TDFCAM: A method for estimating stable isotope trophic discrimination in wild populations

Stable isotope mixing models (SIMMs) are widely used for characterizing wild animal diets. Such models rely upon using accurate trophic discrimination factors (TDFs) to account for the digestion, incorporation, and assimilation of food. Existing methods to calculate TDFs rely on controlled feeding trials that are time‐consuming, often impractical for the study taxon, and may not reflect natural variability of TDFs present in wild populations. We present TDFCAM as an alternative approach to estimating TDFs in wild populations, by using high‐precision diet estimates from a secondary methodological source—in this case nest cameras—in lieu of controlled feeding trials, and provide a framework for how and when it should be applied. In this study, we evaluate the TDFCAM approach in three datasets gathered on wild raptor nestlings (gyrfalcons Falco rusticolus; peregrine falcons Falco perigrinus; common buzzards Buteo buteo) comprising contemporaneous δ13C & δ15N stable isotope data and high‐quality nest camera dietary data. We formulate Bayesian SIMMs (BSIMMs) incorporating TDFs from TDFCAM and analyze their agreement with nest camera data, comparing model performance with those based on other relevant TDFs. Additionally, we perform sensitivity analyses to characterize TDFCAM variability, and identify ecological and physiological factors contributing to that variability in wild populations. Across species and tissue types, BSIMMs incorporating a TDFCAM outperformed any other TDF tested, producing reliable population‐level estimates of diet composition. We demonstrate that applying this approach even with a relatively low sample size (n