Statistical tools for the analysis of stable isotopes to discriminate between pollution sources in coastal water bodies

C and N Stable isotopes analyses have become a popular method for the study of aquatic ecosystems, as they (i) furnish useful information for the understanding of food webs structure and related processes, and (ii) can inform on the origin of assimilated carbon and nitrogen atoms in aquatic organisms. Different nitrogen isotopic values are known to belong to different nitrogen sources, allowing the discrimination between inorganic, organic or synthetic sources. In this work we focus on an experiment carried out in three neighboring coastal lakes located south of Rome (Italy): Caprolace, Fogliano and Sabaudia. The aim of the experiment is to identify the origin of Nitrogen pollutant sources affecting each lake trough the analysis of changes in nitrogen isotopic values (δ15N) in replicated specimens of a single species of green alga (Ulva lactuca), spatially distributed to cover the entire lake surface. Algal specimens used were previously collected from a benchmark, clean site (Circeo). δ15N of each specimens was recorded at the time of collection, by the dissection of 1cm2 of algal surface, and after 48h of submersion in each site of target lake, to calculate the isotopic variation at individual level and avoiding potential confounding effects of intraspecific variability. A Bayesian hierarchical model relating isotopic variations to abiotic covariates, and spatial random effects, is built to understand and describe isotopic variations in the water bodies.