Influence of antecedent hydrologic conditions on patterns of hydrochemical export from a first-order agricultural watershed in Southern Ontario, Canada

The ability to quantify and predict discharge and nutrient export patterns from agricultural catchments is confounded by the effects of antecedent hydrologic conditions (AHC) in upland areas. This paper examines the impacts of AHC on storm hydrologic response and nutrient export for 59 events measured over a 2-year period (year-round). Event-related discharge and nutrient (nitrate, phosphate) export were highly temporally variable. Runoff ratios ranged from 0 to 0.99 over the study period, with the lowest median runoff ratios being observed during the summer and early autumn period and the highest median runoff ratios being observed in March following snowmelt. Runoff ratios were positively correlated with AHC indices such as pre-event stream discharge (1-day) and antecedent (14-day) stream discharge although strong predictive relationships could not be generated (p > 0.01). In general, runoff ratios increased throughout successive events as AHC became wetter; however trends observed among suites of events with having similar AHC did not produce similar patterns of hydrologic or nutrient export response, which clearly demonstrates the non-linear responses to variable AHC in the watershed. Basin nitrate (image), total phosphorus (TP) and soluble reactive phosphorus (SRP) export varied with AHC in the basin, and flow-weighted mean nutrient concentrations (FWMC) ranged considerably (mean, range: 8, 1–18 mg L−1 NO3–N; 18, 3–161 μg L−1 SRP; 75, 15–274 μg L−1 TP). Nutrient export increased with successive events when the basin was on a wetting trend, but decreased with successive events when the basin was on an overall drying trend. These patterns were linked to peak water table position during events and timing of fertilizer application. Overall, this work demonstrates that watershed hydrologic and nutrient export responses to precipitation and AHC are complex and cannot be predicted using simple linear relationships. This study also demonstrates the need to collect larger data sets when examining relationships between watershed hydrologic and nutrient export response.