EFFECTS OF SINGLE-TREE SELECTION HARVESTING ON ROSE-BREASTED

GROSBEAKS (PHEUCTICUS LEUDOVICIANUS) IN A PREDOMINANTLY

FORESTED LANDSCAPE

by

Sonya Richmond

A thesis submitted in conformity with the requirements

for the degree of Doctor of Philosophy, Faculty of Forestry

University of Toronto

© Copyright by Sonya Richmond 2011 Effects of Single-tree Selection Harvesting on Rose-breasted Grosbeaks (Pheucticus leudovicianus) in a Predominantly Forested Landscape

Sonya Richmond

Doctor of Philosophy Faculty of Forestry University of Toronto

2011

ABSTRACT

Single-tree selection harvesting is frequently used in the tolerant hardwood forests of North America but relatively little is known about how this silvicultural system affects wildlife, including many avian species. I investigated Rose-breasted Grosbeak

(Pheucticus leudovicianus) habitat use, survival, and nestling provisioning behaviour in twelve hardwood stands in Algonquin Provincial Park which had been harvested by single-tree selection 0-5, 16-20, 21-25, and > 50 years previously. Density, pairing success, and the number of fledglings per successful nest were all significantly lower in the > 50 years post-harvest stands than in at least one other post-harvest treatment.

Density and pairing success were significantly higher in the 16-20 year post-harvest stands than in other treatments. Neither nest nor fledgling survival differed significantly among post-harvest treatments, but all stands were population sinks except those cut 16-

20 years previously. Nests that were initiated earlier in the season and built in areas with higher basal area were more likely to survive, whereas fledgling survival increased with days since fledging. Nest sites had higher cover from regenerative growth, saplings, and understory, and lower basal area than random locations. During their first week out of the nest, fledglings used locations with significantly higher cover from regenerative growth, saplings, small shrubs, and raspberry and elderberry bushes than were present at

ii random locations. Habitat characteristics at nest and fledgling locations were significantly different, and estimates of nest and fledgling survival were not correlated among harvested stands. Nests attended by after-second-year (ASY) males were initiated significantly earlier, and territory density and productivity were significantly higher for

ASY males than for second-year (SY) males. Nestling provisioning rates, male contribution to nestling provisioning and nest attendance, and mean nestling weights at the time of fledging were also significantly higher at nests attended by ASY males than at nests with SY males. This study found that single-tree selection did not have significant negative effects on Rose-breasted Grosbeaks breeding in a predominantly forested landscape, but like many other species of birds, experienced breeders were more successful than less experienced breeders were.

iii ACKNOWLEDGEMENTS

First and foremost I would like to thank my supervisor, Dr. Jay Malcolm for his helpful advice and feedback, and for patiently and enthusiastically sharing his statistical expertise during the writing phase of this project. Thank you to Dr. Erica Nol and Dr.

Dawn Burke for encouraging me to continue with my studies at the doctoral level and for giving me the opportunity to be part of this project. I am also grateful to Dr. Sandy

Smith for her comments and direction as a committee member, and to Ken Elliott for his role in designing and implement the forestry components of the project. The scientific insight and direction you have all given me has made this a much stronger project on all levels.

I am thankful to Karla Falk and Doug Tozer for their patience, enthusiasm, and skill in coordinating the fieldwork crews, and to Nick Bartok, Alex Hood, J.-P. LeBlanc,

Rhiannon Leshyk, Meghan Noad, Andrea Orr, Eleanor Proctor, Kris Van de Sompel, and especially Phil Wilson, Brad Woodworth, and Gillian Humphries for their help with nest searching and monitoring, banding and radio-tracking fledglings, sitting patiently amidst the blackflies to conduct hour-long nest watches, surveying vegetation, driving long hours, and generally helping out with all the other aspects of fieldwork. Without the dedication and hard work you all generously put in there would have been no data.

I would like to thank the Algonquin Forestry Authority, Bancroft-Minden Forest

Co., Canadian Forest Service, Canadian Wildlife Service, Enhanced Forest Productivity

Science Program, Mazinaw-Lanark Forest Inc., Natural Sciences and Engineering

Research Council of Canada, Ontario Ministry of Natural Resources, Ottawa Valley

Forest Inc., Tembec Inc., and Westwind Forest Stewardship for providing funding and

iv support for this project. I was supported by an NSERC scholarship and grants from the

University of Toronto and Trent University.

Last but by no means least I would like to thank my family for their support, encouragement, and unlimited interest in this project. Without your help this endeavour would definitely not have been possible for me. Thank you especially to Sean Morton for frequently driving “across the province” to visit me during the four month field seasons, for generously volunteering to help in every way possible, and for always being willing to listen. Most of all, thank you all for the endless encouragement and for reminding me why it was important to see this through.

v TABLE OF CONTENTS

ABSTRACT...... ii ACKNOWLEDGEMENTS...... iv TABLE OF CONTENTS ...... vi LIST OF TABLES ...... viii LIST OF FIGURES ...... x LIST OF APPENDICES ...... xi GENERAL INTRODUCTION ...... 1 Background...... 1 Study species ...... 5 Study rationale ...... 7

CHAPTER 1: DEMOGRAPHY OF ROSE-BREASTED GROSBEAKS IN STANDS HARVESTED BY SINGLE-TREE SELECTION...... 11

Abstract...... 11 Introduction ...... 13 Methods ...... 18 Study species ...... 18 Study area ...... 19 Habitat measurements ...... 22 Nest monitoring ...... 22 Density and pairing success ...... 23 Population productivity ...... 24 Statistical Analysis ...... 25 Results ...... 28 Habitat measurements ...... 28 Demographic variables ...... 29 Relationship between habitat and demographic variables ...... 31 Age of male parent ...... 34 Discussion ...... 37 Demographic measures ...... 37 Population productivity ...... 40 Population age structure ...... 42 Conclusions ...... 44

CHAPTER 2: SURVIVAL AND HABITAT USE BY ADULT AND FLEDGLING ROSE-BREASTED GROSBEAKS (PHEUCTICUS LEUDOVICIANUS) IN A FORESTED LANDSCAPE HARVESTED BY SINGLE-TREE SELECTION....45

Abstract ...... 45 Introduction ...... 47 Methods ...... 51 Study species ...... 51 Study area ...... 51

vi Nest searching and monitoring ...... 52 Radio attachment and tracking ...... 53 Habitat measurements ...... 54 Statistical analysis ...... 56 Results ...... 64 Do habitat features differ between used and un-used plots? ...... 64 Do habitat features differ between nest and fledgling plots? ...... 71 Nest success and fledgling survival ...... 73 Are estimates of nest and fledgling survival correlated among post-harvest treatments? ...... 74 How do habitat characteristics influence nest and fledgling survival? ...... 76 Discussion ...... 79 Conclusions ...... 86

CHAPTER 3: NESTLING PROVISIONING RATES AND FORAGING PATTERNS OF ROSE-BREASTED GROSBEAKS IN STANDS HARVESTED BY SINGLE TREE SELECTION...... 90

Abstract ...... 90 Introduction ...... 92 Methods ...... 96 Study species ...... 96 Study area ...... 96 Nest monitoring ...... 97 Nestling provisioning ...... 97 Prey load sizes ...... 99 Nest attendance ...... 99 Nestling weights ...... 99 Statistical analysis ...... 100 Results ...... 102 Nestling provisioning rates ...... 102 Male contribution to nestling provisioning ...... 103 Nest attendance ...... 104 Male contribution to nest attendance ...... 104 Prey load sizes...... 108 Nestling weights at fledging ...... 111 Discussion ...... 111 Single-tree selection harvesting ...... 111 Nestling development ...... 114 Age of male parent ...... 117 Conclusions ...... 120 General Conclusions ...... 121 Literature Cited ...... 127

vii LIST OF TABLES

Table 1.1. Locations and characteristics of the twelve tolerant hardwood stands in Algonquin Provincial Park, Ontario, Canada that were investigated in this study...... 21

Table 1.2. Means of habitat variables (standard error in brackets) and results from two- way ANOVAs and Tukey’s tests comparing sites from four stages in the cutting cycle: 0-5, 16-20, 21-25, and > 50 years post-harvest (n = 3 per stage) and three study years (2006-2008) ...... 30

Table 1.3. Means of demographic variables (standard error in brackets) and results from two-way ANOVAs and Tukey’s tests comparing sites from four stages in the cutting cycle: 0-5, 16-20, 21-25, and > 50 years post-harvest (n = 3 per stage) and three study years (2006-2008) ...... 32

Table 1.4. Population growth rates for Rose-breasted Grosbeaks breeding in 12 tolerant hardwood stands in Algonquin Provincial Park harvested by single-tree selection. Stands represent four stages in the cutting cycle: 0-5, 16-20, 21-25, and > 50 years post-harvest. Models show intrinsic (λ) and exponential (r) population growth rates under best case (high adult and juvenile survival) and worst case (low adult and juvenile survival) scenarios ...... 33

Table 1.5. Means of demographic variables (standard error in brackets) and results from two-way ANOVAs and Tukey’s tests comparing sites from four stages in the cutting cycle: 0-5, 16-20, 21-25, and > 50 years post-harvest (n = 3 per stage) and three study years (2006-2008). Results are shown for second-year (SY) and after- second-year (ASY) males...... 36

Table 2.1. Number of Rose-breasted Grosbeak nests located in four height categories (low, medium, high, and very high) in 12 stands harvested by single-tree selection 0-5, 16-20, 21-25, and > 50 years prior to sampling. Number of nests in each category is shown with row percentages in brackets...... 69

Table 2.2. Support for models predicting daily nest survival rates of Rose-breasted Grosbeaks in Algonquin Provincial Park (2006-2008) as a function of percent cover of regenerative growth, saplings, and understory, basal area (m2/ha), time- since harvest (0-5, 16-20, 21-25, or >50 years), nest concealment (%), nest height (m) and nest initiation date ...... 77

Table 2.3. Support for models predicting daily survival rates of Rose-breasted Grosbeak fledglings in Algonquin Provincial Park 1-21 days after fledging (2006-2008). Predictor variables include percent cover of regenerative growth, saplings, understory, canopy, basal area (m2/ha), time since harvest, concealment (%), height (m), fledging date, and days since fledging...... 78

viii Table 3.1. Mean nestling provisioning rates at Rose-breasted Grosbeak nests in 12 hardwood stands in Algonquin Provincial Park harvested by single-tree selection 0-5, 16-20, 21-25, and > 50 years prior to study (stand errors are in parentheses). Mean nestling provisioning rates are presented for nests attended by second-year (SY) and after-second-year (ASY) males, and for nests with Stage 1 (< 5 days old) and Stage 2 (> 5 days old) nestlings...... 105

Table 3.2. Mean proportion of feeding events contributed by male Rose-breasted Grosbeaks in 12 hardwood stands in Algonquin Provincial Park harvested by single-tree selection 0-5, 16-20, 21-25, and > 50 years prior to study (stand errors are in parentheses). Mean proportion of feeding events are presented for second- year (SY) and after-second-year (ASY) males, and for nests with Stage 1 (< 5 days old) and Stage 2 (> 5 days old) nestlings...... 106

Table 3.3. Mean nest attendance at Rose-breasted Grosbeak nests in 12 hardwood stands in Algonquin Provincial Park harvested by single-tree selection 0-5, 16-20, 21-25, and > 50 years prior to study (standard errors are in parentheses). Mean nest attendance is presented for nests attended by second-year (SY) and after-second- year (ASY) males, and for nests with Stage 1 (< 5 days old) and Stage 2 (> 5 days old) nestlings...... 107

Table 3.4. Mean nest attendance contributed by male Rose-breasted Grosbeaks in 12 hardwood stands in Algonquin Provincial Park harvested by single-tree selection 0-5, 16-20, 21-25, and > 50 years prior to study (standard errors are in parentheses). Mean nest attendance is presented for second-year (SY) and after- second-year (ASY) males, and for nests with Stage 1 (< 5 days old) and Stage 2 (> 5 days old) nestlings...... 109

Table 3.5. Observed frequencies for prey load sizes delivered to Rose-breasted Grosbeak nests in 12 hardwood stands in Algonquin Provincial Park harvested by single- tree selection (row percents are in parentheses). Mean prey load sizes are presented for harvest treatment (0-5, 16-20, 21-25, and > 50 years post-harvest), for nests with > 5 or < 5 days old nestlings, and for nests attended by after- second-year (ASY) and second-year (SY) male parents...... 110

ix LIST OF FIGURES

Figure 1.1. Map of sites used to study the effects of single-tree selection on the demography of Rose-breasted Grosbeaks in Ontario, Canada. The grey polygon indicates Algonquin Provincial Park and the numbers associated with each site indicated the number of years post-harvest...... 20

Figure 1.2. Relationship between seven demographic variables for Rose-breasted Grosbeaks and six habitat features in four post-harvest treatments in Algonquin Provincial Park. Demographic variables are mean clutch size, territory density, number of fledglings per successful nest, pairing success, productivity, proportion of territorial ASY males, and nest survival. Habitat features are percent cover by forbs, regenerative growth, saplings, understory, and canopy, and basal area...35

Figure 2.1. (A) Regeneration cover, (B) sapling cover, (C) understory cover, (D) canopy cover, and (E) basal area at nest and random plots in 12 hardwood stands in Algonquin Provincial Park harvested by single-tree selection 0-5, 16-20, 21-25, and > 50 years previously. Means and standard errors are shown...... 68

Figure 2.2. (A) Regenerative cover, (B) sapling cover, (C) basal area, (D) small shrub (< 2.5 cm dbh), (E) raspberry cover, and (F) elderberry cover at locations occupied by fledglings and random plots in 12 hardwood stands in Algonquin Provincial Park harvested by single-tree selection 0-5, 16-20, 21-25 and > 50 years previously. Means and standard errors are shown...... 70

Figure 2.3. (A) Regeneration cover, (B) understory cover, and (C) basal area at nest sites and locations occupied by fledglings 1-7, 8-14, and 15-21 days post-fledging in 12 hardwood stands harvested by single-tree selection 0-5, 16-20, 21-25, and > 50 years previously. Means and standard errors are shown...... 72

Figure 2.4. Mayfield estimates of period survival rates for Rose-breasted Grosbeak nests and fledglings at three weeks post-fledging in four stand age classes in Algonquin Provincial Park differing in time since single-tree selection logging (0- 5, 16-20, 21-25, and >50 years post-harvest) (2006-2008). Means and standard errors are shown...... 75

x LIST OF APPENDICES

Appendix 2.1. List of candidate models used to explain Rose-breasted Grosbeak nest and fledgling survival in twelve tolerant hardwood stands in Algonquin Provincial Park harvested by single-tree selection 0-5, 16-20, 21-25, and > 50 years previously...... 88

xi GENERAL INTRODUCTION

BACKGROUND

With 347.7 million hectares of forested land, Canada is home to 10% of the world’s forests and 30% of the world’s boreal forest (Natural Resources Canada 2010).

Forested land in this country encompasses twelve forest regions (Rowe 1972), with the boreal forest as the most extensive, followed by the Great Lakes – St. Lawrence. The

Great Lakes – St. Lawrence is a transitional forest zone located between the boreal forest to the north and the deciduous Carolinian forest to the south. It stretches from southeastern Manitoba to the Gaspé, with a gap along the northern shores of Lakes

Superior (Natural Resources Canada 2010), and together with the deciduous forest region it contains the tolerant hardwood forest. Hardwood forests are common in the Acadian region of northern New England and New York, the upper Midwest, and the Maritime provinces of Canada (Seymour et al. 2002). In Ontario, tolerant hardwood forests that consist primarily of sugar maple (Acer saccharum), American beech (Fagus grandifolia), and yellow birch (Betula alleghaniensis) cover 3.6 million hectares and provide approximately 3.2% of the total provincial timber harvest (OMNR 1998). Much of this type of forest is directly affected by anthropogenic activities, such as urbanization and logging.

Forest management in North America has recently shifted away from practices designed solely to maximize timber yield towards a more ecosystem-based approach

(Kohm and Franklin 1997). Within this framework the focus has been on developing and implementing economically, socially, and ecologically sustainable management practices.

Efforts to achieve ecological sustainability have mainly taken the form of management

1 2 practices designed to emulate natural disturbance patterns (Perera et al. 2000, Seymour et al. 2002). In the boreal forest, even-aged silvicultural systems are most often used to emulate the large-scale disturbances characteristic of this region such as fire and wind storms (Perera et al. 2000). In contrast, in tolerant hardwood forests of North America, uneven-aged systems, such as shelterwood or selection cutting are the recommended methods of harvesting (Arboglast 1957, Franklin 1989, OMNR 2000). Disturbances in this region are typically small in scale, removing individual or small groups of trees through insect outbreaks, ice or wind storms, or natural senescence.

The selection harvesting system involves removing individual trees or small groups of trees once every 8-25 years (Hunter 1990, OMNR 2000, Seymour et al. 2002).

Tree marking guidelines and targets for the residual basal area to be retained in maple dominated forests under the selection system are very similar for the Great Lakes jurisdictions (OMNR 1998), Michigan (Tubbs 1977), and Wisconsin (WDNR 2011), but they vary according to stand condition and type. In the maple and beech dominated hardwood forests of the Great Lakes region, no more than one third of the basal area for a stand is removed per entry, and a residual basal area of 16 m2/ha of > 24 cm dbh trees or

20 m2/ha of > 9 cm dbh trees is generally considered appropriate (OMNR 1998, Tubbs

1977, WDNR 2011). The removal of trees during harvesting creates many small gaps or perforations in the canopy which encourage natural regeneration by allowing more light to reach the understory layers (Robinson and Robinson 1999). In the first few years after harvesting stands typically have abundant regenerative growth, low basal area, and sparse sapling, understory, and canopy layers (Robinson and Robinson 1999, Flaspohler et al.

2002, Jobes et al. 2004). By 16-20 years after harvesting, a dense understory layer often 3 forms as shade-tolerant species such as sugar maple and American beech develop

(Finegan 1984, Jobes et al. 2004). As the understory layer grows into the canopy, managed stands develop higher basal area, abundant cover in the seedling, regenerative growth, sapling, and canopy layers, and higher spatial complexity typical of un-harvested hardwood stands (Robinson and Robinson 1999, Kenefic and Nyland 2000, Crow et al.

2002, Flaspohler et al 2002, Jobes et al. 2004, Angers et al. 2005, Vanderwel et al. 2006).

Single-tree selection harvesting is designed to mimic natural disturbance patterns in tolerant hardwood forests, and therefore should have only minimal impacts on wildlife, including migratory songbirds. However, the abundance of many forest interior birds decreases following selection harvesting (Germaine et al. 1997, Robinson and Robinson

1999, Simon et al. 2000, Bourque and Villard 2001, Gram et al. 2003, Jobes et al. 2004,

Doyon et al. 2005, Guénette and Villard 2005, Holmes and Pitt 2007, Cooper et al. 2009,

Haché and Villard 2010, Poulin et al. 2010). Some disturbance-dependent species become more prevalent in harvested areas (Lent and Capen 1995, Annand and Thompson

1997, Simon et al. 2000, Gram et al. 2003, Jobes et al. 2004, Holmes and Pitt 2007,

Robertson and Hutto 2007, Atwell et al. 2008), but habitats with higher densities of birds are not necessarily more productive than other areas, or even equally beneficial (Van

Horne 1983, Vickery et al. 1992). Abundance estimates alone do not provide enough information to assess the impacts of forest management practices on avian populations.

Long-term estimates of habitat quality and reproductive success, while more costly and time consuming to derive, are necessary.

Studies of avian demography in harvested hardwood forests are less prevalent than studies of community composition, but several studies have detected differences in 4 reproductive success (Yahner 2000, Barber et al. 2001, Duguay et al. 2001, Robinson and

Robinson 2001, Rodewald and Yahner 2001, Moorman et al. 2002, Smith et al. 2006,

Dellinger et al. 2007, Cooper et al. 2009, Poulin et al. 2010) and density (Rodewald and

Smith 1988, Kilgo et al. 1999, Kingsley and Nol 1999, Flaspohler et al. 2002, Hagan and

Meehan 2002, Weakland et al. 2002, Jobes et al. 2004) following harvesting.

Silvicultural practices can also affect nest predator communities (Medin and Booth 1989,

Pettersson et al. 1995, Harris and Reed 2002) and prey availability (Morse and Robinson

1999, Marzluff et al. 2000, Sallabanks et al. 2000).

Although we know a little about how silviculture influences nest survival, much less is currently known about habitat use and survival of fledglings in managed forests.

Habitat requirements of breeding adults may be different from those of family groups during the post-fledging period (Rappole and Ballard 1987, Anders et al. 1998, Vega

Rivera et al. 1998, Cohen and Lindell 2004, King et al. 2006). For instance, dense cover near the ground and in the sapling and understory layers may provide important cover for the fledglings of many species regardless of the nest site characteristics selected by adults. Since the offspring of many songbirds spend up to eight weeks on the breeding grounds prior to migration (Anders et al. 1998), and fledgling survival strongly influences population growth rates (Porneluzi and Faaborg 1999, Powell et al. 1999, Woodworth et al. 1999), assessing the impact of silvicultural practices on fledglings is important.

So far, to my knowledge only one study has investigated the survival of songbird fledglings in selection-harvested forests. This study, which was conducted in a fragmented agricultural landscape, did not detect significant differences in juvenile survival among harvested and un-harvested forest stands (Moore et al. 2010). No studies 5 have yet been done on fledgling survival in managed hardwood forests in predominantly forested landscapes, and since silviculture may affect avian populations differently in agricultural and forested landscapes (Sallabanks et al. 2000, Thompson et al. 2000), further research is necessary.

STUDY SPECIES

The Rose-breasted Grosbeak (Aves: Passeriformes: Cardinalidae: Pheucticus leudovicianus) is a Neotropical migratory songbird that breeds across much of eastern and central North America and over winters in central and south America (Wyatt and

Francis 2002). Although this species is still relatively abundant, it has been declining at an annual rate of -0.8% across North America since 1966 (Sauer et al. 2008). In Ontario,

Rose-breasted Grosbeaks have been declining at a rate of approximately -2.4% per year between 1966 and 2007 (Sauer et al. 2008), with an especially sharp decline of – 8.0% per year between 1985 and 2005 (Cadman et al. 2007). This species provides a good opportunity to study the effects of harvesting because, despite its declining numbers, it is still abundant enough in the landscape to allow detection of changes in relation to management practices.

Rose-breasted Grosbeaks nest in primary and secondary mixed and deciduous forests, as well as in thickets, scrublands, urban parks, gardens, and other areas which have been subject to anthropogenic disturbances (Wyatt and Francis 2002). The Rose- breasted Grosbeak is often found in regenerating forests following harvesting, and in some cases this species is more abundant in logged areas than in un-harvested ones

(Holmes et al. 2004, Jobes et al. 2004, Doyon et al. 2005, Guénette and Villard 2005). 6

Although we know that Rose-breasted Grosbeaks tend to nest in the sapling and understory layers, one study found that they use locations with more canopy cover but lower sapling cover than random locations (Smith et al. 2007), whereas others suggest they prefer areas with relatively open canopy and sub-canopy layers (Wyatt and Francis

2002). Apart from studies in fragmented agricultural landscapes (Smith et al. 2006, 2007,

Moore et al. 2010), little is known about Rose-breasted Grosbeak ecology, or how changes in forest structure resulting from silviculture influence the availability or selection of nest sites, the foraging behaviour, or the ability of this species to successfully raise its young once they leave the nest.

Rose-breasted Grosbeaks typically arrive on their breeding grounds in central

Ontario during the first week of May, and nest initiation commences two weeks later.

Both males and females participate in nest building, incubating, brooding, and nestling provisioning, with the males providing approximately one third of the incubation, half of the brooding, and 25-30% of the nestling provisioning (Wyatt and Francis 2002).

Although bi-parental care is common in birds, in many migratory songbird species the females typically do all or most of the incubating and brooding, as well as providing much of the nestling provisioning (Stodola et al. 2010). Male Rose-breasted Grosbeak’s relatively high contribution to parental care may make this species more resilient to habitat alterations than other songbirds, because the trade-off between nest defence and foraging may be less severe. For example, if females must spend more time away from the nest foraging in altered habitats, Rose-breasted Grosbeak males can incubate and defend the nest in their absence, whereas in other species the nest may be left unattended for longer periods and be more likely to be lost to predators. However, since this species 7 is declining, and various forms of silviculture are implemented throughout its entire breeding range in North America, it is important to assess the effectiveness of current forest management guidelines for maintaining good quality habitat for Rose-breasted

Grosbeaks.

STUDY RATIONALE

Changes in Rose-breasted Grosbeak abundance following harvesting have appeared in a number of silvicultural studies from across North America, with mixed results. Rose-breasted Grosbeak abundance increased following single-tree selection harvesting in the northern hardwood forests of Quebec (Doyon et al. 2005) and Ontario

(Jobes et al. 2004), partial harvesting in the deciduous forest fragments of southwestern

Ontario (Holmes et al. 2004), and silvicultural treatments of various intensities in New

Brunswick (Guénette and Villard 2005). However, decreases in abundance were observed in response to selection harvesting and high-grading in fragmented Carolinian forests (Smith et al. 2006, Thompson et al. 2009) and to partial harvesting in boreal forests (Vanderwel et al. 2007, 2009). Other studies detected no statistical difference in abundance following harvesting (Duguay et al. 2001, Holmes and Pitt 2007). Studies of nest survival and productivity, which might provide insight into the inconsistent nature of these responses and help identify silvicultural practices that maintain or improve habitat for Rose-breasted Grosbeaks are still relatively rare.

Most research on Rose-breasted Grosbeak demography has been conducted in the highly fragmented Carolinian forest of southwestern Ontario (Smith et al. 2006, 2007,

Moore et al. 2010). In the hardwood forest fragments of this agricultural landscape, 8 single-tree selection did not significantly affect nest survival (Smith et al. 2006, Moore et al. 2010), but density, nest success, and the ratio of after-second-year to second-year males were slightly lower in the harvested stands than in the un-logged ones (Smith et al.

2006). Fledgling survival in these stands was not significantly affected by harvesting, but it too was slightly lower in the harvested stands (Moore et al. 2010). In this landscape, both harvested and un-harvested stands were classified as population sinks (Smith et al.

2006, Moore et al. 2010). Although single-tree selection harvesting did not appear to significantly alter Rose-breasted Grosbeak survival in forest fragments, no information is available for predominantly forested landscapes. Some songbird populations are above replacement levels in predominantly forested landscapes and large forest fragments but below replacement levels in smaller fragments (Burke and Nol 2000, Falk et al. 2011), suggesting that it is important to investigate the impacts of anthropogenic activities on reproductive success in different landscape contexts. In particular, since a considerable portion of the Rose-breasted Grosbeak’s breeding range falls within managed forest landscapes, it is important to assess whether harvesting affects this species differently in predominantly forested landscapes than in fragmented agricultural ones.

To address this knowledge gap, in this thesis I investigate Rose-breasted

Grosbeak habitat use in tolerant hardwood stands harvested by single-tree selection 0-5,

16-20, 21-25, and > 50 years previously in the predominantly forested landscape of

Algonquin Provincial Park, Ontario, Canada. Algonquin Park was logged for white pine

(Pinus strobes) in the 1800’s and for hemlock (Tsuga canadensis) and yellow birch

(Betula alleghaniensis) during the 1930’s and 1940’s (Tozer and Strickland 1980). Since

1974 approximately 12% of the park has been designated as a Wilderness Zone where no 9 logging activity occurs. The remaining forested area of the park (57% of its area) is managed for timber production at a rate of approximately 2% per year (Ontario Parks

1998). Stands that consist of > 60% hardwood species have been harvested using the selection system since the 1970’s, and some have now been regenerating long enough to allow a second harvest entry.

I used a chronosequence or space-for-time substitution approach (reviewed in

Fukami and Wardle 2005) to study succession following single-tree selection harvesting in tolerant hardwood stands, and to look at its long-term consequences for Rose-breasted

Grosbeaks. I chose stands that were 0-5, 16-20, and 21-25 years post-harvest to investigate three stages in the harvest rotation. The 21-25 years post-harvest stands were chosen to provide insight into forest conditions at the stage when a second harvest is typically carried out, to see if forest structure and reproductive output of Rose-breasted

Grosbeaks had returned to “pre-harvest” levels. There is some concern that if regeneration times are not long enough, cumulative changes could occur. The > 50 years post-harvest stands were chosen to approximate “un-logged” or “pre-harvest” conditions which could be used as a reference point against which to evaluate more recently logged stands because completely unlogged, primary growth hardwood stands were unavailable in my study area within the park.

Chapter one examines whether Rose-breasted Grosbeak density, pairing success, population age structure, nest initiation date, clutch size, nest survival, and productivity differed with time since selection harvesting. Chapter two investigates important habitat characteristics at nest sites and locations used by fledglings, tests whether these characteristics differ among post-harvest treatments, and investigates which habitat 10 variables have the strongest influence on nest and fledgling survival. Chapter three tests whether nestling provisioning rates, prey load sizes, male contributions to nestling provisioning and nest attendance, and nestling weights differed with time since single- tree selection harvesting, stage of nestling development, or age of male parent. Finally, I end with a general discussion on the relevance of these findings to current forest management practices in the tolerant hardwood forests of North America.

I expected considerable changes in forest structure following single-tree selection harvesting, and that as a result of these alterations vegetation characteristics at nest sites and locations used by fledglings would significantly differ among post-harvest treatments. Stands harvested by single-tree selection 16-20 years previously typically have well developed sapling and understory layers, which appear to be preferentially selected by Rose-breasted Grosbeaks. As a result, I predicted that these stands would have higher densities, a greater proportion of after-second-year males, earlier nest initiation dates, higher pairing success, larger clutch sizes, higher nest survival rates, and greater productivity. I also expected that nestling provisioning would be more frequent and/or larger prey loads would be provided to nestlings in these stands and that nestlings would be heavier on average than in the other post-harvest treatments and the uncut stands. CHAPTER 1: DEMOGRAPHY OF ROSE-BREASTED GROSBEAKS IN STANDS HARVESTED BY SINGLE-TREE SELECTION

ABSTRACT

Single-tree selection harvesting is frequently used in the tolerant hardwood forests of North America, but little is known about how it affects the reproductive success of migratory songbirds. Many songbirds that breed in this forest type, including the Rose- breasted Grosbeak (Pheucticus leudovicianus) are currently experiencing population declines across their breeding ranges. I studied the demography of Rose-breasted

Grosbeaks in twelve tolerant hardwood stands in the predominantly forested landscape of

Algonquin Provincial Park. Nine stands had been harvested by single-tree selection 0-5,

16-20, or 21-25 years previously, and three had remained un-harvested for > 50 years.

My main objectives were to test whether density, pairing success, population age structure, nest initiation date, clutch size, nest survival and productivity differed with time since single-tree selection harvesting. My secondary objective was to determine whether any of these demographic parameters varied between after-second-year (ASY) and second-year (SY) breeders. Analysis of variance indicated that density, pairing success, and the number of fledglings per successful nest were all significantly (P < 0.05) lower in the > 50 years post-harvest stands than in at least one other post-harvest treatment, as were clutch size, nest survival, and productivity, although not significantly so. The proportion of ASY males compared to SY males was highest in the > 50 years post- harvest stands and lowest in the 21-25 years post-harvest treatment. Density, pairing success, and productivity were all highest in the 16-20 years post-harvest treatment, and density and pairing success were significantly so. Nests attended by ASY males were

11 12 initiated significantly earlier than nests attended by SY males, and territory density and productivity were significantly higher for ASY males than SY males. Habitat characteristics did not vary significantly between nests attended by ASY and SY males.

This study found that single-tree selection did not have significant negative effects on the reproductive success of Rose-breasted Grosbeaks in predominantly forested landscapes, but that all post-harvest treatments were population sinks except the 16-20 years post- harvest stands, indicating that further research into the causes of this species’ decline is necessary. 13

INTRODUCTION

Over half (57%) of the Neotropical migratory bird species that breed in eastern

North America have been experiencing broad-scale population declines over the past forty years (Kirby et al. 2008). Habitat loss, fragmentation, and degradation are among the most frequently cited causes of declines in migratory songbirds (Donovan and Flather

2002, Sauer et al. 2007, Poulin et al. 2010). Changes in land-use, together with forest management policies intended to suppress natural disturbances such as wildfire, flooding, and insect outbreaks have also reduced the amount of habitat available for some species

(Askins 1993, Brawn et al. 2001, Trani et al. 2001, Brooks 2003, DeGraaf and Yamasaki

2003). Concern over avian population declines has led to research investigating the potential effects of forest management practices on forest breeding birds (Duguay et al.

