The Single, the Whole, and the Pooled: Implications of Tissue Choice for Stable Isotope Analysis

In the exciting era of non-invasive tissue sampling and data reuse via open access databases, it is increasingly challenging to ensure comparability of sample types for stable isotope analysis in ecology. According to the theoretical predictions about the changes in the stable isotope values between prey and consumers and tissue turnover rates, the ecological question alone should dictate the choice of the sample type used (single tissue, whole body, or pooled samples). However, the sample types have also been constrained by the body size, availability, or legal status of the study species, and stable isotope data sets are routinely obtained from a variety of sample types. While there is a general agreement that this practice may bias the ecological conclusions of stable isotope studies, the risks incurred have seldom been reviewed explicitly, let alone quantified. With both ethical aspects and open science opportunities arising in the field, the issues around sample types in stable isotope ecology must be tackled explicitly. We illustrate how comparing stable isotope values from different sample types affects niche analyses and diet reconstructions, and detail the consequences of pooling samples on stable isotope variance. We call for (1) a thorough consideration of the adequation between sample types and ecological questions, including in data harvesting from global databases, and (2) for the development of methodological studies around sample type issues and that of a systematic framework that incorporates the known variations between sample types in statistical analyses.