. a. f. .1.. 5.: W3....:. 1.. ..L .2. ; a... . .wu. ; . A 'QJI- . a9». . J p. 1'... «Waifia .1.. .. .- . .1. 2.... .umfipum... . .1 *1? a. .91 ENE." ) a {a 51L . ; 1:)“. r . I» .. 5:... .. s 1.31:3- www.mfi%u . . .2. 5.1.. r311... . . . “41.1.1.4. 1m ........ ..L . . . . . . . . .. .. . 4.31.... .3“, L.?I . . . 1....th u......!x..m...nr .wuwfi .11... 1 .2 xx“. . I. . .4 1.... .51... man". 1 .. .md. vs... .n... -. . .1. This is to certify that the thesis entitled EFFECT OF PROXIMITY TO FOREST EDGES ON NESTLING GROWTH AND NEST SURVIVAL OF WOOD THRUSH (HYLOCICHLA MUSTELINA) IN SOUTHWESTERN MICHIGAN presented by Sara Ann Kaiser has been accepted towards fulfillment of the requirements for the Master of degree in Zoology and the Science Ecology, Evolutionary Biology and Behavior Fromm \ .4 v~.~ Major Professor’s Signature 4*Foi Date MSU is an Affirmative Action/Equal Opportunity Institution - .— ———v—‘ - ————_.~ l LIBRARY Michigan Btate Universfly PIACE IN RETURN BOX to remove this checkout from your record. TO AVOID FINES return on or before date due. MAY BE RECALLED with earlier due date if requested. DATE DUE DATE DUE DATE DUE SEEP C '3 2008 010709 6/01 c:/ClBC/DateDue.p65-p. 15 EFFECT OF PROXIMITY TO FOREST EDGES ON NESTLING GROWTH AND NEST SURVIVAL OF WOOD THRUSH (HYLOCICHLA MUST ELINA) IN SOUTHWESTERN MICHIGAN By Sara Ann Kaiser A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE Department of Zoology Ecology, Evolutionary Biology and Behavior Program 2004 ABSTRACT EFFECT OF PROXIMITY To FOREST EDGES ON NESTLING GROWTH AND NEST SURVIVAL OF WOOD THRUSH (HYLOCICHLA MUSTELINA) IN SOUTHWESTERN MICHIGAN By Sara Ann Kaiser Nestling growth rates have been used as indicators of habitat degradation because growth rates have been positively correlated with food. However, no previous work has examined growth rates in the context of forest fragmentation. I investigated whether distance to forest edge influenced growth rates of wings, tarsi, and mass of nestling Wood Thrush. 1 also examined whether extrinsic and intrinsic factors explained variability in nestling growth rates. I also report Mayfield’s daily nest survival rates, nest survival rates, and brood parasitism rates in edge and interior forest. The study was conducted in two forested landscapes in southwestern Michigan from May to August in 2002 and 2003. I located 175 Wood Thrush nests and measured nestlings from 58 nests. Tarsi growth rates were most rapid near powerline corridors and recent clearcuts, which may be food-abundant, and mass growth rates increased with total precipitation. Rapid tarsi grth may also be explained by the greater functional role of the tarsus early in the nestling period. There was no relationship between distance to forest edge and nest survival rates. Brood parasitism rates were among the lowest reported for Wood Thrush nesting in Midwestern fragmented forests. My finding that proximity to edge influenced nestling growth rates but not nesting success suggests nestling growth rates may be better indicators of habitat degradation than nesting success when high regional fragmentation levels may overwhelm potential edge-interior differences in local predation patterns. To my family for always supporting my endeavors. Especially to Joshua John Decker for his strength and companionship and to my twin Sister, Michelle, for her gentle guidance and unconditional love. iii ACKNOWLEDGMENTS Funding support for this study was provided by the George J. and Martha C. Wallace Endowed Scholarship Award for Ornithology, the Association of Field Ornithologist’s E. Alexander Bergstrom Memorial Research Award, the Wilson Ornithological Society’s Paul A. Stewart Award, the Willard G. Pierce & Jessie M. Pierce Foundation Grant, two Michigan State University Zoology Departmental Fellowships, and a Michigan State University Graduate School Research Enhancement Award. 1 thank Andrea Battin, Kimberly Berenter, Walter Bialkowski, Nathan Carle, Joshua Decker, Thomas LeBlanc, Andrea Nash, Dave Slager, and Annette Snow for their enthusiastic assistance in the field. I am especially grateful to Gary Pierce of the Pierce Cedar Creek Institute and his staff for providing housing for the duration of this research project. Walter Chomentowski and Eraldo Matricardi assisted in the land-cover analyses and valuable statistical advice was given by Dr. Samuel Riffell, Dr. John Styrsky, Dr. Robert Ricklefs, and Dr. Scott Johnson. I am indebted to my committee members for their guidance and advice, especially during the early planning stages of this project. To my major advisor, Dr. Catherine Lindell, thank you for your support, expertise, and confidence in my abilities. To my committee members, Dr. Thomas Getty and Dr. Scott Winterstein, thank you for your insights and guidance. I am grateful to Emily Cohen for her support and the insightful comments that She provided on earlier versions of the manuscript. I thank the Michigan Department of Natural Resources for permitting me to work on state lands and to Ray Adams for his logistical support. iv TABLE OF CONTENTS LIST OF TABLES ............................................................................................................ vii LIST OF FIGURES ......................................................................................................... viii CHAPTER I: BACKGROUND INFORMATION BACKGROUND INFORMATION ........................................................................ I Overview ...................................................................................................... I Study Species ............................................................................................... 2 Relevant Demographic Research on the Wood Thrush ............................... 3 Potential Mechanisms of Nesting Success Patterns ..................................... 6 Measurements of Nestling Growth ............................................................ IO Derivation of Growth Rates ....................................................................... 11 Nestling Growth as an Indicator of Habitat Quality .................................. I4 Intrinsic Factors Affecting Nestling Growth ............................................. 15 Research Significance ................................................................................ I7 LITERATURE CITED .......................................................................................... 19 CHAPTER 2: EFFECT OF PROXIMITY TO FOREST EDGES ON NESTLIN G GROWTH AND NEST SURVIVAL OF WOOD THRUSH (H YLOCICHLA M UST ELINA) TN SOUTHWESTERN MICHIGAN ABSTRACT ........................................................................................................... 25 INTRODUCTION ................................................................................................. 27 METHODS ............................................................................................................ 29 Study Site ................................................................................................... 29 Study Species ............................................................................................. 33 Nest Searching and Monitoring ................................................................. 33 Nestling Growth Measurements ................................................................ 34 Weather Data ............................................................................................. 35 Distance to Forest Edge ............................................................................. 35 Vegetation Data ......................................................................................... 36 Land-cover ................................................................................................. 36 Data Analyses ............................................................................................ 37 RESULTS .............................................................................................................. 42 Nestling Growth ......................................................................................... 43 DSRS .......................................................................................................... 56 Nest Survival .............................................................................................. 58 Brown-headed Cowbird Parasitism ........................................................... 58 Vegetation Data ......................................................................................... 62 DISCUSSION ........................................................................................................ 66 Variability in Wood Thrush Growth Rates ................................................ 66 Comparisons of Site Survival Rates .......................................................... 70 Comparisons of Forest Edge and Interior DSRS ........................................ 71 Brood Parasitism ........................................................................................ 73 Limitations to this Study ............................................................................ 74 Conservation Implications ......................................................................... 75 LITERATURE CITED .......................................................................................... 77 vi LIST OF TABLES CHAPTER 1: BACKGROUND INFORMATION TABLE 1. Sizes of forest fragments assigned as small or large including the percent forest cover in the landscape and nesting success in small and large fragments. NA=Not Applicable. NR=Not Reported ................................................................. 7 TABLE 2. Reproductive success of Wood Thrush breeding in North American forests. NR=Not Reported .................................................................................................... 8 CHAPTER 2: EFFECT OF PROXIMITY TO FOREST EDGES ON NESTLIN G GROWTH AND NEST SURVIVAL OF WOOD THRUSH (HYLOCICHLA M UST ELINA) IN SOUTHWESTERN MICHIGAN TABLE I. The percentage of variation in the growth response variables explained by the set of predictor variables in each regression tree model and by each predictor variable within each model of the growth of nestling Wood Thrush. Sample Sizes of nests used in regression tree analyses are given in parentheses next to each growth response variable ....................................................................................... 44 TABLE 2. Daily survival rates (DSRS) i SE during the entire nesting period of Wood Thrush nests among plots and within years for each site ....................................... 57 TABLE 3. Daily survival rates (DSRS) i SE during the entire nesting period of Wood Thrush nests found in 2002 and 2003 in edge and interior distance-classes for each Site and year ................................................................................................... 59 TABLE 4. Variation in Wood Thrush nest survival rates among Sites and years ............. 60 TABLE 5. Variation in Wood Thrush nest survival rates between edge and interior distance-classes within sites ................................................................................... 61 TABLE 6. Vegetation structure of each Site in southwestern Michigan .......................... 63 TABLE 7. Vegetation characteristics of each Site in southwestern Michigan ................. 64 TABLE 8. Vegetation composition of each site in southwestern Michigan expressed as percentages ........................................................................................................ 65 vii LIST OF FIGURES CHAPTER 2: EFFECT OF PROXIMITY TO FOREST EDGES ON NESTLING GROWTH AND NEST SURVIVAL OF WOOD THRUSH (H YLOCICHLA MUST ELINA) IN SOUTHWESTERN MICHIGAN Figure 1. Map of two study sites in southwestern Michigan; Allegan State Game Area (ASGA) in Allegan County and Barry State Game Area/Yankee Springs Recreational Area (BSGA) in Barry County ......................................................... 3] Figure 2. Fractional green vegetation cover derived from Landsat 7 ETM+ images over (A) Allegan State Game Area (ASGA) and (B) Barry State Game Area/Yankee Springs Recreational Area (BSGA) in southwestern Michigan. Dark green represents forest with increasing degradation indicated by increasing brightness. Orange represents water. Images in this thesis are presented in color ................... 32 Figure 3. Tree mobile relating growth rate of A) left tarsus and B) right tarsus of nestling Wood Thrush to predictor variables during 2002 and 2003 in southwestern Michigan. ETYP =Edge Type, DIST=Distance from nest to forest edge, TPRE =Total Precipitation received during the nestling period. The tree mobiles had four terminal nodes and a residual relative error of 1.01 for the lefi tarsi and 1.14 for the right tarsi. Images in this thesis are presented in color. ............................. 45 Figure 4. Growth curves of Wood Thrush nestlings grouped by edge type, the most important Splitting variable for (A) left tarsus and (B) right tarsus. Dots indicate the average measurement (mean i SD). Day 0 corresponds to the day the first nestling hatched. Filled circles represent nests nearest to edges bordered by swamps/marshes (M), two-laned roads (TLR), and wildlife openings (W0) (n=3 1) and open circles represent nests nearest to edges bordered by powerline corridors (PC) and recent clearcuts (RC) (n=22) ................................................... 47 Figure 5. Growth curves of Wood Thrush nestlings grouped by edge (0-99 m) and interior (>100 m) distance-classes for (A) left tarsus and (B) right tarsus. Dots indicate the average measurement (mean i SD). Day 0 corresponds to the day the first nestling hatched. Filled circles represent data from nests in edge habitat (n=24) and open circles represent data from nests in interior habitat (n=32) ........ 48 Figure 6. Tree mobile relating growth rate of A) left wing and B) right wing of nestling Wood Thrush to predictor variables during 2002 and 2003 in southwestern Michigan. LTEM =Average Low Temperature during the nestling period, HTEM =Average High Temperature during the nestling period, DIST=Distance from nest to forest edge, TPRE =Total Precipitation received during the nestling period. The tree mobiles had three terminal nodes and a residual relative error of 1.72 for the left wing and a residual relative error of 1.51 for the right wing. Images in this viii thesis are presented in color. .................................................................................. 49 Figure 7. Growth curves of Wood Thrush nestlings grouped by edge (0-99 m) and interior (>100 m) distance-classes for (A) left wing chord and (B) right wing chord. Dots indicate the average measurement (mean i SD). Day 0 corresponds to the day the first nestling hatched. Filled circles represent data from nests in edge habitat (n=24) and open circles represent data from nests in interior habitat (n=3 2) ..................................................................................................................... 51 Figure 8. Tree mobile relating mass growth rate of nestling Wood Thrush to predictor variables during 2002 and 2003 in southwestern Michigan. TPRE =Total Precipitation received during the nestling period, LTEM =Average Low Temperature during the nestling period. The tree mobile for mass had three terminal nodes and a residual relative error of 1.77. Images in this thesis are presented in color. .................................................................................................. 53 Figure 9. Growth curves of Wood Thrush nestlings grouped by total precipitation, the most important splitting variable for mass. Dots indicate the average measurement (mean i SD). Day 0 corresponds to the day the first nestling hatched. Filled circles represent nests that received 32.54 inches of rainfall during the nestling period (n=47) and open circles represent nests that received >2.54 inches of rainfall (n=4) .......................................................................................................... 54 Figure 10. Growth curves of Wood Thrush nestlings grouped by edge (0-99 m) and interior (>100 m) distance-classes for mass. Dots indicate the average measurement (mean 35 SD). Day 0 corresponds to the day the first nestling hatched. Filled circles represent data from nests in edge habitat (n=24) and open circles represent data from nests in interior habitat (n=32) ................................... 