered laying dates (e.g. Visser 1998; Stevenson and Bryant 2000; Visser et al. 2006.). In addition, i f snowmelt is causal in determining breeding initiation dates, variation in snow depth among different aspects and between years may result in stronger temporal variation in breeding schedules in high-elevation habitats, even at the same latitudes (e.g., Moore et al. 2006). Less predictability in the timing of snow melt in the future may also pose problems for high-elevation birds in that it could be difficult to evolve mechanisms to anticipate or track snowmelt and initiate clutches at an ideal time. Despite the strong correlations between brood initiation and snowmelt and temperature at high- and low-elevations, respectively, I could not be certain that weather variables were actually 138 driving the different reproductive schedules between elevations. It was also possible that highl-and low-elevation birds had adapted to their breeding elevation by modifying their responses to increasing spring photoperiod, which is the primary indicator for the transition into the reproductive season in the spring (Rowan 1925). In other words, even though populations experienced identical photoperiods among elevations, they may initiate reproductive development at different critical day lengths. Since photoperiod correlates with changes in weather, different critical day lengths between populations at either elevation could have produced spurious correlations between weather and breeding efforts. This hypothesis was important to consider, because i f populations had adapted to use different critical daylengths, they would be less able to modify breeding schedules in response to rapid changes in weather, and would instead require many generations of selection to modify their critical daylengths, which may take too long when considering the current pace of climate change. Considering that breeding phenologies did not vary inter-annually to the same magnitude as weather did in my study (Table 1.1; Fig . 2.1), selection for different critical daylengths between elevations seemed plausible. Other authors had shown that populations of birds can evolve to respond to different critical day lengths among latitudes (Silverin et al. 1993); however, it was considered less likely over the relatively short spatial scale of an elevation gradient, where isolation by distance was less important. In c h a p t e r 4,1 used a common garden experiment to test whether high- and low-elevation birds responded to different critical day lengths. I collected male and female early fledglings and adults from high- and low-elevation habitats and raised them in a common lab-environment. When photoperiod was increased to mimic the transition from winter to spring in the laboratory, I found that birds collected from high- and low-elevation habitats developed reproductive features on a more similar temporal schedule than they did in the wi ld , suggesting that supplementary cues such as weather and food abundance, which were held constant in the 139 lab, were partially responsible for different reproductive schedules between high- and low-elevation populations in the field. Song sparrows (Melospiza melodia) collected from different breeding elevations also altered their reproductive schedules due to differences in their elevation-specific weather variables rather than different critical day lengths (Perfito et al. 2004). However, the study by Perfito et al. (2004) tested only males collected as adults from different localities, at much lower elevations than investigated here. Although reproductive maturation between high- and low-elevation populations was more similar in the common lab environment than in the field, there were significant differences between age and elevation groups. Older birds (after second year breeders) were capable of breeding before younger (second year) birds in both high- and low-elevation groups. Other studies also showed that breeding performance can increase with age, and that prior photo-stimulation can sensitize birds to future photo-stimulation (Curio 1983; Dawson 2003; Sockman et al. 2004). High-elevation birds also became reproductively capable slightly earlier than low-elevation males and females, the inverse pattern to that observed in the field. Thus, high-elevation birds are physiologically capable of breeding slightly earlier than low-elevation birds given the same environmental contexts, or they are capable of a higher degree of phenotypic plasticity