Defining drivers of nitrogen stable isotopes (δ15N) of surface sediments in temperate lakes

The nitrogen stable isotopic signature (δ15N) of sediment is a powerful tool to understand eutrophication history, but its interpretation remains a challenge. In a large-scale comparative approach, we identified the most important drivers influencing surface sediments δ15N of 65 lakes from two regions of Canada using proxies that reflect watershed nitrogen (N) sources, internal lake microbial cycling and productivity. Across regions, we found that water column total nitrogen (TN),  %N in the sediments and lake morphometric variables were the best predictors of sedimentary δ15N, explaining 66 % of its variation. Significant relationships were also found between sediment δ15N and human-derived N load (Radj.2\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$ R_{{{\text{adj}} .}}^{2} $$\end{document} = 0.23, p < 0.001), the latter being a strong predictor of TN (Radj.2\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$ R_{{{\text{adj}} .}}^{2} $$\end{document}  = 0.68, p < 0.001). Despite a relatively strong overall relationship, variation partitioning revealed an interesting difference in the dominant variable that influenced regional δ15N. Alberta lake sedimentary δ15N signature was dominated by human derived N load. In contrast, internal processing appeared to be more important in Quebec lakes, where sediment δ15N was best explained by  %N in the sediments and lake volume. Overall, our findings support the use of δ15N in paleolimnological investigations to reconstruct changing N sources to lakes but also highlight that regions may have distinctive drivers. Interpretations of sediment δ15N are likely to be strongest when multiple lines of evidence are employed and when placed in a regional context.