Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Community-wide impacts of a generalist brood parasite, the brown-headed cowbird (molothrus ater) De Groot, Krista Leigh 1998

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

Item Metadata

Download

Media
831-ubc_1999-0027.pdf [ 2.34MB ]
Metadata
JSON: 831-1.0088888.json
JSON-LD: 831-1.0088888-ld.json
RDF/XML (Pretty): 831-1.0088888-rdf.xml
RDF/JSON: 831-1.0088888-rdf.json
Turtle: 831-1.0088888-turtle.txt
N-Triples: 831-1.0088888-rdf-ntriples.txt
Original Record: 831-1.0088888-source.json
Full Text
831-1.0088888-fulltext.txt
Citation
831-1.0088888.ris

Full Text

C O M M U N I T Y - W I D E I M P A C T S O F A G E N E R A L I S T B R O O D P A R A S I T E , T H E B R O W N - H E A D E D C O W B I R D (Molothrus ater) by K R I S T A L E I G H D E G R O O T B.Sc. , University of Guelph, 1994 A THESIS S U B M I T T E D I N P A R T I A L F U L F I L L M E N T O F T H E R E Q U I R E M E N T S F O R T H E D E G R E E O F M A S T E R O F S C I E N C E in T H E F A C U L T Y O F G R A D U A T E S T U D I E S (Department of Zoology) We accept this thesis as conforming to the required standard T H E U N I V E R S I T Y O F B R I T I S H C O L U M B I A November 1998 © Krista Leigh De Groot, 1998 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of ,Z(?C>l0^j V The University of British Columbia Vancouver, Canada D a t e Nov. 35> inqs DE-6 (2/88) ABSTRACT M a n y ecologists have searched for species that contribute strongly to the structure and composi t ion of communit ies of organisms. It is w i d e l y bel ieved that the Brown-headed C o w b i r d , a generalist b r o o d parasite, is capable o f changing songbird communit ies . C o w b i r d parasit ism may reduce numbers o f suitable hosts, i.e., songbirds that accept c o w b i r d eggs and raise c o w b i r d young. In contrast, songbird species that have e v o l v e d egg ejection behaviour, nest i n cavities, feed c o w b i r d nestlings an unsuitable diet or are too b i g to parasitize, w i l l generally escape the effects o f c o w b i r d parasit ism. Thus , cowbirds may change the composi t ion of entire songbird communit ies by depressing numbers o f suitable host individuals . I tested this hypothesis using an exist ing c o w b i r d removal program in the state of M i c h i g a n , U S A . T h i s removal program was designed to protect the endangered K i r t l a n d ' s W a r b l e r f r o m high levels o f c o w b i r d parasit ism, throughout its 19 2 0 0 k m 2 breeding range. I compared songbird composi t ion in stands o f y o u n g j a c k pine where cowbirds had been removed for 5-11 years to C o n t r o l sites 5-10 k m f r o m c o w b i r d traps and C o n t r o l sites >10 k m f r o m c o w b i r d traps. I predicted that c o w b i r d R e m o v a l sites w o u l d support greater songbird diversity and a greater proport ion o f suitable host vs. unsuitable host indiv iduals relative to C o n t r o l sites. Results f r o m songbird point counts revealed that species diversity was very s imi lar at c o w b i r d R e m o v a l and C o n t r o l sites but R e m o v a l sites contained 4.0 - 8.7 % more suitable hosts than C o n t r o l sites. I conclude that cowbirds only w e a k l y influence the composi t ion of songbird communit ies i n j a c k pine forests o f M i c h i g a n . It remains to be shown that cowbirds affect songbird c o m m u n i t y compos i t ion more strongly i n other areas, e.g., m i d -western U S A , where cowbirds are more abundant. TABLE OF CONTENTS A B S T R A C T .- i i T A B L E O F C O N T E N T S i i i L I S T O F T A B L E S iv L I S T O F F I G U R E S v A C K N O W L E D G E M E N T S v i i I N T R O D U C T I O N 1 The effects o f long-term removal o f B r o w n - h e a d e d C o w b i r d s on songbird communit ies : hypothesis and predictions 2 M E T H O D S 5 Study sites 5 S o n g b i r d point counts 6 C o w b i r d s counts 7 Nest moni tor ing 8 Habitat measures 8 Statistical A n a l y s i s 9 R E S U L T S i 11 Songbird community composi t ion 11 C o w b i r d numbers as a function of distance f r o m traps 13 Nest parasit ism and rates o f dai ly nest surv iva l 14 Habitat measures 15 D I S C U S S I O N 17 Habitat differences 18 N e s t predation 19 C a n cowbirds exert strong demographic effects on host communit ies? 2 0 C O N C L U S I O N S 25 L I T E R A T U R E C I T E D 26 A P P E N D I X 52 iv LIST O F T A B L E S T A B L E 1. B r i l l o u i n ' s index of species diversity for 1996 and 1997 R e m o v a l sites, C o n t r o l sites 5-10 k m f r o m traps and C o n t r o l sites >10 k m f r o m c o w b i r d traps 33 T A B L E 2. M e a n number of cowbirds detected i n c o w b i r d R e m o v a l sites, C o n t r o l sites 5-10 k m f r o m c o w b i r d traps and C o n t r o l sites >10 k m f r o m c o w b i r d traps 34 T A B L E 3. Tests for differences between 1996 and 1997 m a x i m u m l i k e l i h o o d estimates o f dai ly nest survival in c o w b i r d R e m o v a l sites and C o n t r o l sites 5-10 k m f r o m c o w b i r d traps. N u m b e r o f nests i n brackets 35 T A B L E 4. Incidence of parasitism a m o n g suitable hosts i n c o w b i r d R e m o v a l sites, C o n t r o l sites 5-10 k m and C o n t r o l sites >10 k m f r o m c o w b i r d traps 36 V LIST OF FIGURES Figure 1: M a p o f M i c h i g a n (modif ied f r o m B r e w e r et a l . 1991) s h o w i n g locat ion of s tudy. . . 37 F i g u r e 2: Schematic o f transect l ines through a study area w i t h randomly chosen point count locations 38 F i g u r e 3: Songbird c o m m u n i t y composi t ion i n 1996 at ten c o w b i r d R e m o v a l sites and ten C o n t r o l sites 5-10 k m f r o m c o w b i r d traps 39 Figure 4: Songbird c o m m u n i t y composi t ion i n 1997 at eight c o w b i r d R e m o v a l sites, eight C o n t r o l sites 5-10 k m and eight C o n t r o l sites >10 k m f r o m c o w b i r d traps 4 0 F i g u r e 5: M e a n numbers o f unsuitable hosts in 1996 detected dur ing eight minute point counts i n ten c o w b i r d R e m o v a l sites and ten C o n t r o l sites 5-10 k m f r o m c o w b i r d traps 41 F igure 6: M e a n numbers o f unsuitable hosts i n 1997 detected dur ing ten minute point counts i n eight c o w b i r d R e m o v a l sites, eight C o n t r o l sites 5-10 k m and eight C o n t r o l sites >10 k m f r o m c o w b i r d traps 42 Figure 7: P lot o f species heterogeneity (richness and evennes) i n 1996 at ten R e m o v a l sites and ten C o n t r o l sites 5-10 k m f r o m c o w b i r d traps 43 Figure 8: P lot o f species heterogeneity (richness and evenness) i n 1997 at eight R e m o v a l sites, eight C o n t r o l sites 5-10 k m and eight C o n t r o l sites >10 k m f r o m c o w b i r d traps 44 Figure 9: M a x i m u m l i k e l i h o o d estimates o f dai ly nest survival rates for pooled 1996 and 1997 nests o f suitable and unsuitable hosts 45 F i g u r e 10: M e a n numbers o f trees counted i n 2 0 x 2 0 m plots (2 plots/site) i n ten R e m o v a l sites and ten C o n t r o l sites 5-10 k m f r o m c o w b i r d traps i n 1996. See A p p e n d i x Table 1 for species names 46 F i g u r e 11: M e a n numbers of trees counted i n 20 x 2 0 m plots (6 plots/site) i n eight R e m o v a l sites, eight C o n t r o l sites 5-10 k m and eight C o n t r o l sites >10 k m f r o m c o w b i r d traps i n 1997. See A p p e n d i x Table 1 for species names 47 F i g u r e 12: M e a n percentage of each ground cover type in 1 x 1 m plots (2 plots/site) in ten R e m o v a l sites and ten C o n t r o l sites 5-10 k m f r o m c o w b i r d traps i n 1996. See A p p e n d i x Table 1 for species names 48 Figure 13: M e a n percentage of each ground cover type i n 1 x 1 m plots (24 plots/site) i n eight R e m o v a l sites, eight C o n t r o l sites 5-10 k m and eight C o n t r o l sites >10 k m f r o m c o w b i r d traps in 1997. See A p p e n d i x Table 1 for species names 49 v i F i g u r e 14: M e a n density o f vegetation measured at t w o plots per site at ten R e m o v a l sites and ten C o n t r o l sites 5-10 k m f r o m c o w b i r d traps i n 1996 50 Figure 15: M e a n density of vegetation measured at twelve plots per site at eight R e m o v a l sites, eight C o n t r o l sites 5-10 k m and eight C o n t r o l sites >10 k m f r o m c o w b i r d traps i n 1997 51 ACKNOWLEDGEMENTS I w o u l d l ike to thank m y supervisor, Jamie S m i t h for his support and guidance. I especially appreciate the freedom I had to make m y o w n decisions, and I ' l l probably never f ind a more thorough editor! I w o u l d also l ike to thank the rest o f m y supervisory committee; K a t h y M a r t i n , Charles K r e b s and B i l l N e i l l . I am grateful for the help o f A l a i n Goulet w h o came to G r a y l i n g (and to i led i n the j a c k pine) for both years of the study. N o t only was he a dedicated and tireless f i e l d worker , but also a good fr iend w h o kept m y spirits up dur ing that very stressful first f ie ld season. I w o u l d also l ike to thank M a t t K l e i t c h and R e b e c c a Porte for their assistance i n the second year o f f ie ld work. I owe y o u al l a great deal for y o u r hard work, dedicat ion, support, and friendship. I am also indebted to A n d r e w W e n z e l , C h i n S u n and Jamie S m i t h w h o p r o v i d e d m u c h needed volunteer labour. I w o u l d l ike to thank the K i r t l a n d ' s W a r b l e r R e c o v e r y T e a m , especial ly M i k e D e C a p i t a , P h i l H u b e r , Jerry W e i n r i c h , D o u g M u n s o n , R e x E n n i s and C a r o l B o c e t t i for their advice, encouragement and for taking the t ime to show me the ropes in the first year o f m y study. I also owe thanks to a l l the staff at offices o f the U S D A Forest Service and the M i c h i g a n Department o f Natural Resources for helping me to f i n d and photocopy maps of every section of j a c k pine habitat in the L o w e r Peninsula! F i n a l l y , I owe the most thanks to a l l o f m y friends, to A n d r e w W e n z e l and to m y parents. T h i s thesis is a product o f your bel ie f in me. Research was funded by the Natural Sciences and Engineer ing Research C o u n c i l o f Canada (operating grant to J . N . M . Smith) , U . S . F i s h and W i l d l i f e Service, and S i g m a X i , T h e Scientif ic Research Society. 