55 CHAPTER 1 BACKGROUND INFORMATION Overview Over the last few decades, there has been growing concern over the population declines of Neotropical migrants that breed in mature forests, particularly in the eastern United States. Fragmentation of these forests, and associated deleterious effects of forest edges, has been implicated as the major causes of the decline. Most studies have investigated how nest survival rates vary in different-sized forest fragments and as a function of distance from a forest edge. While nest survival rates effectively measure reproductive success through the nesting period, nestling growth rates can affect nestling and postfledging survivorship and have received little attention in investigations of fragmentation effects on nesting birds. Nestling growth is affected by local environmental conditions and food availability and thus may be reflective of the quality of the habitat selected for nesting. In this chapter, I describe the breeding ecology of my study Species, the Wood Thrush (Hylocichla mustelina), and summarize the results of previous relevant demographic research addressing the effects of forest fragmentation on their nesting success. Next, I discuss the advantages and disadvantages associated with different measures of growth, the derivation of growth rates, and examine how nestling growth rates have been used in other studies to indicate habitat quality. Finally, I discuss intrinsic factors Shown to influence nestling growth for passerine species. Study Species The Wood Thrush (Hylociclzla mustelina) is a Neotropical migrant landbird that breeds throughout the eastern United States and southern Canada in deciduous/mixed forests and woodlands (Roth et al. 1996). A few decades ago, this species was considered common, however, long-term monitoring programs have documented their decline, leaving their future status uncertain (Robbins et al. 1989; Roth et al. 1996). The breeding ecology of the Wood Thrush has been well documented (Roth et al. 1996). At my study Sites in southwestern Michigan, their breeding season is fi‘om early May to late August. Males appear in southern Michigan in late April through early May and establish territories, usually a few days before the females arrive. Males defend their territories against conspecific males and the first breeding pairs are observed mating and building their open cup nests by the second week of May. Wood Thrush are a multi- brooded species capable of compensating for initial nest failures by renesting up to three times in a single breeding season, but typically they attempt to rear only two broods (Friesen et al. 2000; Friesen et al. 2001 ). Their initial clutch Size ranges from three to five eggs with subsequent clutches of two to three eggs. Females lay their eggs asynchronously and initiate incubation once the penultimate egg is laid. Their entire nesting period lasts between 24 and 27 days; 12-14 days in the incubation period and 12- 13 days in the nestling period (Roth et al. 1996; Clement 2000). Nests are built from 1 to 15 m high in small saplings and Shrubs (Ehrlich et al. 1988). Wood Thrush prefer nest sites in mature forests with a diversity of deciduous trees, an open forest floor with a moderate density of saplings and/or Shrubs, moist soil, and sufficient leaf litter cover for foraging (Roth et al. 1996; Hoover and Brittingham 1998). Relevant Demographic Research on the Wood Thrush A probable cause of the decline of forest—interior birds is forest fragmentation, which results in small forest patches and increased amounts of edge habitat (Whitcomb et al. 1981). Reduced reproductive success in small forest patches because of high rates of nest predation and brood parasitism associated with greater proximity to edges may explain the decline of this Species (Brittingham and Temple 1983; Askins et al. 1990). Nesting success of Wood Thrush has been studied extensively across their breeding range. Studies comparing the nesting success of Wood Thrush in forest patches found that nesting success and/or daily survival rates (DSRS) increased with increasing sizes of forest patches in landscapes dominated by agriculture in southcentral Ontario (Burke and No] 2000), northwestern Wisconsin (Donovan et al. 1995), southern Illinois (Robinson 1992; Brawn and Robinson 1996; Trine 1998), Missouri (Donovan et al. 1995), southcentral Pennsylvania (Hoover et al. 1995; Hoover and Brittingham 1998; Wilson et al. 1998), and northern Delaware (Weinberg and Roth 1998). For landscapes dominated by forest in northwestern Wisconsin (Donovan et al. 1995) and Missouri (Donovan et al. 1995) results were Similar. Reproductive success was not influenced by forest patch Size in agricultural landscapes in southwestern Ontario (Friesen et al. 1999), northern Indiana (Fauth 2000; Fauth 2001), or in landscapes dominated by forest in southeastern West Virginia (DeCecco et al. 2000) or southwestern Virginia (DeCecco et al. 2000). Simons et al. (2000) found no relationship between nesting success and forest patch size in the Great Smoky Mountains National Park, the largest contiguous tract of mature forest in the eastern United States, where there is minimal disturbance. There have been fewer studies directly linking the effects of proximity to edge with Wood Thrush reproductive success. The majority of studies, which may bear on edge effects on Wood Thrush, focus on comparing different-Sized forest patches (mentioned above) and determining the source/Sink status of the population in each forest patch. When a population iS self-sustaining and produces excess offspring (source), it may be capable of maintaining smaller populations within the region, which are not self- sustaining (Sinks) through emigration (Pulliam 1988). Several studies in the Midwest determined that small forest patches contained Sink populations (Missouri and Wisconsin, Donovan et al. 1995; Indiana, Ford et al. 2001). However, some small forest fragments produced source populations in most, but not all years (Ontario, Friesen et al. 1999; Indiana, F auth 2001). However, F auth (2000) determined that a large agricultural landscape with several small forest patches in northern Indiana was a Sink, which he described as a regional sink. In the Midwest, large forest patches were sources in Missouri (Donovan et al. 1995). However, even forest patches >1000 ha were Sinks in southern Illinois (Trine 1998). In the east, small forest patches were sinks and larger patches were sources (Pennsylvania, Hoover et al. 1995; Delaware, Weinberg and Roth 1998; Ontario, Burke and No] 2000). However, even small forest patches were capable of producing source populations (Delaware, Roth and Johnson 1993). The largest contiguous forest in the east contained a source population of Wood Thrush (Great Smoky Mountains National Park, Simons et al. 2000). Overall, the relationship between forest patch size and nest success has been more consistent across studies than the identification of large fragments as sources and small fragments as Sinks. Table 1 illustrates the variability in the assignment of forest fragments as small or large and Table 2 summarizes the results of demographic studies of Wood Thrush related to forest patch size and edge effects. In landscapes dominated by agriculture, direct studies of edge effects on nesting success have found no influence of distance to forest edge on nesting success in: southwestern Ontario with five edge classes (0-5 m, 5-25 m, 25-50 m, 50-100 m, and >1 00 m) (Friesen et al. 1999); northern Indiana with three edge classes (0-50 m, 51-100 m, >100 m) (Fauth 2000); southcentral Pennsylvania with continuous distances from edge habitat (Hoover and Brittingham 1998); southcentral Ontario with four edge classes (<50 m, 50-100 m, 101-200 m, >200 m) (Burke and N01 2000); or in landscapes dominated by forest comparing Sites near edges and Sites embedded in the interior forests of the Great Smoky Mountains National Park (Simons et al. 2000). Nest success increased with distance to forest edge in southcentral Pennsylvania but this effect was not evident after controlling for forest patch Size (Hoover et al. 1995). Ford et al. (2001) compared forest interior sites and Sites near agricultural conidors in southcentral Indiana and found that nesting success and nest productivity was lower in sites near agricultural corridors. The differences in experimental design of edge studies make it difficult to definitively determine the effect of proximity to edge on Wood Thrush nesting success. Potential Mechanisms of Nesting Success Patterns The lower rates of nesting success in some small forest patches may be because there are fewer large predators in small forest patches resulting in meso-predator release (Whitcomb et al. 1981) and habitat around fragments may support higher populations of nest predators (Johnson and Temple 1990). Knowledge of Wood Thrush nest predators is limited but Farnsworth and Simons (2000) documented eight nest predators either actively depredating nests or attending already depredated nests in the Great Smoky Mountains National Park: American Crows (Corvus brachyrhynchos), an Eastern Screech Owl (Otus asio), black rat snakes (Elaphe obsolete), a white-footed mouse (Peromyscus Ieucopus), southern flying squirrels (Glaucomys volans), a gray squirrel (Sciurus carolinensis), a least weasel (Mustela rixosa), and black bears (Ursus americanus). Other documented nest predators on Wood Thrush nests are Brown-headed Cowbirds (Molothrus ater), raccoons (Procyon lotor), Great Horned Owls (Bubo virginianus), domestic cats (F elis silvestrus) (Roth et al. 1996), and Common Grackles (Quiscalus quiscula) (Hoover et al. 1995). Blue Jays (C yanocitta cristata) and American Crows were the most abundant nest predators in southcentral Pennsylvania in areas where nesting success was low (Hoover et al. 1995). Predator relative abundance and activity decreased with increasing sizes of forest patches (Hoover et al. 1995) and increased near edge habitat in southcentral Indiana (reported in Ford et al. 2001). Raccoons and eastern chipmunks had higher densities in small forest patches (Wilcove 1985; Small and Hunter 1988) and along forest edges (Forsyth and Smith 1973) than in large forest patches. There was no relationship between nest predator abundance and forest patch size in northern Indiana (Fauth 2000) and no TABLE 1. Sizes of forest fragments assigned as small or large including the percent forest cover in the landscape and nesting success in small and large fragments. NA=Not Applicable. NR=Not Reported. Small Forest Nesting Large Forest Nesting Fragment Cover success Fragment Cover success (ha) (‘Vy (%) (119 670) (%) Rderence 0.2-2.1 15 34.7-58.6 15 15 71.4-76.9 Weinberg and Roth 1998 3-14 14 56 26-140 14 47 friesen et al. 1999 6-500 3 9 29-77 NA NA NA Fauth 2001 7-490 3 9 16-71 NA NA NA Fauth 2000 9-80 21-51 12-43 103- >10,000 97 65-100 Hoover et al. 1995 11.8-2352.8 13.6-53.2 18.9 >10,000 >80 49.6 Burke and N01 2000 16.4- >500 21-51 60 b >ro,ooo 97 60 *’ Hoover and Brittingham 1998 164-1264 NR 67 b >10,000 NR 67 b Wilson et al. 1998 <200 10 NR NA NA NA Brawn and Robinson 1996 551 31 32 29,175 93 60 Donovan et al. 1995 675 32 27 26,794 95 41 lDonovan et al. 1995 NA NA NA >l,000 NR 95.0-98.0 Trine 1998 3 Authors designated fragments of these sizes as both small and medium-sized. b Nesting success of small and large fragments reported as a single percentage. .23an 89a 32580 88.. 588E o 8808 88: some: 2: B 83858 28 888 8882.. 88 8882 88852 288 820 . 82 .8 a 8.8 mz 8» 88888 8 8.8 2 8 an 8.82 22>? 82 8 8 88 m2 ma 8888 8 o 8 8.8 2.2.8 8-82 22:? 88 .8 a 88 .8 m2 88888 8-8 «2 82 mz 8-8 8-82 >2 888 8 8 m2 8 .8888 «2 3 8 82 82 8-82 <> .3 82 8 889.8 8 8 3888 2 8-2. 8.2. 8.8-8.8 83 8-82 20 .5. 888 838.58 o8 mz m2 8888 8A mz 8 8.8-8.8 82 82.8-82 20 .m 82 88 m2 as 82288 mz 878 8.8 8.8-8.8 88? 8.82 a .m 888 88888 88 888 m2 m2 82888 83.: . 8-:- mz 8.8-2 .8 88-8 8-82 .8 .m 82 83cm 8 8:88 m2 . mz 88288 8 82 82 8.8.8.8 218 8-82 <0 .m 88 .8 a 8805 mz 8 88288 mz 8.8 8 82 mz 8-82 >3 .mm 82 .8 8883 m2 v.8 8828 82 8 ._ S .mz 88.88 4.2 8-82 <8 .8 82 888885 88.382 8 m2 8888 8-8 82 s 8 82 88.88 +2 :2 <8 .8 82 .8 a 8.82 .8 .08 8888 8-8 2-8 82-2 82 882A a 8.82 <8 .8 888 82 88 88m 8 .8 8888 8A .8 E 8.2 8.8-8.8 892A .2 8.82 20 .8 28m .8 8 888 m8 m2 8888 a . 8 82 8.8.8 82 8.82 E .8 888 58-2 8 8 88288 a 878 8-2 82 81 8.82 E .32 288 888 m2 m2 3888.8 8 82-3. 8.8 8.8-8.8 83 8.82 E .z 82 8m 88 2883 mz 8» 38888 2 8-8 8-8 82 2-3 8-82 mo .2 82 88888 88 882 m2 8» 828888 2 878 m2 3.3.8 8.2 8-82 .882 d 82 888 88 8829888 m2 m2 8888 8 N a 8 8.8.08 82 8-82 . 822mm 888 8 8 828m 8 8 82288 8 8 2-8 83.8 «2 8-82 . 822mm 82 898 m2 m2 88288 «z o 82 82 .82 8-82 x... .m 82 .8 a 8889 m2 a2 88888 8 E 8 .18 «8 8-82 02 .o 82 8 a 8.582 mz ma 8888 8 o :. 8.8 8.8 8-32 02 .O engage-Bx «goofim 8.8%» ~888on $2 .830 «X» 83% «X» «Vow Mud «as 98o» 22393 thM ~28. “mo-SR 5.83.382 8.8.6:...- ufin Boga-Q -582 math-oz \c smack .3635 «o Zum Z 88.5QO SZHMZ .2888 58:82. 5.82 E ”58005 nus-=E- coo? .«o 8328 o>uonvoaom .m mag“. effect of edges on mammalian nest predator activity in southcentral Ontario (Burke and N01 2000). More work is needed to link specific predators to predation events and to identify whether predator populations are responding to forest patch size and/or proximity to edges. These studies indicate that responses of predators to fragmentation may be species-specific. In addition, whether nest predation levels are causatively linked to regional declines needs to be investigated. Brown-headed Cowbird parasitism pressure varies throughout the range of the Wood Thrush (Hoover and Brittingham 1993; Roth and Johnson 1993). Brown-headed Cowbird abundance decreased with increasing distance from edge habitat in Missouri (Brittingham and Temple 1983) but did not differ between edge and interior habitat in Missouri (Donovan et al. 1997) or in northern Indiana (F auth 2000). Parasitism rates were not higher along forest edges in southwestern Indiana (Ford et al. 2001) or in the Great Smoky Mountains National Park (Simons et al. 2000). Rates were highest within 100 m of forest edges in southcentral Ontario (Burke and No] 2000) and along forest edges in Michigan (Gates and Gysel 1978), Maryland (Chasko and Gates 1982), and southern Wisconsin (Temple and Cary 1988). Cowbird parasitism is higher in the Midwest where agricultural land dominates the landscape. Their occurrence is related to proximity to agriculture, and they respond positively to the creation of edges (Stribley and Haufler 1999). In conclusion, the effect of forest patch size and proximity to edge on the nesting success of Wood Thrush may depend on the surrounding landscape and on the local predation and parasitism pressures. The relationship between the abundance of nest predators, patch size, and proximity to edge are poorly understood but may be species- specific. Brown-headed Cowbirds are generally more abundant in the Midwest region where agricultural land dominates than in the east and respond positively to edges. Clearly, the extreme variation in the reproductive success of Wood Thrush across their geographic range combined with many different experimental designs makes it difficult to generalize the effects of forest fragmentation on the nesting success of Wood Thrush. In contrast to previous studies that have examined fragmentation effects only in the context of nest success, I also examined potential fragmentation effects on nestling growth, which may be linked to fragmentation-induced habitat degradation. Measurements of Nestling Growth Bone growth is typically determined by measuring tarsi, flight feather growth is determined by measuring the lengths of outer primaries or wing chords, and overall body size is determined by weighing individuals throughout the nestling period. The lengths of bones and feathers are highly correlated with the age of the nestlings (Starck and Ricklefs 1998). Therefore, measurements such as size at a particular age, the age a nestling reaches a given size, or growth increments between two ages are useful for intraspecific comparisons of postnatal growth (Ricklefs 1967; Quinney et al. 1986; Riddington and Gosler 1995; Siikaméiki 1996; Bradbury et al. 2003), but these measurements do not describe growth over the entire nestling period (Ricklefs 1968; Ricklefs 1984). Growth increments are especially hard to interpret because growth is nonlinear (van Noordwijk and Marks 1998) and discerning variation in the growth pattern is difficult when measurements are compared at a given age because there may be greater variation in growth earlier or later in the nestling period depending on intrinsic and extrinsic factors, 10 even within a species (Ricklefs 1968). The calculation of grth rate constants should be used in comparisons of intraspecific and interspecific nestling growth because they are independent of the length of the postnatal growth period (Starck and Ricklefs 1998; McCarty 2001). The growth rate constant is proportional to the specific rate of growth until adult body size is attained (Ricklefs 1968) and is essentially the slope of the line tangent to the inflection point of the growth curve (Ricklefs 1967). Only grth rate constants provide information on the magnitude, i.e. net increase in body size during growth, pattern of growth, and the rate at which adult body size is attained (Ricklefs 1968) Derivation of Growth Rates Growth rates of nestlings are calculated by fitting repeated measurements of sizes at particular ages during postnatal growth, such as mass and length of wings and tarsi, to sigrnoidal growth equations using nonlinear regression. There are three equations that can be used to describe avian nestling growth: the logistic, Gompertz, and von Bertalanffy equations (Ricklefs 1968). The advantage of these equations over other alternatives is having only a single derived parameter (Starck and Ricklefs 1998). All three equations estimate the growth rate constant (K), after identifying the miminum size of the measurement at hatch day (M0) and the maximum size of the measurement where size begins to level off, the asymptotic value (A). The equations differ in their specification of the inflection point of the curve when maximum growth is attained (Peters 1983; Starck and Ricklefs 1998). The logistic, Gompertz, and von Bertalanffy growth equations’ inflection points occur when 50, 37, and 30% of the asymptote is attained, respectively 11 (Ricklefs 1968). The Gompertz and von Bertalanffy equations model patterns of growth for birds that grow slowly earlier and have prolonged nestling periods to accommodate this slower grth pattern (Ricklefs 1967), which is characteristic of large-sized, precocial species (Ricklefs 1984). Altricial birds in the temperate region grow more rapidly than same-sized precocial species so that their growth is best approximated by the logistic growth equation (Ricklefs 1968; Ricklefs 1984). The two rarely used alternatives to these three growth functions are the Richards’ and Janoschek equations, which do not require the specification of the inflection point but are more complex with a greater number of derived parameters (Starck and Ricklefs 1998). These equations are usually only applied in studies of species whose growth patterns have not been described. It has repeatedly been shown that the logistic growth equation best describes the altn'cial growth of passerine nestlings (e. g. (Dark-eyed Junco) Smith and Andersen 1982; (Sage and Brewer 's Sparrows) Petersen et al. 1986; (Western Kingbird) Blancher and Robertson 1987; (Magpie) Soler and Soler 1991; (Tree Swallow) McCarty 2001). In a review of interspecific growth rates, 81 of 105 species considered were best fit by the logistic growth equation, including all 60 passeriform species considered (Ricklefs 1968). In addition, the logistic curve was the best fit for measurements of the Wood Thrush and four other thrush species in the genus Hylocichla (Ricklefs 1968). The asymptotic value is important because the estimate of the growth rate constant is only as accurate as the asymptotic value (Ricklefs 1984). Therefore, if the objective is to describe the growth of nestlings from hatching to adult body size, the asymptotic value used would be the maximum adult body size. If the objective is to 12 describe the growth of nestlings from hatching to fledging, the asymptotic value used would be the maximum size of fledglings. In this study, it was not possible to measure nestlings beyond day 10 so nestlings were still below adult body size on the final measurement day. Nestlings may level off in growth and then slowly increase to adult body size after leaving the nest (Ricklefs 1984). Because I was unable to capture and measure fledglings I fit growth curves to the mean measurements of adult body size. Measurements of nestlings from the same brood are not independent (Hurlbert 1984). To avoid pseudoreplication, growth rate constants can be calculated for each nestling within a brood and then averaged to obtain one value for each brood (e. g. Soler and Soler 1991) or one growth rate constant can be calculated for an entire brood using the measurement data of all of the broodmates (Ricklefs 1968; van Noordwijk and Marks 1998). Alternatives to these methods are to include “nest” as a variable in the analyses to account for within brood variation (Stoleson and Beissinger 1997) or to calculate one grth rate constant for an entire population using the mean measurements for each nestling age (McCarty 2001). Each method is valid as long as individual nestlings are treated as subsamples rather than independent samples. 