Winter Habitat Use by Boreal Chickadee Flocks Within a Managed Forest Landscape

ADAM HADLEY

WINTER HABITAT USE BY BOREAL CHICKADEE FLOCKS WITHIN A MANAGED FOREST LANDSCAPE

Mémoire présentée à la Faculté des études supérieures de l’Université Laval dans le cadre du programme de maîtrise en sciences forestières pour l’obtention du grade de de maître ès sciences (M.Sc.)

FACULTÉ DE FORESTERIE ET DE GÉOMATIQUE UNIVERSITÉ LAVAL QUÉBEC

2006

Résumé

On considère que les espèces résidentes d’oiseaux habitant les latitudes nord sont les espèces les plus exposées aux effets de la perte d’habitat et de la fragmentation de la forêt boréale. Nous connaissons très peu l’écologie hivernale des oiseaux boréaux résidents bien que la dynamique de leur population semble être fortement influencée par des événements qui ont lieu en-dehors de la saison de reproduction. Mon objectif était de déterminer comment l’augmentation de la densité des lisières forestières et la réduction de la proportion de forêt boréale mature influencent une espèce résidente d’oiseau. J’ai enregistré les mouvements de 85 volées hivernales de mésanges à tête brune (Poecile hudsonica) non marquées et de sept volées dont les membres étaient marqués individuellement avec des bagues de couleur. De janvier à mars (2004 et 2005), j’ai suivi des volées de mésanges en raquettes à la forêt Montmorency et j’ai enregistré leurs déplacements en temps réel en utilisant un récepteur GPS. Grâce aux volées d’individus marqués, j’ai découvert que les mésanges à tête brune comptent en moyenne 4 oiseaux par volée, occupent un territoire hivernal moyen de 14.7 ha et conservent les mêmes membres dans leur volée pendant l’hiver. À partir des déplacements de volées sur 74 km, les mésanges à tête brune préféraient fortement les peuplements forestiers matures (>7m de hauteur), utilisaient un peu moins les peuplements d’arbres en régénération (4-7 m) et évitaient les jeunes peuplements (<4 m) et les milieux ouverts. Les volées de mésanges ne s’intéressaient pas aux lisières forestières lorsqu’elles utilisaient des peuplements forestiers matures. Par contre, dans les peuplements en régénération, les volées étaient plus près des lisières ouvertes (41±6 m) ainsi que des lisières de peuplements forestiers matures (11±2 m) que prévu. Les volées de mésanges à tête brune n’évitaient pas les lisières exposées durant des conditions hivernales difficiles. Une augmentation de la densité des lisières, due à la coupe totale dans les forêts boréales, ne réduit pas nécessairement la qualité des parties de forêt restantes pour la saison hivernale des mésanges à tête brune et ce, même pendant des températures inclémentes. Par contre, j’arrive à la conclusion que l’exploitation forestière réduira l’habitat hivernal optimal de cette espèce. ii

Abstract

Resident bird species inhabiting northern latitudes are considered to be the species most exposed to the effects of habitat loss and fragmentation of boreal forests. Despite the fact that their population dynamics appear to be strongly determined by events occurring during the non-breeding season, we have little knowledge of the winter ecology of boreal birds. My objective was to determine how increasing edge densities and reducing the proportion of mature boreal forest will affect a resident bird species. I recorded movements of 85 unmarked and seven colour banded winter flocks of the little-known Boreal Chickadee (Poecile hudsonica), in a 66 km2 boreal forest harvested for timber near Quebec City, Quebec, Canada. From January-March (2004 and 2005), I followed flocks on snowshoes and recorded their paths in real time using a handheld GPS receiver. Using marked individuals, I found winter Boreal Chickadee flocks included an average of 4 individuals, occupied a mean winter home range of 14.7 ha and showed stable membership. Based on 74 km of flock movements, Boreal Chickadees strongly preferred mature forest (>7 m in height), used regenerating forest (4-7 m) to a lesser extent and avoided younger stands (<4 m) and open areas. Chickadee flocks showed no response to forest edges when using mature forest stands. However, inside regenerating forest, flocks were significantly closer to both open edges (41 ± 6 m) and mature forest boundaries (11 ± 2 m) than would be expected from random use of the habitat. Boreal Chickadee flocks did not avoid exposed edges during harsh weather conditions. In fact, on colder days, they were found disproportionately more often along edges between mature and regenerating stands. Increasing edge densities, resulting from clearcutting in boreal forest, does not necessarily reduce the winter suitability of remaining forest patches, even under inclement weather. However, I conclude that forest harvesting will result in a reduction of optimal wintering habitat for this species.

iii

Preface

I would like to thank my director André Desrochers for his help with the conception of this project and his assistance with the choice and implementation of the analysis involved. André’s enthusiasm and expertise definitely helped to encourage me throughout the execution of the study. I would also like to thank my family and the Fundy Wolfe Lake bird crew for providing me with the enthusiasm and drive necessary to accomplish the physically demanding fieldwork involved during the study. The Chickadee flocks followed during this project deserve special mention for their unwavering ability to lead me across the steepest ravines and through the toughest terrain at their disposal. Lastly, a special thanks to all of my lab mates who helped me with my French and provided helpful advice.

This mémoire contains two scientific articles submitted to international caliber journals. I will be first author and André Desrochers second author.

CHAPTER 1: Hadley1, A. S., and A. Desrochers1. Winter habitat use by Boreal Chickadee flocks. To be submitted to The Wilson Journal of Ornithology (Formerly the Wilson Bulletin) 1Centre de recherche en biologie forestière, Faculté de foresterie et de géomatique, Université Laval, Québec, Québec, G1K 7P4, Canada.

CHAPTER 2: Hadley1, A. S., and A. Desrochers1. Response of wintering Boreal Chickadees to forest edges: Does weather matter? Submitted to The Auk, February 28, 2006. 1 Centre de recherche en biologie forestière, Faculté de foresterie et de géomatique, Université Laval, Québec, Québec, G1K 7P4, Canada. iv

Eric, Jane and Matthew v

Table of contents

Résumé...... i Abstract...... ii Preface ...... iii Table of contents...... v List of Tables ...... vii List of figures...... viii General Introduction...... 1 Edge effects and landscape structure...... 1 Concerns in boreal regions ...... 2 The importance of winter ...... 2 The Boreal Chickadee ...... 3

CHAPTER 1: Winter habitat use by Boreal Chickadee flocks ...... 5 RÉSUMÉ ...... 6 ABSTRACT...... 7 INTRODUCTION ...... 7 METHODS ...... 9 Study area ...... 9 Use of forest stands...... 9 Home range estimation...... 13 Correlates of home range size ...... 15 RESULTS ...... 16 Use of forest stands...... 16 Flock and home range size ...... 16 Landscape structure and home range size ...... 18 DISCUSSION...... 20 ACKNOWLEDGEMENTS...... 22 LITERATURE CITED...... 23

CHAPTER 2: Response of wintering Boreal Chickadees to forest edges: Does weather matter? ...... 26 RÉSUMÉ ...... 27 INTRODUCTION ...... 28 METHODS ...... 30 Study area ...... 30 Flock following technique...... 31 Response to edges...... 33 RESULTS ...... 34 From mature forest ...... 34 From regenerating forest ...... 35 Effects of weather...... 37 Edge orientation...... 39 DISCUSSION...... 39 vi

Response to edges...... 39 Effects of weather on edge association...... 42 ACKNOLWLEDGEMENTS ...... 44 LITTERATURE CITED...... 45

General Conclusion...... 50 Management implications...... 53 Bibliography for general introduction and conclusion ...... 55 Appendix A...... 59 vii

List of Tables

Chapter 1

Table 1. Relationship between home range size, and landscape composition, flock size and the usage pattern for seven Boreal Chickadee flocks...... 18

Chapter 2

Table 1. Difference between expected and observed distances to habitat edges for Boreal Chickadee flocks...... 35

Table 2. Relationship between mean association with edge and weather conditions while flocks were using regenerating forest ...... 37

Table 3. The effect of orientation on the association with edges of regenerating forest patches ...... 39

viii

List of figures

Chapter 1

Figure 1. Systematic search pattern used to locate flocks ...... 10

Figure 2. Distribution of movement paths from Boreal Chickadee flocks followed at Forêt Montmorency...... 12

Figure 3. Black-capped Chickadee banded during the study showing color bands fitted with flags of electrician’s tape for enhanced visibility ...... 14

Figure 4. Minimum convex polygon estimates of home ranges for seven Boreal Chickadee flocks at Forêt Montmorency...... 17

Figure 5. Association between MCP home range size and proportion of open area within the home range or proportion of mature stands within the home range ...... 19

Chapter 2

Figure 1. Distribution of weather conditions during which flocks were followed ...... 34

Figure 2. Mean locations of Boreal Chickadee flocks with respect to forest edges ...... 36

Figure 3. Response to mature forest edge under different weather conditions when flocks were using regeneration forest ...... 38

Conclusion

Figure 1. The amount of forest habitat available at Forêt Montmorency if edges are not avoided, regions within 30m of open edges are avoided or regions within 50m of open edges are avoided ...... 52

General Introduction

Edge effects and landscape structure

Loss and fragmentation of forests have motivated numerous studies investigating the effects of remaining habitat area, patch size and forest edges on avian distributions. Birds are often sensitive to both amount and configuration of remaining habitat patches. Hence, many studies have tried to categorize bird species as being “area sensitive” (requiring a certain minimal area) and “edge or interior” (being associated with edges or avoiding them respectively) in attempts to simplify management strategies (Villard 1998). Several studies have linked the amount of available habitat with presence and abundance of forest birds (Askins et al. 1987, Trzcinski et al. 1999, Drapeau et al. 2000). Certain bird species are sensitive to habitat patch size and appear to avoid small patches (Robbins et al. 1989). Smaller habitat patches may also confer adverse effects on their users in terms of lower reproductive success (Robinson et al. 1995) and decreased survival (Doherty and Grubb 2002). Configuration of the remaining habitat and habitat fragmentation are also thought to be important. The ratio of edge per unit area tends to increase as remaining forest patches become smaller and more fragmented (Forman 1995). Habitat zones between forest and non-forest habitat often differ from areas deeper within the interior of patches. Several studies have demonstrated changes in vegetation (Chen et al. 1992, Fraver 1994), arthropod abundance (Jokimäki et al. 1998, Haddad and Baum 1999), and local microclimate (Blake and Karr 1987, Chen et al. 1993) with varying distance from edges. These biotic and abiotic changes, referred to as “edge effects”, can in turn affect the animals using regions next to edges. Much of the edge effect research has been conducted using birds. This emphasis has been largely placed on avian species diversity (Yahner 1988), territory location (Yahner 1988), nest predation (Lahti 2001) and brood parasitism (Paton 1994) which have all been shown to differ near edges. Recent studies have demonstrated further effects of edges on avian movements. Edges have been shown to affect foraging patterns within breeding territories (Huhta et al. 1999) and wintering home ranges (Dolby and Grubb 1999, Desrochers and Fortin 2000). 2

Concerns in boreal regions

North American boreal forests comprise one of the largest intact ecosystems on Earth today (Canadian Boreal Initiative 2003). However, technological advances in forest harvesting have enabled the logging of previously un-economically viable forest within this region. Boreal forest regions in Quebec, Canada, are being subjected to intensive forest exploitation and ~300 000 ha are logged annually (Ministère des Ressources naturelles et de la Faune 2006). The extensive logging has resulted in forest habitat loss, fragmentation and landscape conversion. Large tracts of forest are disappearing, being replaced by open habitat and younger sucessional stages (Ministère des Ressources naturelles et de la Faune 2006). Indeed, the regional decline in proportion of mature forest stands has been suggested to be one of the most concerning changes induced by forestry (Imbeau et al. 1999). In Quebec, remaining forest patches are typically linear and restricted to riparian buffer strips or buffers between adjacent clearcuts (Ministère des ressources naturelles du Québec 1996). Despite the increasing conservation concerns associated with boreal regions, we know very little about the requirements of bird species inhabiting them. In Europe, where boreal forests have been intensively harvested for many years, changes in avian abundance have been well documented. Many bird species found in European boreal forests have undergone marked population declines, thought to be linked to forest harvesting (Imbeau et al. 2001). However, the effects that intensifying forest harvesting will have on North American avian populations are little known, particularly during winter.

