Does δ15N in river food webs reflect the intensity and origin of N loads from the watershed?

Stable nitrogen isotope ratios (δ15N) were measured in invertebrates and fish collected from 82 river sites located in the Saint-Lawrence Lowlands in Québec, Canada, to examine the relationship between aquatic biota δ15N and anthropogenic nitrogen (N) loads. Mean δ15N values of all three trophic levels examined (primary consumers, predatory invertebrates and invertebrate-feeding fish) were highly correlated with total anthropogenic N loads on the watershed (kg N km− 2 year− 1; r2 > 0.61, p < 0.0001) and with N loads originating from livestock manure (r2 > 0.62, p < 0.0001), synthetic fertilizers (r2 > 0.45, p < 0.0001), and human population (r2 > 0.29, p < 0.0001), respectively. Significant relationships were also observed between primary consumer δ15N and N loads originating from each of the three livestock species examined (bovines, pigs and poultry; p < 0.0001). Furthermore, all three animal species contributed significantly and independently in elevating primary consumer δ15N (multiple r2 = 0.67, p < 0.0001). Curvilinear regressions were observed at all levels of analysis, δ15N values increasing slowly over a wide range of low levels of N loads, but increasing much faster as N loads grew larger. The three anthropogenic N sources examined were highly correlated with one another, preventing us from statistically isolating their respective effects on δ15N. When these loads were expressed as a proportion of total N load, δ15N of aquatic biota was still highly correlated with N from livestock and fertilizers, but not with N from human population. Overall, these results suggest that δ15N values of aquatic consumers could be used as indicators of the intensity of anthropogenic N loading on watersheds, but not as tracers of the relative importance of individual N sources.