DIFFERENTIAL STOPOVER HABITAT USE AND ITS HEALTH CONSEQUENCES IN FALL MIGRATING LANDBIRDS By Yushi Oguchi A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of Fisheries and Wildl ife Master of Science 2015 A BSTRACT DIFFERENTIAL STOPOVER HABITAT USE AND ITS HEALTH CONSEQUENCES IN FALL MIGRATING LANDBIRDS By Yushi Oguchi Migration is a physiologically - demanding activity, and the ability to locate high - quality stopover hab itat f or refueling has major fitness implications for migrating landbirds. However, stopover habitat is being altered by human land - use changes such as introductions of exotic shrubs , and knowledge is lacking on whether or how differences in stopover habit a t could impact migrating landbirds health . We first tested whether indices of fall migrant use differed between exotic - dominated and native - dominated shrublands at a stopover site in mid - Michigan through banding and fecal analyses in 2012 and 2013, focusi ng on gray catbirds ( Dumetella carolinensis ) and Catharus ustulatus ) (Ch. 1). C apture rate was lower in exotic shrubland than in native shrubland, and habitat switchover rate from exotic to native shrubland was greater than vice versa. Fecal analyses revealed that catbirds and thrushes selected certain native fruits. We then tested whether health status of catbirds and thrushes differed by habitat using blood collected from a subset of captured birds (Ch. 2). Plasma i ndices of r efueling performance (fat metabolites) did not differ by habitat in either species or year, but 3 catbirds using exotic shrub had lower measures of constitutive innate immunity (only in 2013) and antioxidant capacity than conspecifics in native shrubland . Fruit nu trient and fecal data suggested that these health differences may have resulted from differential fruit consumption. This study thus provides the first evidence that differential use of stopover habitat could impact health status of migrating birds and tha t evaluation of stopover habitat quality using an ecophysiology approach should include immune and antioxidan t measures in addition to indices of refueling performance. iii A CKNOWLEDGEMENTS I would first like to thank my Major Advisor Dr. Jen C. Owen and Co - PI Dr. Robert J. Smith for c onceiving and securing funds for this fantastic research project. Their novel idea tested here, that stopover habitat quality may be reflected upon individual health of migrating landbirds , has certainly sparked intense inter est in me when I applied for the M.S. position. Background knowledge on the links between migration and health has o nly recently be en produced, and researchers have just begun to acknowledge the importance of multifaceted physiological approaches in evalua ting stopover habitat quality. T he awareness that we are testing a novel idea has allowed me to renew my motivation s and overcome challenges . I am very honored and privileged to be able to contribute to the body of knowledge in conservation physiology of migrating landbirds as a Master s student . My committee members, Dr. Owen , Dr. Smith, and Dr. David Williams, have provided generous advice , encouragement , and support. T heir diverse areas of expertise ha ve granted me a unique learning experience of i ntegrating ecophysiology , ecoimmunology , bird - habitat relationships , and s patial ecology . In particular, Dr. Owen has taught me invaluable lessons that develop ed my thinking as a scientist , including (1) the hypothesis - prediction paradigm , (2) choosing met hods that directly test the prediction s, and (3 ) appreciating the biological significance/value of negative results. Life and/or philosophical lessons, which I am still in the process of learning, include (1) backward planning with detailed scheduling, (2) moving on to different aspects of work when encountering roadblocks, (3) delegating work/asking for help to qualified individuals, (4) knowing my limits of endurance, and (5) finding/establishing my niche in the realm of science. At the same time, Dr. Owe n and my committee have provided ample iv room t o expand the project , attempt ne w analyses, stumble during the process , and learn from my mistakes ; at the same time, they monitored my progress and were always available to help and guide me when needed. The success of this research project owes tremendously to the assistants and volunteers who committed a significant amount of their time and effort in the field and laboratory. In particular, I thank Zak Pohlen, Carrie Gesmundo, Carrie Gawne , Liz Theile, Kaitl in Clark, Kirsten Johnson, Danielle Boston, Kaitlyn Wilson, Hitomi Matsumura, M egan Porter, Jake Hitchcock, Nathan Martineau , Chris May, Richard , and Ed for their long hours of hard work and enthusiasm. I am thrilled to share with them the fruits of our wo rk. A number of resea rchers from across the United S t ates and world - wide have also graciously shared their knowledge and assay protocols for the benefit of this work and for my learning experience : Dr. K ristina Paxton (University of Hawaii) , Dr. Christoph er Guglielmo (University of Western Ontario, Canada), Dr. Kevin McGraw (Arizona State University) , Dr. David Costantini (University of Glasgow, U.K.) , Dr. Kevin Matson (University of Groningen, Netherlands) , Dr. Peeter Hõrak (University of Tartu, Estonia) , Dr. Tess Killpack (Tufts Medical Center) , Dr. Emily Cohen ( Smithsonian Conservation Biology Institute ), Dr. Scott McWilliams (University of Rhode Island), Dr. Andrew Flies (University of South Australia) , Dr. Lynn Martin (University of South Florida) , Dr. Cynthia Downs (University of Nevada, Reno), and Ms. Julie Craves (University of Michigan - Dearborn). Ms. Rebecca Ives , Ms. Michelle Quigley , Dr. Gary Roloff, and Dr. Robert Montgomery have provided technical support in laboratory or statistic al analyses . D r. Frank Moore (University of Southern Mississippi) and Dr. Rique Campa have also allowed us to use their valuable field equipment. I would like to express my thanks to all of their intellectual and technical contributions. v T he staff with t he Department o f Fisheries and Wildlife has assisted me greatly . In particular, Ms. Marcia Baar and Ms. Jill Cruth have often helped me keep track of the non - requirements and monitored my health and wel l - being. I also thank the custodial service staff, in particular Mr. Michael Munger and Ms. Lucy Quintero, for keeping the lab clean while sharing my enthusiasm for research during their evening work. My academic success s education also owes greatly to my father (Chikara Oguchi), mother (Keiko Oguchi) and grandmother (Yuko Oguchi) in Japan , who have provided many years of love and encouragement . From an early age, they have always encouraged me to pursue my interests and dreams . T hey also endorsed my decision to pursue post - secondary education in the United States . In particular, m y f ather has continue d working after retirement as a self - employed machine - design er to financially assist my education . This work could not have been completed without his generous support. Paris M. Lewis , my significant other, has provided much mental support throughout my M.S. career. As a talented communicator and public speaker, she has also provided me insight into the value of our work for n on - scientists and helped me prepar e for conference presentations. It has been a great pleasure to be able to share my enthusiasm with a bright mind outside of my field , and I appreciate her understanding that my quest for knowledge would often result in le ss time spent together. I would also like extend my thanks to my mentors in physiology research ( Dr. Bill Karasov and Dr. Tess Killpack ) and in banding (Dr. Mara McDonald and Jerry Simons), who taught me critical thinking and key skills during my undergrad uate years at the Unive rsity of vi Wisconsin . They have allowed me to take the first steps toward a career in avian ecophysiology Finally, I thank the following organizati ons for providing f inancial support for this project : USFWS Avian Health and Disease Program (USFWS Migratory Bird Conservation in the Upper Midwest), Hal and Jean Glassen Memorial Foundation, and Michigan State University Undergraduate Research Fund. Than ks to the Michigan Department of Natural Resources for allow ing us to use the Rose Lake State Wildlife area for fieldwork . I greatly appreciate these agencies for recognizing the applied value of ecophysiology and ecoimmunology in wild bird conservation . I t is my sincere hope that ecophysiology and ecoimmunology will con tinue to gain momentum in the realm of conservation biology, and that I will be able to further contribut e to the development of these fields through my future work . vii TABLE OF CONTENTS LI ST OF TABLES ...................................................................................................................... . i x LIST OF FIGURES .......................................................................................................... ........ . ....x CHAPTER 1 : Exotic - and native - dominated shrubland habitat use and fruit consumption in fall migrating landbirds at a stopover site in mid - Michigan . ....... ................................... . ................ 1 ABSTRACT ....................... ................................................................................................2 Introduction ........................................................................................................................3 Methods ............... ................................................................................. ...............................5 Study Site .................................................................................. ............. . ..................5 Study Specie s ..................................................................................... . ......................6 Characterizing Habitats and Resource Attributes ............ . .......... . ...... .. ............ . ......6 Shrub S pecies C omposition and H abitat S tructu re ..... .. ............. ............. .....6 Fruit A bundance .................. ..................................................... . ....... ............ 7 Arthropod A bundance ..................................................................................7 Indexing Bird Use ................................................................ . .... ............. . ................. 8 Indexing Fruit Consumption ....................................................................... . .... ........ 8 Statistical Analy ses ................................................................... ............................... 9 Characteristics of H abitats and R esource A ttributes ........... . .......................9 Indices of B ird U se .............................................. ......................................10 Patterns of F ruit C onsumption ...................................................................12 Results ............................................................................................................ ...................1 2 Characteristics of Habitats and Resource Attributes ............... ................ . ............1 2 Indices of Bird Use ..................................................................................... . ...........13 Patterns of Fruit Consumption ................................................ ..............................1 5 Discussion ........................................................................................... ..............................1 6 Characteristics of H abitat s and Resource Attributes ................. ...........................1 6 Indices of Bird Use ............................................ .. ..................... ..............................1 7 Patterns of Fruit Consumption .......................... ...................... ..............................1 8 Conclusions and Management Implications ........ . .................. ............................... 20 Acknowledgements ............ . ..................................................................... ........................2 2 APPENDIX ..... .................................................................................... ..............................2 3 LITERATURE CITED .................................................................................. . . ...............3 2 CHAPTER 2 : H ealth consequences of differential stopover habitat use and fruit consumption in fall migrating landbirds ................................ ................................ ... .................................................. 3 8 A BSTRACT ......................................................................................................................3 9 Introduction ............................................................................................................. . ........ 40 Methods ............................................................................................................................. 4 3 Study Site and Habitats ................................................................................... . ......4 3 viii Study Species, Capture, and Blood Collection ........................ ..............................4 4 Indices of Refueling Performance .........................................................................4 6 Size - corrected F at M ass ........ .....................................................................46 Plasma Metabolites ....................................................................................46 Indices of Constitutive Innate Immune Function ..................... ....... .......................4 8 Total Leukocyte Count and Granulocyte: l ymphocyte Ratio .....................48 H emolysis and H emagglutination Activities .............................................49 Haptoglobin ........................................... ..................................................... 50 Plasma Total Carotenoids .....................................................................................50 Plasma Total Antioxidant Capacity . ........................................ .......... .................... 51 Dietary Carotenoids and Antioxidant Capacity ....................................................52 Measuring Carotenoids and Antioxidant Capacity of Fruits .....................52 Indices of Fruit Consumption ................. ...................................................53 Statistical Analyses .................................................................. ..............................5 4 Health Parameters ............................................................. .........................54 Dietary Carotenoids and Antioxidant Capacity .........................................56 Results .............................................................................................................. .................5 7 Age and Size by Habitat Type ................................................................................57 Indices of Refueling Performance ........................................... .................. . ............5 7 Indices of Constitutive Innat e Immune Function ........... . ......... ..............................5 9 Plasma Total Carotenoids .....................................................................................61 Plasma Total Antioxidant Capacity .................................... ..... .............................. 61 Dietary Carotenoids and Antioxidant Capacity .............. ........................ . . ............ 6 2 Discussion ........................................................................................... ............... ............... 6 3 Effect of H abitat U se on Indices of Refueling Performance ......... ................ .. .......6 3 Effect of H abitat U se on Indices of Constitutive Innate Immune Function ..... .. ....65 Effect of Habitat Use on Plasma Total Carotenoids .... .........................................67 Effect of Habitat Use on Plasma Total Antioxidant Capacity ......... ..................... . 6 8 Dietary Carotenoids and Antioxidant Capacity ........................ ......................... ...6 9 Immune and Anti oxidant Measures as Indicators of Habitat Quality ................... 71 Acknowledgements ........................................................................... ............................... 7 3 APPENDIX .... ............................................... ....................................................................7 4 LITERATURE CITED ...................................................................................................8 5 ix LIST OF TABLES Table 1.1. Estimated shrub stem densities (ste ms/ha) in exotic - and native - dominated shrubland s near East Lansing, MI, USA. Data ( mean ± SE ) are presented in decreasing order in exotic shrubland and in increasing order in native shrubland, showing profound difference in composition between the two shr ublands. ..... . ........24 Table 2.1. thrushes captured in exotic - or native - dominated shrubland near East Lansing, MI, USA in 2012 and 2013. Sample sizes are listed in parentheses. TRIG = - hydroxybutyrate (mmol/l), WBC = total leukocytes/10,000 erythrocytes, G:L = granulocyte:lymphocyte ratio, HL = hemolysis score, HA = hemagglutination score, Hp = haptoglobin (mg/ml), Car = total carotenoids (µg/ml), OXY = antioxidant capacity (mmol/l of HOCl neutralized). ......................... ............................................................................ .... . ......75 Table 2.2. Summary of variables retained in multiple linear regressions with back ward selection ( P < 0.1). These variables were then included as covariates in ANCOVA - hydroxybutyrate , WBC = total leukocytes, G:L = granulocyte:lymphocyte ratio, HL = hemolysis score, HA = hemagglutination score, Hp = haptoglobin, Car = total carotenoi ds, OXY = antioxidant capacity, D = ordinal date, T = time after sunrise, H = handling time, F = fat score, M = muscle score, W = wing chord, A = age, and N = no covariates. Plus or minus sign denote the direction of effects. Variables with parentheses ind icate those dropped ( P > 0.05) in the final, minimal ANCOVA model that included habitat. ............................................................................................ . .....76 Table 2.3. Number of fecal samples from gray catbirds and Swains in exotic - and native - dominated shrublands with fruit species identified. U.S. Status indicates the origin of fruit species as either exotic or native, while Exotic or Native under bird species indicates shrubland habitat the samples were collected from. Counts of fecal samples were pooled from 2012 and 2013 captures and were interpreted to represent fruit consumption frequencies of individuals in their respective habitat. ............................................................... ................................... . ....77 x LIST OF FIGURES Figure 1.1. Estimated fruit abundance in exotic - and native - dominated shrublands near East Lansing, MI, USA during fall migration of 2012 . Data (mean ± SE) are presented in decreasing orde r in exotic shrubland and in increasing order in native shrubland . Asterisk indicates exotic species. Permutational MANOVA revealed significant difference in fruit composition, as well as composition of total exotic vs. total native fruits between shrublan ds. ..................................... ........................... ... .. ..26 Figure 1.2. Flying arthropod abundance in exotic - and native - dominated shrublands near East Lansing, MI, USA during fall migration of 2012 and 2013 . Arthropods were collected u sing two Malaise traps in each habitat and data are presented as dry biomass ( ± SE). Linear mixed models on ln - transformed data showed no significant habitat difference in arthropod abundance in either year. ........................................ . .27 Figure 1.3. Capture rates of newly - banded landbirds ( ± SE) in exotic - and native - dominated shrublands near East Lansing, MI, USA during fall migration of 2012 and 2013. Birds were passively captured in 15 mist nets in each habitat. Compound Poisson lin ear mixed models revealed significantly greater capture rate in native shrubs as compared to exotic counterparts in both years. ............ ..... ........ . .....28 Figure 1.4. Movement of banded landbirds between exotic - and native - dominated shrublands ne ar East Lasing, MI, USA during fall migratory stopover in 2012 and 2013. S ignificantly greater switchover rate from exotic - to native - dominated shrubland compared to vice versa was found in both years. ............... ............... ... . ....29 Figure 1.5 . Movement of banded gray catbirds, Swainson s thrushes, and magnolia warblers between exotic - and native - dominated shrublands near East Lasing, MI, USA during fall migratory stopover in 2012 and 2013. S ignificantly greater switchover rate from exotic - to native - dominated shrubland compared to vice versa was found in gray catbirds (B) and Swainson s thrushes (D) in 2013, and a similar non - significant trend in gray catbirds in 2012 (A). All other comparison groups (C, E, F) did not show difference in switch over rate. .................................. ...... .... . .......... . ...30 Figure 1.6. Proportions of fruits found in f ecal samples of gray catbirds and Swainson s thrushes captured in shrublands near East Lansing, MI, USA during fall migration. Asterisk denotes exotic species. Site - wide fruit availability data was derived from plot - based fruit counts performed in 2012 (see Figure 1. 