2001). In North America, uneven-aged harvesting practices, such as the shelterwood and selection systems, are becoming more prevalent in mixed and tolerant hardwood forests, because they are believed to mimic natural patterns of small-scale disturbances characteristic of these forest types (Tubbs 1977, OMNR 2000, Seymour et al. 2002). In

Ontario, Canada these practices account for 13% of annual provincial harvests, and they are mostly applied in the Great Lakes – St. Lawrence forest region (Cadman et al. 2007).

Single-tree and group selection harvesting produce uneven-aged stands through the removal of individual or small groups of trees every 15-25 years (Tubbs 1977, Hunter

1990, Seymour et al. 2002). Although single-tree selection emulates natural disturbance patterns in mature hardwood forests better than some large-scale, even-aged silvicultural techniques (Hunter 1990, Seymour et al. 2002), considerable changes in forest structure occur in the years following harvesting. For example, recently logged stands typically have abundant regenerative growth, low basal area, and sparse cover in the sapling, 14 understory, and canopy layers (Robinson and Robinson 1999, Flaspohler et al. 2002,

Jobes et al. 2004). However, increased light penetration immediately following harvesting promotes the development of shade-tolerant sugar maple and American beech seedlings, which form a dense understory layer by about 15 years after harvesting

(Finegan 1984, Jobes et al. 2004). As the understory layer grows into the canopy, managed stands eventually approach pre-harvest conditions, which are characterized by high basal area, abundant cover in the seedling, regenerative growth, sapling, and canopy layers, and high spatial complexity (Robinson and Robinson 1999, Kenefic and Nyland

2000, Crow et al. 2002, Flaspohler et al. 2002, Jobes et al. 2004, Angers et al. 2005,

Vanderwel et al. 2006).

Many studies have documented a decline in the abundance of mature-forest species following selection harvesting (Germaine et al. 1997, Robinson and Robinson

1999, Simon et al. 2000, Bourque and Villard 2001, Gram et al. 2003, Jobes et al. 2004,

Doyon et al. 2005, Guénette and Villard 2005, Holmes and Pitt 2007, Cooper et al. 2009,

Haché and Villard 2010, Poulin et al. 2010). In contrast, the abundance of early- successional species generally increases following selection harvesting (Lent and Capen

1995, Annand and Thompson 1997, Simon et al. 2000, Gram et al. 2003, Jobes et al.

2004, Holmes and Pitt 2007, Robertson and Hutto 2007, Atwell et al. 2008), but the gaps created by single-tree selection may be too small to benefit some species (Flaspohler et al. 2002, Annand and Thompson 1997, Robinson and Robinson 1999). Demographic responses to forest management are mixed, with many species achieving comparable rates of nest survival before and after harvesting (Bourque and Villard 2001, Gram et al.

2003, Smith et al. 2006, Cooper et al. 2009, Moore et al. 2010), but with other species 15 experiencing lower rates of nest success in harvested forests (Robertson and Hutto 2007,

Poulin et al. 2010). Differing responses to selection harvesting among avian species suggests that a more complete understanding of the link between forest management and avian population growth is important for identifying management practices that maintain good quality habitat for wildlife, and are therefore ecologically sustainable.

Since reproductive success can be reduced in areas with high densities of birds

(Arcese and Smith 1988, Vickery et al. 1992), and populations can persist in areas even when reproductive rates are low (Brawn and Robinson 1996), abundance estimates alone are generally not considered sufficient indicators of habitat quality. The ideal free distribution model assumes that individuals will occupy habitats in proportion to their quality, with better habitats supporting more individuals (Fretwell and Lucas 1970,

Fretwell 1972). However, recent (Schmiegelow et al. 1997) or unpredictable (Van Horne

1983) disturbances such as logging can influence avian abundances in forested habitats.

Unreliable indicators of habitat quality (Remes 2003, Weldon and Haddad 2005, Nordby et al. 2009), past reproductive success (Haas 1998, Citta and Lindberg 2007), perceived predation risk (Morton 2005, Fontaine and Martin 2006), and social cues (Doliqez et al.

2004, Fletcher 2007, Seppänen et al. 2007, Ahlering et al. 2010) can also influence how birds select territories. As a result, despite recent evidence that density may be an adequate indicator of avian productivity in some hardwood forest regions (Underwood and Roth 2002, Bock and Jones 2004, Perot and Villard 2009), additional measures of demography should be included in assessments of habitat quality in managed areas

(Sallabanks et al. 2000). 16

Age-related differences in acquisition and defence of high quality territories

(Catterall et al. 1989; Donovan and Stanley 1995, Pärt 2001) may also cause a decoupling of the predicted positive relationship between density and productivity, with older, more experienced birds out-competing younger ones and monopolizing resources (Donazar and

Feijoo 2002). In some species, nest initiation date, clutch size (Reynolds 1997), nest survival (Forschler and Kalko 2006), and number of fledglings per successful nest

(Préault et al. 2005, Brown and Roth 2009) are influenced by the age of the breeding adults. Age-related differences in reproductive success may occur because: 1) inferior phenotypes, which have lower success, experience high mortality at young ages, 2) younger individuals postpone breeding until they can accumulate sufficient resources, 3) reproductive output increases as reproductive value decreases with age, or 4) parental skills improve with age (Forslund and Pärt 1995, Martin 1995, Espie et al. 2000,

Balbontin et al. 2007). Regardless of which hypothesis is supported, if older birds are more successful at obtaining and defending territories in preferred habitats, then the ratio of older birds to younger birds could be used to assess which habitats are perceived as most desirable by breeding individuals. In addition, if older birds experience higher reproductive success than younger birds, areas with higher proportions of experienced individuals may be more productive overall. Clarifying how population age structure influences productivity is useful for better understanding the effects of anthropogenic activities.

The Rose-breasted Grosbeak (Pheucticus leudovicianus) is a Neotropical migrant that breeds in both harvested and un-harvested hardwood forests across North America

(Wyatt and Francis 2002, Smith et al. 2006). This early-successional forest species 17

(Drapeau et al. 2000) uses interior and edge habitats (Freemark and Merriam 1986,

Merrill et al. 1998) as well as managed forests with high tree density, extensive sapling and shrub cover, and well-developed understory layers (Webb et al. 1977, Possardt and

Dodge 1978). Although Rose-breasted Grosbeaks utilize features of regenerating forests, responses to harvesting have been mixed. Estimates of abundance are typically higher in recently harvested stands than in less recently cut (Holmes et al. 2004, Jobes et al. 2004,

Doyon et al. 2005) or unlogged stands (Jobes et al. 2004, Doyon et al. 2005), and Rose- breasted Grosbeaks have responded positively to silvicultural treatments of various intensities in New Brunswick (Guénette and Villard 2005). However, decreases in abundance were observed in response to selection harvesting and high-grading in fragmented Carolinian forests (Smith et al. 2006, Thompson et al. 2009) and to partial harvesting in boreal forests (Vanderwel et al. 2007, 2009), whereas other studies detected no statistical difference in abundance following harvesting (Duguay et al. 2001, Holmes and Pitt 2007). Density of territorial males and the ratio of after-second-year to second- year males were also lower in harvested stands than in un-harvested ones in the fragmented landscape (Smith et al. 2006). No information on most demographic parameters is available for Rose-breasted Grosbeaks nesting in predominantly forested landscapes, even though approximately 40% of its range falls within relatively un- fragmented forests. A better understanding of how silviculture influences Rose-breasted

Grosbeaks is necessary because this species is currently experiencing population declines throughout its breeding range across central and eastern North America (Smith et al.

2006, Cadman et al. 2007, Sauer et al. 2008). 18

Here, I examined whether changes in habitat structure resulting from single-tree selection harvesting influenced the density, population age structure, pairing success, nest initiation dates, and productivity of Rose-breasted Grosbeaks. I predicted that density and productivity would be higher in stands harvested 16-20 years previously than in more recently harvested stands or un-logged ones due to the presence of relatively well- developed sapling and understory layers that appear to be preferentially selected by grosbeaks (Wyatt and Francis 2002). I also expected that older, after-second-year males would out-compete second-year males in these stands and that the older males would experience higher reproductive success. As a result, I predicted that 16-20 years post- harvest stands would have higher densities and greater proportions of after-second-year males, and that the older, more experienced breeders would have earlier nest initiation dates, higher pairing success, larger clutch sizes, nest survival rates, and greater productivity than younger, less experienced males.

METHODS

Study Species

The Rose-breasted Grosbeak is a relatively abundant Neotropical migrant that over winters in central and South America and breeds in first- and second-growth mixed and deciduous forests throughout much of central and eastern North America (Wyatt and

Francis 2002). Rose-breasted Grosbeaks typically arrive on their breeding grounds in central Ontario during the first week of May, where they begin establishing territories ranging from 0.34 – 1.3 ha (mean = 0.77 ha) (Wyatt and Francis 2004). Nesting attempts typically commence during the last two weeks of May (Smith et al. 2006). Incubation 19 takes 12-13 days, and fledging occurs 9-12 days after the eggs hatch (Wyatt and Francis

2002). Most pairs only raise one brood per season, but double-brooding may occasionally take place (Wyatt and Francis 2002).

Study Area

Nest monitoring and territory mapping of Rose-breasted Grosbeaks was carried out in twelve tolerant hardwood stands in Algonquin Provincial Park, Ontario, Canada

(45° 34`N, 78°04`W; Fig. 1.1) in 2006-2008. The higher-elevation west side of this

775,000 ha park consists primarily of mixed hardwood forest, whereas the lower- elevation east side is dominated by white and red pine (Ontario Parks 1998). The tolerant hardwood stands I studied were located in the central and southern part of the park (mean nearest neighbour distance of stands = 9.3 km, range = 1.0 – 34.2 km), which has been harvested using the single-tree selection system since the early 1970’s (OMNR 1998).

This type of forest management removes approximately one-third of the over-story basal area every 15-25 years, resulting in uneven-aged stands with a mix of canopy gaps, regenerative growth, and mature trees (OMNR 1998). Stands used in this study were dominated by sugar maple (Acer saccharum), American beech (Fagus grandifolia), eastern hemlock (Tsuga canadensis), and yellow birch (Betula alleghaniensis) and they represented four stages in the cutting cycle: 0-5, 16-20, 21-25, and > 50 years post- harvest. Three stands (mean stand size = 34 ha) from each post-harvest treatment were studied (Table 1.1). 20

Figure 1.1. Map of sites used to study the effects of single-tree selection on the demography of Rose-breasted Grosbeaks in Ontario, Canada. The grey polygon indicates Algonquin Provincial Park and the numbers associated with each site indicate number of years post-harvest. Solid triangles = 0-5 years post-harvest stands, solid squares = 16-20 years post-harvest stands, solid hexagons = 21-25 years post-harvest stands, and unfilled circles = > 50 years post-harvest stands. 21

TABLE 1.1. Locations and characteristics of the twelve tolerant hardwood stands in Algonquin Provincial Park, Ontario, Canada that were investigated in this study. Site County UTM Last Post-harvest Searched Basal area Percent Percent Percent Code co-ordinates harvest treatment (years area (m2/ha) sugar American other year since harvest) (ha) maple beech tree species (Mh)1 (Be)1 (Sp)1,2 CT Nipissing 716200 5068315 2004.5 0-5 32 22.1 80.0 20.0 0.0 LC Haliburton 714450 5034175 2004.5 0-5 41 19.2 41.7 20.8 37.5 YO Haliburton 713676 5015892 2000.5 0-5 38 23.9 39.1 17.4 43.5 LB Haiburton 714035 5018042 1989.5 16-20 25 26.7 50.0 30.0 20.0 OP Nipissing 712007 5074748 1987.5 16-20 36 26.3 46.1 38.5 15.4 TM Nipissing 701545 5053886 1986 16-20 34 28.3 40.9 22.7 36.4 CF Haliburton 713800 5032150 1980.5 21-25 25 26.9 54.5 9.1 36.4 LW Nipissing 713620 5072340 1984.5 21-25 26 29.0 30.8 15.4 53.8 ZB Nipissing 701560 5056325 1984.5 21-25 43 29.0 47.1 11.8 41.2 BR Nipissing 710659 5051174 1941.5 > 50 32 37.3 50.0 20.0 30.0 CJ Haliburton 713860 5018886 1948.5 >50 29 32.0 29.2 29.2 41.6 TR Nipissing 697000 5049750 1941.5 >50 43 30.5 37.5 25.0 37.5 1Percentage of each tree species in prism sweeps at randomly located plots within each stand 2Other tree species consisted primarily of yellow birch, black cherry, striped maple, eastern hemlock, and balsam fir 22

Habitat Measurements

I measured habitat characteristics in 5 m radius, circular plots in 170 randomly selected locations within my 12 post-harvest stands between 2006 and 2008. In each year, the number of random plots surveyed in each stand was equal to the number of nests monitored in that stand, providing an average of 13 plots/stand and 43 plots/treatment over the three years. Survey plots were selected using a set of randomly generated UTM co-ordinates in each site, and GPS units were used to locate the coordinates on the ground. No restrictions were placed on the locations of plots, as long as they fell within the stand boundaries. Vegetation layers were measured using a modified version of the

James and Shugart (1970) method. At each survey plot, percent cover by 1) forbs/grasses/ferns, 2) regenerative growth (woody growth 0.5-1.3 m tall), 3) saplings

(>1.3 m tall, <2.5 cm dbh), 4) understory (≥ 2.5 cm dbh, <10.0 cm dbh), 5) sub-canopy

(>10-20 m tall), and 6) canopy (>20 m tall) were visually determined, and basal area

(m2/ha) was measured using a 2-factor cruising prism.

Nest Monitoring

I visited each stand once every 3-4 days to search for nests. When a nest was located, a piece of coloured flagging tape was placed within 10 m of the nest and its geographical co-ordinates were taken using a GPS unit. I then monitored nests using a mirror attached to a telescoping pole every 3-4 days to determine nest initiation date, clutch size, hatch date, number of nestlings, and number of fledglings per nest. Mean values for nest initiation date, clutch size, number of fledglings per successful nest, and number of fledglings per nest were calculated for each stand. Whenever possible, exact 23 dates for nest initiation, hatching, and fledging were confirmed by nest checks, but in other instances (e.g., canopy nests) published lengths for nest building (4 days), incubation (12 days), and nestling development (10 days) were used to help calculate important dates (Scott 1998). The number of eggs present in 12 of 52 (23.1%) nests could not be visually confirmed, so clutch size was estimated using the mean number of eggs per nest in the stands where these twelve nests were located. I did not include nests that were initiated on or after June 3rd in my analysis of nest initiation dates (range = May

18th – June 2nd) due to uncertainty as to whether they were first attempts or not. I calculated nest survival for each stand using the Mayfield method (Mayfield 1975).

Density and Pairing Success

In 2007 and 2008 I estimated the density of breeding Rose-breasted Grosbeaks in each stand using a modified version of the spot mapping technique (Bibby et al. 2000). I marked the locations of all Rose-breasted Grosbeaks that were seen or heard during nine visits to all stands between May 10th and June 2nd using a Global Positioning System

(GPS). I stopped mapping on June 2nd due to uncertainty in determining whether individuals observed after this point had relocated after initial nest attempts in other locations (Bibby et al. 2000, Dunford and Freemark 2005). Males that were observed in the same location on three or more occasions were considered territorial. In territories where no nest was located, a male was assumed to be paired if a female was observed in his territory on at least one occasion, and unpaired if no female was observed after a total of 90 minutes of observation during one or more observation periods (Bibby et al. 2000). 24

The proportion of after-second-year (ASY) males relative to second-year (SY) males was also calculated in each stand by recording the age category of each singing male that was mapped over the course of the season. ASY males can be visually distinguished from SY males by their uniformly black primaries, black heads and backs, and well-defined markings (Pyle 1997). In contrast, the head and back of SY males are often streaked with brown, there is contrast between their all-brown or partially-brown primaries and their darker greater coverts, and overall their markings are much less distinct (Pyle 1997). I calculated density (no. males/ ha) and pairing success (no. paired males / total no. males) for each stand.

Population Productivity

I calculated two different measures of productivity. The first measure was used to assess the mean number of Rose-breasted Grosbeak fledglings produced per active nest in each post-harvest treatment (0-5, 16-20, 21-25, and > 50 years post-harvest).

For the second measure of productivity, I estimated whether population growth rates were above or below replacement levels in each harvest treatment. For a population to be self-sustaining, the annual female mortality rate must not exceed the average number of female offspring produced per female that survive to breed (Ricklefs 1973,

Pulliam 1988, Donovan et al. 1995). In other words, the long-term change in population size (λ) must be ≥ 1. I calculated the finite rate of increase for Rose-breasted Grosbeaks in each post-harvest treatment using a two-age-class matrix population model in the

Microsoft Excel add-in program PopTools version 3.2.3 (Hood 2010). Confidence 25 intervals for population λ were calculated using the bootstrapping technique with 100 iterations in the PopTools program.

The two-age-class model for calculating λ incorporates estimates of annual adult and juvenile survival, which I did not directly measure. Following Moore et al. (2010), I used estimates of annual adult survivorship ranging from 0.40 – 0.61 obtained from literature on other Neotropical migratory songbirds. I calculated annual juvenile survivorship for each treatment as the product of fledgling survival and annual adult survival, based on the assumption that once juveniles reach independence their annual survival rate is similar to that of adults (Moore et al. 2010). To calculate annual female fecundity (number of female offspring per female per year) I assumed that Rose-breasted

Grosbeak young occur in a 1:1 sex ratio, that females who initially failed to successfully nest would re-nest only once, and that the probability of nest success for the re-nesters would be similar to that experienced during first nest attempts (i.e., if 30% of nests failed during the first attempt, approximately 30% of nests in the second attempt would fail)

(Smith et al. 2006, Moore et al. 2010). Following Moore et al. (2010) I used these values to calculate population growth rates under best case (high adult and juvenile survival) and worst case (low adult and juvenile survival) scenarios.

Statistical Analyses

Habitat Measurements

I tested for significant (P < 0.05) correlations among habitat variables using the

Spearman rank correlation coefficient. When significant rank correlations were identified between independent variables, the measure I judged to be of lesser ecological 26 importance to Rose-breasted Grosbeak demography was omitted from further analysis

(Burnham and Anderson 2002, Driscoll et al. 2005). During this process seedling cover was omitted due to significant (P < 0.05) correlations with regenerative growth (+), saplings (-), canopy (-) cover, and basal area (-). Percent cover in the canopy and sub- canopy layers were significantly positively correlated, so these variables were combined into one measure (hereafter referred to as canopy cover). Sapling cover was positively correlated with understory cover (r = 0.56, P < 0.05), but both variables were retained because they were judged to be of importance to Rose-breasted Grosbeak ecology.

I used redundancy analysis (RDA; ter Braak and Smilauer 1998) to test whether five habitat variables differed significantly among harvesting treatments (0-5, 16-20, 21-

25, and > 50 years post-harvest) and study years (2006-2008). Habitat variables were then compared among treatments individually using two-way analysis of variance

(ANOVA).

For each response variable a mean value was calculated for each stand in each year. In all tests, each stand mean was weighted according to the number of survey plots collected in that stand. I first tested the full model. Since the two-way interactions between year and post-harvest treatment were clearly not significant (i.e., P > 0.15) for all habitat variables tested, I removed the interaction terms and re-ran the models.

Significant differences among post-harvest treatments were investigated using two-way analysis of variance followed by Tukey’s Studentized post-hoc range test (α = 0.05). For all ANOVAs, model assumptions of homogeneity and normality were verified by examination of residuals, and no transformations were necessary. Correlations and model assumptions were tested in Statistica 7 (StatSoft Inc. 2004), redundancy analysis was 27 performed in CANOCO, and all other analyses were conducted in the PROC GLM module of SAS v. 8.02 (SAS Institute Inc., 1999).

Demographic Variables

I used redundancy analysis to investigate nine demographic variables (territory density, pairing success, proportion of ASY compared to SY males, nest initiation dates, clutch size, number of fledglings per successful nest, daily nest survival, period nest survival, and productivity) as a function of harvesting treatment (0-5, 16-20, 21-25, and >

50 years post-harvest) and study year. Demographic variables were then compared individually among harvest treatments using two-way ANOVAs.

Estimates of nest initiation date, clutch size, and survival were available for all three years (2006-2008), but all other demographic variables were only estimated for

2007 and 2008. For each response variable a mean value was calculated for each stand in each year. In all tests, each stand mean was weighted according to the number of observations collected in that stand. I first tested the full model (treatment, year, and treatment × year interaction), but re-ran the models without the interaction term since it was clearly not significant for any of the variables measured (i.e., P > 0.15). Significant differences among post-harvest treatments were investigated using two-way analysis of variance followed by Tukey’s Studentized post-hoc range test (α = 0.05). For all

ANOVAs, model assumptions of homogeneity and normality were verified by examination of residuals, and no transformations were necessary. 28

Relationship Between Habitat Features and Demographic Variables

I used redundancy analysis (RDA; ter Braak and Smilauer 1998) to examine the overall relationship between demographic variables and habitat characteristics in the four post-harvest treatments. Five habitat variables were retained in the analysis as well as nine demographic variables.

Male Age

I used a procedure similar to the one used for the demographic variables to compare demographic measurements and habitat characteristics at nests attended by ASY and SY males in the four post-harvest treatments. The habitat characteristics I investigated were percent cover by regenerative growth, saplings, understory, and canopy and basal area. Too few SY males were present in most sites to allow separate comparisons for each study year. I pooled measurements across study years by calculating an average for each site in each study year, then taking a weighted average of these values, with each value weighted according to the number of observations in each site. As in the above analysis, I investigated the differences in demographic variables and habitat characteristics among post-harvest treatments and between nests attended by ASY and SY males using two-way ANOVAs followed by Tukey’s Studentized post-hoc range test (α = 0.05).

RESULTS

Habitat Measurements

Redundancy analysis indicated that vegetation variables differed significantly with time since single-tree selection harvesting (F-ratio = 3.642, P = 0.002). Tukey’s 29 tests showed that the removal of trees resulted in significantly lower basal area and lower cover from saplings and understory in 0-5 years post-harvest stands than in other treatments (Table 1.2). As a result of the openings created by harvesting, these sites also had significantly higher cover from regenerative growth than other stands. In the 16-20 years post-harvest stands, regenerative growth had developed into a denser understory layer than was present in other post-harvest treatments. Habitat features 21-25 years after harvesting were not significantly different from those in the 16-20 years post-harvest stands in most study years, but cover from regenerative growth and saplings was significantly lower than in the > 50 years post-harvest stands. The “un-logged” stands (>

50 years post-harvest) had higher basal area and higher sapling and canopy cover than were present in the other post-harvest treatments.

Demographic Variables

Mean density of Rose-breasted Grosbeaks in this study was 0.212 (SE = 0.01) territories / ha, mean pairing success was 0.963 (SE = 0.01), and the mean proportion of

ASY males was 0.629 (SE = 0.11), indicating that more after-second-year males successfully held territories than second-year males. Between 2006 and 2008 I monitored a total of 173 nests (20, 73, and 80 nests in the years 2006-2008 respectively), and I was able to determine exact nest initiation dates for 157 first nest attempts. Mean nest initiation date was May 26th (range = May 18th – June 2nd), with one outlying nest started on May 15th, 2006. Nest contents could be verified for 52 nests, yielding a mean clutch size of 3.81 eggs, and an average of 3.44 fledglings per successful nest. Mean daily survival rate (DSR) for Rose-breasted Grosbeak nests was 0.967 (n = 173 nests), mean 30

TABLE 1.2. Means of habitat variables (standard error in brackets) and results from two-way ANOVAs and Tukey’s tests comparing sites from four stages in the cutting cycle: 0-5, 16-20, 21-25, and > 50 years post-harvest (n = 3 per stage) and three study years (2006- 2008). Time Since Harvest (years)1 P P Variable 0-5 16-20 21-25 >50 (Treatment) (Year) Forb cover (%) 2006 8.5 (3.5) 3.3 (1.4) 23.0 (6.1) 12.0 (0.1) 0.051 0.029 2007 8.6 (2.2) 2.0 (0.6) 4.4 (1.6) 5.2 (1.6) 2008 6.3 (2.0) 2.9 (0.9) 7.2 (0.4) 6.7 (1.6) Regeneration (%) 2006 14.5 (0.5)a 4.3 (0.6)bc 4.3 (0.2)b 12.0 (0.1)ac <0.001 0.061 2007 17.8 (3.2)a 7.1 (3.0)bc 4.2 (1.2)b 11.4 (1.6)ac 2008 23.4 (4.7)a 9.1 (2.9)bc 7.2 (3.3)b 18.4 (5.6)ac Sapling cover (%) 6.3 (0.9)a 16.0 (3.9)ab 9.9 (2.5)a 25.9 (4.1)b <0.001 0.108 Under-story cover (%) 20.8 (2.8)a 34.5 (4.1)b 27.1 (2.9)ab 25.3 (5.4)ab 0.066 0.516 Canopy cover (%) 81.2 (6.8) 90.3 (4.8) 89.4 (3.8) 93. 8 (7.2) 0.376 0.060 Basal area(m2/ha) 22.1 (1.2)a 27.2 (1.2)b 27.0 (0.8)b 28.1 (1.8)b 0.006 0.954 1Superscript letters indicate significant differences between silvicultural treatments from Tukey’s HSD post-hoc test (α = 0.05). 31 period survival rate (PSR) for the entire 24 day period was 0.559, and mean productivity was 2.02 fledglings per nest.

Density, pairing success, and the number of fledglings per successful nest were all significantly lower in the > 50 years post-harvest stands than in at least one other post- harvest treatment, and nest survival and productivity were also lowest in the > 50 years post-harvest stands (Table 1.3). The proportion of ASY males compared to SY males was highest in the > 50 years post-harvest stands and lowest in the 21-25 years post- harvest treatment, but differences were not significant (P = 0.616; Table 1.3). Nests were initiated significantly later in 2008 than in 2006 or 2007, and productivity estimates were significantly higher in 2008 than in 2007.

On average, female fecundity was 1.02 female fledglings per female, with Rose- breasted Grosbeak females producing 1.01 – 1.02 female fledglings / female / year (Table

1.4). Under the worst case scenario (low adult and juvenile survivorship) all post-harvest treatments were population sinks. Under the best case scenario (high adult and juvenile survival), the 16-20 years post-harvest treatment was a source, but all other treatments remained sinks.

Relationship Between Habitat and Demographic Variables

The first two axes of the RDA analysis explained 57% of the variance in demographic variables (41.6% and 15.4% for the first and second axes, respectively; Fig.

1.2). Sapling cover and basal area were strongly associated with the > 50 years post- harvest stands, whereas understory and forbs were weakly associated with this treatment.

Regeneration and canopy cover were weakly associated with 0-5 years post-harvest 32

TABLE 1.3. Means of demographic variables (standard error in brackets) and results from two-way ANOVAs and Tukey’s tests comparing sites from four stages in the cutting cycle: 0-5, 16-20, 21-25, and > 50 years post-harvest (n = 3 per stage) and three study years (2006-2008). Sample sizes (n) show number of nests per treatment, but may vary slightly among demographic measures. Time Since Harvest (years)1 0-5 16-20 21-25 >50 P P Demographic Variable n = 55 n = 34 n = 38 n = 29 (Treatment) (Year) Density (no. territories /ha) 0.246 (0.02)ab 0.254 (0.03)a 0.187 (0.02)ab 0.162 (0.01)b 0.031 0.727 Pairing success (prop. males paired) 0.990 (0.01)a 0.972 (0.02)a 1.000 (0.0)a 0.892 (0.01)b <0.001 0.072 Proportion of territorial ASY males 0.598 (0.08) 0.634 (0.08) 0.550 (0.09) 0.725 (0.11) 0.616 0.473 Nest Initiation Date 2006 May 27 (5.50) May 26 (1.81) May 24 (1.28) May 24 (1.03) 0.430 <0.001 2007 May 24 (0.88) May 24 (1.03) May 22 (0.87) May 25 (1.12) 2008 May 27 (0.46) May 28 (0.56) May 27 (0.87) May 27 (0.76) Clutch size 4.000 (0.07) 3.667 (0.17) 3.625 (0.12) 3.833 (0.40) 0.477 0.283 No. fledglings / successful nest 3.619 (0.13)a 3.333 (0.24)ab 3.500 (0.16)ab 2.833 (0.48)b 0.012 0.060 Mayfield period nest survival rate 0.524 (0.09) 0.598 (0.09) 0.654 (0.12) 0.437 (0.09) 0.350 0.308 Productivity (no. fledglings / nest) 2.305 (0.35) 1.603 (0.28) 2.282 (0.47) 1.383 (0.62) 0.593 0.113 1Superscript letters indicate significant differences between silvicultural treatments from Tukey’s HSD post-hoc test (α = 0.05).

33

TABLE 1.4. Population growth rates for Rose-breasted Grosbeaks breeding in 12 tolerant hardwood stands in Algonquin Provincial Park harvested by single-tree selection. Stands represent four stages in the cutting cycle: 0-5, 16-20, 21-25, and > 50 years post- harvest. Models show intrinsic (λ) and exponential (r) population growth rates under best case (high adult and juvenile survival) and worst case (low adult and juvenile survival) scenarios. Treatments are designated as sources if λ is ≥ 1 and a sink if λ is < 1. Time since Adult survival Annual juvenile Female Population status harvest (years) probability a survival b fecundity c λ Upper and lower 95% CI Best Case Scenario 0-5 0.61 0.337 1.019 0.966 0.864 - 1.070 Sink 16-20 0.61 0.417 1.012 1.025 0.922 - 1.128 source 21-25 0.61 0.249 1.022 0.893 0.790 - 0.996 Sink > 50 0.61 0.198 1.012 0.848 0.745 - 0.951 Sink Worst Case Scenario 0-5 0.40 0.221 1.019 0.713 0.610 - 0.816 Sink 16-20 0.40 0.274 1.012 0.760 0.657 - 0.863 Sink 21-25 0.40 0.163 1.022 0.649 0.546 - 0.752 Sink > 50 0.40 0.130 1.012 0.614 0.511 - 0.717 Sink a Estimates obtained from literature on other Neotropical migratory songbirds b Fledgling survival probability multiplied by annual juvenile survival rate, assumed to be equal to adult survival probability c Mean no. of female offspring / successful nest / year, assuming one attempt at re-nesting following nest failure 34 stands. All demographic variables were weakly associated with harvested stands, except for nest initiation date, the proportion of territorial ASY males compared to SY males, and clutch size, which were not clearly associated with any treatment.

Age of Male Parent

Nests attended by ASY males were initiated significantly earlier than nests attended by SY males (F1,3 = 19.97, P = 0.0002) (Table 1.5). Territory density was significantly higher for ASY males than SY males in all post-harvest treatments (F1,3 =

10.15, P = 0.005), and productivity (no. fledglings produced per active nest) was also significantly higher for ASY males than for SY males (F1,3 = 3.94, P = 0.050).

Differences between nests attended by ASY males and those attended by SY males were not significant for clutch size (F1,3 = 0.02, P = 0.892), mean number of fledglings produced per successful nest (F1,3 = 0.66, P = 0.429), daily survival rate (F1,3 = 1.56, P =

0.226), or period survival rate (F1,22 = 1.32, P = 0.174), but all of these measures were slightly higher at nests attended by ASY males than those attended by SY males.

Multivariate analysis of variance indicated that there were no significant differences in percent cover by regenerative growth, saplings, understory, and canopy cover or in basal area at nests attended by ASY and SY males (F5, 14 = 0.42, P = 0.827).