1 INTRODUCTION Community ecologists are primarily interested in determining factors that maintain or alter community structure. Early observations of the importance of some predators in regulating freshwater aquatic communities (Brooks and Dodson 1965, Hal l et al. 1970, Hurlbert et al. 1972) generated considerable research into the role of predators in other systems (see Power et.al. 1996 and Hurlbert 1997 for reviews). This research expanded to include herbivores, producers and mutualists, often using experimental removal or exclusion to determine the effects of individual species on a community of organisms (Power et al. 1996, Hurlbert 1997). To my knowledge, this is the first study to search for community-wide effects generated by an obligate brood parasite. Generalist brood parasites have the potential to change the composition of entire host communities. Their lack of host specificity allows them to strongly affect a number of host species, without the negative feedback on their own numbers generally associated with single host-parasite interactions (Mayfield 1977, May and Robinson 1985, James and McCul loch 1995). The brood-parasitic Brown-headed Cowbird (Molothrus ater) of North America uses over 220 songbird hosts (Friedmann et al. 1977, Lowther 1993) and has greatly expanded its range during the past two centuries. This range expansion is causing concern among ornithologists, since it has exposed previously naive songbird populations to the threat of brood parasitism (Mayfield 1977, Rothstein 1994). It is widely believed that nest parasitism by cowbirds has contributed to declines in some songbird populations, and that cowbirds can change the composition of entire songbird communities (Mayfield 1977, Brittingham and Temple 1983, Terborgh 1989, Wiens 1989a, Bohning-Gaese et al. 1993, Griffith and Griffith in press). 2 S o n g b i r d communit ies (excluding cowbirds) can be d i v i d e d into two groups, depending on their value to cowbirds as hosts. Suitable hosts accept c o w b i r d eggs and feed their y o u n g a largely animal-based diet. Unsuitable hosts inc lude cavity nesters, species that feed their young a m a i n l y plant-based diet, corvids , and species that reject c o w b i r d eggs f r o m their nests (Rothstein 1975, R o h w e r and S p a w 1988). C o w b i r d pressure on suitable hosts may reduce their abundance, relative to the abundance o f unsuitable hosts w i t h w h i c h the c o w b i r d does not interact strongly. S o m e suitable host species may even be extirpated f r o m communit ies where c o w b i r d pressure is intense. If several host species are affected, c o w b i r d pressure may eventually change the composi t ion o f entire songbird communit ies . H o w e v e r , despite ample evidence that c o w b i r d parasit ism causes reproductive losses for hosts ( W a l k i n s h a w 1983, M a r v i l and C r u z 1989, D o n o v a n et al . 1995, James and M c C u l l o c h 1995, R o m i g and C r a w f o r d 1996, B r a d e n et a l . 1997, S e d g w i c k and Iko in press, Strausberger and A s h l e y 1997, S e d g w i c k and Iko in press), there is st i l l little evidence that c o w b i r d parasit ism regulates songbird populations and communit ies ( M a y and R o b i n s o n 1985, Pease and G r z y b o w s k i 1995, S e d g w i c k and Iko in press). Therefore, I designed a control led experiment to measure the effects o f cowbirds on songbird c o m m u n i t y structure. The effects of long-term removal of Brown-headed Cowbirds on songbird communities: hypothesis and predictions I used c o w b i r d removal to test the f o l l o w i n g hypothesis regarding c o w b i r d - i n d u c e d changes to songbird communit ies: Brown-headed cowbirds change the composition of songbird communities by depressing numbers of suitable host individuals. 3 Three predictions o f this hypothesis are: (1) suitable host indiv iduals w i l l make up a larger proport ion of songbird communit ies in areas where cowbirds have been r e m o v e d on a long-term basis, compared to areas i n s imi lar habitat where c o w b i r d densities are unmanipulated, (2) species diversity w i l l be greater i n c o w b i r d r e m o v a l areas and (3) differences between c o w b i r d removal sites and unmanipulated sites w i l l increase wi th increasing distance f r o m c o w b i r d removal areas. W h i l e support for these predictions can provide evidence that cowbirds change the composi t ion of songbird communit ies through their brood parasitic activities, there c o u l d be alternative explanations for the above patterns. Songbird c o m m u n i t y c o m p o s i t i o n may be predicted by local-scale habitat variables such as vegetation structure (height) diversity ( M a c A r t h u r and M a c A r t h u r 1961, M a c A r t h u r et. a l . 1962, C o d y 1981, but see W i l l s o n 1974) and/or specific plant assemblages (James 1971, K a r r and R o t h 1971, Probst et a l . 1992). If differences i n songbird c o m m u n i t y composi t ion are due to loca l habitat variables, I should f i n d differences in vegetation structure and composi t ion between c o w b i r d r e m o v a l sites and experimental controls, despite attempts to control for habitat variables. Landscape-scale factors, such as p r o x i m i t y to agricultural areas or h u m a n populations, also influence densities o f nest predators ( A m b u e l and T e m p l e 1983, W i l c o v e 1985, A n d r e n and A n g e l s t a m 1988). If high levels o f nest predation strongly l i m i t songbird populations, then nest predators may be more important than b r o o d parasites i n determining the structure o f songbird communit ies ( M a r t i n 1988a, M a r t i n 1988b, W i e n s 1989b). T h i s alternative hypothesis should be considered i f patterns 1 & 2 above are accompanied by a lower rate o f predation o f suitable host nests i n c o w b i r d removal sites compared to experimental controls. 4 T h e extensive c o w b i r d removal program designed to protect the K i r t l a n d ' s W a r b l e r (Dendroica kirrtandii) i n Northern M i c h i g a n provided an experimental f ramework in w h i c h to test the hypothesis that cowbirds change songbirds communit ies . T h e K i r t l a n d ' s W a r b l e r is an endangered neotropical migrant songbird w i t h very specific habitat preferences and a l i m i t e d breeding range. T h e y nest only i n y o u n g j a c k pine forests (1-6 m i n height) i n northern M i c h i g a n ( W a l k i n s h a w 1983). C o n c e r n about K i r t l a n d ' s Warblers heightened after a census i n 1971 recorded only 201 s inging males; a marked decline f r o m 502 counted a decade earlier ( M a y f i e l d 1972). Researchers suggested that h igh levels o f nest parasit ism by cowbirds contributed to this decl ine ( M a y f i e l d 1972, W a l k i n s h a w 1972, R y e l 1981), thus c o w b i r d removal f rom K i r t l a n d ' s W a r b l e r breeding areas began in 1972 and has continued every year since (Walk inshaw 1983, D e C a p i t a in press). T h i s c o w b i r d removal program, represents a classic P R E S S perturbation (Bender et a l . 1984) whereby c o w b i r d removal at indiv idua l sites was sustained for 5-11 years. I compared songbird communit ies on these sites to experimental controls at least 5 k m f r o m c o w b i r d removal areas. 5 METHODS Study sites F i e l d w o r k took place in the Jack pine barrens o f Northern L o w e r M i c h i g a n and encompassed most of the breeding grounds o f the K i r t l a n d ' s W a r b l e r ( F i g . 1). T h i s landscape is heavi ly forested, with a mosaic o f managed stands o f conifers and pockets o f deciduous forest. Study sites were i n stands o f even-aged j a c k pine (Pinus banksiana), often interspersed w i t h oak (Quercus spp.) and p i n cherry (Primus pensylvanica) and occasional ly smal l stands of red pine (Pinus resinosa) or t r e m b l i n g aspen (Populus tremuloides). T h e dominant ground cover was o f grasses (e.g., Andropogon gerardii, Deschampsia flexuosd), sedge (Carex pensylvanica), blueberry (Vaccinium spp.), and other members o f the Heath family (Ericaceae). A p p e n d i x , Table 1 gives a complete list o f trees, shrubs and ground cover found on study sites. C o w b i r d traps were erected and maintained by the U . S . F i s h and W i l d l i f e Service across a 19 200 k m 2 area where most nesting K i r t l a n d ' s Warblers occur ( D e C a p i t a in press). U p to 67 traps (mean = 41 traps) operated between 1972 and 1997. These traps r e m o v e d a total o f 105 309 cowbirds (mean = 4050 cowbirds/per year) f r o m K i r t l a n d ' s W a r b l e r breeding areas ( D e l o r i a and D e C a p i t a 1997). I chose ten c o w b i r d R e m o v a l sites adjacent to active c o w b i r d traps that had been i n operation for 5-11 consecutive years (mean 7.6 years). C h o i c e o f R e m o v a l sites was l i m i t e d by permits w h i c h prohibited m y access to some K i r t l a n d ' s W a r b l e r breeding sites. T e n C o n t r o l sites were chosen to best match the habitat characteristics o f c o w b i r d R e m o v a l sites (total number o f sites in 1996 = 20). I located suitable C o n t r o l areas using survey maps, f o l l o w e d by extensive ground truthing. A l l C o n t r o l sites were at least 5 k m f r o m c o w b i r d traps and f r o m any area that had experienced c o w b i r d removal w i t h i n the past f ive years. In 1997, eight c o w b i r d R e m o v a l sites and eight C o n t r o l sites 5-10 k m f r o m c o w b i r d traps were used. In addit ion, the scale o f the project was expanded to include eight C o n t r o l sites >10 k m from c o w b i r d traps (total number of sites = 24). Site area encompassed a hal f c i rc le o f radius one k m (total area = 1.57 k m 2 ) , adjacent to the c o w b i r d trap, i n the case o f c o w b i r d R e m o v a l sites. A hal f c i rc le was used to enable sites to be monitored without entering K i r t l a n d ' s W a r b l e r territories, as needed to c o n f o r m with conditions o f m y entry permit. Songbird point counts S o n g b i r d point counts were conducted to test the predict ion that songbird communit ies in c o w b i r d R e m o v a l areas differed i n composi t ion f r o m C o n t r o l areas at least 5 k m f r o m c o w b i r d traps. F o u r one k m transect l ines, each 60° apart, were f lagged w i t h i n each site and six permanent count stations were chosen a long these transects ( F i g . 