1 ). Fecal data from 2012 and 2013 were pooled to increase sample size (numbers above bars). Compared to proportions of site - wide fruit abundance , consumption was non - random ( P < 0.001). Post hoc tests revealed disproportionately greater consumption of the native L. benzoin by both bird species, while the exotic E. umbellata was selected against. See text for further details. ... ........... ............. ............. ....... ........... . ...31 xi Figure 2.1. Size - corrected fat mass estimates of gray catbirds captured near East Lansing, MI, USA during fall migration of 2012 and 2013 in relation to time after sunrise and exotic - or nativ e - dominated shrubland habitat. Regression lines through raw data uncorrected for ordinal date are shown. Controlling for ordinal date, significant time after sunrise × habitat interaction was found ( P = 0.02), indicating no fat change over morning feeding hours in native shrubland ( P = 0.23) and fat loss of 0.60 g/hr in exotic shrubland ( P = 0.05). .......................... ... .............. . ..................................... . .78 Figure 2.2. Least square means (± SE) of plasma (A) triglyceride and (B) - refueling in exotic - or native - dominated shrubland near East Lansing, MI, USA during fall migration of 2012 and 2013. Least square means were obtained by controlling for ordinal date , time after sunrise, handling time, fat score, muscle score, wing chord, and age as necessary (see Table 2.2). Significant differences or trends are denoted with P - value while NS describes non - significance. ........ .................... ...79 Figure 2.3. Least square means (± SE) of (A) total leukocyte count per 10,000 erythrocytes refueling in exotic - or native - dominated shrubland near East Lansing, MI, USA during fall migration of 2012 and 2013. Least square means were obtained by controlling for ordinal date, time after sunrise, handling time, fat score, muscle score, wing chord, and age as necessary (see Table 2.2). Significant differences or trends are denoted with P - value wh il e NS describes non - significance ...... ................... . ......80 Figure 2.4. Least square means (± SE) of (A) hemolysis score, (B) hemagglutination score, refueling in exotic - or native - dominated shrubland near East Lansing, MI, USA during fall migration of 2012 and 2013. Least square means were obtained by controlling for ordinal date, time after sunrise, handling time, fat score, muscle score, wing chord, and age as necessary (see Table 2.2). Significant differences or trends are denoted with P - value while NS describes non - significance. ...... ................... . .....81 Figure 2.5. Least square means (± SE) of (A) plasma total carotenoids and (B) antioxidant capacity in gray c - or native - dominated shrubland near East Lansing, MI, USA during fall migration of 2012 and 2013. Least square means were obtained by controlling for ordinal date, time after sunrise, handling time, fat s core, muscle score, wing chord, and age as necessary (see Table 2.2). Significant differences or trends are denoted with P - value while NS describes non - significance. ........ ....... ..................................................... . ...82 Figure 2.6. To tal carotenoids (A) and antioxidant capacity (B) of fruits found in shrublands used by migrating landbirds near East Lansing, MI, USA. Fruit species are order ed from highest to lowest means with centerline of boxes showing median values. Exotic fruits are shown as white boxes and native fruits are barred. Differences in letters above boxes represent significant differences ( P < 0.05), and numbers below the letters denote sample size. ............................ . ........................... . ..83 xii Figure 2.7. Expected dietary means (± SE) of (A) total carotenoids and (B) antioxidant - or native - dominated shrubland in East Lansing, MI, USA. Means were calculated as values of assayed fruits (se e Figure 2.6) weighted by consumption proportion (see Table 2.3). Significant differences are denoted with P - value while NS describes non - significance. ......................................................................................................... ..... . .84 1 CHAPTER 1: Exotic - and native - dominated shrubland habitat use and fruit consumption in fall migrating landbirds at a stopover site in mid - Michigan 2 A BSTRACT The ability to locate high - quality stopover habitat has major implications for fit ness for migrating landbirds, but human land - use change has alter ed stopover habitats, including introductions of non - native shrubs. However, relatively little is known about the value of exotic fruits for fall migrants, and few studies have compared fall migrant use of exotic - dominated shrublands to native - dominated counterparts. We hypothesized that native - dominated shrubland represented superior - quality stopover habitat relative to exotic - dominated shrubland, and predicted that indices of bird use would be greater in the native shrubs and that certain native fruits would be selectively consumed by fall migrating landbirds. We indexed habitat structure, composition of shrubs and fruits, and abundance arthropods in exotic - and native - dominated shrublands an d compared capture rates and habitat switchover rates in g ray c atbirds ( Dumetella carolinensis t hrushes ( Catharus ustulatus ) between these habitats during fall migration of 2012 and 2013. Fecal samples collected from these birds were assaye d to determine patterns of fruit consumption. Habitat structure and total fruit and arthropod abundance were similar between shrublands , with major difference s occurring only in shrub and fruit compositions. Capture rate was significantly greater in the na tive shrubs in both years , and birds captured in exotic shrubs were more likely to switch to and stay in native shrubs than vice versa in 2013. The native c ommon s picebush ( Lindera benzoin ) was disproportionately consumed, whereas the exotic a utumn o live ( Euglenas umbellata ) and European b uckthorn ( Rhamnus cathartica ) showed evidence of being selected against. Our results suggest that exotic shrubland may be of limited importance for fall frugivorous migrants in the presence of adjacent native shrub land , an d that L. benzoin is the single most important shrub for habitat management aiming to improve habitat quality for fall frugivorous migrants in mid - Michigan. 3 I ntroduction Conservation of migratory landbirds requires habitat management throughout the e ntire annual cycle, but relatively little is known about how birds use habitat during migration and the fitness consequence of that use ( Moore et al. 2005 ; Smith et al. 2013 ) . Migration is a demanding activity accompanied by high mortality ( Sillett and Holmes 2002 ) , and because migrating landbirds must stop during migration to rest and refuel, their ability to locate high - quality stopover sites undoubtedly has fitness consequences ( Moore et al. 1995 ; Dierschke 2003 ; Smith and Moore 2003 ; Moore et al. 2005 ; Newton 2006 ) . During fall migration, fruit is a major food source for many landbird species ( Parrish 1997 ; Smith et al. 2007 ; Smith et al. 2013 ) , and the quality and quantity of the fruit offered are likely a key factor determining stopover habitat quality. Unfortunately, stopover habitat is being altered by human land - use changes ( Wilcove a nd Wikelski 2008 ; Klaassen et al. 2012 ) including the introduction of exotic shrub species ( Suthers et al. 2 000 ; Catling 2005 ; Smith et al. 2007 ; Craves 2009 ; Sm ith et al. 2013 ) . In the Midwestern United States, exotics such as a utumn o live ( Elaeagnus umbellata ), h oneysuckle ( Lonicera spp.), m ultiflora r ose ( Rosa multiflora ), and European b uckthorn ( Rhamnus cathartica ) have become common and widespread, possibly owing to their ability to thrive in disturbed landscapes and to inhibit native plant growth by changing the soil chemistry ( Woods 1993 ; Collier et al. 2002 ; Orr et al. 2005 ) . These shrub species bear fruits that are consumed by native landbird migrants ( Suthers et al. 2000 ; Smith et al. 2007 ; Gleditsch and Carlo 2011 ; Smith et al. 2013 ) , but more work is needed on the relative value of these exotic shrubs compared to the native counterparts ( Smith et al. 2013 ) . S tudies on exotic and native fruits ( White and Stiles 1992 ; Stiles 1993 ; Smith et al. 2007 ; Smith et al. 2013 ) indicate that native fruits tend to have greater energetic potential and fat 4 co ntent , whereas exotic fruits tend to contain more sugar and water. Based on energetic potentials of the fruits, it has been estimated that birds may need to consume around twice the amount of exotic fruits (e.g., E . umbellata ) in wet mass than the native c ounterparts (e.g., c ommon w interberry , Ilex verticillata ), to obtain the same amount of energy (Smith et al. 2007). In addition, certain fruits rich in fat may also contain greater antioxidant capacity ( Alan et al. 2013 ) , and dietary antioxidants are beneficial for migrants because they co ntrol tissue - damaging reactive oxygen species produced as a result of strenuous flight ( Costantini et al. 2007 ; Bolser et al. 2013 ; Jenni - Eiermann et al. 2014 ) . Therefore, it is conceivable that consuming native fruits high in energetic potential also have significant physiological benefits for landbird mi grants. However, the relative value of exotic fruits for birds during fall migration is likely context - specific and mixed results have been reported. In general, energetically - rich native fruits appear to be consumed first by migrants ( White and Stiles 1992 ; Stiles 1993 ; Smith et al. 2013 ) , but in one feeding experim ent ( Drummond 2005 ) , energetically poorer Lonicera fruits was preferentially consumed over c ranberrybush ( Viburnum opulus ). Furth ermore, in a matrix of forest, urban, and agricultural land in Pennsylvania, the abundance of birds was shown to be positively correlated with the abundance of Lonicera fruits ( Gleditsch and Carlo 2011 ) . At a mid - Michigan stopover site, long - term capture - recapture data ( Craves 2009 ) revealed that frugivorous migrants that arrived late in the fall were still able to gain fat and mass even after most native fruits had been already depleted. These previous st udies have evaluated use of exotic fruits within a habitat matrix composed of mixed exotic and native shrubs or compared exotic shrubland use to non - shrub habitats. Few studies, on the other hand, have compared migrant use between shrubland habitats that a re predominantly exotic and predominantly native. We compared indices of use in exotic - dominated shrubland (hereafter exotic shrubland) 5 and native - dominated shrubland (hereafter native shrubland) and the consumption patterns of these fruits as part of a l arger study to evaluate relative stopover habitat quality. Based on knowledge on fruit quality, patterns of fruit consumption and stopover biology of landbird migrants, we hypothesized that native shrubland would represent superior - quality habitat than exo tic shrubland. We predicted that: 1) overall bird capture rate would be greater in native shrubland during fall migration, and that primarily frugivorous fall migrants would exhibit 2) greater bias in capture rate toward native shrubs than less frugivorous species, 3 ) greater rates of habitat switching from exotic to native shrubland than vice versa, and 4 ) selection for certain native fruits. M ethods Study Site We conducted fieldwork on a State - managed wildlife area near East Lansing, MI, USA (Rose Lak (170 ha) was largely set within an agricultural and suburban landscape . Exotic and native shrublands existed in similar longitudes and in distinct patches adjacent to oak - maple woods and farml ands. Due to both the orientation and proximity of the two habitats we assumed that migrants using these shrublands wer e from the same population. The most common shrubs in the exotic shrubland were , as determined via vegetation sampling (see below and Tab le 1.1), E. umbellata , Lonicera spp. ( L. tartarica , L. morrowii , and L. x bella [ morrowii x tatarica ]), R. cathartica , and R. multiflora , while the native shrubland was predominantly g ray d ogwood ( Cornus racemosa ), c ommon s picebush ( Lindera benzoin ), I . ve rticillata , and p oison s umac ( Toxicodendron vernix ). 6 Study Species While all landbird species were used for analyses of capture rates, we focused o n the habitat use and fruit consumption of g ray c atbirds ( Dumetella carolinensis t hrushes ( Catharus ustulatus ) due to their high levels of frugivory during fall ( Mack and Yong 2000 ; Smith et al. 2011 ) and large sample size at our study site . In addition, we selected the primarily arthropodivorous m agnolia w arblers ( Dendroica magnolia ) ( Dunn and Hall 2010 ) to evaluate whether habitat - related differences (if any) in capture rate and movement (rate of habitat switch - over) may be more pronounced in fall frugivorous species . Characterizing Habitats and Resource Attribu tes We collected data on attributes of shrub composition to validate our habitat characterization as either native - dominated or exotic - dominated. To further assess habitat differences in cover and fruit resources we compared overall habitat structure, frui t abundance and composition, and flying arthropod abundance between exotic and native shrublands. Shrub S pecies C omposition and H abitat S tructure Plot - based vegetation sampling ( James and Shugart Jr 1970 ) was performed in the spring of 2013. S ample circles (0.04 ha, radi us 11.3 m) were established around each mist net (midpoint at the net center) in both native (n = 14) and exotic (n = 13) shrublands. When two nets were close enough that the circles overlapped, half - circles were sampled around the second net to avoid dupl ication. The half - circles were treated as one independent sampling unit in the analysis. E very assigned to a size class category (A: 8 - 15 cm, B: 15 - 23, C: 23 - 3 8, D: 38 - 53, E: 53 - 69, F: 69 - 84). Next, the number of woody stems with dbh < 8 cm intersecting two - arm - length transect across 7 the circle in each cardinal direction was recorded by species. S aplings of trees were included as shrubs for the purpose of struct ural analysis but were eliminated for shrub compositional analysis. On the same transect presence or absence of ground cover, shrub cover, and canopy cover were recorded by sightings through an ocular tube for 20 locations (taken every 0.56 m) . H eights of the three tallest trees within each plot and three tallest shrubs were estimated using an optical clinometer and triangulation a nd the means were recorded as the canopy and shrub height s (m) respectively . We then estimated tree density (stems/ha) and basal area (m 2 /ha) by size class, shrub stem density (stems/ha) , as well as percen t ground , shrub, and canopy cover. All estimates were calculated on a per - plot basis for statistical comparisons. Fruit A bundance We performed a net - based, direct fruit count ( Blake et al. 1990 ) to provide a baseline index of fruit abundance in mid - September, 2012. The number of all fruits for each shrub species was counted within the 0.04 h a circle vegetation sampling plots using hand - held counters. When direct counting was not possible due to large quantity of fruits (>1,000), we counted all f ruits on a single branch and multiplied the count by the number of similar - sized branches. Immature fruits were counted but removed from the dataset for subsequent analysis. Estimates of fruit abundance per plot w ere calculated as number of fruits per ha f or each species. Arthropod A bundance Flying arthropod biomass was indexed using four Malaise traps ( Smith and Hatch 2008 ) during the fall migration of 2012 and 2013. Two traps were placed at permanent locations in each habitat type. Arthropods were collected and traps reset every three to seven days. The arthropods were stored in 70% ethanol and later dried until a c onstant mass was achieved. The flying arthropod biomass is presented as dry mass collected per day. 8 Indexing Bird Use We passively captured birds in 15 mist nets in each habitat type ( 30 nets total; 30 mm - mesh, 12 m × 2.6 m) from August 13th and extending through October . The beginning of fall migration w as marked by influx of species that have no local breeding record (J. Owen, personal observation) , and field season ended with deer gun hunting season . Mist n ets were operated daily during fall for 4 - 5 hou rs beginning 30 min before sunrise and were checked at 45 minute intervals. Captured b irds were held in individual cloth bags or bird boxes until processing, and were banded with uniquely - numbered U.S. Geological Survey aluminum bands, aged as hatch - year ( HY) or after hatch - year (AHY) based on plumage characteristics or skull ossification ( P yle 1997 ) , and sexed where possible. We measured unflattened wing chord to the nearest 1 mm and mass to the nearest 0.1 g. Fat scores determined based on furcular and abdominal fat storage (0 - 6 scale, 0.5 increments) ( modified fr om Helms and Drury 1960 ) . Indexing Fruit Consumption We assessed diet of g ray c t hrushes using the fecal samples . Captured birds waiting to be processed were stored in boxes lined with white sheets of paper to opportunistically co llect fecal samples. Collected fecal samples were stored at - analysis. Fruits were identified by a combination of seed, color, and texture of the pulp and skin ( Parrish 1997 ) . W e smeared known fruits collected in the field on the same paper liner and stored under the same conditions to use as guide for identification. Arthropod consumption was identified by the presence of exoskeleton remains. Because we were interested in consum ption proportions of fruit and arthropods in our sampling population and volumetric estimations in fecal samples were difficult ( Parrish 1997 ) , we counted samples multiple times when more t han 9 one fruit species or when both fruit and arthropods co - occurred. All fruit were positively identif ied by a single observer blind to which habitat type the bird was captured in. Statistical Analyses We used Sigmaplot Version 11.0 ( Systat Software 2008 ) to analyze habitat switchover rates. All other statistical computations were performed using R Version 3.1.2 ( R Development Core Team 2014 ) . All tests were performed two - tailed and outputs with P < 0.05 deemed as significant and 0.05 < P < 0.1 as a trend unless specified otherwise. Values are presented as means ± SE throughout . Characteristics of H abitats and R e source A ttributes We tested the expectation that exotic and native shrublands differed in shrub species composition (native - dominant vs. exotic dominant) but not with respect to structural attributes. We performed permutational multivariate analysis of var iance (PerMANOVA) using function adonis in R package vegan ( Cohen et al. 2012 ; Oksanen et al. 2015 ) . PerMANOVA uses distance matrices and obviates the necess ity for distributional assumptions that are rarely met in ecological community datasets ( Anderson 2001 ) . We compared shrub composition between habitats for two multivariate responses: (1) stem density estimates of all shrub species (8 exotic and 26 native) to detect species - level difference and (2) summarized total exotic stem density and total native stem density to confirm our habitat categorization of structural metrics of s hrub density, shrub height, canopy height, percent ground cover, percent shrub cover, percent canopy cover, total tree density, and total tree basal area were entered as one multivariate resp onse to test for difference in overall habitat structure. To test for variation in tree size class (A - F) between habitat types (potential confounder of shrubland habitat 10 categorization), we further ran PerMANOVAs on tree density by size class and basal are a by size class. Total fruit abundance was compared with two - sample t - tests after square root transformation. We then used PerMANOVA to test whether fruit composition differed between habitats. To compare indices of arthropod abundance by habitat in each year, we used linear mixed models fit by REML (R package nlme, function lme) with habitat type as the fixed factor and Julian date (continuous random effect) representing pseudoreplication within each trap. Arthropod biomass per day was ln - transformed to meet the normality assumption. Indices of B ird U se We used capture rate as an index for density ( Smith and Hatch 2008 ; Akresh et al. 2009 ) , which is generally pos itively correlated with habitat quality ( reviewed in Johnson 2007 ) . We calculated capture rates as numbers of newly - banded birds per net - hour and compared between exotic and native shrubla nds. C apture rates of g ray c t hrushes, and magnolia warblers were similarly compared between habitat types . We used c ompound Poisson linear mixed model with log link function fit by Laplace approximation using function cpglmm in R packag e cplm ( Zhang 2013 ) with habitat type specified as fixed factor and Julian date (continuous random effect) representing pseudoreplication within each individual net. We then performed single - term 2 - based tests to determine the significance of the habitat effect. In addition, we tested whether age ratio, size, fat mass, and rate of fat gain d iffered by habitat type in g ray c - sample t - tests. Size - corrected fat mass for individuals were estimated u sing a series of regressions ( Owen and Moore 2006 ; 2008 ) . For each species, individuals of the same wing chord length class (1 mm 11 increments) were grouped (data compiled from records at the same