The importance of winter

Among birds, resident species have been proposed to be those most at risk of population declines due to habitat loss and fragmentation in boreal forests (Imbeau et al. 2001, Schmiegelow and Mönkkönen 2002). Furthermore, the population dynamics of resident species inhabiting northern latitudes appear to be strongly determined by events occurring during the non-breeding season (Matthysen 1990, Lahti et al. 1998, Doherty and Grubb 2002). Wind and temperature can interact to affect the metabolic rate and energy 3 expenditure of birds (Porter and Gates 1969, Wolf and Walsberg 1996) and in turn, reduce their survival rate (Mayer et al. 1979). Therefore, careful selection of winter foraging sites is thought to play an important role in reducing the energy costs of birds found in harsh winter environments (Wachob 1996). Changes of microhabitat selection in response to adverse thermal conditions have been well documented in small resident birds. Winter residents have been found to move horizontally less often (Kessel 1976, Grubb 1978) and decrease foraging heights (Grubb 1975, 1977) during cold, windy, conditions. Increasing wind strength and decreasing temperature can also result in a tendency for birds to shift to leeward positions on foraging substrates (Grubb 1977). Birds, in highly fragmented deciduous forests, did not use exposed forest edges during harsh winter weather conditions, and were found further from windward edges of woodlots at high winds and low temperatures (Dolby and Grubb 1999). Increasing edge density or reducing patch size in regions subjected to harsh winters can increase energy expenditures of birds beyond tolerable limits, resulting in a reduction of effective habitat available for use (Blake 1987, Dolby and Grubb 1999). However, most of the studies examining winter weather effects have been conducted in deciduous forests and agricultural landscapes. Hence, abiotic edge effects on northern resident species in conifer- dominated forests remain poorly understood.

The Boreal Chickadee

The Boreal Chickadee (Poecile hudsonica) is a boreal forest resident which exemplifies our relatively weak knowledge of wintering birds within boreal regions. Boreal Chickadees have shown alarming declines along breeding bird survey routes (mean annual change 1966-2004 (%) = -3.59, p-value = 0.0035) in Eastern North America (Sauer et al. 2005) and are listed as highly vulnerable to changes induced by modern forestry (Imbeau et al. 2001). Winter survival is thought to set population limits for this species (Erskine 1977, 1992) and concern has been expressed over winter suitability of the remaining habitat (Erskine 1992, Foss 1994, Cyr and Larivée 1995). Boreal Chickadees have been little 4 studied and despite apparent declines, our knowledge of their wintering requirements is limited (Ficken et al. 1996). The purpose of this study was to examine the winter habitat requirements of Boreal Chickadees and determine their sensitivity to changes in landscape configuration resulting from forest harvesting. The study provided relevant information for a species and a biome of conservation interest, during a little studied season. The first chapter of my study provided information on basic wintering biology of Boreal Chickadee flocks. I was able to collect previously unknown information on winter home range size and characteristics. I documented flock size, composition and stability throughout the winter months. I examined use of habitat by Boreal Chickadee flocks and determined which habitats were considered “optimal” and which were avoided. In my second chapter I used movements of Boreal Chickadee flocks in forest patches left by timber harvest, to determine whether they occur predominantly in the forest interior or near edges. I also examined the effects of winter weather conditions on edge associations within chickadee home ranges.

CHAPTER 1: Winter habitat use by Boreal Chickadee flocks

RÉSUMÉ

Même si on s’intéresse de plus en plus aux questions de conservation dans les régions boréales, on connaît très peu les exigences hivernales des espèces qui habitent cet habitat. Nous avons examiné la taille des volées, l’habitat hivernal préféré et la taille du domaine vital des mésanges à tête brune (Poecile hudsonica), une espèce dont l’écologie hivernale est méconnue. Notre étude s’est étendue de janvier à mars (2004 et 2005) dans une aire de 66 km2 de forêt boréale, sous aménagement forestier, près de la ville Québec, Québec, Canada. Les volées comprenaient en moyenne 4 individus et occupaient un domaine vital moyen de 14.7 ha. Les individus et leur nombre dans une volée ne variaient pas pendant les mois d’hiver. Les mésanges à tête brune préféraient nettement les peuplements forestiers mature de valeur marchande (>7 m de hauteur) et utilisaient de moindre façon les peuplements en régénération (4-7 m). Les mésanges évitaient les jeunes peuplements (<4 m) et les milieux ouverts. La taille des domaines vitaux n’était pas influencée par la composition du paysage. Cependant, les volées qui avaient un domaine vital plus grand, l’utilisaient de façon moins uniforme que celles qui avaient de plus petit domaine vital. Puisque cette espèce préfère les peuplements forestiers matures qui ont une valeur marchande, l’exploitation forestière peut contribuer au déclin apparent de la population de cette espèce. 7

ABSTRACT

Despite the increasing conservation concerns associated with boreal regions, very little is known about winter habitat requirements of bird species inhabiting them. We examined flock size, winter habitat preference and home range size of Boreal Chickadees (Poecile hudsonica), a species whose winter ecology is poorly documented. Our study was conducted from January-March (2004 and 2005) in a 66 km2 boreal forest harvested for timber near Quebec City, Quebec, Canada. Flocks included an average of 4 individuals and occupied a mean winter home range of 14.7 ha. Flock membership and size were stable during the winter months. Boreal Chickadees strongly preferred mature stands of commercial value (>7 m in height) and used regenerating stands (4-7 m) to a lesser extent. Younger stands (<4 m) and open areas were avoided. Home range size was not influenced by landscape composition. Yet, flocks with larger home ranges used their home ranges less evenly than those with smaller home ranges. Since stands of commercial value are preferred by this resident species, logging may contribute to apparent population declines in this species.

INTRODUCTION

Logging is considered the most important threat to species inhabiting boreal forest regions (Imbeau et al. 2001). Logging is rapidly modifying North America’s boreal forests. In Quebec, Canada, alone approximately 300 000 ha of boreal forest have been clearcut annually in recent years (Ministère des Ressources naturelles et de la Faune 2006). These logging practices are leading to a dramatic regional reduction in large tracts of mature forest with a subsequent increase in younger successional stages. Remaining forest patches are typically restricted to riparian buffer strips and buffers between adjacent clearcuts (Ministère des ressources naturelles du Québec 1996). Among birds, resident species are hypothesized to be most exposed to loss and fragmentation of boreal forests (Imbeau et al. 2001, Schmiegelow and Mönkkönen 2002). 8

Furthermore, population dynamics of resident species inhabiting northern latitudes appear to be strongly determined by events occurring during the non-breeding season (Matthysen 1990, Lahti et al. 1998, Doherty and Grubb 2002). Indeed many resident species in habiting European boreal forests have undergone marked population declines hypothesized to result from forest harvesting (Imbeau et al. 2001). Despite the increasing conservation concerns associated with boreal regions, very little is known about winter habitat requirements of bird species inhabiting them. The Boreal Chickadee (Poecile hudsonica) is a ~10 g boreal forest resident which exemplifies the lack of knowledge of wintering birds within these regions. Despite its weak sampling effort for boreal species, the Breeding Bird Survey reports an alarming decline of Boreal Chickadees (mean annual change 1966-2004 (%) = -3.59, p = 0.0035) in Eastern North America (Sauer et al. 2005). Furthermore, the species is listed as highly vulnerable to changes induced by modern forestry (Imbeau et al. 2001). The Boreal Chickadee is often considered the North American ecological equivalent of the Siberian Tit (P. cincta), a species that has also undergone dramatic declines due to the effects of forestry (Imbeau et al. 2001). Flocks of Boreal Chickadees form as soon as the young fledge and persist throughout the non-breeding season (September to late April; Ficken et al. 1996). Mixed species flocks form occasionally with Black-capped Chickadees (P. atricapillus) and Red- breasted Nuthatches (Sitta canadensis) in the study area. Winter home range characteristics and habitat use, despite their obvious implications for management, are virtually unknown for this species (Ficken et al. 1996). Winter survival is thought to set population limits for Boreal Chickadees (Erskine 1977, 1992) and concern has been expressed over winter suitability of the remaining habitat (Erskine 1992, Foss 1994, Cyr and Larivée 1995). Here we examine winter habitat use, flock size and home range size of Boreal Chickadees in a region managed for timber and recreational use. We determined which forest seral stages were preferred or avoided by wintering Boreal Chickadees. We examined flock size, composition and stability during winter months. We tested whether the size of Boreal Chickadee home ranges is associated with changes in landscape composition. If a minimal quantity of optimal habitat is required for winter survival, then flocks should increase home range size in response to inclusion of sub-optimal habitat (Matthysen 1990). We also examined space use within home ranges, with the assumption 9 that the pattern in which individuals use their home range reflects asymmetries in quality. We predicted that flocks with large home ranges would distribute their activities in a few distinct optimal locations (Carr and Macdonald 1986) and flocks with smaller home ranges would distribute their activities more evenly throughout their home range.

METHODS

Study area We collected data during the winters of 2004 and 2005 at the Forêt Montmorency research forest, Quebec, Canada (47o20’-71o10’W). The study area is a 66 km2 boreal forest mosaic managed for timber exploitation and recreational use. Mature coniferous stands (>7 m of height) cover approximately 56% of the study area. Balsam fir (Abies balsamea) and occasionally black spruce (Picea mariana) dominate mature stands, interspersed with white birch (Betula papyrifera). Younger seral stages (4-7 m) are characterized by sapling balsam fir, black spruce and to a lesser extent mixed regeneration. These cover 24% of the study area. Open areas consisting of clearcuts, lakes, rivers, roads >7 m in width and sapling stands < 4 m in height cover 20% of the study area. An extensive road network (2.6 km/ha) crosses the research forest.

Use of forest stands We investigated habitat use by means of 74 km of movement data recorded for 85 unmarked Boreal Chickadee flocks. Seventy-two flocks were studied in the first study winter (Jan 6, 2004 – March 10, 2004) and 13 flocks during the following winter (Feb 14, 2005 - March 24, 2005). We located chickadee flocks each day using randomly selected points on a systematic 1 km spaced grid covering the study area. Grid points were visited only once during the study. We located flocks by snowshoeing a systematic search pattern starting from the selected gridpoint (Figure 1). 10

Figure 1. Systematic search pattern used to locate flocks. Starting from the selected grid point we moved 500 m north, 500 m east, 1000 m south, 500 m west and 500 m north to return to the point of origin.

All flock detections were passive (no use of playback). Although flocks were unmarked, we assume that a new flock was monitored each day, since flock composition differed in all but three cases of adjacent flocks (known to be separate do to simultaneous observations). We followed flocks on snowshoes and logged their locations in real time at one-minute intervals using a hand held TrimbleTM GPS receiver (PDOP < 8). We used a filter based on the time interval and distance between consecutive positions to eliminate possible “spikes” or imprecise positions not representative of the actual flock locations (Appendix A). The standard deviation of point location estimates, based on 1680 stationary points, was found to be 1.37 m under forest canopy. One-minute sampling intervals were used since serial correlation is irrelevant when using the proportion of an animal’s trajectory contained within each habitat type for Compositional Analysis (Aebischer et al. 1993). Frequent sampling more closely approximates the underlying trajectory and provides a more precise estimate of proportional habitat use (Aebischer et al. 1993, Barg et al. 2005). No positions were recorded during the first 2.5 minutes following discovery. We followed each flock for as long as possible up to a limit of 3 hours. Positions were recorded only when we were located at the approximate center of the flock and data logging ceased 11 immediately if observer position no longer represented that of the flock. The following period for a flock was terminated if the observer lost contact. We then moved to another grid sampling point before recommencing the search for different flock. The time for which we followed flocks ranged from 4.7–152 min (mean = 53 min) and path lengths ranged from 90 m to 2125 m (mean = 871 m). The total distance traveled for the 85 flocks was 74 km and the total following time was 75 hours (Figure 2). We observed no discovery bias since there was no relationship between stand type and elapsed time since initial discovery (R2= 0.0011, F = 3.44, P = 0.06).

