BITS: A Bayesian Isotope Turnover and Sampling model for strontium isotopes in proboscideans and its potential utility in movement ecology

Abstract

Abstract Strontium isotope ratios (87Sr/86Sr) of incrementally grown tissues have been widely used to study movement ecology and migration of animals. However, the time scale of 87Sr/86Sr incorporation from the environment into tissue and how it may influence data interpretation are still poorly understood. Using the relocation of a zoo elephant (Loxodonta africana) named Misha, we characterise and model the 87Sr/86Sr turnover process using high‐resolution measurements of its tusk dentine. We seek to develop a framework that can improve quantitative interpretation of 87Sr/86Sr data in tissues. The 87Sr/86Sr transition associated with the relocation is measured using laser ablation inductively coupled plasma mass spectrometry (LA‐ICP‐MS) on a prepared tusk slab. We develop a turnover model (BITS), with a rapidly exchanging central pool and a slowly exchanging peripheral pool, in a Bayesian statistical framework. The measured dentine data are first used to calibrate model parameters. The parameters are then used to estimate possible 87Sr/86Sr input time series from two datasets via model inversion: a fidelity test using Misha's dentine data and a case study using published dentine measurements from an Alaskan Woolly Mammoth (Mammuthus primigenius). The LA‐ICP‐MS data are consistent with a two‐compartment turnover process with equivalent half‐lives of 41 days for the central pool and 170 days for the peripheral pool. The model inversion shows good fidelity when estimating the intake 87Sr/86Sr time series associated with Misha's relocation. In the case study, the model suggests an abrupt pattern of change in, and a much wider range of, intake 87Sr/86Sr values than expressed in the woolly mammoth dentine data themselves. Our framework bridges the gap between environmental 87Sr/86Sr variation and data measured in tusk dentine or other incrementally grown tissues. It could be coupled with movement models and additional isotope tracers to study seasonal residency or the spatial and temporal patterns of movement/migration. The generic turnover processes can be adapted to other isotope systems, additional incremental tissues, or other organisms, thus expanding our modelling toolkit to investigate niche partitioning, life history traits and behavioural patterns in conservation biology, archaeology and paleoecology.