LAUSR

Strontium (Sr) isotope analysis can provide detailed biogeographical and ecological information about modern and ancient organisms. Because Sr isotope ratios (87Sr/86Sr) in biologically relevant materials such as water, soil, vegetation, and animal tissues predominantly reflect local geology, they can be used to distinguish geologically distinct regions as well as identify highly mobile individuals or populations. While the application of Sr isotope analysis to biological research has been steadily increasing, high analytical costs have prohibited more widespread use. Additionally, accessibility of this geochemical tool has been hampered due to limited understanding of (i) the degree to which biologically relevant materials differ in their spatial averaging of 87Sr/86Sr ratios, and (ii) how these differences may be affected by lithologic complexity. A recently developed continental‐scale model that accounts for variability in bedrock weathering rates and predicts Sr isotope ratios of surface water could help resolve these questions. In addition, if this ‘local water’ model can accurately predict 87Sr/86Sr ratios for other biologically relevant materials, there would be reduced need for researchers to assess regional Sr isotope patterns empirically. Here, we compile 87Sr/86Sr data for surface water, soil, vegetation, and mammalian and fish skeletal tissues from the literature and compare the accuracy with which the local water model predicts Sr isotope data among these five materials across the contiguous USA. We find that measured Sr isotope ratios for all five materials are generally close to those predicted by the local water model, although not with uniform accuracy. Mammal skeletal tissues are most accurately predicted, particularly in regions with low variability in 87Sr/86Sr predicted by the local water model. Increasing regional geologic heterogeneity increases both the offset and variance between modelled and empirical Sr isotope ratios, but its effects are broadly similar across materials. The local water model thus provides a readily available source of background data for predicting 87Sr/86Sr for biologically relevant materials in places where empirical data are lacking. The availability of increasingly high‐quality modelled Sr data will dramatically expand the accessibility of this geochemical tool to ecological applications.