Figure 2. Distribution of movement paths from Boreal Chickadee flocks followed at Forêt Montmorency, 2004-2005. Movement paths are shown in blue. Mature forest stands are shown in dark green, regeneration stands are pale green and open areas are in white.

We divided the habitat within the study area into three categories: (1) mature forest (stand >7 m in height), (2) regeneration (4-7 m in height), and open areas (stands <4 m in 13 height or areas devoid of vegetation above snow). Stand seral stages were characterized using existing GIS coverage for the study area (validated in situ, delimited by GPS, and mapped with ArcView 3.3 [ESRI 2002]). In order to establish a measure of available habitat representative of home range size (Jones 2001) we delineated a 200 m buffer for each flock surrounding all locations. We generated a 10 m spaced grid of points within each buffer. The resulting grid points were assumed to represent unbiased samples of the habitat due to the apparent lack of spatial periodicity in the stand types of the study area. In order to determine if used habitat (observed locations) differed form available habitat within 200 m (grid points), we used Compositional Analysis (Aebischer et al. 1993). We replaced missing values in log ratios (available but not used) with 0.001, at least one order of magnitude less than the smallest non-zero value for that habitat type, as suggested in Aebischer (1993). We used SAS (SAS Institute Inc. 2004) and the BYCOMP.SAS macro (Ott and Hovey 1997) to compute the randomization procedure recommended by Aebischer et al. (1993). Only flocks having all three habitat types represented within the “available habitat” were used in the analysis. The latter constraint reduced the sample size to 79 flocks.

Home range estimation We determined home range sizes using data collected from seven color-marked flocks of Boreal Chickadees during the winter of 2004-2005. Twenty-three of the flock members were captured from December 6, 2004 to February 3, 2005 using mist nets and playbacks of chickadees mobbing a stuffed owl. We marked birds with Fish and Wildlife numbered aluminum bands and unique color combinations. To enhance visibility, we fitted color bands with ~ 1 cm long flags of colored electrician’s tape which were visible for up to 10 m away without the use of binoculars (Desrochers 1988; Figure 3).

Figure 3. Black-capped Chickadee banded during the study showing color bands fitted with flags of electrician’s tape for enhanced visibility. This method was also used for all Boreal Chickadees banded in this study. Photo taken by Claude Mailloux at Forêt Montmorency, February 2005.

Each flock received seven visits between February 7, 2005 and March 24, 2005. We randomized the order and time of visits among flocks and successive visits were separated by 2-7 days. We detected flocks using short (5 second) bursts of boreal chickadee calls on a portable speaker audible up to 70 m away and by searching on snowshoes in concentric circles around the initial capture locations. Bursts of playback were short and restricted in frequency to limit effects on flock movements. If answering calls were heard we ceased playbacks and proceeded directly to where the birds were located. At the beginning of each visit we identified flocks by band combinations and flock composition was recorded. During each visit we followed flocks using the same procedure as with unmarked flocks (see above). The time for which we followed marked flocks during each visit ranged from 5-165 min (mean = 43 min). We recorded successive locations when they were separated by at least five minute intervals (Points were considered biologically independent since chickadees could easily move to any point within their home range during this time period (Barg et al. 2005). The number of locations per flock ranged from 48 to 83 (mean = 69 ± 5 points, [mean ± SE]). 15

We determined home range size using both the minimum convex polygon (MCP) and 95% kernel methods. The MCP was used to allow comparability with other studies of parids (Harris et al. 1990). However, the MCP allows little insight into internal configuration of used spaces, is highly affected by peripheral locations and can contain larger areas never used by the organism (Harris et al. 1990, Barg et al. 2005). To address these shortcomings we also used a fixed kernel density estimator to study the density distribution of observations and to construct each flock’s utilization distribution (UD). We used the 95% fixed kernel to define the kernel home range size and the 50% fixed kernel area (containing 50% of the observations) to define the core area for each flock. Fixed kernel density estimations were performed using the Animal Movement extension in ArcView 3.3 (Hooge and Eichenlaub 1997). For each flock we used least squares cross validation (LSCV) to determine the optimal smoothing parameter. Least squares cross validation provides the least biased estimates for smoothing parameters (Worton 1995, Seaman and Powell 1996).

Correlates of home range size We compiled the proportion of different stand seral stages within MCP and kernel home ranges, and examined their association with home range sizes. We also calculated a new variable (USE) representing the distribution of locations within the home range. USE was determined by calculating the difference between the 95% and 50% (CORE) kernel estimations (see Siffczyk et al. [2003] for rationale). The new USE variable provided information on whether the flock locations were concentrated in distinct areas of the home range (high values of use) or evenly distributed throughout the home range (low USE values). Flocks with home ranges containing patchy, widely distributed, resources would be expected to have large USE values (concentrate their time in select locations of optimal habitat) and hence larger home range sizes. Flocks with evenly distributed resources would be expected to have lower USE values (more even use of home range) and smaller home ranges. 16

The association between landscape components and home range size was examined using Spearman rank correlations (SAS Institute Inc. 2004). Effects were considered significant at α = 0.05.

RESULTS

Use of forest stands Boreal Chickadee flocks did not use mature forest, regenerating forest and open areas at random (Wilks’ Lambda = 0.22, F= 134, P= <0.001, n = 79). Mature forest was used more frequently than regenerating forest (mean log-ratio = 7.16, P = <0.01) or open areas (mean log-ratio = 16.44, P = <0.01), and regenerating forest was used more frequently than open areas (mean log-ratio = 3.60, P = <0.01).

Flock and home range size Using data from all 85 flocks of Boreal Chickadees we found that the mean flock size was 4 ± 0.2 (mean ± SE) with a range from three to eight individual boreals. Sixteen of 85 flocks contained at least one Black-capped Chickadee (4 ± 0.4 [mean ± SE], range 1-10 within these 16 flocks) and Red-breasted Nuthatches were present in 27 of 85 flocks (4 ± 0.3 [mean ± SE], range 2-8 within the 27 flocks). Nine flocks contained all three species. Black-capped Chickadees appeared to form cohesive flocks with Boreal Chickadees. They remained in close contact with Boreal Chickadees throughout the entire following period. By contrast, Red-breasted Nuthatches were associated with Boreal Chickadees only as loose (followed at a distance, often leaving and rejoining flock) foraging groups. Using marked individuals we determined that flock membership and size remained stable throughout winter months (January-March). Over a six-week period, we observed no apparent immigration, emigration, or mortality in the seven marked flocks. No banded individuals disappeared or moved between flocks and the total number of individuals within the flocks remained unchanged. Boreal chickadee flocks occupied a mean home range size of 14.7 ± 3.2 ha (MCP; mean ± SE, n = 7) with a range from 7.9 to 30.4 ha (Figure 4). Using the 95% fixed kernel, 17 mean home range size was actually larger, 16.9 ± 3.4 ha (95% kernel; mean ± SE, n = 7) with a range from 7.6 to 33.9 ha. Flocks had a mean core area of 2.3 ± 0.7 ha (50% kernel mean ± SE, n = 7) with a range from 1.0 to 5.8 ha.

Figure 4. Minimum convex polygon estimates of home ranges for seven Boreal Chickadee flocks at Forêt Montmorency. MCPs are represented in blue and flock locations are in black. Mature forest stands are shown in dark green, regeneration stands are pale green and open areas are in white. 18

Landscape structure and home range size On average, boreal chickadee MCP home ranges contained 61 ± 9% mature forest (range 21-94%), 23 ± 10% regeneration forest (range 0-79%) and 15 ± 6% open area (range 0-43%). These values are similar to the distribution of stands within the study area (see Methods, Study area). We found no significant relationship between landscape components and MCP home range size (Table 1). Yet, due to our small sample we feel that the trend of increasing MCP home range size with increasing proportion of open area (Fig 5a) and decreasing home range size with increasing proportion of mature forest (Fig 5b) should be noted. Flocks with larger home ranges concentrated their use in several distinct areas (large differences between 95% and 50% core areas) while flocks with smaller home ranges used their home range more evenly (Table 1). Home range size was not associated with flock size (Table 1).

Table 1. Relationship between home range size, and landscape composition, flock size and the usage pattern within the home range for seven Boreal Chickadee flocks. Correlation coefficients are shown with p-values in parentheses (Spearman rank correlation). Home range size is the 100% minimum convex polygon (ha); USE is the difference between the 95% and 50% kernels (ha); Habitat proportions represent proportions within MCP.

Variable Home range size Proportion of mature stands in home range -0.57 (0.2) Proportion of regeneration stands in home range 0.14 (0.8) Proportion of open area in home range 0.39 (0.4) Flock size 0.26 (0.6) USE 0.89 (0.007)

Figure 5. Association between MCP home range size and a) proportion of open area within the home range and b) proportion of mature stands within the home range.

DISCUSSION

Boreal Chickadee flocks are quite similar to wintering flocks of other northern parid species in terms of size and member stability. Mean flock size in this study (4 ± 0.2 individuals) was slightly larger than Siberian Tit winter flocks, which average between two and three individuals (Virkkala 1990). Willow tits (Parus montanus) form, on average, four-bird flocks (Ekman 1989, Siffczyk et al. 2003), while Black-capped Chickadees usually form slightly larger flocks with six to eight individuals (Smith 1991). Our flocks showed member stability throughout the winter months, as is thought to be the case with most parids (Desrochers and Hannon 1989, Ekman 1989). Like many wintering parids (Hogstad 1987, Smith 1991), Boreal Chickadees did form mixed species flocks, but most often maintained flocks solely composed of conspecifics. Boreal Chickadee flocks preferred mature forest stands and avoided young successional stages or open areas. While mature forest stands (>7 m) were strongly preferred to regeneration stands, flocks frequently did spend time in regenerating forest habitat. Boreal chickadee flocks only rarely spent time in open areas (when lone trees were available or when crossing gaps). Our results are consistent with information on breeding habitat use by Boreal Chickadees. Whitaker et al. (1997) consider the boreal chickadee to be a forest generalist within coniferous wooded areas. Erskine (1977) also showed boreal chickadees to use both mature and young forest during the breeding season. Our flocks had winter home range sizes of 14.7 ha, comparable to those found for Black-capped Chickadees (9.5-14.6 ha, [Smith 1991]; 22.4 ha [Desrochers and Fortin 2000]) and Willow Tits (12.6 ha, [Siffczyk et al. 2003]) during the non-breeding season. The sizes of winter home ranges were not significantly associated with landscape composition. However, our flocks did show similar tendencies of increasing home range size with inclusion of a larger proportion of non-habitat, to those demonstrated previously in other studies (Gjerde and Wegge 1989, Storch 1993, Siffczyk et al. 2003). We conclude that the differences in space use patterns depending on the size of winter home range likely reflect the patchiness of resources within large home ranges. Flocks with large home ranges focused their activity in distinct locations within their home 21 range while flocks with smaller home ranges distributed their activities more evenly across space. These results agree with the resource dispersion hypothesis (RDH), which predicts that home ranges will be large when patches of resources are widely spaced (Carr and Macdonald 1986). Similar results have been shown for Willow Tits (Siffczyk et al. 2003). Boreal forests in Eastern North America are being subjected to intensive forest exploitation (Ministère des Ressources naturelles et de la Faune 2006). The extensive logging has resulted in a reduction in the proportion of mature forest, with a subsequent increase in proportion of young forest or open areas (Imbeau et al. 1999, Ministère des Ressources naturelles et de la Faune 2006). Consequently, forestry practices will result in a substantial reduction of optimal Boreal Chickadee wintering habitat, at least over several decades. Our findings are consistent with the hypothesis that apparent population declines in this species result from the loss of high-quality wintering habitat. 22

ACKNOWLEDGEMENTS We thank David Duchesne, Marc Lamarre and Geneviève d’Anjou for their help in the fieldwork. We are grateful to the staff at Forêt Montmorency (Université Laval) for logistical support. M. Betts, and M. Hadley provided useful comments on the manuscript. This study was funded by a NSERC Discovery grant to A. Desrochers and a NSERC Postgraduate scholarship (PGS A) to A. Hadley. 23

LITERATURE CITED

Aebischer, N. J., P. A. Robertson, and R. E. Kenward. 1993. Compositional analysis of habitat use from animal radio-tracking data. Ecology 74:1313-1325.