site during fall migration in 2009 - 2014). Within each wing chord group, mass was regressed against f at score to obtain the intercept, which is the estimate for size - specific fat free mass (fat score = 0). In the second regression, the size - specific fat - free mass was then regressed against wing chord length to derive a linear model that predicts size - spec ific fat - free mass for a given wing chord length. Based on this model, fat - free mass of individuals were estimated and subsequently subtracted from actual mass to derive size - corrected fat mass (g) of individuals. Habitat differences in fat mass and fat ga in were detected by multiple linear regressions ( modified from Dunn 2000 ) . For each species and year, we fitted models explain ing fat mass by time after sunrise (proxy of foraging time) and habitat type as well as their interactions after controlling for the confounding effect of Julian date (positively related to fat mass in all comparison s , P < 0.05) . A significant interaction between habitat type and time after sunrise indicates different refueling performance between habitats. Non - significant terms ( P > 0.05) were sequentially dropped to test whether fat mass differed by habitat type ( test for condition - bas ed habitat partitioning; Tietz and Johnson 2007 ) . Fat mass of Swainson s thrushes were ln - plus - 5 transformed to attain normality. Using birds with a recapture record we compared the proportion of individual birds that switched from one habitat type to the other ( Akresh et al. 2009 ) . M igrants during stopover tend to move into habitats where food resource s are more abundant or contain more shelter ( Moore et al. 1995 ; Cohen et al. 2012 ) . T hus, a net emigration from poorer - quality habitat and a net immigration into better - quality habitat would be expected from capture - recapture data. We defined a bird as having switched habitat when the individual was recaptured in the other habitat relative to the one in which it was originally captured and was not recaptured again in the habitat where it was first captured. Because we detected few birds that transferred from one habitat to 12 another multiple times over the recapture history (2012: n = 2; 2013: n = 3), these individuals were excluded from further analyses. The proportion of birds that switch ed was calculated for all species combined and for g ray c t hrushes, and m agnolia w arblers. We used 2 contingency test s to compare the proportion of individuals that switched vs. stayed in the exotic Patterns of F ruit C onsumption Fecal samples with identified diet from 2012 (gray cat birds: n = 63, Swainson s thrush: n = 171) and 2013 (gray catbird: n = 42, Swainson s thrush: n = 39) were pooled to increase sample size after confirming that the consumption ranks of fruits (counts of identified fruits found in fecal samples) across habi proportions of fruits vs. arthropods and of fruit species between habitats. In order to assess whether birds consumed certain fruits disproportionate to their abundance within the entire study site, we pooled fruit abundance estimates and fecal samples from both habitats and compared ts showed significant difference, we performed post - hoc exact binomial tests for each of the catego ries against the sum of all other P < 0.0063) ( McDonald 2014 ) . R esults Characteristics of Habitats and Resource Attributes Estimated density of stems showed that the compositions of shrub species were sign ificantly different between exotic and native shrublands (pseudo - F 1, 25 = 20.14, P < 0.001; Table 1.1). 13 Summarized total exotic and native shrub density also differed significantly between the habitat types (pseudo - F 1, 25 = 53.97, P < 0.001). Exotic stems represented 70% of the shrub stems in exotic shrubland but comprised only 12% in the native shrubland. On the other hand, overall habitat structure was similar in these shrublands (pseudo - F 1, 25 = 0.38, P = 0.57), as well as tree density (pseudo - F 1, 25 = 0 .43, P = 0.84) and basal area (pseudo - F 1, 25 = 1.06, P = 0.36) by size class. Total fruit abundance were similar in both habitats ( t 15.94 = - 1.54, P = 0.14). However, fruit composition differed significantly in the two habitat types (pseudo - F 1, 23 = 8.7 3, P < 0.001; Figure 1.1). Summarized total exotic and native fruits also differed significantly (pseudo - F 1, 23 = 6.57, P < 0.001). Exotic fruit represented 50% and 95% of the available fruit in the native and exotic shrublands, respectively. However, R. c athartica , an exotic species, was the most abundant fruit in the native habitat (Figure 1.1). Flying arthropod abundance did not differ by habitat in either 2012 ( F 1, 2 = 4.89, P = 0.16) or 2013 ( F 1, 2 = 1.19, P = 0.39 ; Figure 1.2). Indices of Bird Use We banded a total of 2006 birds including 39 8 g ray c atbirds and 5 76 t hrushes on 45 mornings in 2012 and 21 82 birds including 31 4 g ray c atbirds and 3 60 th rushes on 40 mornings in 2013. Overall c apture rates (all species) were greate r in native shrubs as compared to exotic shrubs by 71% in 2012 ( 2 1 = 30.73, P < 0.001 ; Figure 1.3A) and 92% in 2013 ( 2 1 = 17.74 , P < 0.001; Figure 1.3 B ) . Fall frugivorous g ray c atbirds were 259% (2012 2 1 = 70.67 , P < 0.001 ) and 340% (2013: 2 1 = 118.42 , P < 0.001 ) more likely to be captured in native shrubland than in exotic shrubland. Capture rate of t hrush es differed by habitat in the same direction 14 by 21% ( 2 1 = 66.19, P < 0.001 ) and 29% 2 1 = 9.82, P = 0.002) . We found no difference in capture rate by habitat type in primarily arthropodiv orous m agnolia w arblers in 2 1 = 2.63, P = 0.11) and a non - significant trend toward greater capture rate in native shrubs (by 44%) 2 1 = 2.82, P = 0.09). In both years, over 90% of g ray c t hrushes captured were HY birds. G ray c atbirds did not show habitat differences in age ratio (2012: P = 0.19; 2013: P = 0.63) or wing chord length (2012: t 140.50 = 1.14, P = 0.26; 2013: t 86.52 = 0.55, P = 0.59). Similarly, we did not find age (2012: P = 0.61; 2013: P = 1.0 ) or wing chord differences (2012: t 513.28 = - 1.10, P = 0.27; 2013: t 341.60 = - 1.62, P t hrushes. Based on regression - estimated fat mass, g ray c atbirds in 2012 gained fat with similar rates (by 0.17 g/hr; F 1, 421 = 3.92, P = 0.05) bet ween habitat types (time after sunrise × habitat: F 1, 420 = 0.04, P = 0.83). In 2013 however, fat gain rates of g ray c atbirds differed significantly by habitat type ( F 1. 303 = 4.42, P = 0.04) in that individuals using native shrubs gained fat by 0.25 g/hr ( t 303 = 2.257, P = 0.02) while no fat was gained in exotic shrubland ( t 303 = - 1.33, P = 0.18). The main effect of habitat on fat mass was not found in either 2012 ( F 1, 421 = 1.87, P = 0.17) or 2013 ( F 1, 303 = 0.10, P = 0.75). No habitat - related difference t hrushes were found in either year (2012: F 1, 578 = 0.79, P = 0.37; 2013: F 1, 351 = 0.30, P = 0.58 ); there was no evidence of fat gain in these years (2012: F 1, 579 = 2.44, P = 0.12; 2013: F 1, 352 = 0.71, P = 0.40). Fat mas t hrushes did not differ significantly by habitat in 2012 ( F 1, 580 = 2.43, P = 0.1195) or 2013 ( F 1, 353 = 1.47, P = 0.23). We found that birds (data from all species) initially captured in the exotic shrubs were more likely to be recaptur ed in native shrubs (58% switched) than vice versa (32%) during migratory stopover in 2012 ( 2 1 = 11.79, P < 0.001 ; Figure 1.2A), and similar results were 15 obtained from 2013 (60% switch from exotic to native, 24% switch from native to exotic; 2 1 = 26.85, P < 0.001; Figure 1.4 B ). Examining species - specific movement in 2012, g ray c atbirds trended toward a greater switchover rate from exotic to native shrubs ( 57%, compared to native - to - exotic switch of 19%; P = 0.051 ; Figure1.5A t hrushes ( P = 0.27 ; Figure 1.5C ) did not show habitat - related differences. In 2013, habitat switchover from exotic to native was significantly greater than vice versa in both g ray c atbirds ( 71%, compared to 24% switch from native to exotic; P = 0.02 ; Figure 1.5B ) and S t hrushes ( 71%, compared to 30% switch from native to exotic; P = 0.02 ; Figure 1.5D ) . M agnolia w arblers did not show difference in switching by habitat in 2012 ( P = 1.0; Figure 1.5E) or 2013 ( P = 0.23 ; Figure 1.5 F ). Patterns of Fruit Consumption Gray c t hrush fecal samples mostly contained fruit; arthropod remains were only represented in 6% and 14% of the samples in these species respectively. These fruit - to - arthropod ratios did not differ by habitat in either species ( P = 1 .0). Comparing estimated fruit abundance and fecal samples within the entire study site, we found evidence of non - random selection of certain fruits in both g ray c t hrushes ( P < 0.001; Figures 1.2 and 1.6). Post - hoc tests (see Statist ical Analyses on Bonferroni - corrected P - values) revealed that the native L. benzoin was disproportionately consumed by gray catbirds (319% greater than expected; P < 0.0031 ) and Swainson s thrushes (243% greater than expected; P < 0.0031) , while the exotic E. umbellata was selected against (86% less than expected in gray catbirds and by 88% in Swainson s thrushes; both P t hrushes also showed evidence of selecting against the exotic R. cathartica ( consumed 31% less than expected; P < 0. 0031) and the native I. verticillata ( 54% less than expected; P < 0.0031) . S imilar trends were found in g ray 16 c atbirds ( 40% less than expected consumption of R. cathartica and 69% less of I. verticillata ; both P t hrushes d isproportionately consumed the native c ommon e lderberry ( Sambucus canadensis , by 431%) , and exotic b ittersweet n ightshade ( Solanum dulcamara , by 268%) , whereas R. multiflora was consumed significantly less (by 89%) than expected (all P < 0.0031). D iscu ssion We found that indices of bird use, particularly those of g ray c t hrushes, were significantly greater in native shrubland than in exotic shrubland. Based on the consumption patterns of fruit species, we believe these differences are the result of fall frugivorous birds selecting for certain native fruits, and that native shrubland represent higher - quality stopover habitat in comparison to exotic shrubland. Characteristics of Habitat s and Resource Attributes We found that veg etation structure and flying arthropod abundance were similar in exotic and that native shrublands. The importance of shelter from predators or weather ( Moore et al. 1995 ; Moore et al. 2005 ; Ktitorov et al. 2008 ) and arthropod diet ( Smith et al. 2007 ; Smith et al. 2013 ) during migratory stopover has been well established, and our results suggest that the quality of exotic shrublands as stopover habitat would be similar to native counterparts with respect to these attributes. Similarly, we did not find difference in the total abundance of fruits. On the other hand, these shrublands were significantly different in shrub and fruit species composition and we confirmed that exotic shrubland w as indeed exotic - dominated and native shrubland native - dominated. Therefore, we consider the main difference between exotic shrubland and 17 native shrubland to be in fruit species composition . Indices of Bird Use We predicted that bird use during fall migration would be greater in native shrubs than exotic shrubs, particularly in fall frugivorous species. Consistent to our prediction we found greater capture rate in native shrubland relative to exotic shrubland and greater overall switchover rate from exotic to native shrubland during fall migration of 2012 and 2013. Capture rates in spring and summer at the same study site did not differ between habitat types (Y. Oguchi and J. Owen personal o bservation) suggesting that the difference may be attributed to exotic vs. native fruits available. Indeed, fall frugivorous g ray c t hrushes showed greater capture rate s in native compared to exotic shrubland in both years and greater habitat switchover into native shrubland from exotic shrubland in 2013, whereas the primarily arthropodivorous m agnolia w arbler did not show differences in these measures. Because greater indices of density (e.g., capture rate) in a habitat ( Johnson 2007 ; Smith and Hatch 2008 ; Akresh et al. 2009 ) and net influx into a habitat from adjacent ones ( Johnson 2007 ; Akresh et al. 2009 ; Cohen et al. 2012 ) generally represent superior habitat quality, our results provide evidence for greater quality of native shrublands compared to exotic ones. While the majority of g ray c t hrushes captured were HY birds, we found no age - or size - related differences in habitat use. Thus, we do not consider age - or size - related dominan ce played a significant role in the distribution of these birds. W e also found no evidence for condition - based habitat partitioning ( sensu Tietz and Johnson 2007 ) based on our fat mass estimates. Additionally, w e did not find evidence of differential refueling success (i.e. rates of fat mass gain) during morning foraging in any comparison groups except gray catbirds in 18 2013, which showed fat gain only in native shrubland. Earlier work in Michigan ( Craves 2009 ) also suggested that Swainson s thrushes were able to gain mass at a stopover site composed of mixed exotic and native shrubs after depletion of high - energy native fruits. It is possible that birds could compensate for the lower energetic potential in exotic fruits by increased consumption , but more detailed work will be necessary to compare relative effort and energy gain of consuming exotic vs. native fruits . Patterns of Fruit Consumption Based on theories of optimal foraging ( Emlen 1966 ; MacArthur and Pianka 1966 ) and opti mal migration ( Alerstam and Lindström 1990 ; Alerstam 2011 ) , birds during s topover shou ld prefer food that permits greater energy gain with the same handling effort to maximize refueling performance and minimize predation risk (see also Sih and Christensen 2001 ; Shochat et al. 2002 ) . Because native fruits are generally more energy - dense ( White 1989 ; Smith et al. 2007 ; Smith et al. 2013 ) , we predicted that certain native fruits would be selectively consumed. Despite the inherent difficulties of detection of diet items using fecal samples ( Blake and Loiselle 1992 ; Parrish 1997 ) , our results, along with previous work based on fruit removal ( Smith et al. 2013 ) , suggest that high - energy native fruits are selected for by landbird migrants. In particular, w e found that both g ray c t hrushes disproportionately consumed L. benzoin , a native fruit very rich in fat and energy ( White 1989 ; Smith et al. 2013 ) . As expected, we found evidence for birds selecting against the energy - poor exotic fruits of E. umbellata , R. cathartica ( non - significant trend in g ray c atbirds), and R. multiflora t hru shes). I. verticillata was the only native fruit that was selected against by g ray c atbirds (trend) and t hrushes. Because I. verticillata was among the least energy - dense of the native 19 fruits found at our site ( based on White 1989 ) , birds were likely consuming other abundant native fruits that were higher in energy content. The shrub and fruit composition data showed that the highly - selected L. benzoin was only abundant in native shrubland, and fruits found in exotic sh rubland were predominantly those selected against ( E. umbellata , R. cathartica , and R. multiflora ). Thus, selection of certain fruits explains the pattern of high native shrubland use by g ray c t hrushes. Based on the observation that birds select against some of the most common exotic fruits within the overall study site, there are two possible explanations for the use of exotic shrubland by refueling landbirds: optimal foraging and insufficient information about relative habitat qual ity. Optimal foraging would result in the ideal free distribution in which some individuals occupy lower - quality habitat to avoid competition and attain equal fitness to those using higher - quality habitat ( Emlen 1966 ; MacArthur and Pianka 1966 ; Shochat et al. 2002 ) . But this is based on the assumption that bir ds are familiar with all available habitat, which is not the case during migratory stopover for most landbirds ( Moore et al. 1995 ; Shochat et al. 2002 ) lack of information about relative habitat quality (perceptual limitation) could res ult in use of non - optimal habitats ( Abrahams 1986 ; Shochat et al. 2002 ) . Perceptual limitation is also related to ecological traps in that birds are drawn to sub - optimal habitat as a result of mismatching between cues of habitat quality and actual quality of habitats ( We ldon and Haddad 2005 ) . Our data suggest that both model s could partially explain the observed habitat occupancy. Consistent with optimal foraging, bird density was greater in native shrubs, and avoiding that density by foraging in exotic shrub s d id not appear to compromise refueling success with the except ion of gray catbirds in 2013 . Evidence for perceptual limitation comes from the pattern of greater habitat switchover rate from exotic to native compared to vice versa. Furthermore, from a 20 sub set of g ray c atbirds with blood samples, we found that several measures of health status (antioxidant capacity and immunity) were lower in individuals using exotic shrubland than conspecifics captured in native shrubland (Y. Oguchi and J. Owen personal obs ervation). Future studies should quantitatively test optimal foraging vs. perceptual limitation as explanations for exotic shrubland use. Conclusions and Management Implications Our results suggest that native - dominated shrubland is more important to fal l migrating landbirds relative to exotic - dominated shrubland, and that this importance derives from the value of certain native fruits such as L. benzoin . While primarily arthropodivorous migrants such as m agnolia w arblers may be less impacted by invasions of exotic shrubs, our results suggest that fall frugivorous species such as g ray c t hrushes would select against exotic shrubs and their abundant fruits, moving out into native shrublands if available. These findings should enc ourage managers to conserve and restore L. benzoin across its native range in North America, spanning eastward from Ontario and Texas to the Atlantic coast ( USDA and NRCS 2015 ) . This lipid - rich fruit ( White 1989 ; Smith et al. 2013 ) was disproportionately consumed relative to its abundance in both exotic and native shrublands. We also found L. benzoin fruits to be among the richest in antioxidant capacity among t he 14 different fruits assayed in our study site, with moderate concentration of immunostimulatory carotenoids (Y. Oguchi, personal observation). Thus, we consider L. benzoin to be the single most important fruit for fall frugivorous migrating land birds in mid - Michigan. The relative value of exotic fruits for migrating landbirds would depend on what other food resources are present at a stopover site. E xotic fruits may be of high importance when a 21 stopover site is set in a landscape where few other fruit in g species are present ( Gleditsch and Carlo 2011 ) or for late migrant arriving after native fruits have been depleted ( White and Stiles 1992 ; Craves 2009 ) . In our study site, the possibility remains, therefore, that native fruits of high value could have been depleted after the termination of field season at the end of October, in which case the relative imp ortance of exotic shrubland might have increased. Nevertheless, our results support the view that exotic shrubs are of limited value to fall frugivorous migrants in the presence of native counterparts ( Smith et al. 2013 ) , and our site - wide fruit abundance and consumption data suggests that these birds can select for certain high - energy native fruits even though exotic fruits such as R. cathartica and E. umbellata are more abundant. We thus recommend efforts on preventing encroachment of exotic shrubs into native - dominated shrubland and partially supplanting exotic shr ubs in a predominantly exotic shrubland by native shrubs such as L. benzoin if management priority is to improve habitat quality for fall frugivorous migrants. We conclude that native shrublands represent superior quality habitat compared to exotic counter parts and that this finding is limited to fall frugivorous migrants . Evaluation of exotic shrub impact on native migrating landbirds is complex and may be contingent on phenology over the annual cycle in addition to from inter - species and local variation i n other food resources as described earlier ( reviewed in Rodewald 2012 ) . For example, r ecent evidence indicates that the exotic shrublands dominated by Lonicera spp. may function as an ecological trap for breeding birds ( Rodewald 2009 ; Rodewald et al. 2010 ; Rodewald et al. 2011 ) , while these shrubs have been shown to produce sufficient arthropod abundance that permit successful refueling in spring ( Smith 2006 ; Smith and Hatch 2008 ) . S tudies that associate seasonally - dep endent nutritional opportunities in exotic and native shrubs with food and habitat selection over the entire annual 22 cycle and will likely enhance our understanding of the impact of these shrubs on landbird populations and communities. A cknowledgements We thank many assistants and volunteers for mist netting and banding, particularly, C. Gesmundo, L. Theile, C. Gawne, K. Johnson, K. Clark, D. Boston, K. Wilson, and N. Martineau. In addition to mist - netting, H. Matsumura assisted in habitat sampling and i dentification of woody vegetation. J. Craves, D. Williams, M. Quigley, E. Cohen, and S. McWilliams provided valuable advice and support for data analyses and interpretation. F. Moore and R. Campa lent us valuable equipment. We thank the Michigan Department of Natural Resources for allowing us to use the Rose Lake State Wildlife area. We also thank two anonymous reviewers for their comments that allowed us to improve our manuscript. Funding for our work was provided by USFWS Avian Health and Disease Program, the Hal and Jean Glassen Memorial Foundation, and Michigan State University Undergraduate Research Fund . All work was performed under the following permits: USGS Master Banding (#23629), USFWS Scientific Collection (#MB19470), MI Scientific Collection (#S C1386), Rose Lake Special Use, and Michigan State University IACUC protocol (#07/11 - 145 - 00). 