Phosphorus dynamics across intensively monitored subcatchments in the Beaver River

We report results from a spatially intensive monitoring and modelling study to assess phosphorus (P) dynamics in the Beaver River, a tributary of Lake Simcoe, Ontario. We established multiple monitoring stations (9 flow and 24 water quality stations) from headwaters to near the outflow that were operated for 2 field seasons, complementing longer term data from a flow monitoring site and water chemistry monitoring site. We applied the Branched-INCA-P model, which allows fully distributed simulations supported by highly distributed monitoring data. Using spatially distributed data helped better understand variable P and sediment dynamics across the catchment and identify key model uncertainties and uncertainties related to catchment P management. Measured and modelled total P concentrations often exceeded provisional water quality thresholds in many areas of the catchment and highlight the value of studying water quality across multiple subcatchments rather than at a single site. Total P export coefficients differed widely among subcatchments, ranging from 2.1-21.4 kg km(-2) y(-1) over a single year. Export coefficients were most strongly (negatively) related to the proportion of wetland cover in subcatchments. The INCA-P model captured spatial variation in P concentrations relatively well, but short-term temporal variability in the observed data was not well simulated across sites, in part due to unmodelled hydrological phenomena including beaver activity and unknown drivers of P peaks that were not associated with hydrological events.