Barg, J. J., J. Jones, and R. J. Robertson. 2005. Describing breeding territories of migratory passerines: suggestions for sampling, choice of estimator, and delineation of core areas. Journal of Animal Ecology 74:139-149.

Carr, G. M., and D. W. Macdonald. 1986. The sociality of solitary foragers - a model based on resource dispersion. Animal Behaviour 34:1540-1549.

Cyr, A., and J. Larivée. 1995. Atlas saisonnier des oiseaux du Québec. Presses de l'Université de Sherbrooke et Société de loisir ornithologique de l'Estrie, Sherbrooke, Québec, Canada.

Desrochers, A., and M.-J. Fortin. 2000. Understanding avian responses to forest boundaries: a case study with chickadee winter flocks. Oikos 91:376-384.

Desrochers, A., and S. J. Hannon. 1989. Site-related dominance and spacing among winter flocks of Black-capped Chickadees. Condor 91:317-323.

Doherty, P. F., Jr., and T. C. Grubb, Jr. 2002. Survivorship of permanent-resident birds in a fragmented forested landscape. Ecology 83:844-857.

Ekman, J. 1989. Ecology of non-breeding social systems of Parus. Wilson Bulletin 101:263-288.

Erskine, A. J. 1977. Birds in boreal Canada: communications, densities, and adaptations. Canadian Wildlife Service report series no. 41. Canadian Wildlife Service, Ottawa, Ontario, Canada.

Erskine, A. J. 1992. Atlas of breeding birds of the Maritime Provinces. Nimbus Publications and Nova Scotia Museum, Halifax, NS.

ESRI. 2002. ArcView® 3.3 GIS. Environmental Systems Research Institute, Inc., Redlands, CA, U.S.A.

Ficken, M. S., M. A. McLaren, and J. P. Hailman. 1996. Boreal Chickadee. in The birds of North America, No. 254 (A. Poole, and F. Gill, Eds.). The Academy of Natural Sciences, The American Ornithologists' Union, Philadelphia, PA, Washington, D.C.

Foss, C. R., Ed. 1994. Atlas of breeding birds in New Hampshire. Audubon Society, New Hampshire, Dover. 24

Gjerde, I., and P. Wegge. 1989. Spacing pattern, habitat use and survival of capercaillie in a fragmented winter habitat. Ornis Scandinavica 20:219-225.

Harris, S., W. J. Cresswell, P. G. Forde, W. J. Trewhella, T. Woollard, and S. Wray. 1990. Home-range analysis using radio-tracking data - a review of problems and techniques particularly as applied to the study of mammals. Mammal Review 20:97-123.

Hogstad, O. 1987. Social rank in winter flocks of Willow Tits Parus montanus. Ibis 129:1- 9.

Hooge, P. N., and B. Eichenlaub. 1997. Animal movement extension to arcview, version 2.0. Alaska Biological Science Center, U.S. Geological Survey, Anchorage, AK, U.S.A.

Imbeau, L., M. Mönkkönen, and A. Desrochers. 2001. Long-term effects of forestry on birds of the Eastern Canadian boreal forests: a comparison with Fennoscandia. Conservation Biology 15:1151-1162.

Imbeau, L., J.-P. L. Savard, and R. Gagnon. 1999. Comparing bird assemblages in successional black spruce stands originating from fire and logging. Canadian Journal of Zoology 77:1850-1860.

Jones, J. 2001. Habitat selection studies in avian ecology: a critical review. Auk 118:557- 562.

Lahti, K., M. Orell, S. Rytkonen, and K. Koivula. 1998. Time and food dependence in Willow Tit winter survival. Ecology 79:2904-2916.

Lair, H. 1987. Estimating the location of the focal center in red squirrel home ranges. Ecology 68:1092-1101.

Matthysen, E. 1990. Nonbreeding social organization in Parus. Current Ornithology 7:209- 249.

Ministère des ressources naturelles du Québec. 1996. Règlement sur les normes d'intervention dans les forêts du domaine public. Pages 2750-2786 in Gazette Officielle du Québec, vol. 128.

Ministère des Ressources naturelles et de la Faune. 2006. Ressources et industries forestières : Portrait statistique Édition 2005-2006. Gouvernement du Québec, 1-79 Pp.

Ott, P., and F. Hovey. 1997. BYCOMP.SAS. Clearinghouse fore Ecology Software.

SAS Institute Inc. 2004. SAS software. Version 9.1. SAS Institute Inc., Cary, N.C., U.S.A. 25

Sauer, J. R., J. E. Hines, and J. Fallon. 2005. The North American breeding bird Survey, results and snalysis 1966 - 2004. USGS Patuxent Wildlife Research Center, Laurel, MD, U.S.A.

Schmiegelow, F. K. A., and M. Mönkkönen. 2002. Habitat loss and fragmentation in dynamic landscapes: avian perspectives from the boreal forest. Ecological Applications 12:375-389.

Seaman, D. E., and R. A. Powell. 1996. An evaluation of the accuracy of kernel density estimators for home range analysis. Ecology 77:2075-2085.

Siffczyk, C., L. Brotons, K. Kangas, and M. Orell. 2003. Home range size of Willow Tits: a response to winter habitat loss. Oecologia 136:635-642.

Smith, S. M. 1991. The Black-capped Chickadee: behavioural ecology and natural history. Cornell University Press, Ithaca, N.Y., U.S.A.

Storch, I. 1993. Patterns and strategies of winter habitat selection in alpine capercaillie. Ecography 16:351-359.

Virkkala, R. 1990. Flock size of the Siberian Tit Parus cinctus during the non-breeding season. Ornis Fennica 67:98-99.

Worton, B. J. 1995. Using monte-carlo simulation to evaluate kernel-based home range estimators. Journal of Wildlife Management 59:794-800.

CHAPTER 2: Response of wintering Boreal Chickadees to forest edges: Does weather matter? 27

RÉSUMÉ

La plupart des études sur la réaction des oiseaux aux lisières forestières (délimitant les peuplements) ont examiné la diversité des espèces, la localisation des territoires ou la prédation des couvées. Très peu d'études ont examiné les effets des lisières forestières sur les espèces résidentes pendant l'hiver. Pourtant, en période de vent intense, des conditions microclimatiques sévères peuvent pénétrer profondément dans la forêt, et pourraient rendre les lisières de forêt inutilisables pour les oiseaux. J’ai examiné la réponse d'une espèce résidente de la forêt boréale, la Mésange a tête brune (Poecile hudsonica), aux lisières forestières de la forêt Montmorency, Québec, Canada. Mon objectif était de déterminer si cette espèce se retrouve surtout en forêt profonde, et si ses mouvements sont influencés par les conditions météorologiques. Mes analyses ont porté sur un total de 74 km de mouvements de 85 volées de mésanges, pendant les hivers 2003-2004 et 2004-2005. J’ai suivi les mésanges en raquettes et j’ai enregistré leurs mouvements en temps réel avec un récepteur GPS. Lorsqu’elles étaient dans un peuplement mature (hauteur >7 m), les mésanges a tête brune ne montraient aucune association avec des lisières. Cependant, lorsqu’elles étaient dans un peuplement en régénération (hauteur 4-7m), les mésanges se retrouvaient plus souvent en lisière de ceux-ci. En conditions plus froides que la moyenne, les mésanges des peuplements en régénération se retrouvaient relativement près des peuplements matures. Une augmentation de la densité des lisières, due à la coupe totale dans les forêts boréales, ne réduit pas nécessairement la qualité des parties de forêt restantes pour la saison hivernale des mésanges à tête brune et ce, même pendant des températures inclémentes. 28

ABSTRACT

Avian responses to forest edges have received a great deal of attention in recent years, particularly due to the potential impact of deforestation on the quality of remaining forest patches. However, individual bird responses to forest edges are more often inferred than observed, since most studies emphasize territory placement, with no detail on actual movements of individuals. Thus, our understanding of the effects of edges on movements of forest birds remains limited. We recorded movements of 85 winter flocks of the little- known Boreal Chickadee (Poecile hudsonica), in a 66 km2 boreal forest harvested for timber near Quebec City, Quebec, Canada. From January-March (2004 and 2005), we followed flocks on snowshoes and their trajectory was recorded in real time using a handheld GPS receiver. Chickadee flocks showed no response to forest edges when using mature forest stands. However, flocks mostly used edges of regenerating forest habitat (4-7 m high). Inside regenerating forest, flocks were significantly closer to both open edges (41 ± 6 m) and mature forest boundaries (11 ± 2 m) than would be expected from random use of the habitat. Boreal Chickadee flocks did not avoid exposed edges during harsh weather conditions. In fact, on colder days, they were found disproportionately more often along edges between mature and regenerating stands. We conclude that increasing edge densities, resulting from clearcutting in boreal forest, does not reduce the winter suitability of remaining forest patches, even under inclement weather.

INTRODUCTION

Forest management affects the spatial structure of remaining habitat patches, often increasing the ratio between edge and interior forest (Forman 1995). The boundaries between landscape components (hereafter referred to as edges) can affect neighboring forest vegetation (Chen et al. 1992, Fraver 1994), arthropod abundance (Jokimäki et al. 1998, Haddad and Baum 1999), and local microclimate (Blake and Karr 1987, Chen et al. 1993). These “edge effects” can in turn affect animals. Edge effects on avian species diversity (Yahner 1988), territory location (Yahner 1988), nest predation (Lahti 2001) and brood parasitism (Paton 1994) have received a great deal of attention. Studies in recent 29 years have demonstrated further effects of edges on avian movements and foraging patterns within breeding territories (Huhta et al. 1999) and wintering home ranges (Dolby and Grubb 1999, Desrochers and Fortin 2000). Proper understanding of edge effects is particularly important in determining the suitability of forest fragments for resident birds during the non-breeding season. Population dynamics of resident species inhabiting northern latitudes appear to be strongly determined by events occurring during the non-breeding season (Matthysen 1990, Lahti et al. 1998, Doherty and Grubb 2002). Abiotic conditions including wind and temperature can differ near forest edges compared to interior forest (Murcia 1995). Wind speed can vary greatly with distance to edge and solar radiation may be significantly enhanced near sun-facing edges (Murcia 1995). Furthermore, wind and temperature can interact to affect the metabolic rate and energy expenditure of birds (Porter and Gates 1969, Wolf and Walsberg 1996) and in turn, reduce their survival rate (Mayer et al. 1979). Thus, selection of winter foraging sites may play an important part in reducing the energy costs of a small bird in harsh winter environments (Wachob 1996). Changes of microhabitat selection in response to adverse thermal conditions have been well documented in small resident birds. Winter residents have been found to move horizontally less often (Kessel 1976, Grubb 1978) and decrease foraging heights (Grubb 1975, 1977) during cold, windy, conditions. Grubb (1977) found that with increasing wind strength and decreasing temperature, birds shifted to leeward positions on foraging substrates. In highly fragmented deciduous forests, birds did not use exposed forest edges during harsh winter weather conditions, and were found further from windward edges of woodlots at high winds and low temperatures (Dolby and Grubb 1999). Increasing edge density in regions subjected to harsh winters can increase energy expenditures of birds beyond tolerable limits, resulting in a reduction of effective habitat available for use (Blake 1987, Dolby and Grubb 1999). However, most of the studies of winter weather effects were conducted in deciduous forests and agricultural landscapes, hence the abiotic edge effects remain poorly understood in boreal, conifer- dominated, forests. Boreal forests in Quebec, Canada, are being subjected to intensive forest exploitation and ~300 000 ha are logged annually (Ministère des Ressources naturelles et de la Faune 2006). The extensive logging has resulted in a reduction in large tracts of 30 forest, and an increase in fragmentation and edge density (Ministère des Ressources naturelles et de la Faune 2006). Forest remnants are typically linear and consist largely of riparian buffer strips and buffers between adjacent clearcuts (Ministère des ressources naturelles du Québec 1996). The high proportion of edges in the remaining forest patches could potentially render them functionally unsuitable for birds during the demanding winter months if abiotic edge effects in boreal regions act similarly to those in deciduous forests. Thus, understanding abiotic edge effects in northern boreal forests is important for the conservation of resident species and the maintenance of suitable winter habitats. The Boreal Chickadee is a resident species with its range almost completely restricted to northern boreal forests (Ficken et al. 1996). Boreal Chickadee numbers on Breeding Bird Survey routes have declined alarmingly (annual change (%) = -3.59, p-value = 0.0035) in Eastern North America (Sauer et al. 2005). Despite this decline, Boreal Chickadees have been little studied, particularly in winter (Ficken et al. 1996). Concern over habitat loss has been expressed (Erskine 1992, Foss 1994, Cyr and Larivée 1995) predominantly with respect to winter suitability. Erskine (1977, 1992) suggested that winter survival probably sets population limits for this species. We examined movements of Boreal Chickadee flocks in forest patches left by timber harvest, to determine whether they occur predominantly in the forest interior. We also examined the effects of winter weather conditions on edge associations within their home ranges. We predicted that flocks would avoid exposed (windward or northwest facing) edges in windy conditions, particularly at colder temperatures.