2). L o c a t i o n s of count stations were chosen randomly. H o w e v e r , to a v o i d re-counting i n d i v i d u a l songbirds, it was necessary to restrict randomizat ion so that successive count stations were not c loser than 4 0 0 m . T h e same two observers performed avian point counts i n both 1996 and 1997. T w o additional observers were trained for point counts i n 1997. I trained observers over a four week per iod each year, during w h i c h al l personnel performed simultaneous point counts at each sampl ing point. Results and observations were compared and discussed among observers before leaving count stations. Af ter the training per iod, potential observer biases i n the data 7 were balanced out by ensuring that each observer counted the same number of Removal sites as Control sites. We performed eight minute, unlimited radius point counts between dawn and 10:00 a.m., identifying birds by song, calls or visual observation; noting species, distance from the observer, and time (minute) of first detection. Counts were conducted on most mornings weather permitting, i.e., no heavy rain or constant drizzle and winds < 25 km/hr. Each morning we simultaneously sampled a Removal and a Control site in 1996 and a Removal, a Control site 5 - 1 0 km and a Control site >10 km from traps in 1997. This procedure ensured there were no biases due to differences in weather conditions during count periods. We counted all sites twice between early June and mid-July in 1996 and three times between mid May and early July in 1997. In 1997, I extended the point count duration to ten minutes, after analysis of detection curves from 1996 indicated that the rate of detection was not leveling off in the last few minutes of the 1996 point counts (Appendix, Fig. 1). Cowbirds counts Cowbirds were censused to test the assumption that cowbird densities were greater in Controls than at cowbird Removal sites. We censused cowbirds during point counts, noting the sex of individuals. In 1997, five minutes of playback of cowbird female chatter calls was added after each ten minute point count to improve the likelihood of detecting cowbirds (Miles and Buehler in press, Rothstein and Cook in press), following very low cowbird detection rates in 1996. 8 Nest monitoring Samples o f nests (33 nests i n 1996 and 98 nests i n 1997) were moni tored w i t h i n R e m o v a l and C o n t r o l sites to test for differences in rates o f nest surv iva l and to test the assumption that frequency and intensity o f c o w b i r d parasit ism was higher i n C o n t r o l sites compared to R e m o v a l sites. W e checked nests every 3-5 days i n 1996 and every 3-6 days i n 1997. W e targeted suitable c o w b i r d hosts for moni tor ing , but unsuitable host species' nests were monitored opportunist ical ly to assess their survival rates. Habitat measures Several vegetation variables were measured to test the alternative hypothesis that significant habitat differences between c o w b i r d R e m o v a l and C o n t r o l sites caused differences i n songbird communit ies . W e sampled two count stations per site in 1996 and a l l s ix count stations i n 1997. I randomly selected two (1996) or s ix (1997) 2 0 x 2 0 m plots between 0-100 m f r o m point count stations at each site. W e counted trees, shrubs and snags w i t h i n these plots, noting species and circumference o f trunks at breast height. I later converted c ircumference measurements to diameter at breast height (dbh). W i t h i n each 20 x 2 0 m plot, I randomly selected one (1996) or four (1997) 1 x 1 m plots w i t h i n w h i c h we estimated the proport ion o f each ground cover type (e.g., grasses and sedge, Er icaceous shrubs, leaf litter) to the nearest f ive percent. W e measured ground and understory (0-1 m) vegetation density at one (1996) or two (1997) o f the 1 x 1 m plots us ing a 1 x l m board w i t h a painted g r i d c o m p o s e d o f 100 squares. W e f ixed the board i n a vertical posit ion on the ground and then counted the number 9 of 0.1 x 0.1 m squares on the board that were unobscured, < half obscured, > half obscured or completely obscured by vegetation. Observations were made from 15 m away and from each cardinal direction. I later summed these values and assigned them the following weights; 0 for unobscured, 0.25 for < half obscured, 0.5 for > half obscured or 0.75 for completely obscured. The sum of these weighted values gave an overall vegetation density score. Statistical Analysis I used one-way Analysis of Variance ( A N O V A ) to analyze differences in mean values among Removal sites, Control sites 5-10 km from cowbird traps and Control sites >10 km from traps. Kruskall-Wallis and Mann-Whitney non-parametric tests were applied when data did not meet the assumptions of a parametric Analysis of Variance. In 1997,1 used post-hoc multiple comparison tests (Bonferroni and Dunnet's T3) to determine the location of significant differences among the three groups tested (Removal sites, Controls 5-10 km and Controls >10 km from traps). A significance level of 5% was used. Some additional statistical tests were used as required (see below). D E S C R I P T I O N O F S O N G B I R D C O M M U N I T Y - The six count stations within each site were not considered statistically independent sampling units. Therefore, I used mean values for each site for further analysis (Hurlbert 1984). Songbird individuals, excluding cowbirds, were placed into categories of suitable and unsuitable hosts as outlined above. I tested for differences in the mean proportion of suitable hosts in cowbird Removal and Control sites using Repeated Measures Analysis of Variance (Kuehl 1994). I repeated this procedure using the mean number (vs mean proportion) of suitable hosts as the dependent variable. 10 I investigated differences in relative abundance o f unsuitable hosts in R e m o v a l and C o n t r o l areas by d i v i d i n g unsuitable hosts into the f o l l o w i n g categories: (a) rejecters, (b) cavity nesters, (c) corvids and, (d) species that feed nest l ing cowbirds an unsuitable diet for growth and survival . I then analyzed mean values using One-way A n a l y s i s of Var iance ( A N O V A ) or K r u s k a l l - W a l l i s / M a n n - W h i t n e y tests as outl ined above. I calculated Renkonen (Percent s imi lar i ty) indices to quantify the percentage difference in species composi t ion and used B r i l l o u i n indices to compare species richness and evenness i n R e m o v a l and C o n t r o l sites (Krebs 1989). N E S T D A T A -1 f o l l o w e d methods i n B a r t and R o b s o n (1982) for calculation of m a x i m u m l i k e l i h o o d estimates o f dai ly nest survival rates. Nests monitored i n 1996 and 1997 were pooled to achieve greater statistical power, after testing for between year differences in estimated daily nest survival rates. I computed estimates o f da i ly nest survival rates for suitable hosts and unsuitable hosts separately, for R e m o v a l sites, C o n t r o l sites 5-10 k m f r o m traps and C o n t r o l sites >10 k m f r o m c o w b i r d traps. V E G E T A T I O N - P r i n c i p a l components analyses ( P C A ) were appl ied to attempt to reduce the number o f variables i n the tree and ground cover data sets. H o w e v e r , the new P C A factors d i d not sufficiently explain the variation i n the data according to the broken-stick model (Legendre and Legendre 1983, Jackson 1993). Thus, I used a l l or ig inal variables in Mult ivar iate Analyses o f V a r i a n c e ( M A N O V A ) to test for differences a m o n g R e m o v a l , Controls 5-10 k m and Controls >10 k m f r o m c o w b i r d traps. 11 RESULTS Songbird community composition Proportion of suitable host individuals In 1996, there was a signif icantly greater proport ion o f suitable host individuals i n the songbird communit ies o f c o w b i r d R e m o v a l sites compared to C o n t r o l sites 5-10 k m f r o m c o w b i r d traps (Repeated Measures A N O V A , Fi , ig = 11.76, p = .003). O n average 6 7 . 4 % o f songbirds detected at c o w b i r d R e m o v a l sites were suitable c o w b i r d hosts, whereas only 5 8 . 7 % of the songbird c o m m u n i t y was composed o f suitable host indiv iduals on C o n t r o l sites 5-10 k m f r o m c o w b i r d traps ( F i g . 3; A p p e n d i x Table 2 & 3). In 1997, suitable hosts c o m p r i s e d 6 4 . 4 % of the songbird indiv iduals detected on c o w b i r d R e m o v a l sites, 6 0 . 4 % on C o n t r o l sites 5-10 k m and 5 9 . 1 % on C o n t r o l sites >10 k m f r o m c o w b i r d traps (F ig . 4; A p p e n d i x Table 4 & 5). These differences were not statistically significant (Repeated Measures A N O V A , F 2,2i = 2.86, p = .08). Numbers of suitable and unsuitable hosts T h e increase i n the proport ion o f suitable hosts detected i n R e m o v a l sites compared to C o n t r o l sites (1996 and 1997) was due i n large part to an increase i n numbers of suitable host indiv iduals detected i n R e m o v a l sites. H o w e v e r , the effect was also due to fewer unsuitable host individuals in R e m o v a l sites. In 1996, there were fewer unsuitable host indiv iduals o f al l types, i.e., rejecters, cavity nesters, corvids, and songbirds that feed c o w b i r d y o u n g an unsuitable diet, at c o w b i r d R e m o v a l sites compared to C o n t r o l sites 5-10 k m f r o m c o w b i r d traps ( F i g . 5). H o w e v e r , w i t h 12 the exception o f cavity nesters ( M a n n - W h i t n e y U , z = 3.04, p = .002), these trends were not statistically significant ( M a n n W h i t n e y U , z < .99, p > .05). In 1997, mean numbers o f unsuitable hosts were consistently l o w e r i n c o w b i r d R e m o v a l sites compared to C o n t r o l sites >10 k m f r o m c o w b i r d traps. H o w e v e r , there were no clear trends between mean numbers o f unsuitable host indiv iduals i n C o n t r o l sites 5-10 k m f r o m traps and c o w b i r d R e m o v a l sites ( F i g . 