23 APPENDIX 24 Table 1.1. Estimated shrub stem densities (stems/ha) in exotic - and native - dominated shrubland s near East Lansing, MI, USA. Data ( mean ± SE ) are pres ented in decreasing order in exotic shrubland and in increasing order in native shrubland, showing profound difference in composition between the two shrublands. Scientific name Common name U.S. s tatus E xotic shrubland N ative shrubland Elaeagnus umbella ta Autumn Olive Exotic 1575. 8 ± 221.5 30 .0 ± 22.3 Lonicera spp. Honeysuckle Exotic 1459.8 ± 324.6 0 Rosa multiflora Multiflora Rose Exotic 902. 9 ± 299. 7 409. 5 ± 115.6 Vitis riparia Riverbank Grape Native 722.3 ± 250. 1 157.08 ± 42. 7 Akebia quinata Five - leaf Akebia Exotic 600. 7 ± 455. 6 0 Rhamnus cathartica European Buckthorn Exotic 357. 4 ± 83. 9 171.2 ± 70.2 Rubus occidentalis Black Raspberry Native 146. 4 ± 96. 3 7. 1 ± 4. 8 Parthenocissus quinquefolia Virginia Creeper Native 114.0 ± 24. 2 75.89 ± 26.7 Ru bus allegheniensis Common Blackberry Native 57.0 ± 53.0 0 Rhamnus alnifolia Alderleaf Buckthorn Native 51.3 ± 33. 9 1. 8 ± 1. 8 Rhus typhina Staghorn Sumac Native 34.2 ± 34.2 0 Cornus alternifolia Alternate - leaved Dogwood Native 34.2 ± 19.9 0 Malus spp. Crabapple Native 32.3 ± 19. 7 1. 8 ± 1. 8 Toxicodendron radicans Poison Ivy Native 26.6 ± 15. 2 14.1 ± 9.2 Viburnum trilobum American Cranberrybush Native 26.6 ± 26.6 35.3 ± 31. 5 Rubus idaeus American Red Raspberry Native 22.8 ± 22.8 47. 7 ± 20.6 Viburnum opulus European Cranberrybush Exotic 17.1 ± 10.2 42. 4 ± 20. 5 Rubus flagellaris Northern Dewberry Native 13.3 ± 13.3 0 Sambucus canadensis Common Elderberry Native 7.6 ± 7.6 455. 4 ± 240.2 Amelanchier spp. Serviceberry Native 5.7 ± 5.7 28.2 ± 19. 9 Vaccin ium angustifolium Lowbush Blueberry Native 0 1. 8 ± 1. 8 Prunus virginiana Chokecherry Native 0 10. 6 ± 7.6 Betula pumila Bog Birch Native 0 12. 4 ± 12. 4 Crataegus spp. Hawthorn Native 0 21. 2 ± 21. 2 Berberis thunbergii Japanese Barberry Exotic 0 30 .0 ± 30 . 0 Physocarpus opulifolius Common Ninebark Native 0 130.6 ± 74. 2 Solanum dulcamara Bittersweet Nightshade Exotic 0 169.4 ± 58.8 Cornus sericea Red Osier Dogwood Native 0 176.5 ± 161.9 Viburnum lentago Nannyberry Native 0 194. 2 ± 82.8 Corylus Americana American Hazelnut Native 0 356.5 ± 130. 2 25 Table 1.1 (cont d) Scientific name Common name U.S. s tatus E xotic shrubland N ative shrubland Toxicodendron vernix Poison Sumac Native 0 656. 6 ± 121. 9 Lindera benzoin Common Spicebush Native 0 806.6 ± 167.1 Ilex verticillata Common Winterberry Native 0 1163. 2 ± 284.4 Cornus racemosa Gray Dogwood Native 19 .0 ± 19 .0 1350. 3 ± 422.6 26 Figure 1.1. Estimated fruit abundance in exotic - and native - dominated shrublands near East Lansing, MI, USA during f all migration of 2012 . Data (mean ± SE) are presented in decreasing order in exotic shrubland and in increasing order in native shrubland . Asterisk indicates exotic species. Permutational MANOVA revealed significant difference in fruit composition, as well as composition of total exotic vs. total native fruits between shrublands. 27 Figure 1.2. Flying arthropod abundance in exotic - and native - dominated shrublands near East Lansing, MI, USA during fall migration of 2012 and 2013. Arthropods were collecte d using two Malaise traps in each habitat and data are presented as dry biomass ( ± SE). Linear mixed models on ln - transformed data showed no significant habitat difference in arthropod abundance in either year. 28 Figure 1.3. Capture rates of newly - banded landbirds ( ± SE) in exotic - and native - dominated shrublands near East Lansing, MI, USA during fall migration of 2012 and 2013. Birds were passively captured in 15 mist nets in each habitat. Compound Poisson linear mixed models revealed significantl y greater capture rate in native shrubs as compared to exotic counterparts in both years. 29 Figure 1.4. Movement of banded landbirds between exotic - and native - dominated shrublands near East Lasing, MI, USA during fall migratory stopover in 2012 and 2013. S ignificantly greater switchover rate from exotic - to native - dominated shrubland compared to vice versa was found in both years. 30 Figure 1.5. Movement of banded gray catbirds, Swainson s thrushes, and magnolia warblers between exotic - and native - dominated shrublands near East Lasing, MI, USA during fall migratory stopover in 2012 and 2013. S ignificantly greater switchover rate from exotic - to native - dominated shrubland compared to vice versa was found in gray catbirds (B) and Swainson s thr ushes (D) in 2013, and a similar non - significant trend in gray catbirds in 2012 (A). All other comparison groups (C, E, F) did not show difference in switchover rate. 31 Figure 1.6. Proportions of fruits found in f ecal samples of gray catbirds and Swa inson s thrushes captured in shrublands near East Lansing, MI, USA during fall migration. Asterisk denotes exotic species. Site - wide fruit availability data was derived from plot - based fruit counts performed in 2012 (see Figure 1.1 ). Fecal data from 2012 a nd 2013 were pooled to increase sample size (numbers above bars). Compared to proportions of site - wide fruit abundance , consumption was non - random ( P < 0.001). Post hoc tests revealed disproportionately greater consumption of the native L. benzoin by both bird species, while the exotic E. umbellata was selected against. See text for further details. 32 LITERATURE CITED 33 LITERATURE CITED Abrahams M.V. 1986. Patch choice under perceptual constraints: a cause for departures from an ideal f ree distribution. Behav Ecol Sociobiol 19:409 415. Akresh M.E., K. Dinse, J. Foufopoulos, S.C. Schubel, and T. Kowalczyk. 2009. Passerine breeding and post - fledgling habitat use in riparian and upland temperate forests of the American Midwest. Condor 111:7 56 762. Alan R.R., S.R. McWilliams, and K.J. McGraw. 2013. The importance of antioxidants for avian fruit selection during autumn migration. Wilson J Ornithol 125:513 525. Alerstam T. 2011. Optimal bird migration revisited. J Ornithol 152:5 23. Alerstam T. and Å. Lindström. 1990. Optimal bird migration: the relative importance of time, energy, and safety. Pp. 331 351 in E. Gwinner, ed. Bird Migration: Physiology and Ecophysiology. Springer - Verlag, NY, USA. Anderson M.J. 2001. A new method for non - parametric multivariate analysis of variance. Austral Ecol 26:32 46. Blake J.G., B. Loiselle, T. Moermond, D. Levey, and J. Denslow. 1990. Quantifying abundance of fruits for birds in tropical habitats. Stud Avian Biol 13:73 79. Blake J.G. and B.A. Loiselle. 1992. F ruits in the diets of Neotropical migrant birds in Costa Rica. Biotropica 24:200 210. Bolser J.A., R.R. Alan, A.D. Smith, L. Li, N.P. Seeram, and S.R. McWilliams. 2013. Birds select fruits with more anthocyanins and phenolic compounds during autumn migrati on. Wilson J Ornithol 125:97 108. Catling P.M. 2005. Effects of invasive alien plants on birds: some examples from North America. Biodiversity 6:30 39. Cohen E.B., F.R. Moore, and R.A. Fischer. 2012. Experimental evidence for the interplay of exogenous and endogenous factors on the movement ecology of a migrating songbird. PLoS ONE 7:e41818. Collier M.H., J.L. Vankat, and M.R. Hughes. 2002. Diminished plant richness and abundance below Lonicera maackii , an invasive shrub. The American Midland Naturalist 147 :60 71. Costantini D., M. Cardinale, and C. Carere. 2007. Oxidative damage and anti - oxidant capacity in two migratory bird species at a stop - over site. Comp Biochem Physiol C 144:363 371. 34 Craves J.A. 2009. A fifteen - year study of fall stopover patterns of Catharus thrushes at an inland, urban site. Wilson J Ornithol 121:112 118. Dierschke V. 2003. Predation hazard during migratory stopover: are light or heavy birds under risk? J Avian Biol 34:24 29. Drummond B.A. 2005. The selection of native and invasive p lants by frugivorous birds in Maine. Northeast Nat 12:33 44. Dunn E. and G.A. Hall. 2010. Magnolia Warbler ( Setophaga magnolia ). In A. Poole, ed. The Birds of North America Online. Cornell Lab of Ornithology, Ithaca, NY, USA. doi:10.2173/bna.136. Dunn E.H. 2000. Temporal and spatial patterns in daily mass gain of Magnolia Warblers during migratory stopover. Auk 117:12 21. Emlen J.M. 1966. The role of time and energy in food preference. Am Nat 100:611 617. Gleditsch J.M. and T.A. Carlo. 2011. Fruit quantity of invasive shrubs predicts the abundance of common native avian frugivores in central Pennsylvania. Divers Distrib 17:244 253. Helms C.W. and W.H. Drury. 1960. Winter and migratory weight and fat field studies on some North American buntings. Bird - Banding :1 40. James F.C. and H.H. Shugart Jr. 1970. A quantitative method of habitat description. Audubon Field Notes 24:727 736. Jenni - Eiermann S., L. Jenni, S. Smith, and D. Costantini. 2014. Oxidative stress in endurance flight: an unconsidered factor in bird migration. PLoS ONE 9:e97650. Johnson M.D. 2007. Measuring habitat quality: a review. Condor 109:489 504. Klaassen M., B.J. Hoye, B.A. Nolet, and W.A. Buttemer. 2012. Ecophysiology of avian migration in the face of current global hazards. Philos Trans R So c B 367:1719 1732. Ktitorov P., F. Bairlein, and M. Dubinin. 2008. The importance of landscape context for songbirds on migration: body mass gain is related to habitat cover. Landscape Ecol 23:169 179. MacArthur R.H. and E.R. Pianka. 1966. On optimal use o f a patchy environment. Am Nat 100:603 609. Mack D.E. and W. Yong. 2000. Swainson's Thrush ( Catharus ustulatus ). In A. Poole, ed. The Birds of North America Online. Cornell Lab of Ornithology, Ithaca, NY, USA. doi:10.2173/bna.540. 3 5 McDonald J.H. 2014. Handb ook of Biological Statistics, 3rd ed. Sparky House Publishing, Baltimore, MD, USA. Moore F.R., S.A. Gauthreaux Jr, P. Kerlinger, T.R. Simons, T.E. Martin, and D.M. Finch. 1995. Habitat requirements during migration: important link in conservation. Pp. 121 144 in T.E. Martin and D.M. Finch, eds. Ecology and Management of Neotropical Migratory Birds: A Synthesis and Review of Critical Issues. Oxford University Press, NY, USA. Moore F.R., R.J. Smith, and R. Sandburg. 2005. Stopover ecology of intercontinental migrants: En route problems and consequences for reproductive performance. Pp. 251 261 in R. Greenberg and P.P. Marra, eds. Birds of Two Worlds: The Ecology and Evolution of Migration. Johns Hopkins University Press, Baltimore, MD, USA. Newton I. 2006. Can conditions experienced during migration limit the population levels of birds? J Ornithol 147:146 166. Oksanen J., F.G. Blanchet, R. Kindt, P. Legendre, P.R. Minchin, R.B. O'Hara, G.L. Simpson et al. 2015. vegan: Community Ecology Package. A vignette for R package vegan. Available from http://cran.r - project.org/web/packages/vegan/ . Orr S.P., J.A. Rudgers, and K. Clay. 2005. Invasive plants can inhibit native tree seedlings: testing potential alle lopathic mechanisms. Plant Ecol 181:153 165. Owen J.C. and F.R. Moore. 2006. Seasonal differences in immunological condition of three species of thrushes. Condor 108:389 398. --- . 2008. Relationship between energetic condition and indicators of immune func tion in thrushes during spring migration. Can J Zool 86:638 647. Parrish J.D. 1997. Patterns of frugivory and energetic condition in Nearctic landbirds during autumn migration. Condor:681 697. Pyle P. 1997. Identification Guide to North American Birds, par t 1: Columbidae to Ploceidae. Slate Creek Press, Bolinas, CA, USA. R Development Core Team. 2014. R: a language and environment for statistical computing. R Foundation for Statistical Computing, Vienna. http://www.R - project.org/ . Rodewald A.D. 2009. Urban - associated habitat alteration promotes brood parasitism of Acadian Flycatchers. J Field Ornithol 80:234 241. --- . 2012. Spreading messages about invasives. Divers Distrib 18:97 99. Rodewald A.D., D.P. Shustack, and L.E. Hitchcock. 2010. Exotic shrubs as ephemeral ecological traps for nesting birds. Biol Invasions 12:33 39. 36 Rodewald A.D., D.P. Shustack, and T.M. Jones. 2011. Dynamic selective environments and evolutionary traps in human - dominated landscapes. Ecology 92:1781 1788. Shochat E., Z. Abramsky, B. Pinshow, and M. Whitehouse. 2002. Density - dependent habitat selection in migratory passerines during stopover: what causes the deviation from IFD? Evol Ecol 16:469 488. Sih A. and B. Christensen. 2001. Optimal diet theory: when does it work, and when and why does it fail? Anim Behav 61:379 390. Sillett T.S. and R.T. Holmes. 2002. Variation in survivorship of a migratory songbird throughout its annual cycle. J Anim Ecol 71:296 308. Smith R.J. 2006. Use of forested an d upland shrub/scrub habitats by spring migrating landbirds in northeastern Pennsylvania. Progress Report for Grant Agreement WRCP - 05109. The University of Scranton, Scranton, PA. Smith R.J. and M.I. Hatch. 2008. A comparison of shrub - dominated and foreste d habitat use by spring migrating landbirds in northeastern Pennsylvania. Condor 110:682 693. Smith R.J., M.I. Hatch, D.A. Cimprich, and F.R. Moore. 2011. Gray Catbird ( Dumetella carolinensis ). In A. Poole, ed. The Birds of North America Online. Cornell La b of Ornithology, Ithaca, NY, USA. doi:10.2173/bna.167. Smith R.J. and F.R. Moore. 2003. Arrival fat and reproductive performance in a long - distance passerine migrant. Oecologia 134:325 331. Smith S.B., S.A. DeSando, and T. Pagano. 2013. The value of nativ e and invasive fruit - bearing shrubs for migrating songbirds. Northeast Nat 20:171 184. Smith S.B., K.H. McPherson, J.M. Backer, B.J. Pierce, D.W. Podlesak, and S.R. McWilliams. 2007. Fruit quality and consumption by songbirds during autumn migration. Wilso n J Ornithol 119:419 428. Stiles E. 1993. The influence of pulp lipids on fruit preference by birds. Vegetatio 107:227 235. Suthers H.B., J.M. Bickal, and P.G. Rodewald. 2000. Use of successional habitat and fruit resources by songbirds during autumn migra tion in central New Jersey. Wilson Bull 112:249 260. Systat Software. 2008. SigmaPlot version 11.0. San Jose, California. Tietz J.R. and M.D. Johnson. 2007. Stopover ecology and habitat selection of juvenile Swainson's Thrushes during fall migration along the northern California coast. Condor 109:795 807. 37 USDA and NRCS. 2015. The PLANTS Database. National Plant Data Team, Greensboro, NC, USA. http://plants.usda.gov . Weldon A.J. and N.M. Haddad. 2005. The effects of pat ch shape on Indigo Buntings: evidence for an ecological trap. Ecology 86:1422 1431. White D.W. 1989. North American bird - dispersed fruit: ecological and adaptive significance of nutritional and structural traits. PhD diss. Rutgers University, New Brunswick , NJ. White D.W. and E.W. Stiles. 1992. Bird dispersal of fruits of species introduced into eastern North America. Can J Bot 70:1689 1696. Wilcove D.S. and M. Wikelski. 2008. Going, going, gone: is animal migration disappearing. PLoS Biol 6:e188. Woods K.D . 1993. Effects of invasion by Lonicera tatarica L. on herbs and tree seedlings in four New England forests. Am Midl Nat 130:62 74. Zhang Y. 2013. cplm: Compound Poisson Linear Models. A vignette for R package cplm. Available from http://cran.r - project.org/web/packages/cplm . 38 CHAPTER 2: H ealth consequences of differential stopover habitat use and fruit consumption in fall migrating landbirds 39 ABSTRACT Migration is a physiologically - demandin g activity, and recent studies suggest the importance of migrating birds being able to restore immune and antioxidant status during stopover. However, knowledge is lacking on whether or how differential use of stopover habitat impacts the health status of migrating landbirds. We tested whether frugivorous fall migrating landbirds using exotic - dominated shrubland in central Michigan, USA experience difference in health relative to conspecifics captured in native - dominated shrubland. We assayed blood from mig rating gray catbirds ( Dumetella carolinensis Catharus ustulatus ) captured in exotic or native shrubland in the fall of 2012 and 2013 for indices of refuel ing performance, constitutive innate immunity, total circulating carotenoids and antioxidant capacity . We did not find habitat difference in p lasma triglyceride levels (refueling index) i n either species or year, but three immune indices (granulocyte:lymphocyte ratio, hemagglutination activity and haptoglobin) and plasma total car otenoids were significantly lower in gray catbirds using exotic shrubs than in conspecifics captured in native shrubs in 2013. Gray catbirds in exotic shrubland also showed lower antioxidant capacity in both years. No such differences were found in Swainso Based on the carotenoids and antioxidant capacity of fruits and patterns of consumption, we found limited evidence that differential consumption of exotic and native fruits might have been reflected in the health status of these birds . These results provide the first evidence that differential stopover habitat use may have important health consequences for migrating landbirds and suggest the value of including measures of immunity and antioxidant status in ecophysiology studies evaluating stop over habitat quality, particularly when nutritional difference is suspected between habitats. 40 Introduction Migration involves major modulations of physiological status and trading off immune defense with high energy demands of endurance flight may occur in migrating birds (reviewed in Piersma 1997 ; Altizer et al. 2011 ) . Evidence shows that during migration the spleen , a principal immune organ in birds ( Fänge and Silverin 1985 ) , shrinks in size ( Deerenberg et al. 2002 ) , and circulating leukocyte counts ( Owen and Moore 2006 ) are lower in migrating birds than in conspecifics during breeding. Downregulation of immune activity has also been found in individuals with experimentally - induced migratory disposition ( Owen and Moore 2008b ) . Furthermore, prolonged flight ( Nebel et al. 2012 ; Nebel et al. 2013 ) and limited food supply ( Buehler et al. 2009 ) were shown to result in reduc ed measures of constitutive (non - induced) immunity (but see Hasselquist et al. 2007 ; Hegeman n et al. 2012 for induced response ) , and migrating birds in poor energetic condition may experience greater risk of infection ( Owen and Moore 2008a ) . M igratory flight is also accompanied by oxidative stress result ing from production of large quantities of tissue - damaging reactive oxygen species ( Costantini et al. 2007 ; Costantini 2008 ; Jenni - Eiermann et al . 