The interaction between post-harvest treatment and male age (ASY or SY) was also not significant (F15, 39.05 = 0.78, P = 0.691). 35

FIGURE 1.2. Relationship between seven demographic variables for Rose-breasted Grosbeaks and six habitat features in four post-harvest treatments in Algonquin Provincial Park. Demographic variables are mean clutch size, territory density, number of fledglings per successful nest, pairing success, productivity, proportion of territorial ASY males, and nest survival. Habitat features are percent cover by forbs, regenerative growth, saplings, understory, and canopy and basal area. Filled triangles = 0-5 years post- harvest stands, filled squares = 16-20 years post-harvest stands, filled octagons = 21-25 years post-harvest stands, empty circles = > 50 years post-harvest stands. Inter-sample scaling was used, and the species scores were centered and standardized (RDA on a correlation matrix between demographic variables). The first axis explained 41.6% of the variance in demographic variables and the second axis explained 15.4%. 36

TABLE 1.5. Means of demographic variables (standard error in brackets) and results from two-way ANOVAs and Tukey’s tests comparing sites from four stages in the cutting cycle: 0-5, 16-20, 21-25, and > 50 years post-harvest (n = 3 per stage) and three study years (2006-2008). Results are shown for after-second-year (ASY) and second-year (SY) males. Bold values indicate significant differences between SY and ASY males. Time Since Harvest (years)1 Demographic Variable Male Age 0-5 16-20 21-25 >50 Density (no. territories /ha) ASY 0.122 (0.04) 0.114 (0.04) 0.081 (0.02) 0.091 (0.03) SY 0.071 (0.00) 0.054 (0.00) 0.057 (0.01) 0.029 (0.01) Nest initiation date ASY May 24 (0.35) May 24 (0.78) May 22 (1.49) May 25 (1.65) SY May 27 (1.24) May 27 (0.87) May 28 (0.88) May 28 (1.16) Clutch size ASY 4.000 (0.14) 3.625 (0.09) 3.833 (0.12) 4.000 (0.58) SY 4.000 (0.00) 4.000 (-) 3.800 (0.283) 2.000 (0.58) No. fledglings / successful nest ASY 3.715 (0.17) 3.250 (0.12) 3.667 (0.236) 3.667 (0.33) SY 3.602 (0.23) 4.000 (-) 3.800 (0.283) 2.000 (0.58) Daily nest survival ASY 0.982 (0.00) 0.983 (0.01) 0.993 (0.00) 0.980 (0.00) SY 0.979 (0.00) 0.974 (0.00) 0.983 (0.01) 0.982 (0.00) Period nest survival ASY 0.674 (0.09) 0.699 (0.11) 0.864 (0.05) 0.632 (0.03) SY 0.627 (0.04) 0.545 (0.03) 0.709 (0.16) 0.678 (0.14) Productivity (no. fledglings / pair) ASY 2.833 (0.47) 1.722 (0.57) 2.798 (0.27) 1.710 (0.18) SY 1.526 (0.49) 1.638 (0.48) 2.083 (0.69) 1.000 (0.41) 1Superscript letters indicate significant differences between silvicultural treatments from Tukey’s HSD post-hoc test (α = 0.05). 37

DISCUSSION

Habitat characteristics differed significantly among stands harvested by single- tree selection 0-5, 16-20, 21-25, and > 50 years previously. As in other studies, the most recently harvested stands had significantly higher cover from regenerative growth, lower basal area, and lower sapling, understory, and canopy cover than other post-harvest treatments (Robinson and Robinson 1999, Flaspohler et al. 2002, Jobes et al. 2004,

Moore et al. 2010). Increased light penetration in newly logged stands typically promotes the establishment of shade-tolerant sugar maple and American beech seedlings (Finegan

1984, Jobes et al. 2004), and in this study resulted in significantly higher understory cover by 16-20 years post-harvest. As previously demonstrated, the >50 years post- harvest stands had higher basal area and higher sapling cover than other stands (Robinson and Robinson 1999, Flaspohler et al. 2002, Jobes et al. 2004).

Single-tree selection harvesting did not appear to have a significant negative effect on any of the demographic measures included in this study, and density, pairing success, and the number of fledglings produced per successful nest were all significantly lower in the > 50 years post-harvest treatment than in the harvested stands. However,

Rose-breasted Grosbeaks were below replacement levels in all stands except the 16-20 years post-harvest treatment, even using the most optimistic measures of adult and juvenile survival.

Demographic Measures

In the predominantly forested landscape of Algonquin Provincial Park Rose- breasted Grosbeak density ranged from 0.16-0.24 male territories per hectare, which is 38 similar to the range of densities (0.20-0.24) reported for forest fragments in south-western

Ontario (Smith et al. 2006). However, whereas density was lower in the uncut stands than in the harvested ones in the predominantly forested landscape, the opposite was true in forest fragments in the agricultural landscape, where Rose-breasted Grosbeaks appeared to be attracted to the higher shrub cover, particularly in heavily cut stands

(Smith et al. 2006). Nest predation and brood parasitism by Brown-headed Cowbirds

(Molothus ater) are often higher (Paton 1994, Hoover et al. 1995, Rodewald et al. 2002,

Phillips et al. 2005, Falk et al. 2011), and pairing success is lower (Gibbs and Faaborg

1990, Villard et al. 1993, Van Horn et al. 1995, Burke and Nol 1998) in fragmented agricultural landscapes than in continuously forested regions. As a result, openings in the understory and canopy created by silviculture, which potentially expose songbird nests to visual predators and brood parasites, could have a greater impact in fragmented landscapes than in predominantly forested regions (Andrén 1995, Hanski et al. 1996).

Higher rates of nest predation and brood parasitism following harvesting could result in lower recruitment of individuals into selection harvested stands, and therefore lower densities relative to un-harvested fragments. Rose-breasted Grosbeak nests are parasitized in fragmented agricultural landscapes (Smith et al. 2006, 2007), but were not in my predominantly forested landscape, and estimates of nest survival were considerably lower for the fragmented landscape (0.341-0.390; Smith et al. 2006, 2007, Moore et al.

2010) than in my study area (0.559). These results suggest that the effects of timber harvesting to some extent depend on landscape context, and that adverse effects may be fewer in forested regions than in fragmented landscapes. 39

Rose-breasted Grosbeak productivity, represented by the number of fledglings produced per successful nest, was highest in the 0-5 years post-harvest stands and lowest in the > 50 years post-harvest treatments. The number of fledglings produced per successful nest ranged from 2.8-3.6, which is comparable to, but slightly higher than estimates from southwestern Ontario, which ranged from 2.4 (Smith et al. 2006) to 3.2

(Moore et al. 2010). While more young were fledged on average from successful nests in uncut stands (2.8) than in harvested ones (2.4) in the fragmented landscape (Smith et al.

2006), in the predominantly forested landscape significantly more young were fledged per successful nest in 0-5 years post-harvest stands (3.6) than in uncut sites (2.8).

Estimates of the number of fledglings produced per active nest, which incorporate nest survival rates, were also lower in the > 50 years post-harvest stands than in the more recently harvested treatments in my study landscape, but not significantly so. Together these results suggest that single-tree selection harvesting may have less of a negative impact on Rose-breasted Grosbeak productivity in forested regions than in more fragmented landscapes.

As predicted, density was highest in the 16-20 year post-harvest stands and the population growth model indicated that these stands were at or closer to replacement levels than the other post-harvest treatments. The 16-20 years post-harvest treatment appeared to attract Rose-breasted Grosbeaks and to have high productivity, but in general, most demographic measures were higher in more recently cut stands than in the un-logged ones. Previously harvested stands were associated with higher regenerative growth and understory cover, both of which are features previously identified as important to Rose-breasted Grosbeak demography (Smith et al. 2006, 2007). Other 40 studies have found greater abundances of early-successional species in regenerating forests as well (Annand and Thompson 1997, Simon et al. 2000, Gram et al. 2003,

Holmes et al. 2004, Jobes et al. 2004, Holmes and Pitt 2007, Robertson and Hutto 2007,

Atwell et al. 2008), suggesting that forest management may provide additional nesting habitat and/or food resources for disturbance-dependent songbirds. In my study Rose- breasted Grosbeaks built their nests at a wide range of heights (1.5 – 22 m), and they were observed foraging in vegetation ranging from ground level to the upper layers of the canopy. This species’ ability to demonstrate plasticity in nest site selection and foraging substrate use may help explain why all demographic measures were not clearly higher in one post-harvest treatment than the others.

Population Productivity

Population growth estimates indicated that even under the best case scenario,

Rose-breasted Grosbeaks in my study area were not replacing themselves except in the

16-20 years post-harvest treatments. Although population growth estimates have not previously been determined for Rose-breasted Grosbeaks in predominantly forested landscapes, this species is frequently found in regenerating forests 15 years after harvesting (Pelletier and Dauphin 1996), and often increases in abundance following logging (i.e., Jobes et al. 2004, Doyon et al. 2005). This study indicates that the dense understory and sapling layers characteristic of hardwood stands 16-20 years post-harvest are important for Rose-breasted Grosbeaks, and that access to regenerating forests may provide some benefits for this species in forested areas. 41

Population growth models are highly sensitive to small changes in estimates of annual survival rates for adults and juveniles (Porneluzi and Faaborg 1999, Powell et al.

1999, Woodworth 1999), which despite technological advancements in radio tracking are still difficult to accurately obtain (Greenwood and Harvey 1982, Payne 1991, Paradis et al. 1998, Anders and Marshall 2005). Common responses to this problem are to estimate juvenile survival by dividing adult mortality by the average number of offspring fledged

(Ricklefs 1973), to assume juveniles are half as likely to survive as adults (Greenberg

1980), or to adopt 0.31 as an estimate of juvenile survival (Temple and Carey 1988).

Empirical estimates of juvenile survival from fledging to fall migration range from 0.28 to 0.75 for Juncos (Junco phaenotus), Wood Thrush (Hylocichla mustelina), and Lark

Buntings (Calamospiza melanocorys) (Sullivan 1989, Anders et al. 1997, Powell et al.

2000, Yackel Adams et al. 2001). Annual adult survival is often assumed to be 0.62

(Anders and Marshall 2005), and empirical estimates of annual survival for breeding songbird females range from 0.40-0.55 (Cilimburg et al. 2002, Sillett and Holmes 2002,

Gardali et al. 2003, Stutchbury et al. 2009). I incorporated a range of values for adult and juvenile survival by calculating growth rates under a worst case scenario (adult survival =

0.40, and low juvenile survival) and a best case scenario (adult survival = 0.61, and high juvenile survival), as well as using stand specific values for female fecundity, nest success, and fledgling survival to improve accuracy (Anders and Marshall 2005).

Although these results should be accepted with caution, this study suggests that harvested hardwood stands in predominantly forested landscapes may be population sources for

Rose-breasted Grosbeaks by 16-20 years post-harvest. 42

Population Age Structure

The proportion of territorial ASY males compared to territorial SY males was investigated to determine whether older, more experienced males were out-competing younger, less-experienced ones for “high quality” habitats. Although breeding pairs with

ASY males began nest building significantly earlier than pairs with SY males, my prediction that the 16-20 years post-harvest treatment would have the highest proportion of ASY males was not supported. The proportion of territories held by experienced males was highest in the un-harvested stands, although among-treatment differences were not significant. According to the ideal-free distribution theory, a higher proportion of older, more experienced males in the > 50 years post-harvest stands indicates that these stands may provide better quality habitat for Rose-breasted Grosbeaks. However, since overall density, pairing success, and productivity were lowest in this treatment, this interpretation seems unlikely. Instead, I propose that ASY males may have been out- competing SY males in this treatment for a more limited number of available territories than were offered in harvested stands. Older males were less productive in the > 50 years post-harvest stands than in other treatments, but even in these stands their productivity was higher than that of younger males in most other treatments. The opposing patterns in density, nest initiation dates, productivity, and proportion of older, more experienced males among the post-harvest treatments indicates that the factors which influence reproductive success and habitat quality are complicated, and as a result, assessments of the effects of harvesting or other disturbances should be based on more than one demographic measure whenever possible. 43

Rose-breasted Grosbeak nests attended by after-second-year males were started significantly earlier, and had significantly higher productivity than nests attended by second-year males, suggesting that more experienced breeders had a reproductive advantage over less experienced birds. First-time breeders of other species have shown lower nest survival than older individuals (Pärt 2001, Forschler and Kalko 2006) and they sometimes occupy poorer quality habitats (Greenwood and Harvey 1982, Møller 1991,

Holmes et al. 1996). In some cases, second-year birds may have limited access to high- quality territories as a result of later arrival on the breeding grounds, incomplete knowledge of where to find good territories, and despotic distribution of more experienced breeders (Sherry and Holmes 1989, Holmes et al. 1996, Bayne and Hobson

2001, Part 2001a). However, in my study habitat characteristics at nests attended by SY males did not significantly differ from those at nests with ASY males, suggesting that differences in territory quality were likely not the main cause of differences in reproductive success. Lack of experience or differences in timing may be more likely explanations for the lower reproductive success of first-time Rose-breasted Grosbeak breeders.

Single-tree selection harvesting did not have a significantly negative effect on any of the demographic measures investigated in this study. However, since fewer birds were present in the > 50 years post-harvest stands, sample sizes for most demographic measures were smaller in this treatment than in the others. In addition, nests in the 21-25 and > 50 years post-harvest stands tended to be higher up, making precise estimates of nest initiation dates, brood sizes, and nest outcomes more difficult to obtain. Weighted analyses were used in an effort to address these potential differences in sample size, but it 44 is possible they still had an effect on among-treatment comparisons. Finally, a small proportion of territorial males (3-5%) could not be accurately aged, and although the proportion was consistent among treatments, this source of uncertainty may have affected the among-treatment comparisons. Increasing the number of stands studied might be one way for future studies to increase the power of the tests and confirm that no significantly negative effects are occurring as a result of single-tree selection cutting.

Conclusions

Density, pairing success, and productivity were higher in the more recently harvested stands than in the > 50 years post-harvest treatment, suggesting that in tolerant hardwood stands in predominantly forested landscapes, single-tree selection harvesting does not appear to have negative effects on Rose-breasted Grosbeaks. However, the proportion of older, more experienced males was highest in the > 50 years post-harvest stands, indicating that these may have been perceived as providing higher quality habitat by Rose-breasted Grosbeaks than more recently harvested stands. In addition, population growth rates calculated from a wide range of realistic estimates of juvenile and adult survival were below replacement levels in my study landscape, except in the 16-20 years post-harvest stands, suggesting that further research is needed to help identify the underlying causes of the decline in this species’ numbers. Selection harvesting influenced Rose-breasted Grosbeaks differently in the predominantly forested region than in the fragmented landscape, underscoring the importance of considering landscape and regional contexts when designing silvicultural prescriptions (Thompson et al. 1995). CHAPTER 2: SURVIVAL AND HABITAT USE BY ADULT AND FLEDGLING ROSE-BREASTED GROSBEAKS (PHEUCTICUS LEUDOVICIANUS) IN A FORESTED LANDSCAPE HARVESTED BY SINGLE-TREE SELECTION

ABSTRACT

Many songbirds that nest in the managed hardwood forests of North America are currently declining, but although fledgling survival can have a considerable influence on population growth, few studies have investigated how silvicultural practices influence fledglings. I studied nest and fledgling survival and habitat use of Rose-breasted Grosbeaks

(Pheucticus leudovicianus) in twelve hardwood stands in a predominantly forested landscape that had been harvested by single-tree selection 0-5, 16-20, 21-25, and > 50 years previously.

Mean nest and fledgling survival in each treatment were estimated using the Mayfield method, and I used an information theoretic approach to assess relative support for models explaining nest and fledgling survival as a function of habitat characteristics. Nest sites had higher cover from regenerative growth, saplings, and understory, and lower basal area than random locations. During their first week out of the nest, fledglings used locations with significantly higher cover from regenerative growth, saplings, small shrubs, and raspberry and elderberry bushes than were present at random locations. Mayfield daily survival for

Rose-breasted Grosbeak nests (n = 173) was 0.967 and period survival rate was 0.559. Daily fledgling survival (n = 46) was 0.920 and period survival over the 21 day observation period was 0.551. Neither nest nor fledgling survival differed significantly among post-harvest treatments (P > 0.05). The best model for explaining nest survival included basal area, nest initiation date, and sampling year, and suggested that nests initiated earlier in the season and built in locations with higher basal area had a better chance of survival. The best model for explaining fledgling survival included the number of days since fledging, indicating that as

45 46 fledglings age their probability of surviving increases. Habitat characteristics at nest and fledgling locations were significantly different, and estimates of nest and fledgling survival were not correlated among harvested stands, suggesting that assessments of habitat quality for songbirds should not be limited to the nest stage. The lack of significant difference in survival among post-harvest treatments suggests that single-tree selection harvesting is likely not contributing to the observed declines in Rose-breasted Grosbeak populations. 47

INTRODUCTION

During the past decade there has been a shift in North America toward developing and implementing forest management practices that mimic natural disturbance patterns

(Seymour et al. 2002). Under these guidelines, single-tree selection has been implemented in much of the tolerant hardwood forests in northeastern North America (Tubbs 1977, OMNR

1998, Seymour et al. 2002). Although this low-intensity method of harvesting should have minimal long-term impacts on wildlife (OMNR 2000), many migratory songbirds that breed in this forest type are currently experiencing population declines (Newton et al. 2004,

Blancher et al. 2009). For some passerines nest survival rates remain stable before and after single-tree selection harvesting (Bourque and Villard 2001, Duguay et al. 2001, Gram et al.

2003, Smith et al. 2006, Dellinger et al. 2007, Cooper et al. 2009, Moore et al. 2010), whereas others experience lower rates of nest success in harvested forests (Barber et al. 2001,

Robinson and Robinson 2001, Robertson and Hutto 2007, Poulin et al. 2010). Nest survival and habitat requirements have been well described for many North American birds (Akresh et al. 2009), but far fewer studies have investigated fledgling survival and habitat requirements once the young leave the nest (Anders and Marshall 2005, King et al. 2006,

Yackel Adams et al. 2006, Poulin et al. 2010). Since evaluating the long-term impacts of forest management and other factors on declining songbird populations requires accurate population models (Stutchbury et al. 2009) that are highly sensitive to estimates of juvenile survival (Porneluzi and Faaborg 1999, Powell et al. 1999, Woodworth et al. 1999), studies of fledgling survivorship are necessary.

The fledglings of many songbirds that nest in managed hardwood forests across North

America spend up to eight weeks on the breeding grounds prior to migration (Anders et al.

1998). Mortality during this period is typically high, especially during the first week out of 48 the nest when the young are unable to fly very well (Anders et al. 1997, 1998, Vega Rivera et al. 1998, Monrós et al. 2002, Kershner et al. 2004, Yackel Adams 2006, Rush and Stutchbury

2008). Recent studies have shown that the habitat requirements of breeding adults may be different from those of family groups during the post-fledging period (Rappole and Ballard

1987, Anders et al. 1998, Vega Rivera et al. 1998, Cohen and Lindell 2004, King et al. 2006; but see Berkeley et al. 2007, Ciudad et al. 2009). For example, the amount and structure of understory cover is often important during the first week post-fledging (King et al. 2006,

Mitchell et al. 2010), and the young of several species move out of the open, mature forest used by adults for nest building into edge or early- to mid-successional forests (Anders et al.

1998, Vega Rivera et al. 1998, Pagen et al. 2000, Vitz and Rodewald 2006). These differences suggest that conclusions based solely on studies of nest success may be insufficient to assess the long-term impacts of forest management on songbird populations.

Recent studies have found correlations between fledgling survival and the amount of coarse woody debris (Mitchell et al. 2010), the amount and structure of understory cover

(Mitchell et al. 2010), shrub cover (Moore et al. 2010), fledging date (Monros et al. 2002,

Yackel Adams et al. 2006, Moore et al. 2010), and weather (Yackel Adams et al. 2006). A number of studies found that fledglings use areas with dense cover (Anders et al. 1998, Vega

Rivera et al. 1998), locations with greater than average vertical structure 0-3 m above ground

(King et al. 2006), and regenerating clear-cuts (Pagen et al. 2000, Vitz and Rodewald 2006).

One hypothesis to explain these patterns of resource use is the predator avoidance hypothesis, which suggests that young birds seek habitats with more cover because they provide protection from potential predators, especially during pre-migration molting (Anders et al.

1998, Vega Rivera et al. 1998). Alternatively, the optimal foraging hypothesis suggests that juveniles’ movements are primarily determined by the need to find food (Vega Rivera et al. 49

1998). We are beginning to identify some of the habitat types and features that are important for songbird fledglings, but we do not yet have a good understanding of the primary factors driving resource use during the post-fledging period.

Studies of fledgling survival and habitat requirements have recently become more prevalent, but to my knowledge only two studies have investigated fledgling survival in managed forests (Lang et al. 2002, Moore et al. 2010). Thinning and burning in a > 30 year old loblolly pine (Pinus taeda) forest had no effect on habitat use prior to parental independence or on dispersal distances of Wood Thrush fledglings (Hylocichla mustelina)

(Lang et al. 2002). Similarly, the survival of Rose-breasted Grosbeak (Pheucticus leudovicianus) fledglings in deciduous forest fragments in an agricultural landscape was not significantly affected by single-tree or diameter-limit harvesting (Moore et al. 2010). Single- tree selection cutting results in stands which are characterized by abundant regenerative growth, low basal area, and sparse cover in the sapling, understory, and canopy layers a few years after harvesting (Robinson and Robinson 1999, Flaspohler et al. 2002, Jobes et al.

2004). By 16-20 years post-harvest a dense understory layer typically forms (Finegan 1984,

Jobes et al. 2004), and eventually stands re-develop higher basal area, abundant cover in the seedling, regenerative growth, sapling, and canopy layers, and high spatial complexity

(Robinson and Robinson 1999, Kenefic and Nyland 2000, Crow et al. 2002, Flaspohler et al.

2002, Jobes et al. 2004, Angers et al. 2005, Vanderwel et al. 2006). Since fledglings appear to benefit from increased cover near the ground, it is reasonable to expect that single-tree selection would not have a significant negative impact on their survival in a predominantly forested landscape.

Here I investigated the influence of single-tree selection on survival probability and habitat use of adult and fledgling Rose-breasted Grosbeaks (Pheucticus leudovicianus) in a 50 predominantly forested landscape. This species has been declining at an annual rate of approximately - 2.4% on its breeding grounds in Ontario between 1966 and 2007 (Sauer et al. 2008), with especially sharp declines (- 8% per year) between 1985 and 2005 (Cadman et al. 2007). The Rose-breasted Grosbeak is a Neotropical migrant that nests in deciduous and mixed forests throughout central and eastern North America (Wyatt and Francis 2002), primarily in the sapling and understory layers in sites with relatively open canopy and sub- canopy layers compared to random locations (Wyatt and Francis 2002). In the highly fragmented deciduous forests of southwestern Ontario, neither nest survival (Smith et al.

2006) nor fledgling survival (Moore et al. 2010) were affected by timber harvesting, but both harvested and un-harvested sites were population sinks (Smith et al. 2006, Moore et al.

2010). Although Rose-breasted Grosbeaks are declining, to my knowledge no studies have investigated the demography of this species in managed stands in predominantly forested landscapes, or investigated whether populations are self-sustaining in these areas.

I took a chronosequence approach to investigate the response of Rose-breasted

Grosbeaks to the structural changes that follow single-tree selection. Specifically, my main objectives were to 1) identify important habitat characteristics at nest sites and locations used by fledglings, and determine whether these differed among post-harvest treatments, 2) estimate rates of nest and fledgling survival, and test whether these differed among post- harvest treatments, and 3) determine which habitat features had the strongest influence on nest success and fledgling survival in harvested stands. I expected habitat features to differ among post-harvest treatments due to forest growth and development, and that as a result vegetation characteristics at nest sites and fledgling locations would also differ among post- harvest treatments. I predicted that nest survival would initially decrease following harvesting, due to a reduction in regenerative growth and cover in the understory and canopy 51 layers, but that it would increase again by 16-20 years post-harvest with the development of a dense understory layer. In contrast, I expected that fledgling survival would increase shortly after harvesting due to an increase in cover near the ground from slash, raspberry bushes, and saplings, but that it would decrease by 16-20 years post-harvest as the forest floor became more open with the development of the understory and canopy layers.

METHODS

Study Species

Rose-breasted Grosbeaks typically arrive on their breeding grounds in central Ontario during the first week of May. In the highly fragmented deciduous forests of southwestern

Ontario, Rose-breasted Grosbeaks nest in locations with more canopy cover but lower sapling cover and basal area than are available at random sites (Smith et al. 2007). In these fragmented forests nest survival is negatively associated with nest concealment, but positively associated with canopy cover, sapling cover, and nest height (Smith et al. 2007).

Both males and females incubate and care for the young, which usually fledge 9-12 days after hatching (Scott 1998). Little is known about the post-fledging period, but young are unable to fly at the time of fledging and remain dependent on the adults for approximately three weeks after leaving the nest (Wyatt and Francis 2002).

Study Area

The study was conducted in twelve tolerant hardwood stands in Algonquin Provincial

Park, Ontario, Canada (45° 34`N, 78°04`W) between 2006 and 2008 (Fig. 1.1). The west side of this 775, 000 ha park is dominated by mixed hardwood forest, whereas the east side consists primarily of white and red pine (Ontario Parks 1998). The stands I studied were 52 located in the central and southern part of the park (mean nearest neighbour distance among stands = 9.3 km, range = 1.0 – 34.2 km). Tolerant hardwood stands in this part of the park have been harvested using the single-tree selection system since the early 1970’s (OMNR

1998). This silvicultural system removes approximately one-third of the over-story basal area every 20-25 years, resulting in uneven-aged stands with a mix of canopy gaps, regenerative growth, and mature trees (OMNR 1998). Dominant tree species in the stands I studied included sugar maple (Acer saccharum), American beech (Fagus grandifolia), eastern hemlock (Tsuga canadensis), and yellow birch (Betula alleghaniensis).

In this study, three temporal stages in the cutting cycle were studied: 0-5, 16-20, and

21-25 years post-harvest. Stands that had not been logged for > 50 years were studied to represent “un-logged” control conditions. Three sites from each temporal stage in the cutting cycle were studied (mean stand size = 34 ha) (Table 1.1).

Nest Searching and Monitoring

Nest searching and monitoring were undertaken from May to August, 2006-2008.

Nests were monitored every 3-4 days using a mirror attached to a telescoping pole to determine clutch size, hatching date, number of nestlings and fledglings, nest outcome, and age of nestlings at the time of fledging. For nests that were too high to visually confirm hatching dates, nestling age was estimated using the observed nest initiation date and clutch size, published lengths for nest building (4 days) and incubation (12 days) (Scott 1998), and the assumption that one egg was laid per day. 53

Radio Attachment and Tracking

For lower nests (≤ 2 m), all nestlings were removed from the nest nine days after hatching (1-2 days prior to fledging) and given a uniquely-numbered aluminium US Fish and

Wildlife Service leg band and a unique combination of three coloured leg bands. All nestlings were weighed using a SC-HT-S-232 digital scale with an accuracy of 0.01 g, and a

Holohil Systems (Carp, Ontario) BD-2 radio-transmitter was attached to the nestling with the greatest mass. The transmitters weighed 0.6 g, (approximately 2.4% of a fledgling’s body weight on day nine post-hatching) which is less than the 5% recommended by Caccamise and

Hedin (1985). Leg harnesses made from thin cotton thread were used to attach the transmitters following the method developed by Rappole and Tipton (1991). Once all the nestlings had been weighed, banded, and a transmitter had been attached they were returned to the nest. For nests that were located too high to reach, fledglings were located and captured on the ground immediately below their nest at the time of fledging, and bands and a radio-transmitter were attached. For these nests, transmitters were attached to the heaviest fledgling captured. Nestlings in lower nests, which were tagged prior to fledging, did not leave their nests at a earlier age than young that were tagged post-fledging (mean age at fledging = 10.1 and 10.3 days respectively), suggesting that radio-tag attachment likely did not cause early fledging.

Once out of the nest, radio-tagged juveniles were located every second day for a period of three weeks (11 possible observations per individual, mean = 7.5 observations per individual) using programmable receivers and three-element Yagi antennas. Juveniles were tracked on the ground, and the exact location of each radio-tagged individual was determined visually, unless the bird was hidden by canopy foliage. The place where the bird was first sighted was marked using a GPS unit and flagging tape, and the height and level of 54 concealment of the bird were recorded. In some instances determining the exact location of a bird took several minutes, but this usually only happened when birds were located in the canopy and not visible from below. I did not observe any instances when the tracker’s movements caused the fledgling to change locations. In cases where the fledgling was first seen flying from one location to another, the spot it was leaving from was marked. Juveniles were considered dead if mortality was visually confirmed (i.e. a whole or partial carcass was found) or were considered to have an unknown fate if their harness was located intact without signs of predation, or if the signal was lost.

Habitat Measurements

Habitat characteristics were measured at nest sites, fledgling locations, and at a corresponding number of random points. Random plots used for comparison with nests were selected using a random number generator to produce a set of UTM co-ordinates located within each stand, and a GPS unit was used to locate those co-ordinates on the ground.

Random surveys were not done in predominantly coniferous patches or swampy areas, but otherwise there were no restrictions on the location of random plots as long as they fell within the study stands.

For each fledgling that was tracked, habitat characteristics were surveyed at three locations, one that was occupied during the first 7 days post-fledging, one that was occupied between days 8 and 14, and one that was occupied between days 15 and 21. For each fledgling, the location to be surveyed from each week-long period was randomly chosen from amongst their known locations. Random points used for comparison with fledgling locations were chosen to represent the type of habitat that would have been encountered if the fledgling had moved the same linear distance since it was last observed, but in a different 55 direction. To determine the locations of random plots, I calculated the linear distance between the fledgling location of interest (i.e., day 9 post-fledging) and its location when last observed (i.e., day 7 post-fledging). I then walked this distance in a straight line from the fledgling’s location when last seen (i.e. day 7 post-fledging) in a direction randomly chosen by blindly spinning my compass. This random point was paired with the fledgling location of interest (i.e. day 9 post-fledging). Due to fledgling mortality, sample sizes were smaller for habitats surveyed during the second and third weeks post-fledging than during the first week.

At fledgling locations, nests, and corresponding random points vegetation layers were measured in 5 m radius circular plots using a modified version of the James and Shugart

(1970) method. Percent cover by: 1) forbs/grasses/ferns, 2) regenerative growth (woody growth 0.5-1.3 m tall), 3) saplings (>1.3 m tall, <2.5 cm dbh), 4) understory (≥2.5 cm dbh,

<10.0 cm dbh), 5) sub-canopy (>10 m tall), and 6) canopy (>20 m tall) were visually estimated, and basal area (m2/ha) was measured using a 2-factor cruising prism. At nest sites, percent concealment and nest height were visually estimated, and fledgling concealment and height were also visually estimated. At fledgling locations and corresponding plots percent cover by small shrubs (< 2.5 cm dbh), large shrubs (2.6-8 cm dbh), wild red raspberries

(Rubus spp), and red elderberry (Sambucus spp) plants were also assessed. These variables were selected for measurement because they have been shown to be correlated with the nest and fledgling locations of songbirds elsewhere (Wyatt and Francis 2002, Smith et al. 2006,

Moore et al. 2010). 56

Statistical Analyses

Habitat Measurements

I used the Spearman rank correlation coefficient to test for significant (P < 0.05) correlations among seven habitat measures, and a suite of relatively uncorrelated variables that were judged to be of ecological importance to Rose-breasted Grosbeaks was retained for further analysis (Burnham and Anderson 2002, Driscoll et al. 2005). Seedling cover was significantly correlated with basal area (-), percent cover by regenerative growth (+), and understory cover (-), and so was omitted from the analysis. Percent cover by forbs was omitted due to significant correlations with regenerative growth (-), and understory (-), whereas sub-canopy and canopy cover were combined into a single measure (hereafter referred to as canopy cover) due to a significant positive correlation. Understory cover was positively correlated with regenerative growth (r = 0.60), but both variables were retained because they were judged to be of importance to Rose-breasted Grosbeaks.

Do habitat features differ between used and un-used plots?

With the exception of canopy cover, vegetation characteristics at nest sites and random plots differed significantly among post-harvest treatments (F12, 135.22 = 2.42, P =

0.007), but not among study years (F8, 102 = 0.87, P = 0.542), and interactions between post- harvest treatment and year were not significant (F30, 202 = 0.54, P = 0.975). Canopy cover differed significantly among study years (F1,3 = 9.58, P = 0.010), so each year was analyzed separately. All other habitat characteristics at nest sites and random plots were pooled across study years by calculating average values for each site in each year, and then taking a weighted average of those values. When data on vegetation characteristics at locations occupied by fledglings was broken down by year and treatment, sample sizes were too small 57

(n < 3 in some cases) to allow meaningful analysis of habitat use. As a result, habitat use data for fledglings were also pooled across study years, by calculating average values for each site in each year, and then taking a weighted average of those two values.

I used weighted MANOVA to test for overall differences in habitat characteristics among post-harvest treatments and between nest and random locations. Habitat variables were then tested individually using split-plot analysis of variance (ANOVA). The residuals for each habitat variable were examined, and as a result I square root transformed canopy cover and log-transformed regenerative growth (log10 + 0.05). Tukey’s Studentized post-hoc range test was performed on the results of split-plot ANOVAs that identified significant among treatment differences.

I used a similar procedure to test for differences between habitat characteristics at points used by fledglings and random locations, but separate comparisons were made for habitat use during the first, second, and third weeks post-fledging. To obtain normality and homoscedasticity I log-transformed regenerative growth (log10 + 0.05), and square-root transformed sapling, understory, and canopy cover.

Do habitat features differ between nest and fledgling plots?