METHODS

Study area o o We conducted fieldwork at Forêt Montmorency, Quebec (47 20’-71 10’W) during the winters of 2004 and 2005. The study area is a 66 km2 managed boreal forest mosaic comprised of mature conifer stands (56%), mixed regeneration (24%) and open areas (20%). Balsam fir (Abies balsamea) and sometimes black spruce (Picea mariana) dominate the mature stands and are interspersed with white birch (Betula papyrifera). Early 31 serial stages tend to be dominated by coniferous regeneration or white birch/balsam fir stands. An extensive road network (2.6 km/ha) crosses the study area. Mean annual temperature at Forêt Montmorency is 0.3oC and mean annual precipitation is 158cm, 39% of which falls as snow (Environment Canada 2005). For analysis of response to edges, were characterized forest seral stages using existing GIS coverage for the study area (validated in situ, delimited by GPS, mapped with ArcView 3.3 [ESRI 2002]). We combined landscape elements within the study area into three classes: (1) All stands > 7 m in height were considered to be mature forest. These stands were almost exclusively coniferous, occasionally interspersed with birch. (2) Forested landscape elements between 4 m and 7 m in height were considered to be regeneration. This class included a few young spruce plantations, but consisted largely of naturally regenerating balsam fir, black spruce or mixed stands. (3) Open areas included forest stands < 4 m in height, clearcuts, gravel pits, lakes, rivers and roads > 7 m in width. Mature/regeneration was the most prevalent edge type found in the study area and had density of 55 m/ha. Mature/open edges were present at 44 m/ha and regeneration/open edges at 13 m/ha.

Flock following technique

We followed 72 Boreal Chickadee flocks in the first winter (6 January 2004 – 10 March 2004) and 13 flocks during the following winter (14 February 2005- 24 March 2005). Mean flock size was 4 ± 0.2 (mean ± SE) during both winter seasons and ranged from three to eight birds. We located chickadee flocks each day using a regular grid of points spaced by 1 km covering the study area. Grid points were randomly selected and visited only once during the study. Beginning from the selected grid point, we used a standard search pattern to minimize habitat bias in flock detection. We snowshoed 500 m north, 500 m east, 1,000 m south, 500 m west and 500 m north to return to the point of origin (sketched in Chapter 1, Figure 1). Detection of flocks was passive using only sight and sound. We did not revisit the widely spaced sample grid points in order to limit the chances of re-sampling the same flock. Despite being unmarked, flocks were considered to 32 be separate, independent units since flock composition differed in all but 3 cases (known to be different due to simultaneous observations) of adjacent flocks. Based on color-banding, we found that flock composition was stable throughout winter months (Chapter 1), as is the case in most other parids (Ekman 1989). We chose direct monitoring of the flocks over radio telemetry because the rugged topography limits the number of attainable fixes and reduces the resolution and accuracy of point locations (Ibarzabal and Desrochers 2004). Following the birds directly allowed a greater quantity of data to be collected for more flocks than would have been possible with telemetry methods. Once detected, flocks were followed on snowshoes and their locations were logged in real time at one-minute intervals using a hand held TrimbleTM GPS receiver. No positions were recorded during the first 2.5 minutes following discovery. GPS locations were recorded using a maximum point dilution of precision (PDOP) of 8 and were differentially corrected with data collected at a base station located near the study area. We followed the first flock detected along the search transect for as long as possible up to a limit of three hours. Flock cohesion varied, with all members of a flock foraging in the same tree in some instances while at other times individuals were spread as far as 20 m apart. To account for this, positions of the flock were recorded only when the observer was located at the approximate center of the flock and data logging ceased immediately if the observer’s position no longer reflected that of the flock. If contact with the flock was lost, the following period for that flock was terminated and we moved to a remote location before recommencing the search for a different flock. The periods for which flocks were followed ranged from 4.7 – 152 min (mean = 53 min). Path lengths during the observation periods for flocks ranged from 90 m to 2125 m (mean = 871 m). The total distance traveled for the 85 flocks was 74 km and the total following time was 75 hours. We found no discovery bias since there was no relationship between distance to forest edge and elapsed time since initial discovery (R2= 0.0003, F = 1.9, P = 0.28).

Response to edges

We examined response to edges at two spatial scales: (a) < 200 m (home range), and (b) < 30 m (micro-habitat). For each flock the mean distance to edge was calculated for four different situations: (1) points in mature forest to the nearest regeneration edge, (2) points in mature forest to the nearest open edge, (3) points in regeneration to the nearest mature edge, (4) points in regeneration to the nearest open edge. Only flock locations separated by at least one-minute intervals were used in this analysis. In order to assess whether boreal chickadee flocks moved throughout these habitats at random distances to edges, we created a quantitative null hypothesis specific to each flock. First, we delineated a 200 m (approximates home range) and 30 m (approximates available microhabitat) buffer for each flock surrounding all locations (Jones 2001). Secondly, we generated a 10 m spaced grid of points within each buffer. The resulting grid points were assumed to represent unbiased samples of the surrounding habitat due to the apparent lack of periodicity in spatial vegetation patterns of the study area. Thirdly, using the four habitat/edge situations mentioned above, we determined the mean distance to edge for each buffer. If boreal chickadee flocks respond to edges while moving through their winter habitat, we would expect the mean distance to edge calculated from observed locations to differ from the mean distance calculated using the “available” grid points within the corresponding buffer. We attempted to follow boreal chickadee flocks under all weather conditions encountered during the study period. Weather conditions were recorded hourly at an Environment Canada weather station located on the study site (Environment Canada 2005). Conditions during which flocks were followed covered a broad spectrum (Figure 1). The mean temperature during which flocks were followed was –9.6oC (range: –30oC to +3oC). Seventy-nine of the eighty-five flocks were followed on days with wind speed greater than 5 km/h (mean 10.8 km/h, range 5.8 km/h to 30 km/h).

Figure 1. Distribution of weather conditions during which flocks were followed; (a) temperature and (b) wind speed.

Overall differences between observed and expected distances to edge were analyzed using paired-sample t-tests. We used weighted least squares for statistical tests, with the inverse of the error variance of distance estimates for a weight factor, in order to account for the variation in sampling effort among flocks. The effects of wind, temperature and the interaction wind*temperature on distance to edge were examined using linear models, with Type III sums of squares (SAS Institute Inc. 1993). Effects were considered significant at α = 0.05.

RESULTS

From mature forest

Eighty-two of the 85 Boreal Chickadee flocks were observed in mature forest and flocks spent more than 80 percent of their time within this habitat. Chickadees used mature forest independently of distance to edge. Flocks were neither closer to, nor further from, regeneration or open edges than would be expected from random use of mature forest 35

(Table 1; Fig. 2a and 2b). This result was equivalent for both the home range (200 m) and microhabitat scales (30 m) (Table 1).

Table 1. Difference between expected and observed distances to habitat edges for Boreal Chickadee flocks. Positive values denote closer distances to edge than expected. (Paired t- test)

Available buffer (m) From habitat (N) Edge type Distance (m) SE t P

200 Mature (82) Regeneration -2.0 3.3 0.59 0.6 Open 0.6 2.7 -0.20 0.8

Regeneration (43) Mature 11.1 1.8 -6.92 <0.001 Open 40.9 6.4 -6.30 <0.001

30 Mature (82) Regeneration -0.7 0.96 0.73 0.5 Open -1.0 0.79 1.22 0.2

Regeneration (43) Mature 3.6 0.81 -4.07 <0.001 Open 11.7 2.03 -5.67 <0.001

From regenerating forest

Only 43 of the 85 flocks used regenerating forest. Chickadees avoided going deep into regenerating stands. When in regenerating forest chickadee flocks located themselves relatively close to edges bordering mature forest (Table 1). Only 9 of those 43 flocks were located further from mature forest edges than would be expected (Fig. 2c). The tendency of flocks to locate themselves close to adjoining mature forest when in regenerating forest was the same for both the 200 m and 30 m scales (Table 1).

Figure 2. Flock locations with respect to forest edges. Each point represents the mean distance available minus observed for individual flocks. Points above the dashed reference lines indicate flocks with a tendency to forage closer to the edge than expected from forest types within 200m. Note that scales are different among panels.

Flocks in regeneration were found considerably closer to open edges than expected (Table 1). Only 13 of 43 flocks using regeneration were further from open edges than expected within 200 m (Fig. 2d). In fact, chickadee flocks in regenerating forest were located much closer to open edges (41 ± 6 m [mean ± SE]) than they were to mature forest edges. As with distance to mature forest edges, the association of flocks to open edges was consistent across both scales examined (Table 1).

Effects of weather

When flocks were located in mature forest we found no significant effect of temperature, wind or their interaction on chickadee location in relation to edges with either regeneration forest or open areas. However, weather did have a statistically significant effect on the locations of boreal chickadees in regenerating forest (Table 2). With decreasing temperature chickadee flocks in regeneration tended to be more strongly associated with mature forest edges than during warmer weather (Fig. 3a). At high wind speeds, flocks in regenerating stands were less associated with mature forest edges (Fig. 3b). The interaction wind*temperature was also significant with flocks showing the strongest association with mature forest edges during cold, calm, conditions (Fig. 3c). In the interest of space, results were presented only for the 200 m (home range) scale; however, the trends were equivalent at the 30 m (microhabitat) scale. Weather conditions had no significant effect on the proportion of time spent in different habitat types (Temperature: R2= 0.012, F =1.03, P = 0.31; Wind: R2= 0.021, F =1.7, P = 0.19).

Table 2. Relationship between mean association with edge and weather conditions while flocks were using regenerating forest.

Edge type Weather variable Association SE P Regeneration/mature Temperature 2.96 0.45 <0.001 Wind -1.35 0.41 0.001 Temperature*Wind -0.19 0.04 <0.001 Regeneration/open Temperature -1.68 2.15 0.4 Wind -3.13 1.25 0.02 Temperature*Wind 0.06 0.13 0.6

Figure 3. Response to mature forest edge under different weather conditions when flocks were using regeneration forest. Response to edge is the difference between the mean distance available and the mean distance observed for individual flocks.

