Vol.:(0123456789) 1 3 Oecologia https://doi.org/10.1007/s00442-018-4218-z BEHAVIORAL ECOLOGY – ORIGINAL RESEARCH Individual specialization in a migratory grazer refects long-term diet selectivity on a foraging ground: implications for isotope-based tracking Jordan A. Thomson 1,2  · Elizabeth R. Whitman 2  · Maria I. Garcia-Rojas 1  · Alecia Bellgrove 1  · Merrick Ekins 3  · Graeme C. Hays 1  · Michael R. Heithaus 2 Received: 17 December 2017 / Accepted: 30 June 2018 © Springer-Verlag GmbH Germany, part of Springer Nature 2018 Abstract Stable isotope analysis (SIA) can be a useful tool for tracking the long-distance movements of migratory taxa. However, local- scale sources of isotopic variation, such as differences in habitat use or foraging patterns, may complicate these efforts. Few studies have evaluated the implications of local-scale foraging specializations for broad-scale isotope-based tracking. Here, we use > 300 h of animal-borne video footage from green turtles (Chelonia mydas) paired with SIA of multiple tissues, as well as fine-scale Fastloc-GPS satellite tracking, to show that dietary specialization at a single foraging location (Shark Bay, Western Australia) drives a high level of among-individual δ 13 C variability (δ 13 C range = 13.2‰). Green turtles in Shark Bay were highly omnivorous and fed selectively, with individuals specializing on different mixtures of seagrasses, macroal- gae and invertebrates. Furthermore, green turtle skin δ 13 C and δ 15 N dispersion within this feeding area (total isotopic niche area = 41.6) was comparable to that from a well-studied rookery at Tortuguero, Costa Rica, where isotopic dispersion (total isotopic niche area = 44.9) is known to result from large-scale (> 1500 km) differences in foraging site selection. Thus, we provide an important reminder that two different behavioral dynamics, operating at very different spatial scales, can produce similar levels of isotopic variability. We urge an added degree of caution when interpreting isotope data for migratory species with complex foraging strategies. For green turtles specifically, a greater appreciation of trophic complexity is needed to bet- ter understand functional roles, resilience to natural and anthropogenic disturbances, and to improve management strategies. Keywords Green turtle · Movement ecology · Niche partitioning · Seagrass · Stable isotope analysis Introduction Stable isotope analysis (SIA) is often used to elucidate the long-distance movements of migratory animals that are dif- ficult or expensive to track by other means (Gannes et al. 1998; Hobson 1999; Rubenstein and Hobson 2004). Isotope- based tracking is possible because of the presence of broad- scale isotopic gradients in nature, and because movements along these gradients by consumers may be recorded as vari- ation in isotope ratios of certain elements (e.g., C, N, S, H, O) in their tissues. For example, broad-scale latitudinal gra- dients in the ratio 13 C/ 12 C (δ 13 C) exist in terrestrial (Körner et al. 1991) and marine (Sackett et al. 1965) ecosystems. This can help us identify the approximate latitudes used by migratory animals for foraging based on tissues sampled at breeding sites or vice versa (e.g., Chamberlain et al. 1996). Despite its usefulness, SIA is not a panacea for animal tracking. A key challenge is that consumer isotope signatures Communicated by Helene Marsh. Electronic supplementary material The online version of this article (https://doi.org/10.1007/s00442-018-4218-z) contains supplementary material, which is available to authorized users. * Jordan A. Thomson jordy.thomson@deakin.edu 1 School of Life and Environmental Sciences, Centre for Integrative Ecology, Deakin University, Warrnambool Campus, P.O. Box 423, Warrnambool, VIC 3280, Australia 2 Department of Biological Sciences, School of Environment, Arts and Society, Florida International University, 3000 NE 151st St., North Miami, FL 33181, USA 3 Queensland Museum, South Brisbane, PO Box 3300, Brisbane, QLD 4101, Australia