Hysteresis Patterns During Storms Suggest that Vegetative Cover Mediates Nitrate Export in Two Agricultural Watersheds

Abstract:

Alteration of the Midwestern U.S. landscape, including changing land cover, excess nutrient inputs from fertilizer, and modified hydrology via agricultural drainage, has influenced solute transport in streams. A clear understanding of the mechanisms driving nutrient export from agricultural watersheds will be critical given anticipated shifts in hydrology associated with a changing climate. Specifically, more frequent, intense precipitation and altered snow patterns are predicted for the upper Midwest. However, the impact of interactions between changing precipitation and land cover on nitrate runoff patterns are largely unknown, yet could be critical for mitigating nutrient pollution. We used four years of high-frequency nitrate sensor data from two tile-drained, agricultural watersheds in Indiana to explore hysteresis and flushing patterns for >100 storm events. We used two indices, the hysteresis index (HI) and flushing index (FI), to characterize the magnitude, timing, and source behavior of nitrate loss in order to explore physicochemical controls of nitrate export from two contrasting watersheds. We observed mostly negative HIs, due to widespread nitrate availability from fertilizer application in both watersheds. In contrast, patterns in FI were variable, suggesting antecedent precipitation likely controlled the transport of nitrate from the landscape to streams. In fact, pre-event discharge (one day prior to storm), reflecting overall basin wetness preceding an event, explained 23-46% of the variation in storm nitrate export within a watershed across the study period. Our results also show increased watershed land cover in winter, planted as ryegrass cover crops when farm fields would normally be fallow, can decouple the tight relationship between nitrate mass loss and runoff that is typical of tile-drained, agricultural watersheds. Overall, we found that high-frequency nitrate data obtained via real-time sensors provided unique insights into the seasonality and drivers of storm-induced nitrate loss from agricultural watersheds.