In agricultural watersheds across the world, decades of commercial fertilizer application and intensive livestock production have led to elevated stream nutrient levels and problems of eutrophication in both inland and coastal waters. Despite widespread implementation of a range of strategies to reduce nutrient export to receiving water bodies, expected improvements in water quality have often not been observed. It is increasingly understood that long time lags to seeing reductions in stream nutrient concentrations can result from the existence of legacy nutrient stores within the landscape. However, it is less understood how spatial heterogeneity in legacy nutrient dynamics might allow us to target implementation of appropriate management practices. In this thesis, we have explored the dominant controls of legacy nitrogen accumulation in a predominantly agricultural 6000-km2 mixed-landuse watershed. First, we synthesized a 216 year (1800 – 2016) nitrogen (N) mass balance trajectory at the subbasin scale accounting for inputs from population, agriculture, and atmospheric data, and output from crop production using a combination of census data, satellite imagery data, and existing model estimates. Using these data, we calculated the N surplus, defined as the difference between inputs to the soil surface from manure application, atmospheric deposition, fertilizer application, and biological N fixation, and outputs primarily from crop production. We then used the ELEMeNT-N model, with the estimates of the N mass balance components as the model inputs, to quantify legacy accumulation in the groundwater and soil in the study basin and 13 of its subbasins. Our results showed that from 1950, N surplus across the study site rose dramatically and plateaued in 1980. Agricultural inputs from fertilizer and biological nitrogen fixation were the dominant drivers of N surplus magnitude in all areas of the watershed. Model results revealed that 40% of the N surplus to the watershed since 1940 is stored as legacy N, and that the proportion of N surplus that is stored as legacy vary across the watershed, ranging from 33% to 69%. Where legacy tends to accumulate also varies across the watershed, ranging from 49% - 72% stored in soil, and 28% - 51% stored in groundwater. Through correlation analysis, we found that soil N accumulation tends to occur where there is high agricultural N surplus, and groundwater N accumulation tends to occur where mean groundwater travel times are long. We also found that using the model calibrated mean groundwater travel times as an indication of lag times, we can identify the length of lag time in various regions in the watershed to help inform long-term management plans. Our modeling framework provides a way forward for the design of more targeted approaches to water quality management.

A data driven approach was used in this study to investigate the drivers of nutrient water quality across the Laurentian Great Lakes drainage basin. Monitored time series of nutrient water quality and discharge were modelled using a dynamic regression-based model. Random forest machine learning was used as a framework to assess drivers of nutrient water quality, using mean annual flow-weighted concentrations (FWCs) and ratios calculated from modelled water quality, combined with spatial factors from monitored watersheds. Analysis revealed that landscape variables of developed land use, tile drained land, and wetland area played important roles in controlling nitrate and nitrite (DIN) and soluble reactive phosphorus (SRP) FWCs, while soil type and wetland area was important for controlling particulate phosphorus (PP) FWCs. Fertilizer and manure practices were important controls in nutrient ratios of SRP:Total Phosphorus (TP), and DIN:TP, with developed land use, manure application, and tile drained land important for the former, and developed land use and manure application (vs synthetic fertilizer application) important for the latter. Plots of feature contribution were generated to isolate the effect that spatial variables had in machine learning models and revealed underlying behaviour of important controls in driving nutrient water quality across the basin. Random forest models were further developed to predict FWCs and ratios of nutrients across all watersheds within the Great Lakes drainage basin. Modelled results revealed hot spots of high DIN, SRP and PP in the watersheds along the southeastern shores of Lake Huron, on the eastern watersheds of the Huron-Erie corridor, and in the southwestern watersheds of Lake Erie. High SRP:TP ratio hot spots were seen in watersheds along the southeastern shores of Lake Huron and along the eastern side of the Huron-Erie corridor. Hot spots of low DIN:TP ratios with high nutrient export were seen in the southwestern watersheds of Lake Erie, which has implications for harmful algal growth. Nutrient ratios across the Great Lakes watersheds compared similarly to other heavily human impacted catchments of the Baltic Sea and western Europe. Annual basin loads of DIN, SRP, and TP were estimated from random forest models for each year from 2000-2016. Calculated annual nutrient loadings of SRP and TP were consistent with other published values of Great Lakes watershed estimates and revealed highest loadings during 2011 when the largest recorded algal bloom in Lake Erie occurred to date. Overall, this data-driven analysis of nutrient water quality reinforces and refines our process understanding of nutrient pollution dynamics across the Great Lakes drainage basin.