overall. M y results suggest the exciting possibility that high-elevation birds are adapting through countergradient genetic evolution (reviewed in Conover and Schultz 1995), or by countergradient perinatal/maternal developmental influences (that are elevation specific), whereby the influence of selection (or the early maternal/perinatal environment) runs counter to the influence o f the environment. Simply worded, i f the environment slows a bird down, selection or perinatal/maternal effects may be deployed to speed the bird up; the balance of these opposing forces is what I observe in nature. Thus, when I removed the environmental influences in the lab, I was able to see an underlying countergradient force. High-elevation populations may need countergradient mechanisms or they would be unable to finish breeding before late season 140 deterioration of weather. In other words, i f high-elevation birds had to wait for the snow to melt and for the growing degrees to reach an equivalent value of those at which low-elevation birds initiate most of their broods, they might simply run out of time. A n interesting subsequent study would be to conduct a common garden experiment at colder temperatures mimicking those experienced at high-elevations. In such an experiment, I would predict that high-elevation birds would develop reproductive features earlier than low-elevation birds, however, I would expect this pattern to be much larger in magnitude (i.e. high-elevation birds breeding much earlier than low-elevation birds in the lab) similar to the patterns seen in frogs collected from high- and low-elevations and raised in the lab at temperatures mimicking high-elevation conditions (Berven et al. 1979). Unt i l the end of chapter 4,1 focused on comparing life-history strategies and mechanisms for timing of reproduction between elevations. In chapter 5,1 extended my investigation to more components of the avian phenotype. I focused on how avian morphologies change with elevation, as morphological traits are also subject to elevation-specific selective pressures, constraints, and trade offs between life-history strategies and morphological traits exist in energetically challenging environments (e.g. Altshuler et al. 2004). I investigated whether the morphologies of high-elevation birds differed between elevations, and whether elevation-specific morphotypes were also a compromise between countergradient forces and constraints as reproductive phenology data suggested. I found that rectrix lengths were longer in male juncos at high-elevations, similar to other studies that have noted differences in rectrix shapes at higher elevations in other taxa (Landmann and Winding 1993, 1995). I also found that wing lengths were longer in female juncos at high-elevations, which has also been observed in interspecific comparative studies (Landmann and Winding 1993; 1995) and in hummingbirds (Altshuler et al. 2004). When I collected young and adult birds from high- and low- elevation habitats and raised them for a full molt cycle in a common lab environment, fat, weight, rectrix lengths, and wing lengths did not converge between 141 elevation groups. In other words, phenotypic plasticity alone was not underlying the morphological differences between elevational populations in the field. However, both populations were capable of phenotypic plasticity, as they reacted to the abundant food resources and benign common lab conditions by increasing the sizes of at least some morphological traits relative to values observed at comparable elevations in the field. However, large differences in trait values between elevational groups remained or increased in the laboratory. This result is very interesting in comparison to the results of chapter 4. Recall that, in chapter 4,1 suggested that phenotypic plasticity in response to environmental factors was very important in shaping the different breeding schedules between elevations, but that a countergradient force (either genetic or perinatal/maternal) was also likely working to oppose the constraining effect of the high-elevation environment on breeding initiation. In chapter 5, however, I found that the elevation-specific genetic or perinatal/maternal effects were likely more important than phenotypic plasticity to elevation-specific conditions in shaping morphological traits between elevations. In other words, the fact that high-elevation birds in nature have longer tail feathers and wings, despite the equivalent or greater