TWEL Andrea Lindsay Thesis

Seasonal events and associated carry-over effects in a neotropical migratory songbird, the yellow warbler
Andrea M. Lindsay, MS
Advisor: Paul Rodewald
The ability to determine how events during a bird’s life cycle are linked and to identify ecological factors that limit populations has received much recent attention due to declining populations of many species of migratory songbirds. Ecological events occurring during any stage of a migratory bird’s life cycle have the potential to carry-over and affect an individual during a subsequent season. As a result, evidence of relationships between breeding and non-breeding (i.e., winter and migration) events has become increasingly important in advancing conservation efforts for migratory songbirds.
The quality of habitat used by songbirds during the winter has been shown to have important consequences for migratory birds. For example, the quality of winter territories (as measured by moisture and resource availability) held by American Redstarts (Setophaga ruticilla) determines physical condition in wintering areas and timing of migratory departure (Marra and Holmes 2001), as well as arrival date and physical condition upon reaching the breeding grounds (Webster et al. 2002; Hobson 2005b).
Male plumage coloration and ornamentation functions in territoriality, dominance, and aggression in male-male interactions, and signals attractiveness, individual quality, and reproductive strategy, status, and potential to females. Birds wintering in higher quality habitat should be more able to allocate resources to the formation of intensely colored alternate feathers grown in wintering areas (Saks et al. 2003), which in turn should allow males to more effectively attract females and increase their breeding success (Hill et al. 1998).
Stable-carbon isotope analysis measures the ratio of 13C to 12C (δ13C) in plant and animal tissues, and can reflect local vegetation types, diet, and habitat quality and usage where these tissues are grown (Hobson et al. 2004). Values of δ13C are transferred up the food web and incorporated into growing tissues, so that analysis of inert tissues (e.g., feathers) can provide information about habitat conditions during previous seasons. Stable-carbon isotope ratios in feathers grown in Neotropical wintering areas during prealternate molt will reflect habitat conditions of birds across a mesic to xeric habitat gradient (Marra et al. 1998). Using stable-carbon isotope analysis of alternate feathers, it is possible to examine relationships between winter habitat conditions and events occurring in previous seasons. My research focused on seasonal events and associated carry-over effects in a Neotropical migratory songbird, the Yellow Warbler (Dendroica petechia). My research objectives were to 1) assess winter habitat conditions of breeding Yellow Warblers using stable-carbon isotope analysis of feathers grown in tropical wintering areas, 2) examine direct and indirect relationships among winter conditions, plumage characteristics, molt status, and reproductive parameters in Yellow Warblers, and 3) determine how age, plumage coloration and ornamentation, and nest initiation date are associated with reproductive success.
This study was conducted along the southwestern shore of Lake Erie at the Ottawa National Wildlife Refuge, Ottawa County, Ohio from late April through June of 2007 and 2008. To determine reproductive success of Yellow Warblers, nests (n = 252) were located and monitored in willow and dogwood shrubland within 5 km of the Lake Erie shoreline. I estimated daily survival rates of nests using logistic exposure models, and recorded number of young fledged per adult. Adults males (n = 154) and females (n = 148) from each nest were captured and banded, morphological measurements were taken, and greater coverts replaced during prealternate molt were counted. A series of digital photographs of each adult bird was taken to quantify streaking and yellow coloration on the breast using the computer programs ImageJ and AdobePhotoshop, respectively. The two innermost greater coverts replaced in prealternate molt were collected for stable-carbon isotope analysis. Stable-carbon isotope values ranged from -25.25‰ to -18.10‰, which broadly overlaps with δ13C values reported in other studies of seasonal interactions in migratory birds (e.g., Marra et al. 1998; Bearhop et al. 2004; Norris et al. 2004a), except that approximately 25% of values fell in a more xeric range. I used path analysis to examine direct and indirect effects of winter habitat conditions on reproductive measures. Path analysis indicated that for males, winter habitat conditions had direct positive effects on streaking and molt but no effects on number of young fledged per male. Streaking and fledging date had direct negative effects on number of young fledged. The pathways explained 23% of the variation in number of young fledged from each nest. For females, path analysis indicated that streaking had a direct negative effect and molt had a direct positive effect on first egg date, but winter habitat conditions had no effects on plumage variables or reproductive measures. The pathways for the female model explained only 8% of the variation in number of young fledged from each nest. The path models for both males and females provided a good fit (males: χ2 = 2.52, d.f. = 8, p = 0.962; females: χ2 = 4.90, d.f. = 4, p = 0.297) when observed values were compared to predicted values.
I used an information theoretic approach and multi-model averaging to determine which models were important in explaining variation in daily nest survival rates and number of young fledged per adult for both males and females. I found that age (β = 1.127, SE = 0.562, Relative Importance Value (RVI) = 0.562) was important in explaining variation in daily nest survival for males, but that none of the measured variables was important in explaining variation in daily nest survival for females. Breast streaking (β = -0.0000360, SE = 0.0000151, RVI = 0.790) and hue (β = -0.213, SE = 0.104, RVI = 0.790) were most effective in explaining variation in number of young fledged for males, but none of the measured variables was important in explaining variation in number of young fledged per female, despite high relative variable importance values for hue (RVI = 0.594) and molt (RVI = 0.594).
Previous research has shown that male Yellow Warblers with more breast streaking are more territorial and achieve more extra-pair copulations, whereas males with less breast streaking spend more time caring for young at the nest and are cuckolded more often (Studd and Robertson 1988; Yezerinac and Weatherhead 1997b). I suggest that since males in this study with less breast streaking fledged more young, females may assess streakiness of males and choose a mate based on his perceived potential to fledge young, regardless of who sires the young. Because breast streaking, yellow coloration, and age of males helped explain variation in reproductive measures, males may employ multiple messages in signaling quality to females and to other males (i.e., yellow breast coloration signals age and reproductive experience while breast streaking signals reproductive strategy).
My study indicates that plumage coloration and age may be important factors in predicting reproductive success in male Yellow Warblers, but that these attributes had little apparent impact on reproduction in females. I found neither a direct effect nor strong indirect effects of winter habitat conditions on reproduction in Yellow Warblers. However, I did observe direct effects of winter habitat conditions on plumage characteristics and of plumage characteristics on reproductive measures in males. Seasonal interactions, although moderate, were detected in this population of Yellow Warblers, and this underscores the importance of conserving habitats used throughout the annual life cycle of migratory birds.