6). N o n e of the differences among groups were statistically significant i n 1997 ( K r u s k a l l W a l l i s , x2.05,2 < 2.65, p > .05). R e m o v a l of unsuitable hosts f r o m the analysis by using absolute numbers (vs. proportions) o f suitable hosts st i l l reveals a statistically significant difference between R e m o v a l and C o n t r o l sites i n 1996 (Repeated Measures A N O V A , F M 8 = 7.50, p = .01), conf i rming that changes i n suitable hosts numbers influenced differences i n songbird community composi t ion i n R e m o v a l and C o n t r o l sites. In 1997, analysis o f numbers (vs. proportions) o f suitable hosts continues to y i e l d non-signif icant differences (Repeated Measures A N O V A , F 2 , 2 i = 1.88, p = .18) a m o n g R e m o v a l , C o n t r o l sites 5-10 k m and C o n t r o l sites >10 k m f r o m c o w b i r d traps. Community similarity and species diversity T h e shift to a greater proport ion o f suitable c o w b i r d hosts i n the songbird communit ies o f c o w b i r d R e m o v a l sites was a result o f smal l responses o f many host populations rather than a qualitative change i n species composi t ion . There were no species absent on C o n t r o l sites that were abundant on R e m o v a l areas and vice versa ( A p p e n d i x , Tables 2-5). Rather, the shift i n proportions o f suitable hosts was a result o f smal l posit ive shifts i n the abundance o f host 13 individuals in cowbird Removal sites, compared to Control sites. Removal sites were 80.8% similar to Control sites in 1996, and 83% similar to Control sites 5-10 km from cowbird traps and Control sites >10 km from cowbird traps in 1997. Control sites 5-10 km and Control sites >10 km from cowbird traps were 87% similar. Species evenness was also very similar across Removal sites, Control sites 5-10 km from traps and Control sites >10 km from cowbird traps in both 1996 (Fig. 7, Table 1) and 1997 (Fig. 8, Table 1). Species richness was only marginally higher in Removal sites compared to Controls in both years of the study (Table 1). Cowbird numbers as a function of distance from traps Cowbird traps were highly effective at reducing cowbird abundance at Removal sites. In 1996, 0.025 male cowbirds were detected per count station at Removal sites and no female cowbirds were detected during point counts on Removal or Control sites (Table 2). Male cowbird numbers increased over six fold at Control sites 5-10 km from cowbird traps compared to cowbird Removal sites (Table 2; Mann Whitney U , z = 2.17, p = .03). In 1997, the number of female and male cowbird detections during the five minute playback period differed significantly among Removal, Control sites 5-10 km and Control sites >10 km from cowbird traps (Table 2; Kruskall Wallis, females X-osa = 11.01, p = .004; males X-o5,2 = 18.79, p<0.001). Mean number of female cowbird detections more than doubled and male numbers increased more than nine-fold from Removal sites to Controls 5-10 km from traps. However multiple comparison tests reveal that these differences are statistically significant for male abundance only (Table 2; females Dunnett T3, Mean Difference = .21, 14 p > .05; males Dunnett T 3 , M e a n Dif ference = 1.37, p = .001). T e n times more female cowbirds and twenty times more males were counted at C o n t r o l sites >10 k m f r o m c o w b i r d traps compared to R e m o v a l areas. (Table 2; females Dunnett T 3 , M e a n Difference = 1.21, p = .01; males Dunnett T 3 , M e a n Dif ference = 3.29, p = .003). F e m a l e c o w b i r d numbers increased four-fold and male abundance doubled f r o m C o n t r o l sites 5-10 k m f r o m traps to Controls >10 k m (Table 2; females Dunnett T 3 , M e a n Dif ference = 1.00, p = .025; males Dunnett T 3 , M e a n Dif ference = 1.92, p = .046 ). Nest parasitism and rates of daily nest survival There were no significant differences between m a x i m u m l i k e l i h o o d estimates o f dai ly nest surv iva l i n 1996 and 1997 for suitable or unsuitable hosts (see Table 3). Therefore, I pooled data f r o m 1996 and 1997 nests for analyses o f nest survival rates f r o m c o w b i r d R e m o v a l and C o n t r o l sites 5-10 k m f r o m c o w b i r d traps. T h e 41 nests o f suitable hosts i n c o w b i r d R e m o v a l sites s u r v i v e d at a rate o f 0.957 per day. T h i s was sl ightly lower than the survival rate o f 35 suitable host nests i n C o n t r o l sites 5-10 k m (0.975/day) and 20 suitable host nests i n Contro ls >10 k m f r o m c o w b i r d traps (0.974/day) ( F i g . 9). H o w e v e r , this result was not statistically signif icant ( Z = 1.53, p = 0.063; Z = 1.19, p = 0.117). Differences in dai ly nest surv iva l estimates, 0.979/day, 0.982/day and 0.982/day o f 15 unsuitable host nests in c o w b i r d R e m o v a l sites, 16 nests at C o n t r o l sites 5-10 k m and 6 nests on Controls >10 k m f r o m c o w b i r d traps respectively, were smal l and also not statistically significant ( Z = 0.22, p = 0.41; F i g . 9). N o parasitized nests were located in 1996 (Table 4) nor were any c o w b i r d f ledglings detected w h i l e performing other w o r k on the sites or dur ing careful observation o f 29 f ledgl ing 15 famil ies of suitable hosts. In 1997, no parasitized nests were located on c o w b i r d R e m o v a l sites. H o w e v e r , six parasitized nests (25%) containing a total o f 10 c o w b i r d eggs were located on C o n t r o l sites 5-10 k m f r o m c o w b i r d traps and f ive parasit ized nests (25%) containing 7 c o w b i r d eggs were located on C o n t r o l sites >10 k m f r o m c o w b i r d traps (Table 4). W h e n samples o f nests f r o m 1996 and 1997 are pooled, observed parasit ism rates increase f r o m 0 % in c o w b i r d R e m o v a l sites, 17.1% i n C o n t r o l sites 5-10 k m f r o m c o w b i r d traps to 2 5 % i n C o n t r o l sites >10 k m f r o m c o w b i r d traps (Table 4). T h e primary host at a l l C o n t r o l sites was the H e r m i t T h r u s h (Catharus guttatus). F i v e o f eight (62.5%) thrush nests were parasit ized on Contro ls 5-10 k m f r o m traps and four o f six (66.6%) were parasitized on Controls >10 k m f r o m c o w b i r d traps. These nine nests contained an average o f 1.9 c o w b i r d eggs. I only observed parasit ism o f two other host species. One O v e n b i r d (Seiurus aurocapillus) nest and one S o n g S p a r r o w (Melospiza melodia) nest were found containing a single c o w b i r d egg. H o w e v e r , l o w sample sizes for host species other than H e r m i t Thrushes made it dif f icult to assess the frequency o f host use on m y study sites. H a b i t a t m e a s u r e s N u m b e r s and composi t ion o f tree species counted on 2 0 x 2 0 m plots d i d not differ signif icantly between c o w b i r d R e m o v a l and C o n t r o l sites in either 1996 ( F i g . 10, P i l l a i ' s Trace multivariate test, F M 8 = 1.59, p > .05) or 1997 ( F i g . 11, R o y ' s Largest R o o t mult ivariate test, F 2 i 2 1 = 2.82, p > .05). G r o u n d cover composi t ion measured in 1 x 1 m plots was not s ignif icantly different between R e m o v a l and Controls i n 1996 ( F i g . 12, P i l l a i ' s Trace multivariate test, F 7 / i 2 = 0.93, 16 p > .05). There was however, a smal l but statistically significant difference i n ground cover compos i t ion among 1997 R e m o v a l sites, C o n t r o l sites 5-10 k m f r o m traps and C o n t r o l sites >10 k m f r o m traps ( F i g . 13, R o y ' s Largest R o o t mult ivariate test, F 3 j 2 0 = 11.00, p = 0.03). H o w e v e r , one-way analysis o f variance on each of the 2 0 variables revealed that no single ground cover variable differed signif icantly among R e m o v a l , Controls 5-10 and Controls >10 k m f r o m traps ( F 2 j 2 i < 2 .03, p > .05). In 1996, there were no significant differences i n vegetation density f r o m 0 - 1 m between c o w b i r d R e m o v a l and C o n t r o l sites 5-10 k m f r o m traps ( F i g . 14, M a n n - W h i t n e y U , ground cover z = .79, p > .05, l o w cover z = .95, p > .05, h i g h cover z = .45 p > .05). There were no significant differences i n vegetation density i n 1997 among R e m o v a l , C o n t r o l sites 5-10 k m and C o n t r o l sites >10 k m f r o m c o w b i r d traps ( F i g . 15, one-way A N O V A , ground cover F 2 j 2 3 = 1.07, p > .05, l o w cover F 2 > 2 3 = 1.16, p > .05, h igh cover F 2 > 2 3 = 1.13, p > .05). 17 DISCUSSION There is considerable concern that the brood-parasitic activities of Brown-headed cowbirds are contributing to declines in some songbird populations, and that cowbirds generate widespread changes in songbird community composition (Mayfield 1977, Brittingham and Temple 1983, Terborgh 1989, Wiens 1963, Bohning-Gaese et al. 1993). I used an existing cowbird removal program to measure the effects of cowbirds on songbird communities. I compared the composition of songbird communities in cowbird Removal areas to Control sites 5-10 km from cowbird traps and Controls >10 km from traps to test the predictions that songbird communities in cowbird Removal areas would contain: (1) a greater proportion of suitable vs. unsuitable host individuals, (2) greater songbird diversity, and that (3) these effects would strengthen with increasing distance from cowbird Removal sites. Contrary to my predictions, the proportion of suitable host individuals did not increase strongly from sites 5-10 km from traps to sites >10 km from cowbird traps and I found no support for the prediction that songbird species diversity would increase as a result of long-term (>5 years) cowbird removal. Species richness and evenness were very similar across Removal sites and all Control sites in both years of the study. While my results do support the first prediction, the magnitudes of these changes are small and support is weak in the second year of the study. The songbird communities in cowbird Removal sites supported 8.7 % more suitable host individuals compared to Control sites 5-10 km from traps in 1996. In the following year, there was a 4.0 % difference observed between cowbird Removal sites and Controls 5-10 km from traps and a 5.3% difference between Removal sites and Controls >10 km from cowbird traps. This result was statistically significant in 1996 only. On average, 18 c o w b i r d R e m o v a l sites contained 6 % more suitable host indiv iduals than a l l C o n t r o l sites i n the two years o f the study. These differences are smal l when compared to the c o m m u n i t y - w i d e effects generated i n some w e l l - k n o w n removal studies. R e m o v a l o f a sea star predator resulted i n clearly v is ible dominance of mussels in rocky intertidal communit ies , compared to the diverse assemblages of species i n experimental controls (Paine 1974). E x p e r i m e n t a l removal o f a predaceous fire ant f r o m corn and squash plants resulted i n 0.17 - 49 f o l d increases in abundance of 35 arthropod species, relative to controls ( R i s c h and C a r r o l l 1982). H o w e v e r , these two examples come f r o m relatively simple systems. T h e regulation of avian c o m m u n i t y composi t ion may be so complex that removal o f a single factor only results i n smal l observable effects. Before I consider the l imits to cowbird- induced changes on songbird communit ies i n detail , I first discuss the alternative explanations for the smal l differences i n songbird communit ies observed in m y study. H a b i t a t d i f ferences S o n g b i r d c o m m u n i t y compos i t ion is often correlated w i t h the structure and compos i t ion o f vegetation ( M a c A r t h u r and M a c A r t h u r 1961, M a c A r t h u r et a l . 1962, James 1971, K a r r and R o t h 1971, W i l l s o n 1974, C o d y 1981, Probst et a l . 