I used weighted MANOVA to test for overall differences in habitat characteristics at nest sites and locations used by fledglings. Habitat variables were then tested individually using split-plot ANOVAs, and Tukey’s Studentized post-hoc range tests were performed on variables that significantly differed among treatments. 58

Nest and Fledgling Survival

Nest and fledgling survival were determined for each post-harvest treatment using the

Mayfield estimator (Mayfield 1961, 1975). Daily survival rate (DSR) for nests was calculated as 1 – no. of failed nests / no. of exposure days, where exposure days were equal to the number of days the nest was under observation. Period survival rate (PSR) was calculated as (DSR)d, where d represented the number of days between the onset of incubation and fledging (24 days for Rose-breasted Grosbeaks). DSR estimates the probability that a nest will survive one day, and PSR estimates the probability that a nest will survive until the young fledge. Similar estimates were calculated for fledglings, but for DSR the number of fledglings that died was substituted for the number of failed nests, and for

PSR, d represented the 21 day observation period for the fledglings.

Are estimates of nest and fledgling survival correlated among harvest treatments?

To test whether single-tree selection harvesting had a similar influence on nest and fledgling survival over the course of the harvesting rotation, I calculated a Spearman rank correlation coefficient for nest and fledgling survival.

How do habitat characteristics influence nest and fledgling survival?

I constructed a set of 42 models containing 1-3 variables to explain nest survival and a set of 75 models containing 1-4 variables to explain fledgling survival (Appendix 2.1).

Each model represented a hypothesis to explain nest and fledgling survival based on literature reviews, significant differences in vegetation between points occupied by nests or fledglings and random locations, and my own observation of the study species. The 59 following variables were included in the set of candidate models explaining nest and fledgling survival:

1) REGENERATIVE GROWTH – In this study, percent cover from regenerative

growth at Rose-breasted Grosbeak nest locations was significantly higher than at

random locations. Minimum nest height was 1.5 m, suggesting that regenerating

vegetation (0.5 – 1.3 m tall) may help conceal nests from potential predators on the

ground. Similarly, this vegetation layer can provide cover for fledglings, especially

during their first week out of the nest when they are learning to fly (King et al. 2006).

Percent cover from regenerative growth was included in models for both nest and

fledgling survival.

2) SAPLING COVER – In forest fragments percent cover by saplings was positively

associated with Rose-breasted Grosbeak nest survival, even though nests were placed

in areas with fewer saplings than were present at random locations (Smith et al.

2007). In my study, Rose-breasted Grosbeaks used nest sites with significantly

higher sapling cover than was present at random locations, suggesting this variable

may be important to survival. I expected saplings to provide cover for fledglings

during the first and second weeks out of the nest when movement is limited and

mortality is high (i.e., Yackel Adams et al. 2006, Moore et al. 2010). Percent cover

by saplings was included in models explaining both nest and fledgling survival.

3) UNDERSTORY COVER – Understory cover was significantly higher at Rose-

breasted Grosbeak nest locations than at random points in my study suggesting that

the understory layer was important for nest placement and camouflage. Structure and

amount of understory vegetation have also been related to fledgling survival (King et 60

al. 2006, Mitchell et al. 2010). I included percent cover by understory in models of

nest and fledgling survival.

4) CANOPY COVER – In forest fragments Rose-breasted Grosbeaks selected nest sites

with lower canopy cover than was present at random locations, even though this

variable was positively associated with nest survival (Smith et al. 2007). In my study,

mean nest height for Rose-breasted Grosbeaks was 9.6 m, suggesting that the sub-

canopy and canopy layers would be important for nest placement, concealment, and

survival. Although canopy cover has not been directly related to fledgling survival to

my knowledge, juveniles were frequently observed feeding in the canopy once they

could fly, so I expected this variable to be important for fledgling survival as well,

and included it in models explaining both nest and fledgling survival.

5) BASAL AREA – Basal area was lower around Rose-breasted Grosbeak nests than at

random locations in forest fragments (Smith et al. 2007) and in my study. Although

the connection between fledgling survival and basal area has not been previously

established to my knowledge, basal area was included in models of both nest and

fledgling survival to determine whether habitat requirements differ before and after

fledging.

6) POST-HARVEST TREATMENT – Neither nest survival (Smith et al. 2006) nor

fledgling survival (Moore et al. 2010) of Rose-breasted Grosbeaks were significantly

influenced by selection harvesting in forest fragments. Harvest treatment was

included in the models for both nest and fledgling survival to test whether Rose-

breasted Grosbeaks responded similarly to selection cutting in a predominantly

forested landscape. Models including each variable, treatment, and an interaction

term were also tested for both nest and fledgling survival. 61

7) CONCEALMENT – For some species, nests that are more camouflaged from

potential nest predators may be more likely to succeed (Martin and Roper 1988,

Filliater et al. 1994, Colombelli-Négrel and Kleindorfer 2009). In forest fragments

Rose-breasted Grosbeak nest survival was positively associated with nest

concealment, even though nests were placed in areas with less than average cover

(Smith et al. 2007). Fledglings of several species use areas with greater concealment

from dense vegetation (Anders et al. 1998, Vega Rivera et al. 1998, Marshall et al.

2003), likely to avoid predation. Percent concealment was included in models of nest

and fledgling survival.

8) HEIGHT – For Rose-breasted Grosbeaks in forest fragments in southern Ontario nest

height is positively associated with nest survival (Smith et al. 2007), and these two

variables are correlated in several other species as well (Burhans et al. 2002, Prokop

2004, Colombelli-Negrel and Kleindorfer 2009). Although I am unaware of any

studies linking fledgling height to survival, I expected that as fledglings learned to fly

and moved up into the canopy their chances of survival would increase. Height was

included in models explaining nest and fledgling survival.

9) AGE OF MALE PARENT – First time breeders of some birds have lower nest

survival than older individuals (Pärt 2001, Forschler and Kalko 2006), as well has

producing smaller clutch sizes (Reynolds 1997) and fewer fledglings (Préault et al.

2005, Brown and Roth 2009). Age of male parent (second-year or after-second-year)

was included in models of nest survival, but since only two of the fledglings I tracked

came from nests with second-year males it was not included in models explaining

fledgling survival. 62

10) DATE – Nests initiated earlier in the breeding season can have higher survival rates

than nests initiated later on (i.e., Spanhove et al. 2009), so nest initiation date was

included in models for nest survival. Date of hatching (Monrós et al. 2002) and date

of fledging (Yackel Adams et al. 2006, Middleton and Green 2008, Moore et al.

2010) can be correlated with fledgling survival, so date of fledging was included in

models explaining fledgling survival.

11) DAYS SINCE FLEDGING – For many songbird species, fledgling mortality is

highest during the first 1-7 days out of the nest, when fledglings are not yet competent

fliers (Sullivan 1989, Anders et al. 1997, Naef-Daenzer et al. 2001, Cohen and

Lindell 2004, Kershner et al. 2004, King et al. 2006, Yackel Adams et al. 2006,

Moore et al. 2010). Time since fledging was included in models for fledgling

survival.

12) YEAR –There is often yearly variation in survival estimates, so year was included in

models for both nest and fledgling survival.

13) TIME SINCE HATCHING – The time between hatching and fledging ranges from 9-

12 days for Rose-breasted Grosbeaks (Wyatt and Francis 2002). If older nestlings are

larger and more fully developed at the time of fledging than younger ones, they could

be expected to have higher post-fledging survival, especially if their feathers and

wing muscles are more fully developed. Time since hatching was included in models

of fledgling survival.

14) WEIGHT AT FLEDGING – The body condition of nestlings at the time of fledging

can influence their ability to find suitable habitats for foraging and predator

avoidance, and nestlings that are in poorer condition may have lower chances of

survival (Magrath 1991, Naef-Daenzer et al. 2001, Vitz and Rodewald 2010). 63

Nestling weight at the time of fledging was included as a measure of body condition

in models explaining fledgling survival.

15) RASPBERRY COVER – Although raspberry bushes did not produce ripe fruit during

the three week period when fledglings were tracked, dense patches of raspberries

could have provided camouflage and protection for fledglings before they were

proficient fliers. Percent cover by raspberries was included in models explaining

fledgling survival.

16) ELDERBERRY COVER – Rose-breasted Grosbeaks have been observed feeding red

elderberries (Sambucus racemosa) to their fledglings (Stutchbury et al. 2005), and

elderberries appeared to be an important food source for fledglings in my study.

Percent cover by elderberry plants was included as a rough estimate of elderberry

abundance in models explaining fledgling survival.

17) SHRUB COVER – Survival of Ovenbird (Seiurus aurocapilla) (King et al. 2006),

Hooded Warbler (Wilsonia citrina) (Rush and Stutchbury 2008), and Rose-breasted

Grosbeak (Moore et al. 2010) fledglings was positively associated with the amount of

shrub cover, likely because small shrubs provide shelter from potential predators.

Percent cover by small (< 2.5 cm dbh) shrubs was included as an explanatory variable

in models of fledgling survival.

Shaffer’s logistic-exposure method (2004) was used to compare the set of models containing variables explaining nest survival. I considered each variable separately (11 models), together with sampling year (10 models), and combined with sampling year and date of nest initiation (9 models). I also included a set of models containing each variable separately, post-harvest treatment, and their interaction term (10 models). A null model 64

(containing the intercept only) and a global model (containing all the variables under consideration) were also included in my set of 42 candidate models containing 1-3 variables explaining nest survival. Shaffer’s logistic-exposure method uses Akaike’s Information

Criteria (corrected for small sample sizes: AICc) (Burnham and Anderson 2002) to compare the relative strength of competing models and was implemented in the PROC GENMOD module of SAS (v. 8.02).

Fledgling survival was analyzed using the nest survival model in Program MARK (v.

6.0) (White and Burnham 1999, Schmidt et al. 2008). For each fledgling only one record was included, which corresponded to the last time that individual had been observed alive or dead. This approach to modeling survival assumes that habitat quality determines survival, and reduces bias resulting from decreasing sample sizes with time. Individual records were coded with the date of fledging, the last day the bird was known to be alive, the last day the fledgling was tracked, the fledgling’s fate (0 = survived, 1 = died), and 16 explanatory variables. I considered each variable separately (16 models), together with sampling year (15 models), combined with sampling year and time since fledging (14 models), and in models containing sampling year, date of fledging, and time since fledging (13 models). I also included models containing each variable separately, post-harvest treatment, and their interaction term (15 models), as well as a null model and global model in the set of 75 candidate models containing 1-4 variables explaining fledgling survival.

RESULTS

Do habitat features differ between used and un-used plots?

Nest Sites. – Habitat characteristics were measured at 94 Rose-breasted Grosbeak nests and at 94 corresponding random plots. Multivariate analysis of variance showed 65

significant differences in vegetation characteristics among post-harvest treatments (F15, 149.47 =

4.54, P < 0.0001) and between nest and random plots (F5, 54 = 7.56, P < 0.0001) but the interaction between treatment and plot type was not significant (F15, 149.47 = 0.83, P = 0.642).

Regenerative growth was significantly higher at Rose-breasted Grosbeak nest sites

(18.7 ± 2.8%) than at random locations (12.8 ± 1%; F1,3 = 6.37, P = 0.014; Fig. 2.1a).

Sapling cover was higher at nest sites (17.7 ± 1.5%) than at random locations (12.8 ± 1.8%), except in the > 50 years post-harvest stands, but differences were not significant (F1,3 = 2.71,

P = 0.105; Fig. 2.1b). Understory cover was also significantly higher at nest sites (41.7 ±

2.9%) than at random locations (26.7 ± 2.0%; F1,3 = 18.60, P < 0.0001; Fig. 2.1c). In contrast, Rose-breasted Grosbeaks used nest sites with significantly lower basal area (21.7 ±

2 2 0.6 m /ha) than was available (25.6 ± 0.7 m /ha; F1,3 = 21.21, P < 0.0001; Fig. 2.1d). Canopy cover was higher at random locations (90.9 ± 5.0%) than at nest sites (70.0 ± 10.2%) in 2006

(F1,3 = 9.92, P = 0.010), but differences were not significant in 2007 (F1,3 = 0.35, P = 0.565) or 2008 (F1,3 = 1.62, P = 0.221).

Nest height was higher on average in the > 50 years post-harvest treatment (mean =

11.08 m) than in the other stands (0-5: 8.60 m, 16-20: 9.19 m, 21-25: 9.89 m), but differences were not significant (F1,3 = 1.31, P = 0.27). Low (0-4 m) nests accounted for 21.1% of the nests in my sample, and the highest number of these nests (16) was found in the 0-5 years post-harvest treatment (Table 2.1). Nests of medium height (4.5-9 m) comprised 33.3% of the nests I located, and occurred in approximately equal proportions in all post-harvest treatments except the > 50 years post-harvest stands, which had a lower proportion (18.8%).

Nests in the sub-canopy (10-20 m) and canopy (> 20 m) layers accounted for 45.5% of the nests I monitored. High nests occurred in approximately equal proportions in all treatments except the 0-5 years post-harvest treatment which had fewer (35.1%). The > 50 years post- 66 harvest treatment had the highest proportion of very high nests (9.4%). In general, in the 0-5 years post-harvest stands 63.1% of the nests were located below 9 m, whereas in the 16-20,

21-25, and > 50 years post-harvest treatments the proportions were 56.8, 46.8, and 46.9% respectively (Table 2.1).

Plots in the 0-5 years post-harvest stands had significantly lower sapling (F1,3 = 5.39,

P = 0.002) and understory (F1,3 = 4.48, P = 0.007) cover and lower basal area (F1,3 = 8.57, P

< 0.001) than plots in the other post-harvest treatments (Fig. 2.1b-d). In contrast, canopy openings created by the removal of trees resulted in significantly higher cover from regenerative growth (F1,3 = 8.60, P < 0.001) in the 0-5 post-harvest stands than in the other post-harvest treatments (Fig. 2.1a). By 16-20 years after harvesting, regenerative growth had developed into a dense understory layer, giving these stands significantly lower regenerative growth (F1,3 = 8.60, P < 0.001) and higher understory (F1,3 = 4.48, P = 0.007) cover than the other post-harvest treatments (Fig. 2.1a, c). Habitat features 21-25 years after harvesting were not significantly different from those in the 16-20 and > 50 years post-harvest stands.

Nest and random plots in the > 50 years post-harvest stands had significantly higher basal area (F1,3 = 8.57, P < 0.001) than those in the 16-20 and 21-25 years post-harvest stands, and regenerative growth (F1,3 = 8.60, P < 0.001) and sapling cover (F1,3 = 5.39, P = 0.002) were also high (Fig. 2.1a,b,d).

Fledgling Locations. – Habitat characteristics were measured at 85 fledgling plots and

85 random locations (n = 44, 22, and 19 in weeks 1, 2, and 3 post-fledging respectively).

Weighted multivariate analysis of variance showed significant differences in vegetation characteristics among post-harvest treatments in week one post-fledging (Week 1: F18, 201.3 =

2.95, P = 0.0001; Week 2: F18, 88.17 = 1.60, P =0.070; Week 3: F15, 61.1 = 1.40, P = 0.175) and between fledgling and random plots in week one post-fledging (Week 1: F6, 71 = 2.09, P = 67

0.050; Week 2: F6, 31 = 0.490, P = 0.808; Week 3: F5, 22 = 1.57, P = 0.209). Interactions between treatment and plot type were not significant in any week post-fledging (Week 1: F18,

201.3 = 0.43, P = 0.981; Week 2: F18, 88.167 = 1.38, P = 0.160; Week 3: F15, 61.13 = 0.59, P = 0.872).

During the first week post-fledging, regenerative growth was significantly higher at locations used by Rose-breasted Grosbeak fledglings (17.4 ± 2.9%) than at random plots (9.2

± 2.8%; F1,3 = 4.78, P = 0.032; Fig. 2.2a). Percent cover by saplings was also significantly higher at fledgling locations (17.4 ± 2.8%) than at random plots (10.7 ± 1.6%; F1,3 = 4.17, P

= 0.045; Fig. 2.2b). During their first week out of the nest, fledglings used locations with more abundant small shrubs (11.3 ± 3.8%) than were available at random locations (3.8 ±

0.9%; F1,3 = 4.88, P = 0.030; Fig. 2.2d). Raspberry and elderberry cover were also more abundant at fledgling locations (8.1 ± 3.7% and 1.3 ± 0.6% respectively) than at random points (0.8 ± 0.4% and 0.1 ± 0.08% respectively) during the first week post-fledging

(raspberry: F1,3 = 4.13, P = 0.045; elderberry: F1,3 = 4.11, P = 0.042; Fig. 2.2e,f).

During the second and third weeks out of the nest, habitat use by fledglings was highly variable and most vegetation measurements did not significantly differ between used and un-used plots. However, during both the second and third weeks out of the nest fledglings used locations with significantly lower basal area (22.5 ± 1.6 m2/ha and 21.2 ± 1.6 m2/ha respectively) than was present at random locations (23.9 ± 1.3 m2/ha and 26.2 ± 1.7

2 m /ha respectively; F1,3 = 2.86, P = 0.020 and F1,3 = 3.65, P = 0.014 respectively).

Fledgling and random plots in the 0-5 years post-harvest stands had significantly lower sapling cover (F1,3 = 3.84, P = 0.013) and basal area (F1,3 = 8.47, P < 0.0001) than other post-harvest treatments, and significantly higher cover from small shrubs (F1,3 = 4.65, P

= 0.005) and raspberry bushes (F1,3 = 6.38, P = 0.0007) than were present in all other stands 68

35 ) A) % (

30 r e

v 25 o c 20 e v i t 15 a r

e 10 n e

g 5 e

R 0 Nest Random B) 35 Plot type

) 30 % (

r 25 e v 20 o c

g 15 n i l

p 10 a

S 5 0 Nest Random C) 60 Plot type ) %

( 50

r 0-5 e

v 40 o c

16-20

y 30 r o t 21-25 s 20 r e

d 50

n 10 U 0 Nest Random D) 35 Plot type )

a 30 h

/

q 25 s

m 20

(

a 15 e r a

l 10 a s

a 5 B 0 Nest Random Plot type

Figure 2.1. (A) Regeneration cover, (B) sapling cover, (C) understory cover, (D) basal area at nest and random plots in 12 hardwood stands in Algonquin Provincial Park harvested by single-tree selection 0-5, 16-20, 21-25, and > 50 years previously. Means and standard errors are shown. 69

TABLE 2.1. Number of Rose-breasted Grosbeak nests located in four height categories (low, medium, high, and very high) in 12 stands harvested by single-tree selection 0-5, 16-20, 21- 25, and > 50 years prior to sampling. Number of nests in each category is shown with row percentages in brackets. Row totals represent the number of nests in each post-harvest treatment and column totals represent the number of nests in each height category. Nest height (m) Time since Low Medium High Very High Row Total harvest (0-4.0 m) (4.5 – 9 m) (10-20 m) (> 20 m) 0-5 16 (28.1) 20 (35.1) 20 (35.1) 1 (1.8) 57 (100.0) 16-20 5 (11.4) 20 (45.5) 19 (43.2) 0 (0.0) 44 (100.0) 21-25 8 (17.0) 14 (29.8) 23 (48.9) 2 (4.3) 47 (100.0) > 50 9 (28.1) 6 (18.8) 14 (43.8) 3 (9.4) 32 (100.0) Column Total 38 (21.1) 60 (33.3) 76 (42.2) 6 (3.3) 180 (100.0) 70

A) D) )

) 50

% 50 % (

(

r

r 40 e

40 e v v o o c 30

c

30 e b v i u t

r 20

a 20 h r s e

l l n 10 10 a e g m e

0 S 0 R Random 1-7 days since 8-14 15-21 Random 1-7 days post- 8-14 15-21 fledging B) fledging E) Time since fledging (days) Time since fledging (days) 60 50 0-5

50 ) %

) 40 16-20 (

r % ( e

40 r v 30 21-25 e o v c

30 o y c r 20

r > 50 g

20 e n i b l p p

s 10 a 10 a S R 0 0 Random 1-7 days since 8-14 15-21 Random 1-7 8-14 15-21 fledging Time since fledging (days) C) Time since fledging (days) F)

35 10 ) ) % a

30 (

h

r 8 /

25 e v m ( o 6 20 c a

e y r r r a 15 4

e l b a

10 r s e a 2 d l B 5 E 0 0 Random 1-7 8-14 15-21 Random 1-7 8-14 15-21 Time since fledging (days) Time since fledging (days)

Figure 2.2. (A) Regenerative cover, (B) sapling cover, (C) basal area, (D) small shrub (< 2.5 cm dbh), (E) raspberry cover, and (F) elderberry cover at locations occupied by fledglings and random plots in 12 hardwood stands in Algonquin Provincial Park harvested by single-tree selection 0-5, 16-20, 21-25 and > 50 years previously. Means and standard errors are shown. 71

(Fig. 2.2b-e). Habitat characteristics in the 16-20 and 21-25 years post-harvest stands were generally intermediate between the 0-5 and > 50 years post-harvest stands, but sapling cover was significantly higher (F1,3 = 3.84, P = 0.013) in the 16-20 years post-harvest stands than in the 0-5 years post-harvest treatment (Fig. 2.2b). The > 50 years post-harvest treatment had higher sapling (F1,3 = 3.84, P = 0.013) and understory cover (F1,3 = 3.18, P = 0.029) and higher basal area (F1,3 = 8.47, P < 0.0001) than other stands.

Do habitat features differ between nest and fledgling plots?

Weighted multivariate analysis of variance indicated that habitat characteristics at nest sites and fledgling locations differed significantly among post-harvest treatments (F15,

367.56 = 1.91, P < 0.021), and between nest and fledgling plots (F15, 367.56 = 3.06, P = 0.0001).

Interactions between treatment and plot type were not significant (F45,598.04 = 1.02, P = 0.445).

Individual split-plot ANOVAs indicated that nest plots had significantly higher percent cover from understory than was present at locations used by fledglings during their first (F3,3 = 4.89, P = 0.003) and second (F3,3 = 3.89, P = 0.005) weeks out of the nest (Fig.

2.3b). Nest plots also had significantly lower basal area than was present at locations used by fledglings during their first week out of the nest (F3,3 = 9.18, P < 0.0001) (Fig. 2.3c).

Nest and fledgling plots in 0-5 years post-harvest stands had significantly higher percent cover from regenerative growth than similar plots in the 16-20 and 21-25 years post- harvest stands (F3,3 = 4.90, P = 0.003) (Fig. 2.3a). Nest and fledgling plots in 0-5 years post- harvest stands also had significantly lower basal area than similar plots in 21-25 and > 50 years post-harvest stands (F3,3 = 9.18, P < 0.0001) (Fig. 2.3c). 72

A)

50 ) %

( 40

r e v

o 30 c

n o i t

a 20 r e n e

g 10 e R 0 Nest 1-7 days 8-14 days 15-21 days B) Plot Type 70 0-5 60 ) 16-20 % (

r 50 21-25 e v

o 40 > 50 c

y r 30 o t s r

e 20 d n

U 10 0 Nest 1-7 days 8-14 days 15-21 days C)

40 Plot Type ) a h / 30 q s

m (

a 20 e r a

l a

s 10 a B

0 Nest 1-7 days 8-14 days 15-21 days Plot Type

FIGURE 2.3. (A) Regeneration cover, (B) understory cover, and (C) basal area at nest sites and locations occupied by fledglings 1-7, 8-14, and 15-21 days post-fledging in 12 hardwood stands harvested by single-tree selection 0-5, 16-20, 21-25, and > 50 years previously. Means and standard errors are shown. 73

Nest Success and Fledgling Survival

Nest Success. – A total of 173 Rose-breasted Grosbeak nests were monitored over three years (20, 73, and 80 nests respectively in the years 2006-2008), 97 (56.0%) of which were successful. Of the 73 nests that failed, 32 (43.8%) failed during the nestling stage, 22

(30.1%) failed during the egg stage, and 19 (26%) failed either very late in the egg stage or very early in the nestling stage. I also observed 25 nest building attempts that were abandoned before eggs were laid. During both the egg and nestling stages, the primary cause of nest failure was predation (68 of the 73 nests), presumably by small mammals or avian nest predators. The secondary cause of nest failure was nest abandonment during the egg stage (5 of the 73 nests), for unknown reasons.

Fledgling Survival. – Over three years 46 fledglings were radio-tagged of which 18

(38%) survived the three week-tracking period, 17 (40%) were confirmed dead, and 11

(23%) were lost during the tracking period, presumably due to malfunctioning radio-tags (n =

10) or harness failure (n = 1). The 10 cases where radio-tag failure was assumed all occurred during a two week period in 2008 when signals were lost for ten individuals that were 3-6 days post-fledging and would not have been capable of flying large distances. In all years, fledgling mortality was highest during the first week (1-7 days) post-fledging, with 12 confirmed deaths (67%) occurring during this interval. Fledgling survival to independence was similar between 2006 (67%; n = 3, 1 confirmed death) and 2007 (65%; n = 23, 8 confirmed deaths total). Unfortunately, in 2008 only 2 fledglings were tracked for the entire

21 day period (10 were lost as a result of malfunctioning radio-tags, 9 were confirmed dead).

The primary causes of fledgling mortality appeared to be predation by hawks and mammals. Four transmitters were recovered amid piles of feathers on top of elevated 74 perches, presumably having been left by raptors (Moore et al. 2010). One transmitter was found on the ground directly below an active Red-tailed Hawk’s (Buteo jamaicensis) nest, and the transmitter from another fledgling ended up in a Broad-winged Hawk’s (Buteo platypterus) nest. One transmitter was found in the scat of a mammal, and the remains of 5 fledglings were found buried approximately 6 cm below the leaf litter, having been partially consumed presumably by a mammal. The five remaining transmitters showed signs of predation (partial remains of fledglings, or heavily chewed transmitter harnesses), but predator identity was unclear.

Are estimates of nest and fledgling survival correlated among post-harvest treatments?

Mean Mayfield daily survival rate (DSR) for Rose-breasted Grosbeak nests during the egg and nestling stages was 0.967 (n = 173 nests) and the mean period survival rate

(PSR) for the entire 24 day nesting period was 0.559. The model containing post-harvest treatment did not have any support as an explanatory variable for nest survival (AICc weight

= 0.00), indicating that single-tree selection does not strongly influence nest success. When nest survival estimates for each post-harvest treatment were combined across study years, nest survival was highest in the 21-25 years post-harvest stands and lowest in the > 50 years post-harvest treatment (Fig. 2.4).

Mean daily survival rate for Rose-breasted Grosbeak fledglings was 0.920 (n = 47), and the probability of a fledgling surviving the entire 21 day tracking period was 0.510. The model containing post-harvest treatment had little support as an explanatory variable for fledgling survival (AICc weight = 0.01), suggesting that fledgling survival is not strongly influenced by single-tree selection harvesting. Like nest survival, fledgling survival was 75

n = 34, 8 0.9 n = 55, 8 n = 38, 16 0.8 l

a n = 29, 8

v 0.7 i v r

u 0.6 s

d o

i 0.5 Nest r e p

0.4 Fledgling d l e

i 0.3 f y a 0.2 M 0.1 0 0-5 16-20 21-25 >50 Time since harvest (years)

FIGURE 2.4. Mayfield estimates of the probability of Rose-breasted Grosbeak nests surviving to fledge young and fledglings surviving their first three weeks out of the nest in 12 stands in Algonquin Provincial Park harvested by single-tree selection 0-5, 16-20, 21-25, and > 50 years previously. Treatment means calculated from weighted site means and standard error are shown. Samples sizes (n) are shown for the number of nests and fledglings respectively that were monitored in each treatment during 2007 and 2008. 76 lowest in the > 50 years post-harvest stands, but it was highest in the 16-20 years post- harvest treatment (Fig. 2.4).

The Spearman rank correlation coefficient indicated that nest and fledgling survival were not significantly correlated in harvested stands (r = 0.28, P > 0.05; n = 12).

How do habitat characteristics influence nest and fledgling survival?

Nest Survival. – The model with the best support for explaining variation in daily nest survival included nest initiation date, sampling year, and basal area (Table 2.2). The top nine of forty-two models tested included nest initiation date and sampling year, with weights ranging from 0.313-0.001. The model averaged estimate (regression coefficient averaged across models in the candidate set) for basal area was positive, indicating that nests were more likely to survive in areas with greater basal area (Table 2.2). Nest initiation date was negatively associated with nest survival, indicating that earlier nests had a higher probability of survival.

Fledgling Survival. – The model with the strongest support for explaining variation in fledgling survival included number of days since fledging (Table 2.3). The model containing days since fledging and year also had some support, and these two variables appeared in the top nineteen of seventy-five models, with weights ranging from 0.280-0.020. The model averaged estimate for days since fledging was positive, suggesting that older fledglings had a higher probability of continued survival. 77

TABLE 2.2. Support for models predicting daily nest survival rates of Rose-breasted Grosbeaks in Algonquin Provincial Park (2006- 2008) as a function of percent cover of regenerative growth, saplings, and understory, basal area (m2/ha), time-since harvest (0-5, 16- 20, 21-25, or >50 years), nest concealment (%), nest height (m) and nest initiation date. Shown are model averaged parameter estimates, unconditional standard error, number of model parameters (k), deviance, Akaike’s Information Criteria corrected for small sample sizes (AICc), Δi (AICc modeli – AICc model minimum), and Akaike model weights for models that fall within 2 AICc units of the best fit model. Statistics for the global model (including all variables) and the null model (intercept only) are also shown.

Model Predictors Model Unconditional K Deviance AICc Δi Wi Averaged SE Estimate 1 Basal area 0.0088 0.0140 5 526.334 536.355 0.000 0.313 Date -0.1056 0.0135 Year (2006) 1.3371 0.5875 (2007) -0.3959 0.2428 (2008) 0.0000 0.0000 2 Date -0.1056 0.0135 4 529.186 537.200 0.845 0.205 Year (2006) 1.3371 0.5875 (2007) -0.3959 0.2428 (2008) 0.0000 0.0000 9 Global - - 13 520. 905 547.030 10.675 0.001 10 Null (intercept only) - - 1 600.223 602.224 65.870 0.000 78

TABLE 2.3. Support for models predicting daily survival rates of Rose-breasted Grosbeak fledglings in Algonquin Provincial Park 1- 21 days after fledging (2006-2008). Predictor variables include percent cover of regenerative growth, saplings, understory, canopy, small shrubs, raspberries, and elderberries, basal area (m2/ha), time since harvest, concealment (%), height (m), weight (g), fledging date, and days since fledging, days since hatching, and sampling year. Shown are model averaged parameter estimates, unconditional standard error, number of model parameters (k), deviance, Akaike’s Information Criteria corrected for small sample sizes (AICc), Δi (AICc modeli – AICc model minimum) and Akaike model weights for models that fall within 2 AICc units of the best fit model. Statistics for the global model (including all variables), the null model (intercept only), and the model with the most support containing post-harvesting treatment are also shown.

Model Predictors Model Averaged Unconditional K Deviance AICc Δi Wi Estimate SE 1 Days since fledging 0.555 0.286 2 22.779 24.789 0.000 0.280 2 Days since fledging 0.555 0.286 3 22.288 26.318 1.529 0.130 Year -0.003 0.003 7 Post-harvest treatment -0.677 0.566 5 20.124 28.225 3.435 0.050 Year -0.003 0.003 Fledging date 0.129 0.192 Days since fledging 0.555 0.286 20 Global - - 17 38.545 39.933 5.749 0.011 22 Null (intercept only) - - 1 39.339 41.346 16.557 0.000 79

DISCUSSION

Habitat characteristics at Rose-breasted Grosbeak nest sites and locations used by fledglings differed significantly among stands harvested by single-tree selection 0-5, 16-20,

21-25, and > 50 years post-harvest. As expected, nest and fledgling plots in the most recently harvested stands had lower basal area, lower sapling, understory and canopy cover, and more shrubs, raspberry, and elderberry bushes than plots in other post-harvest treatments.

In general, nest and fledgling locations in the 16-20 and 21-25 years post-harvest treatments had lower cover from regenerative growth and higher understory cover than plots in other treatments. In the > 50 years post-harvest treatment basal area was higher than in other stands, and cover from regenerative growth, saplings, and canopy were also high. Despite these differences, which are generally consistent with structural changes reported in other long-term studies of single-tree selection (Crow et al. 2002, Kenefic and Nyland 2000), neither nest success nor fledgling survival significantly differed among post-harvest treatments. Rose-breasted Grosbeaks also exhibited plasticity in nest site selection without showing differences in survival rates in managed forest fragments in southern Ontario (Smith et al. 2007), suggesting that single-tree selection harvesting likely does not have a significant negative effect on this species.

Previous research has suggested that single-tree selection may provide suitable habitat for many forest songbirds, and may have only limited effects on their reproductive success

(Barber et al. 2001, Duguay et al. 2001, Lang et al. 2002, Smith et al. 2006, Poulin et al.