1 calculated growth constants for each brood because I was interested in interbrood variation. Because Wood Thrush eggs hatch asynchronously, I was unable to measure last-hatched nestlings at least five times because of logistical constraints so that there were too few measurements to fit a curve to the last-hatched nestling. Hence, information on variation in nestling growth within a brood would have been lost if I had averaged growth rate constants for only 60-75% of the brood. l3 Nestling Growth as an Indicator of Habitat Quality Measurements of nestling growth may be useful indicators of nesting habitat quality (e. g. Pierotti 1982; McCarty 2001). Specifically, nestling growth has been used as a measure of food supply (Quinney et al. 1986; Stoleson and Beissinger 1997). Forested habitats may vary with respect to food availability (Schowalter et a1. 1981; Niemeléi et al. 1993), which in turn may be affected by several factors including age of forests (Bertin 1977; Jolivet 1986; Niemeléi et al. 1993; Greenberg and McGrane 1996), proximity to forest edge (Duguay et al. 2000), and environmental conditions, such as ambient temperature (e. g. McCarty and Winkler 1999) and rainfall (e. g. Bradbury et al. 2003). Growth rate has been shown to positively correlate with food availability in several passeriform species (e. g. (House Martin) Bryant 1975; (Tree Swallow) Quinney et al. 1986; (Great Tit) Keller and van Noordwijk 1994; (Great Tit) Eeva et al. 1997; (Tree Swallow) McCarty and Winkler 1999; (Wood Thrush) Duguay et al. 2000). Riddington and Gosler (1995) found that nestling Great Tits were larger at fledging in mature woodland than in gardens and hedgerows, which were considered marginal habitat because they contained food lower in diet quality, determined by scoring the quality of food types collected from fecal sacs. Quinney et al. (1986) determined that site explained 51% of the variation in the growth of nestling Tree Swallows. Nestlings grew more rapidly at this site because of greater food abundance resulting in greater food provisioning rates by parents (Quinney et al. 1986). Western Kingbirds had more rapid growth rates at sites with higher insect biomass (Blancher and Robertson 1987), and nestling growth rates of Black-throated Blue Warblers were reduced because of a decrease in caterpillar abundances (Rodenhouse and Holmes 1992). 14 Adverse weather has both indirect and direct effects on nestling growth. Adverse weather affects growth directly by increasing energy requirements for thermoregulation and maintenance and indirectly by decreasing the foraging efficiency of parents (Siikamaki 1996) and the abundance of prey (McCarty and Winkler 1999; Bradbury et al. 2003). Pied Flycatcher nestling masses were negatively affected by total rainfall received during the rearing period and positively affected by mean and maximum temperatures (Siikamaki 1996). The impact of adverse weather on growth rate was greater during the final few days of the nestling period when the demand for food was highest (Siikaméiki 1996). Nestling Skylark (A lauda arvensis), Chaffinch (Fringilla coelebs), and Yellowhammer (Emberiza citrinella) growth were negatively affected by precipitation and low temperatures attributed to increased demands for thermoregulation, reduced rates of food delivery, and reduced availability of food through decreased invertebrate activity (Bradbury et al. 2003). Cold and wet weather negatively influenced growth of Tree Swallows (McCarty and Winkler 1999) likely by affecting their food supplies (McCarty and Winkler 1999). Intrinsic Factors Affecting Nestling Growth Nestling growth rates may be affected by intrinsic factors, such as species-specific brood size, date of hatching, and parental quality. Nestlings from larger broods may have reduced growth rates and fledging masses (Kunz and Ekman 2000) because of the inability of parents to feed each nestling at the same rate as parents of smaller broods (N aef-Daenzer and Keller 1999) and the greater amount of energy expended by nestlings in intrabrood competitions for food (Schifferli 1978). Typically, this pattern is seen in 15 species that exhibit extreme variation in their brood sizes. In a study by Podlesak and Blem (2001), the brood size of Prothonotary Warblers ranged from 1 to 6 chicks. They found that the growth rates and fledging masses of Prothonotary Warblers were lower in larger broods (Podlesak and Blem 2001). Likewise, nestling Blue Tits from experimentally enlarged broods had reduced growth rates and fled ging masses (Kunz and Ekman 2000). The results of Podlesak and Blem’s study (2001) were attributed to interactions of brood size with date of hatching and female age. Middle-aged female Prothonotary Warblers produced nestlings with higher growth rates than did one-year old females and females >3 years old (Podlesak and Blem 2001). Older females initiated laying earlier than younger females (Blem et al. 1999). The date of first hatching was positively associated with nestling growth rates of Willow Tits (Pravosudov and Pravosudova 1996) and Tree Swallows (McCarty and Winkler 1999). All authors attributed these patterns to seasonal changes in food abundance. Finally, parents may differ in their breeding experience, ability to acquire high- quality territories, and ability to find and deliver food (van Noordwijk et al. 1995). Nestling growth rates of Great and Blue Tits were positively related to parental food provisioning rates (Naef-Daenzer and Keller 1999), which were positively correlated with prey biomass (N aef-Daenzer et al. 2000). Quinney et al. (1986) found this same result with Tree Swallows. Hence, it is difficult to conclude that variation in nestling growth rates is solely a result of habitats that differ in food availability because a parent’s ability to find and provide food to their chicks may also influence growth rates (Naef- Daenzer et a1 2000). Nestling growth rates, food abundance, and parental provisioning l6 appear to interact but few studies have been designed to effectively investigate these factors separately. Observational studies have generally ignored the genetic component of growth. Several genes control nestling growth and these genes interact with extrinsic factors to determine growth rates (van Noordwijk and Marks 1998). The focus of this study was to document patterns in growth and to describe the variability in the growth of three measures of nestling size with respect to the proximity of nests to forest edges. 1 did not address to what extent the observed variation in nestling growth was inherited, the genetic component of growth. Research Significance Although Wood Thrush are ofien chosen as the focal species in investigations of forest fragmentation, the effect of forest edges on Wood Thrush reproductive success remains unclear. Previous studies differed widely in their definition of edges and at what distance to measure edge effects. In this study, I compare nesting success and nestling growth rates as functions of continuous distances from the forest edge to capture patterns that may have been masked in studies relying on discrete edge classes. Detection of differences in growth rates as a function of distance to forest edge will indicate differences in habitat quality, or, potentially, the quality of the individuals that nest at different distances to edge. Such information will lead to greater understanding of how forest fragmentation may lead to the creation of optimal and suboptimal habitat, which may affect the future prospects of a population. An understanding of the effects of forest edges on nestling growth and the nesting success of 17 Wood Thrush will identify potential interactions between distance to forest edge and measures of reproductive success, thus, helping biologists design effective conservation measures for Wood Thrush and other declining Neotropical migrants that breed in fragmented forested landscapes. l8 LITERATURE CITED Artman, V. L. and J. F. Downhower. 2003. Wood Thrush (Hylocichla mustelina) nesting ecology in relation to prescribed burning of mixed—oak forest in Ohio. Auk 120:874-882. Askins, R. A., J. F. Lynch, and R. Greenberg. 1990. Population declines in migratory birds in eastern North America. Current Ornithology 7:1-57. Bertin, R. I. 1977. Breeding habitats of the Wood Thrush and Veery. Condor 79:303-31 1. Blancher, P. J. and R. J. Robertson. 1987. Effect of food supply on the breeding biology of Western Kingbirds. Ecology 68:723-732. Blem, C. R., L. B. Blem, and C. I. Barrientos. 1999. Relationships of clutch size and hatching success to age of female Prothonotary Warblers. Wilson Bulletin 111:577-581. Bradbury, R. B., J. D. Wilson, D. Moorcrofi, A. J. Morris, and A. J. Perkins. 2003. Habitat and weather are weak correlates of nestling condition and growth rates of four UK farmland passerines. Ibis 145:295-306. 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. Bryant, D. M. 1975. Environmental influences on growth and survival of nestling House Martins Delichon urbica. Ibis 120:271-283. Brittingham, M. C. and S. A. Temple. 1983. Have cowbirds caused forest songbirds to decline? Bioscience 33:31-35. Burke, D. M. and E. N01. 2000. Landscape and fragment size effects on reproductive success of forest-breeding birds in Ontario. Ecological Applications 10: 1 749-1761. Carrie, N. R. 1999. Brown-headed Cowbird parasitism of Wood Thrush nests in eastern Texas. Journal of Field Ornithology 70:263-267. Chasko, G. G. and J. E. Gates. 1982. Avian habitat suitability along a transmission-line corridor in an oak-hickory forest region. Wildlife Monographs 82:1-41. Clement, P. 2000. Wood Thrush. Pages 319-322 in Thrushes. Princeton University Press, Princeton, New Jersey. DeCecco, J. A., M. R. Marshall, A. B. Williams, G. A. Gale, R. J. Cooper. 2000. Comparative seasonal fecundity of four Neotropical migrants in middle Appalachia. Condor 102:653-663. 19 Donovan, T. M., P. W. Jones, E. M. Annand, and F. R. Thompson, 111. 1997. Variation in local-scale edge effects: Mechanisms and landscape context. Ecology 78:2064-2075. Donovan, T. M., F. R. Thompson, 111, J. F aaborg, and J. R. Probst. 1995. Reproductive success of migratory birds in habitat sources and sinks. Conservation Biology 9: 1380- 1395. 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 28:1 123-1131. Eeva, T., E. Lehikoinen, and T. Pohjalainen. 1997. Pollution-related variation in food supply and breeding success in two hole-nesting passerines. Ecology 78:1 120-1131. Ehrlich, P. R., D. S. Dobkin, and D. Wheye. 1988. Wood Thrush. Page 456 in The Birder’s Handbook: A Field Guide to the Natural History of North American Birds. Simon & Schuster Inc., New York, New York. F arnsworth, G. L., and T. R. Simons. 2000. Observations of Wood Thrush nest predators in a large contiguous forest. Wilson Bulletin 112282-87. Fauth, P. T. 2000. Reproductive success of Wood Thrushes in forest fragments in northern Indiana. Auk 1 17:194-204. Fauth, P. T. 2001. Wood Thrush populations are not all sinks in the agricultural Midwestern United States. Conservation Biology 15:523—527. Ford, T. B., D. E. Winslow, D. R. Whitehead, and M. A. Koukol. 2001. Reproductive success of forest-dependent songbirds near an agricultural corridor in south-central Indiana. Auk 1182864-873. Forsyth, D. J. and D. A. Smith. 1973. Temporal variability in home ranges of Eastern Chipmunks ( T amias-Striatus) in a southeastern Ontario woodlot. American Midland Naturalist 90: 107-1 17. Friesen, L. E., M. D. Cadman, and M. L. Allen. 2001. Triple brooding by southern Ontario Wood Thrushes. Wilson Bulletin 113:237-239. Friesen, L., M. D. Cadman, and R. J. MacKay. 1999. Nesting success of Neotropical migrant songbirds in a highly fragmented landscape. Conservation Biology 13:338-346. Friesen, L. E., V. E. Wyatt, M. D. Cadman, R. J. MacKay, E. D. Cheskey, M. L. Allen, and D. Ramsay. 2000. Extent of double-brooding and seasonal movement of nesting females in a northern population of Wood Thrushes. Wilson Bulletin 112:505-509. Gates, J. E., and L. W. Gysel. 1978. Avian nest dispersion and fledging success in field- forest ecotones. Ecology 59:871-883. 20 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. Hoover, J. P. and M. C. Brittingham. 1993. Regional variation in cowbird parasitism of Wood Thrushes. Wilson Bulletin 1052228-238. Hoover, J. P. and M. C. Brittingham. 1998. Nest-site selection and nesting success of Wood Thrushes. Wilson Bulletin 110:375-383. Hoover, J. P., M. C. Brittingham, and L. J. Goodrich. 1995. Effects of forest patch size on nesting success of Wood Thrushes. Auk 1122146-155. Hurlbert, S. H. 1984. Pseudoreplication and the design of ecological field experiments. Ecological Monographs 54: 1 87-21 1 . Johnson, R. G. and S. A. Temple. 1990. Nest predation and brood parasitism of tallgrass prairie birds. Journal of Wildlife Management 54:106-11 1. Jolivet, P. 1986. Insects and plants: parallel evolution and adaptations in Flora and Fauna (E. R. Brill, Ed.). Flora & Fauna Publications, New York, New York. Keller, L. F. and A. J. van Noordwijk. 1994. Effects of local environmental conditions of nestling growth in the Great Tit Parus major L. Ardea 82:349-362. Kunz, C. and J. Ekman. 2000. Genetic and environmental components of growth in nestling Blue Tits (Parus caeruleus). Journal of Evolutionary Biology 13:199-212. McCarty, J. P. 2001. Variation in growth of nestling Tree Swallows across multiple temporal and spatial scales. Auk 1 18: 176-190. McCarty, J. P. and D. W. Winkler. 1999. Relative importance of environmental variables in determining the growth of nestling Tree Swallows T achycineta bicolor. Ibis 1412286- 296. Naef-Daenzer, B. and L. F. Keller. 1999. The foraging performance of Great and Blue Tits (Parus major and P. caeruleus) in relation to caterpillar development, and its consequences for nestling growth and fledging weight. Journal of Animal Ecology 68:708-718. 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. 21 Niemeléi, J ., D. Langor, and J. R. Spence. 1993. Effects of clear-cut harvesting on boreal ground-beetle assemblages (ColeopterazCarabidae) in western Canada. Conservation Biology 7:551-561. Noordwijk, A. J. van and H. L. Marks. 1998. Genetic Aspects of Growth. Pages 305-323 in Avian Growth and Development (J. M. Starck and R. E. Ricklefs, Eds.). Oxford University Press, New York, New York. Noordwijk, A. J. van, R. H. McCleery, and C. M. Perrins. 1995. Selection for the timing of Great Tit breeding in relation to caterpillar growth and temperature. Journal of Animal Ecology 64:451-458. Peters, R. H. 1983. Chapter 8: Production: Growth and reproduction. Pages 122-124 in The Ecological Implications of Body Size (E. Beck, H. J. B. Birks, E. F. Connor, Eds). Cambridge University Press, New York, New York. Petersen, K. L., L. B. Best, and B. M. Winter. 1986. Growth of nestling Sage Sparrows and Brewer’s Sparrows. Wilson Bulletin 98:535-546. Pierotti, R. 1982. Habitat selection and its effect on reproductive output in the Herring Gull in Newfoundland. Ecology 63:854-868. Podlesak, D. W. and C. R. Blem. 2001. Factors associated with growth of nestling Prothonotary Warblers. Wilson Bulletin 113:263-272. Powell, L. A., M. J. Conroy, D. G. Krementz, and J. D. Lang. 1999. A model to predict breeding-season productivity for multibrooded songbirds. Auk 116:1001-1008. Pravosudov, V. V. and E. V. Pravosudova. 1996. The breeding biology of the Willow Tit in northeastern Siberia. Wilson Bulletin 108:80-93. Pulliam, H. R. 1988. Sources, sinks, and population regulation. American Naturalist 1321652-661. Quinney, T. E., D. J. T. Hussell, and C. D. Ankney. 1986. Sources of variation in growth of Tree Swallows. Auk 103:389-400. Ricklefs, R. E. 1967. A graphical method of fitting equations to growth curves. Ecology 48:978-983. Ricklefs, R. E. 1968. Patterns of growth in birds. Ibis 110:419-451. Ricklefs, R. E. 1984. The optimization of grth rate in altricial birds. Ecology 65:1602- 1616. 22 Riddington, R. and A. G. Gosler. 1995. Differences in reproductive success and parental qualities between habitats in the Great Tit Parus major. Ibis 137:371-378. Robbins, C. S., J. R. Sauer, R. S. Greenberg, and S. Droege. 