Embracing uncertainty in time: A Bayesian dynamic mixing model for isotopic diet estimation

Abstract Isotopic mixing models are powerful tools in ecology for estimating diets. Since their initial formulation, these models have evolved into two distinct branches: Bayesian mixing models and dynamic mixing models. Bayesian mixing models significantly improve uncertainty management but remain static assuming a stable state. In contrast, dynamic mixing models are deterministic and incorporate isotopic dynamics, making them suitable for non‐stable systems. However, they lack uncertainty management. To bridge these two approaches, we developed a Bayesian dynamic mixing model. This model integrates isotopic temporal dynamics with robust uncertainty management. In this study, we confronted our model and existing dynamic mixing model to simulations and real‐world case studies. The results revealed that the Bayesian dynamic model outputs exhibit high stability with respect to data variability. It can also partially compensate for inaccuracies in isotopic turnover rate inputs, leading to more reliable diet estimations. This framework provides an accessible and reliable method for conducting diet studies using stable isotopes regardless of the system stability.