Quantifying pelagic phosphorus regeneration using three methods in lakes of varying productivity

Phosphorus (P) is often a limiting nutrient in freshwater ecosystems, and understanding P dynamics in lakes is critical for eutrophication management. Pelagic P regeneration can support a large fraction of primary production in stratified freshwaters. Various techniques have been used to quantify pelagic P regeneration including (1) P mass balance supply-demand, (2) regression using total P as a predictor, and, more recently, (3) whole-lake metabolism calculated from highfrequency dissolved oxygen (DO) data. To our knowledge no study comparing these methods in multiple lakes has been performed. To compare these 3 approaches, we investigated 3 Global Lake Ecological Observatory Network (GLEON) lakes that differ in productivity: Acton, a Midwestern USA hypereutrophic reservoir; and 2 Northeastern USA glacial lakes, oligotrophic Giles and mesotrophic/dystrophic Lacawac. In Acton, we used all 3 methods, but for Giles and Lacawac we used only the total P regression and metabolism techniques. Our results show the best agreement among methods in the mesotrophic lake, whereas the metabolism approach underestimated regeneration in the oligotrophic lake and overestimated regeneration in the hypereutrophic reservoir compared with other methods. P regeneration rates for the hypereutrophic reservoir were the most sensitive to the metabolism-based input parameters. Our study illustrates a novel use of high-frequency DO data, which are commonly collected on many GLEON buoys, to understand lake nutrient dynamics.