Statistical regression relationships between the hydrogen (H) and oxygen (O) isotope ratios (δ2H and δ18O, respectively) of animal organic tissues and those of environmental water have been widely used to reconstruct animal movements, paleoenvironments, and diet and trophic relationships. In natural populations, however, tissue-environment isotopic relationships are highly variable among animal types and geographic regions. No systematic understanding of the origin(s) of this variability currently exists, clouding the interpretation of isotope data. Here, we present and apply a model, based on fundamental metabolic relationships, to test the sensitivity of consumer tissue H and O isotope ratios, and thus tissue-environment relationships, to basic physiological, behavioral, and environmental parameters. We then simulate patterns in consumer tissue isotopic compositions under several 'real-world' scenarios, demonstrating that the new model can reproduce-and potentially explain-previously observed patterns in consumer tissue H isotope ratios, including between-continent differences in feather-precipitation relationships and 2H-enrichment with trophic level across species. The model makes several fundamental predictions about the organic O isotope system, which constitute hypotheses for future testing as new data are obtained. By highlighting potential sources of variability and bias in tissue-environment relationships and establishing a framework within which such effects can be predicted, these results should advance the application of H and O isotopes in ecological, paleoecological, and forensic research.
Keywords: First-order process; Gradient; Sensitivity; Trophic level; Water.