constraints on growth (as shown by constrain release patterns in the lab) likely means that these traits are very important for life at high-elevations, possibly for efficient flight in high winds (Landmann and Winding 1993, 1995), and are maintained by strong countergradient forces that override, environmental constraints. Alternatively the lower investment in reproduction at high-elevations may leave more energy for feather production at high-elevations sites in the field. Future studies that quantify gene flow between elevations wi l l aid in determine how genetically independent high- and low- elevation populations are from one another and therefore the strength of selection that would be required to overcome gene flow. Sequencing of D-loop mitochondrial D N A may also help in determining whether populations were historically segregated, and whether current patterns could be vestigial remnants leftover from historical selection or isolation events. 142 The results of my dissertation contribute to a more complete understanding of organisms breeding at different locations along steep environmental gradients. The themes that emerge here may be relevant to other species breeding along other clines where strong selection pressures run counter to constraint on those same traits, and where phenotypes represent a compromise between the environment and countergradient genetic or maternal/perinatal forces. I have shown an interesting set of phenotypic shifts between the extremes of an environmental gradient, and that populations of the same species organize their life-history schedules in relation with different variables across their breeding range. I have suggested the somewhat daunting idea that the phenotypes observed in nature may be deceiving, and the true phenotypic potential of the organism may be masked by influences of their environment. Populations residing at the energetically challenging extremes of their breeding ranges appear to be well adapted to these conditions; they may be playing a completely different game by different rules, and realizing far different advantages. A s the environment becomes more physiologically constraining along an environmental gradient, however, populations may also need to counteract those constraints through genetic or maternal/perinatal adaptive changes. Such adaptations may not be easily detected in wi ld populations, and it may not be easy to predict the direction of the adaptations i f they run counter to environmental influences. Thus, populations that live within the most physiologically challenging extremes of a species' environmental gradient/range may be undergoing phenotypic diversification and selection inconspicuously. In turn, the challenging extreme of a species range may act to create "tougher" phenotypes through necessity, which creates the selective pressure required to change the mean phenotype of the population. In essence, populations breeding within the harshest extremes of their ranges may be like marathon runners that train at altitude. We all know that in the Boston Marathon, the Kenyans that train at high-elevation training camps always win. However, it would be impossible to predict that such Kenyans have superior endurance by simply comparing running times of Kenyans running a 143 marathon in high elevation sites within Kenya to Boston natives running a marathon in Boston. We might even conclude wrongly that Kenyans were less fit than Boston natives i f we were to compare Kenyans running a marathon at the Mount Everest base camp to Boston-dwellers running a marathon at sea level. If we imagine that Kenyans and Boston natives represent two populations of the same species, and that endurance level has evolved based on their local elevation (to counteract constraints of the training environment in the case of the high elevation Kenyan), we can see the problem with conclusions drawn from phenotypic comparisons made in the wild. Only when the proverbial "blanket of the environment" is pulled out from underneath populations can one really begin to observe and understand their phenotypic potential or a group of animals, or the direction in which their phenotypes are trying to move through selective pressure. When the Kenyan and Bostoner race in Boston we can observe this idea in action; there is no question that the high-elevation training environment has pushed the Kenyan via necessity to a higher endurance level, as they have been forced to acquire a more efficient oxygen