2010). This appears true for Rose-breasted Grosbeaks as well, who showed no significant differences in nest success (Smith et al. 2006, 2007, Moore et al. 2010) or fledgling survival

(Moore et al. 2010) in managed stands in the highly fragmented agricultural landscape of 80 southern Ontario. Although single-tree selection harvesting does not appear to negatively influence Rose-breasted Grosbeak survival in either fragmented or predominantly forested landscapes, there were differences in survival rates among the two landscape types.

Estimates of nest survival, which ranged from 0.341 (Moore et al. 2010) to 0.390 (Smith et al. 2006, 2007) in the fragmented landscape, were considerably lower than my estimate

(0.559) in the predominantly forested landscape. In contrast, fledgling survival was slightly higher (0.620) in the fragmented landscape (Moore et al. 2010) than in the predominantly forested one (0.510). The net effect of these measures suggests that overall survival is lower in the fragmented landscape, likely due to higher levels of nest predation. The contrasting patterns of nest and fledgling survival indicate either that Rose-breasted Grosbeaks respond differently to selective pressures during the nest and fledgling stages, or that nest and fledgling survival are most strongly influenced by different variables.

Estimates of nest and fledgling survival in my study were not significantly correlated among post-harvest treatments, supporting the idea that Rose-breasted Grosbeak responses to environmental factors differ between reproductive phases. Seventy percent of the fledgling deaths I observed were attributed to predation by hawks and small mammals, and 67% of these deaths occurred in the 21-25 and > 50 years post-harvest sites. Raptors and chipmunks have previously been identified as common predators of Wood Thrush (Schmidt et al. 2008) and Rose-breasted Grosbeak fledglings (Moore et al. 2010). One study found that partial harvesting in hardwood stands had relatively little impact on the density of eastern chipmunks (Holloway and Malcolm 2006), but this species is positively associated with many features present in un-logged stands, such as the number of declining trees, the density of understory cover, and the amount of downed woody debris (Bennet et al. 1994, Mahan and 81

Yahner 1996, Zollner and Crane 2003, Holloway and Malcolm 2006). Mature forest may also provide more open or better quality habitat for forest-breeding birds of prey such as

Broad-winged Hawks, which nest in tall, large diameter trees and typically remain below a well developed canopy layer when hunting for insects, amphibians, small mammals, and birds (Goodrich et al. 1996). Higher predation by raptors and chipmunks on fledglings, but not necessarily nests, in less recently harvested stands may help explain differences in nest and fledgling survival rates among post-harvest treatments.

Significant differences also existed between the habitat characteristics present at nest sites and fledgling locations, which may help explain differences in survival patterns.

Specifically, nest plots had significantly lower basal area and higher percent cover from understory than were present at locations used by fledglings during their first and second weeks out of the nest. There was substantial variation associated with the habitat measurements at locations occupied by fledglings, likely because they moved around in search of food and shelter. The high degree of movement and variation in habitat use made it difficult to identify important habitat features for fledglings during their second and third weeks out of the nest. The fledglings of some species move into habitat types that contrast significantly with those used by nesting adults (Lang et al. 2002, White et al. 2005), whereas in other species habitat use during the nesting and post-fledging periods is similar (Berkeley et al. 2007, Ciudad et al. 2009). Rose-breasted Grosbeak fledglings appeared to use similar habitats to nesting adults with only a few significant differences, but this could partially be the result of the relative similarity of habitat types available in the predominantly forested study landscape. 82

Rose-breasted Grosbeaks tended to use nest sites with lower basal area and higher cover from regenerative growth, saplings, and understory than was available at random locations in all post-harvest treatments. In the fragmented woodlots of southern Ontario

Rose-breasted Grosbeaks also used nest sites with lower basal area and higher sapling cover than was available at random points (Smith et al. 2007). Although grosbeaks consistently build nests in locations with lower basal area, nest survival was positively associated with this habitat feature in my study. Previous research has shown that although Rose-breasted

Grosbeak nest survival was positively associated with canopy cover, adults used nest sites with lower canopy cover than was present at random locations (Smith et al. 2007). Together these studies suggest that Rose-breasted Grosbeaks do not necessarily select nest sites with habitat features that maximize reproductive success. Although there is no obvious explanation for this, Smith et al. (2007) have suggested this species may be experiencing an ecological trap. Locations with lower basal area and canopy cover may appear attractive because they are typically associated with other desirable features, such as denser understory and sapling layers, which may not be present in managed forests during all stages in the harvest rotation.

The best models for explaining nest success included basal area, nest initiation date, and sampling year. Nests that were started earlier in the season and built in locations with higher basal area were more likely to be successful. Higher survival rates for nests initiated earlier in the season have been reported for several other avian species (e.g. Middleton and

Green 2008, Spanhove et al. 2009). Older, more experienced birds often begin breeding earlier than less experienced adults, and higher rates of nest and juvenile survival earlier in the season are often attributed to age related differences in the timing of nest initiation 83

(Hatchwell 1991, Middleton and Green 2008). Older birds may have higher reproductive success because they are more skilful than younger birds at selecting high quality nest sites, foraging, defending their offspring, and working together in pairs (Forslund and Part 1995,

Martin 1995, Espie et al. 2000, Balbontin et al. 2007). In the fragmented forests of southern

Ontario Rose-breasted Grosbeak nest survival was best explained by a negative association with nest concealment, and positive relationships with nest height and canopy and sapling cover (Smith et al. 2007). Differences in the composition and foraging tactics of predator communities in the two landscapes may account for the importance of different habitat features in explaining nest survival in the two landscapes.

Nest survival was also positively associated with basal area, which was significantly higher in the > 50 years post-harvest treatment than in the more recently harvested stands.

However, despite the positive association between nest survival and basal area, mean nest success was slightly lower in the > 50 years post-harvest stands than in the other post-harvest treatments, suggesting that additional factors or more complex relationships among habitat measures may influence nest survival. For example, in the 16-20 and 21-25 years post- harvest treatments, which had the highest mean nest survival rates, nest sites had relatively high understory cover as well as high basal area. In comparison, the > 50 years post-harvest treatment had the highest basal area but relatively low understory cover, suggesting understory cover may have been important as well. Although the model containing understory cover did not fall within 2 units of the best fit model explaining nest survival, it was ranked third and had an AICc weight of 0.11. Alternatively, higher basal area may be beneficial up to a certain point, after which nest survival begins to decrease. Although the exact nature of the relationship between basal area and nest survival may still be uncertain, 84 my study suggests that if basal area is further reduced in hardwood stands following a second selection harvest, the chances of nest survival may also decrease.

In contrast to nesting adults, during their first week out of the nest, fledgling Rose- breasted Grosbeaks used locations with higher cover from regenerative growth, saplings, small shrubs, raspberry and elderberry bushes. A similar pattern of habitat use was observed in Wood Thrush (Hylocichla mustelina) fledglings, which moved into areas with more dense forest cover relative to the open, mature forest used by adults for nest building (Anders et al.

1998, Vega Rivera et al. 1998). Ovenbird fledglings (Seiurus aurocapilla) also occupied locations with greater vertical structure 0-3 m above ground than was present at nest or random plots (King et al. 2006). Due to fledgling movement the habitat characteristics observed at Rose-breasted Grosbeak fledgling locations were highly variable, but this study suggests that in predominantly forested landscapes Rose-breasted Grosbeak fledglings use areas with higher than average cover.

The fledglings of many mature-forest bird species have been observed in regenerating clear-cuts during the post-fledging period (Pagen et al. 2000, Vitz and Rodewald 2006), possibly because these areas provide increased cover for thermoregulation, protection from predators (Weathers and Sullivan 1989, Morton et al. 1991), and abundant food sources

(Anders et al. 1998, Vega Rivera et al. 1998, Vitz and Rodewald 2007). In this study, during the first week after leaving their nests Rose-breasted Grosbeak fledglings used sites with higher cover from raspberry bushes than was present at random locations, and although differences were not significant, fledglings also selected areas with higher than average cover from elderberry bushes during all three weeks post-fledging. Rose-breasted Grosbeaks have been videotaped feeding on red elderberries in the wild (Stutchbury et al. 2005), and both 85 adults and fledglings were frequently observed feeding extensively on both elderberries and raspberries in my study area. Availability of food resources has been identified as an important determinant of fledgling movement patterns during the post-fledging period for several forest-breeding songbird species (Stutchbury et al. 2005, Vitz and Rodewald 2007,

Ciudad et al. 2009). Movement patterns of Rose-breasted Grosbeak families during the post- breeding season may be partially determined by the availability of berry bushes, and since recently harvested stands had higher cover from raspberries, this treatment may represent preferred habitat during this period.

The best model to explain fledgling survival included the number of days since fledging, indicating that the probability of survival increased with fledgling age. In forest fragments, Rose-breasted Grosbeak fledgling survival was best explained by positive associations with date of fledging, patch size, and percent shrub cover (Moore et al. 2010).

Many studies of forest songbirds have reported higher fledgling mortality during the first 1-7 days out of the nest (Sullivan 1989, Anders et al. 1997, Naef-Daenzer et al. 2001, Cohen and

Lindell 2004, Kershner et al. 2004, King et al. 2006, Yackel Adams et al. 2006, Moore et al.

2010) when fledglings are unable to fly and capable of only limited movement. Other factors which have been commonly associated with fledgling survival prior to independence include habitat type (Krementz et al. 1989, Anders et al. 1997, Vega Rivera et al. 1998), amount of coarse woody debris, amount and structure of understory cover (Mitchell et al. 2010), age of parents (Rush and Stutchbury 2008), hatching date and fledgling mass (Monrós et al. 2002), date of fledging, and weather (Yackel Adams 2006). Although age at fledging is variable for

Rose-breasted Grosbeaks (9-12 days post-hatching), and the model containing days since hatching was ranked as the third best-fit model explaining fledgling survival, the model 86 containing days since fledging had more support. The results of my study are consistent with the growing body of evidence that indicates that fledgling survival increases considerably with age, and it indicates that for Rose-breasted Grosbeaks the relationship between survival and age may be more strongly influenced by days since fledging than the length of time spent in the nest.

Overall, single-tree selection harvesting did not appear to have a significant negative effect on Rose-breasted Grosbeak nest or fledgling survival. However, since fewer nests and fledglings were monitored in the > 50 years post-harvest stands, sample sizes were smaller in this treatment than in the others. Differences in samples sizes were addressed using weighted analyses, but it is possible they influenced among-treatment comparisons. Incorporating a larger number of unharvested stands into future studies so as to increase sample sizes in this treatment may help provide insight into the relationship between basal area and nest survival.

In addition, a relatively large set of candidate models was ranked to explain the survival rates of a relatively small number of fledglings. Increasing the number of fledglings tracked might be one way for future studies to increase the power of the tests and confirm that single-tree selection harvesting does not negatively influence nest or fledgling survival.

Conclusions

Despite significant differences in habitat features at nest sites and fledgling locations in stands harvested 0-5, 16-20, 21-25, and > 50 years previously, single-tree selection did not appear to negatively influence Rose-breasted Grosbeak nest or fledgling survival.

Differences in habitat characteristics between nest sites and locations used by fledglings indicate that creating a mosaic of forest stands in different stages of the harvest rotation 87 might provide Rose-breasted Grosbeaks with access to better quality habitats during all stages in the breeding cycle. Un-even aged silvicultural techniques applied over relatively small areas likely provide higher quality habitat overall for Rose-breasted Grosbeaks than even-aged practices applied over larger areas, because they provide easier access to habitats in different stages of regeneration. Differences in nest and fledgling survival among post- harvest treatments also suggest that future studies attempting to assess the impact of forest management techniques or other habitat alterations on avian populations should consider both nest and fledgling requirements. APPENDIX 2.1. List of candidate models used to explain Rose-breasted Grosbeak nest and fledgling survival in twelve tolerant hardwood stands in Algonquin Provincial Park harvested by single-tree selection 0-5, 16-20, 21-25, and > 50 years previously. Model Predictor variables included Plots for which no. model was used Models with one variable 1 Regenerative growth Nest and fledgling 2 Sapling cover Nest and fledgling 3 Understory cover Nest and fledgling 4 Canopy cover Nest and fledgling 5 Basal area Nest and fledgling 6 Post-harvest treatment Nest and fledgling 7 Concealment Nest and fledgling 8 Height (nest, fledgling) Nest and fledgling 9 Date (nest initiation, fledging) Nest and fledgling 10 Year Nest and fledgling 11 Age of male parent Nest 12 Small shrub cover Fledgling 13 Raspberry cover Fledgling 14 Elderberry cover Fledgling 15 Days since fledging Fledgling 16 Fledgling weight Fledgling 17 Time since hatching Fledgling Models with two variables 18 Regenerative growth + year Nest and fledgling 19 Sapling cover + year Nest and fledgling 20 Understory cover + year Nest and fledgling 21 Canopy cover + year Nest and fledgling 22 Basal area + year Nest and fledgling 23 Post-harvest treatment + year Nest and fledgling 24 Concealment + year Nest and fledgling 25 Height + year Nest and fledgling 26 Date + year Nest and fledgling 27 Age of male parent + year Nest 28 Small shrub cover + year Fledgling 29 Raspberry cover + year Fledgling 30 Elderberry cover + year Fledgling 31 Days since fledging + year Fledgling 32 Fledgling weight + year Fledgling 33 Age at fledging + year Fledgling Models with three variables 34 Regenerative growth + year + date Nest and fledgling 35 Sapling cover year + year + date Nest and fledgling 36 Understory cover + year + date Nest and fledgling 37 Canopy cover + year + date Nest and fledgling 38 Basal area + year + date Nest and fledgling 39 Post-harvest treatment + year + date Nest and fledgling

88 89

APPENDIX 2.1 Cont’d Model Predictor variables included Plots for which no. model was used 40 Concealment + year + date Nest and fledgling 41 Height + year + date Nest and fledgling 42 Age of male parent + year + date Nest 43 Small shrub cover + year Fledgling 44 Raspberry cover + year + date Fledgling 45 Elderberry cover + year + date Fledgling 46 Days since fledging + year + date Fledgling 47 Fledgling weight + year + date Fledgling 48 Age at fledging + year + date Fledgling Models with four variables 49 Regenerative growth + year + date + dsfledging Fledgling 50 Sapling cover year + year + date + dsfledging Fledgling 51 Understory cover + year + date + dsfledging Fledgling 52 Canopy cover + year + date + dsfledging Fledgling 53 Basal area + year + date + dsfledging Fledgling 54 Post-harvest treatment + year + date + dsfledging Fledgling 55 Concealment + year + date + dsfledging Fledgling 56 Height + year + date + dsfledging Fledgling 57 Small shrub cover + year + dsfledging Fledgling 58 Raspberry cover + year + date + dsfledging Fledgling 59 Elderberry cover + year + date + dsfledging Fledgling 60 Fledgling weight + year + date + dsfledging Fledgling 61 Age at fledging + year + date + dsfledging Fledgling Models with an interaction term 62 Regenerative growth, treatment, treatment * regen Nest and fledgling 63 Sapling cover, treatment, treatment * saplings Nest and fledgling 64 Understory cover, treatment, treatment * understory Nest and fledgling 65 Canopy cover, treatment, treatment * canopy Nest and fledgling 66 Basal area, treatment, treatment * basal area Nest and fledgling 67 Concealment, treatment, treatment * concealment Nest and fledgling 68 Height, treatment, treatment * height Nest and fledgling 69 Date, treatment, treatment * date Nest and fledgling 70 Year, treatment, treatment * year Nest and fledgling 71 Age of male, treatment, treatment * age of male parent Nest 72 Small shrubs, treatment, treatment * small shrubs Fledgling 73 Raspberry cover, treatment, treatment * raspberries Fledgling 74 Elderberry cover, treatment, treatment * elderberries Fledgling 75 Days since fledging, treatment, treatment* dsfledging Fledgling 76 Fledgling weight, treatment, treatment * fledgling weight Fledgling 77 Age at fledging, treatment, treatment * age Fledgling CHAPTER 3: NESTLING PROVISIONING RATES AND FORAGING PATTERNS OF ROSE-BREASTED GROSBEAKS IN STANDS HARVESTED BY SINGLE TREE SELECTION

ABSTRACT

Prey availability in hardwood forests influences many aspects of songbird reproduction, but few studies have investigated how timber harvesting in these forests influences the ability of breeding birds to provide food for their offspring. I conducted one hour feeding watches at Rose-breasted Grosbeak (Pheucticus leudovicianus) nests in twelve hardwood stands in Algonquin Provincial Park which had been harvested using single-tree selection 0-5, 16-20, 21-25, and > 50 years previously. My main objectives were to test whether nestling provisioning rates (# trips/nestling/hr), male contribution to nestling provisioning, nest attendance, male contribution to nest attendance, prey load sizes, and nestling weights at the time of fledging differed with time since harvest, stage of nestling development (< 5 or > 5 days old), or age of male parent (SY or ASY). None of the response variables differed significantly among harvesting treatments in either study year (P > 0.08), but they differed with stage of nestling development in 2007 (P =

0.006), and with age of male parent in both years (P < 0.04). Specifically, in 2007 provisioning was less frequent (P < 0.01) for < 5 day old nestlings than for > 5 day old nestlings. Nest attendance was higher (P < 0.01) for < 5 day old nestlings than for > 5 day old nestlings, except in the > 50 years post-harvest treatment. In 2008, nestling provisioning was higher at nests attended by ASY males than at those attended by SY males (P = 0.002), and ASY males contributed more to nestling provisioning (P = 0.02) and to nest attendance (P < 0.01) than SY males did. Mean nestling weight was higher at nests attended by ASY males than SY males in both study years, with the difference

90 91 significant (P = 0.04) in 2008, suggesting that younger breeders could not compensate for lower provisioning rates with larger prey load sizes. This study suggests that the age of the male parent is more important in determining nestling provisioning rates than time since single-tree selection harvesting. 92

INTRODUCTION

During the nesting season, most songbirds can be classified as ‘central place foragers’ because they carry prey back to centrally located nests (Orians and Pearson

1979). According to this model, adults forage close to their nests until available resources become depleted and then exploit food sources further away (Naef-Daenzer

2000). To maximize nestling provisioning, there should be a positive correlation between the distance travelled while foraging and the prey load size delivered to the nestlings

(Grieco 2002). Predator avoidance (e.g., Newman and Caraco 1988) and conspecific competition (Giraldeau et al. 1994) can influence the relationship between time spent foraging and prey load size (Sejberg et al. 2000). Assessments of the relationship between foraging rate and prey load size can be used to test whether changes in forest structure resulting from silvicultural practices influence invertebrate abundance, prey availability, or the foraging patterns of breeding songbirds.

Previous studies in hardwood forests have shown that selection harvesting alters invertebrate communities, with some taxa increasing in abundance and others decreasing relative to un-harvested stands (Werner and Raffa 2000, Latty et al. 2006, Nol et al. 2006,

Shields et al. 2008). Changes in the abundance of understory plants (Heliolä et al. 2001,

Heyborne et al. 2003, Latty et al. 2006), quantity of coarse woody debris (Martikainen et al. 2000, Latty et al. 2006), availability of large diameter woody material (Hammond et al. 2004), snag volume, and gap area (Latty et al. 2006) have been linked to changes in arthropod abundance following harvesting. For insectivorous songbirds, prey abundance can influence the timing of nest initiation (Marshall et al. 2002), clutch size (Zanette et al.

2006, Robertson 2009), brood size (Preston and Rotenbury 2006), number of broods per season (Rodenhouse and Holmes 1992, Nagy and Holmes 2005), population density 93

(Gordon 1997, Lyons 2005), reproductive success (Lyons 2005), nestling provisioning rates (Nagy and Smith 1997), and nestling growth rates (Rodenhouse and Holmes 1992,

Duguay et al. 2000).

Prey availability depends on prey abundance (Holmes and Schultz 1988), prey characteristics (Cooper and Whitmore 1990, Holmes and Schultz 1988), a bird’s morphology and foraging strategies (Hutto 1990, Holmes and Robinson 1988, Holmes and Schultz 1988), and vegetation structure (Robinson and Holmes 1982, Holmes and

Schultz 1988). Foraging behaviour of songbirds is strongly associated with certain aspects of forest structure, including volume and height of trees, amount of dead wood, tree species composition (Naka 2003), and canopy characteristics (Bohm and Kalko

2009). Large diameter trees with highly structured bark, large canopy volume, and large amounts of dead wood typically sustain more insects than small diameter trees, smaller volume canopies, and lesser amounts of dead wood (Mariani and Manuwal 1990). As a result, changes in forest structure and composition resulting from silvicultural practices such as single-tree selection could influence insectivorous passerines.

Structural changes following single-tree selection harvesting include fewer large diameter (dbh >39.1 cm) live trees (Kenefic and Nyland 2000, Crow et al. 2002, Angers et al. 2005), fewer snags (Angers et al. 2005, Vanderwel et al. 2006), lower basal area

(Crow et al. 2002) and canopy closure (Kenefic and Nyland 2000), and reduced mean tree age (Angers et al. 2005, Vanderwel et al. 2006) and spatial complexity (Kenefic and

Nyland 2000). Single-tree selection also maintains (Johnson 1984, Della-Bianca and

Beck 1985, Busing and White 1997, Kenefic and Nyland 2000, Raymond et al. 2006) or favours the regeneration of shade-tolerant species such as sugar maple (Acer saccharum) 94 and American beech (Fagus grandifolia) (Poulson and Platt 1989, Kelty et al. 2003), whereas shade-intolerant species such as yellow birch (Betula alleghaniensis) and black cherry (Prunus serotina) are out-competed (Solomon and Grove 1999, Crow et al. 2002,

Webster and Lorimer 2002). If prey becomes less abundant or more difficult to detect as a result of these changes, adults may spend more time searching for food, leave their nests undefended for longer periods of time (Arcese and Smith 1988, Wolf et al. 1990,

Martin 1992), and potentially obtain less food for themselves and their nestlings. In addition, younger, less experienced birds may not be able to respond to these structural changes as efficiently as older birds (Ratcliffe et al. 1998).

The Rose-breasted Grosbeak (Pheucticus leudovicianus) is a Neotropical migrant that frequently breeds in both harvested and un-harvested hardwood forests across North

America (Wyatt and Francis 2002, Smith et al. 2006). Although it is still relatively abundant, this species is now of conservation concern in Ontario (Smith et al. 2006) due to annual population declines of -2.4% in Ontario between 1966 and 2007 (Sauer et al.

2008) and an especially sharp decline (-8% per year) between 1985 and 2005 (Cadman et al. 2007). Rose-breasted Grosbeaks respond inconsistently to harvesting, reacting negatively to selection harvesting in the Carolinian forest (Smith et al. 2006, Thompson et al. 2009) and partial harvesting in the boreal (Vanderwel et al. 2007, 2009), positively to various silvicultural treatments in New Brunswick (Guénette and Villard 2005), but showing no difference in abundance following selection harvesting (Holmes and Pitt

2007) and clear cutting (Duguay et al. 2001) in other regions. Although we know that

Rose-breasted Grosbeaks feed primarily on invertebrate prey during the nesting season 95

(Wyatt and Francis 2002), little information is available on how selection harvesting influences the feeding ecology of this declining species.

Here, I tested whether changes in habitat structure resulting from single-tree selection harvesting influence nestling provisioning rates in Rose-breasted Grosbeaks.

My main objectives were to test whether nestling provisioning rates, male contribution to nestling provisioning and nest attendance, prey load sizes, and nestling weights differed as a function of 1) time since single-tree selection harvesting, 2) stage in the nestling’s development, and 3) age of the male parent. I predicted that nestling provisioning rates would be highest, and prey load sizes would be largest by 16-20 years post-harvest, because these stands have the well-developed sapling and understory layers (Finegan

1984, Jobes et al. 2004) preferred by this species (Wyatt and Francis 2002). Even in unaltered habitats, the energy costs of nestling care are high for adult songbirds

(Thompson et al. 1998, Reid et al. 2002), and although both male and female Rose- breasted Grosbeaks provision their nestlings, females of many species provide most of the parental care (Stodola et al. 2010). If food items are easier to find and locate in the

16-20 year post-harvest stands, I would expect male contribution to nestling provisioning and nest attendance to be lower in this treatment than in the others. I also expected that nestling provisioning rates and prey load sizes would increase with nestling age (Norment

2003, Mitrus et al. 2010), and that nestlings cared for by older, more experienced males would be provisioned more frequently and receive larger prey load sizes than nestlings attended by younger males (Wunderle 1991, Limmer and Becker 2009). Finally, I predicted that if selection harvesting negatively affected the feeding ecology of Rose- 96 breasted Grosbeaks, nestling mass at the time of fledging would be higher in the 16-20 years post-harvest stands than in the other treatments.

METHODS

Study Species

During 2006 – 2008 Rose-breasted Grosbeaks arrived on their breeding grounds during the first week of May, began nest building by the last week of May, and cared for their nestlings throughout June and July. During the nestling provisioning period, Rose- breasted Grosbeak diets consist primarily of insects, including Coleoptera (McAtee 1908) and Lepidoptera larvae (Robinson and Holmes 1982, Holmes and Schultz 1988). Much of the time, this songbird perches and gleans invertebrate prey from the periphery of foliage and branches throughout the canopy, but it also collects invertebrates from branches and tree trunks while hovering (Holmes and Schultz 1988) and occasionally hawks for insects or forages at ground level (Wyatt and Francis 2002). When Rose- breasted Grosbeaks forage in the canopy, invertebrates are obtained from the underside of leaves during more than 65% of feeding attempts, making them lower-leaf specialists

(Holmes and Schultz 1988). Both male and female Rose-breasted Grosbeaks incubate and provision their young, which typically fledge 9-12 days after hatching (Scott 1998).

Study Area

Fieldwork was carried out during May-August in 2007 and 2008 in twelve tolerant hardwood forest stands in Algonquin Provincial Park, Ontario, Canada (45°

34`N, 78°04`W). The higher-elevation west side of this 775,000 ha park consists 97 primarily of mixed hardwood forest, whereas the lower-elevation east side is dominated by white and red pine (Ontario Parks 1998). The tolerant hardwood stands I studied were located in the central and southern part of this park (mean nearest neighbour distance among stands = 9.3 km, range = 1.0 – 34.2 km) (Fig. 1.1).

The hardwood forest stands were dominated by sugar maple (Acer saccharum),

American beech (Fagus grandifolia), eastern hemlock (Tsuga canadensis), and yellow birch (Betula alleghaniensis), and had been harvested using single-tree selection since the early 1970’s (OMNR 1998). This uneven-aged silvicultural system removes approximately one-third of the over-story basal area every 21-25 years (OMNR 1998). I studied four temporal stages in the single-tree selection cutting cycle: 0-5, 16-20, and 21-

25 years post-harvest and stands that had not been logged for >50 years. Three sites

(mean stand size = 34 ha) from each temporal stage were studied.

Nest Monitoring

Nests were monitored every 3-4 days using a mirror attached to a telescoping pole to determine date of nest initiation, clutch size, hatching date, number of nestlings, and fledging success.

Nestling Provisioning

Provisioning data were collected when the nestlings were 1-4 days old (n = 54; hereafter Stage 1 nestlings) and again when they were 6-9 days old (n = 33; hereafter

Stage 2 nestlings). Altricial passerine nestlings can be aged based on whether their eyes are open or closed, they can vocalize or not, and the developmental stage of their pin 98 feathers (Jongsomjit et al. 2007). By day 5-6 after hatching Rose-breasted Grosbeak nestlings open their eyes and are capable of making audible noises (Wyatt and Francis

2002). Nestlings were confirmed to be less than five days old if they were blind and inaudible, and more than five days old if they could see and make sounds. To avoid ambiguous estimates of nestling age I did not observe nestlings when they were thought to be exactly 5 days old.

Nests were observed for 1 h periods between 06:00 and 12:00 h when nestlings were 1-4 days old (n = 54, mean age = 2.2 d) and 6-9 days old (n = 33, mean age = 6.5 d).

Thirty-four nests were observed only during the first developmental stage, thirteen nests were only observed when nestlings were 6-9 days old, and twenty nests were observed during both developmental stages. Since feeding rate can vary with weather (Keller and van Noordwijk 1994) and time of day (Cowie and Hinsley 1988), no observations were collected if it was windy or raining, and I ensured that the proportion of observations taken in the early (06:00-08:00), mid (08:00-10:00), and late (10:00-12:00) morning were approximately equal among post-harvest treatments.

Observations were made using binoculars or a 20X telescope from concealed locations at distances of approximately 15-30 m from the nest. I recorded the sex of the bird bringing the food, the age if the bird was male (after-second-year or second-year), the time of each feeding event, and the size of the prey delivered relative to the size of the adult’s beak. After-second-year (hereafter ASY) males can be distinguished from second-year (hereafter SY) males by visible differences in their plumage (Pyle 1997).

While ASY males have uniformly black primaries, black heads and backs, and well- defined markings, SY males have less defined markings and often have brown streaking 99 on their heads, backs, and primary feathers (Pyle 1997). At the end of each 1 h observation period nest contents were checked using a telescoping pole with a mirror attached to the end to count the number of nestlings present. Standardized feeding rates were then calculated as the number of provisioning trips/ nestling/ hour.

Prey Load Sizes

Prey load sizes were visually assessed and categorized as small (barely visible in the adult’s beak), medium (visible in less than one-half the length of the adult’s beak), or large (filling more than one-half the adult’s beak) (Schaefer et al. 2004). When possible,

I visually identified prey types into broad taxonomic groups (e.g., insect order).

Nest Attendance

During the one-hour nest observations, I recorded the total amount of time that an adult was present at the nest (either brooding the nestlings or sitting within c. 30 cm of the nest) and the sex of the adult in attendance. I calculated the proportion of time the nest was attended, and the proportion of the total nest attendance that was contributed by the male.

Nestling Weights

I weighed nestlings at the time of fledging to the nearest 0.01g using a SC-HT-S-

232 digital scale. Whenever possible, multiple fledglings from each nest were weighed and an average weight for the brood was calculated. For some higher nests (22.6%; 12 of

53 nests) only one weight measurement was obtained because I was unable to capture 100 more than one fledgling once they had left the nest. These nests were distributed approximately equally among the 0-5 (3.8%), 16-20 (5.7%), and 21-25 (5.6%) years post- harvest stands, whereas the > 50 years post-harvest treatment had a slightly higher proportion (7.5%).

Statistical Analyses

I used split-split plot analysis of variance to investigate four response variables as a function of harvesting treatment (0-5, 16-20, 21-25, and > 50 years post-harvest), stage of nestling development (< 5 or > 5 d old), and adult male age (SY or ASY). Harvesting treatment was the whole-plot factor and the latter two were split-plot factors (Littell et al.

1996). Response variables were 1) nestling provisioning rate (number of trips/nestling/h), 2) male contribution to nestling provisioning, 3) nest attendance, and 4) male contribution to nest attendance. I first used multivariate analysis of variance to test for significant relationships between the three predictor variables and each of the response variables, and then I used univariate split-split plot analyses of variance to investigate each response variable separately.

For each response variable a mean value was calculated for each stand from the data collected during the 1 h nest observation periods. In all tests, each stand mean was weighted according to the number of observations collected in that stand. Separate analyses were conducted for 2007 and 2008 due to significant (P < 0.002) among year differences in nestling provisioning rates, and model assumptions of homogeneity and normality were verified by examination of residuals. I obtained homogeneity of variance for nestling provisioning rates using a log transformation (log(x + 1)). All statistical 101 analyses were conducted in the PROC GLM module of SAS (v. 8.02). I first tested the full model. If the 3-way interaction was clearly not significant (i.e., P > 0.15) it was removed from the analysis and the model rerun. Similarly, any 2-way interactions that were clearly not significant in the second run were excluded from the final run.

I used contingency table analysis to investigate prey load size (small, medium, or large) as a function of harvesting treatment (0-5, 16-20, 21-25, and > 50 years post- harvest), stage of nestling development (< 5 or > 5 d old), and adult male age (SY or

ASY). Model significance was determined using the Mantel-Haenszel Chi-square test statistic that is appropriate when 50% or more of the expected cell frequencies are less than 5 (SAS 2001). Analyses were conducted in the PROC FREQ module of SAS (v.

8.02).

I could not use split-split plot analyses of variance to investigate nestling weight at the time of fledging as a function of harvesting treatment (0-5, 16-20, 21-25, and > 50 years post-harvest) and male age (SY or ASY) due to insufficient sample size (n = 63).

