WATER QUALITY IN NORTH AMERICAN LAKES: PARTITIONING MACRO- AND MICRO-SCALE ECOLOGICAL PROCESSES USING NON-STATIONARY SPATIAL MODELS
Freshwater ecosystems are vulnerable to anthropogenic environmental changes that occur at broad spatial extents, leading to widespread nutrient pollution and poor water quality. Recent studies suggest that drivers of water quality vary across individual lakes and regions, but elucidating the multiple mechanisms behind those differences remains challenging. Employing new statistical methods, such as non-stationary models, can tell us whether a covariate (e.g., land use, climate) is associated with macro-scale ecological patterns through its importance in the mean function, with micro-scale patterns through its importance in the covariance function, or with both macro- and micro-scale patterns. We estimated non-stationary models using HMC in Stan for water quality data for thousands of lakes in the LAGOS-NE database. Our results show that precipitation, agriculture, and lake depth explained broad-scale variation in mean water clarity, and we found evidence of anisotropy and a non-stationary spatial variance: spatial SD increased by a factor of 1.39 per SD increase in agriculture in the watershed. Overall, these results suggest that both macro- and micro-scale drivers are important for understanding and managing eutrophication.
Sarah Collins (Primary Presenter/Author), University of Wyoming, firstname.lastname@example.org;
Pavel Chernyavskiy (Co-Presenter/Co-Author), University of Wyoming, email@example.com ;
Charlotte Narr (Co-Presenter/Co-Author), Colorado State University, firstname.lastname@example.org;
Marie-Agnes Tellier (Co-Presenter/Co-Author), University of Wyoming, email@example.com;