2014 ) and depletion of antioxidants ( Costantini 2008 ; Costantini et al. 2008 ) . A reduction in immune status may play a role in the high mortality that occurs during migration ( see Sillett and Holmes 2002 ) , and - quality stopover sites to rest and refuel may further exacerbate their condition and have major fitness conseque nces ( Moore et al. 1995 ; Smith and Moore 2003 ; Moore et al. 2005 ) . Recently, physiological parameters related to individual performance have been identified as valuable tools to evaluate habitat quality ( Albano 2012 ) . T he a nalyses of plasma fat metabolites , whose profiles change within hours of foraging ( Jenni and Jenni - Eiermann 1992 ; Jenni - Eiermann and Jenni 1994 ) , have gained popularity in c ompar ing individual refueling 41 performance between different habitat types (e.g., Guglielmo et al. 2005 ; Cerasale and Guglielmo 2010 ; Smith and McWilliams 2010 ; Evans Ogden et al. 2013 ; Liu and Swanson 2014 ) . However, refueling success is only one aspect of performance, and evaluations of immune and antioxidant status are also likely germane to habitat assessment ( Albano 2012 ) . Indeed, the need to understand factors influencing health status of migratory birds facing anthropogenic landscape changes has been identified as a research priority for avian ecophysiologists ( Klaassen et al. 2012 ) . Limited evidence indicates that successful stopover involves recover y or upregulat ion of immune function ( Owen and Moor e 2008a ; Buehler et al. 2010 ; Nebel et al. 2012 ) and circulating antioxidants ( Costantini et al. 2007 ; Alan et al. 2013 ) that were reduced duri ng flight. However, to our knowledge it is unknown whether stopover habitat quality has any consequences on immune and antioxidant status. It is conceivable that differential stopover habitat use manifests as difference s in health status, particularly whe n dietary quality differs between habitats . Here, we focused on the difference in the fruit quality between exotic and native shrubs as part of a larger study to evaluate stopover habitat quality between predominantly exotic and predominantly native shrubl ands . Introductions of exotic shrubs due to human land - use change has been identified as a potential conservation concern for migrating landbirds ( Suthers et al. 2 000 ; Catling 2005 ; Craves 2009 ) , particularly because exotic fruits common in Midwestern and N orth eastern United States such as autumn olive ( E laeagnus umbellata ), European buckthorn ( Rhamnus cathartica ) , and honeysuckle ( Lonicera spp.) generally have higher sugar contents and lower energetic potential compared to the native counterparts ( White 1989 ; Smith et al. 2007a ; Smith et al. 2013 ) . Thus, consuming the high - fat, high - energy native fruits such as common spiceb ush ( Lindera benzoin ), gray dogwood ( Cornus racemosa ), and Virginia creeper ( Parthenocissus quinquefolia ) ( White 42 1989 ; Smith et al. 201 3 ) could be a d vantageous for refueling ( Smith et al. 2007a ; Smith et al. 2013 ) . These nutritional differences and observed selection of certain native fruits has led previous work ( Smith et al. 2013 ) to conclude that exotic shrubs may represent poorer stopover habitat than native counterparts during fall migration. Refueling performance may also impact recovery of overal l health status because superior energetic condition has been linked to stronger immune function ( Owen and Moore 2008a ; Buehler et al. 2010 ) and antioxidant capacity ( Costantini et al. 2007 ) during stopover. But exotic and native fruits also vary by species in their antioxidant capacity and carotenoid contents ( Alan et al. 2013 ) , and carotenoids are well - known for their strong immun ostimulatory effects in birds ( McGraw and Ardia 2003 ; McGraw et al. 2006 ; McGraw et al. 2011 ; Simons et al. 2012 ) . Because birds select fruits based on a combination of different characteristics including fat and antioxidant contents ( Stiles 1993 ; Alan et al. 2013 ; Bolser et al. 2013 ; Smith et al. 2013 ; Pierce and McWilliams 2014 ) , the combined effect of fruit quality and consumption likely impacts a migrant s health status. However, the health consequences of consuming exotic fruits relative to consuming native counterpart s are largely unknown. In the present study, we investigated whether health status of fall migrating landbirds using exotic - dominated shrubland differed from conspecifics using native - dominated counterpart. Because migrants need to refuel and restore thei r health during stopover (sensu Costantini et al. 2007 ; Costantini et al. 2008 ; Owen and Moore 2008a ; Buehler et al. 2010 ; Nebel et al. 2013 ) , a stopover habitat in which individuals exhibit greater indices o f fat gain, immune function, and antioxidant capacity could be considered superior - quality. We assessed (1) refueling performance , (2) cellular (leukocytes) and humoral (plasma protein) constitutive innate immunity, (3) plasma total carotenoids, and (4) pl asma antioxidant capacity of two species of fall 43 frugivorous migrants . As indices of immunity we focused on the constitutive innate branch b ecause it represent the first line of defense for birds encountering novel pathogens during migration and deemed mos t relevant during short stopover when acquired immunity is too slow to respond ( Schmid - Hempel and Ebert 2003 ; Buehler et al. 2010 ) . The immun e parameters selected for this study ( Davis 2005 ; Sepp et al. 2010 ; van de Crommenacker et al. 2010 ; Nebe l et al. 2012 ) , as well as circulating carotenoids ( Hõrak et al. 2004 ) and antioxidant capacity ( van de Crommenacker et al. 2010 ) appear to change within a range of several h ours to several days, well within the stopover lengths of fall migrating landbirds ( Woodrey and Moore 1997 ; Craves 2009 ) . Additionally, we investigated whether differential fruit quality and consumption were reflected as habitat - related differences in health parameters . W e measured total carotenoid concentration and antioxidant capacity in exotic and native fruits found in our study site and c ombin ed these data with consumption proportions as found in fecal samples to test for habitat differences in expected dietary means of carotenoids and antioxidant capacity. This integrative approach provides a more comp rehensive picture of how differential stopover habitat use may have important health consequences for migrating landbirds. Methods Study Site and Habitats We conducted fieldwork during fall migration of 2012 and 2013 on a s tate - managed wildlife area near East Lansing, MI, USA (Rose Lake State Wildlife The site was set largely within an agricultural and suburban landscap e but n o significant geographical barrier was present in surrounding areas . Both exotic - and native - dominated shrublands were at similar lon gitudes and occurred as distinct patches divided by 44 mature oak - maple woodland around 250 m wide. Woody vegetation structure was similar between shrublands (Y. Oguchi and J. Owen, unpublished data). Native - dominated shrubland (hereafter native shrubland) was primarily composed of C. racemosa , L. benzoin , common winterberry ( Ilex verticillata ), poison sumac ( Toxicodendron vernix ), common elderberry ( Sambucus canadensis ), and various saplings of canopy species. Exotic - dominated shrubland (hereafter exotic sh rubland) was composed largely of E . umbellata , honeysuckles ( Lonicera tartarica , L. morrowii , and L. x bella [ morrowii x tatarica ]), and R. multiflora , all introduced from East Asia. Although multiflora rose and the exotic R . cathartica ) occurred in simila r frequencies in both habitats, greater than 70% of the shrub stems in exotic shrubland were exotic, whereas the exotic stem frequency was only 12% in native shrubland (Y. Oguchi and J. Owen, unpublished data). Study Species, Capture, and Blood Collectio n We selected the gray catbird ( Dumetella carolinensis Catharus ustulatus ) as focal species due to their presence and abundance in both shrublands and their high consumption of fruits during fall migration ( fruits present in >93% and >85% in gray catbird and Swainson's thrush fecal samples respectively, Y. Oguchi and J. Owen unpublished data; also see Parrish 1997 ) . Birds were captured passively using 30 mist nets (30 mm - mesh, 12m x 2.6m), 15 in each habitat type, from August 13 to September 30 in 2012 and 2013. Nets were checked at 45 - min intervals beginning 30 min before sunrise for 4 - 5 h daily. Capture time was recorded ( + 1 min) as the time the bird was first detected in the net. We held birds in individual cloth bags or bird boxes until banding and blood collection. Bird boxes were lined with a sheet of paper to collect fecal samples for fruit consumption analysis. Birds were banded with U.S. Geological 45 Survey aluminum bands and aged as hatch - year (HY) or after hatch - year (AHY) using plumage or skull ossification ( Pyle 1997 ) . Both species are sexually monomorphic ( Pyle 1997 ) so sex could not be determined. Unflattened wing chord was measured to the nearest mm and mass to the nearest 0.1 g. We visually assigned fat scores based on furcular and abdomina l fat storage (0 - 6 scale, 0.5 increments) ( modified from Helms and Drury 1960 ) and pectoral muscle scores (0 - 3 scale, 0.5 increments) ( modified from Bairlein et al. 1995 ) . We collected 30 0 µl of blood ( <1% of body mass representing <10% of blood volume; see Owen 2011 ) from the brachial vein directly into heparinized capillary tubes. Time of blood sampling ( ± 1 min) was recorded and handling time was calculated by subtracting the capture time (see above) from the time of bleeding. Upon collection we made a single - cell layer slide which was air dried, fixed in 100% methanol, and later stained with Wright - Giemsa solution (Fisher Scientific Hema3 # 22 - 122 - 911 , Hampton, NH, USA ) . The r emaining blood was stored over ice in the field until centrifuged. After centrifugation , plasma was collected and stored in - 8 Capture rates of study species were lower in exotic compared to native shrubland (Y. Oguchi and J. Owen, unpublished data), and in total, we sampled 212 ( exotic shrubland: 46; native shrubland: 166) and 196 (exotic shrubland: 46; native shrubland: 150) gray catbirds in 2012 and 2013 respectively, and 203 (exotic shrubland: 82, native shrubland: 121) and 229 (exotic shrubland: 108; native shrubland: 12 1) Swainson s thrushes in 2012 and 2013 respectively . Sample sizes varied for each assay due to inter - individual variation in the blood volume we were able to collect (see means ± SE and sample size for health parameters in Table 2.1). All work was perform ed under the following permits: USGS Master Banding (#23629), USFWS Scientific Collection (#MB19470), MI Scientific Collection (#SC1386), Rose Lake 46 Special Use, and Michigan State University IACUC protocol (#07/11 - 145 - 00). Indices of Refueling Performanc e Size - corrected F at M ass S ize - corrected fat mass ( Owen and Moore 2006 ; 2008a ) , regressed against time after sunrise, can be used to detect habitat differences in fat gain or loss ( see Statistical Analyses and Dunn 2000 ) . We used this method in addition to plasma metabolites (see below) because it provides estimates of actual rate of fat mass change and because plasma metabolites can be sensitive to capture effects ( Jenni and Jenni - Eiermann 1992 ; Guglielmo et al. 2002 ) and may be affected by dietary differences in carbohydrates and fats ( Seaman et al. 2005 ; Smith et al. 2007b ; Smith and McWilliams 2009 ) . To estimate size - corrected fat mass, we first grouped conspecific individuals by 1 - mm - increment wing chord class and regressed body mass on fat score using data of the study species captured at our site during fall of 2009 - 2014. The intercept (fat score = 0) was taken as the fat - free body mass for that wing chord class. We then regr essed size - specific fat - free mass on wing chord to produce a linear equation, which was used to calculate size specific fat - free mass of each individual. The size - corrected fat mass was calculated by subtracting the fat - free mass from the actual mass. Pla sma Metabolites Plasma triglycerides ( triacylglycerol: - hydroxybutyrate (BUTY) levels are common ly applied to index stopov er habitat quality (e.g., Cerasale and Guglielmo 2010 ; Evans Ogden et al. 2013 ) . In general, high levels of TRIG represent greater fattening rates within a few hours prior to capture whereas higher BUTY levels represent greater rates of fat catabolism ( Jenni and Jenni - Eiermann 1992 ; Jenni - Eiermann and Jenni 1994 ; 1996 ; 1998 ) . While plasma 47 metabolites do not translate to actual fat mass being gained, the advantage of this approach is that these metabolites are very precise and time - sensitive ( Guglielmo et al. 2005 ) . Previous studies (e.g., Guglielmo et al. 2002 ; Guglielmo et al. 2005 ; Smith and McWi lliams 2010 ; Evans Ogden et al. 2013 ) suggest that TRIG is the most reliable and informative plasma lipid metabolite for comparing habitat differences in refueling performance, while some doubt has been placed on the utility of BUTY for this purpose (see Acevedo Seaman et al. 2006 ; Smith and McWilliams 2010 ; Evans Ogden et al. 2013 ) . We used commercially - available reagents to measure both plasma TRIG and BUTY concentrations (mmol/l) as previously described ( Guglielmo et al. 2002 ; Guglielmo et al. 2005 ) . Plasma samples (10 µl) were serially diluted (1:2 and 1:4) in 10 µl of 0.9% NaCl solution . Two - fold and four - fold dilutions w ere used for TRIG and BUTY respectively, which yielded concentrations that fell within the range of the standard curves. TRIG was assayed sequentially using an endpoint assay. Briefly, 2.5 µl of diluted plasma w ere pipetted into a flat - bottom 96 - well assay plate (Thermo Nunc #47509 4, Waltham, MA, USA) with 120 µl of g lycerol r eagent . After taking an initial reading with , 30 µl of t riglyceride r e agent ( converts TRIG to glycerol, Sigma # T2449, St. Louis, MO, USA) w ere added, incubated, and read under the same conditions. The concentration TRIG was obtained by subtracting the concentration of free glycerol (first reading) from total glycerol (secon d reading). V alues were multiplied by 2 to correct for the dilution. Standard curves were obtained by serially diluting (5.64, 2.82, 1.41, 0.705, 0.3535, and 0 mmol/l) the g lycerol s tandard (Sigma, #G7793, St. Louis, MO, USA). Assays were run in duplicate. BUTY assays were performed using a kinetic endpoint assay (R - Biopharm, 48 #10907979035, Marshall, MI, USA). Briefly, 5 µl of diluted plasma w ere pipetted into a flat - bottom 96 - well plate with 200 µ l of working solution of reagents prepared as per manufactu rer s directions. After 2 min of incubation at room temperature, - hydroxybutyrate dehydrogenase suspension and three times the change in absorbance between 30 - 40 min (background) was subtracted from the change between 0 - 40 min. This change , linea rly related to BUTY concentration, was calculated against the change in standards of 4, 2, 1, 0.5, 0.25, and 0 mmol/l of BUTY (Sigma #166898, St. Louis, MO, USA) . The measures were multiplied by 4 to correct for the dilution . Assays were run in duplicate. Indices of Constitutive Innate Immune Function Total Leukocyte Count and Granulocyte: l ymphocyte Ratio Total leukocyte count have been widely used as a measure of immunological readiness (e.g., Campbell and Dein 1984 ; Kilgas et al. 2006a ; Owen and Moore 2006 ; Bueh ler et al. 2010 ) , and various stressors ( Maxwell 19 93 ; Davis et al. 2008 ) including migration ( Owen and Moore 2006 ; 2008a ) are linked to elevated heterophil:lymphocyte (H:L) or granulocyte:lymphocyte (G:L) ratio s . We performed d ifferential leukocyte counts on blood smears by observing 100 fi elds of view under ×1 , 000 magnification (oil - immersion) representing an average of 200 erythrocytes ( Campbell and Dein 1984 ; Owen and Moore 2006 ; Owen 2011 ; Owen et al. 2013 ) . Total leukocyte count was obtained as the tally of all leukocyte types (heterophil, eosinoph il, lymphocyte, monocyte, and basophil) divided by 2 and is hence express ed as the number of leukocytes per 10,000 red blood cells. G:L ratio was calculated by dividing the number of granulocytes (tally of heterophils, eosinophils, and basophils ) by the nu mber of lymphocytes. 49 H emolysis and H emagglutination Activities We modified the hemolysis - hemagglutination assay ( Matson et al. 2005 ) to index activit ies of natural antibodies and complement in the plasma. This technique chara cterizes the ability of natural antibodies to agglutinate foreign red blood cells and to activate the complement system , which results in lysi s and w ere pipetted into column 2 of a rou nd - bottom 96 - well plate (Corning Coaster #3795, Corning, MO, USA) w ere pipetted into columns 2 - 12 , and samples were serially diluted (1:2) through column 12 using a mu lti - channel pipette showed hemolysis at 1:2 dilution of plasma, 10 µl of PBS w ere pipetted into column 1 as negative control (instead of undiluted plasma in the original protocol ). A 1% rabbit red blood cell suspension was prepared from a commercially available source (HemoStat Laboratories #RBA050, Dixon, CA) and ere added to each well. Chicken serum was assayed in row 1 of each plate as a quality control standard. Plates were gently shaken, covered with an acetate plate sealer (Thermo #3501, Milford, MA), and then incubated in a 37 °C water bath for 90 min. After removal from the bath, plates were tilted at a 60 - degree angle for 20 min at room temperature to enhance the visualization of hemagglutination. Plates were scanned at 300 dpi using Epson Perfection 4180 Photo Scanner for h emagglutination scor ing. A hemolysis scan was then taken after an additional 70 min of incubation at room temperature. From the scans hemolysis and hemagglutination were score d blindly by a single observer as the last well (highest dilution) exhibit ing the respective activity. Thus in principle, scores represent log - 2 transformed (1:2 serial dilutions) measures of hemolysis or hemagglutination activity . 50 Haptoglobin Haptoglobin is an acute phase protein that binds to freely - circulating heme (iron ) preventing it from becoming a nutrie nt to pathogens ( Delers et al. 1988 ) , and its baseline level generally predicts readiness of a n acute phase response ( Matson et al. 2012a ) . In addition, because heme induces oxidative stress during endurance exercise ( Cooper et al. 2002 ) , the role of haptoglobin as an antioxidant ( Gutteridge 1987 ) may play an important role during migration. We assayed haptoglobin using a commercially - available kit (Tridelta Development #TP801, Maynooth, County Kildare, Ireland). Plasma samples remaining from previous assays were used because one freeze - thaw cycle does not measurably degrade haptoglobin ( Brinc et al. 2012 ) . Following the manual method described by the manufacturer, 50 µl of Reagent 1 were added to 3.75 µl of plasma samples in a 96 - well flat - 630 nm, 70 µl of Reagent 2 was added and incubated at room temperature for 5 min. Then, the plate was immediately read at the same absorban ce. T o correct for the initial variation in plasma color or turbidity, the background was subtracted from the final reading . The standard curve was generated from serial dilutions (2.5, 1.25, 0.625, 0.312, and 0 mg/ml) of the calibr ator serum. Concentratio ns of samples are reported as mg/ml. Plasma Total Carotenoids Carotenoids are dietary - derived pigments that play a minor role as antioxidants ( Costantini and Møller 2008 ; Simons et al. 2012 ) and have strong stimula tory effects on various branches of immunity ( McGraw and Ardia 2003 ; McGraw et al. 2006 ; McGraw et al. 2011 ; Simons et al. 2012 ) . We used a spectrophotometric assay to measure plasma total carotenoid concentration ( Tella et al. 1998 ; Ninni et al. 2004 ; Hõrak et al. 2006 ; McGraw et al. 2008 ) . Plasma samples 51 remaining from previous assays (one prior freeze - thaw cycle) were used ( Nierenberg 1985 ; Comstock et al. 2001 ) . Briefly, 90 µl of absolute ethanol w ere added to 10 µl of plasma (1:10 dilution) to extract carotenoids and precipitate p roteins. The mixture was vortexed for 10 sec and centrifuged for 1 min under 16,800× g . Then, 70 µl of the supernatant was pipetted into a flat - bottom 96 - well assay plate. Clear plastic tape was placed over each row to prevent evaporation of ethanol until r eading at xanthophyll standard (Sigma #X6250, St. Louis, MO, USA) serially diluted (20, 10, 5, 2.5, 1.25, and 0 µg/ml) in absolute ethanol . M easurements of plasma total carotenoids were multiplied by 100 to correct for the dilution and were expressed as µg/ml. Plasma Total Antioxidant Capacity Antioxidants protect migrating birds against oxidative damage resulting from large quantities of reactive oxygen species produced during strenuous flight ( Costantini et al. 2007 ; Costantini et al. 2008 ; Jenni - Eiermann et al. 2014 ) . Because those oxygen free radicals are also generated by phagocytes during inflammation, antioxidants play a role in preventing immunopathology (reviewed in Matés and Sánchez - Jiménez 1999 ; Costantini 2008 ) . We used the OXY - Adsorbent test (Diacron I nternational # MC435, Grosseto, Italy) to measure the plasma total antioxidant capacity ( Costantini et al. 2006 ; Saino et al. 2011 ) . This functional assay measures the ability of plasma samples to neutralize hypochlorous acid (HOCl), a powerful and pathologically - relevant oxidant in biological systems. Briefly, 5 µl of plasma were diluted 1:100 with distilled wat er (495 µl). Then, 2.5 µl of diluted plasma were incubated with a 100 µl of titered HOCl solution - diethyl - p - phenylenediamine solubilized in a chromogenic mixture were rapidly added , and after gently sha ken the plates were immediately 52 residual HOCl is inversely related to the plasma total antioxidant capacity. A calibrator serum of known antioxidant capacity (350 m mol/l of HOCl neutralized, diluted 1:100) and distilled water (blank) were used as control reference, and calculations were performed as per directions by the manufacturer. Measurements were obtained as mmol/l of HOCl neutralized. Dietary Carotenoids and Antioxidant Capacit y Measuring Carotenoids and Antioxidant Capacity of Fruits For dietary carotenoid and antioxidant capacity assays we sampled both non - native and native fruits in late September 2013. Exotic fruits collected were E. umbellata , Lonicera spp., R. multiflora , Rhamnus cathartica , European cranberrybush ( Viburnum opulus ), Japanese barberry ( Berberis thunbergii ), and bittersweet nightshade ( Solanum dulcamara ). Native fruits collected were C. racemosa , I. verticillata , L. benzoin , T. vernix , S. canadensis , nannyberry ( Viburnum lentago ), and Virginia creeper ( Parthenocissus quinquefolia ). One fruit per shrub (minimum of 5 fruits per species) was collected, and for fruit species found in both exotic and native shrubland ( E. umbellata , Lonicera spp ., R. multiflora , R. cathartica , V. opulus , and L. benzoin ), a minimum of 5 fruits per species were collected from each habitat type. Fruit samples were stored in - freezer until laboratory assays on a later date. We extracted fruit carotenoids in eth anol for spectrophotometric assay ( Rodriguez - Amaya 2001 ; Alan et al. 2013 ) . After removing the seeds, approximately 100 mg of wet fruit, weighed to the nearest 0.01 mg, was placed in 1 ml absolute ethanol (1:10 dilution). Samples were vortexed for 2 min and sonicated for 5 min (Fisher Scientific Sonic Dismembrator Model 100, Hampton, NH, USA). The mixture was then centrifuged for 5 min under 16,800×g, and 70 53 µl of the supernatant was placed in a flat - bottom 96 - We generated a standard curve from 70 µl of serially diluted (50, 25, 12.5, 6.25, 3.125, 1.5625, 0.78125, and 0 µg/ml) xanthophyll standard. Carotenoid concentrations were obtained as µ g/g of wet fruit. We were unable to quantify carotenoids in poison sumac because the supernatant obtained was a turbid, waxy gel. We modified the OXY - Adsorbent test described for bird dietary analysis ( Costantini 2010 ) . After removal of seeds, a pproximately 100 mg of wet fruits ( weighed to the nearest 