Since harvesting treatment did not have a significant effect on any of the other response variables tested, I collapsed harvesting treatment into just two levels: harvested (i.e., 0-25 years post-harvest) and unharvested (> 50 years post-harvest) for inclusion into the analyses of nestling weights. Split-plot analysis of variance was conducted in SAS using the procedure described above to investigate nestling weight as a function of harvest treatment and male age. 102

RESULTS

Multivariate analysis of variance indicated that the response variables differed significantly among study years (F23.4, 48.75 = 8.89, P = 0.002), so separate analyses were performed for 2007 and 2008. Multivariate analysis of variance indicated that there were no significant differences in nestling provisioning rates, male contribution to nestling provisioning, nest attendance, or male contribution to nest attendance among post-harvest treatments in either study year (2007: F12, 26.75 = 1.48, P = 0.192; 2008: F12, 39.98 = 1.83, P =

0.076). In 2007, response variables differed significantly with stage of nestling development (F4, 10 = 6.87, P = 0.006) but there was a significant interaction between stage of nestling development and harvesting treatment (F12, 26.75 = 3.95, P = 0.002) for nest attendance and male contribution to nest attendance. In 2008, response variables did not differ significantly between stages of nestling development (F4, 15 = 1.27, P = 0.327).

Response variables differed significantly with age of male parent during both study years

(2007: F4, 10 = 3.64, P = 0.044; 2008: F4,15 = 6.44, P = 0.003) but there was a significant interaction between harvest treatment and male age in 2007 (F12, 26.75 = 2.89, P = 0.011) for nest attendance, and an almost significant interaction between harvest treatment and male age in 2008 (F12, 39.98 = 1.92, P = 0.061) for nestling provisioning rates.

Nestling Provisioning Rates

In total I completed 1 h nest watches at 130 nests (90 in 2007 and 40 in 2008).

Nestlings were provisioned on average at a rate of 0.67 trips/nestling/h (SE = 0.07, range

= 0-4.33 trips/nestling/h, n = 130). Nestling provisioning rates did not differ significantly among harvest treatments in either year (2007: F1,3 = 0.21, P = 0.887; 2008: F1,3 = 2.77, P 103

= 0.078), although provisioning rates were lowest in the > 50 years post-harvest stands in both study years (Table 3.1). Provisioning was less frequent for Stage 1 nestlings (< 5 days old) than for Stage 2 nestlings (> 5 days old) in all post-harvest treatments in 2007

(F1,3 = 9.79, P = 0.005), but in 2008 differences were not significant (F1,3 = 1.72, P =

0.209), and the mean provisioning rate in the 0-5 years post-harvest stands was higher for

Stage 1 nestlings than for Stage 2 nestlings (interaction: F1,3 = 3.67, P = 0.036). Nestling provisioning was more frequent at nests attended by ASY males than at those attended by

SY males in all post-harvest treatments in 2007, although not significantly so (F1,3 = 3.47,

P = 0.078). Mean provisioning rates were significantly higher at nests attended by ASY males in all stands in 2008 (F1,3 =13.45, P = 0.002) as well, except the 21-25 years post- harvest treatment (interaction: F1,3 = 3.67, P = 0.063) (Table 3.1).

Male Contribution to Nestling Provisioning

On average, males contributed 35.2 (± 3.2 SE) % of the feeding events. Male contribution to nestling provisioning did not differ significantly among post-harvest treatments in either year (2007: F1,3 = 1.54, P = 0.243; 2008: F1,3 = 1.27, P = 0.31), and among-treatment differences were not consistent between years (Table 3.2). Male contribution to nestling provisioning did not differ significantly with nestling age in either year (2007: F1,3 = 1.50, P = 0.238; 2008: F1,3 = 0.01, P = 0.952), but in both years it tended to be lower for Stage 1 than Stage 2 nestlings in all stands except the > 50 years post-harvest stands. In 2007, contributions from ASY and SY males did not consistently differ among post-harvested treatments (F1,3 = 1.52, P = 0.236), but in 2008 ASY males 104

contributed significantly more (F1,3 = 6.68, P = 0.017) than SY males in all post-harvest treatments except the 16-20 years post-harvest stands.

Nest Attendance

Nests were attended by adults 66.7 (± 3.2) % of the time, with male attendance accounting for only 18.8 (± 1.9) % of that time. Nest attendance did not significantly differ among harvest treatments in either year (2007: F1,3 = 2.32, P = 0.123; 2008: F1,3 =

0.47, P = 0.709), and trends were not consistent among study years (Table 3.3). In 2007, nest attendance was higher when nestlings were < 5 days old than when nestlings were >

5 days old (F1,3 = 8.18, P = 0.013), except in the > 50 years post-harvest treatment

(interaction: F1,3 = 4.35, P = 0.025). In 2008, nest attendance was also generally higher for < 5 day old nestlings, but not significantly so (F1,3 = 1.23, P = 0.281). In 2007, ASY males attended their nests less than SY males (F1,3 = 0.19, P = 0.673) except in the > 50 years post-harvest treatment (interaction: F1,3 = 3.95, P = 0.033). In 2008, nest attendance by ASY and SY males did not differ significantly among harvest treatments

(F1,3 = 1.23, P = 0.281).

Male Contribution to Nest Attendance

Nest attendance was observed at 102 nests with ASY males and 28 nests with SY males. The proportion of time males spent at the nest relative to females differed significantly with time since harvest in both study years (2007: F1,3 = 3.80, P = 0.031;

2008: F1,3 = 4.53, P = 0.013). In 2007, male contribution was significantly lower in the

16-20 year post-harvest treatment than in the other treatments, but in 2008 male 105

TABLE 3.1. Mean nestling provisioning rates at Rose-breasted Grosbeak nests in 12 hardwood stands in Algonquin Provincial Park harvested by single-tree selection 0-5, 16-20, 21-25, and > 50 years prior to study (standard errors and sample sizes are in parentheses). Mean nestling provisioning rates (no. trips/ nestling/ h) are presented for nests attended by second-year (SY) and after-second-year (ASY) males, and for nests with Stage 1 (< 5 days old) and Stage 2 (> 5 days old) nestlings. Time since harvest (years) 0-5 16-20 21-25 >50 2007 Overall treatment mean 0.676 (0.175, 34) 0.667 (0.191, 14) 0.520 (0.142, 26) 0.459 (0.177, 16) Nestling development Stage 1 0.391 (0.081, 22) 0.002 (0.001, 6) 0.314 (0.082, 16) 0.250 (0.164, 8) Stage 2 1.193 (0.447, 12) 1.167 (0.189, 8) 0.850 (0.330, 10) 0.667 (0.309, 8) Male Age ASY 0.766 (0.193, 30) 0.800 (0.249, 10) 0.507 (0.154, 23) 0.612 (0.220, 12) SY 0.002 (0.001, 4) 0.335 (0.193, 4) 0.670 (0.001, 3) 0.001 (0.001, 4) 2008 Overall treatment mean 0.767 (0.290, 10) 0.765 (0.178, 10) 1.137 (0.252, 11) 0.667 (0.190, 9) Nestling development Stage 1 1.083 (0.293, 6) 0.570 (0.115, 7) 0.811 (0.220, 7) 0.667 (0.202, 7) Stage 2 0.436 (0.172, 4) 1.220 (0.485, 3) 1.707 (0.493, 4) 0.665 (0.665, 2) Male Age ASY 1.767 (0.290, 8) 0.582 (0.145, 4) 1.585 (0.338, 6) 1.000 (0.001, 6) SY 0.500 (0.001, 2) 0.887 (0.281, 6) 0.600 (0.214, 5) 0.141 (0.001, 3) 106

TABLE 3.2. Mean proportion of feeding events contributed by male Rose-breasted Grosbeaks in 12 hardwood stands in Algonquin Provincial Park harvested by single-tree selection 0-5, 16-20, 21-25, and > 50 years prior to study (stand errors and sample sizes are in parentheses). Mean proportion of feeding events are presented for second-year (SY) and after-second-year (ASY) males, and for nests with Stage 1 (< 5 days old) and Stage 2 (> 5 days old) nestlings. Time since harvest (years) 0-5 16-20 21-25 >50 2007 Overall treatment mean 0.466 (0.068, 34) 0.338 (0.081, 14) 0.286 (0.101, 26) 0.125 (0.060, 16) Nestling development Stage 1 0.405 (0.082, 22) 0.154 (0.067, 6) 0.222 (0.088, 16) 0.167 (0.110, 8) Stage 2 0.580 (0.118, 12) 0.522 (0.131, 8) 0.400 (0.245, 10) 0.082 (0.054, 8) Male Age ASY 0.496 (0.069, 30) 0.309 (0.082, 10) 0.250 (0.115, 23) 0.167 (0.077, 12) SY 0.001 (0.000, 4) 0.500 (0.289, 4) 0.500 (0.001, 3) 0.001 (0.000, 4) 2008 Overall treatment mean 0.285 (0.099, 10) 0.300 (0.113, 10) 0.497 (0.068, 11) 0.472 (0.138, 9) Nestling development Stage 1 0.274 (0.107, 6) 0.357 (0.148, 7) 0.486 (0.110, 7) 0.500 (0.159, 7) Stage 2 0.300 (0.200, 4) 0.167 (0.167, 3) 0.518 (0.017, 4) 0.375 (0.375, 2) Male Age ASY 0.342 (0.110, 8) 0.292 (0.172, 4) 0.578 (0.087, 6) 0.708 (0.109, 6) SY 0.001 (0.000, 2) 0.305 (0.163, 6) 0.400 (0.100, 5) 0.001 (0.000, 3) 107

TABLE 3.3. Mean nest attendance at Rose-breasted Grosbeak nests in 12 hardwood stands in Algonquin Provincial Park harvested by single-tree selection 0-5, 16-20, 21-25, and > 50 years prior to study (standard errors and sample sizes are in parentheses). Mean nest attendance is presented for nests attended by second-year (SY) and after-second-year (ASY) males, and for nests with Stage 1 (< 5 days old) and Stage 2 (> 5 days old) nestlings. Time since harvest (years) 0-5 16-20 21-25 >50 2007 Overall treatment mean 0.694 (0.053, 34) 0.581 (0.084, 14) 0.713 (0.081, 26) 0.595 (0.120, 16) Nestling development Stage 1 0.719 (0.059, 22) 0.793 (0.065, 6) 0.998 (0.002, 16) 0.500 (0.189, 8) Stage 2 0.645 (0.108, 12) 0.422 (0.110, 8) 0.257 (0.101, 10) 0.690 (0.152, 8) Male Age ASY 0.673 (0.055, 30) 0.464 (0.091, 10) 0.689 (0.086, 23) 0.793 (0.108, 12) SY 1.000 (0.001, 4) 0.875 (0.072, 4) 1.000 (0.001, 3) 0.002 (0.001, 4) 2008 Overall treatment mean 0.715 (0.099, 10) 0.727 (0.102, 10) 0.663 (0.092, 11) 0.576 (0.152, 9) Nestling development Stage 1 0.647 (0.131, 6) 0.860 (0.578, 7) 0.721 (0.115, 7) 0.593 (0.167, 7) Stage 2 0.810 (0.155, 4) 0.417 (0.252, 3) 0.560 (0.162, 4) 0.515 (0.485, 2) Male Age ASY 0.658 (0.110, 8) 0.820 (0.148, 4) 0.705 (0.090, 6) 0.385 (0.182, 6) SY 1.000 (0.001, 2) 0.665 (0.143, 6) 0.612 (0.183, 5) 0.957 (0.043, 3) 108

contribution was significantly higher in this treatment than in the others (Table 3.4). In 2007, males contributed significantly more to nest attendance when nestlings were < 5 days old than when they were > 5 days old (F1,3 = 8.54, P = 0.010) except in the > 50 years post- harvest treatment (interaction: F1,3 = 5.33, P = 0.010). In 2008 nest attendance by males was also higher when nestling were < 5 days old, but not significantly so (F1,3 = 3.36, P = 0.081).

ASY males contributed significantly more to nest attendance than SY males in both study years (2007: F1,3 = 13.45, P = 0.002; 2008: F1,3 = 17.31, P < 0.001).

Prey Load Sizes

During 37 of 73 (51%) feeding events it was possible for observers to accurately identify the type of food item being delivered to the nestlings. Insect matter which could not be identified to family was delivered to nestlings during 16 feeding events (22%), adult

Lepidoptera were brought during 11 visits (15%), and larval Lepidoptera were brought during 8 feeding events (11%). Rose-breasted Grosbeak adults were also observed feeding dragonflies (1%) and leaf buds (1%) to their nestlings.

Small prey loads were the most frequently delivered in all post-harvest treatments except the 16-20 years post-harvest stands, where large loads were the most frequently provided, although differences were not significant (X2 = 3.098, df = 1, P = 0.078) (Table

3.5).

The age of the male parent was not significantly associated with the prey load sizes delivered (X2 = 1.576, df = 1, P = 0.209). Small prey loads were most frequently delivered at nests attended by ASY males, followed by large and then medium loads. At nests attended 109

TABLE 3.4. Mean nest attendance contributed by male Rose-breasted Grosbeaks in 12 hardwood stands in Algonquin Provincial Park harvested by single-tree selection 0-5, 16-20, 21-25, and > 50 years prior to study (standard errors and sample sizes are in parentheses). Mean nest attendance is presented for second-year (SY) and after-second-year (ASY) males, and for nests with Stage 1 (< 5 days old) and Stage 2 (> 5 days old) nestlings. Time since harvest (years) 0-5 16-20 21-25 >50 2007 Overall treatment mean 0.282 (0.045, 34) 0.056 (0.182, 14) 0.220 (0.037, 26) 0.162 (0.045, 16) Nestling development Stage 1 0.388 (0.053, 22) 0.280 (0.103, 6) 0.298 (0.025, 16) 0.132 (0.053, 8) Stage 2 0.082 (0.040, 12) 0.085 (0.033, 8) 0.080 (0.049, 10) 0.192 (0.075, 8) Male Age ASY 0.301 (0.046. 30) 0.211 (0.065, 10) 0.215 (0.043, 23) 0.217 (0.051, 12) SY 0.001 (0.000, 4) 0.025 (0.014, 4) 0.250 (0.000, 3) 0.001 (0.000, 4) 2008 Overall treatment mean 0.094 (0.043, 10) 0.820 (0.053, 10) 0.125 (0.043, 11) 0.070 (0.042, 9) Nestling development Stage 1 0.137 (0.068, 6) 0.204 (0.070, 7) 0.160 (0.062, 7) 0.900 (0.052, 7) Stage 2 0.034 (0.034, 4) 0.130 (0.098, 3) 0.065 (0.009, 4) 0.001 (0.000, 2) Male Age ASY 0.113 (0.050, 8) 0.360 (0.037, 4) 0.138 (0.045, 6) 0.105 (0.059, 6) SY 0.001 (0.000, 2) 0.063 (0.031, 6) 0.110 (0.079, 5) 0.001 (0.000, 3) 110

TABLE 3.5. Observed frequencies for prey load sizes delivered to Rose-breasted Grosbeak nests in 12 hardwood stands in Algonquin Provincial Park harvested by single-tree selection (row percents are in parentheses). Mean prey load sizes are presented for harvest treatment (0-5, 16-20, 21-25, and > 50 years post-harvest), for nests with < 5 or > 5 days old nestlings, and for nests attended by after-second-year (ASY) and second-year (SY) male parents.

Prey Load Size Small Medium Large Years Since Harvest 0-5 35 (51.5) 19 (27.9) 14 (20.6) 16 – 20 13 (35.1) 9 (24.3) 15 (40.5) 21 – 25 29 (46.8) 17 (27.4) 16 (25.8) > 50 14 (63.6) 2 (9.1) 6 (27.3) Nestling Age < 5 days 45 (60.0) 23 (30.7) 7 (9.3) > 5 days 45 (39.8) 24 (21.2) 44 (38.9) Male Age ASY 81 (50.0) 38 (23.4) 43 (26.5) SY 14 (48.3) 9 (31.0) 6 (20.7) 111

by SY males small prey loads were also most frequently delivered, followed by medium and then large loads.

Nestling Weights at Fledging

Mean nestling weight at the time of fledging was 31.1 g (± 0.4 g) (range = 22.5 – 37.5 g, n = 63). Mean nestling weight did not significantly differ with harvest treatment in 2007

(harvested: 30.6g; unharvested: 30.3g; F1,1 = 0.19, P = 0.665) or in 2008 (harvested: 31.5g; unharvested 32.3g; F1,1 = 0.19, P = 0.672). Mean nestling weight was higher at nests attended by ASY males than SY males in both 2007 (ASY: 30.8g; SY: 29.6g) and 2008

(ASY: 32.0g; SY: 30.3g) but only significantly so in 2008 (2007: F1,1 = 1.87, P = 0.180;

2008: F1,1 = 4.80, P = 0.043).

DISCUSSION

Single-tree Selection Harvesting

Nestling provisioning rates, nest attendance, the size of prey loads being delivered to nestlings, and nestling weights did not significantly differ among post-harvest treatments in either study year. Previous research has found that harvesting can alter prey availability for insectivorous songbirds (Pettersson et al. 1995, Duguay et al. 2000), but most studies have focused on the effects of clear-cutting on ground-dwelling arthropods (Niemelä et al. 1993,

Greenberg and McGrane 1996, Marra and Edmonds 1998, Heliӧlä et al. 2001, Koivula et al.

2002). Studies of invertebrate responses to less-intensive silvicultural methods are less frequent (Koivula 2002, Moore et al. 2002, Siira-Pietikainen et al. 2003, Pihlaja et al. 2006), 112

and most focus on changes in diversity and species richness (i.e., Werner and Raffa 2000,

Ulyshen et al. 2004, Latty et al. 2006, Shields et al. 2008), making potential changes in prey availability for songbirds difficult to assess (Halaj et al. 2008). Two studies concluded that there was no difference in overall abundance of ground invertebrates following selection harvesting (Atlegrim and Sjoberg 1996, Moore et al. 2002), but one study found that in the first fifteen years after single-tree selection harvesting there was an increase in the abundance of generalist and open habitat ground beetles, but similar species richness compared to unharvested hardwood stands (Vance and Nol 2003). I am unaware of any studies which have investigated the consequences of single-tree selection for nestling provisioning rates or nestling condition at the time of fledging.

Nestling growth rates have been linked to variations in food abundance resulting from natural causes (Price 1985, Dias and Blondel 1996, Holmes et al. 1996) and from clear-cut harvesting (Duguay et al. 2000). My study suggests that if there were any changes in arthropod abundance or prey availability resulting from single-tree selection harvesting they did not appear to have a significantly negative influence on the Rose-breasted Grosbeak’s ability to find food during the nestling phase.

Studies investigating changes in arthropod diversity and species richness following selection harvesting show that responses are largely taxon specific, with groups that favour more open habitat increasing whereas mature forest groups decrease in abundance (Werner and Raffa 2000, Latty et al. 2006, Shields et al. 2008). In particular, spiders (Arachnida) decreased following harvesting (Moore et al. 2002), millipedes (Polydesmida), snails

(Stylommatophora) (Strayer et al. 1986, Moore et al. 2002), hoverflies (Diptera), bees 113

(Hymenoptera), and click beetles (Coleoptera) (Nol et al. 2006) were more abundant, and there was no difference in the abundance of Carabid beetles (Coleoptera) (Moore et al. 2002) and crickets (Orthoptera) (Moore et al. 2002) before and after harvesting. In addition, wood- boring insects (Ulyshen et al. 2004, Shields et al. 2008), grasshoppers (Orthoperta) and wolf- spiders (Araneae) (Shields et al. 2008) were more abundant in canopy gaps than in surrounding areas with higher canopy cover, but weevils (Coleoptera) (Shields et al. 2007) and rove beetles (Coleoptera) (Yu et al. 2006) were less abundant in these areas. If any taxon-specific changes in invertebrate abundance occurred after single-tree selection, Rose- breasted Grosbeaks were able to compensate sufficiently to maintain nestling provisioning rates and nestling masses at the time of fledging.

The absence of significant differences in nestling provisioning rates, nest attendance, and nestling weights among post-harvest treatments could be the result of minimal changes in arthropod abundance or prey availability following single-tree selection harvesting

(Atlegrim and Sjoberg 1996, Moore et al. 2002). Alternatively, while there is evidence to suggest that food is limiting for songbirds during the breeding season (Martin 1986), it is possible that Rose-breasted Grosbeaks were able to compensate for any changes in food availability by altering their foraging behaviour. For example, Coquerel’s Coua (Coua coquereli) were able to maintain their rate of prey capture despite structural changes following logging by adjusting their feeding activities (Chouteau 2009). Feeding rate can also vary with weather (Keller and van Noordwijk 1994), time of day (Cowie and Hinsley

1988), and intensity of nestling begging (Ottosson et al. 1997, Burford et al. 1998, Greico

2001), and although I do not think it is likely given the relatively large sample sizes here, 114

variations in feeding rates caused by these factors may have obscured differences among post-harvest treatments.

A lack of significant differences in the proportion of time Rose-breasted Grosbeaks allocated to nest attendance, or in the proportion of nest attendance provided by males in the varying post-harvest treatments also supports the conclusion that single-tree selection had minimal impacts on Rose-breasted Grosbeak feeding ecology. Fedy and Martin (2009) demonstrated that time spent foraging was negatively correlated with time spent brooding by females of Red-faced Warblers (Cardellina rubrifrons) and Gray-headed Juncos (Junco hyemalis caniceps). When males were not in attendance females spent more time being vigilant, their foraging efficiency decreased, and they spent less time on the nest (Fedy and

Martin 2009). In my study nest attendance and nestling provisioning rates were both slightly higher in more recently harvested stands than in un-harvested ones, which suggests that prey were more abundant or more easily located and captured after selection cutting. Since male contribution to nest attendance was also slightly higher in more recently harvested stands, an alternative hypothesis is that predation pressure, and the need for nest defence, were also higher following selection harvesting. However, no aspect of Rose-breasted Grosbeak feeding ecology that I studied showed consistently significant differences among post-harvest treatments.

Nestling Development

Nestling provisioning was less frequent and nest attendance was higher for < 5 day old nestlings than for > 5 day old nestlings, but differences were only significant in 2007, 115

possibly as a result of larger sample sizes during that year. Larger nestlings have greater energy requirements, and in many species both males and females increase the frequency of feeding as nestlings mature (e.g. Haggerty 1992, Norment 2003; Mitrus et al. 2010). In addition, parents adjust their provisioning rates and the amount of food they bring according to the frequency and intensity of nestling begging, and older nestlings are capable of more vigorous and sustained demands (Haggerty 1992). Also as expected, Rose-breasted

Grosbeak nest attendance was higher when nestlings were <5 days old than when nestlings were >5 days old during both study years. In many passerines the amount of time spent brooding decreases with the age of the nestlings, partially because more time is spent foraging and thermoregulatory requirements of nestlings are lower (Moreno 1987, Sanz and

Moreno 1995, Sanz and Tinbergern 1999, Chastel and Kersten 2002).

In both study years male Rose-breasted Grosbeaks contributed less to nestling provisioning and nest attendance than females did during both stages of nestling development. According to Evolutionary Stable Strategy (ESS) models, males and females of monogamous species with biparental care should make approximately equal contributions to parental care (Maynard-Smith 1977, Winkler 1987). However, in situations where paternity is uncertain or the opportunity for additional fertilizations is high, males may provide less parental care than females (Maynard-Smith 1977, Breitwisch et al. 1986, Smith et al. 1988, Westneat 1988). Males contribute slightly less to nestling provisioning than females in Northern Mockingbirds (Mimus polyglottos) (Breitwisch et al. 1986), Tree

Swallows (Tachycineta bicolour) (Leffelaar and Robertson 1986), Pied Flycatchers,

(Ficedula hypoleuca) (Alatalo et al. 1988), Great Tits (Parus Major) (Smith et al. 1988) and 116

Eastern Phoebes (Sayornis phoebe) (Conrad and Robertson 1993), and research has shown that up to 40% of nests contain nestlings sired through extra-pair copulations in some species of songbirds (Stutchbury et al. 1997). Future assessments of paternity in Rose-breasted

Grosbeak nestlings are needed to determine if high rates of extra-pair copulations can account for the lower contribution to parental care by the males relative to the females of this species.

Rose-breasted Grosbeak males’ contribution to provisioning almost doubled for > 5 day old nestlings (39.5%) relative to < 5 day old nestlings (23%), but differences were not significant in either study year, possibly as a result of small sample sizes and high variation among post-harvest treatments. No significant differences in the proportion of nestling provisioning provided by males between the two stages suggests that male and female Rose- breasted Grosbeaks increase their feeding effort at the same rate during nestling development. In some species, including Great Tits (Parus major) (Sanz and Tinbergen

1999), Eastern Kingbirds (Tyrannus tyrannu) (Morehouse and Brewer 1968), Gray Catbirds

(Dumatella carolinensis) (Johnson and Best 1982), and Eastern Phoebes (Sayornis phoebe)

(Conrad and Robertson 1993) males contribute more to nestling provisioning during the early stages, while the females brood. Rose-breasted Grosbeak males contributed significantly more to nest attendance during the early stages (when nestlings were < 5 days old) than later on, but did not alter their relative contribution to nestling provisioning. This suggests that when females cannot adequately meet the conflicting demands of brooding and foraging,

Rose-breasted Grosbeak males will contribute proportionally more to nest attendance rather than to nestling provisioning. 117

Age of Male Parent

Nestling provisioning rates were lower at nests attended by younger, less experienced males than at nests with older males, and the younger adults were unable to compensate by providing larger prey load sizes. As a result, nestlings raised by younger males weighed less at the time of fledging than nestlings reared by older males. Other studies have found that the age of both male and female parents can influence the date of clutch initiation, clutch size

(Reynolds 1997), number of fledglings produced (Préault et al. 2005, Brown and Roth 2009), nestling condition (Préault et al. 2005, but see Wheelwright and Schultz 1994), and quality of parental care (Préault et al. 2005). The body condition of nestlings at the time of fledging can influence their ability to find suitable habitats for foraging and predator avoidance, and as a result, fledglings in poorer condition may have lower chances of survival (Magrath 1991,

Naef-Daenzer et al. 2001, Vitz and Rodewald 2010; but see Anders et al. 1997). The correlation between body condition and fledgling survival suggests that for Rose-breasted

Grosbeaks, the offspring of first-time breeders, which weigh less at the time of fledging, may have a fitness disadvantage relative to offspring raised by more experienced adults.

Several hypotheses have been proposed to explain differences in reproductive success between SY and ASY adults (reviewed in Curio 1983, Forslund and Pärt 1995, Martin 1995).

The selection hypothesis suggests that reproductive success appears higher in older individuals because inferior phenotypes, which would have lower success, experience high mortality at young ages. The recruitment hypothesis proposes that younger individuals may postpone breeding in order to accumulate sufficient resources to successfully breed later in life. According to the restraint hypothesis, older individuals should have higher reproductive 118

output because survival and reproductive value decrease with age. Finally, the constraint hypothesis suggests that older individuals have higher reproductive success because parental skills such as nest-site selection, foraging ability, coordination between partners, and offspring defence improve with age (Forslund and Part 1995, Martin 1995, Espie et al. 2000,

Balbontin et al. 2007). These hypotheses are not mutually exclusive (Curio 1983, Forslund and Pärt 1995, Martin 1995, Brown and Roth 2009), and I did not specifically test any of them. The lack of difference I observed between habitat characteristics at nests with older and younger Rose-breasted Grosbeaks suggests that lack of experience (constraint hypothesis) was more likely a cause of differences in nestling provisioning rates than unequal habitat quality, but further research is required to clarify the causes of lower reproductive success in less experienced Rose-breasted Grosbeaks.

Previous studies have found that younger birds of many species cannot forage as efficiently as older, more experienced adults (e.g. Janson 1990, Wunderle 1991, Daunt et al.

1999, Limmer and Becker 2009; but see Heise and Moore 2003). Differences in foraging ability (Daunt et al. 1999), diet choice (Rutz et al. 2006), the ability to capture large (Catterall et al. 1989, Desrochers 1992) or energy rich (Limmer and Becker 2009) prey items, timing of nest initiation (Stutchbury and Roberston 1988, Hatch and Westneat 2007; but see Daunt et al. 1999) and access to high quality foraging territories (Catterall et al. 1989, Pärt 2001) may contribute to the positive relationship between age and the amount or quality of food obtained. Evidence that the foraging skills and reproductive success of young birds improve with age (Stuchbury and Robertson 1988, Desrochers 1992, de Forest and Gaston 1996,

Catry and Furness 1999) supports the hypothesis that age-related differences in reproductive 119

performance can be attributed to a lack of previous experience in younger birds (Yasukawa

1981, Nol and Smith 1987, Wheelwright and Schultz 1994).

As a result of age-related differences in foraging ability, in environments and years when food resources are particularly scarce, younger birds may be more strongly affected than older birds (Hannon and Smith 1984, Bedard and LePointe 1985, Hussell and Quinney

1987, Stutchbury and Robertson 1988) or they may be unable to provide enough food to successfully fledge young (Sydeman et al. 1991). In my study nestlings raised by younger male parents weighed less than nestlings raised by older parents, but differences were only significant in 2008. Most Rose-breasted Grosbeaks have nestlings during June, and there was considerably more rain in 2008 than in 2007 during this month, which suggests that younger Rose-breasted Grosbeak parents may have more difficulty adjusting to adverse weather conditions than more experienced adults.

In addition to age-related differences in nestling provisioning rates, nest attendance and male contribution to nest attendance also varied with the age of the male parent. Overall nest attendance was higher at nests attended by younger, less experienced males than at nests with older, more experienced males. Higher nest attendance could be expected to increase nest survival, but nests with younger males were not more successful on average than nests with older males (Table 1.5, Chapter 1). In addition, while nest attendance was higher at nests with younger males, the SY males contributed less to nestling provisioning than older males did during both stages in the nestlings’ development. This suggests that females who breed with second-year males may provide more parental care than females paired with older, more experienced males. The body condition of female songbirds decreases rapidly at 120

the onset of brooding and remains low until the young fledge (Curlee and Beissinger 1995,

Sanz and Moreno 1995, Chastel and Kersten 2002). Incubation and nestling provisioning are both energetically costly for females (Thomson et al. 1998, Reid et al. 2002), which suggests that female Rose-breasted Grosbeaks who are mated with SY males may have much lower body condition than females mated with more experienced males who provide more parental care.

Conclusions

This study suggests that single-tree selection does not negatively influence Rose- breasted Grosbeak’s ability to find food, provision their nestlings, or attend their nests in hardwood forests. None of the response variables differed significantly among post-harvest treatments in either study year, but they differed with stage of nestling development and with age of male parent. Provisioning was less frequent and nest attendance was higher for < 5 day old nestlings than for > 5 day old nestlings. Fledglings from nests attended by younger

(SY) males weighed less than offspring from nests attended by older (ASY) males, and may have been at a fitness disadvantage. Further research is needed to determine why younger breeders were less effective at nestling provisioning. GENERAL CONCLUSIONS

In the predominantly forested landscape of Algonquin Provincial Park, single-tree selection did not appear to have a significant negative effect on the reproductive success of Rose-breasted Grosbeaks. Density, pairing success, and the mean number of fledglings per successful nest were significantly higher in more recently harvested stands than in the > 50 years post-harvest stands, and nest success, fledgling survival, and productivity were also higher in the harvested stands, although not significantly so.

These demographic trends partially support my prediction that single-tree selection would benefit Rose-breasted Grosbeaks. However, the proportion of ASY males was slightly higher in the > 50 years post-harvest treatment than in the other stands, which could indicate that the “unharvested” stands provided higher quality territories, assuming that older, more experienced birds select and successfully defend better quality territories than younger, less experienced ones. Since overall density was significantly lower and the proportion of ASY males was not significantly higher in the > 50 years post-harvest treatment, it seems unlikely that the unharvested stands provided significantly better habitat. Rather, ASY males may have been out-competing SY ones for a more limited number of high quality territories in these stands than were available in other treatments.

The lack of agreement among demographic measures indicates that the factors influencing Rose-breasted Grosbeak reproductive success are complex, and as a result, relying on any single demographic measure to assess the effects of habitat management may result in misleading conclusions.

Single-tree selection harvesting may not have a significant negative effect on

Rose-breasted Grosbeaks in predominantly forested landscapes, but population

121 122 productivity was below replacement levels in all stands except in the 16-20 years post- harvest stands. This suggests that forest management policies should be implemented with caution in areas where the welfare of this species is of concern, at least until the causes of the Rose-breasted Grosbeak’s decline can be identified. My study suggests that the dense understory and sapling layers characteristic of forests regenerating after single- tree selection harvesting may slightly increase Rose-breasted Grosbeak productivity, but some other forest management practices may have more negative consequences. For example, forest management policies intended to suppress natural disturbances such as wildfires, flooding, and insect outbreaks have reduced the amount of early-seral forest across North America (Askins 1993, Brawn et al. 2001, Trani et al. 2001, Brooks 2003,

DeGraaf and Yamasaki 2003). Creating forested landscapes which provide access to a range of forest types and successional stages may be beneficial for mitigating the population declines of the Rose-breasted Grosbeak and other disturbance-dependent species, but silvicultural applications should be selected and implemented with caution.

