Modeling urban phosphorus export to receiving waters: magnitudes, speciation, and management implications

Understanding phosphorus (P) dynamics in urban landscapes is an emerging research topic as P export from urban landscapes towards aquatic ecosystems causes eutrophication-related challenges in these environments. We investigated P export and forms in four research sites in Ontario, Canada, including three urban catchments and a stormwater pond, all located within the Great Toronto Area in the drainage basin of Lake Ontario. We conducted P speciation laboratory analyses on runoff and suspended sediment samples to measure total P (TP), total dissolved P (TDP), dissolved reactive P (DRP), dissolved unreactive P (TDP–DRP), and PP (TP–TDP). Multiple linear regression (MLR) models were also developed to quantify annual loadings of these P species. Models indicated that P loadings in our sites were close to the lower limit of values reported in the literature, with the simulated range of 0.2—0.46 kg ha^-1 yr^-1 for TP export, 0.06—0.168 kg ha^-1 yr^-1 for TDP, 0.011—0.073 kg ha^-1 yr^-1 for DRP, 0.026—0.095 kg ha^-1 yr^-1 for DUP, and 0.163—0.288 kg ha^-1 yr^-1 for PP. In our MLR models, precipitation explained a large fraction of variability in loadings with the median of 58% across all models. Moreover, we realized that as the proportion of residential land within the drainage area increased, larger amounts of P loadings were exported at the catchment scale. Results also implied that pond served as a major P sink, with annual retention of 82, 93, 91, 94, and 77% for TP, TDP, DRP, DUP, and PP, respectively. Mass balance analyses based on sequential P extraction in the sediment core samples revealed that P retention was attributed to sedimentation in the ponds, as well as chemical precipitation of P with calcium mineral phases. In terms of P composition, most of P export in our sites (72—88%) were in particulate form. Besides, the ratio between dissolved forms and TP were the highest in the catchment with the largest amount green spaces. This study demonstrates that, as land-use characteristics impose variations in constituent loadings, urban P management options also have to be varying from a catchment to another. However, sediment removing practices such as the use of ponds will certainly be a reliable P retention approach as most of urban P could be sediment-bound. Furthermore, enhancing the formation of calcium phosphate and other redox-stable mineral phases could be explored as a best management practice in existing and new ponds for improving P retention.