Edge orientation

We examined edge orientation to determine if the effects of weather on the association with edges in regenerating forest were dependent on direction. We divided edges into windward/shaded and leeward/sunny edges based on prevailing northwest winds in the study area and sun position during following periods (Environment Canada 2005). Roads were excluded from for this analysis due to the narrowness of the canopy gaps they create. Windward/shaded edges between regeneration and open areas were strongly avoided (Table 3). Flocks showed strong association with both leeward/sunny and windward/shaded edges mature forest edges. However, flocks were more closely associated with windward/shaded edges (Table 3).

Table 3. The effect of orientation on the association with edges of regenerating forest. Positive values denote closer distances to edge than expected.

Edge type Orientation Distance to edge (m) SE P Regeneration/mature Leeward/Sunny 12.5 1.8 <0.001 Windward/Shaded 25.6 4.0 <0.001 Regeneration/open Leeward/Sunny -7.6 4.3 0.08 Windward/Shaded -12.1 4.7 0.01

DISCUSSION

Response to edges Boreal Chickadee flocks did not avoid forest edges during daily movements within their winter home ranges and were in fact strongly associated with certain edge types. Our results are in accordance with two other studies of parid movements that have shown edge association (Desrochers and Fortin 2000, Brotons and Herrando 2003). However, others 40 have found edge independence (Germaine et al. 1997, Brand and George 2001) or even edge avoidance (Dolby and Grubb 1999), so it may not be possible to generalize at the level of this family. Boreal Chickadee edge association was dependent on the habitat type being used by the flock and unaffected by the scales examined (< 30 m and < 200 m). Flocks used mature forest irrespective of distance to edges. However, Boreal Chickadee flocks avoided areas deep into regenerating forest patches.

It is unlikely that the Boreal Chickadee association with edges while using regenerating forest can be attributed to a single main reason. We believe that at least three main factors must be addressed as possible drivers of the association with mature forest. First, changes in food or vegetation near mature forest edges may result in a more favorable foraging microhabitat than areas deeper into regeneration. Changes in arthropod abundance may occur with increasing distance from patch edges (Jokimäki et al. 1998) and seed crops have been shown to be greater near edges (Brotons and Herrando 2003). Food hoarding has also been shown to differ depending on proximity to edges, with a tendency to hoard food away from more exposed areas (Brotons et al. 2001). However, evidence for changes in winter food abundance with distance from edges is lacking (Desrochers and Fortin 2000). If Boreal Chickadees store food found in regenerating forest closer to mature forest than where it was collected, then recovery of these cached items later in the winter could drive the apparent edge association. Secondly, predation risk on adult animals has been considered an important driving force behind edge effects (Rodríguez et al. 2001, Turcotte and Desrochers 2003). Small birds need to trade-off energy gains with predation risk (Lima and Dill 1990) and their movements are largely determined by perceived risk of predation (Lima 1998). Habitat mediated predation risk often operates through the amount of available cover provided by vegetation that can be used to shelter from predator attacks (Lima and Dill 1990). Regenerating forest in our study area is much more open and exposed than mature forest. Birds are easily visible in this habitat type since they often forage on the outer branches of the young trees where they are much more exposed to view than within the crown of mature trees. Foraging in these more exposed habitats may make flocks more vulnerable to predation. Parid species will usually exhibit escape tactics based on flying into woody cover (Ekman 1987, Lima 1993), thus to mitigate the risks posed by predation, chickadees 41 may avoid venturing too far from the shelter offered by neighboring mature forest. Our flocks also appeared to forage in more cohesive groups when using regeneration forest and other studies have shown that group cohesion increases under “risky” conditions (Rodríguez et al. 2001). Predators of adult Boreal Chickadees are little known, yet there is evidence for predation risk within the study area. Northern Shrike (Lanius excubitor), Northern Hawk Owl (Surnia ulula) and Northern Goshawk (Accipiter gentilis) are potential diurnal predators present during the winter months. All three of these species were observed on occasion during the study and northern shrikes are considered to be a major predator of chickadees during winter (reviewed by Smith 1991). One Boreal Chickadee flock was observed mobbing a Northern Hawk Owl for >5 minutes and in another instance, predation of a Black-capped Chickadee by a shrike was observed (A. Hadley pers. obs.). Despite the relatively low abundance of these predators during the study period, animals have been shown to overcompensate for the threat of predation and behave as if a predator was actually present when there is just a perception of vulnerability (Bouskila and Blumstein 1992). A third explanation for edge responses by chickadee flocks could be that sites close to mature forest edges offer a more favorable microclimate than exposed areas deeper into regeneration patches. Microclimate selection has been shown to be important for winter survival in resident species (Wachob 1996). We found some support for this microclimate hypothesis since Boreal Chickadee flocks avoided the exposed areas far from mature forest and tended to focus their activities closer to mature forest on cold days. However, the effects of weather on Boreal Chickadee response to edges were not pronounced and association with mature forest still occurred under comparatively favorable weather conditions. Unlike response to mature forest edges, Boreal Chickadee association with edges adjoining open areas while using regenerating forest is likely attributable to a single factor. Edges sometimes act as movement barriers channeling movements in birds (Desrochers and Fortin 2000), mammals (Bider 1968, Desrochers et al. 2003) and insects (Haddad 1999) resulting in a disproportional amount of time spent along edges. While flocks did cross open areas, (14 of 85 flocks crossed gaps > 15 m in width, mean width = 53 ± 7 m, max= 120 m) flocks often showed reluctance to cross even small (~10 m) gaps with usually 1 or 2 42 individuals crossing first followed by the remaining flock members a few minutes later. The flocks were able to skirt around the edges of open areas moving through regeneration forest in order to move from one mature forest patch to another.

Effects of weather on edge association

Boreal Chickadees’ movements, like those of most resident species investigated to date (Grubb 1975, Lens 1996, Wachob 1996, Dolby and Grubb 1999) were affected by winter weather conditions. However, the effects of adverse weather conditions in our study were not pronounced and depended on both habitat type and edge type. Winter weather conditions caused no change in the distance of Boreal Chickadees to habitat edges when they were using mature forest. In regenerating forest Boreal Chickadee flocks were affected by inclement weather and were more strongly associated with edges adjoining mature forest during cold conditions. The evident effect of weather in regeneration stands relative to older forest is not surprising since regeneration in the study area is generally more open and exposed with larger gaps between trees than mature stands. Contrary to our predictions, and unlike other studies (Grubb 1977, Dolby and Grubb 1999, but see Desrochers and Fortin 2000) we did not find a strong avoidance of edges adjoining open areas during harsh weather. However, we did find that when edges with large open areas were divided by orientation, windward/shaded edges were generally avoided. The lack of strong weather effects is perhaps not so surprising and can be explained by at least two main factors. First, most studies investigating the effects of weather on edge association were conducted in deciduous forests, which let the wind penetrate easily in winter due to their lack of leaves. In the boreal forests of our study area the dense coniferous boughs retained snow cover and would easily dampen the effects of wind within a very short distance from edges. Other studies have shown that the density of vegetation next to edges is an important factor in mitigating the effects of wind and allowing the birds to use regions close to windy edges (Grubb 1977). 43

Secondly, the weak effect of weather might reflect potential behavioral and physiological adaptations that enable Boreal Chickadees to compensate for the increased exposure next to edges. Most other studies (e.g. Dolby and Grubb 1999) have dealt with species near the northern limits of their ranges. Boreal Chickadees, however, can be found in regions much further north than represented by our study area and it can be expected that they have developed the ability to cope with extreme conditions. Bent (1946) found that Black-capped Chickadees died under extremely cold conditions (below –45oC) while Boreal Chickadees at the same site were apparently unaffected. While we found little in terms of edge response due to harsh weather, Boreal Chickadees appear to reduce their foraging heights under windy conditions, particularly when using regeneration forest (A. Hadley pers. obs.). Vertical shifts in foraging locations have been well documented in other species of parids (Grubb 1975, 1977; Dolby and Grubb 1999) and it is possible that vertical shifts in foraging locations by Boreal Chickadees are more important than horizontal shifts as a means to compensate for adverse weather. Grubb (1977) also found that using the leeward portion of foraging substrates can allow birds to compensate for increasingly strong winds. Resident bird species, such as the Boreal Chickadee, have been proposed to be the species most at risk of population declines due to habitat loss and fragmentation (Bender et al. 1998, Schmiegelow and Mönkkönen 2002). As a result, any information on the use of landscapes by resident boreal forest species is of practical interest. Species adversely affected by fragmentation or patch size are expected to occur predominantly in the interior of patches with an avoidance of edges (Bender et al. 1998). Our results provide evidence against a strong effect of fragmentation since Boreal Chickadees did not avoid edges within their home ranges. Therefore, any population declines in this species should not be automatically attributed to forest fragmentation or edge effects. 44

ACKNOLWLEDGEMENTS We thank the staff at Forêt Montmorency (Université Laval) for logistical support. M. Betts, Y. Turcotte, Y. Aubry, G. Rompré, and M. Hadley provided useful comments on the manuscript. This study was funded by a NSERC Discovery grant to A. Desrochers and a NSERC Postgraduate scholarship (PGS A) to A. Hadley. 45

LITTERATURE CITED

Bender, D. J., T. A. Contreras, and L. Fahrig. 1998. Habitat loss and population decline: a meta-analysis of the patch size effect. Ecology 79:517-533.

Bent, A. C. 1946. Life Histories of North American Crows, Jays and Titmice. Smithsonian Institution United States National Museum Bulletin 191:322-338.

Bider, J. R. 1968. Animal activity in uncontrolled terrestrial communities as determined by a sand transect technique. Ecological Monographs 38:269-308.

Blake, J. G. 1987. Species-area relationships of winter residents in isolated woodlots. Wilson Bulletin 99:243-252.

Blake, J. G., and J. R. Karr. 1987. Breeding birds of isolated woodlots: area and habitat relationships. Ecology 68:64-71.

Bouskila, A., and D. T. Blumstein. 1992. Rules of thumb for predation hazard assessment - predictions from a dynamic-model. American Naturalist 139:161-176.

Brand, L. A., and T. L. George. 2001. Response of passerine birds to forest edge in coast redwood forest fragments. Auk 118:678-686.

Brotons, L., A. Desrochers, and Y. Turcotte. 2001. Food hoarding behaviour of Black- capped Chickadees (Poecile atricapillus) in relation to forest edges. Oikos 95:511- 519.

Brotons, L., and S. Herrando. 2003. Effect of increased food abundance near forest edges on flocking patterns of Coal Tit (Parus ater) winter groups in mountain coniferous forests. Bird Study 50:106-111.

Chen, J., J. F. Franklin, and T. A. Spies. 1992. Vegetation responses to edge environments in old-growth Douglas-fir forests. Ecological Applications 2:387-396.

Chen, J., J. F. Franklin, and T. A. Spies. 1993. Contrasting microclimates among clearcut, edge, and interior old-growth Douglas-fir forest. Agricultural and Forest Meteorology 63:219-237.

Cyr, A., and J. Larivée. 1995. Atlas saisonnier des oiseaux du Québec. Presses de l'Université de Sherbrooke et Société de loisir ornithologique de l'Estrie, Sherbrooke, Québec, Canada.

Desrochers, A., and M.-J. Fortin. 2000. Understanding avian responses to forest boundaries: a case study with chickadee winter flocks. Oikos 91:376-384.

Desrochers, A., I. Hanski, and V. Selonen. 2003. Siberian flying squirrel responses to high- and low-contrast forest edges. Landscape Ecology 18:543-552.

Doherty, P. F., Jr., and T. C. Grubb, Jr. 2002. Survivorship of permanent-resident birds in a fragmented forested landscape. Ecology 83:844-857.

Dolby, A. S., and T. C. Grubb. Jr. 1999. Effects of winter weather on horizontal and vertical use of isolated forest fragments by bark-foraging birds. Condor 101:408- 412.

Ekman, J. 1987. Exposure and time use in Willow Tit flocks: the cost of subordination. Animal Behaviour 35:445-452.

Ekman, J. 1989. Ecology of non-breeding social systems of Parus. Wilson Bulletin 101:263-288.