0.01 mg ) were placed in around 1 ml of distilled water (measured to the nearest µ l) to attain 1:10 dilution, vortexed for 2 min, and sonicated for 5 min. T his emulsion (100 µ l) was further diluted in 400 µl of distilled water (total dilution of 1:100) and vortexed. Then 2.5 µl of this dilution were placed in a flat - bot tom 96 - well assay plate with 100 µl of HOCl. Because the coloration and turbidity of this mixture differed among fruit, initial background absorbance was immediately same setting after 5 min absorbance and the difference used to calculate antioxidant capacity. Using the cali brator serum provided in the kit, the antioxidant 5 capacity was obtained as mmol of HOCl neutralized per kg of wet fruit. Indices of Fruit Consumption We used fecal samples collected from birds held in bird boxes to identify fruits consumed. Identificatio n of fruits relied on a combination of seed, color, and texture of the pulp ( Parrish 1997 ) . We smeared fruits collected in the field on the same paper liner and used them as reference. Frui t remains were identif ied by a single observer blind to which habitat type the bird was captured in, and samples that did not allow positive identification of fruits were removed 54 from further analysis. Because we were interested in relative consumption of fruit species, for each bird species, the number of occurrence of fruits species in the fecal samples was tallied by habitat (i.e., samples with two fruit species were counted twice to represent both fruit species). In total, 63 (exotic shrubland: 6, nativ e shrubland: 57) fecal samples from gray catbirds and 171 (exotic shrubland: 56, native shrubland: 115) samples from Swainson s thrushes in 2012. In 2013, 42 (exotic shrubland: 6, native shrubland: 34) samples from gray catbirds and 39 (exotic shrubland: 2 1, native shrubland: 18) samples from Swainson s thrushes allowed identification. Statistical Analyses All statistical analyses were performed using R Version 3.1.2 ( R Development Core Team 2014 ) . Significance level was set as P < 0.05 for all tests with 0.05 < P < 0.1 deemed as a trend unless otherwise specified. All tests were performed two - tailed and data were transformed as necessary to meet normality assumption. For graphical purposes, least square means (± SE) are presented. Prior to a nalyses of health parameters, we tested for habitat - related differences in age ratio ( 2 contingency test ) and structural size (i.e. wing chord; two - sample t - test ) in gray catbirds and to evaluate whether variations in these potential confounders may have affected our results. Health Parameters We perfor med multiple linear regression s t o test w h ether the rate of change in size - corrected fat mass differed between the two habitat types during the morning foraging hours in each species and year ( Dunn 2000 ; Guglielmo et al. 2005 ) . Specifically, after controlling for ordinal date (positively related to fat mass in all comparison groups, P < 0.05), habitat - related difference in rate of fat mass change was detected as a significant interaction between habitat and time after 55 sunrise. F at mass of Swainson s thrushes were ln - plus - 5 transformed to attain normality and handling time was not significantly related to fat mass ( P > 0.05) and thus not included in any model. When the interaction was non - significant, then we removed the interacti on to test if birds gained or lost mass within the overall study site (main effect of time after sunrise) or whether fat mass differed by habitat type. We then tested whether each of the health parameters differed by habitat in gray catbirds s thrushes in each year. We used ANCOVAs to test habitat - related differences in the health parameters and to calculate least square means for graphing. The distributions for hemolysis score s , hemagglutination scores in gray catbirds, haptoglobin (except in 2012 gr ay catbirds ) , and antioxidant capacity were normal and no transformations were necessary. We transformed TRIG (ln - plus - 1), BUTY (sqrt), total leukocytes (ln), hemagglutination scores of n), and total carotenoids (sqrt) to bring the distributions to normal , but for graphical purposes hemagglutination and haptoglobin data are shown untransformed . Because no transformation was appropriate for G:L ratios, we fitted compound Poisson GLMs (func tion cpglm, package cplm) with log link function s ( Zhang 2013 ) . We first fitted a multiple linear model explaining each health parameter by ordinal date, time after sunrise at capture, handling time, fat score, muscle score, wing chord, and age as predictors. We kept f at and muscle scores as separate covariates b ecause many parameters were only relate d to either fat or muscle (see Table 2.2) used to select predictor variables (summarized in Table 2.2; Guglielmo et al. 2005 ; Liu and Swanson 2014 ) . These variables were retained as covariates in ANCOVA models explicitly testing habitat effect along with covariate × habitat interaction terms. In particular, significant interaction between time after sunrise and habitat in TRIG, similarly to fat mass, is considered a n 56 indicator of differential refueling performance between habitats ( Evans Ogden et al. 2013 ) . To produce the minimum model that included habitat, we eliminated all non - significant covariates ( P > 0.05) with backward selection. All procedures were similar in the compound Poisson GLMs on G:L ratios, but least square means for graphing was obtained by re - fitting the final models as Markov chain Monte Carlo GLMs (function MCMCglmm, package MCMCglmm) due to incompatibility of R packages of cplm and lsmeans ( Hadfield 2014 ; Lenth and Hervé 2015 ) . Multicollinearity was not detected in any of the final models (vif < 10). Dietary Carotenoids and Antioxidant Capacity Fruit total carotenoids and antioxidant capacity were analyzed using one - way ANOVA to test whether there were differences among f ruit species. Post - comparisons analyses were used to compare these measures between each pair of fruit species. None of the 6 fruit species collected from both exotic and native shrublands showed significant differences in caroteno ids or antioxidant capacity by habitat type ( P > 0.05) so these data were pooled by species. W e also pooled fecal data from in 2012 and 2013 by bird and fruit species to increase sample size (see Table 2.3) after confirming that consumption ranks of fruit species within habitat types were similar between years in both bird species (Friedman s rank sum test for year with fruit species as block, P > 0.05) . We then evaluated whether habitat - related variation in carotenoids and antioxidant capacity in birds co uld be attributable to differential fruit quality and consumption. We calculated expected means of carotenoids and antioxidant capacity in a unit mass of fruit consumed in each habitat type by combining the fruit assay fecal data . Specifically, the expecte d values were derived as mean carotenoids or antioxidant capacity weighted by consumption proportion in e ach habitat . Conceptually, for each bird species in each habitat, the e xpected mean 57 dietary carotenoids ( µg/g of wet fruit ) = (carotenoid contents of each fruit its consumption proportion). Similar approach was taken for the expected mean dietary antioxidant capacity ( mmol HOCl neutralized/kg of wet fruit ). These habitat - based dietary means could be then statistically com pared using weighted two - sample t - tests ( function wtd.t.test in package weights in R ), with weights as fecal counts rather than proportions ( Pasek et al. 2014 ) . But because data for carotenoid contents and antioxidant capacity were obtained from multiple fruits with varying sample sizes (range: n = 5 - 11), weight for each fruit required standardization by the sample size of these assays. For example, gray catbirds captured in native shrubland had 63 fecal samples contain ing L. benzoin , with 10 samp les assayed for carotenoids ; the weight for each L. benzoin fruit in native shrubland was entered as 63/10 = 6.3. Results Age and Size by Habitat Type The majority of our sampled birds represented HY birds in both in gray catbirds (HY: AHY = 190: 22 in 205: 24 in 2013). We confirmed that age ratio did not differ by habitat type in gray catbirds 2 1 = 0.18, P 2 1 = 1.06, P 2 1 = 0.97, P 2 1 = 0.08, P = 0.77) in either year. Similarly, wing chord did not differ between habitats in either species or year (gray catbirds, 2012: t 71.27 = 1.04, P = 0.30; 2013: t 69.55 = 1.08, P t 1 48.28 = - 1.54, P = 0.13; 2013: t 218.18 = - 1.30, P = 0.20). Indices of Refueling Performance Based on size - corrected fat mass change during the morning foraging hours, gray catbirds in 58 2012 gained fat in both habitat types by 0.33 g/hr (time after sunrise : F 1, 207 = 7.16 , P = 0.008 ) but the rate of gain did not differ by habitat (time after sunrise × habitat: F 1, 208 = 0.115, P = 0.735). Significant interaction between time after sunrise and habitat ( F 1, 191 = 5.4 3, P = 0.02 ) was found in 2013 gray catbird s (Figure 2.1): fat was lost by 0.60 g/hr in exotic shrubland ( t 191 = - 2.002 , P = 0.0 5) while no change in fat mass was found in native shrubland ( t 191 = 1.206, P = 0.2 3). Rate of fat mass change of Swainson s thrushes did not differ by habitat type in eit her year (2012: F 1, 198 = 0.080 , P = 0.7 8; 2013: F 1, 224 = 0.081 , P = 0.7 8). Additionally, in 2012 fat mass of Swainson s thrushes in native shrubland was greater by 1.20 g ( F 1, 199 = 11.78, P < 0.001) and fat was gained in both habitats by 1.05 g/hr ( F 1, 199 = 5.3 8, P = 0.02 ), while in 2013 no difference in fat mass was found between habitat types ( F 1, 225 = 0.88 9, P = 0.3 5), and no gain or loss in fat was detected across both habitats ( F 1, 225 = 0.32 , P = 0.57 ). Plasma TRIG levels of both study specie s increased with time after sunrise in consistently in 2012 and 2013 ( P < 0.05, Table 2.2), but we found no significant habitat effect in gray catbirds ( 2012 : F 1, 172 = 0.30, P = 0.59 ; 2013 : F 1, 180 = 0.02, P = 0.90) or thrushes (2012: F 1, 159 = 0.002, P = 0.96; 2013: F 1, 217 = 2.22, P = 0.14 ) in either year ( Figure 2.2A). Interactions between time after sunrise and habitat were also non - significant in all comparison groups ( P > 0.05). Plasma BUTY decreased with time after sunrise in both species and years ( P < 0.05, Table 2.2). Gray catbird BUTY in exotic shrubs in 2012 showed a lower trend (by 1.29 mmol/l) than in native shrubs ( F 1, 174 = 3.21, P = 0.07) without significant time after sunrise × habitat interactions ( P > 0.05 ; Figure 2.2B). In 20 13, BUTY of gray catbirds using exotic shrubland decrease d more rapidly with time after sunrise than in conspecifics captured in native shrubs, as indicated by significant time after sunrise × habitat interaction ( F 1, 179 = 4.52, P = 0.03), although the m ain effect of habitat was non - significant ( F 1, 179 = 0.29, P = 59 0.59). This interaction effect was small, however, translating to BUTY decreasing by only an additional 0.039 mmol/l per hour in exotic compared to in native shrubland. A model without the inte raction term also yielded minimal BIC (396.30, as opposed to 396.90 with interaction) so we graphically present least square means by habitat type based on this reduced model (BUTY lower by 1.39 mmol/l in exotic relative to native shrubland; F 1, 180 = 5.17 , P = 0.02 ; Figure 2.2B). Plasma BUTY of F 1, 160 = 1.17, P = 0.28) or 2013 ( F 1, 218 = 1.95, P = 0.16), with no significant interactions between time after sunrise and habitat ( P > 0.05 ; Figure 2.2B). A part from t ime after sunrise , fat score had a consistent ly positive effect on TRIG in both species and year s (Table 2.2). Handling time, if significant, showed a consistent negative effect on TRIG and a positive effect on BUTY (Table 2.2). Indices of Con stitutive Innate Immune Function Total leukocyte counts did not differ by habitat in gray catbirds (2012: F 1, 137 = 0.02, P = 0.89; 2013: F 1, 73 = 0.03, P F 1, 85 = 0.52, P = 0.47; 2013: F 1, 92 = 0.08, P = 0.78) in eith er year (Figure 2.3A). Handling time, whenever significant, negatively impacted total leukocyte counts (Table 2.2). We found a trend toward higher G:L ratio in gray catbirds using exotic shrubland compared to those captured in native shrubland in 2012 ( F 1 , 139 = 3.02, P = 0.08) and a significant difference in the same direction in 2013 ( F 1, 72 = 4.36, P = 0.04 ; Figure 2.3B). - related difference in G:L ratio in either year (2012: F 1, 83 = 1.33, P = 0.25; 2013: F 1, 92 = 0.06, P = 0.81 ; Figure 2.3B). N o covariate consistently explained variations in G:L ratio between species and years (Table 2.2). Hemolysis scores of gray catbirds did not show habitat - related difference in 2012 ( F 1, 162 60 = 1.29, P = 0.26), but in 2013, a trend toward lower hemolysis was observed in individuals captured in exotic shrubland ( F 1, 179 = 3.17, P = 0.08) compared to conspecifics in native shrubland (Figure 2.4A). This trend implies 27% lower hemolysis activity in gray catbirds using exotic sh rubs than in conspecifics captured in native shrubs (back transforming log - 2 scores to activity) F 1, 176 = 0.27, P = 0.61) or 2013 ( F 1, 218 = 1.66, P = 0.20 ; Figure 2.4A). Ordinal date, whenever significant, had a positive effect on hemolysis scores (Table 2.2). Hemagglutination scores of gray catbirds did not differ by habitat in 2012 ( F 1, 163 = 0.05, P = 0.83 ) , but was significantly lower in individuals captured in exotic shrubland tha n conspecifics in native shrubland in 2013 ( F 1, 177 = 4.51, P = 0.03 ; Figure 2.4B). This difference would translate to 44% reduction on average in hemagglutination activity in exotic shrubs from that in native shrubs (back transforming log - 2 scores) . Time after sunrise had a significantly positive effect on hemagglutination in this comparison group (Table 2.2) . Significant habitat F 1, 117 = 2.01, P = 0.16; 2013: F 1, 218 = 1.66, P = 0.19 ; Figure 2.4B). Ordinal date either had a significantly positive effect or a positive trend on hemagglutination across species and year s (Table 2.2). Plasma haptoglobin levels of gray catbirds did not differ significantly by habitat in 2012 ( F 1, 134 = 1. 83, P = 0.18), but in 2013 we found 0.152 mg/ml or 17% lower haptoglobin levels in individuals using exotic shrubs relative to conspecifics captured in native shrubland ( F 1, 121 = 7.56, P = 0.007 ; Figure 2.4C). No habitat - related differences were found in in 2012 ( F 1, 151 = 0.80, P = 0.37), but a non - significant trend toward lower haptoglobin in exotic shrub s (by 0.054 mg/ml or 10% ) than in native shrubs was found in 2013 ( F 1, 178 = 2.89, P = 0.09 ; Figure 2.4C). We did not find consisten t patterns in significant covariates explaining 61 haptoglobin between species and years (Table 2.2). Plasma Total Carotenoids While we did not find significant differences by habitat in plasma total carotenoids of gray catbirds in 2012 ( F 1, 160 = 0.58, P = 0 .45), in 2013 carotenoid concentrations of individuals in exotic shrubs were 4.55 µ g/ml lower in exotic shrubs had significantly lower carotenoid concentrations than conspecifics in native shrubs ( F 1, 165 = 9.03, P = 0.003 ; Figure 2.5A). No such difference F 1, 177 = 0.99, P = 0.32) or 2013 ( F 1, 211 = 1.74, P = 0.89 ; Figure 2.5A). Fat score had consistent positive effects on plasma total carotenoids (Table 2.2). Plasma Total Antioxidant Capacity We found that anti oxidant capacity of gray catbirds was consistently lower in exotic shrubs than in native counterparts by 6% in both 2012 ( mean difference: 16.0 mmol/l of HOCl neutralized ; F 1, 167 = 8.32, P = 0.004) and 2013 ( mean difference: 14.9 mmol/l of HOCl neutralize d; F 1, 151 = 5.48, P = 0.02 ; Figure 2.5B). But significant covariates differed between years in gray catbirds: in 2012 h andling time, fat score, and wing chord length had significant positive effects on antioxidant capacity, whereas 2 01 3 only ordinal date was positively related to antioxidant F 1, 167 = 0.01, P = 0.91; 2013: F 1, 203 = 0.36, P = 0.55 ; Figure 2.5B). Significant covariates also did not show consistent patterns in Swainson s thrushes between year s (Table 2.2). 62 Dietary Carotenoids and Antioxidant Capacity Our measures of fruit carotenoids showed substantial variation among species ( F 12, 77 = 88.06, P < 0.001 ; Figure 2.6A). Carotenoid concentration wa s highest in exotic fruits of R. multiflora , S. dulcamara , and Lonicera spp., with L. benzoin as the highest ranking native fruit (rank 5) (Figure 2.6A). We also found significant differences in the measure of antioxidant capacity among fruit species ( F 13, 91 = 37.73, P < 0.001 ; Figure 2.6B). The exotic R. cathartica , E. umbellata and the native C. racemosa and L. benzoin had the highest levels of antioxidant capacity (Figure 2.6B). Based on the number of fruits identified in fecal samples (Table 2.3), the exotic R. cathartica was the most frequently consumed fruit in the exotic shrubland, accounting for 21% and 29% of the diet in gray catbirds and Swainson s thrushes respectively. In the native shrubland, the native L. benzoin ranked first and was found in 69% and 64% of fecal samples from gray catbirds and Swainson s thrushes respectively. In contrast, the carotenoid - rich exotic fruits of Rosa multiflora and Lonicera spp. were rarely consumed by either bird species in either habitat . Integrating fruit q uality and fecal data, t he expected mean total carotenoids obtained through diet was significantly lower in gray catbirds using exotic shrubs relative to using native shrubs ( by 10.0 µ g/g wet fruit; t 19.84 = - 2.21, P ( t 109.37 = - 0.07, P = 0.95 ; Figure 2.7A). We did not find significant difference in the expected mean dietary antioxidant capacity in gray catbirds ( t 14.25 = - 1.36, P t 165.37 = 0.36, P = 0.72 ; Figure 2.7B). 63 Discuss ion We tested whether frugivorous landbirds during fall migration experience difference in energetic and immunological condition between individuals using exotic - dominated shrubland compared to conspecifics using native - dominated shrub habitat. Collectivel y our finding s indicate that exotic habitat may be suboptimal for some species of fall migrating landbirds. Despite it being a complex story, our study provides a unique approach that highlights the value of using multiple health parameters as a way of ass essing stopover habitat quality. Effect of H abitat U se on Indices of Refueling Performance Contrary to our expectations based on nutrient properties of exotic and native fruits , we did not find habitat differences in refueling performance in most comparis on groups. Although native fruits generally have higher fat and energy contents than exotic counterparts ( White 1989 ; Smith et al. 2007a ; Smith et al. 2013 ) , birds gained fat mass through the morning only occurred in 2012 and the rate did not differed by habitat. While we observed mass loss in gray catbirds captured i n exotic habitat in 2013 this relationship was not replicated in SWTH, nor did it occur in 2012. Likewise, our plasma metabolite results also do not indicate refueling differences between habitats for either species. Both species regardless of habitat and year showed a general pattern of TRIG increasing and BUTY decreasing during the morning, a pattern consistent to successful foraging (e.g., Jenni - Eiermann a nd Jenni 1994 ; Guglielmo et al. 2005 ; Smith and McWilliams 2010 ) . Although in 2013 gray catbird BUTY decreased at faster rate in exotic shrubland tha n in native shrubland , this metabolite is considered less reliable and informative in inferring habitat differences refueling performance than TRIG ( Acevedo Seaman et al. 2006 ; Smith and McWilliams 2010 ; Evans Ogden et al. 2013 ) , which did not vary by habitat in this 64 comparison group. These results are consistent with previous work in mid - Michigan ( Craves 2009 ) showing that Swainson s thrushes primarily consuming exotic fruits could still gain mass during stopover even aft er the depletion of native fruits. Taken together our findings on the estimated rates of fat mass change and plasma metabolites, we consider that refueling performance of birds using exotic shrub was equivalent to conspecifics foraging in native shrubland . We found minor incongruenc e between refueling performance as inferred from rate of fat mass change and that based on plasma metabolites in 2013 gray catbirds. Fat was lost only in exotic habitat while we failed to detect habitat differences in TRIG. Ther e are two potentially confounding factors that may explain this incongruenc e ; they are (1) time between birds hitting the net and time of extraction and (2) effect of dietary macronutrient composition on TRIG and BUTY. Although we corrected for the effect of handling time on plasma metabolites, we do not know the precise time of birds hitting the net (birds extracted every 45 min). Because plasma metabolite levels, particularly BUTY, are sensitive to delay in blood sampling ( Jenni and Jenni - Eiermann 1992 ; Guglielmo et al. 2002 ; Guglielmo et al. 2005 ; Smith and McWilliams 2010 ) , variable extraction delay likely added noise. Differences in macronutrient composition in exotic (high sugar) and native (high fat) fruits might also explain the lack of differences in ou r TRIG results ( but see Cerasale et al. 2006 ) . In feeding exper iments using controlled diet ( Seaman et al. 2005 ) , TRIG was found to be higher and BUTY lower in birds on low - fat diet than o n high - fat diet, even though mass gain was similar. Similar results were found in experiments using isoenergetic diet that differed in these macronutrient composition ( Smith and McWilliams 2009 ) . Although not