deliver system and enzymatic pathway to counteract the lower barometric pressures in high- elevation training camps. Inversely, i f the Kenyan and Bostoner were to run a marathon at the Mount Everest base camp, we would expect the Kenyan to do far better than the Boston runner, who would likely suffer from altitude sickness. This analogy illustrates that what you see when comparing two populations in their native, differentially constraining environment, is not always an accurate representation of the phenotypic prowess of an animal. Therefore gradients that challenge, push, and necessitate adaptive changes and different tolerance limits among populations of a species, where the necessity for evolving or producing a trait via plasticity is strong enough such that the species must find ways of overriding their greater constraints may be valuable for conserving species, as we cannot know which traits and tolerance limits w i l l be required by species in the future on this rapidly changing planet. 144 6.1 R E F E R E N C E S Al len , J .A. 1877. The influence of physical conditions in the genesis of species. Radical Review 1: 108-140. Altshuler, D . , Dudley, L . R . and McGuire , J .A. 2004. Resolution of a paradox: Hummingbird flight at high elevations does not come without a cost. - PNAS101: 17731-17736. Angeilletta, M.(Jr.), Niewiarowski, P .H . , Dunham, A . E . , Leache, A . D . , and Porter, W.P . 2004. Bergmann's clines in ectotherms: Illustrating a life-history perspective with Sceloporine Lizards. Am. Nat. 164: 168-183. Ashton, K . G . 2001. Are ecological and evolutionary rules being dismissed prematurely? Diversity and Distributions 7: 289-295. Badyaev, A . V . 1997. Avian life history variation along altitudinal gradients: A n example with cardueline finches. - Oecologia 111: 365-374. Badyaev, A . V and Ghalambor, C K . 2001. Evolution of life histories along elevational gradients: trade-off between parental care and fecundity. - Ecology 82:2948-2960. Bears, H . 2002. How and why a small songbird breeds over a steep elevation gradient. M.Sc . Thesis, University of British Columbia. Bears, H . , Smith, J . N . M . , and Wingfield, J .C. 2003. Adrenocortical sensitivity to stress in Dark-eyed Juncos (Junco hyemalis oregonus) breeding in low and high elevation habitat. -Ecoscience 10: 127-133. Bears, H . 2004. Parasite prevalence in dark-eyed juncos (Junco hyemalis) breeding at different elevations. - Can. Field Nat. 118: 235-238. Beason, R . C . 1995. Horned Lark (Eremophila alpestris):The birds of North America, No . 195. The Academy of Natural Sciences; Philadelphia, U S A (Poole, A„ and G i l l , F., Eds.). Pp 1-24. Bergmann, C. 1847. Ueber die Verhaltnisse der Warmeokonomie der Thiere zu ihrer Grosse. -145 - Gottinger Studien 3: 595-708. Berven, K . A . , G i l l , D . E . and Smith-Giils, S.J. 1979. Countergradient selection in the Green Frog, Rana clamitans. - Evolution 33: 519-623. Cristol, D . A . , Nolan, V(Jr)., and Ketterson, E . D . 1990. Effect of prior residence on dominance status of dark-eyed juncos, Junco hyemalis. - Anim. Behav. 40: 580-586. Conover, D.O. , and Schultz, E.T. 1995. Phenotypic similarity and the evolutionary significance of countergradient variation. - TREE 10: 248-252. Curio, E . 1983. Why do young birds reproduce less well? Ibis 125: 400-404. Dawson, A . 2003. A comparison of the annual cycles in testicular size and moult in captive European starlings Sturnus vulgaris during their first and second years. - J. Avian Biol. 34: 119-123. Grasso, M . J . , Savalli, U . M . , and Mumme, R. 1996. Status signalling in dark-eyed juncos: perceived status of other birds affects dominance interactions. - Condor 98:636-639. Hannon, S.J. 1978. Timing of reproduction in 2 populations of wil low ptarmigan in northern Canada. Auk 105: 330-338. Hegelbach, J. 2001. Water temperature and phytophenology indicate the earlier onset of oviposition in Eurasian Dipper (Cinclus cinclus) from the Swiss Lowlands. - J. Fiir Ornithol. 142: 284-294. Hendricks, P. 2003. Spring snow conditions, laying date, and clutch size in an alpine population of American Pipits. - J. Field Ornithol. 74:423-429. Inouye, D.W. , Barr, W . A . , Armitage, K . B . , and Inouye, B . D . 2000. Climate change is affecting altitudinal migrants and hibernating species. - PNAS 97: 1630-1633. Koll insky, C. and Landmann, A . 1996. Altitudinal distribution of male Black Redstarts: are there age-dependent patterns? - Bird Study 43: 103-107. Krementz, D . G . and Handford, P. 1984. Does avian clutch size increase with Elevation? -Oikos 43: 256-259. Lack, D . 