STABLE ISOTOPE EVIDENCE LINKS BREEDING GEOGRAPHY AND MIGRATION TIMING IN WOOD WARBLERS (PARULIDAE)

Auk, The, Apr 2006 by Kelly, Jeffrey F

The relationship between timing of migration and breeding origins can be evaluated using extrinsic and intrinsic markers of migrants captured en route (e.g. legbands and stable-hydrogen isotope ratios; Rubenstein and Hobson 2004). Smith et al. (2003) used this approach to evaluate timing of Sharp-shinned Hawk (Accipiter striatus) migration. Among wood warblers, Kelly et al. (2002) used stable-hydrogen isotope ratios to demonstrate that during autumn, Wilson's Warblers (Wilsonia pusilla) from northern latitudes pass through New Mexico earlier, on average, than conspecifics that breed at lower latitudes.

My objective was to determine what patterns of migration timing exist in other Parulidae (hereafter "wood warblers"). In particular, is there a relationship between breeding location and timing of autumnal migratory passage among wood warblers? To address this question, I measured stable-hydrogen isotope ratios of feathers collected from three species of migratory wood warblers captured en route.

METHODS

As part of a songbird-migration banding project, technicians collected feathers from migratory wood warblers from 7 August through 30 October 1998 at the Bosque del Apache National Wildlife Refuge (NWR), New Mexico (33°48'N, 106°52'W). Technicians removed a primary feather from each bird before banding and releasing them. I focused on three species of longdistance migrant wood warblers: Orangecrowned Warbler (Vermivora celata), Yellow Warbler (Oendroica petechia), and Common Yellowthroat (Geothlypis trichas). I expected comparisons among these species to be instructive for several reasons. First, because the species are widely distributed, the stable-hydrogen isotope gradient between southern and northern breeding birds should be large. second, these species were commonly captured in autumn across a broad range of dates (Yong and Finch 2002). Third, because these species are almost exclusively passage migrants at the Bosque del Apache NWR (i.e. neither breeding nor wintering on site), captured birds were migrants rather than residents. Finally, these species represent three different genera within the wood warblers. Consequently, if there is variation in the fundamental form of the relationship of migration timing with breeding origin, we might reasonably expect to see this variation in this sample.

Analysis of stable-hydrogen isotope ratios was done via continuous-flow isotope-ratio mass spectrometry (CFIRMS) using pyrolysis combustion. Feather-sample preparation followed the methods of Kelly et al. (2002). I washed feathers in detergent and thoroughly rinsed them to remove oil, dirt, and residual detergent. Before isotopic analysis, feathers were oven-dried at 100°C to remove water. I removed 0.1-0.2 mg of the distal end of each feather and wrapped them in a silver capsule. All isotope analyses were conducted at the University of New Mexico's Stable Isotope Laboratory. Feathers were loaded into an autosampler from which they were dropped into a high-temperature reduction furnace (Finnigan TC/EA, Thermo Electron Corporation, Waltham, Massachusetts) interfaced through an open split (Finnigan MAT Conflo II) with a mass spectrometer (Finnigan MAT Delta plus XL). The reduction furnace was used to pyrolize feather samples at 1,450°C. Wassenaar and Hobson (2000b) have shown that ~23% of hydrogen in feathers is exchangeable with atmospheric hydrogen. I present δD values of bulk feather tissues, not corrected for the exchangeable fraction of hydrogen contained in the feathers (Wassenaar and Hobson 2000b, 2003). Values of δD that are corrected for exchangeable hydrogen are more useful than uncorrected values; however, data reported here were collected in the winter of 2002, before a method of online comparative equilibration was widely known or available. All samples were air-equilibrated for two weeks prior to sampling, so that differential exchange was not a problem. I express the ratio of stable-hydrogen isotopes (H^sub 2^/H^sub 1^) in a sample as the parts per thousand ([per thousand]) deviation from standard mean ocean water (vSMOW = [per thousand]). We report deviation from these standards in delta notation: δD = ([stable-hydrogen isotope ratio^sub sample^/stable-hydrogen isotope ratio^sub standard^] - 1) × 1,000. δD is in parts per thousand ([per thousand]) deviation from standard mean ocean water (vSMOW) with an analytical precision of ±2.0[per thousand].