as energetically efficient as directly assimilating dietary fat, birds have exceptionally high capacity for de novo synthesis of fatty acids from carbohydrates ( Stevens 65 1996 ; Klasing 1998 ) , and thus a sugar - rich diet (i.e. exotic fruits) may serve just as well as fat - rich diet (i.e. native fruits) for refueling. Therefore, we caution that macronutrient variation between habitats may well confound plasma metabolite data ( Seaman et al. 2005 ; Smith and McWilliams 2009 ) and that sampling should be done as soon as possible post capture in the net ( Guglielmo et al. 2002 ; Guglielmo et al. 2005 ; Smith and McWilliams 2010 ) . Measurements of c onstitutive immunity ( Buehler et al. 2008 ) and antioxidant capacity ( Costantini et al. 2007 ) may provide a more robust approach for dealing with handling stress or sampling delay. Effe ct of H abitat U se on Indices of Constitutive Innate Immune Function Increased G:L or H:L ratio occurs as circulating heterophils increase and lymphocytes decrease in response to elevated glucocorticoid stress hormone levels ( Maxwell 1993 ; Davis et al. 2008 ) . We found that gray catbirds captured in exotic habitat had higher G:L ratios than conspecifics in native shrubs in 2013 with a similar, but no n - significant trend in 201 2 . However, no habitat related differences were observed for total leukocyte counts f or either species or G:L ratios in The habitat - related difference in G:L ratio in 2013 gray catbirds could be interpreted as a sign of stress and/or flight - induced reduction of lymphocytes. Food restriction ( Maxwell et al. 1992 ) , migration ( Owen and Moore 2006 ) , and intense flight ( Matson et al. 2012b ; but see Nebel et al. 2012 ) have all been suggest ed to be important stressors based on G:L or H:L ratios . But elevated G:L ratio could be also due to reduced lymphocytes in circulation that may represent adaptive tradeoff b etween acquired immunity and flight ( Owen and Moore 2006 ; 2008a ) . Because 2013 gray catbirds showed fat loss and higher G:L ratio without significant reduction in total leukocyte counts in exotic shrubs compared to conspecifics in native habitat, the difference in G:L ratio seems more consiste nt with higher stress in exotic 66 shrubland, possibly resulting from suboptimal foraging opportunities. Likewise, gray catbirds using exotic shrubs in 2013 had significantly lower hemagglutination activities and haptoglobin levels with a non - significant tr end toward lower hemolysis activities compared to conspecifics in native shrubs. Habitat effects on any of those measures did not . Similarly to the lymphocyte responses to flight and stopover , recent experimental w ork on European starlings ( Sturnus vulgaris ) showed reduction in measures of hemolysis, hemagglutination, and haptoglobin as a result of prolonged flight; starlings rested and foraged for 2 d post flight exhibited greater hemolysis and trends toward greate r hemagglutination and haptoglobin compared to individuals immediately sampled after flight ( Nebel et al. 2012 ) . E vidence of recovery or upregulation of hemolysis and haptoglobin , contingent on refueling success, has also been found in red knots ( Calidris canutus ) at a stopover site ( Buehler et al. 2010 ) . We therefore consider that habitat - related differences in gray catbird immunity captured in 2013 may have reflect ed differential immune recovery. Migration is thought to be the phas mortality occurs ( Sillett a nd Holmes 2002 ) , and small differences in immune status resulting from varying quality of stopover habitat may have significant implications for future disease resistance and survival ( Klaassen et al. 2012 ) . Low H:L ratio has been linked to greater resistance to bacterial infection ( Al - Murrani et al. 2002 ; Krams et al. 2012 ) and high hemagglutination activity to lower ectoparasite load ( Whiteman et al. 2006 ) ). These or similar measures of constitutive immunity also appear to be positively related to survival in nestlings and adult birds ( Kilgas et al. 2006b ; Parejo and Silva 2009 ; Townsend et al. 2010 ) . Having robust immunity during stopover is likely adaptive in the defense against pathogens i ngested 67 during foraging ( Buehler and Piersma 2008 ; Buehler et al. 2010 ; Altizer et al. 2011 ) , and it is possible that birds leaving stopover with lower immunity may face greater risk of future infection ( Owen and Moore 2008a ) . Constitutive innate immunity may be particularly important in immunologically naïve HY birds experiencing their first fall migration. In addition, baseline haptoglobin levels prior to flight have been positively linked to flight duration ( Nebel et al. 2013 ) . Taken together, it is conceivable that some species using ex otic shrubs in certain years suffer a relative fitness disadvantage compared to conspecifics stopping over in native shrubs. We are uncert ain as to why these habitat differences in immunological condition were found in only one year (2013) in only species (gray catbirds). However, it was only this comparison group that showed fat loss and lower circulating carotenoids (see below) in exotic h abitat. Furthermore, all immune parameters except total leukocyte counts in these gray catbirds showed habitat effect toward the same direction. We thus suspect that the measured immunological parameters are linked to energetic condition and/or carotenoids , and that loss of fat and/or lower carotenoids in 2013 gray catbirds captured in exotic shrubs may have contributed to their poorer immunological condition than in native shrubs. Future studies should further investigate under what nutritional conditions birds may manifest habitat differences in immunity, and whether these immune parameters vary in sync in migrating landbirds during stopover . At the same time, fruit abundance and consumption data by year will be necessary to elucidate inter - annual variatio ns in these links. Effect of Habitat Use on Plasma Total Carotenoids We found significantly lower plasma total carotenoid concentrations in gray catbirds using exotic shrubland in 2013 than those in native shrubland, but no other comparison groups showed this 68 pattern. Carotenoids are immunostimulatory pigments entirely derived from diet ( McGraw and Ardia 2003 ; McGraw et al. 2006 ; McGraw et al. 2011 ; Simons et al. 2012 ) , thus may link habitat - dependent carotenoid acquisition to immune status (see below for detail) . B ecause p lasma carotenoids may fluctuate rapidly due to feeding and resting ( see Hõrak et al. 2004 ) , and sustained high levels of circulation may be required to i mpact immunity (see experimental designs in McGraw and Ardia 2003 ; McGraw et al. 2006 ) , it is difficult to directly link plasma total carotenoids with immunity during the relatively short time scale of migratory stopover. However, the only comparison group that exhibited significant habitat effect in multiple measures of immunity (2013 gray catbirds) also differed in circulating carotenoid levels in the same direction. It is thus possible that these gray catbirds could attain sustained high levels of carotenoids in native shrubs and t hereby could enhance their immune parameters. Effect of Habitat Use on Plasma Total Antioxidant Capacity Antioxidant capacity of gray catbirds using exotic shrubland was consistently lower than conspecifics in native shrubland in 2012 and 2013, althoug h no such relationships were present catbirds suggest that these differences reflect inherent differences in habitat quality for this species. The importance of a ntioxidants in birds subject to oxidative damage resulting from strenuous flight is well - established ( Costantini et al. 2007 ; Costantini et al. 2008 ; Jenni - Eiermann et al. 2014 ) . During stopover body condition was shown to be positively related to antioxidant capacity ( Costantini et al. 2007 ) , which may indicate recovery of antioxidant status during stopover. Recent findings that birds preferentially consume fruits with high antioxidant levels ( Alan et al. 2013 ; Bolser et al. 2013 ) also suggest the need to restore antioxidant status during 69 migratory stopover. Having a sufficient antioxidant barrier may be important for combatting infection, because antioxidants may be required to control reactive oxygen species produced d uring inflammatory response ( Costantini 2008 ; Costantini and Møller 2009 ) . Overall, the ability to successfully control oxidative damage likel y has significant fitness implications. In a long - term study in barn swallows ( Hirundo rustica ), antioxidant capacity was found to be positively related to return rate, with an increase of 1 mmol/l of HOCl neutralized corresponding to 0.6% increase in surv ival ( Saino et al. 2011 ) . While their findings are not directly applicable to our study, the magnitude of habitat effect observed in gray catbirds (15 mmol/l of HOCl neutralized) suggests a relativ e fitness disadvantage of using exotic shrubland. Dietary Carotenoids and Antioxidant Capacity We found that both total carotenoids and antioxidant capacity of exotic and native fruits were highly variable among species. Unlike the pattern of fa t and energy contents reported on exotic and native fruits ( Smit h et al. 2013 ) , we did not find a clear pattern that suggests native fruits are greater in measures of carotenoids and antioxidant capacity. Instead, certain exotic fruits (e.g., R. multiflora ) had exceptionally high carotenoid contents, similarly to pre viously described ( Alan et al. 2013 ) . However, the fecal data show that birds rarely consumed the carotenoid - rich exotic fruits of R. multiflor a , S. dulcamara , or Lonicera spp. (see also Suthers e t al. 2000 ; Alan et al. 2013 ; Bolser et al. 2013 ; Smith et al. 2013 ) . A bundant fruits of the exotic E. umbel lata with moderately high carotenoid contents were also rare in fecal samples (see also Smith et al. 2007a ; Smith et al. 2013 ) . These exotic fruits, therefore, likely contributed minimally to carotenoid acquisition. The fruits frequently consumed in our study were the native L. benzoin and the exotic R. cathartica 70 within their respective shrublands . Fruits from these two species had very high antioxidant s and moderate levels of carotenoids and were likely major contribut ors of dietary acquisition o f these nutrients. In addition, L. benzoin fruits have very high fat and energy contents ( White 1989 ; Smith et al. 2013 ) and was the single most consumed fruit acros s habitat types . As b irds select fruits based on multiple factors including fat, energy, carotenoids, and antioxidant contents ( Smith et al. 2007a ; Alan et al. 2013 ; Bolser et al. 2013 ; Pierce and McWilliams 2014 ) , L. benzoin may provide the best combination of nutrients c ompared to other fruits in our study site, and conserving this shrub appears importan t for fall frugivorous migrants. I ntegrating fruit consumption and nutrient data of carotenoids and antioxidants , w e found limited evidence that differential fruit consumption may have been reflected in the immune and antioxidant status for either species. Given the fecal data we would predict that g ray catbirds using exotic shrubland could have lower carotenoid levels in their average fruit diet (per unit mass) tha W e were not able to detect significant habitat differences in the expected dietary means of antioxidant capacity in either species ; small fecal sample size (only 14 samples from exotic shr ubland) along with relatively large intra - and inter - species variation in fruit antioxidant capacity likely hampered detection. While we present these data with caution due to limitations of fecal sample size and inherent difficulties of dietary analyses based on fecal sampling ( Blake and Loiselle 1992 ; Parrish 1997 ) , we consider it is plausible that differential fruit consumption be tween habitats impacted health status in gray catbirds. S imultaneous intake of fat has been suggested to improve absorption of carotenoids and other antioxidants ( P ietta 2000 ; van het Hof et al. 2000 ; Surai et al. 2001 ) , and the high fat content of native fruits ( White 1989 ; Smith et al. 2007a ; Smith et al. 71 2013 ) could enhance antioxidant absorption. This effect could lead to a greater habitat difference in the actual acquisition than what is predictable through our calculations. We encourage future studies to take a more rigorous approach in linking dietary carotenoids and antioxidant capacity to these measures in circ ulation in birds. Immune and Antioxidant Measures as Indicators of Habitat Quality Applications of ecophysiology can provide information on individual performance related to fitness and hence have great potentials as indicators of habitat quality ( Albano 2012 ) . However, the utility of immune and antioxidant indices for assessing habitat quality depends on whether the observed habitat - related differences can be interpreted as consequences of differential habitat use (see Johnson 2007 ; Albano 2012 ) . Based on our results, we do argu e that habitat differences in health status may well represent consequences of differential use. Habitat - related differences in populations or pathogen exposure were unlikely due to the proximity of the two shrublands within the study area. However, altern ative explanations should be considered. One such explanation is that individuals with superior health status were more likely to use native shrubland. For example, spatial segregation by health is conceivable if there was greater predation risk in native shrubland and healthier birds more willing to forage in that risker habitat . However, this explanation is unlikely because vegetation structure (i.e. cover) was similar between exotic and native shrublands (Y. Oguchi and J. Owen, unpublished data). The sec ond explanation is that individuals sampled in exotic shrubland simply represented more recent arrivals , and that on average, bird s captured in native shrubland had more time to recover or upregulate their health status. Consistent with this explanation, o ur capture - recapture record s how s that birds initially captured in exotic shrubland were more likely to switch to native 72 shrubland than vice versa (Y. Oguchi and J. Owen, unpublished data). H owever, we believe that such variations in stay length was at lea st partially controlled for by inclusions of condition indices (fat and muscle scores) in ANCOVA models because stay length is generally linked to both condition and health indices ( Owen and Moore 2008a ; Buehler et al. 2010 ) . I n addition, sever al indices of immunity and antioxidant capacity were related to time after sunrise in at least some comparison groups (see Table 2.2), indicating these measures were sensitive to short - term change and reflected current health status ( see also van de Crommenacker et al. 2010 ) . Results from thi s study provide the first evidence that differential use of stopover habitat may impact the health status of migrating landbirds, and conversely, that stopover habitat quality could be inferred from measures of health in birds. In light of the recent revie ws that highlight the need for advancing knowledge on factors influencing migratory bird health ( Albano 2012 ; Klaassen et al. 2012 ) , we encourage future studies comparing habitat quality through measures of individual performance to include indices of immunity and antioxidant status. In addition, inclusions of immune and antioxidant measures may reveal differences in habitat quality that may not be reflected in plasma metabolites, such as habitat variation in nutrient acquisition. We conclude that exotic shrubs in our study site may represent suboptimal stopover habitat compared to native counterparts for some , but not all, species of fall frugivorous migrants , and that the degree of habitat quality difference likely differ between years. We recommend future studies to test these same hypotheses through controlled laboratory experiments, which will help isolate and confirm the link among nutritional attributes of exotic and native fruits , refueling performance, immunity , and antioxi dant capacity of individual birds. 73 Acknowledgements We thank USFWS Avian Health and Disease Program (USFWS Migratory Bird Conservation in the Upper Midwest) for funding to J.C.O. and R.J.S. Funds were also provided from Hal and Jean Glassen Memorial F oundation to Y.O. through Michigan State University. We thank the Michigan Department of Natural Resources for allowing us to use the Rose Lake State Wildlife Research Area. Thanks to many assistants and volunteers who provided field and laboratory support , particularly Z. Pohlen, C. Gesmundo, C. Gawne, L.Theile, K. Clark, K. Johnson, D. Boston, K. Wilson, H. Matsumura, M. Porter, J. Hitchcock, and N. Martineau. We thank the following individuals for providing valuable advice on assays or data interpretatio n: K. Paxton, C. Guglielmo K. McGraw, D. Costantini, K. Matson, L. Martin, P. Hõrak, T. Killpack, A. Flies, C. Downs, R. Ives, S. McWilliams, M. Quigley, and D. Williams. 74 APPENDIX 75 Table 2.1. Summary of health parameters (means ± SE) in gray catbirds an - or native - dominated shrubland near East Lansing, MI, USA in 2012 and 2013. Sample sizes are listed in parentheses. TRIG = - hydroxybutyrate (mmol/l), WBC = total leukocytes/10,000 ery throcytes, G:L = granulocyte:lymphocyte ratio, HL = hemolysis score, HA = hemagglutination score, Hp = haptoglobin (mg/ml), Car = total carotenoids (µg/ml), OXY = antioxidant capacity (mmol/l of HOCl neutralized). Para - Gray catbird, 2012 Gray catbird, 2 013 meter Exotic Native Exotic Native Exotic Native Exotic Native TRIG .67 ± .04 (44) .77 ± .03 (134) .62 ± .07 (43) .57 ± .03 (142) 1.29 ± .06 (61) 1.46 ± .17 (103) 1.02 ± .06 (103) 1.18 ± .09 (119) BUTY 9.36 ± .78 (44) 10.10 ± .41 (134) 6.56 ± .66 (43) 8.48 ± .37 (142) 6.87 ± .63 (61) 7.98 ± .49 (103) 6.43 ± .47 (103) 7.02 ± .41 (119) WBC 61.00 ± 7.27 (25) 57.74 ± 3.48 (116) 78.93 ± 12.35 (14) 81.80 ± 5.84 (61) 42.21 ± 3.51 (33) 48.65 ± 4.03 (54) 43.43 ± 5.30 (44) 43.35 ± 3.21 (51) G:L .37 ± .07 (25) .26 ± .02 (116) .50 ± .19 (14) .24 ± .04 (61) .50 ± .13 (33) .33 ± .07 (54) .25 ± .04 (44) .26 ± .03 (51) HL 5.53 ± .23 (36) 5.29 ± .13 (129) 4.33 ± .27 (40) 4.78 ± .11 (141) 4.12 ± .15 (70) 3.84 ± .13 (11 0) 3.89 ± .08 (104) 4.06 ± .10 (116) HA 7.07 ± .36 (36) 7.17 ± .21 (129) 6.45 ± .38 (40) 7.17 ± .18 (141) 5.21 ± .24 (70) 4.70 ± .14 (110) 4.92 ± .05 (104) 5.33 ± .17 (116) Hp .78 ± .04 (25) .83 ± .02 (113) .73 ± .06 (24) .88 ± .02 (99) .39 ± .02 (61) .4 2 ± .02 (94) .50 ± .02 (86) .54 ± .02 (95) Car 20.24 ± 1.83 (37) 22.89 ± .96 (126) 16.93 ± 1.84 (36) 19.87 ± .83 (133) 34.28 ± 1.82 (72) 40.22 ± 1.86 (109) 36.99 ± 1.82 (100) 43.29 ± 1.98 (116) OXY 239.7 ± 5.0 (40) 255.3 ± 2.9 (132) 226.6 ± 7.1 (34) 240. 4 ± 3.3 (120) 228.8 ± 4.1 (70) 226.9 ± 3.2 (100) 221.2 ± 2.6 (95) 225.3 ± 2.8 (111) 76 Table 2.2. Summary of variables retained in multiple linear regressions with backward selection ( P < 0.1). These variables were then included as covariates in ANCOVA te - hydroxybutyrate , WBC = total leukocytes, G:L = granulocyte:lymphocyte ratio, HL = hemolysis score, HA = hemagglutination score, Hp = haptoglobin, Car = total carotenoids, OXY = antioxidant capacity, D = ordinal date, T = time after sunrise, H = handling time, F = fat score, M = muscle score, W = wing chord, A = age, and N = no covariates. Plus or minus sign denote the direction of eff ects. Variables with parentheses indicate thos e dropped ( P > 0.05) in the final, minimal ANCOVA model that included habitat. Parameter Gray catbird, 2012 Gray catbird, 2013 TRIG +D, +T, - H, +F, (+A) +D, +T, +F +T, - H, +F +T, ( - H), +F BUTY - T, +H - T, + H, - F - T, +H, ( - A) - D, - T, +H, - F WBC - D, - H N (+T), ( - H) - H G:L (+A) (+D), - M +D, +H - W HL +D, (+A) (+D), ( - A) +D, ( - H), - F, ( - A) (+H) HA (+D) (+D), +T, ( - H), - A +D, ( - H), ( - A) (+D), +W Hp - D, - T, (+M) N - D, +H - H Car +F +D, +F +D, +F +D, +F, (+M), (+W), - A OXY +H, +F, (+M), +W +D, ( - A) +T, (+H) (+D), +F 77 - and native - dominated shrublands with fruit species identified. U.S. Status indicates the origin of fruit species as either exotic or native, while Exotic or Native under bird species indicates shrubland habitat the samples were collected from. Counts of fecal samples were pooled from 2012 and 2013 captures and were interpreted to represent fruit cons umption frequencies of individuals in their respective habitat. U.S. Gray catbird Scientific name Common name Status Exotic Native Exotic Native Lindera benzoin Common spicebush Native 3 63 22 86 Rhamnus cathartica European buckt horn Exotic 5 15 28 18 Solanum dulcamara Bittersweet nightshade Exotic 0 3 7 6 Ilex verticillata Common winterberry Native 1 3 4 8 Viburnum opulus European cranberrybush Exotic 1 3 7 2 Elaeagnus umbellata Autumn olive Exotic 1 3 5 2 Sambucus canadensi s Common elderberry Native 1 1 3 5 Parthenocissus quinquefolia Virginia creeper Native 2 0 0 5 Rosa multiflora Multiflora rose Exotic 0 0 1 0 Lonicera spp. Honeysuckle Exotic 0 0 0 1 Total 14 91 77 133 78 Figure 2.1. Size - corrected fat mass es timates of gray catbirds captured near East Lansing, MI, USA during fall migration of 2012 and 2013 in relation to time after sunrise and exotic - or native - dominated shrubland habitat. Regression lines through raw data uncorrected for ordinal date are show n. Controlling for ordinal date, significant time after sunrise × habitat interaction was found ( P = 0.02), indicating no fat change over morning feeding hours in native shrubland ( P = 0.23) and fat loss of 0.60 g/hr in exotic shrubland ( P = 0.05). 