1968. Ecological adaptations for breeding in birds. Methuen, London. Landmann, A . , and Winding, N . 1993. Niche segregation in high-elevation Himalayan chats (Aves, Turdidae): does morphology match ecology. - Oecologia 95: 506-519. Landmann, A , and Winding, N . 1995. Gui ld organization and morphology of high-elevation granivorous and insectivorous birds: convergent evolution in an extreme environment.. Oikos 73: 237-250. Lapp, S., Byrne, J., Townshend, I., and Kienzle, S. 2005. Climate warming impacts on snowpack accumulation in an alpine watershed. - Int. J. Climatol. 25: 521-536. Laugen, A . T . , Laurila, A . , Rasanen, K . , and Meri la , J. 2003. Latitudinal countergradient variation in the common frog (Rana temporaria) development rates - evidence for local adaptation. - E v o l . Biol. 16: 996-1005. Leary, J., Sullivan, K . A . , and Hillgarth, N . 1999. Relationships among dominance, foraging proficiency, and condition in juvenile dark-eyed juncos. - Auk 116: 1136-1141. Marcell , C. 2002. The life-history basis of latitudinal diversity gradients: how do species traits vary from the poles to the equator? - J . Anim. Ecol. 71: 79-87. Martin, K . and Wiebe, K . L . 2004. Coping mechanisms of alpine and arctic breeding birds: Extreme weather and limitations to reproductive resilience. - Int. Comp. Biol. 44: 177-185. Moore, I.T., Perfito, N . , Wada, H . , Sperry, T.S., and Wingfield, J .C. 2002. Latitudinal variation in plasma testosterone levels in birds of the genus Zonotrichia. - Gen. Comp. Endocrinol. 129: 13-19. Moore, I.T., Bentley, G.E . , Wotus, C. and J.C. Wingfield. 2006. Photoperiod-independent changes in immunoreactive brain gonadotropin-releasing hormone (GnRH) in a free-living, tropical bird. Brain Behav. Evol. 68: 37-44. 147 Morton, M . L . , Pereyra, M . E . , Crandall, J.D., MacDougall-Shackleton, E . A . , and Hahn, T.P. 2004. Reproductive effort and return rates in the Mountain White-crowned Sparrow. Condor 106: 131-138. Naef-Daenzer L , Nager, R . G . , Keller, L . F . , and Naef-Daenzer, B . 2004. Are hatching delays a cost or a benefit for Great Tit Parus major parents? - ARDEA 92: 229-237. Perfito, N . , Tramontin, A . D , Meddle, S., Sharp, P, Af ik , D. , Gee, J., Ishii, S., Kikuchi , M . , and Wingfield, J.C. 2004. Reproductive development according to elevation in a seasonally breeding male songbird. - Oecologia 140:201-210. Rohwer, S. 2004. Using age ratios to infer survival and despotic breeding dispersal in hybridizing warblers. - Ecology 85: 423-431. Rowan, W. 1925. Relation of light to bird migration and developmental changes. - Nature 115: 494-495. Sandercock, B . K . , Martin, K . , and Hannon, S.J. 2005a. Life history strategies in extreme environments: comparative demography of arctic and alpine ptarmigan. - Ecology 86: 2176-2186. Sandercock, B . K . , Martin, K , and Hannon, S.J. 2005b. Demographic consequences of age- structure in extreme environments: population models for arctic and alpine ptarmigan. - Oecologia 146: 13-24. Silverin, B . , Massa, R., and Stokkan, K . A . 1993. Photoperiodic adaptation to breeding at different latitudes in great tits. - Gen and Comp. Endocrinol. 90: 14-22. Sockman, K . W . , Will iams, T.D. , Dawson, A . , and Ba l l , G.F. 2004. Prior experience with photostimulation enhances photo-induced reproductive development in female European starlings: A possible basis for the age-related increase in avian reproductive performance. -Biol. Reprod. 71: 979-986. Stenseth, N . C . , and Mysterud, A . 2002. Climate, changing phenology, and other life-history traits: Non-linearity and match-mismatch to the environment. - PNAS 99: 13379-13381. Stevenson, I.R., and Bryant, D . M . 2000. Av ian phenology- climate change and constraints on breeding. - Nature 406: 366-367. Verbeek, N . A . M . 1967. Breeding biology and ecology of the horned lark in alpine tundra. -Wilson Bulletin 79: 208-217. Visser, M . E . 1998. Warmer Springs lead to mistimed reproduction in great tits (Parus major). -Proc. Roy. Soc.B 265: 1867-1970. Visser, M . E . , Holleman, L . J . M . , and Gienapp, P. 2006. Shifts in caterpillar biomass phenology due to climate change and its impact on the breeding biology of an insectivorous bird. -Oecologica 147: 164-172. Widmer, M . 