1992), therefore I carefully matched the habitat of experimental Controls to c o w b i r d R e m o v a l sites. H o w e v e r , I also measured habitat variables in detail , to test the alternative hypothesis that observed differences i n songbird c o m m u n i t y structure were the result o f differences in habitat. T h e majority o f suitable hosts i n the c o m m u n i t y are ground nesters, thus, compos i t ion of ground cover is l i k e l y to be the most important component o f vegetation, f o l l o w e d by 19 composi t ion and density o f surrounding trees and shrubs, and number o f snags. B a s e d on detailed sampl ing w i t h i n each study plot, structure and c o m p o s i t i o n of vegetation were not discernibly different i n R e m o v a l and C o n t r o l sites. H o w e v e r , the intensity o f vegetation sampl ing on the 1996 study sites was m u c h lower than i n 1997. M o r e cavity nesters were detected on C o n t r o l sites relative to c o w b i r d R e m o v a l areas i n 1996, suggesting that C o n t r o l sites may have had significantly more snags than R e m o v a l sites. Therefore, habitat differences might sti l l account for some o f the observed differences in songbird c o m m u n i t y composi t ion between c o w b i r d R e m o v a l sites and Controls 5-10 k m f r o m c o w b i r d traps i n 1996. I confidently reject this alternative hypothesis for the 1997 data, and consider that I successfully matched habitats at C o n t r o l and R e m o v a l sites. N e s t p r e d a t i o n I checked for differences i n rates o f nest survival i n R e m o v a l and C o n t r o l sites to test the alternative hypothesis that nest predation rather than nest parasi t ism was d r i v i n g songbird community composi t ion ( M a r t i n 1988a, M a r t i n 1988b, W i e n s 1989a). I f nest predators were responsible for the sl ightly lower proportion of suitable host i n d i v i d u a l s found on C o n t r o l sites, then I expected to f i n d a lower rate o f dai ly nest survival i n C o n t r o l areas compared to c o w b i r d R e m o v a l sites. Estimates o f dai ly nest survival were actually s l ightly h i g h e r in Controls sites 5-10 k m f r o m traps and Controls >10 k m f r o m c o w b i r d traps c o m p a r e d to c o w b i r d R e m o v a l sites; clearly refuting this alternative hypothesis. 20 Can cowbirds exert strong demographic effects on host communities? T h e absence of strong shifts i n songbird c o m m u n i t y compos i t ion may be due to one or more o f the f o l l o w i n g reasons: (1) the large spatial scale o f effective c o w b i r d removal and an insufficient t ime scale over w h i c h to detect host demographic changes, (2) source-sink populat ion dynamics o f host populations and (3) an inabi l i ty o f cowbirds at moderate densities, to l i m i t host populations. I n o w discuss each o f these possibi l i t ies . Spatial and temporal scale of cowbird removal Paradoxical ly , the absence o f strong shifts i n songbird c o m m u n i t y composi t ion in m y data may be m a i n l y due to a strong treatment effect. C o w b i r d removal was so effective that it reduced c o w b i r d densities to near zero on R e m o v a l sites as w e l l as on C o n t r o l sites 5-10 k m f r o m c o w b i r d traps. I was able to verify m y assumption that cowbirds were present i n very l o w densities i n c o w b i r d R e m o v a l areas and i n higher densities on experimental Controls . H o w e v e r , there was a strong gradient effect, whereby cowbirds were almost e l iminated f r o m the area directly adjacent to the c o w b i r d trap, densities increased s l ightly at sites 5-10 k m f r o m traps, and increased strongly at sites >10 k m f r o m c o w b i r d traps. These data suggest that the network o f c o w b i r d traps i n Northern L o w e r M i c h i g a n affects within-year c o w b i r d densities at least 5 k m f r o m c o w b i r d removal areas. T h i s is a plausible result g iven that cowbirds may commute up to 7 k m f r o m breeding to feeding ranges (Rothstein et a l . 1984, T h o m p s o n 1994). H o w e v e r , in this heavi ly forested region o f Northern M i c h i g a n , cowbirds are l i k e l y to commute shorter distances than the recorded m a x i m u m . Areas w e l l beyond 10 k m f r o m c o w b i r d traps may have supported even greater c o w b i r d densities and thus lower abundances 21 o f suitable host indiv iduals . H o w e v e r , habitat at this range is too different f r o m habitat on R e m o v a l areas to test this predict ion adequately. Despite a marked increase in female c o w b i r d numbers at sites >10 k m f r o m c o w b i r d traps, c o w b i r d densities on these sites are st i l l l o w relative to some regions i n N o r t h A m e r i c a . T w i c e as many female cowbirds were detected i n shorter count intervals (6 m i n vs. 10 m i n point counts performed i n this study) i n forests o f I l l i n o i s ( R o b i n s o n et a l . in press). B r e e d i n g B i r d Survey ( B B S ) data indicate that cowbirds are only about a third as abundant in Northern M i c h i g a n (10-30 cowbirds per B B S route) relative to regions o f midwestern U S A (30 - 100 or >100 cowbirds per B B S route) (Peterjohn et a l . in press). T h e magnitude o f changes i n host demography is also l i k e l y to be a function o f the duration o f c o w b i r d removal . D u e to the shifting nature o f suitable K i r t l a n d ' s W a r b l e r breeding habitat, no c o w b i r d R e m o v a l areas had been trapped for the f u l l 26 years o f the removal program. C o w b i r d R e m o v a l sites censused in this study had been trapped annually for 5-11 years. A marked shift in c o m m u n i t y composi t ion w o u l d be more l i k e l y in areas that support higher c o w b i r d densities, and where c o w b i r d removal is continuous at the same location over a longer period. Source-sink dynamics of host populations M a n y authors have suggested that c o w b i r d pressure can drive host populat ion declines ( M a y f i e l d 1977, B r i t t i n g h a m and T e m p l e 1983, T e r b o r g h 1989, B o h n i n g - G a e s e et a l . 1993). M y results do not strongly support this c l a i m . H o w e v e r , dispersal o f indiv iduals can act to l i m i t the impact o f cowbird- induced populat ion changes. Immigrat ion f r o m productive "source" populations may compensate for reduced recruitment i n " s i n k " populations (Pul l iam 22 1988). There is increasing evidence that these processes can operate in avian populations ( P u l l i a m and Danie lson 1991, B r a w n and R o b i n s o n 1996). M a n y host species present i n the j a c k pine ecosystem are habitat generalists w i t h an extensive range across N o r t h A m e r i c a . T h e K i r t l a n d ' s W a r b l e r is a notable exception, in that it has very specific habitat requirements and a l i m i t e d breeding range. It is therefore possible that source-sink population dynamics reduce the magnitude o f cowbird- induced changes in host populations. Songbirds nesting w i t h i n c o w b i r d R e m o v a l areas may even contribute immigrants to less productive areas outside o f c o w b i r d R e m o v a l sites. E m p i r i c a l data on host dispersal distances and further study on host seasonal product iv i ty i n these populations are required to test these hypotheses. The effects of cowbird parasitism on host communities and populations M y data on songbird c o m m u n i t y composi t ion and c o m m u n i t y - w i d e parasit ism rates suggest that moderate c o w b i r d pressure is not sufficient to generate strong demographic effects i n host communit ies in the j a c k pine forests o f Northern M i c h i g a n . T h i s statement begs the question: what levels o f parasit ism are required to generate detectable changes i n songbird communities? It is clear that the cost o f parasit ism varies f r o m host to host w i t h i n a c o m m u n i t y (Friedmann 1963, Rothstein 1975). A crit ical level o f parasitism logical ly exists (and may vary geographical ly) for each host species, above w h i c h host population declines w i l l ensue without steady immigrat ion f r o m source populations. M a x i m u m sustainable levels o f parasit ism for i n d i v i d u a l host species are u n k n o w n , although there have been efforts to model the consequences o f nest parasit ism on host demography ( M a y and R o b i n s o n 1985, Pease and G r z y b o w s k i 1995, G r z y b o w s k i and Pease in press b). M a y f i e l d (1977) suggested that some 23 smal l hosts w o u l d be i n danger i f rate o f parasit ism exceeded 3 0 % o f nests. S m i t h (in press) predicts that many otherwise healthy host populations c o u l d sustain parasit ism levels o f 6 0 % . T h i s estimate is supported by m o d e l i n g o f cowbird-host demography ( G r z y b o w s k i and Pease in press a). H o w e v e r , no e m p i r i c a l data exist to determine whether these estimates are reasonable, or h o w these estimates should vary depending on host size, number of broods per breeding season, and incubation period ( M a y f i e l d 1977, S m i t h in press). M a n y managers and researchers report signif icantly l o w e r rates o f nest parasitism on nests o f indiv idua l host species, f o l l o w i n g the onset o f c o w b i r d removal (Stutchbury 1997, D e C a p i t a in press, Gr i f f i th and Gr i f f i th in press, H a y den et al . in press, W h i t f i e l d in press). N u m b e r s o f K i r t l a n d ' s Warblers stabil ized after c o w b i r d trapping began in 1972, however, numbers d i d not increased significantly unt i l a large tract o f breeding habitat became available in the 1990's ( D e C a p i t a in press). Unfortunately, it is impossible to determine whether c o w b i r d trapping arrested the decl ine o f K i r t l a n d ' s W a r b l e r or whether this was due to other causes on the breeding and/or winter ing range (Probst 1986, James and M c C u l l o c h 1995, H a n e y et al . 1998, Rothste in and C o o k in press). Gr i f f i th and G r i f f i t h (in press), H a y d e n et al . (in press) and W h i t f i e l d (in press) a l l report that c o w b i r d trapping resulted in growth o f Least B e l l ' s V i r e o (Vireo belliipusillus), B l a c k - c a p p e d V i r e o (Vireo atricapillus) and Southwestern W i l l o w Flycatcher (Empidonax trailii extimus) populations respectively. H o w e v e r , only W h i t f i e l d ' s data are part o f a control led experiment, i n w h i c h c o w b i r d removal areas are compared to reference areas wi th no c o w b i r d removal (Rothstein and C o o k in press). Therefore there is sti l l l i m i t e d information available on the extent to w h i c h cowbirds affect host populations. 