79 - exotic - or native - dominated shrubland near East Lansing, MI, USA during fall migration of 2012 and 2013. Least square means were obtained by controlling for ordinal date, time after sunrise, handling time, fat score, muscle score, wing chord, and age as necessary (see Table 2.2). Significant differences or trends are denoted with P - value while NS de scribes non - significance. 80 Figure 2.3. Least square means (± SE) of (A) total leukocyte count per 10,000 erythrocytes in exotic - or native - dominated shrubland nea r East Lansing, MI, USA during fall migration of 2012 and 2013. Least square means were obtained by controlling for ordinal date, time after sunrise, handling time, fat score, muscle score, wing chord, and age as necessary (see Table 2.2). Significant diff erences or trends are denoted with P - value while NS describes non - significance. 81 Figure 2.4. Least square means (± SE) of (A) hemolysis score, (B) hemagglutination score, ling in exotic - or native - dominated shrubland near East Lansing, MI, USA during fall migration of 2012 and 2013. Least square means were obtained by controlling for ordinal date, time after sunrise, handling time, fat score, muscle score, wing chord, and a ge as necessary (see Table 2.2). Significant differences or trends are denoted with P - value while NS describes non - significance. 82 Figure 2.5. Least square means (± SE) of (A) plasma total carotenoids and (B) antioxidant capacity in gray catbirds and S - or native - dominated shrubland near East Lansing, MI, USA during fall migration of 2012 and 2013. Least square means were obtained by controlling for ordinal date, time after sunrise, handling time, fat score, muscle score, wing chord, and age as necessary (see Table 2.2). Significant differences or trends are denoted with P - value while NS describes non - significance. 83 Figure 2.6. Total carotenoids (A) and antioxidant capacity (B) of fruits found in shrublands used by migrating landbirds near East Lansing, MI, USA. Fruit species are order ed from highest to lowest means with centerline of boxes showing median values. Exotic fruits are shown as white boxes and native fruits are barred. Differences in letters above box es represent significant differences ( P < 0.05), and numbers below the letters denote sample size. 84 Figure 2.7. Expected dietary means (± SE) of (A) total carotenoids and (B) antioxidant exotic - or native - dominated shrubland in East Lansing, MI, USA. Means were calculated as values of assayed fruits (see Figure 2.6) weighted by consumption proportion (see Table 2.3). Significant differences are denoted with P - value while NS describes non - s ignificance. 85 LITERATURE CITED 86 LITERATURE CITED Acevedo Seaman D.A., C.G. Guglielmo, R.W. Elner, and T.D. Williams. 2006. Landscape - scale physiology: site differences in refueling rates indicated by plasma metabolite analysis in free - living migratory sandpipers. Auk 123:563 574. Al - Murrani W.K., I.K. Al - Rawi, and N.M. Raof. 2002. Genetic resistance to Salmonella typhimurium in two lines of chickens selected as resistant and sensitive on the basis of heterophil/lymphocyte ratio. Br Poult Sci 43:501 507. Alan R.R., S.R. McWilliams, and K.J. McGraw. 2013. The importance of antioxidants for avian fruit selection during autumn migration. Wilson J Ornithol 125:513 525. Albano N. 2012. Conservation physiology tools: their role in assessin g habitat quality in birds. Ardeola 59:197 216. Altizer S., R. Bartel, and B.A. Han. 2011. Animal migration and infectious disease risk. Science 331:296 302. Bairlein F., L. Jenni, A. Kaiser, L. Karlsson, A. van Noordwijk, W. Peach, A. Pilastro, F. Spina, and G. Walinder. 1995. European - African Songbird Migration Network: Manual of Field Methods. Institut für Vogelforschung/Vogelwarte Helgoland, Wilhelmshaven, Germany. Blake J.G. and B.A. Loiselle. 1992. Fruits in the diets of Neotropical migrant birds in C osta Rica. Biotropica 24:200 210. Bolser J.A., R.R. Alan, A.D. Smith, L. Li, N.P. Seeram, and S.R. McWilliams. 2013. Birds select fruits with more anthocyanins and phenolic compounds during autumn migration. Wilson J Ornithol 125:97 108. Brinc D., M.K. Cha n, A.A. Venner, M.D. Pasic, D. Colantonio, L. Kyriakopolou, and K. Adeli. 2012. Long - term stability of biochemical markers in pediatric serum specimens stored 826. Buehler D.M., N. Bhola, D. Barjaktarov, W. Goymann, I. Schwabl, B.I. Tieleman, and T. Piersma. 2008. Constitutive immune function responds more slowly to handling stress than corticosterone in a shorebird. Physiol Biochem Zool 81:673 681. Buehler D.M., F. Encinas - Viso, M. Petit, F. Vézina, B.I. Ti eleman, and T. Piersma. 2009. Limited access to food and physiological trade - offs in a long - distance migrant shorebird. II. Constitutive immune function and the acute - phase response. Physiol Biochem Zool 82:561 571. Buehler D.M. and T. Piersma. 2008. Trave lling on a budget: predictions and ecological evidence 87 for bottlenecks in the annual cycle of long - distance migrants. Philos Trans R Soc B 363:247 266. Buehler D.M., B.I. Tieleman, and T. Piersma. 2010. Indices of immune function are lower in Red Knots ( Ca lidris canutus ) recovering protein than in those storing fat during stopover in Delaware Bay. Auk 127:394 401. Campbell T.W. and F.J. Dein. 1984. Avian hematology. The basics. Vet Clin N Am: Small Anim Pract 14:223 248. Catling P.M. 2005. Effects of invasi ve alien plants on birds: some examples from North America. Biodiversity 6:30 39. Cerasale D.J. and C.G. Guglielmo. 2010. An integrative assessment of the effects of tamarisk on stopover ecology of a long - distance migrant along the San Pedro River, Arizona . Auk 127:636 646. Cerasale D.J., C.G. Guglielmo, and K. Hobson. 2006. Dietary effects on prediction of body mass changes in birds by plasma metabolites. Auk 123:836 846. Comstock G.W., A.E. Burke, E.P. Norkus, G.B. Gordon, S.C. Hoffman, and K.J. Helzlsoue r. 2001. Effects of repeated freeze - thaw cycles on concentrations of cholesterol, micronutrients, and hormones in human plasma and serum. Clin Chem 47:139 142. Cooper C.E., N.B.J. Vollaard, T. Choueiri, and M.T. Wilson. 2002. Exercise, free radicals and ox idative stress. Biochem Soc Trans 30:280 284. Costantini D. 2008. Oxidative stress in ecology and evolution: lessons from avian studies. Ecol Lett 11:1238 1251. --- . 2010. Complex trade - offs in the pigeon ( Columba livia ): egg antioxidant capacity and femal e serum oxidative status in relation to diet quality. J Comp Physiol B 180:731 739. Costantini D., M. Cardinale, and C. Carere. 2007. Oxidative damage and anti - oxidant capacity in two migratory bird species at a stop - over site. Comp Biochem Physiol C 144:3 63 371. Costantini D., S. Casagrande, S. De Filippis, G. Brambilla, A. Fanfani, J. Tagliavini, and G. Falco tinnunculus ). J Comp Physiol B 176:329 337. Costantini D., G. Dell'Ariccia , and H. - P. Lipp. 2008. Long flights and age affect oxidative status of homing pigeons ( Columba livia ). J Exp Biol 211:377 381. Costantini D. and A.P. Møller. 2008. Carotenoids are minor antioxidants for birds. Funct Ecol 22:367 370. 88 --- . 2009. Does immune response cause oxidative stress in birds? A meta - analysis. Comp Biochem Physiol A 153:339 344. Craves J.A. 2009. A fifteen - year study of fall stopover patterns of Catharus thrushes at an inland, urban site. Wilson J Ornithol 121:112 118. Davis A.K. 2005. Effect of handling time and repeated sampling on avian white blood cell counts. J Field Ornithol 76:334 338. Davis A.K., D.L. Maney, and J.C. Maerz. 2008. The use of leukocyte profiles to measure stress in vertebrates: a review for ecologists. Funct Ecol 2 2:760 772. Deerenberg C., H. Biebach, and U. Bauchinger. 2002. Spleen size variation during long - distance migration in the garden warbler Sylvia borin . Avian Sci 2:217 226. Delers F., G. Strecker, and R. Engler. 1988. Glycosylation of chicken haptoglobin: isolation and characterization of three molecular variants and studies of their distribution in hen plasma before and after turpentine - induced inflammation. Biochem Cell Biol 66:208 217. Dunn E.H. 2000. Temporal and spatial patterns in daily mass gain of M agnolia Warblers during migratory stopover. Auk 117:12 21. Evans Ogden L.J., K. Martin, and T.D. Williams. 2013. Elevational differences in estimated fattening rates suggest that high - elevation sites are high - quality habitats for fall migrants. Auk 130:98 106. Fänge R. and B. Silverin. 1985. Variation of lymphoid activity in the spleen of a migratory bird, the pied flycatcher ( Ficedula hypoleuca ; Aves, Passeriformes). J Morphol 184:33 40. Guglielmo C.G., D.J. Cerasale, and C. Eldermire. 2005. A field valida tion of plasma metabolite profiling to assess refueling performance of migratory birds. Physiol Biochem Zool 78:116 125. Guglielmo C.G., P.D. O'Hara, T.D. Williams, and C. Blem. 2002. Extrinsic and intrinsic sources of variation in plasma lipid metabolites of free - living western sandpipers ( Calidris mauri ). Auk 119:437 445. Gutteridge J.M.C. 1987. The antioxidant activity of haptoglobin towards haemoglobin - stimulated lipid peroxidation. Biochim Biophys Acta 917:219 223. Hadfield J. 2014. MCMCglmm: MCMC Generalised Linear Mixed Models. A vignette for R package MCMCglmm. Available from http://cran.r - project.org/web/packages/ MCMCglmm . Hasselquist D., Å. Lindström, S. Jenni - Eiermann, A. Koolhaas, and T. Piersma. 2007. Long flights do not influence immune responses of a long - distance migrant bird: a wind - tunnel 89 experi ment. J Exp Biol 210:1123 1131. Hegemann A., K.D. Matson, M.A. Versteegh, and B.I. Tieleman. 2012. Wild skylarks seasonally modulate energy budgets but maintain energetically costly inflammatory immune responses throughout the annual cycle. PLoS ONE 7:e363 58. Helms C.W. and W.H. Drury. 1960. Winter and migratory weight and fat field studies on some North American buntings. Bird - Banding:1 40. Hõrak P., P.F. Surai, I. Ots, and A.P. Møller. 2004. Fat soluble antioxidants in brood - rearing great tits Parus major : relations to health and appearance. J Avian Biol 35:63 70. Hõrak P., M. Zilmer, L. Saks, I. Ots, U. Karu, and K. Zilmer. 2006. Antioxidant protection, carotenoids and the costs of immune challenge in greenfinches. J Exp Biol 209:4329 4338. Jenni - Eiermann S. and L. Jenni. 1994. Plasma metabolite levels predict individual body - mass changes in a small long - distance migrant, the Garden Warbler. Auk:888 899. --- . 1996. Metabolic differences between the postbreeding, moulting and migratory periods in feeding an d fasting passerine birds. Funct Ecol:62 72. --- . 1998. What can plasma metabolites tell us about the metabolism, physiological state and condition of individual birds? An overview. Biol Conserv Fauna 102:312 319. Jenni - Eiermann S., L. Jenni, S. Smith, and D. Costantini. 2014. Oxidative stress in endurance flight: an unconsidered factor in bird migration. PLoS ONE 9:e97650. Jenni L. and S. Jenni - Eiermann. 1992. Metabolic patterns of feeding, overnight fasted and flying night migrants during autumn migration . Ornis Scand 23:251 259. Johnson M.D. 2007. Measuring habitat quality: a review. Condor 109:489 504. Kilgas P., R. Mänd, M. Mägi, and V. Tilgar. 2006a. Hematological parameters in brood - rearing great tits in relation to habitat, multiple breeding and sex. Comp Biochem Physiol A 144:224 231. Kilgas P., V. Tilgar, and R. Mänd. 2006b. Hematological health state indices predict local survival in a small passerine bird, the great tit ( Parus major ). Physiol Biochem Zool 79:565 572. Klaassen M., B.J. Hoye, B.A. N olet, and W.A. Buttemer. 2012. Ecophysiology of avian migration in the face of current global hazards. Philos Trans R Soc B 367:1719 1732. Klasing K.C. 1998. Comparative Avian Nutrition. CABI, NY, USA. 90 Krams I., J. Vrublevska, D. Cirule, I. Kivleniece, T. Krama, M.J. Rantala, E. Sild, and P. Hõrak. 2012. Heterophil/lymphocyte ratios predict the magnitude of humoral immune response to a novel antigen in great tits (Parus major). Comp Biochem Physiol A 161:422 428. Lenth R.V. and M. Hervé. 2015. lsmeans: Leas t - Squares Means. A vignette for R package lsmeans. Available from http://cran.r - project.org/web/packages/lsmeans/ . Liu M. and D.L. Swanson. 2014. Physiological evidence that anthropogenic wood lots can substitute for native riparian woodlands as stopover habitat for migrant birds. Physiol Biochem Zool 87:183 195. Matés J.M. and F. Sánchez - Jiménez. 1999. Antioxidant enzymes and their implications in pathophysiologic processes. Front Biosci 4:d339 d345. Matson K.D., N.P. Horrocks, M.A. Versteegh, and B.I. Tieleman. 2012a. Baseline haptoglobin concentrations are repeatable and predictive of certain aspects of a subsequent experimentally - induced inflammatory response. Comp Biochem Physiol A 162:7 15. Matson K.D., N.P.C. Horrocks, B.I. Tieleman, and E. Haase. 2012b. Intense flight and endotoxin injection elicit similar effects on leukocyte distributions but dissimilar effects on plasma - based immunological indices in pigeons. J Exp Biol 215:3734 3741. M atson K.D., R.E. Ricklefs, and K.C. Klasing. 2005. A hemolysis hemagglutination assay for characterizing constitutive innate humoral immunity in wild and domestic birds. Dev Comp Immunol 29:275 286. Maxwell M.H. 1993. Avian blood leucocyte responses to str ess. World Poultry Sci J 49:34 43. Maxwell M.H., P.M. Hocking, and G.W. Robertson. 1992. Differential leucocyte responses to various degrees of food restriction in broilers, turkeys and ducks. Br Poult Sci 33:177 187. McGraw K.J. and D.R. Ardia. 2003. Caro tenoids, immunocompetence, and the information content of sexual colors: an experimental test. Am Nat 162:704 712. McGraw K.J., O.L. Crino, W. Medina - Jerez, and P.M. Nolan. 2006. Effect of dietary carotenoid supplementation on food intake and immune functi on in a songbird with no carotenoid coloration. Ethology 112:1209 1216. McGraw K.J., P.M. Nolan, and O.L. Crino. 2011. Carotenoids bolster immunity during moult in a wild songbird with sexually selected plumage coloration. Biol J Linn Soc 102:560 572. McGr aw K.J., E.A. Tourville, and M.W. Butler. 2008. A quantitative comparison of the commonly used methods for extracting carotenoids from avian plasma. Behav Ecol Sociobiol 62:1991 2002. 91 Moore F.R., S.A. Gauthreaux Jr, P. Kerlinger, T.R. Simons, T.E. Martin, and D.M. Finch. 1995. Habitat requirements during migration: important link in conservation. Pp. 121 144 in T.E. Martin and D.M. Finch, eds. Ecology and Management of Neotropical Migratory Birds: A Synthesis and Review of Critical Issues. Oxford University Press, NY, USA. Moore F.R., R.J. Smith, and R. Sandburg. 2005. Stopover ecology of intercontinental migrants: En route problems and consequences for reproductive performance. Pp. 251 261 in R. Greenberg and P.P. Marra, eds. Birds of Two Worlds: The Ecolog y and Evolution of Migration. Johns Hopkins University Press, Baltimore, MD, USA. Nebel S., U. Bauchinger, D.M. Buehler, L.A. Langlois, M. Boyles, A.R. Gerson, E.R. Price, S.R. McWilliams, and C.G. Guglielmo. 2012. Constitutive immune function in European starlings, Sturnus vulgaris , is decreased immediately after an endurance flight in a wind tunnel. J Exp Biol 215:272 278. Nebel S., D.M. Buehler, A. MacMillan, and C.G. Guglielmo. 2013. Flight performance of western sandpipers, Calidris mauri, remains unco mpromised when mounting an acute phase immune response. J Exp Biol 216:2752 2759. - carotene levels measured with an improved method of high - performance liquid chromatography. J Chromatogr B 339:273 284. Ninni P., F. De Lope, N. Saino, C. Haussy, and A.P. Møller. 2004. Antioxidants and condition - dependence of arrival date in a migratory passerine. Oikos 105:55 64. Owen J.C. 2011. Collecting, processing, and storing avian blood: a review. J Field Ornithol 82:339 354. Ow en J.C., E.A. Cornelius, D.A. Arsnoe, and M.C. Garvin. 2013. Leukocyte response to eastern equine encephalomyelitis virus in a wild passerine bird. Avian Dis 57:744 749. Owen J.C. and F.R. Moore. 2006. Seasonal differences in immunological condition of thr ee species of thrushes. Condor 108:389 398. --- . 2008a. Relationship between energetic condition and indicators of immune function in thrushes during spring migration. Can J Zool 86:638 647. --- reduced immune function. J Ethol 26:383 388. Parejo D. and N. Silva. 2009. Immunity and fitness in a wild population of Eurasian kestrels Falco tinnunculus . Naturwissenschaften 96:1193 1202. Parrish J.D. 1997. Patterns of frugivory and energetic condition in Nearctic landbirds during autumn migration. Condor:681 697. 92 Pasek J., A. Tahk, G. Culter, and M. Schwemmle. 2014. weights: Weighting and Weighted Statistics. A vignette for R package weights. Available from https://cran.r - project.org/web/packages/weight s/. Pierce B.J. and S.R. McWilliams. 2014. The fat of the matter: how dietary fatty acids can affect exercise performance. Integr Comp Biol 54:903 912. Piersma T. 1997. Do global patterns of habitat use and migration strategies co - evolve with relative inve stments in immunocompetence due to spatial variation in parasite pressure? Oikos 80:623 631. Pietta P. - G. 2000. Flavonoids as antioxidants. J Nat Prod 63:1035 1042. Pyle P. 1997. Identification Guide to North American Birds, part 1: Columbidae to Ploceidae . Slate Creek Press, Bolinas, CA, USA. R Development Core Team. 2014. R: a language and environment for statistical computing. R Foundation for Statistical Computing, Vienna. http://www.R - project.org/ . Rodriguez - Am aya D.B. 2001. A Guide to Carotenoid Analysis in Foods. ILSI Press, Washington DC, USA. Saino N., M. Caprioli, M. Romano, G. Boncoraglio, D. Rubolini, R. Ambrosini, A. Bonisoli - Alquati, and A. Romano. 2011. Antioxidant defenses predict long - term survival i n a passerine bird. PLoS ONE 6:e19593. Schmid - Hempel P. and D. Ebert. 2003. On the evolutionary ecology of specific immune defence. Trends Ecol Evol 18:27 32. Seaman D.A., C.G. Guglielmo, and T.D. Williams. 2005. Effects of physiological state, mass change and diet on plasma metabolite profiles in the western sandpiper Calidris mauri . J Exp Biol 208:761 769. Sepp T., E. Sild, and P. Hõrak. 2010. Hematological condition indexes in greenfinches: effects of captivity and diurnal variation. Physiol Biochem Zool 83:276 282. Sillett T.S. and R.T. Holmes. 2002. Variation in survivorship of a migratory songbird throughout its annual cycle. J Anim Ecol 71:296 308. Simons M.J., A.A. Cohen, and S. Verhulst. 2012. What does carotenoid - dependent coloration tell? Plasma c arotenoid level signals immunocompetence and oxidative stress state in birds a meta - analysis. PLoS ONE 7:e43088. Smith R.J. and F.R. Moore. 2003. Arrival fat and reproductive performance in a long - distance passerine migrant. Oecologia 134:325 331. 93 Smith S. B., S.A. DeSando, and T. Pagano. 2013. The value of native and invasive fruit - bearing shrubs for migrating songbirds. Northeast Nat 20:171 184. Smith S.B., K.H. McPherson, J.M. Backer, B.J. Pierce, D.W. Podlesak, and S.R. McWilliams. 2007a. Fruit quality a nd consumption by songbirds during autumn migration. Wilson J Ornithol 119:419 428. Smith S.B. and S.R. McWilliams. 2009. Dietary macronutrients affect lipid metabolites and body composition of a migratory passerine, the white - throated sparrow ( Zonotrichia albicollis ). Physiol Biochem Zool 82:258 269. --- . 2010. Patterns of fuel use and storage in migrating passerines in relation to fruit resources at autumn stopover sites. Auk 127:108 118. Smith S.B., S.R. McWilliams, and C.G. Guglielmo. 2007b. Effect of d iet composition on plasma metabolite profiles in a migratory songbird. Condor 109:48 58. Stevens L. 1996. Avian Biochemistry and Molecular Biology. Cambridge University Press, Cambridge, UK. Stiles E. 1993. The influence of pulp lipids on fruit preference by birds. Vegetatio 107:227 235. Surai P., B. Speake, and N. Sparks. 2001. Carotenoids in avian nutrition and embryonic development. 1. Absorption, availability and levels in plasma and egg yolk. J Poult Sci 38:1 27. Suthers H.B., J.M. Bickal, and P.G. Rod ewald. 2000. Use of successional habitat and fruit resources by songbirds during autumn migration in central New Jersey. Wilson Bull 112:249 260. Tella J.L., J.J. Negro, M.G. Forero, F. Hiraldo, M.C. Blazquez, R. Rodríguez - Estrella, and G. Blanco. 1998. A comparison of spectrophotometry and color charts for evaluating total plasma carotenoids in wild birds. Physiol Biochem Zool 71:708 711. Townsend A.K., A.B. Clark, K.J. McGowan, A.D. Miller, and E.L. Buckles. 2010. Condition, innate immunity and disease mo rtality of inbred crows. Proc R Soc B 282:2875 2883. van de Crommenacker J., N.P.C. Horrocks, M.A. Versteegh, J. Komdeur, B.I. Tieleman, and K.D. Matson. 2010. Effects of immune supplementation and immune challenge on oxidative status and physiology in a m odel bird: implications for ecologists. J Exp Biol 213:3527 3535. van het Hof K.H., C.E. West, J.A. Weststrate, and J.G. Hautvast. 2000. Dietary factors that affect the bioavailability of carotenoids. J Nutr 130:503 506. White D.W. 1989. North American bir d - dispersed fruit: ecological and adaptive significance of 94 nutritional and structural traits. PhD diss. Rutgers University, New Brunswick, NJ. Whiteman N.K., K.D. Matson, J.L. Bollmer, and P.G. Parker. 2006. Disease ecology in the Galapagos Hawk ( Buteo gal apagoensis ): host genetic diversity, parasite load and natural antibodies. Proc R Soc B 273:797 804. Woodrey M.S. and F.R. Moore. 1997. Age - related differences in the stopover of fall landbird migrants on the coast of Alabama. Auk 114:695 707. Zhang Y. 201 3. cplm: Compound Poisson Linear Models. A vignette for R package cplm. Available from http://cran.r - project.org/web/packages/cplm .