This study was conducted in stands that were regenerating after their first harvest, and some of them were reaching the stage when a second harvest could be carried out.

Silvicultural guidelines for tolerant hardwood forests in the states just south of the Great

Lakes suggest that a second harvest can be made using single-tree selection 8-15 years after the first (Tubbs 1977), while guidelines for Ontario suggest waiting 15-20 years

(OMNR 1998). Although basal area was not significantly lower in the 16-20 or 21-25 years post-harvest stands than in the > 50 years post-harvest treatment, other measures of forest structure were significantly different. Rose-breasted Grosbeak density and pairing success were significantly higher in the 16-20 years post-harvest treatment than in the > 123

50 years post-harvest stands, and pairing success was significantly higher in the 21-25 years post-harvest stands than in the “unharvested” stands. This suggests that neither forest structure nor Rose-breasted Grosbeak demography had completely returned to

“pre-harvest” conditions even by 25 years after harvesting, and that shorter harvest rotations may lead to cumulative changes in forest structure and avian ecology.

Specifically, basal area, which was positively associated with nest survival, was still slightly lower in the 21-25 years post-harvest stands than in the > 50 years post-harvest treatment. However, further study of forest structure and Rose-breasted Grosbeak demography following the second cutting would be necessary to test for cumulative changes.

Although the first single-tree selection harvest rotation did not appear to have a significant effect on nest survival, my study suggests that basal area, sampling year, and timing of nest initiation were important variables explaining nest survival, with nests built in areas with higher basal area and started earlier in the season having a greater chance of survival. The tendency for older, more experienced birds, which typically have higher reproductive success, to begin nesting earlier may partially explain the positive association between early breeding and nest survival. However, it is possible that the hatching dates of young from nests initiated earlier in the season are more closely synchronized with the emergence of insect prey, and parents at these nests may be able to devote more time to nest defence rather than foraging. Many studies have observed a positive correlation between earlier egg-laying dates and increasing temperatures (i.e.,

Crick et al. 1997, Crick and Sparks 1999, Dunn 2004, Torti et al. 2005). If nests that are initiated earlier allow birds to take advantage of earlier insect peak abundances, this could 124 indicate that Rose-breasted Grosbeaks may be flexible enough to adapt to climate change as well as to changes in forest structure resulting from forest management practices.

However, as this species is currently declining across its breeding range, potentially negative effects of climate change, such as more variable weather conditions (IPCC

2002) or changes in the emergence time of insect prey should not be ignored.

When assessing the influence of climate change, habitat loss or degradation, forest management practices, or other factors it is important to study both nest and fledgling survival and habitat requirements. Although single-tree selection harvesting did not significantly alter nest or fledgling survival, habitat characteristics at nests sites and locations used by fledglings were significantly different, and patterns of nest and fledgling survival among the post-harvest treatments were not strongly correlated. This suggests that conclusions regarding the influence of various factors on the growth rates of avian populations should not be based solely on estimates of nest survival.

My study also suggests that population age structure should be taken into consideration in future studies of avian demography and population growth rates. Overall productivity of after-second-year males was significantly higher than that of second-year males, nestling provisioning was more frequent at nests with older, more experienced males, and nestlings weighed more on average at the time of fledging in nests with older males. After-second-year males also contributed more to nestling provisioning and nest attendance than second-year males did. The density of ASY males was significantly higher than the density of SY males in all post-harvest treatments, but if management practices or other factors altered these proportions to favour more SY adults at the expense of ASY breeders, there could be negative consequences for population 125 productivity. Variation in reproductive success between more and less experienced parents should be taken into consideration when investigating the consequences of management practices because all species may not exhibit the same patterns or responses.

There are several areas for improvement in this study. First, the population models I used relied on estimates of annual adult mortality from the literature, which may not have been appropriate for this particular species. Although accurate estimates of annual survival are extremely difficult to obtain, future research should focus on improving these estimates to allow more reliable assessments of population growth rates.

Second, increased sample sizes and a larger number of study sites would have improved the statistical power of my results, and may have provided additional insight into suggested results. Finally, investigations of nestling provisioning rates and considerations of prey availability would have been complimented by direct measurement of insect and fruit abundance.

Future research should focus on areas outside of Ontario, and should aim to identify habitat types and areas where Rose-breasted Grosbeak populations are above replacement levels. Although the population estimates derived in this study suggested

Rose-breasted Grosbeak populations were sinks in most post-harvest treatments, and similar results were reported for forest fragments (Smith et al. 2006, Moore et al. 2010), single-tree selection harvesting did not appear to be having a negative effect on Rose- breasted demography. Future studies could also consider factors that might affect this species during migration and on the wintering grounds, such as weather, climate change, land-use change, and habitat loss or degradation. Long-term studies should also be a priority for future research, to allow a better understanding of temporal variation in nest 126 survival and productivity, and determine how these parameters respond to selection harvesting over the long-term.

This study suggests that single-tree selection harvesting can maintain and even improve habitat for Rose-breasted Grosbeaks in predominantly forested landscapes if it is strictly implemented according the guidelines set out by the Ontario Ministry of Natural

Resources. It should be emphasized that single-tree selection is not equally beneficial for all migratory songbirds, and may even be detrimental to some forest interior species.

Nonetheless, for species such as the Rose-breasted Grosbeak, which benefit from habitat features characteristic of regenerating forests, this system can provide an economically and ecologically sustainable system of timber harvesting. These findings should be of value to municipalities, conservation authorities, landowners, land managers, and consultants involved in assessing or developing forest management plans, silvicultural prescriptions, or stewardship plans. LITERATURE CITED

Ahlering, M.A., D. Arlt, M.G. Betts, and R.J. Fletcher, Jr. 2010. Research needs and

recommendations for the use of conspecific-attraction methods in the

conservation of migratory songbirds. Condor 112: 252-264.

Akresh, M.E., K. Dinse, J. Foufopoulos, S.C. Schubel, and T. Kowalczyk. 2009.

Passerine breeding and post-fledgling habitat use in riparian and upland temperate

forests of the American Midwest. Condor 111: 756-762.

Alatalo, R.V., K. Gottlander, and A. Lundberg. 1988. Conflict or cooperation between

parents in feeding nestlings in the Pied Flycatcher Ficcedula hypoleuca. Ornis

Scandinavica 19: 31-34.

Anders, A.D., D.C. Dearborn, J. Faaborg, and F.R. Thompson III. 1997. Juvenile survival

in a population of Neotropical migrant birds. Conservation Biology 11: 698-707.

Anders, A.D., J. Faaborg, and F.R. Thompson III. 1998. Postfledging dispersal, habitat

use, and home-range size of juvenile Wood Thrushes. Auk 115:349-358.

Anders, A.D., and M.R. Marshall. 2005. Increasing the accuracy of productivity and

survival estimates in assessing landbird population status. Conservation Biology

19: 66-74.

Andrén, H. 1995. Effects of landscape composition on predation rates at habitat edges.

Pages 225-255 in L. Hansson, L. Fahrig, and G. Merriam, eds. Mosaic landscapes

and ecological processes. Chapman and Hall, New York.

Angers, V.A., C. Messier, M. Beaudet, and A. Leduc. 2005. Comparing composition and

structure in old-growth and harvested (selection and diameter limit cuts) northern

hardwood stands in Quebec. Forest Ecology and Management 217: 275-293.

127 128

Annand, E.M. and F.R. Thompson, III. 1997. Forest bird response to regeneration

practices in central hardwood forests. Journal of Wildlife Management 61: 159-

171.

Arbogast, C. 1957. Marking guides for northern hardwoods under the selection system.

U.S. Forest Service Station Paper 56. Lake States Forest Experiment Station, St.

Paul, MN. 20 p.

Arcese, P. and J.N.M. Smith. 1988. Effects of population density and supplemental food

on reproduction in song sparrows. Journal of Animal Ecology 57: 119-136.

Arlt, D. and T. Part. 2007. Nonideal breeding habitat selection: a mismatch between

preference and fitness. Ecology 88: 792-801.

Askins, R.A. 1993. Population trends in grassland, shrubland, and forest birds in eastern

North America. Current Ornithology 11: 1-34.

Askins, R.A., B. Zuckerberg, and L. Novak. 2007. Do the size and landscape context of

forest openings influence the abundance and breeding success of shrubland

songbirds in southern New England? Forest Ecology and Management 250: 137-

147.

Atlegrim, O. and K. Sjoberg. 1996. Effects of clear-cutting and single-tree selection

harvests on herbivorous insect larvae feeding on bilberry (Vaccinium myrtillus) in

uneven-aged boreal Picea abies forests. Forest Ecology and Management 87:

139-148.

Atwell, R.C., L.A. Schulte, and B.J. Palik. 2008. Songbird response to experimental

retention harvesting in red pine (Pinus resinosa) forests. Forest Ecology and

Management 255: 3621-3631. 129

Balbontin, J., I.G. Hermosell, A. Marzal, M. Reviriego, F. de Lope, and A.P. Møller.

2007. Age-related changes in breeding performance in early life associated with

an increase in competence in the migratory barn swallow Hirundo rustica. Journal

of Animal Ecology 76: 915-925.

Barber, D.R., T.E. Martin, and M.A. Melchiors. 2001. Nesting success of birds in

different silvicultural treatments in southeastern U.S. pine forests. Conservation

Biology 15: 196-207.

Bayne, E.M. and K.A. Hobson. 2001. Effects of habitat fragmentation on pairing success

of Ovenbirds: importance of male age and floater behaviour. Auk 118: 380-388.

Bedard, J. and G. LaPointe. 1985. Influence of parental age and season on Savannah

sparrow reproductive success. Condor 87: 106-110.

Bennet, A.F., K. Henein, and G. Merriam. 1994. Corridor use and the elements of

corridor quality: chipmunks and fencerows in a farmland mosaic. Biological

Conservation 68: 155-165.

Berkeley, L.I., J.P. McCarty, and L.L. Wolfenbarger. 2007. Postfledging survival and

movement in Dickcissels (Spiza americana): implications for habitat management

and conservation. Auk 124: 396-409.

Bibby, C.J., N.D. Burgess, D.A. Hill, and S.H. Mustoe. 2000. Bird census techniques. 2nd

edition. Academic Press, San Diego, California.

Blancher, P.J., R.D. Phoenix, D.S. Badzinski, M.D. Cadman, T.L. Crewe, C.M. Downes,

D. Fillman, C.M. Francis, J. Hughes, D.J.T. Hussel, D. Lepage, J.D. McCracken,

D.K. McNicol, B.A. Pond, R.K. Ross, R. Russell, L.A. Venier, and R.C. Weeber. 130

2009. Population trend status of Ontario’s forest birds. Forestry Chronicle

85:184-201.

Bock, C.E., and Z.F. Jones. 2004. Avian habitat evaluation: should counting birds count?

Frontiers in Ecology and the Environment 2: 403-410.

Bohm, S.M. and E.K.V. Kalko. 2009. Patterns of resource use in an assemblage of birds

in the canopy of a temperate alluvial forest. Journal of Ornithology 150:799-814.

Bourque, J. and M.-A. Villard. 2001. Effects of selection cutting and landscape-scale

harvesting on the reproductive success of two neotropical migrant bird species.

Conservation Biology 15: 184-195.

Brawn, J.D. and S.K. Robinson. 1996. Source-sink population dynamics may complicate

the interpretation of long-term census data. Ecology 77: 3-12.

Brawn, J.D., S.K. Robinson, and F.R. Thompson III. 2001. The role of disturbance in the

ecology and conservation of birds. Annual Review of Ecology and Systematics

32: 251-276.

Breitwisch, R., P.G. Merritt, and G.H. Whitesides. 1986. Parental investment by the

Northern Mockingbird: male and female roles in feeding nestlings. Auk 103: 152-

159.

Brooks, R.T. 2003. Abundance, distribution, trends, and ownership patterns of early-

successional forests and native shrublands in the northeastern U.S. Forest Ecology

and Management 185: 65-74.

Brown, W.P. and R.R. Roth. 2009. Age-specific reproduction and survival of individually

marked Wood Thrushes, Hylocichla mustelina. Ecology 90: 218-229.

Burford, J.E., T. J. Friedrich, and K. Yasukawa. 1998. Response to playback of nestling 131

begging in the red-winged blackbird, Agelaius phoeniceus. Animal Behaviour 56:

555-561.

Burhans, D.E., D. Dearborn, F.R. Thompson III, and J. Faaborg. 2002. Factors affecting

predation at songbird nests in old fields. Journal of Wildlife Management 66:240-

249.

Burke, D.M., and E. Nol. 1998. Influence of food abundance, nest-site habitat, and forest

fragmentation on breeding Ovenbirds. Auk 115: 96-104.

Burke, D.M., and E. Nol. 2000. Landscape and fragment size effects on reproductive

success of forest-breeding birds in Ontario. Ecological Applications 10: 1749-

1761.

Burnham, K.P., and D.R. Anderson. 2002. Model selection and multimodel inference: a

practical information-theoretic approach. 2nd Edition. Springer-Verlag, New York,

U.S.A.

Busing, R.T. and P.S.White.1997. Species diversity and small-scale disturbance in an

old-growth temperate forest: a consideration of gap partitioning concepts. Oikos

78: 562-568.

Cadman, M.D., D.A. Sutherland, G.G. Beck, D. Lepage, and A.R. Couturier (eds.). 2007.

Atlas of the Breeding Birds of Ontario 2001-2005. Birds Studies Canada,

Environment Canada, Ontario Field Ornithologists, Ontario Ministry of Natural

Resources, and Ontario Nature, Toronto, Ontario.

Cassamise, D. F., and R. S. Hedin. 1985. An aerodynamic basis for selecting transmitter

loads inbirds. Wilson Bulletin 97:306–318.

Catterall, C.P., J. Kikkawa, and C. Gray. 1989. Inter-related age-dependent patterns of 132

ecology and behaviour in a population of Silvereyes (Aves: Zosteropidae). Journal

of Animal Ecology 58: 557-570.

Catry, P., and R.W. Furness. 1999. The influence of adult age on territorial attendance by

breeding great skuas Catharacta skua: an experimental study. Journal of Avian

Biology 30: 399-406.

Chastel, O., and M. Kersten. 2002. Brood size and body condition in the House Sparrow

Passer domesticus: the influence of brooding behaviour. Ibis 144: 284-292.

Chouteau, P. 2009. Impact of logging on the foraging behaviour of two sympatric species

of Couas (Coua coquereli and Coua gigas) in the western dry forest of

Madagascar. Ecology 332: 567-58.

Cilimburg, A.B., M.S. Lindberg, J.J. Tewksbury, and S.J. Hejl. 2002. Effects of dispersal

on survival probability of adult yellow warblers (Dendroica petechia). Auk 119:

778-789.

Citta, J.J., and M.S. Lindberg. 2007. Nest-site selection of passerines: effects of

geographic scale and public and personal information. Ecology 88: 2034-2046.

Ciudad, C., H. Robles, and E. Matthysen. 2009. Postfledging habitat selection of juvenile

middle spotted woodpeckers: a multi-scale approach. Ecography 32: 676-682.

Cohen, E.B., and C.A. Lindell. 2004. Survival, habitat use, and movements of fledgling

White-throated Robins (Turdus assimilis) in a Costa Rican agricultural landscape.

Auk 121: 404-414.

Colombelli-Négrel, D., and S. Kleindorfer. 2009. Nest height, nest concealment, and

predator type predict nest predation in superb fairy-wrens (Malurus cyaneus).

Ecological Research 24: 921-928. 133

Conrad, K.F., and R.J. Robertson. 1993. Patterns of parental provisioning by Eastern

Phoebes. Condor 95: 57-62.

Cooper, R.J. and R.C. Whitmore. 1990. Arthropod sampling methods in ornithology.

Studies in Avian Biology 13:29-37.

Cooper, R.J., L.A. Wood, J.J. Gannon, and R.R. Wilson. 2009. Effects of timber harvest

and other factors on a floodplain forest indicator species, the Prothonotary

Warbler. Wetlands 29: 574-585.

Cowie, R.J., and S.A. Hinsley. 1988. Feeding ecology of great tits (Parus major) and

blue tits (Parus caeruleus) breeding in suburban gardens. Journal of Animal

Ecology 57: 611-626.

Crick, H.Q.P., C. Dudley, D.E. Glue, and D.L. Thomson. 1997. UK birds are laying eggs

earlier. Nature 388: 526.

Crick, H.Q.P., and T.H. Sparks. 1999. Climate change related to egg-laying trends.

Nature 399: 423.

Crow, T.R., D.S. Buckley, E.A. Nauertz, and J.C. Zasada. 2002. Effects of management

on composition and structure of northern hardwood forests in Upper Michigan.

Forest Science 48: 129-145.

Curio, E. 1983. Why do young birds reproduce less well? Ibis 125: 400-404.

Curlee, A.P., and S.R. Beissinger. 1995. Experimental analysis of mass change in female

Green-rumped Parrotlets (Forpus passerinus): the role of male cooperation.

Behavoural Ecology 2: 192-198.

Daunt, F., S. Wanless, M.P. Harris, and P. Monaghan. 1999. Experimental evidence that 134

age-specific reproductive success is independent of environmental effects.

Proceedings of the Royal Society of London, B: 266: 1489-1493.

DeGraaf, R.M., and M. Yamasaki. 2003. Options for managing early-succssional forest

and shrubland bird habitats in the northeastern United States. Forest Ecology and

Management 185: 179-191.

Della-Bianca, L., and D.E. Beck. 1985. Selection management in southern Appalachian

hardwoods. Southern Journal of Applied Forestry 9: 191-196.

Dellinger, R.L., P.B. Wood, and P.D. Keyser. 2007. Occurrence and nest survival of four

thrush species on a managed central Appalachian forest. Forest Ecology and

Management 243: 248-258.

Desrochers, A. 1992. Age and foraging success in European blackbirds: variation

between and within individuals. Animal Behaviour 43: 85-894.

Dias, P.C., and J. Blondel. 1996. Breeding time, food supply and fitness components of

blue tits Parus caeruleus in Mediterranean habitats. Ibis 138: 644-649.

Doliqez, B., T. Part, E. Danchin, J. Colbert, and L. Gustafsson. 2004. Availability and use

of public information and conspecific density for settlement decisions in the

collard flycatcher. Journal of Animal Ecology 73: 75-87.

Donazar, J.A., and J.E. Feijoo. 2002. Social structure of Andean condor roosts: influence

of sex, age, and season. Condor 104: 832-837.

Donovan, T,M., and C.M. Stanley. 1995. A new method of determining Ovenbird age on

the bases of rectrix shape. Journal of Field Ornithology 66: 247-252.

Donovan, T.M., F.R. Thompson III, J. Faaborg, and J.R. Probst. 1995. Reproductive 135

success of migratory birds in habitat sources and sinks. Conservation Biology 9:

1380-1395.

Doyon, F. D. Gagnon, and J.-F. Giroux. 2005. Effects of strip and single-tree selection

cutting on birds and their habitat in a southwestern Quebec northern hardwood

forest. Forest Ecology and Management 209: 101-115.

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.

Driscoll, M.J.L., T. Donovan, R. Mickey, A. Howard, and K.K. Fleming. 2005.

Determinants of wood thrush nest success: a multi-scale, model selection

approach. Journal of Wildlife Management 69: 699-709.

Duguay, J.P., P.B. Wood, and G.W. Miller. 2000. Effects of timber harvests on

invertebrate biomass and avian nest success. Wildlife Society Bulletin 28:1123-

1131.

Duguay, J.P., P.B. Wood, and J.V. Nichols. 2001. Songbird abundance and avian nest

survival rates in forests fragmented by different silvicultural treatments.

Conservation Biology 15: 1405-1415.

Dunford, W., and K. Freemark. 2005. Matrix matters: effects of surrounding land uses on

forest birds near Ottawa, Canada. Landscape Ecology 20:497-511.

Dunn, P. 2004. Breeding and dates and reproductive performance. In Moller, A., P.

Berthold, and W. Fiedler (eds). Birds and climate change: advances in ecological

research. Elsevier Academic Press.

Espie, R.H.M., L.W. Oliphant, P.C. James, I.G. Warkentin, and D.J. Lieske. 2000. Age- 136

dependent breeding performance in Merlins (Falco columbarius). Ecology 81:

3404-3415.

Falk, K.J., E. Nol, and D.M. Burke. 2011. Weak effect of edges on avian nesting success

in fragmented and forested landscapes in Ontario, Canada. Landscape Ecology

26: 239-251.

Fedy, B.C., and T.E. Martin. 2009. Male songbirds provide indirect parental care by

guarding females during incubation. Behavioural Ecology 20:1034-1038.

Filliater, T.S., R. Breitwisch, and P.M. Nealen. 1994. Predation on Northern Cardinal

nests: does choice of nest site matter? Condor 96: 761-768.

Finegan, B. 1984. Forest succession. Nature 312: 109-114.

Flaspohler, D.J., C.J. F. Huckins, B.R. Rub, and P.J. Van Dusen. 2002. Temporal patterns

in aquatic and avian communities following selective logging in the Upper Great

Lakes Region. Forest Science 48: 339-349.

Fletcher, R.J. Jr. 2007. Species interactions and population density mediate the use of

social cues for habitat selection. Journal of Animal Ecology 76: 598-606.

Fontaine, J.J., and T.E. Martin. 2006. Habitat selection responses of parents to offspring

predation risk: an experimental test. American Naturalist 168: 811-818. de Forest, L.N., and A.J. Gaston. 1996. The effect of age on timing of breeding and

reproductive success in the thick-billed murre. Ecology 77:1501-1511.

Forschler, M.L., and E.K.V. Kalko. 2006. Age-specific reproductive performance in citril

finches Carduelis citrinella. Ardea 94: 275-279.

Forslund, P., and T. Pärt. 1995. Age and reproduction in birds: hypotheses and tests.

Trends in Ecology and Evolution 10: 374-378. 137

Franklin, J.F. 1989. Toward a new forestry. American Forester 95: 37-44.

Freemark, K.E., and H.G. Merriam. 1986. Importance of area and habitat heterogeneity to

bird assemblages in temperate forest fragments. Biological Conservation 36: 115-

141.

Fretwell, S.D. 1972. Populations in a seasonal environment. Princeton, N.J. University

Press.

Fretwell, S.D., and H.L. Lucas, Jr. 1970. On territorial behaviour and other factors

influencing habitat distribution in birds. I. Theoretical Development. Acta

Biotheoretica 19: 16-36.

Fukami, T., and D.A. Wardle. 2005. Long-term ecological dynamics: reciprocal insights

from natural and anthropogenic gradients. Proceedings of the Royal Society B.

272: 2105-2115.

Gardali, T., D.C. Barton, J.D. White, and G.R. Geupel. 2003. Juvenile and adult survival

of Swainson’s Thrush (Catharus ustulatus) in coastal California: annual estimates

using capture-recapture analyses. Auk 120: 1188-1194.

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.

Gibbs, J.P., and J. Faaborg. 1990. Estimating the viability of Ovenbird and Kentucky

Warbler populations in forest fragments. Conservation Biology 4: 193-196.

Giraldeau, L.-A., D. L. Kramer, I. Deslandes, and H. Lair. 1994. The effect of

competitors and distance on central place foraging eastern chipmunks, Tamias

striatus. Animal Behaviour 47: 621-632.

Goodrich, L.J., S.C. Crocoll, and S.E. Senner. 1996. Broad-winged Hawk (Buteo 138

platypterus), The Birds of North America Online (A. Poole, Ed.). Ithaca: Cornell

Lab of Ornithology; Retrieved from the Birds of North America Online:

http://bna.birds.cornell.edu/bna/species/218 doi:10.2173/bna.218.

Gordon, D.M. 1997. The population consequences of territorial behaviour. Trends in

Ecology and Evolution 12:63-66.

Gram, W.K., P.A. Porneluzi, R.L. Clawson, J. Faabory, and S.C. Richter. 2003. Effects

of experimental forest management on density and nesting success of bird species

in Missouri Ozark forests. Conservation Biology 17: 1324-1337.

Greenberg, R. 1980. Demographic aspects of long-distance migration. Pages 493-504 in

A. Keast and E.S. Morton, eds. Migrant birds in the Neotropics: ecology,

behaviour, distribution, and conservation. Smithsonian Institution Press,

Washington, D.C.

Greenberg, C.H., and A. McGrane. 1996. A comparison of relative abundance and

biomass of ground-dwelling arthropods under different forest management

practices. Forest Ecology and Management 89: 31-41.

Greenwood, P.J., and P.H. Harvey. 1982. The natal and breeding dispersal of birds.

Annual Review of Ecology and Systematics 13: 1-21.

Grieco, F. 2001. Short-term regulation of provisioning rate and effect on prey size in the

blue tit, Parus caeruleus. Animal Behaviour 62: 107-116.

Grieco, F. 2002. Time constraint on food choice in provisioning blue tits, Parus

caeruleus: the relationship between feeding rate and prey size. Animal Behaviour

64: 517-526.

Guénette, J.-S., and M.-A. Villard. 2005. Thresholds in forest bird response to habitat 139

alteration as quantitative targets for conservation. Conservation Biology 19: 1168-

1180.

Haas, C.A. 1998. Effects of prior nesting success on site fidelity and breeding dispersal:

an experimental approach. Auk 115: 929-936.

Haché, S., and M.-A. Villard. 2010. Age-specific response of a migratory bird to an

experimental alteration of its habitat. Journal of Animal Ecology 79: 897-905.

Haggerty, T.M. 1992. Effects of nestling age and brood size on nestling care in the

Bachman Sparrow (Aimophila aestivalis). American Midland Naturalist 128: 115-

125.

Halaj, J., C. B. Halpern, and H. Yi. 2008. Responses of litter-dwelling spiders and

carabid beetles to varying levels and patterns of green-tree retention. Forest

Ecology and Management 255: 887-900.

Hammond, H.E.J., D.W. Langor, and J.R. Spence. 2004. Saproxylic beetles (Coleoptera)

using Populus in boreal aspen stands of western Canada: spatiotemporal variation

and conservation of assemblages. Canadian Journal of Forest Research 34: 1-19.

Hannon, S.J., and J.N.M. Smith. 1984. Factors influencing age-related reproductive

success in the willow ptarmigan. Auk 101: 848-854.

Hanski, I.K., T.J. Fenske, and G.J. Niemi. 1996. Lack of edge effect in nesting success of

breeding birds in managed forest landscapes. Auk 113: 578-585.

Hatch, M.I., and D.F. Westneat. 2007. Age-related patterns of reproductive success in

house sparrows Passer domesticus. Journal of Avian Biology 38: 603-611.

Hatchwell, B.J. 1991. An experimental study of the effects of timing of breeding on the 140

reproductive success of common guillemots (Uria aalge). Journal of Animal

Ecology 60: 721-736.

Heise, C.D., and F.R. Moore. 2003. Age-related differences in foraging efficiency, molt,

and fat deposition of Gray Catbirds prior to autumn migration. Condor 105: 496-

504.

Heliolä, J., M. Koivula, and J. Niemelä. 2001. Distribution of carabid beetles (Coleoptera,

Carabidae) across a boreal forest-clearcut ecotone. Conservation Biology 15:

370-377.

Heyborne, W.H., J.C. Miller, and G.L. Parsons. 2003. Ground dwelling beetles and forest

vegetation changes over a 17-year-period, in western Oregon, USA. Forest

Ecology and Management 179: 123-134.

Holloway, G.L., and J.R. Malcolm. 2006. Sciurid habitat relationships in forests managed

under selection and shelterwood silviculture in Ontario. Journal of Wildlife

Management 70: 1735-1745.

Holmes, S.B., D.M. Burke, K.A. Elliott, M.D. Cadman, and L. Friesen. 2004. Partial

cutting of woodlots in an agriculture-dominated landscape: effects on forest bird

communities. Canadian Journal of Forest Research 34: 2467-2476.

Holmes, R.T., P.P. Marra, and T.W. Sherry. 1996. Habitat-specific demography of

breeding black-throated blue warblers (Dendroica caerulescens): implications for

population dynamics. Journal of Animal Ecology 65: 183-195.

Holmes, S.B., and D.G. Pitt. 2007. Response of bird communities to selection harvesting

in a northern tolerant hardwood forest. Forest Ecology and Management 238:

280-292. 141

Holmes, R.T., and S.K. Robinson. 1988. Spatial patterns, foraging tactics, and diets of

ground-foraging birds in a northern hardwood forest. Wilson Bulletin 100: 377-

394.

Holmes, R.T., and J.C. Schultz. 1988. Food availability for forest birds: effects of prey

distribution and abundance on bird foraging. Canadian Journal of Zoology 66:

720-728.

Hood, G. M. 2010. PopTools version 3.2.3. Available on the internet. URL

http://www.poptools.org

Hoover, J.P., M.C. Brittingham, and L.J. Goodrich. 1995. Effects of forest patch size on

nesting success of Wood Thrushes. Aike 112: 146-155.

Hunter, M.L. 1990. Wildlife, forests, and forestry: principles of managing forests for

biological diversity. Prentice-Hall, New Jersey.

Hussell, D.J.T., and T. E. Quinney. 1987. Food abundance and clutch size of tree

swallows Tachycineta bicolour. Ibis 129: 243-258.

Hutto, R.L. 1990. Measuring the availability of food resources. Studies in Avian Biology

13:20-28.

IPCC. 2002. Climate change and biodiversity. Gitay, H. A., R.T. Suarez, and O. Watson

(eds). Technical paper V, IPCC Working Group II, Technical Support Unit.

James, F.C., and H.H. Shugart, Jr. 1970. A quantitative method of habitat description.

Audubon Field Notes 24: 727-736.

Jansen, A. 1990. Acquisition of foraging skills by Heron Island silvereyes Zosterops

lateralis chlorocephala. Ibis 132: 95-101.

Jobes, A.P., E. Nol, and D.R. Voigt. 2004. Effects of selection cutting on bird 142

communities in contiguous eastern hardwood forests. Journal of Wildlife

Management 68: 51-60.

Johnson, E.J., and L.B. Best. 1982. Factors affecting feeding and brooding of Gray

Catbird nestlings. Auk 99: 148-156.

Jongsomjit, D., S.L. Jones, T. Gardali, G.R. Geupel, and P.J. Gouse. 2007. A guide to

nestling development and aging in altricial passerines. U.S. Fish and Wildlife

Service, Biological Technical Publication BTP-R6008-2007.

Keller, L.F., and A.J. van Noordwijk. 1994. Effects of local environmental conditions on

nestling growth in the great tit Parus major L. Ardea 82: 349-362.

Kelty, M.J., D.B. Kittredge, Jr., T. Kyker-Snowman, and A.D. Leighton. 2003. The

conversion of even-aged stands to uneven-aged structure in southern New

England. Northern Journal of Applied Forestry 20: 109-116.

Kenefic, L.S., and R. D. Nyland. 2000. Habitat diversity in uneven-aged northern

hardwood stands: a case study. United States Department of Agriculture, Forest

Service Research Paper NE-714.

Kershner, E.L., J.W. Wal, and R.E. Warner. 2004. Postfledging movements and survival

of juvenile Eastern Meadowlarks (Sturnella magna) in Illinois. Auk 121:1146-

1154.

Kirby, J.S., A.J. Stattersfield, M.I. Evans, R. Grimmett, I. Newton, J. O’Sullivan, and G.

Tucker. 2008. Key conservation issues for migratory birds in the world’s major

flyways. Bird Conservation International 18: 49-73.

Krementz, D.G., J.D. Nichols, and J.E. Hines. 1989. Post-fledging survival of European

Starlings. Ecology 70: 646-655. 143

King, D.I., R.M. Degraaf, M.-L. Smith, and J.P. Buonaccors. 2006. Habitat selection and

habitat-specific survival of fledgling Ovenbirds (Seiurus aurocapilla). Journal of

Zoology 269:414-421.

Kohm, K.A., and J.F. Franklin, eds. 1997. Creating a forestry for the 21st century: the

science of ecosystem management. Island Press, Washington, D.C.

Koivula, M. 2002. Alternative harvesting methods and boreal carabid beetles

(Coleoptera, Carabidae). Forest Ecology and Management 167: 103-121.

Koivula, M., J. Kukkonen, and J. Niemelä. 2002. Boreal carabid-beetle (Coleoptera,

Carabidae) assemblages along the clear-cut originated succession gradient.

Biodiversity and Conservation 11: 1269-1288.

Lang, J.D., L.A. Powell, D.G. Krementz, and M.J. Conroy. 2002. Wood Thrush

movements and habitat use: effects of forest management for red-cockaded

woodpeckers. Auk 119: 109-124.

