2380 Reports Ecology, 86(9), 2005, pp. 2380–2385 q 2005 by the Ecological Society of America NONBREEDING HABITAT OCCUPANCY AND POPULATION PROCESSES: AN UPGRADE EXPERIMENT WITH A MIGRATORY BIRD COLIN E. STUDDS 1,2,3 AND PETER P. MARRA 2 1 Program in Behavior, Ecology, Evolution, and Systematics, University of Maryland, College Park, Maryland 20742 USA 2 Smithsonian Environmental Research Center, P.O. Box 28, Edgewater, Maryland 21037 USA Abstract. Evidence is accumulating that winter habitats occupied by migratory birds produce differences in individual condition that can carry over into subsequent stages of the annual cycle. Despite strong observational evidence, experimental work is needed to strengthen support for this hypothesis. We experimentally upgraded individual American Redstarts (Setophaga ruticilla) from low-quality second-growth scrub habitat to high-quality mangrove forest habitat by permanently removing behaviorally dominant, primarily adult males from mangrove, allowing females and immature males from scrub to colonize vacated territories. Prior to the manipulation, upgraded and control redstarts had stable-carbon isotope values in their blood indicative of scrub habitat occupancy and were comparable in body mass. Relative to control birds that overwintered exclusively in scrub, upgraded redstarts incorporated mangrove isotopic signatures, maintained body mass from winter to spring, departed earlier on spring migration, and returned at a higher rate in the following winter. Furthermore, insect biomass on upgrade territories was significantly greater than on control territories, suggesting food availability as a proximate mechanism underlying gradients of nonbreeding habitat suitability. Findings here demonstrate that winter habitat occupancy can be an important determinant of individual performance in migratory birds. Restricted access to food-rich winter habitats may limit survival of females and immature males, an outcome that could be an important driver of population structure and dynamics. Key words: American Redstart; carryover effects; experimentally upgraded territories; migratory birds; nonbreeding winter habitat; population dynamics; removal experiments; seasonal interactions; Setophaga ruticilla; stable-carbon isotopes. INTRODUCTION Migratory animals move between geographically disparate habitats to meet seasonally changing needs, a process that may produce a suite of pressures on population dynamics. Nearctic–Neotropical migratory passerine birds spend three to four months on their temperate breeding grounds, about one to two months each on spring and fall migration, and six to seven months on their tropical winter quarters. Because of the spatially disjunct nature of these habitats and the difficulty of tracking individuals between seasons, con- tention remains about when population limitation oc- curs in the annual cycle (Rappole and McDonald 1994, Latta and Baltz 1997, Newton 2004). Much of the ev- idence for breeding-season limitation comes from re- search on forest loss and fragmentation, including ef- fects on food supply (Burke and Nol 1998, Zanette et al. 2000), nest predation and parasitism (Robinson et al. 1995, Rodewald and Yahner 2001), and annual sur- vival (Bayne and Hobson 2002). Other evidence sug- gests that migratory birds may be limited by events Manuscript received 19 July 2004; revised 19 January 2005; accepted 14 March 2005. Corresponding Editor: J. R. Walters. 3 Address correspondence to Colin E. Studds, SERC, P.O. Box 28, Contees Wharf Rd., Edgewater, Maryland 21037-0028 USA. E-mail: studdsc@si.edu during migration (Moore et al. 1995, Sillett and Holmes 2002) or by food availability on tropical winter areas (Strong and Sherry 2000, Latta and Faaborg 2002). A more holistic model for examining migratory bird population dynamics has emerged recently with the demonstration of carryover effects (Marra et al. 1998, Gill et al. 2001, Møller et al. 2004, Norris et al. 2004a, b). Carryover effects occur when ecological factors limit individual condition in one part of the annual cycle and thereby alter the timing or capacity of an individual to migrate, reproduce, or survive in a sub- sequent stage (Runge and Marra 2005, Marra et al., in press). Recent work demonstrates that individuals with multiple failed nest attempts, or those that fledge off- spring in late summer, may delay molt until during fall migration (Norris et al. 2004b). However, most studies to date have examined carryover effects originating during the winter portion of the annual cycle. Occu- pancy of poor-quality winter habitats has been shown to adversely affect physical condition during migration (Bearhop et al. 2004), arrival date and condition at breeding sites (Marra et al. 1998, Gill et al. 2001), and reproductive success (Norris et al. 2004a). Elucidating how winter habitats determine individual performance and the process by which they produce carryover ef- fects therefore may be critical for understanding pop- ulation dynamics of migratory birds.