1989. Population declines in North American birds that migrate to the Neotropics. Proceedings of the National Academy of Science 86:7658-7662. Robinson, S. K. 1992. Population dynamics of breeding Neotropical migrants in a fragmented Illinois landscape. Pages 408-418 in Ecology and Conservation of Neotropical migrant landbirds (J. M. Hagan and D. W. Johnston, Eds). Smithsonian Institution Press, Washington, DC. 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 73:3 57-372. Roth, R. R., and R. K. Johnson. 1993. Long-term dynamics of a Wood Thrush population breeding in a forest fragment. Auk 110237-48. Roth, R. R., M. S. Johnson, and T. J. Underwood. 1996. Wood Thrush (Hylocichla mustelina). The Birds of North America, No. 246 (A. Poole and F. Gill, Eds.). The Academy of National Science, Philadelphia, and The AOU, Washington, DC. Schifferli, L. 1978. Experimental modification of brood size among House Sparrows Passer domesticus. Ibis 120:365-369. Schowalter, T. D., J. W. Webb, and D. A. Crossley, Jr. 1981. Community structure and nutrient content of canopy arthropods in clearcut and uncut forest ecosystems. Ecology 62:1010—1019. Siikamaki, P. 1996. Nestling growth and mortality of Pied F lycatchers F icedula hypoleuca in relation to weather and breeding effort. Ibis 138:471-478. Simons, T. R., G. L. Famsworth, and S. A. Shriner. 2000. Evaluating Great Smoky Mountains National Park as a population source for the Wood Thrush. Conservation Biology 14:1133-1144. Small, M. F. and M. L. Hunter. 1988. Forest fragmentation and avian nest predation in forested landscapes. Oecologia 76:62-64. Smith, K. G. and D. C. Andersen. 1982. Food, predation, and reproductive ecology of the Dark-eyed Junco in northern Utah. Auk 99:650-661. 23 Soler, M. and J. J. Soler. 1991. Growth and development of Great Spotted Cuckoos and their Magpie host. Condor 93 249-54. Starck, J. M. and R. E. Ricklefs. 1998. Avian Growth Rate Data Set. Pages 381-382 in Avian Growth and Development (J. M. Starck and R. E. Ricklefs, Eds.). Oxford University Press, New York, New York. Stoleson, S. H. and S. R. Beissinger. 1997. Hatching asynchrony, brood reduction, and food limitation in a Neotropical parrot. Ecological Monographs 67:131-154. Stribley, J. M. and J. B. Haufler. 1999. Landscape effects on cowbird occurrences in Michigan: Implications to research needs in forests of the inland west. Studies in Avian Biology 18:68-72. Temple, S. A. and J. R. Cary. 1988. Modeling dynamics of habitat-interior bird populations in fragmented landscapes. Conservation Biology 2:340-347. Trine, C. L. 1998. Wood Thrush population sinks and implications for the scale of regional conservation strategies. Conservation Biology 12:576-585. Weinberg, H. J ., and R. R. Roth. 1998. Forest area and habitat quality for nesting Wood Thrushes. Auk 115:879-889. Whitcomb, R. F ., J. F. Lynch, M. L. Klimkiewicz, C. S. Robbins, B. L. Whitcomb, and D. Bystrak. 1981. Effects of forest fragmentation on avifauna of the eastern deciduous forest. Pages 125-205 in Forest Island Dynamics in Man-dominated Landscapes (R. L. Burgess and D. M. Sharpe, Eds.). Springer-Verlag, New York, New York. Wilcove, D. S. 1985. Nest predation in forest tracts and the decline of migratory songbirds. Ecology 66:12] 1-1214. Wilson, G. R., M. C. Brittingham, and L. J. Goodrich. 1998. How well do artificial nests estimate success of real nests? Condor 100:357-364. 24 CHAPTER 2 EFFECT OF PROXIMITY TO FOREST EDGES ON NESTLING GROWTH AND NEST SURVIVAL OF WOOD THRUSH (H YLOCICHLA M US T ELINA) IN SOUTHWESTERN MICHIGAN ABSTRACT Many previous studies investigating fragmentation effects on forest-nesting birds have focused on nest predator and brood parasite responses to edges. However, forest- nesting birds may also be affected by forest edges because edge habitat is degraded, thus influencing microclimate and, potentially, food supplies. Nestling grth rates have been used as indicators of habitat degradation because growth rates have been positively correlated with food. However, previous work has not examined growth rates in the context of forest fragmentation. I investigated whether distance to forest edge influenced growth rates of wings, tarsi, and mass of nestling Wood Thrush. I also examined whether distance to forest edge, edge type, fractional cover, average high and low temperatures, total precipitation, brood size, and date of hatching explained some of the variability in growth rates. To compare my results to previous studies, which have focused on measures of nest success to assess Wood Thrush responses to fragmentation, I also report Mayfield’s daily nest survival rates, nest survival rates, and brood parasitism rates in edge and interior forest. The study was conducted in two forested landscapes in southwestern Michigan from May to August in 2002 and 2003. I located 175 Wood Thrush nests and measured nestlings from 58 nests. Tarsi growth rates were most rapid near powerline corridors and recent clearcuts, which may be food-abundant, and mass growth rates increased with total precipitation. Growth was more rapid for body 25 structures, like tarsi, that may play an important functional role early in the nestling period, as opposed to wing chords. There was no relationship between distance to forest edge and nest survival rates. Brood parasitism rates were among the lowest reported for Wood Thrush nesting in fragmented forests of the Midwest. My finding that proximity to edge influenced tarsi growth rates but not nesting success suggests nestling growth rates may indicate habitat degradation at edges even when high regional fragmentation levels may overwhelm potential edge-interior differences in local predation patterns and nesting SUCCESS. 26 INTRODUCTION Fragmentation of forests has been repeatedly indicated as a probable cause of the decline of forest-interior Neotropical migrant species, particularly in the Midwest (Robbins et al. 1989b; Donovan et al. 2002). As forests become increasingly fragmented, the size of fragments decreases and a greater proportion of their area contains edge habitat (Whitcomb et al. 1981). The majority of the work to date investigating fragmentation effects on forest- nesting birds has focused on nest predator and brood parasite responses to edges. The greater abundance and activity of brood parasites and nest predators near edges and throughout small forest patches often reduces reproductive success of nesting birds (Brittingham and Temple 1983; Askins et al. 1990; Andrén 1992; Robinson 1992; Roth and Johnson 1993; Hoover et al. 1995; Hoover and Brittingham 1998). Edge habitat may also support a greater diversity of nest predators (Johnson and Temple 1990). However, effects of edges on forest—nesting birds may not be restricted to those caused by nest predators and brood parasites. Edge habitat can experience modified microclimatic conditions (Chen et al. 1993; Murcia 1995), which can affect the abundance of invertebrates, the primary food source of many forest-nesting species (Schowalter et al. 1981). Therefore, territories in edge habitat may differ in their food supplies compared to territories in the forest interior (Burke and N01 2000). Hence, edge effects may become evident not only in reduced nest success because of nest predators and brood parasites but also in reduced nestling growth rates because of food limitation. In my study species, the Wood Thrush (Hy/ocichla mustelina), nestling growth rates have 27 been positively correlated with the biomass of their primary food source, soil invertebrates (Duguay et al. 2000), indicating that nestling growth rates may reflect differences in nesting habitat quality. An understanding of the effects of forest edges on nestling growth and the nesting success of Wood Thrush would identify potential effects of fragmentation mediated through habitat quality differences between edge and interior. The Wood Thrush is a Neotropical migrant that has shown population declines over its entire breeding range (Robbins et al. 1989b; Peterjohn et al. 1995; Sauer et al. 1996; Holmes and Sherry 2001). Wood Thrush often nest near habitat edges and in small woodlots (Roth et al. 1996) but have been classified as area-sensitive based on several studies that found nesting success increased with increasing size of forest patches and with increasing distance to forest edge (Robbins et al. 1989a; Donovan et al. 1995; Hoover et a1. 1995; Brawn and Robinson 1996; Hoover and Brittingham 1998; Trine 1998; Weinberg and Roth 1998; Wilson et a1. 1998; Burke and N01 2000). The effect of forest patch size on their nesting success has been consistent across their breeding range. However, many studies have found no influence of distance to forest edge on nesting success, although no previous studies have examined the influence of distance to edge on nestling growth rates. In addition to examining relationships between distance to forest edge and nestling growth rates, I examined the influence of vegetation characteristics and weather conditions on growth rates. One difficulty in measuring potential vegetation differences between edge and interior forest is the binary classification of forest and non-forest used in most land-cover classification schemes. Binary classes ofien misrepresent secondary- growth areas as forested areas. These kinds of errors in describing vegetation at and near 28 nest sites could conceal the effect of forest cover on measures of reproductive success. Fractional forest cover maps use continuous fields of forest density rather than discrete landcover classes. Using hi gh-resolution ETM+ data, fractional cover maps reveal patterns of deforestation and forest degradation (Skole and Qi 2001). The use of fractional cover enables the identification and differentiation of forested areas from secondary growth, typically found along forest edges. In this study I determined how nestling growth rates vary in forest edge and interior habitats. I also examined the effects of different edge types on nestling growth rates because information about the influence of edge types is limited, although species apparently vary in their response to different edge types (Sisk and Battin 2002). The specific objectives of this study were to 1) document how nestling growth rates varied with distance to forest edges, fractional cover, temperature and precipitation patterns, brood size, and date of hatching; 2) determine whether Mayfield’s daily survival rates and overall nest survival rates were reduced in edge habitat compared to interior habitat; and 3) compare rates of brood parasitism in edge and interior distance-classes. I report Mayfield’s daily nest survival rates and overall nest survival rates to facilitate comparisons with previous studies that have focused on measures of nest success to assess Wood Thrush responses to fragmentation. METHODS Study sites—The two study sites are located in southwestern Michigan. Allegan State Game Area (ASGA) (42°34’ N, 85°58’ W) is approximately 20,235 ha in extent in 29 Allegan County and is located 56 km west of Barry State Game Area/Yankee Springs Recreational Area (BSGA) (42°36’ N, 85°27’ W), approximately 8300 ha in size, in Barry County (Fig. 1). At each site in 2002, two plots that ranged from 75 to 80 ha were established in contiguous, mature forest tracts >300 ha in size (Fig 2). Each plot was bordered by a two- laned road on one side. The four plots were chosen because they had similar 1) vegetation composition; 2) vegetation structure; and 3) management history. Forest cover at each site is contiguous, although there are more residential openings in state land holdings at BSGA. Both game areas are surrounded primarily by agricultural land. Disturbance patterns at ASGA and BSGA are similar; both employ timber sales and maintain openings by cutting or mowing. Plots were in areas managed for mature forests and were undisturbed by management practices over the course of this two-year study. ASGA plots were poorly drained, generally flat slopes in bottomland forests on sandy lakeplain and ranged from approximately 210 to 230 m in elevation while the BSGA plots were on well-drained undulating hills in upland forests on interlobate deposits and ranged from approximately 240 to 320 m in elevation. Based on my vegetation measurements, ASGA plots were dominated in the upperstory by red maple (A cer rubrum) (35.6%) and white oak (Quercus alba) (24.5%), in the midstory by red maple (22.2%), ironwood (Carpinus caroliniana) (14.7%), and flowering dogwood (Cornusfloria’a) (14.7%) and in the understory by ironwood (25.7%) and flowering dogwood (16.7%). BSGA plots were dominated in the upperstory by black oak (Quercus velutina) (44.7%), red maple (25.2%), and white oak (10.3%), in the midstory by red 30 Allegan County Barry County Fig. 1. Map of two study sites in southwestern Michigan; Allegan State Game Area (ASGA) in Allegan County and Barry State Game Area/Yankee Springs Recreational Area (BSGA) in Barry County. 31 .560 E coucomoa 2m $.85 WE“ E momma: .353 3:82ro uwamco .wmocfimtn wEmmEQE .3 @8365 cosmonaut @6385 5:5 528 Ecomoaou 5on fan— aoE >93 some Co 3:36:09 2: 0:250 moxom .SwwEBE 5283558 5 20mm: 3?. Ecocmohuom mwctam ooxcm>R2< oEmO 25m 53m Am: 28 Ao mamas: +2Hm b Ham—2:5 Eob no>toe .550 cozflowg :oouw Baotofim .N .wE 32 maple (38.8%) and flowering dogwood (17.6%), and in the understory by red maple (30.7%) and sassafras (Sassafras albidum) (14.6%). Study species—The Wood Thrush was chosen as the focal species because 1) it is considered a forest-interior species (Whitcomb et al. 1981; Brittingham and Temple 1983; Robbins et al. 198%); 2) substantial populations nest at the study sites and so may be important source populations in the Midwest (The Nature Conservancy and MI DNR 2000); 3) their nests can be located and monitored in the forest understory and midstory; 4) previous studies have found that they are negatively affected by forest fragmentation (Brittingham and Temple 1983; Donovan et al. 1995; Hoover et al. 1995; Robinson et al. 1995; Hoover and Brittingham 1998; Weinberg and Roth 1998; Wilson et al. 1998; Burke and N01 2000; F auth 2000; Ford et al. 2001); and 5) comparable demographic research has been conducted on Wood Thrush in other regions (reviewed by Hahn and Hatfield 1995) Nest searching and monitoring.— Active nests of Wood Thrush were located in 2002 and 2003 from early May through late August. Nests were found by searching plots every two days and observing singing males, listening for female vocalizations, and following birds carrying nesting material or food. Locations of nests were marked with flagging at a distance 210 m from the nest and their coordinates recorded using a global positioning system (GPS) (Garmin 12XL, GARMIN Corporation, Olathe, Kansas) using the position averaging function to improve accuracy to within 5 m of the nest’s true location. 33 Nests were checked every two days for predation and brood parasitism events until the young fledged or the nest failed. In accordance with Ralph et al. (1993), nests were categorized as abandoned, successful, depredated, or failures resulting from weather. Nests were considered successful if they fledged at least one young. Adults were unmarked. Hence, renesting attempts were not identifiable. The following parameters of reproductive success were calculated: 1) daily nest survival rates (DSRS); and 2) overall nest survival rates, derived by the Mayfield method (Mayfield 1975) for the entire nesting period. Nestling growth measurements. — Mass and measurements of tarsi and wing chords were recorded each time the nest was checked until the last time the nestlings were found in the nests, or until two days before the projected fledging date (Roth et al. 1996). Because Wood Thrush typically hatch asynchronously, I waited to measure nestlings until all eggs had hatched to ensure my attending the nest did not reduce hatching success. I identified individual nestlings by marking nestlings on a unique toe with a nontoxic marker. The marks were not permanent and had to be reapplied at each visit. The ages of nestlings are given in brood days, where brood day 0 is the hatching day. All nestlings were weighed with a digital scale (ProScale 250, My Weigh Scale Company, Phoenix, Arizona) to the nearest 0.1 g. The unflattened right and left wing chords were measured to the nearest 0.5 mm using a steel wing ruler with a feather stop, and the right and left tarsi were measured to the nearest 0.1 mm using dial calipers. On the final visit to the nest, nestlings were fitted with US. Fish & Wildlife metal bands. Nestlings were only measured from nests that could be reached with a camouflaged 3 m stepladder; typically 34 those nests were S6 m high. The reachable nests constituted 55% (58 of 105) of the active nests that survived the incubation period. Weather Data—I obtained weather data because growth rates of several species have been shown to be negatively affected by heavy rainfall (McCarty and Winkler 1999; Bradbury et al. 2003) and cold weather (McCarty and Winkler 1999) and positively affected by warm weather (Bradbury et al. 2003). Daily precipitation (in) and maximum and minimum temperatures (°F) were obtained from weather stations located nearest to ASGA and BSGA through the Midwestern Climate Information System (MICIS) obtained from Michigan State University’s Agricultural Weather Office. For each nest for which I had growth measurements, I calculated the average maximum and minimum temperatures and the total amount of precipitation received during each nest’s nestling period. Distance to Forest Edge.