Latty, E.F., S.M. Werner, D.J. Mladenoff, K.F. Raffa, and T.A. Sickley. 2006. Response

of ground beetle (Carabidae) assemblages to logging history in northern

hardwood-hemlock forests. Forest Ecology and Management 222: 335-347.

Leffelaar, D., and R.J. Robertson. 1986. Equality of feeding roles and the maintenance of

monogamy in Tree Swallows. Behavioural Ecology and Sociobiology 18: 199-

206.

Lent, R.A., and D.E. Capen. 1995. Effects of small-scale habitat disturbance on the

ecology of breeding birds in a Vermont (USA) hardwood forest. Ecography 18:

97-108.

Leps, J., and P. Smilauer. 2003. Multivariate analysis of ecological data using CANOCO. 144

Cambridge University Press, U.K.

Limmer, B., and P.H. Becker. 2009. Improvement in chick provisioning with parental

experience in a seabird. Animal Behaviour 77: 1095-1101.

Lyons, J.E. 2005. Habitat-specific foraging of Prothonotary Warblers: deducing habitat

quality. Condor 107:41-49.

Magrath, R.D. 1991. Nestling weight and juvenile survival in the blackbird, Turdus

merula. Journal of Animal Ecology 60: 335-351.

Mahan, C.G., and R.H. Yahner. 1996. Effects of forest fragmentation on burrow-site

selection by the eastern chipmunk (Tamias striatus). American Midland Naturalist

144:286-296.

Mariani, J.M., and D.A. Manuwal. 1990. Factors influencing Brown Creeper (Certhia

americana) abundance patterns in the southern Washington cascade range.

Studies in Avian Biology 13: 53-57.

Marra, J.L., and R.L. Edmonds. 1998. Effects of coarse woody debris and soil depth on

the density and diversity of soil invertebrates on clearcut and forested sites on the

Olympic Peninsula, Washington. Environmental Entomology 27: 1111-1124.

Marshall, M.R., R.J. Cooper, J.A. DeCecco, J. Strazanac, and L. Butler. 2002. Effects of

experimentally reduced prey abundance on the breeding ecology of the Red-Eyed

Vireo. Ecological Applications 12:261-280.

Martikainen, P., J. Siitonen, P. Punttila, L. Kaila, and J. Rauh. 2000. Species richness of

Coleoptera in mature managed and old-growth boreal forests in southern Finland.

Biological Conservation 94: 199-209.

Martin, T.E. 1986. Competition in breeding birds: on the importance of considering 145

processes at the level of the individual. Current Ornithology 4: 181-210.

Martin, T.E. 1992. Interaction of nest predation and food limitation in reproductive

strategies. Current Ornithology 9: 163-197.

Martin, K. 1995. Patterns and mechanisms for age-dependent reproduction and survival

in birds. American Zoologist 35: 340-348.

Martin, T.E., and J.J. Roper. 1988. Nest predation and nest-site selection of a western

population of the Hermit Thrush. Condor 90: 51-57.

Marshall, M.R., J.A. DeCecco, A.B. Williams, G.A. Gale, and R.J. Cooper. 2003. Use of

regenerative clearcuts by late-successional bird species and their young during the

post-fledging period. Forest Ecology and Management 183: 127-135.

Mayfield, H.F. 1961. Nesting success calculated from exposure. Wilson Bulletin 73: 255-

261.

Mayfield, H.R. 1975. Suggestions for calculating nest success. Wilson Bulletin 87: 456-

466.

Maynard-Smith, J. 1977. Parental investment - a prospective analysis. Animal

Behaviour 25: 1-9.

McAtee, W.L. 1908. Food habits of the grosbeaks. USDA Biological Survey Bulletin 32.

Merrill, S.B., F.J. Cuthbert, and G. Oehlert. 1998. Residual patches and their contribution

to forest-bird diversity on northern Minnesota aspen clearcuts. Conservation

Biology 12: 190-199.

Middleton, H.A., and D.J. Green. 2008. Correlates of postfledging survival, the timing of

dispersal, and local recruitment in American Dippers. Canadian Journal of

Zoology 86: 875-881. 146

Mitchell, G.W., P.D. Tayor, and I.G. Warkentin. 2010. Multiscale postfledging habitat

associations of juvenile songbirds in a managed landscape. Auk 127: 354-363.

Mitrus, C., J. Mitrus, and M. Sikora. 2010. Sex differences in the rate of food

provisioning to nestling red-breasted flycatchers (Ficedula parva). Annales

Zoologici Fennici 47: 144-148.

Møller, A.P. 1991. Clutch size, nest predation, and distribution of avian unequal

competitors in a patch environment. Ecology 72: 1336-1349.

Monrós, J.S., E.J. Belda, and E. Barba. 2002. Post-fledging survival of individual Great

Tits: the effect of hatching date and fledgling mass. Oikos 99: 481-488.

Moore, J., R. Ouimet, C. Camire, and D. Houle. 2002. Effects of two silvicultural

practices on soil fauna abundance in a northern hardwood forest, Quebec, Canada.

Canadian Journal of Soil Science 82: 105-113.

Moore, L.C., B.J.M. Stutchbury, D.M. Burke, and K.A. Elliott. 2010. Effects of forest

management on postfledging survival of Rose-breasted Grosbeaks (Pheucticus

ludovicianus). Auk 127: 185-194.

Morehouse, E.L., and R. Brewer. 1968. Feeding of nestling and fledgling Eastern

Kingbirds. Auk 85: 44-45.

Morton, E.S. 2005. Predation and variation in breeding habitat use in the Ovenbird, with

special reference to breeding habitat selection in northwestern Pennsylvania.

Wilson Bulletin 117: 327-335.

Morton, M.L., M.W. Wakamatsu, M.E. Pereyra, and G.A. Morton. 1991. Postfledging

dispersal, habitat imprinting, and philopatry in a montane, migratory sparrow.

Ornis Scandinavica 22: 98-106. 147

Naef-Daenzer, L., B. Naef-Daenzer, and R.G. Nager. 2000. Prey selection and foraging

performance of breeding Great Tits Parus major in relation to food availability.

Journal of Avian Biology 31:206-214.

Naef-Daenzer, L., F. Widmer, and M. Nuber. 2001. Differential post-fledgling survival of

Great and Coal tits in relation to their condition and fledging date. Journal of

Zoology (London) 269: 317-330.

Nagy, L.R., and R.T. Holmes. 2005. Food limits annual fecundity of a migratory

songbird: an experimental study. Ecology 86:675-681.

Nagy, L.R., and K.G. Smith. 1997. Effects of insecticide-induced reduction in

lepidopteran larvae on reproductive success of Hooded Warblers. Auk 114:619-

627.

Naka, L.N. 2003. Structure and organization of canopy bird assemblages in Central

Amazonia. Auk 121:88-102.

Natural Resources Canada. 2010. The state of Canada’s forests. Annual report 2010.

Natural Resources Canada, Forestry Service, Headquarters, Ottawa.

Newman, J.A., and T. Caraco. 1988. Foraging, predation hazard and patch use in grey

squirrels. Animal Behaviour 35: 1804-1813.

Newton, I. 2004. Population limitation in migrants. Ibis 146:197-226.

Niemelä, J., D. Langor, and J.R. Spence. 1993. Effects of clear-cut harvesting on boreal

ground-beetles assemblages (Coleoptera: Carabidae) in western Canada.

Conservation Biology 7: 551-561.

Nol, E., H. Douglas, and W.J. Crins. 2006. Responses of Syrphids, Elaterids and bees to

single-tree selection harvesting in Algonquin Provincial Park, Ontario. Canadian 148

Field Naturalist 120: 15-21.

Nol, E., and J.N.M. Smith. 1987. Effects of age and breeding experience on seasonal

reproductive success in the Song Sparrow. Journal of Animal Ecology 56: 301-

313.

Nordby, J.C., A.N. Cohen, and S.R. Beissinger. 2010. Effects of a habitat-altering

invader on nesting sparrows: an ecological trap? Biological Invasions 11: 565-

575.

Norment, C.J. 2003. Patterns of nestling feeding in Harris’s Sparrows, Zonotrichia

querula and White-crowned Sparrows, Z. leucophrys, in the Northwest

Territories, Canada. Canadian Field-Naturalist 117: 203-208.

OMNR. 1998. A silvicultural guide for the tolerant hardwood forest in Ontario. Queen’s

Printer for Ontario, Ontario, Canada.

OMNR. 2000. A silvicultural guide to managing southern Ontario forests, Version 1.1

Ontario Ministry Natural Resources. Queen’s Printer for Ontario, Toronto.

Ontario Parks. 1998. Algonquin Provincial Park Management Plan. Queen’s Printer for

Ontario, Ontario, Canada.

Orians, G.H., and N.E. Pearson. 1979. On the theory of central place foraging. In:

Analysis of Ecological Systems (Eds. Horn, D.J., R. Mitchell, and G.R. Stair), pp.

155-177. Ohio State University Press, Columbus, Ohio.

Ottosson, U., J. Bäckman, and J.G. Smith. 1997. Begging affects parental effort in the

pied flycatcher, Ficedula hypoleuca. Behavioural Ecology and Sociobiology 41:

381-384.

Pagen, R.W., F.R. Thompson III, and D.E. Burhans. 2000. Breeding and post-breeding 149

habitat use by forest migrant songbirds in the Missouri Ozarks. Condor 102: 738-

747.

Paradis, E., S.R. Braille, W.J Sutherland, and R.D. Gregory. 1998. Patterns of natal and

breeding dispersal in birds. Journal of Animal Ecology 67: 518-536.

Pärt, T. 2001. Experimental evidence of environmental effects on age-specific

reproductive success: the importance of resource quality. Proceedings of the

Royal Society of London, Series B 268: 2267-2271.

Pärt, T. 2001a. The effects of territory quality on age-dependent reproductive

performance in the northern wheatear (Oenanthe oenanthe). Animal Behaviour

62: 379-388.

Pärt, T., D. Arlt, and M. Villard. 2007. Empirical evidence for ecological traps: a two-

step model focusing on individual decisions. Journal of Ornithology 148:S327-

S332.

Paton, P.W.C. 1994. The effect of edge on avian nest success: how strong is the

evidence? Conservation Biology 8: 17-26.

Payne, R.B. 1991. Natal dispersal and population structure in a migratory songbird, the

Indigo Bunting. Evolution 45: 49-62.

Pelletier, R., and D. Dauphin. 1996. The breeding birds of Quebec. Pages 954-957 in

Gauthier, J. and Y. Aubry, eds. The breeding birds of Quebec: atlas of the

breeding birds of southern Quebec. Canadian Wildlife Service, Environment

Canada, Montreal, Quebec.

Perera, A.H., D.L. Euler, and I.D. Thompson, eds. 2000. Ecology of a managed terrestrial 150

landscape: patterns and processes of forest landscapes in Ontario. UBC Press,

Vancouver, B.C.

Perot, A., and M.-A. Villard. 2009. Putting density back into the habitat-quality equation:

case study of an open-nesting forest bird. Conservation Biology 23: 1550-1557.

Pettersson, R.B., J.P. Ball, K.E. Renhorn, P.A. Esseen, and K. Sjoberg. 1995. Invertebrate

communities in boreal forest canopies as influenced by forestry and lichens with

implications for passerine birds. Biological Conservation 74: 57-63.

Phillips, J., E. Nol, D. Burke, and W. Dunford. 2005. Impacts of housing developments

on Wood Thrush nesting success in hardwood forest fragments. Condor 107: 97-

106.

Pihlaja, M., M. Koivula, and J. Niemelä. 2006. Response of boreal carabid beetle

assemblages (Coleoptera, Carabidae) to clear-cutting and top-soil preparation.

Forest Ecology and Management 222: 182-190.

Porneluzi, P.A., and J. Faaborg. 1999. Season-long fecundity, survival, and viability of

ovenbirds in fragmented and unfragmented landscapes. Conservation Biology 13:

1151-1161.

Possardt, E.E., and W.E. Dodge. 1978. Stream channelization impacts on songbirds and

small mammals in Vermont. Wildlife Society Bulletin 6: 18-24.

Poulin, J.-F., M.-A. Villard, and S. Haché. 2010. Short-term demographic response of an

old forest specialist to experimental selection harvesting. Ecoscience 17: 20-27.

Poulson, T.L., and W.J. Platt. 1989. Gap light regimes influence canopy tree diversity.

Ecology 70: 553-555.

Powell, L.A., M.J. Conroy, D.G. Krementz, and J.D. Lang. 1999. A model to predict 151

breeding-season productivity for multibrooded songbirds. Auk 116: 1001-1008.

Powell, L.A. J.D. Lang, M.J. Conroy, and D.G. Krementz. 2000. Effects of forest

management on density, survival, and population growth of wood thrushes.

Journal of Wildlife Management 64: 11-23.

Préault, M., O. Chastel, and F. Cézilly. 2005. Male bill colour and age are associated with

parental abilities and breeding performance in blackbirds. Behavioural Ecology

and Sociobiology 58: 497-505.

Preston, K.L., and R.T. Rotenberry. 2006. Independent effects of food and predator-

mediated processes on annual fecundity in a songbird. Ecology 87:160-168.

Price, T. 1985. Reproductive responses to varying food supply in a population of

Darwin’s finches: clutch size, growth rates and hatching synchrony. Oecologia

66: 441-416.

Prokop, P. 2004. The effect of nest usurpation on breeding success of the black-billed

magpie (Pica Pica). Biologia 59: 213-217.

Pulliam, H.R. 1988. Sources, sinks, and population regulation. American Naturalist 132:

652-661.

Pyle, P. 1997. Identification guide to North American birds, part 1. Slate Creek Press,

Bolins, California. Pp. 609-611.

Rappole, J.H., and K. Ballard. 1987. Post-breeding movements of selected species of

birds in Athens, Georgia. Wilson Bulletin 99: 475-480.

Rappole, J. H., and A. Tipton. 1991. New harness design for attachment of radio

transmitters to small passerines. J. Field Ornith. 62:335–337.

Ratcliffe, N., R.W. Furness, and K.C. Hamer. 1998. The interactive effects of age and 152

food supply on the breeding ecology of great skuas. Journal of Animal Ecology

67: 853-862.

Raymond, P., A.D. Munson, J-C. Ruel, and J.D. Coates. 2006. Spatial patterns of soil

microclimate, light, regeneration, and growth within silvicultural gaps of mixed

tolerant hardwood – white pine stands. Canadian Journal of Forest Research 36:

639-651.

Reid, J.M., G.D. Ruxton, P. Monaghan, and G.M. Hilton. 2002. Energetic consequences

of clutch temperature and clutch size for a uniparental intermittent incubation: the

starling. Auk 119: 54-61.

Remes, V. 2003. Effects of exotic habitat on nesting success, territory density, and

settlement patterns of the Blackcap. Conservation Biology 17: 1127-1133.

Reynolds, J. 1997. Body condition, territory ownership and age-related reproductive

performance in Spruce Grouse Dendragapus Canadensis hens. Ibis 139: 646-651.

Ricklefs, R.L. 1973. Fecundity, mortality, and avian demography. Pages 366-435 in

Breeding biology of birds (D.S. Farner, Ed.). National Academy of Science,

Philadelphia, Pennsylvania, USA.

Robertson, B.A. 2009. The influence of spatio-temporal variation in food availability and

nest-predation risk on clutch-size decisions. Condor 111:523-533.

Robertson, B.A., and R.L. Hutto. 2006. A framework for understanding ecological traps

and an evaluation of existing evidence. Ecology 87: 1075-1085.

Robertson, B.A., and R.L. Hutto. 2007. Is selectively harvested forest an ecological trap

for Olive-sided flycatchers? Condor 109: 109-121.

Robinson, S.K., and R.T. Holmes. 1982. Foraging behavior of forest birds: the 153

relationships among search tactics, diet and habitat structure. Ecology 63:1918-

1931.

Robinson, W.D., and S.K. Robinson. 1999. Effects of selective logging on forest bird

populations in a fragmented landscape. Conservation Biology 13: 58-66.

Robinson, S.K., and W.D. Robinson. 2001. Avian nesting success in a selectively

harvested north temperate deciduous forest. Conservation Biology 15: 1763-1771.

Rodenhouse, N. L., and R.T. Holmes. 1992. Results of experimental and natural food

reductions for breeding Black-throated Blue Warblers. Ecology 73:357-372.

Rodewald, A.D. 2002. Nest predation in forested regions: landscape and edge effects.

Journal of Wildlife Management 66: 634-640.

Rowe, J.S. 1972. Forest Regions of Canada. Publication no. 1300. Canadian Forestry

Service, Department of the Environment, Ottawa.

Rush, S.A., and B.J.M. Stutchbury. 2008. Survival of fledgling Hooded Warblers

(Wilsonia citrina) in small and large forest fragments. Auk 125: 183-191.

Rutz, C., M.J. Whittingham, and I. Newton. 2006. Age-dependent diet choice in an avian

top predator. Proceedings of the Royal Society of London, Series B. 273: 579-

586.

Sallabanks, R., E.B. Arnett, and J.M. Marzluff. 2000. An evaluation of research on the

effects of timber harvest on bird populations. Wildlife Society Bulletin 28: 1144-

1155.

Sanz, J.J., and J. Moreno. 1995. Mass loss in brooding female Pied Flycatcher Ficedula

hypoleuca: no evidence for reproductive stress. Journal of Avian Biology 26: 313-

320. 154

Sanz, J.J., and J.M. Tinbergen. 1999. Energy expenditure, nestling age, and brood size: an

experimental study of parental behaviour in the Great Tit Parus major.

Behavioural Ecology 10: 598-606.

SAS Institute. 2001. SAS Institute Inc., Help and Documentation,Version 8.02. SAS

Institute Inc., Cary, N.C.

Sauer, J.R., J.E. Hines, and J. Fallon. 2008. The North American breeding bird survey,

results and analysis 1966-2007. Version 5.15.2008. U.S. Geological Survey

Patuxent Wildlife Research Center, Laurel, M.D.

Schaefer, R.R., R.N. Conner, D.C. Rudolph, and D. Saenz. 2004. Red-Cockaded

Woodpecker nestling provisioning and reproduction in two different pine habitats.

Wilson Bulletin 116: 31-40.

Schmidt, K.A., S.A. Rush, and R.S. Ostfeld. 2008. Wood Thrush nest success and post-

fledging survival across a temporal pulse of small mammal abundance in an oak

forest. Journal of Animal Ecology 77: 830-837.

Schmiegelow, F.K.A., C.S. Machtans, and S.J. Hannon. 1997. Are boreal birds resilient

to forest fragmentation? An experimental study of short-term community

responses. Ecology 78: 1914-1932.

Scott, D.M. 1998. Laying hours and other nestling data of Rose-breasted Grosbeaks.

Ontario Birds 16: 88-93.

Sejberg, D., S. Bensch, and D. Hasselquist. 2000. Nestling provisioning in polygynous

great reed warblers (Acrocephalus arundinaceus): do males bring larger prey to

compensate for fewer nest visits? Behavioural Ecology and Sociobiology 47:213-

219. 155

Seppänen, J.T., J.T. Forsman, M. Monkkӧnen, and R.L. Thompson. 2007. Social

information use is a process across time, space, and ecology, reaching

heterospecifics. Ecology 88: 1622-1633.

Seymour, R.S., A.S. White, and P.G. deMaynadier. 2002. Natural disturbance regimes in

northeastern North America – evaluating silviculture systems using natural scales

and frequencies. Forest Ecology and Management 155: 357-367.

Shaffer, T.L. 2004. A unified approach to analyzing nest success. Auk 121: 526-540.

Sherry, T.W., and R.T. Holmes. 1989. Age-specific social dominance affects habitat use

by breeding American redstarts (Stephaga ruticilla): a removal experiment.

Behavioural Ecology and Sociobiology 25: 327-333.

Shields, J.M., C.R. Webster, and L.M. Nagel. 2007. Factors influencing tree species

diversity and Betula alleghaniensis establishment in silvicultural openings.

Forestry 80: 293-307.

Shields, J.M., C.R. Webster, and A.J. Storer. 2008. Short-term community-level response

of arthropods to group selection with seed-tree retention in a northern hardwood

forest. Forest Ecology and Management 255: 129-139.

Siira-Pietikäinen, A., J. Haimi, and J. Siitonen. 2003. Short-term responses of soil

macroarthropod community to clear felling and alternative forest regeneration

methods. Forest Ecology and Management 172: 339-353.

Sillett, T.S., and R.T. Holmes. 2002. Variation in survivorship of a migratory songbird

throughout its annual cycle. Journal of Animal Ecology 71: 296-308.

Simon, N.P.P., F.E. Schwab, and A.W. Diamond. 2000. Patterns of breeding bird 156

abundance in relation to logging in western Labrador. Canadian Journal of Forest

Research 30: 257-263.

Smith, L.A., D.M. Burke, E. Nol, and K.A. Elliott. 2006. The effects of partial cutting on

the Rose-breasted Grosbeak: abundance, food availability, and nest survival.

Canadian Journal of Forest Research 36: 1087-1096.

Smith, L.A., E. Nol, D.M. Burke, and K.A. Elliot. 2007. Nest-site selection of Rose-

breasted Grosbeaks in southern Ontario. Wilson Journal of Ornithology 119:151-

160.

Smith, H.G., H. Källander, K. Fontell, and M. Ljungstrom. 1988. Feeding frequency and

parental division of labour in the double-brooded Great Tit Parus major. Effects

of manipulating brood size. Behavioural Ecology and Sociobiology 22: 447-453.

Solomon, D.S. and J.H.Grove. 1999. Effects of uneven-age management intensity on

structural diversity in two major forest types in New England. Forest Ecology and

Management 114: 265-274.

Spanhove, T., V. Lehouck, P. Boets, and L. Lens. 2009. Forest fragmentation relaxes

natural nest predation in an Afromontane forest. Animal Conservation 12: 267-

275.

Stodola, K.W., E.T. Linder, D.A. Buehler, K.E. Franzreb, D.H. Kim, and R.J. Cooper.

2010. Relative influence of male and female care in determining nestling mass in

a migratory songbird. Journal of Avian Biology 41: 515-522.

Strayer, D., D.H. Pletscher, and S.P. Hamburg. 1986. The effects of forest disturbance on

land gastropod communities in northern New England. Canadian Journal of

Zoology 64: 2094-2098. 157

Stutchbury, B.J., and R.J. Robertson. 1988. Within-season and age-related patterns of

reproductive performance in female tree swallows (Tachycineta bicolour).

Canadian Journal of Zoology 66: 827-834.

Stutchbury, B.J.M., W.H. Piper, D.I. Neudorf, S.A. Tarof, J.M. Rhymer, G. Fuller, and

R.C. Fleischer. 1997. Correlates of extra-pair fertlilization success in Hooded

Warblers. Behavioural Ecology and Sociobiology 40: 119-126.

Stutchbury, B.J.M., B. Capuano, and G.S. Fraser. 2005. Avian frugivory on a gap-

specialist, the red elderberry (Sambucus racemosa). Wilson Bulletin 117: 336-

340.

Stutchbury, B.J.M., J.R. Hill III, P.M. Kramer, S.A. Rush, and S.A. Tarof. 2009. Sex and

age-specific annual survival in a Neotropical migratory songbird, the purple

martin (Progne subis). Auk 126: 278-287.

Sullivan, K.A. 1989. Predation and starvation: age-specific mortality in juvenile juncos

(Junco phaenotus). Journal of Animal Ecology 58: 275-286.

Sydeman, W.J., J.F. Penniman, T.F. Penniman, P. Pyle, and D.G. Ainley. 1991. Breeding

performance in the Western gull, effect of parental age, timing of breeding and

year in relation to food supply. Journal of Animal Ecology 60: 135-149.

Temple, S.A., and J.R. Carey. 1988. Modeling dynamics of habitat-interior bird

populations in fragmented landscapes. Conservation Biology 2: 340-347.

Ter Braak, C.J.F., and P. Smilauer. 1998. Canoco reference manual and user’s guide to

Canoco for Windows: software for canonical community ordination (version 4).

Microcomputer Power Ithaca, New York, U.S.A.

Thompson, I.D. 2000. Forest vegetation of Ontario: factors influencing landscape change. 158

Pages 30-53 in A.H. Perera, D.L. Euler, and I.D. Thompson, eds. Ecology of a

managed terrestrial landscape. University of British Columbia Press, Vancouver,

B.C.

Thompson, F.R. III, J.R. Probst, and M.G. Raphael. 1995. Impacts of silviculture:

overview and management recommendations. Pages 201-219 in T.E. Martin and

D.M Finch, eds. Ecology and management of neotropical migratory birds. Oxford

University Press, New York, N.Y.

Thompson, F.R. III, J.D. Brawn, S. Robinson, J. Faaborg, and R.L. Clawson. 2000.

Approaches to investigate effects of forest management on birds in eastern

deciduous forests: how reliable is our knowledge? Wildlife Society Bulletin 28:

111-1122.

Thompson, I.D., J.A. Baker, S.J. Hannon, R.S. Rempel, and K.J. Szuba. 2009. Forest

birds and forest management in Ontario: status, management, and policy. Forestry

Chronicle 85: 245-257.

Thomson, D.L., P. Monaghan, and R.W. Furness. 1998. The demands of incubation and

avian clutch size. Biological Reviews of the Cambridge Philosophical Society 73:

293-304.

Torti, V.M., P. Dunn, and R. Scott. 2005. Variable effects of climate change on six

species of North American birds. Oecologia 145: 486.

Tozer, R., and D. Strickland. 1980. A pictorial history of Algonquin Provincial Park.

Ontario Ministry of Natural Resources, Ontario, Canada.

Trani, M.K., R.T. Brooks, T.L. Schmidt, V.A. Rudis, and C.M. Gabbard. 2001. Patterns 159

and trends in early-successional forests in the eastern United States. Wildlife

Society Bulletin 29: 413-424.

Tubbs, C.H. 1977. Manager’s handbook for northern hardwoods in the north central

states. Forest Service, U.S. Department of Agriculture, Minnesota, U.S.A.

Underwood, T.J., and R.R. Roth. 2002. Demographic variables are poor indicators of

Wood Thrush productivity. Condor 104: 92-102.

Ulyshen, M.D., J.L. Hanula, S. Horn, J.C. Kilgo, and C.E. Moorman. 2004. Spatial and

temporal patterns of beetles associated with coarse woody debris in managed

bottomland hardwood forests. Forest Ecology and Management 199: 259-272.

Vance, C.C. and E. Nol. 2003. Temporal effects of selection logging on ground beetle

communities in northern hardwood forests of eastern Canada. Ecoscience 10: 49-

56.

Vanderwel, M.C., J.P. Caspersen, and M.E. Woods. 2006. Snag dynamics in partially

harvested and unmanaged northern hardwood forests. Canadian Journal of Forest

Research 36: 2769-2779.

Vanderwel, M.C., J.R. Malcolm, and S.C. Mills. 2007. A meta-analysis of bird responses

to uniform partial harvesting across North America. Conservation Biology 21:

1230-1240.

Vanderwel, M.C., S.C. Mills, and J.R. Malcolm. 2009. Effects of partial harvesting on

vertebrate species associated with late-successional forests in Ontario’s boreal

region. Forestry Chronicle 85: 91-104.

Van Horne, B. 1983. Density as a misleading indicator of habitat quality. Journal of

Wildlife Management 47: 893-901. 160

Van Horn, M.A., R.M. Gentry, and J. Faaborg. 1995. Patterns of pairing success of the

Ovenbird (Seiurus aurocapillus) within Missouri forest fragments. Auk 112: 98-

106.

Vega Rivera, J.H., J.H. Rappole, W.J. McShea, and C.A. Haas. 1998. Wood Thrush

postfledging movements and habitat use in Northern Virginia. Condor 100:69-78.

Vega Rivera, J.H., W.J. McShea, J.H. Rappole, and C.A. Haas. 1999. Postbreeding

movements and habitat use of adult Wood Thrushes in northern Virginia. Auk

116: 458-466.

Vickery, P.D., M.L. Hunter, Jr., and J.V. Wells. 1992. Is density an indicator breeding

success? Auk 109: 706-710.

Villard, M.-A., P.R. Martin, and C.G. Drummond. 1993. Habitat fragmentation and

pairing success in the Ovenbird (Seiurus aurocapillus). Auk 110: 759-768.

Vitz, A.C., and A.D. Rodewald. 2006. Can regenerating clearcuts benefit mature-forest

songbirds? An examination of post-breeding ecology. Biological Conservation

127:477-486.

Vitz, A.C., and A.D. Rodewald. 2007. Vegetative and fruit resources as determinants of

habitat use by mature-forest birds during the postbreeding period. Auk 124: 494-

507.

Vitz, A.C., and A.D. Rodewald. 2010. Movements of fledgling Ovenbirds (Seiurus

aurocapilla) and Worm-eating Warblers (Helmitheros vermivorum) within and

beyond the natal home range. Auk 127: 364-371.

WDNR (Wisconsin Department of Natural Resources). 2011. Wisconsin Forest

Management Guidelines PUB-FR-226 2011. Available online at: 161

http://dnr.wi.gov/forestry/publications/guidelines/PDF

Weathers, W.W., and K.A. Sullivan. 1989. Juvenile foraging proficiency, parental effort,

and avian reproductive success. Ecological Monographs 59: 223-246.

Webb, W.L., D.F. Behrend, and B. Saisorn. 1977. Effect of logging on songbird

populations in a northern hardwood forest. Wildlife Monographs 55:1-35.

Webster, C.R., and C.G. Lorimer. 2002. Single-tree versus group selection in hemlock-

hardwood forests: are smaller openings less productive? Canadian Journal of

Forest Research 32: 591-604.

Weldon, A.J., and N.M. Haddad. 2005. The effects of patch shape on Indigo Buntings:

evidence for an ecological trap. Ecology 86: 1422-1431.

Welsh, C.J.E., and W.M. Healy. 1993. Effects of even-aged timber management

on bird species diversity and composition in northern hardwoods of New

Hampshire. Wildlife Society Bulletin 21: 143-154.

Werner, S.M., and K.F. Raffa. 2000. Effects of forest management practices on the

diversity of ground-occurring beetles in mixed northern hardwood forests of the

Great Lakes Region. Forest Ecology and Management 139: 135-155.

Westneat, D.F. 1988. Male parental care and extrapair copulations in the Indigo Bunting.

Auk 105: 149-160.

Wheelwright, N.T., and C.B. Schultz. 1994. Age and reproduction in Savannah sparrows

and tree swallows. Journal of Animal Ecology 63: 686-702.

White, G.C., and K.P. Burnham. 1999. Program MARK: Survival estimation from

populations of marked animals. Bird Study 46 (Supplement): S120-S139.

White, J.D., T. Gardali, F.R. ThompsonIII, and J. Faaborg. 2005. Resource selection by 162

juvenile Swainson’s Thrushes during the post-fledging period. Condor 107:388-

401.

Winkler, D.W. 1987. A general model for parental care. American Naturalist 130: 526-

543.

Wolf, L., E.D. Ketterson, and V. Nolan, Jr. 1990. Behavioural response of female dark-

eyed juncos to the experimental removal of their mates: implications for the

evolution of male parental care. Animal Behaviour 39: 125-134.

Woodworth, B.L. 1999. Modeling population dynamics of a songbird exposed to

parasitism and predation and evaluating management options. Conservation

Biology 13: 67-76.

Wunderle, J.M. 1991. Age-specific foraging proficiency in birds. Current Ornithology 8:

273-324.

Wyatt, V.E., and C.M. Francis. 2002. Rose-breasted Grosbeak (Pheucticus ludovicianus).

In The Birds of North America. The Birds of North America, Inc., Philadelphia,

Pa.

Yackel Adams, A.A., S.K. Skagen, and J.A. Savidge. 2006. Modeling post-fledging

survival of Lark Buntings in response to ecological and biological factors.

Ecology 87: 178-188.

Yasukawa, K. 1981. Male quality and female choice of mate in the red-winged blackbird

(Agelaius phoeniceus). Ecology 62: 922-929.

Yu, X.D., T.H. Luo, J. Yang, and H.Z. Zhou. 2006. Distribution of ground-dwelling

beetles (Coleoptera) across a forest-clearcut ecotone in Wolong Natural Reserve,

southwestern China. Insect Science 13: 127-135. 163

Zanette, L., M. Clinchy, and J.N.M. Smith. 2006. Food and predators affect egg

production in Song Sparrows. Ecology 87:2459-2467.

Zollner, P.A., and K.J. Crane. 2003. Influence of canopy closure and shrub coverage on

travel along coarse woody debris by eastern chipmunks (Tamias striatus).

American Midland Naturalist 150: 151-157.