24 A recent s imulat ion model o f the effects o f c o w b i r d parasit ism on songbird communit ies provides a theoretical f ramework w i t h w h i c h to understand cowbird-host interactions and suggests that c o m m u n i t y - w i d e effects are possible ( G r z y b o w s k i and Pease in press b). H o w e v e r , many key parameters required i n this and other host demographic models are u n k n o w n or vary tremendously among host species ( G r z y b o w s k i and Pease in press b). It is also clear that many factors other than parasit ism by cowbirds can influence songbird population dynamics and thus shape songbird communit ies . S o n g b i r d recruitment depends not only on nesting success but also on survival unti l the next breeding season. F o o d avai labi l i ty , weather, predation, and habitat deterioration can a l l influence surv iva l throughout the year. ( M a r t i n 1988a, M a r t i n 1988b, W i e n s 1989b, N e w t o n 1994, Rotenberry et a l . 1995, H o l m e s et al . 1992, Cote and Sutherland 1997) T h i s study documented small differences i n songbird c o m m u n i t y composi t ion i n areas where cowbirds were removed for 5 - 1 1 years compared to control areas w i t h moderate c o w b i r d densities. I believe that c o w b i r d impacts on other songbird communit ies are l imited , and w i l l be roughly proportional to their relative abundance i n the c o m m u n i t y . Thus large effects are only l i k e l y i n landscapes where cowbirds are dominant members (>6%) of the songbird c o m m u n i t y . I recommend further tests o f this hypothesis us ing exist ing c o w b i r d removal programs c o m b i n e d w i t h experimental controls, in landscapes that support greater densities o f cowbirds. 25 CONCLUSIONS This is the first experimental study to investigate the demographic consequences of cowbird parasitism on entire songbird communities. I used an existing cowbird removal program to test the hypothesis that cowbirds change the composition of songbird communities by reducing numbers of suitable host individuals. I compared the songbird communities of sites where cowbirds had been removed annually for 5 - 11 years to Control sites 5-10 km from cowbird traps and Control sites >10 km from cowbird traps. Cowbird traps were very effective at removing cowbirds from the area 0 - 1 km from traps. Cowbird densities increased moderately at sites 5 - 1 0 km from cowbird traps and increased sharply at sites >10 km from traps. M y study is the first to document landscape-scale effects of a network of cowbird traps on cowbird densities. Songbird census data support the prediction that songbird communities in cowbird Removal areas contain a greater proportion of suitable vs. unsuitable host individuals relative to Control areas. However, the size of this effect is small (4.0 - 8.7%) and it was not statistically significant in the second year of the study. Contrary to predictions, I did not observe a significantly greater proportion of suitable hosts in the songbird communities of Control sites >10 km from traps compared to Controls 5-10 km from cowbird traps. I also found no support for the prediction that songbird diversity would increase in cowbird Removal areas relative to Control sites. I conclude that the influence of cowbirds on communities of hosts is generally weak, however I recommend further investigation of the effects of cowbird parasitism on songbird community structure in landscapes supporting greater densities of cowbirds. 26 LITERATURE CITED A m b u e l , B . and S . A . Temple . 1983. Area-dependent changes i n b i r d communit ies and vegetation of southern W i s c o n s i n forests. E c o l o g y 64: 1057-1068. A n d r e n , H . and P. A n g e l s t a m . 1988. E l e v a t e d predation rates as an edge effect i n habitat islands: experimental evidence. E c o l o g y 69: 544-547. Bart , J . and D . S . R o b s o n . 1982. E s t i m a t i n g surv ivorship when the subjects are v is i ted per iodical ly . E c o l o g y 63: 1078-1090. Bender , E . A . , T . J . Case, and M . E . G i l p i n . 1984. Perturbation experiments in c o m m u n i t y ecology: theory and practice. E c o l o g y 65: 1-13. B o h n i n g - G a e s e , K . , M L . Taper, and J . H . B r o w n . 1993. A r e declines i n N o r t h A m e r i c a n insectivorous songbirds due to causes on the breeding range? Conservat ion B i o l o g y 7: 76-86. B r a d e n , G . T . , R . L . M c k e r n a n , and S . M . P o w e l l . 1997. Effects o f nest parasi t ism by the Brown-headed C o w b i r d on nesting success o f the C a l i f o r n i a Gnatcatcher. C o n d o r 99: 858-865. B r a w n , J . D . and S . K . R o b i n s o n . 1996. Source-s ink populat ion dynamics may complicate the interpretation of long-term census data. E c o l o g y 7 7 : 3 - 1 2 . B r e w e r , R. , G . A . M c P e e k and R . J . A d a m s Jr. 1991. T h e A t l a s o f B r e e d i n g B i r d s o f M i c h i g a n . M i c h i g a n State U n i v e r s i t y Press, East L a n s i n g . B r i t t i n g h a m , M . C . and S . A . T e m p l e . 1983. H a v e cowbirds caused forest songbirds to decline? B i o s c i e n c e 3 3 : 3 1 - 3 5 . B r o o k s , J . L . and S.I. D o d s o n . 1965. Predat ion, b o d y size, and c o m p o s i t i o n of p lankton. Science 150:28-35 . C o d y , M . L . 1981. Habitat selection i n birds: the roles o f vegetation structure, competitors, and productivity . B i o s c i e n c e 31: 107-113. 27 Cote , I . M . and W . J . Sutherland. 1997. T h e effectiveness o f r e m o v i n g predators to protect b i r d populations. Conservat ion B i o l o g y 11 :395-405. D e C a p i t a , M . E . in press. B r o w n - h e a d e d c o w b i r d control on Kir t land 's W a r b l e r nesting areas i n M i c h i g a n , 1972 - 1975. in T h e b i o l o g y and management o f cowbirds and their hosts. J . N . M . S m i t h , T . L . C o o k , S.I. Rothste in , S . K . R o b i n s o n and S . G . Sealy , eds. U n i v e r s i t y o f Texas Press, A u s t i n , Texas. D e l o r i a , C M . and M . E . D e C a p i t a . 1997. C o n t r o l o f brown-headed cowbirds on Kir t land 's W a r b l e r nesting areas i n Northern M i c h i g a n . U . S . F i s h and W i l d l i f e Service Report. D o n o v a n , T . M . , F . R . T h o m p s o n III, J . Faaborg , and J . R . Probst. 1995. R e p r o d u c t i v e success of migratory birds i n habitat sources and sinks. Conservat ion B i o l o g y 9: 1380-1395. F r i e d m a n n , H . 1963. H o s t relations o f the parasitic cowbirds . U S N a t i o n a l M u s e u m B u l l e t i n N o . 233. F r i e d m a n n , H , L . F . K i f f , and S.I. Rothstein. 1977. A further contr ibut ion to the knowledge of the host relations o f the parasitic cowbirds . Smithsonian Contr ibut ions to Z o o l o g y 235: 1-75. G r i f f i t h , J . T . and J . C . Gr i f f i th , in press. C o w b i r d control and the endangered Least B e l l ' s V i r e o : a mangement success story, in T h e b io logy and management o f cowbirds and their hosts. J . N . M . S m i t h , T . L . C o o k , S.I. Rothste in , S . K . R o b i n s o n and S . G . Sealy , eds. U n i v e r s i t y o f Texas Press, A u s t i n , Texas. G r z y b o w s k i , J . A . and C M . Pease, in press a. C o m p a r i n g the relative effects o f b r o o d parasit ism and nest predation on seasonal fecundity in passerine birds, in T h e b io logy and management o f cowbirds and their hosts. J . N . M . S m i t h , T . L . C o o k , S.I. Rothstein, S . K . R o b i n s o n and S . G . Sealy, eds. U n i v e r s i t y o f Texas Press, A u s t i n , Texas. G r z y b o w s k i , J . A . and C M . Pease, in press b. A model o f the dynamics o f cowbirds and their host communit ies . A u k . H a l l , D . J . , W . E . Cooper , and E . E . W e r n e r . 1970. A n experimental approach to the production dynamics and structure o f freshwater animal communit ies . L i m n o l o g y and Oceanography 15: 839-928. 28 Haney, J.C., D.S. Lee, and M . Walsh-McGehee. 1998. A quantitative analysis of winter distribution and habitats of Kirtland's Warblers in the Bahamas [review]. Condor 100: 201-217. Hayden, T.J., D.J . Tazik and R . H . Melton, in press. Cowbird control program on Fort Hood, Texas: Lessons for mitigation of cowbird parasitism on a landscape scale, in The biology and management of cowbirds and their hosts. J . N . M . Smith, T . L . Cook, S.I. Rothstein, S.K. Robinson and S.G. Sealy, eds. University of Texas Press, Austin, Texas. Holmes, R.T., T .W. Sherry, P.P. Marra, and K . E . Petit. 1992. Multiple brooding and productivity of a neotropical migrant the Black-throated Blue Warbler Dendroica caerulescens in an unfragmented temperate forest. Auk 109: 321-333. Hurlbert, S.H. 1984. Pseudoreplication and the design of ecological field experiments. Ecological Monographs 54: 187-211. Hurlbert, S.H. 1997. Functional importance vs keystoneness: reformulating some questions in theoretical biocenology. Australian Journal of Ecology 22:369-382. Hurlbert, S.H., J. Zedler, and D . Fairbanks. 1972. Ecosystem alteration by mosquitofish (Gambusia affinis) predation. Science 175:639-641. Jackson, D . A . 1993. Stopping rules in Principle Components Analysis: a comparison of heuristical and statistical approaches. Ecology 74: 2204-2214. James, F .C . 1971. Ordinations of habitat relationships among breeding birds. Wilson Bulletin 83: 215-236. James, F .C . and C . E . McCul loch . 1995. The strength of inferences about causes of trends in populations, pp. 40-51 in Ecology and Management of Neotropical Migratory Birds. T .E. Martin and D . M . Finch, eds. Oxford University Press, Oxford. Karr, J.R. and R.R. Roth., 1971. Vegetation structure and avian diversity in several New World areas. American Naturalist 105:423-435. Krebs, C.J . 1989. Ecological Methodology. Harper & Row Publishers, New York. 29 K u e h l , R . O . 1994. Repeated measures designs, pp. 499-528 in Statistical Pr inciples o f Research D e s i g n and A n a l y s i s . D a x b u r y Press, B e l m o n t , C a l i f o r n i a . Legendre, L . and P . Legendre. 1983. N u m e r i c a l E c o l o g y . E l s e v i e r Scient i f ic P u b l i s h i n g C o m p a n y , A m s t e r d a m . L o w t h e r , P . E . 1993. B r o w n - h e a d e d C o w b i r d (Molothrus ater). in T h e B i r d s o f N o r t h A m e r i c a , N o . 47. A . Poole and F . G i l l , eds. T h e A c a d e m y of Natura l Sciences, Phi ladelphia and T h e A m e r i c a n Ornithologists ' U n i o n , Washington, D . C . M a c A r t h u r , R . H . and J . W . M a c A r t h u r . 1961. O n b i r d species diversity. E c o l o g y 4 2 : 5 9 4 -598. M a c A r t h u r , R . H . , J . W . M a c A r t h u r , and J . Preer. 