Environment Canada. 2005. Canadian Climate Normals, 1971-2005 (Quebec). Publication of the Canadian Climate Program.

Erskine, A. J. 1977. Birds in boreal Canada: communications, densities, and adaptations. Canadian Wildlife Service report series no. 41. Canadian Wildlife Service, Ottawa, Ontario, Canada.

Erskine, A. J. 1992. Atlas of Breeding Birds of the Maritime Provinces. Nimbus Publications and Nova Scotia Museum, Halifax, NS.

ESRI. 2002. ArcView® 3.3 GIS. Environmental Systems Research Institute, Inc., Redlands, CA, U.S.A.

Ficken, M. S., M. A. McLaren, and J. P. Hailman. 1996. Boreal Chickadee. In The Birds of North America, no. 254 (A. Poole, and F. Gill, Eds.). Academy of Natural Sciences, Philadelphia, and American Ornithologists' Union, Washington, D.C.

Forman, R. T. T. 1995. Land mosaics. Cambridge University Press, Cambridge, U.K.

Foss, C. R., Ed. 1994. Atlas of Breeding Birds in New Hampshire. Audubon Society, New Hampshire, Dover.

Fraver, S. 1994. Vegetation responses along edge-to-interior gradients in the mixed hardwood forests of the Roanoke River basin, North Carolina. Conservation Biology 8:822-832.

Germaine, S. S., S. H. Vessey, and D. E. Capen. 1997. Effects of small forest openings on the breeding bird community in a Vermont hardwood forest. Condor 99:708-718.

Grubb, T. C., Jr. 1975. Weather-dependent foraging behavior of some birds wintering in a deciduous woodland. Condor 77:175-182. 47

Grubb, T. C., Jr. 1977. Weather-dependent foraging behavior of some birds wintering in a deciduous woodland - horizontal adjustments. Condor 79:271-274.

Grubb, T. C., Jr. 1978. Weather-dependent foraging rates of wintering woodland birds. Auk 95:370-376.

Haddad, N. M. 1999. Corridor use predicted from behaviors at habitat boundaries. American Naturalist 153:215-227.

Haddad, N. M., and K. A. Baum. 1999. An experimental test of corridor effects on butterfly densities. Ecological Applications 9:623-633.

Huhta, E., J. Jokimäki, and P. Rahko. 1999. Breeding success of Pied Flycatchers in artificial forest edges: the effect of a suboptimally shaped foraging area. Auk 116:528-535.

Ibarzabal, J., and A. Desrochers. 2004. A nest predator's view of a managed forest: Gray Jay (Perisoreus canadensis) movement patterns in response to forest edges. Auk 121:162-169.

Jokimäki, J., E. Huhta, J. Itämies, and P. Rahko. 1998. Distribution of arthropods in relation to forest patch size, edge, and stand characteristics. Canadian Journal of Forest Research 28:1068-1072.

Jones, J. 2001. Habitat selection studies in avian ecology: a critical review. Auk 118:557- 562.

Kessel, B. 1976. Winter activity patterns of Black-capped Chickadees in interior Alaska. Wilson Bulletin 88:36-61.

Lahti, D. C. 2001. The "edge effect on nest predation" hypothesis after twenty years. Biological Conservation 99:365-374.

Lahti, K., M. Orell, S. Rytkonen, and K. Koivula. 1998. Time and food dependence in Willow Tit winter survival. Ecology 79:2904-2916.

Lens, L. 1996. Wind stress affects foraging site competition between Crested Tits and Willow Tits. Journal of Avian Biology 27:41-46.

Lima, S. L. 1993. Ecological and evolutionary perspectives on escape from predatory attack - a Survey of North-American Birds. Wilson Bulletin 105:1-47.

Lima, S. L. 1998. Stress and decision making under the risk of predation: Recent developments from behavioral, reproductive, and ecological perspectives. Pages 215-290 in Stress and Behavior, vol. 27. Academic Press Inc., San Diego.

Lima, S. L., and L. M. Dill. 1990. Behavioral decisions made under the risk of predation: a review and prospectus. Canadian Journal of Zoology 68:619-640.

Matthysen, E. 1990. Nonbreeding social organization in Parus. Current Ornithology 7:209- 249.

Mayer, L., S. Lustick, and T. C. Grubb., Jr. 1979. Energetic control of behavior - foraging in Carolina Chickadees. Comparative Biochemistry and Physiology a-Physiology 63:577-579.

Ministère Des Ressources Naturelles Du QuÉbec. 1996. Règlement sur les normes d'intervention dans les forêts du domaine public. Pages 2750-2786 in Gazette Officielle du Québec, vol. 128.

Ministère des Ressources naturelles et de la Faune. 2006. Ressources et industries forestières : Portrait statistique Édition 2005-2006. Gouvernement du Québec, 1-79 Pp.

Murcia, C. 1995. Edge effects in fragmented forests: implications for conservation. Trends in Ecology & Evolution 10:58-62.

Paton, P. W. C. 1994. The effect of edge on avian nest success: how strong is the evidence? Conservation Biology 8:17-26.

Porter, W. P., and D. M. Gates. 1969. Termodynamic equilibria of animals with environment. Ecological Monographs 39:227-224.

Rodríguez, A., H. Andrén, and G. Jansson. 2001. Habitat-mediated predation risk and decision making of small birds at forest edges. Oikos 95:383-396.

SAS Institute Inc. 1993. SAS/STAT Software: The genmod procedure. SAS Technical Report P-243. SAS Institute Inc., Cary, N.C., U.S.A.

Sauer, J. R., J. E. Hines, and J. Fallon. 2005. The North American Breeding Bird Survey, Results and Analysis 1966 - 2004. USGS Patuxent Wildlife Research Center, Laurel, MD, U.S.A.

Schmiegelow, F. K. A., and M. Mönkkönen. 2002. Habitat loss and fragmentation in dynamic landscapes: avian perspectives from the boreal forest. Ecological Applications 12:375-389.

Smith, S. M. 1991. The Black-capped Chickadee: Behavioural ecology and natural history. Cornell University Press, Ithaca, N.Y., U.S.A.

Turcotte, Y., and A. Desrochers. 2003. Landscape-dependent response to predation risk by forest birds. Oikos 100:614-618.

Wachob, D. G. 1996. The effect of thermal microclimate on foraging site selection by wintering Mountain Chickadees. Condor 98:114-122.

Wolf, B. O., and G. E. Walsberg. 1996. Thermal effects of radiation and wind on a small bird and implications for microsite selection. Ecology 77:2228-2236.

Yahner, R. H. 1988. Changes in wildlife communities near edges. Conservation Biology 2:333-339.

General Conclusion

There is added value to studying little known species, such as the Boreal Chickadee, since it allows elements of both unapplied (curiosity-driven) and applied (policy-driven) research to be addressed. Boreal Chickadees are found in boreal forests across Canada, however, our knowledge of this species and its ecology relies heavily on a single study

(McLaren 1975) or is inferred from studies of similar parids (Ficken et al. 1996). For wide ranging, yet little studied species such as the Boreal Chickadee, it is important to conduct further descriptive or “unapplied” research in order to gain a more complete understanding of their natural history. Such research also facilitates interspecific comparative analyses, quantitative reviews and other syntheses, which are often focused on only a few common, well-studied, species. Additionally, my research provides applications to the policy world because it is directly relevant to the issue of managing through clearcut size and pattern.

Thus, choosing a little-known species for an applied study allows for a simultaneous gain in knowledge of the specie’s natural history.

In this study I was able to address several important gaps in our knowledge of

Boreal Chickadee natural history. Previously, knowledge of winter flock size and stability had been limited to a few observations (Ficken et al. 1996). I was also able provide much needed information pertaining to Boreal Chickadee winter home range size and habitat us.

In addition, my work added to our limited knowledge of the responses or northern resident forest birds to edges.

I found that Boreal Chickadees are not strictly interior forest specialists, and in certain habitats, flocks even appear to be strongly associated with edges. Stand 51 configuration of optimal mature forest habitat appeared to be of little importance. All portions of optimal mature stands appear to be equally usable irrespective of distance to edge or edge type. However, movements within regeneration stands were constrained to regions next to edges and flocks avoided venturing deep into sub-optimal patches.

Unlike resident birds inhabiting deciduous forests (Dolby and Grubb 1999), abiotic edge effects under adverse weather conditions do not appear to greatly restrict the use of boreal forest habitat near edges. When using optimal, mature stands, flock movements near stand edges were unaffected by weather conditions. However, weather conditions did affect edge association within sub-optimal, regenerating stands. Flocks were most strongly associated with edges adjoining mature stands under cold conditions.

Despite the fact that I did not observe edge avoidance by flocks during the study, the effects of increasing edge densities and changing boreal forest configuration on northern residents need to be investigated further. For species or conditions where edges are avoided, the amount of usable habitat can decrease substantially (Figure 1). The effects of edges and fragmentation are thought to be greater as the total amount of habitat within the landscape decreases (Andrén 1994). Yet, Forêt Montmorency remains largely forested, with 80% of its surface area covered by forest stands > 4 m in height. The width of open areas (across which winds blow) is also important in determining the speed at which wind strikes the opposing forest edge (Zeng et al. 2004). Clearcuts at Forêt Montmorency are usually small (Bélanger et al. 1991). Therefore, in heavily fragmented forest landscapes, with very large clearcuts, wind may penetrate further into adjacent forest stands. Wind speeds may also be greater in landscapes with flatter topography than found within Forêt

Montmorency. Hence, abiotic edge effects in boreal forests should continue to be investigated, particularly in landscapes with limited suitable habitat. 52

Figure 1. The amount of forest habitat available at Forêt Montmorency if (a) edges are not avoided, (b) regions within 30m of open edges are avoided or (c) regions within 50m of open edges are avoided. Forest stands are shown in green and open areas in white.

Further effects of landscape configuration, other than those associated with edge densities, should also be investigated. Some studies have linked forest fragmentation to increases in animal home range size (Gjerde and Wegge 1989, Storch 1993, Siffczyk et al.

2003). Home range extension is thought to be associated with a variety of costs to the individual. With an increase in home range size, locomotive and territorial defense costs are likely to increase (Matthysen 1990). Costs of accessing, learning and relocating the presence of food would potentially increase within the larger area involved (Matthysen

1990). Pilfering rates of food supplies or hoarded food caches could also increase (Siffczyk et al. 2003). Increase in home range size can potentially even lead to decreased survival

(Gjerde and Wegge 1989, Doherty and Grubb 2002). Furthermore, the requirement for 53 larger territories in fragmented landscapes may have critical impacts at the population level.

The number of birds that a region can support would be reduced leading to a subsequent population decrease (Siffczyk et al. 2003).

Management implications Species adversely affected by fragmentation or patch size are expected to occur predominantly in the interior of patches with an avoidance of edges (Bender et al. 1998).

Hence, my results provide evidence against a strong effect of fragmentation since Boreal

Chickadees did not avoid edges within their home ranges. I conclude that increasing edge densities, resulting from clearcutting in boreal forest, does not in itself reduce the winter suitability of remaining forest patches, even under inclement weather. Therefore, any population declines in this species should not be automatically attributed to forest fragmentation or edge effects per se. Rather, apparent population declines in Boreal

Chickadees are more likely attributable to habitat loss.

My study confirms concerns stated by Ficken et al. (1996) surrounding the effects of forest harvesting on the winter suitability of remaining boreal forest stands for Boreal

Chickadees. Mature coniferous stands of commercial value comprise optimal chickadee wintering habitat. Thus, as clearcutting and forest harvesting continue, the proportion of optimal mature habitat within boreal regions will undergo substantial declines. The extensive logging has already resulted in a reduction in the proportion of mature stands, with a subsequent increase in proportion of young forest or open areas (Imbeau et al. 1999;

Ministère des Ressources naturelles et de la Faune 2006); thereby increasing the frequency of sub-optimal stands and non-habitat within the landscape. Consequently, current forestry 54 practices will result in a substantial reduction of optimal Boreal Chickadee wintering habitat, at least over several decades.

