Separating flow components in tile drainage and streamflow in a headwater agricultural catchment

Authors: GOSPODYN, LARISSA - University Of Waterloo; JARVIE, HELEN - University Of Waterloo; Williams, Mark; PLACH, JAINA - University Of Waterloo; MACRAE, MERRIN - University Of Waterloo

Submitted to: Journal of Hydrology
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 12/22/2025
Publication Date: 1/6/2026
Citation: Gospodyn, L., Jarvie, H., Williams, M.R., Plach, J., Macrae, M. 2026. Separating flow components in tile drainage and streamflow in a headwater agricultural catchment. Journal of Hydrology. https://doi.org/10.1016/j.ejrh.2025.103081.
DOI: https://doi.org/10.1016/j.ejrh.2025.103081

Interpretive Summary: Water carries nutrients from farm fields into nearby streams, rivers, and lakes. To improve farm productivity and water quality, we must understand how, when, and where water moves. ARS scientists in West Lafayette, Indiana in collaboration with the University of Waterloo assessed water movement at sites in Ontario, Canada. Hydrologic tracers were measured during 10 storm events in a drained field and small watershed. Using this data, scientists were able to identify different sources of water. This includes separating ‘new water’ falling as recent rainfall from ‘old water’ that was stored in the soil or groundwater. Results showed that during wet, spring conditions, identified water sources were similar among tracers and methods. However, during snowmelt, results were variable. Considering logistics and cost, electrical conductivity was a reliable tracer of water sources, expect for storm events following fertilizer application. Findings suggest that multiple tracers should therefore be used to identify water sources during critical flow conditions. Understanding the best methods to track water movement can help guide nutrient management decisions on the farm.

Technical Abstract: Hydrograph separation has become a common approach to infer water sources and nutrient pathways to stream discharge but is less frequently used to separate tile drain effluent. The identification of hydrologic and nutrient pathways into both streams and tiles is important in agricultural management and modelling. We separated ten event hydrographs for one contributing tile and stream outlet site of a southern Ontario headwater, agricultural catchment over a one-year period using two graphically-based approaches and a mass balance approach using a variety of common geochemical tracers (electrical conductivity (EC), d18O, d2H, Na+, Mg2+, Ca2+ and Cl-). Although these methods differ in their operational definition of baseflow, we observed similar baseflow estimates during wet, spring conditions, while snowmelt periods resulted in diverging estimates. However, variability in total flow of each component across methods was similar in the stream and tile drain. Variability in baseflow estimates was greater within the graphically-based methods than the tracer methods, but this was largely driven by user-calibration. When calibrating our graphically-based method with tracers, the resulting hydrographs became less flashy, lagged and had smaller peak baseflow amounts compared to the tracer-based separation alone and is therefore not recommended when intra-event dynamics are of importance. This work shows that EC can be successfully used in headwater agricultural streams and tile drainage discharge to separate flow components, but sampling additional tracers during critical periods, such as snowmelt or following fertilizer application, is recommended to ensure adequate end-member representation is acquired.