— Forest edges were defined as the outer perimeter of a forest tract, including two-lane roads (TLR) and powerline corridors (PC). Wildlife openings (WO) maintained by mowing, recent clearcuts (RC) (_<_ 5 years old), and swamps/marshes (M) were classified as edges if their diameters were three times the height of the adjacent forest canopy (Paton 1994) and if there was an abrupt change in vegetation structure between them and the adjacent forest (Flaspohler et al. 2001). The shortest distance from each nest to the nearest forest edge was determined using GPS locations of nests and l m resolution 1998 Digital Orthophoto Quadrangles in ArcView (ESRI, Inc. 1998). Aerial imagery was obtained from the Michigan Department of Natural Resources. I ground- 35 truthed all plots and determined that land-cover had not changed significantly between 1998 and 2001, which was later verified by land managers. Vegetation data—The vegetation structure and composition of each study plot was assessed in early July of 2002 and 2003 after vegetation had reached maximum growth, at six randomly chosen sampling points per plot. I sampled the same six points in each year. Centered on each point, a sample circle of radius 10 m was established. Vegetation within each circle was divided into four main layers, upperstory (>15 m), midstory (5-15 m), understory (0.5-5 m), and ground cover (<05 m) (Nott 1999) and vegetation characteristics were measured by using a modified method of James and Shugart (1970). For each plot the cumulative data from the six sample circles were used to determine relative dominance, tree density/hectare, and total basal area in m2 by species for the upperstory, midstory, and understory vegetative layers. In addition, I recorded estimated canopy height, leaf litter depth, and percent canopy and ground cover (James and Shugart 1970) at each sampling point. The mean of each vegetation measure for each plot was calculated from the 2002 and 2003 data. Land-cover.—-—Remotely sensed imagery acquired with Landsat 7 ETM+ on August 31, 2002 was used to derive fractional green vegetation cover (fc) for each study area. The spatial resolution of these images was 30 m. Before derivation of fractional cover, the images were corrected for atmospheric effects to reduce the atmosphere-induced noise in the estimation of vegetation dynamics caused by solar angle and cloud cover. After atmospheric corrections, I used a linear mixture model to relate fractional cover with the normalized difference vegetation index (NDVI), a spectral vegetation index indicating 36 the amount of greenness in a pixel, following Qi et al. (2000): (076-090 nm (Near lnfrared))—(0.63—0.69 nm (Red)) 0.76-0.90 nm (Near Infrared) NDVI =( _ Now—Now,oil ° NDVIVeg—NDWsoi. where NDVISOn is the NDVI value of an area of bare soil (0) and NDVIvcg is the NDVI of a fully vegetated pixel (1). Therefore, fractional cover estimates range from 0-1. In order to calculate fractional cover I identified areas I knew to be 100% forest cover and areas that lacked vegetation. NDVIveg was determined to be 0.7, therefore, NDVIveg _>_0.7 represented 100% forest cover. NDVISOH was determined to be 0.3, therefore, NDVlson $0.3 represented 0% forest cover. These adjustments to my definition were made because there was no land completely void of vegetation in August in my study area to use as a reference for 0% forest cover, only sparsely vegetated agricultural fields, which had an NDVI of 0.3. Fractional cover images were overlaid on point coverages of nest locations and 50 m radius buffer regions were created around each nest location. Percent fractional cover of each pixel within the buffer region was averaged to obtain one value of fractional cover for each nest. Data Analyses—Logistic growth curves were fit to measurements of nestlings by brood day and growth rate constants were calculated for mass, right and left wing chord, and right and lefi tarsus length for each brood (Ricklefs 1967) following Peters (1983): A M(x) = A 1+ [— — l]e(_K*x) M(0) 37 where x is the nestling age, M(x) is the measurement at age x, A is the asymptotic value, M(O) is the initial measurement on brood day 0, and K is the grth rate constant. I used the nonlinear regression procedure (SAS Institute Inc. 2002) to fit the curves. To avoid pseudoreplication, grth rate constants were calculated for each brood because measurements of nestlings from the same nest cannot be treated as independent (Hurlbert 1984). Curves were fit to observations of the growth of all nestlings of a brood from brood day 0 to 10. However, nestlings may still be increasing in mass on day 10. Since I did not capture adults at my study sites I used adult asymptotic values for the mass, wing, and tarsus lengths from data of after-hatching year males and females from Roth et a1. (1996). There are no reports of geographic variation in Wood Thrush adult body size (Clement 2000). The same adult asymptotic value for each measurement was entered into the logistic growth equation for every brood (Starck and Ricklefs 1998). Regression trees were used to initially investigate which predictor variables were important for each growth variable and to identify interactions between predictor variables (CART® 5.0, Salford Systems, 2003). I used regression trees to examine patterns of variability in the growth rate of nestlings, which is an alternative to multiple regression and analysis of variance models (Roff and Roff 2003). Regression trees make fewer assumptions than procedures that rely on parametric statistics. Ecological data ofien contain categorical and numeric predictor variables, interactions and multicollinearity between variables, missing values, and nonlinear relationships (De’ath and Fabricius 2000). Regression trees make no assumptions about data distributions or the relationship between predictor and response variables, while still allowing for the identification of interactions (Andersen et al. 2000; Fan et al. 2003). Trees enable the 38 exploration and identification of patterns not detected by standard regression models and helps to limit the number of variables used in traditional regression models (Andersen et al. 2000; Grubb et a1. 2002; De’ath and Fabricius 2000). I specified 11 potential predictor variables in the regression tree models for nestling growth. The predictor variables included in the model were year (YEAR), date of hatching (DOH), plot, site, distance to forest edge (DIST), edge type (ETYP), fractional cover (F RAC), brood size (BS), average high (HTEM) and low temperatures (LTEM) during the nestling period, and total precipitation (TPRE) during the nestling period. The data were sequentially split using the least squares fitting method based on the sum of the residual deviations. The optimal tree size was chosen using a series of 10-fold cross- validations to select the tree size with the lowest estimated relative error (Breiman et al. 1984) and to assure the reliability of the fitted tree (Andersen et al. 2000). Cross- validation partitions 90% of the data into a training dataset used to fit the tree and 10% of the data into a validation dataset and is recommended for datasets with moderate sample sizes (Andersen et al. 2000). Tree mobiles were produced for each growth response variable and I report the relative error of each model. The importance of each variable was determined by the model’s improvement with the inclusion of the predictor variables as splitters and the percent variance explained by each predictor variable for each model. Interactions between variables are identified on tree mobiles when there is more than one split node on the same branch. In the language of regression tree models, the root node of the tree is the very top node of the tree that contains all observations of brood growth rates. The root node is subsequently partitioned into split nodes based on the explanatory power of the specified 39 predictor variables. The splitting criterion is the decision rule that is used to partition the observations at a split node into either two terminal nodes or nodes with further binary splitting. A terminal node is the endpoint when variability in brood growth rates cannot be explained with improvement to the model by additional splits (Breiman et al. 1984). Trees were assessed for potential outliers, which were identified as terminal nodes containing a single observation (Breiman et al. 1984). The improvement to the model was compared with and without the potential outlier. Outliers may have a large influence on least-squares estimation so we removed them if they resulted in a significant reduction in the residual sum of squares (Breiman et al. 1984). To test for significant relationships between each growth variable and the predictor variables and interactions identified in the regression trees I used stepwise multiple regression. I restricted the analysis to include only those variables and interactions identified in the regression trees to ensure that there were at least ten observations per growth variable. The significance level specified for the selection of predictor variables to enter the stepwise regression model was 0.2 and 0.05 to stay in the model. Assumptions of normality and multicollinearity were tested by examination of normal probability plots and variance inflation factors. T -tests were used to compare weather variables between sites and years, fractional cover between distance-classes, and brood sizes for nests for which we were able and unable to obtain measurements of nestlings. All statistical analyses were conducted with SAS procedures unless otherwise indicated (SAS Institute, Inc. 2002). I considered P $0.05 significant. 40 Daily survival rates and nest survival rates were estimated for each year for each plot and site and for each year for both distance-classes within sites by the Mayfield method (Mayfield 1975) with exposure terminated using the midpoint approach for nests of known fate (Manolis et al. 2000). I used a 13—day incubation period and a 12-day nestling period to determine exposure days (Roth et al. 1996; Clement 2000). Standard errors and confidence intervals for Mayfield estimates were calculated according to Johnson (1979). To compare daily survival rates of nests located in forest edge and interior habitats, nests were grouped into two distance-classes, 0-99 m (edge) and 2100 m (interior) from the nearest forest edge, which contained approximately the same number of nests. Previous research examining the effect of distance to forest edge on the nesting success of Wood Thrush compared four distance-classes (550 m, 51-100 111, 101-200 m, and >200 m) (Burke and N01 2000) and three distance-classes (£50 m, 51-100 m, and >100 m) (Hoover et al. 1995; Fauth 2000). However, I had too few nests within 50 m of an edge to include this additional distance-class. Daily survival rates for the entire nesting period were compared between plots within a year, between years within a site, between sites, and also between years within a distance-class for each site and between distance-classes for each site with x2 tests in the software program CONTRAST (Hines and Sauer 1989). Abandoned nests were excluded from these analyses. When rates were not different between, for example, plots within a year, I combined exposure days from plots to obtain one rate for that site and year. I combined exposure days from 2002 and 2003 for each site when rates did not differ between years. 41 Cox proportional hazard regression was used to test the influence of year, site, distance to forest edge, and fractional cover on nest survival (Allison 1997). Exposure days were evaluated as risk factors. Nest fate was included as a censor variable (0 for successful and uncertain, 1 for failure). In the regression model, the hazard of any nest is a fixed proportion of the hazard for any other nest and there is no underlying distribution assumption (Allison 1997). The main advantages of Cox regression are that constant mortality is not assumed and exposure days, rather than nest fate, are used as the response variable (Manolis et al. 2000). Hence, nests observed merely during the incubation or nestling period, alone, were not given the same weight as nests observed throughout the entire nesting period. RESULTS I found 175 Wood Thrush nests at the two study sites over the two seasons of the study. Thirty-nine nests were located at ASGA in 2002 and 49 in 2003. Thirty-five nests were located at BSGA in 2002 and 52 in 2003. F orty-six percent of nests (l 8 of 39) were successful at ASGA and 49% (17 of 35) were successful at BSGA in 2002. F ifty-nine percent of nests (29 of 49) were successful at ASGA and 39% (20 of 52) at BSGA in 2003. All nesting failures were attributed to predation, with the exception of one failure because of brood parasitism at ASGA in 2002. Twenty-two nests were depredated during the incubation period and seven during the nestling period at ASGA over both years. Twenty-nine nests were depredated during 42 the incubation period and 15 during the nestling period at BSGA over both years. Ten nests were abandoned prior to egg laying at ASGA, as were six nests at BSGA. Nestling growth. —- I measured 91 nestlings from 28 nests at ASGA; 35 nestlings from 1 l nests in 2002 and 56 nestlings from 17 nests in 2003. At BSGA, 102 nestlings were measured from 33 nests; 65 nestlings from 20 nests in 2002 and 37 nestlings from 13 nests in 2003. Nestlings from three nests, two at ASGA and one at BSGA, were excluded from analyses because early depredation resulted in too few measurements of nestlings to fit a growth curve. The percentage of variation in the growth response variables explained by the set of predictor variables in each regression tree model and by each predictor variable within each model is given in Table l. The trees constructed for growth rates of the left and right tarsi were very similar to each other and included three predictor variables, indicating edge type, distance to forest edge, and total precipitation influenced nestling growth (Table 1, Fig. 3). One observation was identified as an outlier and removed from both the left and right tarsi trees. Nests were first split into two major groups based on different edge types. For those nests nearest to forest edges formed by swamps/marshes, two-laned roads, and wildlife openings (split node 2), growth rates were significantly more rapid S3 77 m from the forest edge, compared to nests >377 m from the forest edge. For nests adjacent to powerline corridors or recent clearcuts edge types, growth was more rapid for nests receiving 30.95 inches of rainfall (terminal node 3) than nests receiving >0.95 inches of rainfall. Growth of both the left and right tarsi was most rapid in nests adjacent to powerline corridors and recent clearcuts receiving £0.95 inches of rainfall during the 43 TABLE 1. The percentage of variation in the growth response variables explained by the set of predictor variables in each regression tree model and by each predictor variable within each model of the growth of nestling Wood Thrush. Sample sizes of nests used in regression tree analyses are given in parentheses next to each growth response variable. a Lift Wing (51) Right Wing (52)Left Tarsus (45)Right Tarsus (45) Mass (51) 53.3 Overall 12.6 Overall 33.4 Overall 35.0 Overall 36.2 Overall 40.4 HTEM 10.1 TPRE 1 18.3 ETYP 18.2 ETYP 19.7 TPRE 12.9 LTEM 2.5 DIST 7.7 TPRE 8.9 TPRE 116.5 LTEM % 7.4 DIST 7.9 DIST a HTEM=Average High Temperature during the nestling period, LTEM=Average Low Temperature during the nestling period, TPRE=Total Precipitation received during the nestling period, DIST=Distance from nest to forest edge, ETYP=Edge Type. 44 .228 E comcomoa 8m $85 was E momma: .65 Em: 2: 28 E .2 was ES a2 2: 28 S; «o Hobo 03228 .9662 a Ea move: 35828 58 was 2.2508 85 2:. 60:3 wfizmoc 05 waist 3388 cougfiooi gob" man; .030 588 8 “mo: Boa oogumanhma .09? omwmn 351m gamma—22 €282,558 E 88 can NOON wad—6 83225.» 886.123 9 smash. woo? wfiumon mo mamas Em: Am can mama: a2 A< mo 82 538w mace—2 259: on; .m .wE N212 mnz mnz $12 ~72 mnz muz 812 E: u a}. 84.0 n 94 33 u 9i 2.2 " o>< $3 1 9i S; u 93 92 n 34 $3 1 03 «mod n Em 33 u Em am: 1 Em :3 n Em mm: 1 Em mm: 1 Em ~82 n Em $3 1 Em v 252 m 082 N 082 _ 082 v 082 m 082 N 082 F 082 $252.82. BEES. BEES. BEES. SEEDS, iigoh BEES issuer - - — - - 1 u mac A man: 3o 1 m2“: Em? A Ema .35 1 ma mm o A #2: mad 3 #E Ems A Ema :35 3 ma .141. .Jl. .l_lll. ri_|1. 32 «~12 8.12 «~12 a; u 94 82 n 93. as; u 9.4 mag n .23 mag u Em Sod “ Em 53. u Em 53 n Em m 082 N 082 m 082 m 082 u - - u 629: 1 math 83.5»..5 u mews 82.0”: n math 852:2; u merm T _ _ . _ _ m4 1 2 mv u 2 22 u 9i 23 u 94 a; u Em 83. u Em _ £52 _ 082 82E £92 a mama... :3 2 45 nestling period. Daily mean lengths of the right and left tarsi of the nestlings were plotted for edge type, the most important splitting variable (Fig. 4), and for edge (0—99 m) and interior (>100 m) distance-classes (Fig. 5) to illustrate the growth pattern of tarsi. Given the shape of the curve and the asymptotic value of 31 mm, tarsi had reached adult length by brood day 10. Edge type, distance to forest edge, total precipitation, the interaction between edge type and distance to forest edge, and the interaction between edge type and total precipitation were regressed against left tarsi growth rate constants. Stepwise multiple regression indicated a negative correlation between the interaction between edge type and distance to forest edge and left tarsi growth rate constants, r2=0.16, P=0.01 and right tarsi growth rate constants, r2=0.12, P=0.03; and a negative correlation between edge type and left tarsi growth rate constants, r2=0.10, P=0.03 and right tarsi growth rate constants, r2=0.09, P=0.05; no other variables entered into the model. Broods in nests adjacent to edges formed by swamps/marshes, two-laned roads, and wildlife openings had slower left tarsi growth rates than those adjacent to powerline corridors and recent clearcuts. The interaction between edge type and distance to forest edge indicates that the effect of distance to forest edge on left tarsi growth rates depends on the nearest edge type. The tree constructed for nestling growth rate of the left wing included two predictor variables, suggesting average low and high temperatures influenced nestling growth of the left wing (Table 1, Fig. 6). The first split divided nests with an average low temperature during the nestling period of £55°F from those with an average low temperature higher than 55°F. The low temperature nests were then divided at Split node 2 into nests that had an average high temperature during the nestling period of £82°F and 46 A) Left Tarsus 40 T I I 1 I E 2 _ $.30 1 1 i (D 3 e 1 1 (U j. . ‘5 20- 1 1 a _J 1 1 ETYP f . M,TLR,WO 1o 1 1 -1 1 1 . PC,RC 0 2 4 6 8 10 12 Brood Days B) Right Tarsus 40 1 1 1 I I (A) O l l——1.