Bayesian tracer mixing models and isotope data elephant diet

In the supplementary information we explain why we did not use Bayesian tracer mixing models in the final paper. Isotope data: We air-dried the faecal and vegetation samples in an air-drying cabinet. After transportation, we dried samples for a further 24 h at 70 °C and ground to pass through a 1 mm mesh at the Okavango Research Institute laboratory. Samples were then shipped to the stable isotope laboratory housed in the archaeology department at the University of Cape Town, South Africa. There they were weighed in to tin cups to an accuracy of 1 mg on a Sartorius M2P microbalance.Sample weights were 2.5 mg for samples with potentially low nitrogen content, 2.1–2.2 mg for legumes and agricultural samples and 2.3 mg for other sample types. Samples were combusted in a Flash 2000 elemental analyser interfaced to a Delta V Plus isotope ratio mass spectrometer (IRMS) via a Conflo IV gas control unit (Thermo Scientific,Bremen, Germany). The in-house standards used were: Sucrose(“Australian National University (ANU)”sucrose), MG (Merck Gel), Acacia (Acacia saligna, Glencairn). All the in-house standards were calibrated against IAEA (International Atomic Energy Agency) standards,either at UCT or by other labs. Nitrogen was expressed in terms of its value relative to atmospheric nitrogen, while carbon was expressed in terms of its value relative to Pee-Dee Belemnite (VPDB).In our analyses we use stable isotopic values from faeces and their association to isotopic values of most likely consumed browse and grass species as proxies for elephant diet. In reality, diet content can vary from isotopic content due to issues as fractionation of isotopes during digestion, this is why we correct the results with fractionation values (Codron and Codron, 2009).