Multi-tissue stable-isotope analyses can identify dietary specialization

Alexander L. Bond, Timothy D. Jardine, Keith A. Hobson, Matthew Davey (Editor)

    Research output: Contribution to journalArticlepeer-review

    42 Citations (Scopus)


    Individual specialization along one or more niche axes is now recognized as an integral and ubiquitous aspect of populations. A major challenge, however, is quantifying the level of specialization using robust metrics that are applicable across species and ecosystems. Measuring stable‐isotope values in multiple tissues with different isotopic turnover rates could be one mechanism for quantifying specialization.
    We used simulation studies of stable‐isotope values to investigate how the recently proposed relative index of specialization varies in relation to variance in prey isotope values, diet–tissue discrimination factors, specialist group size and tissue half‐life, and applied specialization metrics to two systems – Australian freshwater fish and marine birds in the Canadian Arctic.
    In all simulations, populations comprised entirely of generalists were easily separated from those with even small amounts (5%) of individual specialization. In some cases, however, specialization measured using isotope values with bimodal distributions may appear similar to those with univariate distributions, but this can be detected by examining the original data. All fish and bird species examined showed varying degrees of individual specialization.
    Analysing stable isotopes in multiple tissues can provide a useful index of the degree of specialization within a population that can be compared to the same metric measured in other groups or species.
    Original languageEnglish
    Pages (from-to)1428-1437
    Number of pages9
    JournalMethods in Ecology and Evolution
    Issue number12
    Publication statusPublished - 1 Dec 2016


    Dive into the research topics of 'Multi-tissue stable-isotope analyses can identify dietary specialization'. Together they form a unique fingerprint.

    Cite this