-'.-'—H l—f‘t—i M 1+“. 1 Right Tarsus [mm] M O l H—o—H O M, TLR, W0 10 1 1 1 1 1 . PC,RC 0 2 4 6 81012 Brood Days i I 1 ETYP Fig. 4. Growth curves of Wood Thrush nestlings grouped by edge type, the most important splitting variable for (A) left tarsus and (B) right tarsus. Dots indicate the average measurement (mean i SD). Day 0 corresponds to the day the first nestling hatched. Filled circles represent nests nearest to edges formed by swamps/marshes (M), two-laned roads (TLR), and wildlife openings (WO) (n=31) and open circles represent nests nearest to edges formed by powerline corridors (PC) and recent clearcuts (RC) (n=22) 47 A) Left Tarsus 40 I 1 1 1 E30- 1 I i - (D 3 8 1 1 (U j. 8 20— i 1 — _j .. 10 . 1 l L l O 2 4 6 8 1O 12 BroodDays B) Right Tarsus 40 T I 1 I E 30— i i - . ~ 1 3 U) a 1 1 1— ., £220— 1 - m 1 10 . l l l l O 2 6 8 1O 12 Brood Days EDGE 0 EDGE S INTERIOR EDGE 0 EDGE INTERIOR Fig. 5. Growth curves of Wood Thrush nestlings grouped by edge (0-99 m) and interior (>100 m) distance-classes for (A) left tarsus and (B) right tarsus. Dots indicate the average measurement (mean _+_ SD). Day 0 corresponds to the day the first nestling hatched. Filled circles represent data from nests in edge habitat (n=24) and open circles represent data from nests in interior habitat (n=32). 48 .2200 E @2585 08 £85 £5 E memes: .mE3 Emc E2 28 En; mo 5:0 03:28 13282 a Ea mE3 $2 05 8.2 an; mo more 03232 12668 2 ES 860: 35828 085 9E 8:908 om: E H dozen mEzmo: o5 macaw ©3602 coumzfiooi :3an man: .omco $88 8 “mo: 80c oocfimauhfla .motoa mam—2mm: 2: mEEc 82259228. Ema om8o>>Em_m Am 953 to; 2 49 nests that had an average high temperature greater than 82°F. The tree indicates nestling growth of the left wing was most rapid in nests with average low temperatures $55°F and average high temperatures exceeding 82°F (terminal node 2). Average low temperature, average high temperature, and the interaction between average low and high temperature were regressed against left wing growth rate constants. No variable met the 0.2 Significance level for entry into the model. The tree constructed for growth rate of the right wing included two predictor variables, suggesting nestling growth of the right wing was influenced by distance to forest edge and total precipitation (Table 1, Fig. 6). Analyses split nests into two major groups based on a distance of 282 m from the forest edge. Nests located 5282 m from the forest edge were Split from nests >282 m from the forest edge (split node 2). Those nests greater than 282 m from the forest edge were further split into two groups, one of which received $0.69 inches of precipitation where growth rate was higher and one of which received >0.69 inches. Nestling growth of the right wing was more rapid in nests greater than 282 m from the forest edge and receiving £0.69 inches of rainfall during the nestling period (terminal node 2) compared to nests less than 282 m from the forest edge. Daily mean lengths of the right and left wing chords of the nestlings were plotted as a function of edge (0-99 m) and interior (>100 m) distance-classes (Fig. 7) to illustrate the growth pattern of wing chords. Given the shape of the curve and the asymptotic value of 106 mm wings had reached 56% of adult length by brood day 10. Distance to forest edge, total precipitation, and the interaction between distance to forest edge and total precipitation were regressed against right wing growth rate constants. Stepwise multiple regression indicated a negative correlation between total 50 A) Left Wing 70 1 I I I F 60- .. - 01 O I H—O—H H—H 10%-I 1 0) O F I—CrH Left Wing Chord [mm] A O I 1—+—-O—H l-—+—O—+—I l N O 1 1+1! Hod—I l E _ EDGE 0 EDGE '0 INTERIOR .3 O Il'lll O _ _ _ _ __ Brood Days B) Right Wing 70 I I I f I 60“ r 1 50— I I r 40— 1 30— I 1 — i 20 1 Right Wing Chord [mm] 10 g I 1 EDGE “2 * . EDGE 0 1 1 1 1 1 c; INTERIOR 0 2 4 6 8 10 12 Brood Days Fig. 7. Growth curves of Wood Thrush nestlings grouped by edge (0-99 m) and interior (>100 m) distance-classes for (A) left wing chord and (B) right wing chord. Dots indicate the average measurement (mean i SD). Day 0 corresponds to the day the first nestling hatched. Filled circles represent data from nests in edge habitat (n=24) and open circles represent data from nests in interior habitat (n=32). 51 precipitation and right wing growth rate constants, r2=0.06, P=0. l 8; no other variables entered into the model. However, total precipitation did not meet the 0.05 significance level to stay in the model. The tree constructed for nestling mass growth included two predictor variables, suggesting total precipitation and average low temperature influenced nestling mass growth (Table 1, Fig. 8). The first split divided nests in areas that received >2.54 inches during the nestling from those that received £2.54 inches of precipitation. Nests in areas that received 32.54 inches of precipitation were further split into those that had average low temperatures during the nestling period of S62°F and those that had average low temperatures of >62°F. Mass growth rates were higher for nestlings in nests receiving 32.54 inches of precipitation with average low temperatures during the nestling period >62°F (terminal node 2). Daily mean masses of the nestlings were plotted for total precipitation, the most important splitting variable (Fig. 9), and for edge (0-99 m) and interior (>100 m) distance-classes (Fig. 10) to illustrate the growth pattern of mass. Given the shape of the curve and the asymptotic value of 50g, mass had reached 70% of adult weight on brood day 10. Total precipitation, average low temperature, and the interaction between total precipitation and average low temperature were regressed against mass growth rate constants. Stepwise multiple regression indicated a positive correlation between total precipitation and mass growth rate constants, r2=0.28, P=0.001; no other variables entered into the model. Nests receiving higher amounts of total precipitation during the nestling period had more rapid mass growth rates than those with lower amounts of total precipitation. 52 mi 154 ' " 1 i . N ode 2 Terminal Node 3 STD x 0.097 Av - 0494 STD: 0.105 ii.- 47 Avg . 0.674 N x 4 I I I LTEIR: 61.82 LTEH ’1 61.82 Terminal Terminal Node 1 Node 2 sm = 0.080 STD = 0.167 Avg = 0.484 Avg x 0.712 N:45 N=2 Fig. 8. Tree mobile relating mass growth rate of nestling Wood Thrush to predictor variables during 2002 and 2003 in southwest Michigan. TPRE =Total Precipitation received during the nestling period, LTEM =Average Low Temperature during the nestling period. The tree mobile for mass had three terminal nodes and a residual relative error of 1.77. Images in this thesis are presented in color. 53 50 I I I l I 00 O l I—+--o-H 1—-o—H 1—+--—-I—o—-I :11 I—H 1 Mass [9] N O l II—O-I I—I-H l 10” I TPRE ”—‘ . <=2.54 in o 1 1 1 1 1 >254 in 0 2 4 6 8 10 12 Brood Days Fig. 9. Growth curves of Wood Thrush nestlings grouped by total precipitation, the most important splitting variable for mass. Dots indicate the average measurement (mean i SD). Day 0 corresponds to the day the first nestling hatched. Filled circles represent nests that received 52.54 inches of rainfall during the nestling period (n=47) and open circles represent nests that received >2.54 inches of rainfall (n=4). 54 50 I I I I I Mass [9] (A) O 1 H+4—I I—I—O—Il I——+—-o—+—-1 H~+—1 H—O—II l N O l 10L 1 ‘ EDGE ‘5 . EDGE 0 1 1 1 1 1 INTERIOR 0 2 4 6 8 1o 12 Brood Days Fig. 10. Growth curves of Wood Thrush nestlings grouped by edge (0-99 m) and interior (>100 m) distance-classes for mass. Dots indicate the average measurement (mean :1: SD). Day 0 corresponds to the day the first nestling hatched. Filled circles represent data from nests in edge habitat (n=24) and open circles represent data from nests in interior habitat (n=32) 55 Average low temperature was significantly lower in 2003 at ASGA compared to 2002 (t=2.09, df=215, P<0.04). All other weather variables did not vary between years at either site, which is consistent with my finding no yearly variation in nestling growth. Year was not selected as an important predictor variable in the regression trees for any of the growth response variables. Brood Sizes of nests used in growth rate analyses were Si gnificantly larger (mean=3.22) than those in nests for which we could not measure nestlings (mean=2.83) (t=2.42, dfi94, P<0.02). DSRs.— For all nests combined within the nesting period, DSRS did not differ between years for each plot or between plots within years at ASGA or BSGA (Table 2). Therefore, data from plots were pooled within sites. DSRS did not differ between years at ASGA (2002: 0.97 i 0.01; 2003: 0.98 i 0.01; x2=2.10, df=1, P=0.15) or BSGA (2002: 0.97 i 0.01; 2003: 0.95 i 0.01; x2=2.76, df=1, P=0.10). After pooling years for each site, DSRS did not differ between sites (ASGA: 0.97 2 0.01; BSGA: 0.96 2 0.01; x2=3.13, df—‘l, P=0.08). DSRS in ASGA edge habitat did not differ between years (x2=2.86, df=1, P=0.09) but yearly differences in DSRS in BSGA edge habitat were significant (x2=6.30, df=l, P=0.01). DSRS did not differ between years in ASGA interior habitat (x2=0.004, df=1, P=0.95) or BSGA interior habitat (12:0.12, df=1, P=0.73). Therefore, years were pooled for ASGA edge and interior distance-classes and for BSGA interior habitat to make comparisons between edge and interior distance-classes within sites. Pooled data showed no significant difference between DSRS in edge and interior distance-classes at ASGA 56 TABLE 2. Daily survival rates (DSRS) 2 SE during the entire nesting period of Wood Thrush nests among plots and within years for each site. * Year Site Plot Exposure Days DSRS x 2 P 2002 ASGA l 1 10.5 0.95 2 0.02 1.10 0.29 ‘_.2002--_.A.S_GAW.2 ._ -3699”- V9:9729:91-.--”_ ._ 2002 BSGA 3 291.5 0.98 i 0.01 0.86 0.35 -._2092H---B.SGAW .41 , 204-5., .. 0:96.209!__....... 2003 ASGA 1 121.0 0.97 i 0.02 0.77 0.38 2903-83913- .2 ._ .503,-.0.._ . . ,,..Q-,9.8,=#.9-91__..... 2003 BSGA 3 272.0 0.94 d: 0.01 0.81 0.37 2003 BSGA 4 307.5 0.96 d: 0.01 * Standard errors (SE) calculated according to the methods of Johnson (1979). 57 (Table 3). In 2002 and 2003, DSRS did not differ between edge and interior distance- classes at BSGA (Table 3). Nest survival. — Overall nesting success varied between years and between sites (Table 4) and between edge and interior distance-classes within sites (Table 5). The risk of nest failure was significantly influenced by site (x2=3.91, df=1, P=0.04) in the Cox regression and not influenced by year, distance to forest edge, or fractional cover. The risk of nest failure was 42.4% greater at BSGA than at ASGA over both years, or in other words, nests at BSGA were 1.7 times more likely to fail than nests at ASGA. 1 found no difference in fractional cover of nest sites between distance-classes (t=0.43, df=138, P=0.67). Brown-headed Cowbird parasitism. — I only included nests observed during the incubation period in the calculation of parasitism frequency. The inclusion of nests found during the nestling period may lead to underestimation of parasitism levels if the eggs of Brown-headed Cowbirds do not hatch or if predation rates differ among parasitized and unparasitized nests (Robinson and Robinson 2001). Parasitism frequency was higher at ASGA (21.7%; 15 of 69 nests) than at BSGA (1.5%; l of 68 nests). In ASGA, six nests were parasitized in edge habitat and nine nests in interior habitat. In BSGA, only one nest was parasitized in interior habitat. Seven of the 15 nests parasitized at ASGA successfully fledged at least one Brown-headed Cowbird, as well as Wood Thrush young, and the one parasitized nest at BSGA was depredated late in the nestling period. 58 TABLE 3. Daily survival rates (DSRS) i SE during the entire nesting period of Wood Thrush nests found in 2002 and 2003 in edge and interior distance-classes for each Site and year. * Exposure Days (# nests lost) DSRS Year Site Edge Class Interior Class Edge Class Interior Class I 2 P 2002 ASGA 236.5 (9) 241.0 (8) 0.96 i 0.01 0.97 i 0.01 0.08 0.77 2002 BSGA 152.0 (2) 344 (13) 0.99 2 0.01 0.96 i 0.01 3.17 0.07 2003 ASGA 406.5 (6) 217.5 (7) 0.99 d: 0.01 0.97 2 0.01 1.68 0.19 2003 BSGA 133.5 (10) 446.0 (19) 0.93 :1: 0.02 0.96 d: 0.01 1.71 0.19 Pooled ASGA 643.0 (15) 458.5 (15) 0.98 :1: 0.01 0.97 d: 0.01 0.84 0.36 Pooled BSGA 285.5412) 790.0 Q2) 0.96 2 0.01 0.96 :t 0.01 —- —- * Standard errors (SE) calculated according to the methods of Johnson (1979). 59 TABLE 4. Variation in Wood Thrush nest survival rates among sites and years. a N0. of active Nesting Success N0. of active Nesting Success Nesting Site nests 2002 2002 b nests 2003 2003 b success 2002-03 ASGA 36 0.408 (17) 42 0.575 (13) 0.498 Plot 1 8 0.228 (6) 8 0.448 (4) 0.331 Plot 2 28 0.464 (1 l) 34 0.606 (9) 0.543 BSGA 32 0.465 (15) 49 0.297 (29) 0.358 Plot 3 19 0.533 (7) 25 0.239 (16) 0.357 Plot 4 13 0.339 (8) 24 0.356 (13) 0.354 a Nesting survival rates calculated according to Mayt'ield (1975). b . . . Numbers of failed nests are given in parentheses. 60 TABLE 5. Variation in Wood Thrush nest survival rates between edge and interior distance- classes within sites. a N0. of active Nesting Success N0. of active Nesting Success Nesting Site nests 2002 2002 b nests 2003 2003 b success 2002-03 ASGA 0-99 m 19 0.383 (9) 23 0.685 (6) 0.555 >100 m 16 0.430 (8) 19 0.389 (7) 0.424 BSGA 0-99 m 8 0.707 (2) 15 0.168 (10) 0.360 >100 m 24 0.382 (13) 34 0.358(19) 0.363 a Nest survival rates calculated according to Mayfield (1975). b Numbers of failed nests are given in parentheses. 61 Vegetation data. — The vegetation structure of the two sites was similar and is summarized in Table 6. Vegetation characteristics and composition of the upperstory, midstory, understory, and ground cover are summarized in Tables 7 and 8. 62 TABLE 6. Vegetation structure of each site in southwestern Michigan. 8 SITE b Variable ASGA 11 2) BSGA (12) Percent Canopy Cover 93 86 Percent Ground Cover 56 38 Mean Canopy Height [m] 17.8 17.2 Mean Shrub Height [m] 0.4 0.3 Mean Herbaceous Vegetation Height [m] 0.4 0.2 Mean Leaf Litter Depth [cmL 2.3 2.5 a Numbers represent means. Numbers of vegetation sample pomts wrthm each Site are given m parentheses. 63 TABLE 7. Vegetation characteristics of the four vegetative layers for each site in southwestern Michigan. SITE " ASGA BSGA Variable (12) (12) Upperstory (>15 m) Tree Density/Ha 305 253 Total Basal Area [m2] 1 1.0 11.3 Midstory (5-15 m) Tree Density/Ha 546 556 Total Basal Area [m2] 1.4 1.6 Understog (0.5-5 m) Tree Density/Ha 2066 2135 Total Basal Area [mg] 0.3 0.7 M Tree Density/Ha 2918 2944 Total Basal Area [m2] 12.8 13.5 8 Numbers of vegetation sample points within each site are given in parentheses. 64 TABLE 8. Vegetation composition of the four vegetative layers for each site in southwestern Michigan expressed as percentages. a SITE b Vegetation Iypes ASGA (12) BSGA ( I 2) Upperstog (>15 m) Conifer 0.1 0.4 Broad-leaved 99.9 99.6 Forbs & Ferns 0.0 0.0 Grass-like 0.0 0.0 Midstory (5-15 m) Conifer 0.3 3.8 Broad-leaved 99.7 96.2 Forbs & Ferns 0.0 0.0 Grass-like 0.0 0.0 Understory (0.5-5 m) Conifer 0.6 3.4 Broad-leaved 79.0 93.0 Forbs & Ferns 20.4 3.5 Grass-like 0.0 0.1 Ground Cover (<05 m) Live Vegetation 52.1 37.0 Dead Vegetation 45.1 59.9 Total Non- 2.8 3. 1 Live Vegetation (<05 m) Woody 40.0 55.7 Nonvascular 8.4 8.0 Forbs & Ferns 32.3 24.0 Grass-like 19.3 12.3 a Numbers represent means. Numbers of vegetation sample points within each site are given in parentheses. 65 DISCUSSION Variability in Wood Thrush growth rates. — Food availability has been recognized as the most important and most limiting environmental factor influencing nestling growth (Martin 1987). The majority of the environmental influences on growth variability that have been identified in the literature relate to food abundance (Gebhardt-Henrich and Richner 1998). There are several reasons to believe nest sites near edges may have reduced food supplies for nestling Wood Thrush when compared to nest sites in the forest interior. Changes in forest structure at edges may affect leaf litter cover and microclimatic conditions, such as temperature and soil moisture (J olivet 1986; Niemeléi et al. 1993; Murcia 1995; Greenberg and McGrane 1996), resulting in changes in the abundance and composition of invertebrate food supplies available for nesting birds (Schowalter et al. 1981; Niemela et al. 1993). For example, litter-dwelling invertebrate biomass was higher in unharvested forested habitat than in clearcut habitat, which had reduced forest cover (Duguay et al. 2000). For Wood Thrush, which primarily forage for soil invertebrates in leaf litter on the forest floor, it has been shown that they prefer nest sites with canopies of mature trees (Hoover and Brittingham 1998), moist soils, shade, and decaying leaf litter (Roth et a1. 1996; Artman and Downhower 2003). Nest sites with high canopies may have increased amounts of leaf litter available for foraging and moister soils (Bertin 1977). Nest sites in the forest interior are more likely to have these conditions than nest sites near forest edges where there is a greater amount of secondary forest regrowth composed of shrubs and immature trees (Hoover and Brittingham 1998). 66 Wood Thrush nestlings growing in edge habitat may have access to less food compared to nestlings growing in interior habitat, accelerating growth of structures important to procuring food and fledging the nest. Rapid grth of the tarsi may assist in nestling begging, perching, and competitive interactions for food with their siblings (O’Connor 1977). In addition, the first two weeks out of the nest, Wood Thrush perch quietly in dense undergrowth near the ground with minimal short flights (Vega Rivera et al. 1998) and will hop, rather than fly, away from pursuers (pers. obs.). They depend on their legs during the postfledging period for movement when they remain incapable of long flights because wing chords do not fully develop until approximately 15 days postfledging (Roth et a1. 