Additionally, there is growing interest in the use of partial harvesting practices including pre-commercial thinning, commercial thinning and selective harvesting (Hunter

1990). Larger stands of forest are left following partial harvesting than clearcuting, however, the remaining stand structure is greatly altered. The effects of these harvesting techniques on boreal residents need to be investigated. Reducing stand density following partial harvesting practices may alter Boreal Chickadee use of both mature and regenerating stands. Thinned stands are much more open and wind penetrates more deeply (Gardiner et al. 1997). These more open forests may also increase vulnerability to predators (Rodríguez et al. 2001) or change food availability (van Wilgenburg et al. 2001). Therefore, the suitability of selectively harvested stands, for winter resident species, should to be examined. 55 Bibliography for general introduction and conclusion

Andrén, H. 1994. Effects of habitat fragmentation on birds and mammals in landscapes with different proportions of suitable habitat: a review. Oikos 71:355-366.

Askins, R. A., M. J. Philbrick, and D. S. Sugeno. 1987. Relationship between the regional abundance of forest and the composition of forest bird communities. Biological Conservation 39:129-152.

Bélanger, L., L. Bertrand, P. Bouliane, and L.-J. Lussier. 1991. Plan général d'aménagement de la Forêt Montmorency. Université Laval, Faculté de foresterie et de géomatique, Sainte-Foy, Québec, Canada.

Bender, D. J., T. A. Contreras, and L. Fahrig. 1998. Habitat loss and population decline: a meta-analysis of the patch size effect. Ecology 79:517-533.

Blake, J. G. 1987. Species-area relationships of winter residents in isolated woodlots. Wilson Bulletin 99:243-252.

Blake, J. G., and J. R. Karr. 1987. Breeding birds of isolated woodlots: area and habitat relationships. Ecology 68:64-71.

Canadian Boreal Initiative. 2003. The Boreal Forest at Risk: A Progress Report. Pages 1-18. Senate Committee on Agriculture and Forestry, Ottawa, ON.

Chen, J., J. F. Franklin, and T. A. Spies. 1992. Vegetation responses to edge environments in old-growth Douglas-fir forests. Ecological Applications 2:387- 396.

Chen, J., J. F. Franklin, and T. A. Spies. 1993. Contrasting microclimates among clearcut, edge, and interior old-growth Douglas-fir forest. Agricultural and Forest Meteorology 63:219-237.

Cyr, A., and J. Larivée. 1995. Atlas saisonnier des oiseaux du Québec. Presses de l'Université de Sherbrooke et Société de loisir ornithologique de l'Estrie, Sherbrooke, Québec, Canada.

Desrochers, A., and M.-J. Fortin. 2000. Understanding avian responses to forest boundaries: a case study with chickadee winter flocks. Oikos 91:376-384.

Doherty, P. F., Jr., and T. C. Grubb, Jr. 2002. Survivorship of permanent-resident birds in a fragmented forested landscape. Ecology 83:844-857.

Dolby, A. S., and T. C. Grubb. 1999. Effects of winter weather on horizontal and vertical use of isolated forest fragments by bark-foraging birds. Condor 101:408-412. 56

Drapeau, P., A. Leduc, J.-F. Giroux, J.-P. L. Savard, Y. Bergeron, and W. L. Vickery. 2000. Landscape-scale disturbances and changes in bird communities of boreal mixed-wood forests. Ecological Monographs 70:423-444.

Erskine, A. J. 1977. Birds in boreal Canada: communications, densities, and adaptations. Canadian Wildlife Service report series no. 41. Canadian Wildlife Service, Ottawa, Ontario, Canada.

Erskine, A. J. 1992. Atlas of breeding birds of the Maritime Provinces. Nimbus Publications and Nova Scotia Museum, Halifax, NS.

Forman, R. T. T. 1995. Land mosaics. Cambridge University Press, Cambridge, U.K.

Foss, C. R., Ed. 1994. Atlas of breeding birds in New Hampshire. Audubon Society, New Hampshire, Dover.

Fraver, S. 1994. Vegetation responses along edge-to-interior gradients in the mixed hardwood forests of the Roanoke River basin, North Carolina. Conservation Biology 8:822-832.

Gardiner, B. A., G. R. Stacey, R. E. Belcher, and C. J. Wood. 1997. Field and wind tunnel assessments of the implications of respacing and thinning for tree stability. Forestry 70:233-252.

Gjerde, I., and P. Wegge. 1989. Spacing pattern, habitat use and survival of capercaillie in a fragmented winter habitat. Ornis Scandinavica 20:219-225.

Grubb, T. C. 1975. Weather-dependent foraging behavior of some birds wintering in a deciduous woodland. Condor 77:175-182.

Grubb, T. C. 1977. Weather-dependent foraging behavior of some birds wintering in a deciduous woodland - Horizontal Adjustments. Condor 79:271-274.

Grubb, T. C. 1978. Weather-dependent foraging rates of wintering woodland birds. Auk 95:370-376.

Haddad, N. M., and K. A. Baum. 1999. An experimental test of corridor effects on butterfly densities. Ecological Applications 9:623-633.

Huhta, E., J. Jokimäki, and P. Rahko. 1999. Breeding success of Pied Flycatchers in artificial forest edges: the effect of a suboptimally shaped foraging area. Auk 116:528-535. 57

Hunter, M. L. J. 1990. Wildlife, forests, and forestry, principles of managing forests for biological diversity. Prentice Hall, Princeton, New Jersey.

Imbeau, L., M. Mönkkönen, and A. Desrochers. 2001. Long-term effects of forestry on birds of the Eastern Canadian boreal forests: a comparison with Fennoscandia. Conservation Biology 15:1151-1162.

Imbeau, L., J.-P. L. Savard, and R. Gagnon. 1999. Comparing bird assemblages in successional black spruce stands originating from fire and logging. Canadian Journal of Zoology 77:1850-1860.

Kessel, B. 1976. Winter activity patterns of Black-capped Chickadees in interior Alaska. Wilson Bulletin 88:36-61.

Lahti, D. C. 2001. The "edge effect on nest predation" hypothesis after twenty years. Biological Conservation 99:365-374.

Lahti, K., M. Orell, S. Rytkonen, and K. Koivula. 1998. Time and food dependence in Willow Tit winter survival. Ecology 79:2904-2916.

Matthysen, E. 1990. Nonbreeding social organization in Parus. Current Ornithology 7:209-249.

Mayer, L., S. Lustick, and T. C. Grubb. 1979. Energetic control of behavior - Foraging in Carolina Chickadees. Comparative Biochemistry and Physiology a-Physiology 63:577-579.

McLaren, M. A. 1975. Breeding biology of the Boreal Chickadee. Wilson Bulletin 87:344-354.

Ministère des ressources naturelles du Québec. 1996. Règlement sur les normes d'intervention dans les forêts du domaine public. Pages 2750-2786 in Gazette Officielle du Québec, vol. 128.

Ministère des Ressources naturelles et de la Faune. 2006. Ressources et industries forestières : Portrait statistique Édition 2005-2006. Gouvernement du Québec, 1-79 Pp.

Paton, P. W. C. 1994. The effect of edge on avian nest success: how strong is the evidence? Conservation Biology 8:17-26.

Porter, W. P., and D. M. Gates. 1969. Termodynamic equilibria of animals with environment. Ecological Monographs 39:227-224. 58

Robbins, C. S., D. K. Dawson, and B. A. Dowell. 1989. Habitat area requirements of breeding forest birds of the Middle Atlantic States. Wildlife Monographs 103:1- 34.

Robinson, S. K., F. R. Thompson, T. M. Donovan, D. R. Whitehead, and J. Faaborg. 1995. Regional forest fragmentation and the nesting success of migratory birds. Science 267:1987-1990.

Rodríguez, A., H. Andrén, and G. Jansson. 2001. Habitat-mediated predation risk and decision making of small birds at forest edges. Oikos 95:383-396.

Sauer, J. R., J. E. Hines, and J. Fallon. 2005. The North American Breeding Bird Survey, Results and Analysis 1966 - 2004. USGS Patuxent Wildlife Research Center, Laurel, MD, U.S.A.

Schmiegelow, F. K. A., and M. Mönkkönen. 2002. Habitat loss and fragmentation in dynamic landscapes: avian perspectives from the boreal forest. Ecological Applications 12:375-389.

Siffczyk, C., L. Brotons, K. Kangas, and M. Orell. 2003. Home range size of Willow Tits: a response to winter habitat loss. Oecologia 136:635-642.

Storch, I. 1993. Patterns and strategies of winter habitat selection in Alpine Capercaillie. Ecography 16:351-359.

Trzcinski, M. K., L. Fahrig, and G. Merriam. 1999. Independent effects of forest cover and fragmentation on the distribution of forest breeding birds. Ecological Applications 9:586-593. van Wilgenburg, S. L., D. F. Mazerolle, and K. A. Hobson. 2001. Patterns of arthropod abundance, vegetation, and microclimate at boreal forest edge and interior in two landscapes: Implications for forest birds. Écoscience 8:454-461.

Villard, M.-A. 1998. On forest-interior species, edge avoidance, area-sensitivity, and dogmas in avian conservation. Auk 115:801-805.

Wachob, D. G. 1996. The effect of thermal microclimate on foraging site selection by wintering Mountain Chickadees. Condor 98:114-122.

Wolf, B. O., and G. E. Walsberg. 1996. Thermal effects of radiation and wind on a small bird and implications for microsite selection. Ecology 77:2228-2236.

Yahner, R. H. 1988. Changes in wildlife communities near edges. Conservation Biology 2:333-339.

Zeng, H. C., H. Peltola, A. Talkkari, A. Venalainen, H. Strandman, S. Kellomaki, and K. Y. Wang. 2004. Influence of clear-cutting on the risk of wind damage at forest edges. Forest Ecology and Management 203:77-88.

59 Appendix A

The SAS programming used to filter inaccurate GPS positions. The point filter is based on the time interval and distance between consecutive positions and was used to eliminate possible “spikes” or imprecise positions not representative of the actual flock locations. Using this filter, consecutive positions that could not be reached at a snowshoeing speed of three meters per second were considered to be spikes and thus filtered from the data set.

*Spike Filter; *sas program that uses the northing and easting, and GPS time to calculate the speed between points and eliminate impossible spikes; proc import out=ah datafile="Z:\Adam Hadley\Table1.DBF" dbms=DBF replace; run; data ah2; retain ok lastokx lastoky lastokt; set ah (keep= gps_second northing easting point_id); prevOK = ok; prevx = lag(easting); prevy = lag(northing); prevt = lag(GPS_second); if prevok=0 then do; prevx = lastOKx; prevy = lastOKy; prevt = lastOKt; end;

* Determine if OK; dist = ((easting-prevx)**2+(northing-prevy)**2)**0.5; elapsed = (GPS_second-prevt); if elapsed = 0 then elapsed=0.001; if _N_ ne 1 then speed = dist/elapsed; else speed = 0; if not (speed = .) then if (speed lt 3) or (elapsed lt 0) or (elapsed gt 1000) or (dist lt 3) then ok = 1; else ok = 0; if prevOK = 1 and ok = 0 then do; lastOKx = prevx; lastOKy = prevy; lastOKt = prevt; end;

if dist lt 5 then distcat = round(dist,1); 60

else if dist lt 100 then distcat=round(dist,10); else distcat = 1000; if elapsed lt 5 then elapscat = round(elapsed,1); else if elapsed lt 100 then elapscat = round(elapsed,10); else elapscat = 1000; run; proc print data=ah2; where point_id lt 75; var speed dist easting prevx ok prevok lastOKx; run; symbol i = join v=dot; proc gplot data=ah2; plot northing*easting=ok; where point_id gt 35 and point_id lt 70; run; proc freq data=ah2; table distcat*ok elapscat*ok/nopercent norow nocol; run; data ok; set ah2 (keep=point_id ok); run; PROC EXPORT DATA= ok OUTFILE= "Z:\ok.dbf" DBMS=DBF REPLACE; run;