Fig. 7.4.

Example of a mixing model using δ13C and δ15N to estimate the importance of estuarine producers to a hypothetical caridean shrimp. (a) Firstly, the stable isotope composition of all potential sources and shrimp (corrected using appropriate trophic discrimination factors) is plotted to confirm that corrected consumer values fall within the polygon delimited by all sources. (b) The Bayesian mixing model simmr (Parnell 2019) is run to estimate the importance of each source. In this example, results suggest that mangroves have the least importance, contributing only 3–18% (95% credibility interval) and that seagrass contributes 18–35%. For phytoplankton and microphytobenthos, the resulting probability distributions are very wide, so their importance could not be resolved. (c) These two aquatic sources are therefore combined a posteriori to provide a more precise estimate of their combined contribution, with results showing that, combined, these contribute the most (57–70%) to shrimp nutrition.