1996). Dutta et al. (1998) found that nestling leg bones of House Wrens had approximately as much calcium as adult bones at fledging, whereas wing bones were not nearly as calcified at first flight. Their results suggest that nestlings may selectively allocate calcium to critical bones to enhance survival (Nilsson and Svensson 1996; Dutta et al. 1998). In food-limited edge habitat this may result in rapid tarsi growth, at the expense of wing chord growth. The most parsimonious hypothesis explaining rapid growth rates is that edges are food-abundant. I found a significant relationship between specific edge types and the growth responses of the left and right tarsi. Tarsi growth was depressed in nests located in forest nearest to edges formed by swamp/marshes, two-lane roads, and wildlife openings, and more rapid in nests nearest to edges formed by powerline corridors and recent clearcuts. Powerline corridors and recent clearcuts contain dense shrubbery at their edges and by mid-season were transformed from abrupt edges into more gradual edges as ground vegetation reached maximum growth. Territories near these edges may have had 67 more food resources than territories near the other edge types, resulting in more rapid growth of nestlings. 1f edges are food-abundant, these results indicate that edges may enhance growth rates and that food availability depends on edge type. In a study conducted in New Jersey, Wood Thrush were more abundant near edges bordered by powerline corridors than narrow, unpaved roads suggesting important differences in edge habitat selection (Rich et al. 1994). Mass and right wing growth rates were related to total precipitation received during postnatal growth. Weather conditions have been shown to directly affect nestling growth and development (Podlesak and Blem 2001). Keller and van Noordwijk (1994) showed that precipitation indirectly influenced the growth of nestling Great Tits through parental adjustments in food provisioning rates when food was either in short supply or foraging time was reduced because of the adverse weather. I found mass growth was significantly more rapid for nestlings developing in nests receiving a high total amount of precipitation during the nestling period. This result suggests that food may have been plentiful because of the rainfall. Growth of the right wing was depressed for nests receiving greater precipitation during the nestling period, but this was only true for nests located greater than 282 m from a forest edge. During periods of food shortage, energy may be allocated to the growth of structures like tarsi with more immediate importance. An alternative explanation is that delivery of food by parents was reduced because of the weather as Keller and van Noordwijk (1994) showed with the Great Tits. However, the stepwise multiple regression analysis did not indicate that total precipitation influenced right wing growth contrary to the regression tree results. 68 McCarty (2001) found a relationship between food supply and mean temperature across an entire breeding season. Specifically, grth rates were higher in warm years when insect abundance was high. I found that mass growth was lower in nests in areas with the lowest total amount of precipitation received during the nestling period and lowest average temperatures. Colder and wetter weather requires nestlings to expend energy on thermoregulation, increasing their food requirements to maintain normal growth. However, the stepwise multiple regression analysis did not indicate that average low temperature influenced mass growth contrary to the regression tree results. Nestling growth rate variables were not affected by brood size or date of hatching. Nestlings from larger broods may suffer reduced rates of growth (Ricklefs 1968; Howe 1976; O’Connor 1978; Kunz and Ekman 2000) because increased intrabrood competition for food reduces the amount of energy allocated to growth (Schifferli 1978) and larger broods have greater food requirements. In my study, brood size ranged from one to five nestlings and declined over the breeding season. A lack of a brood size effect is not surprising because the reachable nests, for which I was able to calculate brood growth rates, most often contained three to four chicks, so my samples do not reflect the variability in brood sizes that I observed during the breeding season. I verified this observation and found that brood sizes of nests used in growth rate analyses were significantly larger than brood sizes of nests we were not able to reach. Hence, my nestling growth results may represent a biased sample of Wood Thrush nests. It is also possible that females may reduce their brood size in response to environmental conditions. Hence, small broods could indicate poor local food conditions and, under these conditions, high growth rates are not expected (Schew and Ricklefs 1998). 69 I found that fractional cover of nest sites did not affect growth. Fractional cover did not vary greatly at my sites, likely because plots had fairly similar vegetation structure and composition and nests were typically built where canopy cover was high. Rather, concealment in the understory and midstory may increase the probability of nest survival (Hoover and Brittingham 1998) and better explain variation in growth. Fractional cover between nests in the edge and interior distance-classes did not differ and was not related to any of the growth parameters. It is possible that the fractional cover algorithm could not detect differences that would be important to nesting Wood Thrush. Comparisons of site survival rates. — One of the strengths of this study is that I assess the risk of nest failure as a function of continuous distances to forest edges and further measure nest survival rates as a function of discrete distance to forest edge classes. In a review of the literature on the effects of edges on nesting success, Paton (1994) emphasized that the experimental design of edge studies varied considerably in both artificial and natural nest studies. In a separate review, Murcia (1995) also indicated inconsistency in design as the major factor in why patterns in edge effects have been hard to discern. Specifically, authors differed widely in their definition of edges and at what distances to investigate edge effects. Most often, experimental designs involve making decisions about the intervals of discrete distance-classes in which to compare variables like daily survival rates in forest edge and interior habitat (Hoover et al. 1995; Hoover and Brittingham 1998; Friesen et al. 1999; Burke and N01 2000; Fauth 2000; Simons et al. 2000; Ford et al. 2001). This approach may conceal important patterns. Investigating 70 variables as a function of distance from edge is easily replicable from one study to the next Daily survival rates for the entire nesting period for Wood Thrush in ASGA and BSGA were comparable to the rates previously reported in other parts of their range (Burke and No] 2000; Donovan et al. 1995; Hoover et al. 1995; Trine 1998; Famsworth and Simons 1999; Simons et a1. 2000; F auth 2000; Fauth 2001; Ford et al. 2001; Robinson and Robinson 2001; Artman and Downhower 2003). I identified no differences in daily survival rates between my two study sites, ASGA and BSGA. However, the Cox regression model identified site as an explanatory variable that influenced the risk of nest failure. The risk of nest failure was much greater at BSGA than at ASGA when data from both years were combined. Cox proportional hazard regression may have been more sensitive to differences in nest survival because constant mortality is not assumed, which was important given that twice as many nests failed during the nestling period at BSGA than at ASGA. BSGA contains younger forest compared to ASGA and has an understory that is less dense. BSGA also has more white pine in the understory and midstory which was used more often as a substrate for nests than at ASGA. Nests in white pine may have been more conspicuous to avian predators because the needles conceal the nest less than deciduous leaves and nests were usually built on the highest branch near the main stem of the tree, an easily detected location (pers. obs.). Comparisons of forest edge and interior DSRs. —I found few edge effects on DSRS at either site. Contrary to my expectations, in 2002, edge habitat had higher DSRs during 71 the nestling period than interior habitat at BSGA. However, this trend was absent when data from both years were combined. In the Cox regression, I treated edge as a continuous variable of distances from each nest to the nearest forest edge rather than discrete distance-classes of <99 m (edge) and >100 m (interior) and found that distance to forest edge did not influence the risk of nest failure. While southwestern Michigan contains forested sites, such as ASGA and BSGA, these landscapes are highly fragmented and surrounded by agricultural land. Land-cover can influence the abundance and diversity of nest predators, which could constrain the importance of proximity to edge effects on nest predation (Thompson et al. 2002). In a study comparing the detection of factors affecting nesting success at multiple scales, Thompson et al. (2002) found that landscape context explained why some studies documented edge effects on predation patterns and why others might have failed to find edge effects. Thus, high regional fragmentation levels may overwhelm potential differences in local edge effects on nest predation. Wood Thrush are highly mobile (Powell 2000) and adults were frequently observed crossing two-lane roads, powerline corridors, and wildlife openings to forage and search for nesting material. In addition, males chose song perches along these edges to attract mates throughout the breeding season. Wood Thrush require different habitats types during different stages of their life cycle, including edge habitats. F ledglings will disperse to secondary-growth forests and forest edges from mature forests where fruits are more available and there is greater vegetative concealment (Anders et al. 1998; Vega Rivera et al. 1998; Vega Rivera et al. 1999; Marshall et al. 2003). In addition, Wood Thrush are increasingly selecting nest sites in residential or urban habitats (Clement 72 2000) and early-aged deciduous habitats, which were previously not used for nesting, have now been considered marginal habitat for breeding (Dettmers et al. 2002). These findings suggest that with the regional loss of habitat and increased fragmentation of existing forests, Wood Thrush may be using suboptimal habitat. While potential differences in local predation patterns between edge and interior habitat may be overwhelmed by the effects of regional fragmentation, local scale effects on nestling growth rates may still be detectable. Brood parasitism. — Longer nesting periods may reduce nestling survival by prolonging the time that nestlings are susceptible to brood parasitism (Petit and Petit 2000). Previous studies have shown that the presence of nestling brood parasites may affect growth and the hatching and fledging success of host young for commonly parasitized species (Marvil and Cruz 1989; Soler and Soler 1991; Dearbom et al. 1998; Burhans and Thompson 2000). No studies have yet determined whether the presence of nestling brood parasites has detrimental effects on the growth of Wood Thrush. However, my sample sizes for parasitized nests were too small to conduct this analysis. Brood parasitism can significantly reduce nesting success of Wood Thrush (Brittingham and Temple 1983; Fauth 2001'; Donovan et a1. 1995). Parasitism rates at ASGA and BSGA were among the lowest reported for Wood Thrush breeding in the Midwest. In a study comparing Brown-headed Cowbird parasitism of Wood Thrush, cowbird abundance and parasitism rates were highest in the Midwest and lowest in the Northeast (Hoover and Brittingham 1993). Therefore, my results are inconsistent with the 73 prediction that large forest tracts located in agricultural landscapes of the Midwest will have high rates of brood parasitism (Robinson et al. 1995). I had too few nests parasitized to examine patterns of parasitism in relation to distance from nests to the forest edge and edge type, but there was no evident pattern. Brittingham and Temple (1983) concluded that Brown-headed Cowbird parasitism rates increased with distance from deciduous forest edges, but other studies found no significant relationship (Gates and Gysel 1978; Chasko and Gates 1982). Parasitism pressures range widely throughout their distribution (Hoover and Brittingham 1993; Roth and Johnson 1993; Hahn and Hatfield 1995; Hoover et al. 1995; Friesen et al. 1999; F amsworth and Simons 1999; Ross 1999; Burke and No] 2000; and DeCecco et al. 2000) and especially within the Midwest region (Donovan et al. 1995; Trine 1998; Wilson et al. 1998; Fauth 2000; Trine 2000; Fauth 2001; Ford et al. 2001; and Robinson and Robinson 2001). Therefore, caution should be made when extrapolating parasitism data from Wood Thrush to aid in the management of mature interior forests for nesting birds. Limitations to this study. — The first limitation to my study is that I used data from adults nesting in the northeastern United States to determine my asymptotic values to calculate growth constants because I did not capture adults at my sites and local data were not available. Geographic variation of adult body size in Wood Thrush has not been reported (Clement 2000) so I expect that adult Wood Thrush data from the northeastern region are comparable to adults breeding in Michigan. Only mass data from adults captured during the breeding season were used for calculation since weights from postbreeding adults 74 may be reduced and weights from migrants may be elevated (Johnson et al. 1990; Conway et al. 1994; Leberg et al. 1996). However, if differences do exist between adults in these two regions, 1 used the same asymptotic value when fitting all curves so that the relative differences would remain informative even if the absolute growth rate constants were slightly biased. Another limitation to my study is that I did not measure food abundance directly in edge and interior forest. Wood Thrush forage for invertebrates, small vertebrates, and fruits on the forest floor and in multiple layers of the forest (Holmes and Robinson 1988). Conservation Implications. —The Midwest is characterized by fragmented forested habitat of variable quality for birds that nest in the interior of forests. While daily survival rate and nesting success provide information about the importance of nest predators and brood parasites in fragmented forests, growth rates provide information about the environmental degradation of the habitat that results from fragmentation and may influence the survivorship and reproductive success of individuals in the future. Growth rates reflect environmental conditions such as food availability (Barrett et a1. 1987; Bryant 1975; Blancher and Robertson 1987; Tinbergen and Boerlijst 1990; McCarty 2001) and weather (Bryant 1975; Konarzewski and Taylor 1989; Keller and van Noordwijk 1994; McCarty 2001), and parental quality (Riddington and Gosler 1995; Kunz and Ekman 2000), and are good indicators of the future survival of nestlings (Bolton 1991; Kersten and Brenninkmeijer 1995). While previous research has identified several factors that influence variation in nestling growth, including food supplies, weather, and parental effort, the complex interactions of these factors remain poorly 75 understood. My finding that distance to forest edge influenced nestling growth suggests that the quality of the habitat or individuals occupying those habitats may change with increasing distances from the forest edge. I also found that growth may be more rapid for body structures, like tarsi, that may play an important functional role early in the nestling period, as opposed to wing chords. The availability of food and parental provisioning rates may constrain the development of nestlings. To strengthen our understanding of the proximate causes of variability in growth, future work should quantify parental provisioning rates when it is not possible to directly measure food abundance while controlling food availability. In fragmented habitats, individuals of interior-adapted species that are lower in quality may be forced into edge habitat, so that their nestlings may grow more slowly because there is less food or food availability is unpredictable and/or their parents are less proficient at providing resources. I did not address the effects of parental age or experience on nestling growth. I treated nests as independent in this study but it is possible that I observed re-nesting attempts by parents, which would mean that the brood growth rates of a few nests were autocorrelated because they shared the same parents. Growth has a strong genetic component (O’Connor 1975; Galbraith 1988; Rhymer 1988), which is a source of variation that needs more investigating. Cross-fostering designs have the potential to determine to what extent variation in nestling growth is caused by genetic and environmental factors (van Noordwijk and Marks 1998; Kunz and Ekman 2000). 76 LITERATURE CITED Allison, P. D. 1997. Survival analysis using the SAS system: A practical guide. SAS Institute, Inc., Cary, North Carolina. Anders, A. D., J. Faaborg, and F. R. Thompson 111. 1998. Postfledging dispersal, habitat use, and home-range size of juvenile Wood Thrushes. Auk 115:349-358. Andersen, M. C., J. M. Watts, J. E. Freilich, S. R. Yool, G. I. Wakefield, J. F. McCauley, and P. B. Fahnestock. 2000. Regression-tree modeling of desert tortoise habitat in the central Mojave Desert. Ecological Applications 10:890-900. 1 Andre’n, H. 1992. Corvid density and nest predation in relation to forest fragmentation: A landscape perspective. Ecology 73:794-804. Artman, V. L. and J. F. Downhower. 2003. Wood Thrush (Hylocichla mustelina) nesting l, ecology in relation to prescribed burning of mixed-oak forest in Ohio. Auk 120:874—882. g Askins, R. A., J. F. Lynch, and R. Greenberg. 