1999. Altitudinal variation of migratory traits of the garden warbler Sylvia borin. PhD Thesis, University of Zurich, Zurich, Switzerland. Zammuto, R . M . , and Mil lar , J.S. 1985a. Environmental predictability, variability, and Spermophilus columbianus life history over an elevational gradient. - Ecology 66: 1784-1789. Zammuto, R . M . , and Mil la r , J.S. 1985b. A consideration of bet-hedging in Spermophilus columbianus. - J. Mammal. 66: 652-660. 149 APPENDICES Appendix I Beta estimates of survival (phi) link function parameters of the top-ranking model (Phi(elevation+t) /?(elevation) Design Matrix} ± 95% Confidence Interval from Program M A R X . The elevation effect is significantly different than zero (i.e., the confidence interval does not overlap zero). The high-elevation birds have a higher survival rate than the low-elevation birds, as the high-elevation birds are always at least 1.42 units higher than the high-elevation birds. Parameter Beta Standard Lower Upper Error 1 :Phi Intercept 3.40 3.75 -3.94 10.75 2:Phi L o w -1.42 0.37 -2.14 -0.69 3:Phi t l -3.48 3.87 -11.07 4.11 4:Phi t2 15.99 0.00 15.99 15.99 5:Phi t3 -2.13 3.69 -9.37 5.10 6:Phi t4 -0.05 3.83 -7.56 7.46 7:Phi t5 -1.01 4.16 -9.16 7.14 8:Phi t6 -0.98 3.82 -8.48 6.51 9:Phi t7 -2.99 3.79 -10.42 4.45 10:Phi t8 0.89 3.86 -6.67 8.45 l l : P h i t 9 -4.22 3.82 -11.70 3.26 12:Phit lO -0.18 4.15 -8.31 7.96 13:Phi t l 1 -3.39 3.77 -10.78 3.99 14:Phi t l2 -2.75 3.88 -10.35 4.86 15:Phi t l3 15.60 970.21 -1886.02 1917.22 16:p Intercept -0.95 0.18 -1.30 -0.59 17:pLow -0.17 0.35 -0.85 0.52 150 Appendix II Day length, a primary factor thought to drive reproductive development in temperate, seasonally breeding songbirds by date in high- and low-elevation sites in Jasper National Park. Day lengths did not differ between elevations. Hobo Pro© light meters (1 per elevation) were also installed in open areas in 2004 to record photoperiod. Photoperiod was the same between elevations. Date 151 Appendix III Parameter estimates of fixed effects of linear mixed models depicting eight morphological traits of dark-eyed juncos (Junco hyemalis) as a function of elevation of origin, rearing condition, and their interaction. Gender refers to male (M) or female (F). Age refers to second-year (SY) and after second-year ( A S Y ) . Elevation refers to the average difference between high- and low-elevations, such a positive number indicates H > L . Rearing refers to the average difference between field- and aviary, such that a positive number indicates F > A . Interaction refers to the interaction between elevation and rearing effects, n/s means non-significant, and parameters are from main effects model only, t refers to estimate/standard error, and parameters where the absolute value of t > 1.96 are significantly different from 0, at a = 0.05. Age/Gender: A S Y Females S Y Females A S Y Males S Y Males Measure Parameter Estimate SE t Estimate SE t Estimate SE t Estimate SE t Wing Length Intercept 74.4 0.6 125.4 74.8 0.6 121.1 78.0 0.3 269.7 77.7 0.3 229.0 Elevation -2.1 0.8 -2.8 -2.0 0.8 -2.6 -0.5 0.4 -1.1 -0.3 0.5 -0.6 Rearing 1.7 0.8 2.1 2.2 0.8 2.8 1.0 0.7 1.5 1.3 0.6 2.2 Interaction n/s n/s n/s n/s Tail Length Intercept 60.5 1.2 52.4 59.7 1.1 54.0 62.3 0.5 116.6 62.1 0.5 121.2 Elevation -1.3 1.4 -0.9 -1.9 1.3 -1.4 -2.4 0.8 -3.2 -3.1 0.7 -4.2 Rearing -1.8 1.5 -1.2 -0.2 1.1 -0.2 1.0 1.1 0.9 1.2 0.9 1.3 Interaction n/s n/s n/s n/s Weight Intercept 18.2 0.3 68.7 2.3 0.4 6.6 18.4 0.1 158.8 18.3 0.1 140.5 Elevation -0.4 0.3 -1.2 -0.5 0.4 -1.1 0.1 0.2 0.9 0.3 0.2 1.3 Rearing 0.4 0.3 1.0 0.6 0.4 1.3 1.7 0.4 4.4 0.9 0.3 2.5 Interaction n/s n/s -2.2 0.5 -4.1 -0.9 0.4 -2.1 Fat Score Intercept 1.8 0.3 6.3 2.3 0.4 6.6 1.8 0.1 16.2 1.7 0.2 10.4 Elevation 0.9 0.4 2.3 -0.5 0.4 -1.1 0.04 0.2 0.2 0.1 0.2 0.3 152 Rearing Interaction 0.7 n/s 0.4 1.9 Tarsus Length Intercept Elevation 19.7 -0.2 0.2 109.2 0.3 -0.9 Beak Depth Intercept Elevation 5.4 0.0 191.1 -0.01 0.02 -0.5 Beak Length Intercept Elevation 7.9 0.1 0.1 0.1 77.8 0.4 Beak Width Intercept Elevation 5.6 0.0 0.1 0.1 87.4 0.5 153 0.6 0.4 1.3 n/s 19.4 0.2 126.9 0.4 0.2 2.0 5.4 0.0 125.0 0.03 0.1 0.3 7.7 0.1 76.1 0.5 0.1 3.6 5.6 0.1 55.8 0.1 0.1 0.7 0.3 0.3 1.2 n/s 20.2 0.1 206.2 0.3 0.1 2.0 5.4 0.0 211.4 0.1 0.0 1.5 8.0 0.0 184.9 -0.1 0.1 -0.8 5.8 0.0 168.0 0.1 0.1 1.6 1.2 0.2 4i8 n/s 20.4 0.1 162.7 0.1 0.2 0.7 5.4 0.0 114.5 0.0 0.1 0.4 8.0 0.1 129.5 0.04 0.1 -0.4 5.8 0.0 135.8 0.1 0.1 0.9 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            data-media="{[{embed.selectedMedia}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.831.1-0074960/manifest

Comment

Related Items