1962. O n b i r d species diversity II. Predict ion o f b i r d census f rom habitat measurement. A m e r i c a n Natural ist 96: 167-174. M a r t i n , T . E . 1988a. Habitat and area effects on forest b i r d assemblages: is nest predation an influence? E c o l o g y 69: 74-84. M a r t i n , T . E . 1988b. Processes organiz ing open-nesting b i r d assemblages: is nest predation an influence? Evolut ionary E c o l o g y 2: 37-50. M a r v i l , R . E . and A . C r u z . 1989. Impact o f B r o w n - h e a d e d C o w b i r d parasit ism on the reproductive success o f the Sol i tary V i r e o . A u k 106: 476-480. M a y , R . M . and S . K . R o b i n s o n . 1985. Populat ion dynamics o f avian brood parasit ism. A m e r i c a n Natural ist 126: 475-494. M a y f i e l d , H . 1972. T h i r d decennial census o f the Kir t land 's W a r b l e r . A u k 89: 263-268. M a y f i e l d , H . 1977. Brown-headed C o w b i r d : agent o f extermination? A m e r i c a n B i r d s 31: 107-113. M i l e s , R . K . and D . A . Buehler . in press. A n evaluation o f point-count and playback techniques for censusing B r o w n - h e a d e d C o w b i r d s . in T h e b io logy and management o f cowbirds and their hosts. J . N . M . S m i t h , T . L . C o o k , S.I. Rothste in , S . K . R o b i n s o n and S . G . Sealy, eds. U n i v e r s i t y o f Texas Press, A u s t i n , Texas. 30 Newton, I. 1994. Experiments on the limitation of bird breeding densities: a review. Ibis 136: 397-411. Paine, R.T. 1974. Intertidal community structure. Experimental studies on the relationship between a dominant competitor and its principal predator. Oecologia 15: 93-120. Pease, C M . and J.A. Grzybowski. 1995. Assessing the consequences of brood parasitism and nest predation on seasonal fecundity in passerine birds. Auk 112: 343-363. Peterjohn, B . G . , J.R. Sauer and S. Schwartz, in press. Temporal and geographic patterns in population trends in brown-headed cowbirds. in The biology and management of cowbirds and their hosts. J . N . M . Smith, T . L . Cook, S.I. Rothstein, S.K. Robinson and S.G. Sealy, eds. University of Texas Press, Austin, Texas. Power, M . E . , D . Tilman, J .A. Estes, B . A . Menge, W.J . Bond, L . S . M i l l s , G . Daily, J.C. Castilla, J. Lubchenco, and R.T. Paine. 1996. Challenges in the quest for keystones. Bioscience 46: 609-620. Probst, J.R. 1986. A review of factors limiting the Kirtland's Warbler on its breeding grounds. American Midland Naturalist 116: 87-100. Probst, J.R., D.S. Rakstad, and D.J . Rugg. 1992. Breeding bird communities in regenerating and mature broadleaf forests in the U S A Lake States. Forest Ecology and Management 49: 43-60. Pulliam, H.R. 1988. Sources, sinks and population regulation. American Naturalist 132: 652-661. Pulliam, H.R. and B.J . Danielson. 1991. Sources, sinks and habitat selection: a landscape perspective on population dynamics. American Naturalist 137 (supplement): S50-S66. Risch, S.J. and C R . Carroll. 1982. Effect of a keystone predaceous ant, Solenopsis geminata, on arthropods in a tropical agroecosystem. Ecology 63: 1979-1983. Robinson, S.K., J.P. Hoover and J.R. Herkert. in press. Cowbird parasitism in a fragmented landscape: effects of tract size, habitat and abundance of cowbirds and hosts, in The biology and management of cowbirds and their hosts. J . N . M . Smith, T . L . Cook, S.I. Rothstein, S.K. Robinson and S.G. Sealy, eds. University of Texas Press, Austin, Texas. 31 R o h w e r , S. and C D . Spaw. 1988. E v o l u t i o n a r y lag versus b i l l - s i ze constraints: a comparative study of the acceptance of c o w b i r d eggs by o l d hosts. E v o l u t i o n a r y E c o l o g y : 27-36. R o m i g , G . P . and R . D . C r a w f o r d . 1996. C l a y - c o l o r e d Sparrows i n N o r t h D a k o t a parasit ized by B r o w n - h e a d e d C o w b i r d s . Pra ir ie Natural ist 27: 193-203. Rotenberry, J . T . , R . J . Cooper , J . M . W u n d e r l e and K . G . S m i t h . 1995. W h e n and h o w are populations l imited? The roles o f insect outbreaks, f ire, and other natural perturbations, pp. 55-84 in E c o l o g y and M a n a g e m e n t o f N e o t r o p i c a l M i g r a t o r y B i r d s . T . E . M a r t i n and D . M . F i n c h , eds. O x f o r d U n i v e r s i t y Press, O x f o r d . Rothstein, S.I. 1975. A n experimental and te leonomic investigation o f avian brood parasit ism. C o n d o r 77: 250-271 . Rothste in, S.I. 1994. T h e cowbird's invas ion o f the far west: history, causes and consequences experienced by host species. Studies i n A v i a n B i o l o g y 15: 301-315. Rothste in , S.I. and T . L . C o o k , in press. Sect ion introduct ion: C o w b i r d management, host populat ion l imitat ion, and efforts to save endangered species, in T h e b io logy and management o f cowbirds and their hosts. J . N . - M . S m i t h , T . L . C o o k , S.I. Rothste in, S . K . R o b i n s o n and S . G . Sealy, eds. U n i v e r s i t y of Texas Press, A u s t i n , Texas. Rothste in, S.I., J . Verner , and E . Stevens. 1984. R a d i o - t r a c k i n g conf irms a unique diurnal pattern o f spatial occurence i n the parasitic brown-headed c o w b i r d . E c o l o g y 65: 77-88. R y e l , L . A . 1981. Populat ion change i n the Kir t land 's W a r b l e r . J a c k - P i n e W a r b l e r 5 9 : 7 7 - 9 0 . S e d g w i c k , J . A . and W . M . Iko. in press. Costs o f B r o w n - h e a d e d C o w b i r d parasit ism to W i l l o w Flycatchers: an overview. Studies in A v i a n B i o l o g y . S m i t h , J . N . M . in press. T h e basis for c o w b i r d management: host selection, impacts on hosts, and criteria for taking management action. Studies i n A v i a n B i o l o g y . Strausberger, B . M . and M . V . A s h l e y . 1997. C o m m u n i t y - w i d e patterns o f parasit ism o f a host "generalist" brood-parasitic c o w b i r d . O e c o l o g i a 112:254-262. 32 Stutchbury, B . J . M . 1997. Effects o f female c o w b i r d removal on reproductive success o f hooded warblers. W i l s o n B u l l e t i n 1 0 9 : 7 4 - 8 1 . Terborgh, J . 1989. W h e r e have a l l the birds gone? Princeton U n i v e r s i t y Press, Pr inceton, N e w Jersey. T h o m p s o n , F . R . I l l 1994. T e m p o r a l and spatial patterns o f breeding B r o w n - h e a d e d C o w b i r d s i n the midwestern U n i t e d States. A u k 111: 979-990. W a l k i n s h a w , L . H . 1972. Kir t land 's W a r b l e r - endangered. A m e r i c a n B i r d s 2 6 : 3 - 9 . W a l k i n s h a w , L . H . 1983. Kirt land's W a r b l e r : the natural history o f an endangered species. Cranbrook Institute o f Science, B l o o m f i e l d H i l l s , M i c h i g a n . W h i t f i e l d , M J . in press. Results o f a B r o w n - h e a d e d C o w b i r d control program for the Southwestern-Wi l low Flycatcher, in T h e b io logy and management o f cowbirds and their hosts. J . N . M . S m i t h , T . L . C o o k , S.I. Rothste in, S . K . R o b i n s o n and S . G . Sealy, eds. U n i v e r s i t y o f Texas Press, A u s t i n , Texas. W i e n s , J . A . 1963. Aspects o f c o w b i r d parasit ism in Southern O k l a h o m a . W i l s o n B u l l e t i n 7 5 : . 130-139. W i e n s , J . A . 1989a. B e y o n d competi t ion: other factors inf luencing c o m m u n i t y structure, pp. 89-140 in T h e ecology of b i r d communit ies V o l u m e 2 Processes and variations. Cambridge Univers i ty Press, Cambridge , E n g l a n d . W i e n s , J . A . 1989b. T h e ecology of b i r d communit ies V o l u m e 2 Processes and variations. Cambridge U n i v e r s i t y Press, Cambridge , E n g l a n d . W i l c o v e , D . S . 1985. Nest predation in forest tracts and the decl ine o f migratory songbirds. E c o l o g y 66: 1211-1214. W i l l s o n , M . F . 1974. A v i a n c o m m u n i t y organization and habitat structure. E c o l o g y 55: 1017-1029. 33 T A B L E 1. B r i l l o u i n ' s index of species diversity for 1996 and 1997 R e m o v a l sites, C o n t r o l sites 5-10 k m f r o m traps and C o n t r o l sites >10 k m f r o m c o w b i r d traps. 1996 1997 Removals Contro ls R e m o v a l s Contro ls C o n t r o l s 5-10 k m 5-10 k m > 1 0 k m B r o u i l l o u i n ' s diversity (H) 4.377 4.334 4.623 4.583 4.576 -bits/individual Evenness 0.791 0 .880 0.811 0.798 0 .802 34 T A B L E 2. M e a n number o f cowbirds detected i n c o w b i r d R e m o v a l sites, C o n t r o l sites 5-10 k m f r o m c o w b i r d traps and C o n t r o l sites >10 k m f r o m c o w b i r d traps. 1996 1997 8 m i n point counts 10 m i n point counts 5 m i n playback females males females males females males R E M O V A L M e a n 0 0.025 0 0.063 0.021 0.028 S I T E S S E 0 0.018 0 0.034 0.015 0.015 C O N T R O L M e a n 0 0.167 0.069 0.326 0.056 0.257 S I T E S 5-10 k m S E 0 0.069 0.025 0.062 0.021 0.035 C O N T R O L M e a n _ _ 0.174 0.583 0.222 0.576 S I T E S > 1 0 k m S E - - 0.062 0.081 0.047 0.103 35 T A B L E 3. Tests for differences between 1996 and 1997 m a x i m u m l i k e l i h o o d estimates o f dai ly nest survival in c o w b i r d R e m o v a l sites and C o n t r o l sites 5-10 k m f r o m c o w b i r d traps. N u m b e r o f nests i n brackets. D a i l y nest P survival S E of estimate Z (one-tailed rate test) 1996 1997 1996" 1997 Removal sites Controls 5-10 km suitable 0.953 0.959 0.019 0.010 0.255 0.397 hosts (12) (29) unsuitable 0.974 0.982 0.018 0.010 0.368 0.386 hosts (4) (11) suitable 0.987 0.970 0.009 0.009 1.276 0.100 hosts (11) (24) unsuitable 0.974 0.986 0.018 0.009 0.603 0.274 hosts (6) (10) T A B L E 4. Incidence o f parasit ism among suitable hosts i n c o w b i r d R e m o v a l sites, C o n t r o l sites 5-10 k m and C o n t r o l sites >10 k m f r o m c o w b i r d traps. R e m o v a l sites # of suitable host nests parasitized % parasit ized nests 1996 0 / 1 2 0 % 1997 0 / 2 9 0 % ' 9 6 & '97 combined 0 / 4 1 0 % Contro ls 5-10 k m # of suitable host nests parasitized % parasit ized nests 0 / 1 1 0 % 6 / 2 4 2 5 % 6 / 3 5 17.1% C o n t r o l s > 1 0 k m # of suitable host nests parasitized % parasit ized nests 5 / 2 0 2 5 % F i g . 1: M a p of M i c h i g a n (modif ied f r o m B r e w e r et al . 1991) s h o w i n g location o f study. F i g . 2: Schematic o f transect lines through a study area w i t h randomly chosen point count locations. 70 68 66 64 62 60 58 + 56 54 52 50 Removal sites Controls 5-10 km F i g . 3: S o n g b i r d community composi t ion in 1996 at ten c o w b i r d R e m o v a l sites and ten C o n t r o l sites 5-10 k m f r o m c o w b i r d traps 4 0 F i g . 4: Songbird c o m m u n i t y composi t ion in 1997 at eight c o w b i r d R e m o v a l sites, eight C o n t r o l sites 5-10 k m and eight C o n t r o l sites >10 k m f r o m c o w b i r d traps. 