1990. Population declines in migratory birds in eastern North America. Current Ornithology 7: 1-57. Barrett, R. T, N. T. Anker-Nilssen, F. Rikardsen, K. Valde, N. Rev, and W. Vader. 1987. The food, growth and fledging success of Norwegian Puffin chicks F ratercula arctica in 1980-1983. Omis Scandinavica 18:73-83. Bertin, R. I. 1977. Breeding habitats of the Wood Thrush and Veery. Condor 79:303-311. Blancher, P. J. and R. J. Robertson. 1987. Effect of food supply on the breeding biology of Western Kingbirds. Ecology 682723-732. Bolton, M. 1991. Determinants of chick survival in the Lesser Black-backed Gull: Relative contributions of egg size and parental quality. Journal of Animal Ecology 60:949-960. Bradbury, R. B., J. D. Wilson, D. Moorcroft, A. J. Morris, and A. J. Perkins. 2003. Habitat and weather are weak correlates of nestling condition and growth rates of four UK farmland passerines. Ibis 145:295-306. Brawn, J . D. and S. K. Robinson. 1996. Source-sink population dynamics may complicate the interpretation of long-term census data. Ecology 7723-12. Breiman, L., J. H. Friedman, R. A. Olshen, and C. G. Stone. 1984. Classification and Regression Trees. Wadsworth International Group, Belmont, California. 77 Brittingham, M. C. and S. A. Temple. 1983. Have cowbirds caused forest songbirds to decline? Bioscience 33 :3 1-35. Bryant, D. M. 1975. Environmental influences on growth and survival of nestling House Martins Delichon urbica. Ibis 120:271-283. Burhans, D. E. and F. R. Thompson 111. 2000. Costs of parasitism incurred by two songbird species and their quality as cowbird hosts. Condor 102:364—3 73. Burke, D. M. and E. N01 1998. Influence of food abundance, nest-site habitat, and forest fragmentation on breeding Ovenbirds. Auk 115296-104. Burke, D. M. and E. No]. 2000. Landscape and fragment size effects on reproductive success of forest-breeding birds in Ontario. Ecological Applications 10:1749-1761. Carrie, N. R. 1999. Brown-headed Cowbird parasitism of Wood Thrush nests in eastern Texas. Journal of Field Ornithology 70:263-267. Chasko, G. G. and J. E. Gates. 1982. Avian habitat suitability along a transmission line corridor in an oak hickory forest region. Wildlife Monographs 8221-41. Chen, J., J. F. Franklin, and T. A. Spies. 1993. Contrasting microclimates among clearcut, edge, and interior old-growth Douglas—fir forest. Agricultural and Forest Meteorology 632219-23 7. Clement, P. 2000. Wood Thrush. Pages 319-322 in Thrushes. Princeton University Press, Princeton, New Jersey. Conway, C. J ., W. R. Eddleman, and K. L. Simpson. 1994. Seasonal changes in fatty acid composition of the Wood Thrush. Condor 96:791-794. De’ath, G. and K. E. Fabricius. 2000. Classification and regression trees: A powerful yet simple technique for ecological data analysis. Ecology 81 2317 8-3192. Dearbom, D. C., A. D. Anders, F. R. Thompson 111, and J. Faaborg. 1998. Effects of cowbird parasitism on parental provisioning and nestling food acquisition and growth. Condor 100:326-334. DeCecco, J. A., M. R. Marshall, A. B. Williams, G. A. Gale, R. J. Cooper. 2000. Comparative seasonal fecundity of four Neotropical migrants in middle Appalachia. Condor 102:653—663. Dettmers, R., D. A. Buehler, and K. E. Franzreb. 2002. Testing habitat-relationship models for forest birds of the southeastern United States. Journal of Wildlife Management 66:417-424. 78 Donovan, T. M., F. R. Thompson 111, J. Faaborg, and J. R. Probst. 1995. Reproductive success of migratory birds in habitat sources and sinks. Conservation Biology 921380- 1395. Donovan, T. M., C. J. Beardmore, D. N. Bonter, J. D. Brawn, R. J. Cooper, J. A. Fitzgerald, R. Ford, S. A. Gauthreaux, T. L. George, W. C. Hunter, T. E. Martin, J. Price, K. V. Rosenberg, P. D. Vickery, and T. B. Wigley. 2002. Priority research needs for the conservation of Neotropical migrant landbirds. Journal of Field Ornithology 73:329-339. 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. Dutta, C., S. Johnson, D. Larkin, and L. P. Mangurian. 1998. Skeletal development at the time of fledging in House Wrens. Condor 1002568-573. Fan, 2., S. R. Shifley, M. A. Spetich, F. R. Thompson 111, and D. R. Larsen. 2003. Distribution of cavity trees in midwestem old-growth and second-growth forests. Canadian Journal of Forest Research 33:1481-1494. Famsworth, G. L. and T. R. Simons. 1999. Factors affecting nesting success of Wood Thrushes in Great Smoky Mountains National Park. Auk 116: 1075-1082. F auth, P. T. 2000. Reproductive success of Wood Thrushes in forest fragments in northern Indiana. Auk 1 172194-204. Fauth, P. T. 2001. Wood Thrush populations are not all sinks in the agricultural Midwestern United States. Conservation Biology 15:523-527. Flaspohler, D. J., S. A. Temple, and R. N. Rosenfield. 2001. Species-specific edge effects on nest success and breeding bird density in a forested landscape. Ecological Applications 11:32-46. Ford, T. B., D. E. Winslow, D. R. Whitehead, and M. A. Koukol. 2001. Reproductive success of forest-dependent songbirds near an agricultural corridor in south-central Indiana. Auk 118:864-873. Friesen, L., M. D. Cadman, and R. J. MacKay. 1999. Nesting success of Neotropical migrant songbirds in a highly fragmented landscape. Conservation Biology 13:338-346. Galbraith, H. 1988. Effects of egg size and composition on the size, quality and survival of lapwing Vanellus vanellus chicks. Journal of Zoology (London) 214:383-398. Gates, J. E. and L. W. Gysel. 1978. Avian nest dispersion and fledging success in field- forest ecotones. Ecology 59:871-883. 79 Gebhardt-Henrich, S. and H. Richner. 1998. Causes of Growth Variation and Its Consequences for Fitness. Pages 324-339 in Avian Growth and Development (J. M. Starck and R. E. Ricklefs, Eds.). Oxford University Press, New York, New York. 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. Grubb, T. G., W. L. Robinson, and W. W. Bowerrnan. 2002. Effects of watercraft on Bald Eagles nesting in Voyageurs National Park, Minnesota. Wildlife Society Bulletin 30:156-161. Hahn, D. C. and J. S. Hatfield. 1995. Parasitism at the landscape scale: Cowbirds prefer forests. Conservation Biology 921415-1424. Hines, J. E. and J. R. Sauer. 1989. Program CONTRAST-A general program for the analysis of several survival or recovery rate estimates. US. Fish & Wildlife Service, Fish & Wildlife Technical Report 24, Washington, DC. Holmes, R. T. and S. K. Robinson. 1988. Spatial patterns, foraging tactics, and diets of ground-foraging birds in a northern hardwoods forest. Wilson Bulletin 100:377-394. Holmes, R. T. and T. W. Sherry. 2001. Thirty-year bird population trends in an unfragrnented temperate deciduous forest: Importance of habitat change. Auk 1182589- 609. Hoover, J. P. and M. C. Brittingham. 1993. Regional variation in cowbird parasitism of Wood Thrushes. Wilson Bulletin 105:228-238. Hoover, J. P. and M. C. Brittingham. 1998. Nest-site selection and nesting success of Wood Thrushes. Wilson Bulletin 1102375-383. Hoover, J. P., M. C. Brittingham, and L. J. Goodrich. 1995. Effects of forest patch size on nesting success of Wood Thrushes. Auk 1122146-155. Howe, H. F. 1976. Egg size, hatching asynchrony, sex, and brood reduction in the Common Grackle. Ecology 5721 195-1207. Hurlbert, S. H. 1984. Pseudoreplication and the design of ecological field experiments. Ecological Monographs 54:187-21 1. James, F. C. and H. H. Shugart, Jr. 1970. A quantitative method of habitat description. Audubon Field Notes 24:727-736. Johnson, D. H. 1979. Estimating nest success: The Mayfield method and an altemative. Auk 96:651-661. 80 Johnson, R. K., R. R. Roth, and J. T. Paul, Jr. 1990. Mass variation in breeding Wood Thrushes. Condor 92:89-96. Johnson, R. G., and S. A. Temple. 1990. Nest predation and brood parasitism of tallgrass prairie birds. Journal of Wildlife Management 54:106-111. Jolivet, P. 1986. Insects and plants: parallel evolution and adaptations in Flora and Fauna (E. R. Brill, Ed.). Flora & Fauna Publications, New York, New York. Keller, L. F. and A. J. van Noordwijk. 1994. Effects of local environmental conditions of nestling growth in the Great Tit Parus major L. Ardea 82:349-362. Kersten, M. and A. Brenninkmeijer. 1995. Growth, fledging success and post-fledging survival of juvenile Oystercatchers Haematopus ostralegus. Ibis 137:396-404. Konarzewski, M. and J. R. E. Taylor. 1989. The influence of weather conditions on growth of Little Auk Alle alle chicks. Omis Scandinavica 20:112-116. Kunz, C., and J. Ekman. 2000. Genetic and environmental components of growth in nestling Blue Tits (Parus caeruleus). Journal of Evolutionary Biology 132199-212. Leberg, P. L., T. J. Spengler, and W. C. Barrow, Jr. 1996. Lipid and water depletion in migrating passerines following passage over the Gulf of Mexico. Oecologia 10621-7. Manolis, J. C., D. E. Andersen, and F. J. Cuthbert. 2000. Uncertain nest fates in songbird studies and variation in Mayfield estimation. Auk 1172615-626. Marshall, M. R., J. A. DeCecco, A. B. Williams, G. A. Gale, and R. J. Cooper. 2003. Use of regenerating clearcuts by late-successional bird species and their young during the post-fledging period. Forest Ecology and Management 183:127-135. Martin, T. E. 1987. Food as a limit on breeding birds: A life-history perspective. Annual Review of Ecology and Systematics 18:453-487. Marvil, R. E., and A. Cruz. 1989. Impact of Brown-headed Cowbird parasitism on the reproductive success of the Solitary Vireo. Auk 106:476-480. Mayfield, H. F. 1975. Suggestions for calculating nest success. Wilson Bulletin 87:456- 466. McCarty, J. P. 2001. Variation in growth of nestling tree swallows across multiple temporal and spatial scales. Auk 118: 1 76—190. McCarty, J. P. and D. W. Winkler. 1999. Relative importance of environmental variables in determining the growth of nestling Tree Swallows T achycineta bicolor. Ibis 141 :286— 296. 81 Murcia, C. 1995. Edge effects in fragmented forests: Implications for conservation. TREE 10:58-62. Niemeléi, J ., D. Langor, and J. R. Spence. 1993. Effects of clear-cut harvesting on boreal ground-beetle assemblages (ColeopterazCarabidae) in western Canada. Conservation Biology 7:551-561. Nilsson, J. and M. Svensson. 1996. Sibling competition affects nestling growth strategies in Marsh Tits. Journal of Animal Ecology 65:825-836. Noordwijk, A. J. van and H. L. Marks. 1998. Genetic Aspects of Growth. Pages 305-323 in Avian Growth and Development (J. M. Starck and R. E. Ricklefs, Eds.). Oxford University Press, New York, New York. Nott, P. 1999. Monitoring avian productivity and survivorship (MAPS) habitat structure assessment (HAS) protocol: Describing vertical and horizontal spatial habitat patterns at MAPS stations. The Institute for Bird Populations. O’Connor, R. J. 1975. Initial size and subsequent growth in passerine nestlings. Bird- Banding 46:329-340. O’Connor, R. J. 1978. Brood reduction in birds: Selection for fratricide, infanticide, and suicide? Animal Behavior 26:79-96. O’Connor, R. J. 1977. Differential growth and body composition in altricial passerines. Ibis 119:147-166. Paton, P. W. C. 1994. The effect of edge on avian nest success: How strong is the evidence? Conservation Biology 8: 17-26. Peterjohn, B. G., J. R.’ Sauer, and C. S. Robbins. 1995. Population trends from the North American Breeding Bird Survey and population trends of Neotropical migrant birds. Pages 3-39 in Ecology and Management of Neotropical Migratory Birds (D. M. Finch and T. E. Martin, Eds.). Oxford University Press, New York, New York. Peters, R. H. 1983. Chapter 8: Production: Growth and reproduction. Pages 122-124 in The Ecological Implications of Body Size (E. Beck, H. J. B. Birks, E. F. Connor, Eds.). Cambridge University Press, New York, New York. Petit, L. J. and D. R. Petit. 2000. Brown-headed Cowbird parasitism of migratory birds: Effects of forest area and surrounding landscape (Smith, J. N. M., T. L. Cook, S. 1. Rothstein, S. K. Robinson, S. G. Sealy, Eds.). University of Texas Press, Austin, Texas. Podlesak, D. W. and C. R. Blem. 2001. Factors associated with growth of nestling Prothonotary Warblers. Wilson Bulletin 113:263-272. 82 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 64:1 1-23. Qi, J., R. C. Marsett, M. S. Moran, D. C. Goodrich, P. Heilman, Y. H. Kerr, G. Dedieu, A. Chehbouni, and X. X. Zhang. 2000. Spatial and temporal dynamics of vegetation in the San Pedro River basin area. Agricultural and Forest Meteorology 105:55-68. Ralph, C. J., G. R. Geupel, P. Pyle, T. E. Martin, and D. F. DeSante. 1993. Handbook of Field Methods for Monitoring Landbirds. Rhymer, J. M. 1988. The effect of egg size variability on thermoregulation of Mallard (A nas platyrhynchos) offspring and its implications for survival. Oecologia 75:20-24. Rich, A. C., D. S. Dobkin and L. J. Niles. 1994. Defining forest fragmentation by corridor width: The influence of narrow forest-dividing corridors on forest-nesting birds in southern New Jersey. Conservation Biology 821 109-1 121. Ricklefs, R. E. 1967. A graphical method of fitting equations to growth curves. Ecology 48:978-983. Ricklefs, R. E. 1968. Patterns of growth in birds. Ibis 110:419-451. Riddington, R. and A. G. Gosler. 1995. Differences in reproductive success and parental qualities between habitats in the Great Tit Parus major. Ibis 137:371-3 78. Robbins, C. S., D. K. Dawson, and A. Dowell. 1989a. Habitat area requirements of breeding forest birds of the Middle Atlantic States. Wildlife Monographs No. 103:1-34. Robbins, C. S., J. R. Sauer, R. S. Greenberg, and S. Droege. 1989b. Population declines in North American birds that migrate to the Neotropics. Proceedings of the National Academy of Science 86:7658-7662. Robinson, S. K. 1992. Population dynamics of breeding Neotropical migrants in a fragmented Illinois landscape. Pages 408-418 in Ecology and Conservation of Neotropical migrant landbirds (J. M. Hagan and D. W. Johnston, Eds). Smithsonian Institution Press, Washington, DC. Robinson, S. K. and W. D. Robinson. 2001. Avian nesting success in a selectively harvested north temperate deciduous forest. Conservation Biology 15:1763-1771. Robinson, S. K., F. R. Thompson 111, T. M. Donovan, D. R. Whitehead, and J. Faaborg. 1995. Regional forest fragmentation and the nesting success of migratory birds. Science 267: 1987-1990. 83 Roff, D. A. and R. J. Roff. 2003. Of rats and Maoris: a novel method for the analysis of patterns of extinction in the New Zealand avifauna before European contact. Evolutionary Ecology Research 5:759—779. Roth, R. R. and R. K. Johnson. 1993. Long-term dynamics of a Wood Thrush population breeding in a forest fragment. Auk 110:37-48. Roth, R. R. M. S. Johnson, and T. J. Underwood. 1996. Wood Thrush (Hylocichla mustelina). The Birds of North America, No. 246 (A. Poole and F. Gill, Eds.). The Academy of National Science, Philadelphia, and The AOU, Washington, DC. SAS Institute. 2002. SAS, version 8. SAS Institute, Inc., Cary, North Carolina. Sauer, J. R., G. W. Pendleton, B. G. Peterjohn. 1996. Evaluating causes of population change in North American insectivorous songbirds. Conservation Biology 10:465-478. Schew, W. A. and R. E. Ricklefs. 1998. Developmental plasticity. Pages 381-382 in Avian Growth and Development (J. M. Starck and R. E. Ricklefs, Eds.). Oxford University Press, New York, New York. Schifferli, L. 1978. Experimental modification of brood size among House Sparrows Passer domesticus. Ibis 120:365-369. Schowalter, T. D., J. W. Webb, and D. A. Crossley, Jr. 1981. Community structure and nutrient content of canopy arthropods in clearcut and uncut forest ecosystems. Ecology 62:1010—1019. Simons, T. R., G. L. Famsworth, and S. A. Shriner. 2000. Evaluating Great Smoky Mountains National Park as a population source for the Wood Thrush. Conservation Biology 14:1133-1144. Sisk, T. D. and J. Battin. 2002. Habitat edges and avian ecology: Geographic patterns and insights for western landscapes. Studies in Avian Biology 25:30-48. Skole, D. L. and J. Qi. 2001. Optical remote sensing for monitoring forest and biomass change in the context of the Kyoto Protocol. CGCEO/RAO] -01/w. Michigan State University, East Lansing, Michigan. Soler, M. and J. J. Soler. 1991. Growth and development of Great Spotted Cuckoos and their Magpie host. Condor 93:49-54. Starck, J. M. and R. E. Ricklefs. 1998. Avian Growth Rate Data Set. Pages 381-382 in Avian Growth and Development (J. M. Starck and R. E. Ricklefs, Eds.). Oxford University Press, New York, New York. 84 The Nature Conservancy and Michigan Department of Natural Resources. 2000. Allegan State Game Area Biodiversity Conservation Plan. Unpublished report by The Nature Conservancy, Allegan, MI. Thompson, F. R. 111, T. M. Donovan, R. M. DeGraaf, J. Faaborg, and S. K. Robinson. 2002. A multi-scale perspective of the effects of forest fragmentation on birds in eastern forests. Studies in Avian Biology 25:8-19. Tinbergen, J. M. and M. C. Boerlijst. 1990. Nestling weight and survival in individual Great Tits (Parus major). Journal of Animal Ecology 59:1113-1 127. Trine, C. L. 1998. Wood Thrush population sinks and implications for the scale of regional conservation strategies. Conservation Biology 12:576-585. Trine, C. L. 2000. Effects of multiple parasitism on cowbird and Wood Thrush nesting success in Ecology and Management of Cowbirds and Their Hosts (Smith, J. N. M., T. L. Cook, S. I. Rothstein, S. K. Robinson, S. G. Sealy, Eds.). University of Texas Press, Austin, Texas. 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 1 16:458- 466. 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. Weinberg, H. J. and R. R. Roth. 1998. Forest area and habitat quality for nesting Wood Thrushes. Auk 115:879-889. Whitcomb, R. F., J. F. Lynch, M. L. Klimkiewicz, C. S. Robbins, B. L. Whitcomb, and D. Bystrak. 1981. Effects of forest fragmentation on avifauna of the eastern deciduous forest. Pages 125-205 in Forest Island Dynamics in Man-dominated Landscapes (R. L. Burgess and D. M. Sharpe, Eds.). Springer-Verlag, New York, New York. Wilson, G. R., M. C. Brittingham, and L. J. Goodrich. 1998. How well do artificial nests estimate success of real nests? Condor 100:357-3 64. 85