41 • Removal sites 0 Controls 5-10 km Rejecters Cavity Unsuitable Corvias Nesters nestling diet F i g . 5: M e a n numbers o f unsuitable hosts i n 1996 detected dur ing eight minute point counts i n ten c o w b i r d R e m o v a l sites and ten C o n t r o l sites 5-10 k m f r o m c o w b i r d traps. 42 Rejecters Cavity Unsuitable Corvids Nesters nestling F i g . 6: M e a n numbers o f unsuitable hosts in 1997 detected dur ing ten minute point counts in eight c o w b i r d R e m o v a l sites, eight C o n t r o l sites 5-10 k m and eight C o n t r o l sites >10 k m f r o m c o w b i r d traps. 43 c F i g . 7: P lot o f species heterogeneity (richness and evenness) i n 1996 at ten R e m o v a l sites and ten C o n t r o l sites 5-10 k m f r o m c o w b i r d traps. 4 4 Species in rank order F i g . 8: P lot o f species heterogeneity (richness and evenness) i n 1997 at eight R e m o v a l sites, eight C o n t r o l sites 5-10 k m and eight C o n t r o l sites >10 k m f r o m c o w b i r d traps. 45 F i g . 9: M a x i m u m l i k e l i h o o d estimates o f dai ly nest survival for pooled 1996 and 1997 nests o f suitable and unsuitable hosts. 4 6 30 25 20 15 10 5 JO— 1M, LJ I tBL-• Removal sites 0 Controls 5-10 km 16 E a 5 x o CNI .£ 4 •D O 1 3 u u) 2 -f r 1-£i E i o d i 11 J j • Removal sites 0 Controls 5-10 km n | l~l | • i n _ 16 14 12 10 4 + 2 + i • Removal sites 0 Controls 5-10 km m r-~ in csi V in csi V c csi hawthc o sbe hawthc hawtt servic< i in cm) Tree species and size classes (dbh i  Fig. 10: Mean numbers of trees counted in 20 x 20 m plots (2 plots/site) in ten Removal sites and ten Control sites 5-10 km from cowbird traps in 1996. See Appendix Table 1 for species names. 47 o vt cr ° OJ Q -E i s •2 « 40 35 30 25 20 15 10 5 0 J 1- V I" T-• Removal sites 0 Controls 5-10 km • Controls >10 km 25 E ! = 20 15 10 I ™ 11 C\l A CO i as n H h " 1 — H - — H eg JL • Removal sites 0 Controls 5-10 km • Controls >10 km A 0) Q. E IO CM i n V V ±- A a> o sp CO CU uS ^ .c io x: co •a CD co X3 CD Tree species and size class (dbh in cm) Fig. 11: Mean numbers of trees counted in 20 x 20 m plots (6 plots/site) in eight Removal sites, eight Control sites 5-10 km and eight Control sites >10 km from cowbird traps in 1997. See Appendix Table 1 for species names. 48 F i g . 12: M e a n percentage of each ground cover type i n 1 x 1 m plots (2 plots/site) i n ten R e m o v a l sites and ten C o n t r o l sites 5-10 k m f r o m c o w b i r d traps. See A p p e n d i x T a b l e 1 for species names. 49 Fig. 13: Mean percentage of each ground cover type in 1 x 1 m plots (24 plots/site) in eight Removal sites, eight Controls 5-10 km and eight Controls >10 km from cowbird traps in 1997. See Appendix Table 1 for species names. 50 F i g . 14: M e a n density o f vegetation measured at two plots per site at ten R e m o v a l sites and ten C o n t r o l sites 5-10 k m f r o m c o w b i r d traps i n 1996. 51 F i g . 15: M e a n density of vegetation measured at twelve plots per site at eight R e m o v a l sites, eight C o n t r o l sites 5-10 k m and eight C o n t r o l sites >10 k m f r o m c o w b i r d traps i n 1997. 52 APPENDIX T A B L E 1. Trees, shrubs and ground vegetation found in study plots. 53 TREES AND SHRUBS Jack pine (Pinus banksiana) Red pine (Pinus resinosa) Eastern white Pine (Pinus strobus) Northern pin oak (Quercus ellipsoidalis). Pin cherry (Prunus pensylvanica) Trembling aspen (Populus tremuloides) Large-toothed aspen (Populus grandidentata) Willow (Salix spp.) White spruce (Picea glauca) Balsam fir (Abies balsamea) Northern white cedar (Thuja occidentalis) Larch (Larix laricina) Red maple (Acer rubrum) Black alder (Alnus glutinosa) White birch (Betula papyrifera) Hawthorn (Crataegus spp.) Serviceberry (Amelanchier arborea) Witch Hazel (Hamamelis virginiana) GROUND COVER Blueberry (Vaccinium angustifolium) Common bearberry (Arctostaphylos uva-ursi) Wintergreen (Gaultheria procumbens) Hairgrass (Deschampsia flexuosa) Big bluestem grass (Andropogon gerardii) Sedge (Carex pensylvanica) Sweet Fern (Comptonia peregrina) Bracken Fern (Gaultheria procumbens) Reindeer Moss (Cladonia mitis) 54 T A B L E 2. Mean number of suitable host individuals detected at ten Removal and ten Control sites 5-10 km from cowbird traps in 1996. Removal Control 5-10 km Suitable Kirtland's Warbler (Dendroica kirtlandii) 4.5 0 (33 spp.) Yellow-Rumped Warbler {Dendroica coronata) 29 16 Nashville Warbler (Vermivora ruficapilla) 97 74.5 Pine Warbler (Dendroica pinus) 0.5 1 Ovenbird (Seiurus aurocapillus) 44.5 61.5 American Redstart (Setophaga ruticilla) 0.5 1 Common Yellowthroat (Geothlypis trichas) 5 1 Black-and-white Warbler (Mniotilta varia) 0.5 0.5 Black-throated Green Warbler (Dendroica virens) 0.5 0.5 Red-eyed Vireo (Vireo olivaceus) 7 18.5 Eastern Wood-pewee (Contopus virens) 11.5 22.5 Eastern Phoebe (Sayornis phoebe) 1 0 Olive-sided Flycatcher (Contopus borealis) 0.5 0.5 Least Flycatcher (Empidonax minimus) 0 2.5 Chipping Sparrow (Spizella passerina) 63.5 62.5 Vesper Sparrow (Pooecetes gramineus) 27 17.5 Lincoln Sparrow (Melospiza lincolnii) 14.5 11.5 Clay-coloured Sparrow (Spizella pallida) 2.5 1 White-throated Sparrow (Zonotrichia albicollis) 10.5 8 Song Sparrow (Melospiza melodia) 5.5 0 Field Sparrow (Spizella pusilla) 31 17.5 Swamp Sparrow (Melospiza georgiana) 2.5 0.5 Savannah Sparrow (Passerculus sandwichensis) 0.5 0 Dark-eyed Junco (Junco hyemalis) 17.5 13.5 Eastern Towhee (Pipilo erythrophthalmus) 13.5 6 Indigo Bunting (Passerina cyanea) 7.5 7 Rose-breasted Grosbeak (Pheucticus ludovicianus) 6.5 10 Hermit Thrush (Catharus guttatus) 131 118 Veery (Catharus fuscescens) 0 2.5 Red-winged Blackbird (Agelaius phoeniceus) 13.5 1 Common Grackle (Quiscalus quiscula) 2 0 Scarlet Tanager (Piranga olivacea) 2 6 T O T A L 553 482.5 55 T A B L E 3. Mean number of unsuitable host individuals detected at ten Removal and ten Control sites 5-10 km from cowbird traps in 1996. Removal Control 5-10 km Unsuitable Great Crested Flycatcher (Myiarchus crinitus ) 6.5 12 (18 spp.) Eastern Kingbird (Tyrannus tyrannus) 3 2 Brown Thrasher {Toxostoma rufum) 24 10 American Robin (Tardus migratorius) 15 28.5 American Goldfinch (Carduelis tristis) 10.5 15 Mourning Dove (Zenaida macroura) 33.5 36 Cedar Waxwing (Bombycilla cedrorum) 8.5 19 White-breasted Nuthatch (Sitta carolinensis) 3 5.5 Red-breasted Nuthatch (Sitta canadensis) 4.5 7 Winter Wren (Troglodytes troglodytes) 0.5 1.5 Eastern Bluebird (Sialia sialis) 2 3.5 Black-capped Chickadee (Parus atricapillus) 37 74.5 Brown Creeper (Certhia americana) 3.5 0 Blue Jay (Cyanocitta cristata) 79 68 Tree Swallow (Tachycineta bicolor) 1.5 1.5 Common Raven (Corvus corax) 12.5 9.5 American Crow (Corvus brachyrhyncos) 32 58.5 Baltimore Oriole (Icterus galbula) 1 0.5 TOTAL 279.5 356.5 56 T A B L E 4. M e a n number of suitable host indiv iduals detected at eight R e m o v a l sites, eight C o n t r o l sites 5-10 k m and eight C o n t r o l sites >10 k m f r o m c o w b i r d traps i n 1997. Removal Control 5-10 km Control >10km Suitable Kirtland's Warbler (Dendroica kirtlandii) 12.7 1.7 2.0 (41 spp.) Yellow-Rumped Warbler (Dendroica coronata) 37.3 34.3 31.3 Nashville Warbler (Vermivora ruficapilla) 97.0 88.7 74.0 Pine Warbler (Dendroica pinus) 1.7 1.0 4.3 Ovenbird (Seiurus aurocapillus) 46.3 63.0 70.7 American Redstart (Setophaga ruticilla) 0.7 1.7 0.0 Common Yellowthroat (Geothlypis trichas) 15.3 4.7 14.7 Black-and-white Warbler (Mniotilta varia) 2.0 2.3 2.0 Black-throated Green Warbler (Dendroica virens) 1.7 0.3 0.0 Chestnut-sided Warbler (Dendroica pensylvanica) 0.3 1.3 0.7 Black-throated Blue Warbler (Dendroica caerulescens) 0.3 0.0 0.0 Canada Warbler (Wilsonia canadensis) 0.0 1.7 0.0 Northern Waterthrush (Seiurus noveboracensis) 0.0 3.3 0.0 Mourning Warbler (Oporornis Philadelphia) 0.0 0.3 0.0 Red-eyed Vireo (Vireo olivaceus) 2.0 7.3 8.0 Solitary Vireo (Vireo solitarius) 0.3 1.3 0.3 Eastern Wood-pewee (Contopus virens) 6.7 6.3 15.7 Eastern Phoebe (Sayornis phoebe) 1.7 0.7 0.7 Olive-sided Flycatcher (Contopus borealis) 2.0 0.0 0.0 Least Flycatcher (Empidonax minimus) 0.3 2.3 0.7 Alder Flycatcher (Empidonax alnorum) 0.0 1.7 0.3 Chipping Sparrow (Spizella passerina) 62.0 46.7 58.3 Vesper Sparrow (Pooecetes gramineus) 25.0 27.3 24.7 Lincoln Sparrow (Melospiza lincolnii) 34.0 12.3 9.0 Clay-coloured Sparrow (Spizella pallida) 1.3 3.7 1.7 White-throated Sparrow (Zonotrichia albicollis) 19.0 12.0 4.3 Song Sparrow (Melospiza melodia) 4.7 4.0 6.3 Field Sparrow (Spizella pusilla) 36.7 26.3 35.0 Swamp Sparrow (Melospiza georgiana) 5.3 0.7 0.7 Dark-eyed Junco (Junco hyemalis) 18.0 6.7 10.7 Eastern Towhee (Pipilo erythrophthalmus) 9.0 11.3 16.3 Indigo Bunting (Passerina cyanea) 2.7 5.7 5.3 Rose-breasted Grosbeak (Pheucticus ludovicianus) 5.3 18.3 4.0 Northern Cardinal (Cardinalis cardinalis) 0.0 0.0 0.3 Hermit Thrush (Catharus guttatus) 111.7 96.3 103.0 Veery (Catharus fuscescens) 0.3 2.7 1.0 Wood Thrush (Hylocichla mustelina) 0.0 0.0 0.3 Red-winged Blackbird (Agelaius phoeniceus) 34.3 10.0 9.0 Brewer's Blackbird (Euphagus cyanocephalus) 0.3 0.0 0.7 Common Grackle (Quiscalus quisculd) 3.0 11.0 8.3 Scarlet Tanager (Piranga olivacea) 1.3 3.3 8.3 TOTAL 603 522.3 532.7 57 T A B L E 5. Mean number of unsuitable host individuals detected at eight Removal sites, eight Control sites 5-10 km and eight Control sites >10 km from cowbird traps in 1997. Removal Control 5-10 km Control >10km Unsuitable Great Crested Flycatcher (Myiarchus crinitus ) 8.7 7.3 21.0 (20 spp.) Eastern Kingbird (Tyrannus tyrannus) 5.0 2.3 1.0 Gray Catbird (Dumetella carolinensis) 0.0 0.3 0.0 Brown Thrasher (Toxostoma rufum) 30.7 19.7 20.3 American Robin (Turdus migratorius) 22.3 35.3 30.3 American Goldfinch (Carduelis tristis) 18.7 23.7 23.7 Pine Siskin (Carduelis pinus) 0.0 0.0 0.3 Mourning Dove (Zenaida macroura) 38.0 33.3 50.7 Cedar Waxwing (Bombycilla cedrorum) 15.7 15.3 25.0 White-breasted Nuthatch (Sitta carolinensis) 1.0 2.3 2.3 Red-breasted Nuthatch (Sitta canadensis) 12.0 3.7 4.0 Winter Wren (Troglodytes troglodytes) 0.0 2.7 0.3 Eastern Bluebird (Sialia sialis) 4.7 0.7 3.3 Black-capped Chickadee (Parus atricapillus) 47.3 52.7 46.0 Brown Creeper (Certhia americana) 1.0 0.3 0.7 Blue Jay (Cyanocitta cristata) 81.3 91.3 79.3 Tree Swallow (Tachycineta bicolor) 9.7 3.3 6.7 Common Raven (Corvus corax) 9.7 7.3 6.7 American Crow (Corvus brachyrhyncos) 27.7 46.7 58.3 Baltimore Oriole (Icterus galbula): 0.3 1.0 0.3 T O T A L 342 356.3 386.3 58 F i g . 1: M e a n detection curves for point count observers, representing cumulat ive proport ion of total songbird detections over t ime. 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.831.1-0088888/manifest

Comment

Related Items