Seasonal and event-based concentration-discharge relationships to identify catchment controls on nutrient export regimes

Abstract

The analysis of concentration-discharge (C-Q) relationships provides useful information on the processes controlling the mobilization and delivery of chemical elements into streams as well as biogeochemical transformations in river networks. Previous metrics developed to characterize export regimes seldom considered the possibility for the C response to Q dynamics to differ between short-term Q variations during storm events and seasonal Q variations during baseflow periods. Here, we present the “C-Qquick-slow” model, which considers the possibility for C-Q relationships to vary across temporal scales. This model was applied in 219 French catchments with various sizes (11–2500 km²), land use and hydrological contexts. We evidenced contrasting export regimes for nitrate (NO3−), total phosphorus (TP) and soluble reactive phosphorus (SRP), and surprisingly consistent C-Q patterns at the seasonal scale for each parameter. For instance, NO3−-Q relationships were positive at the seasonal scale in 75% cases and relationships during storms showed either a dilution pattern (24% cases), a non-significant pattern (50%), or a mobilization pattern (12%). TP and SRP relationships with Q at the seasonal scale were almost systematically negative (95%), and patterns during storm events were in most cases mobilization for TP (77%) or non-significant for SRP (69%). We linked the different C-Q relationships with catchment descriptors and found that indicators of diffuse source loading determined NO3− seasonal amplitudes, and hydrological drivers could explain the behavior during storms. By contrast, point sources determined P seasonal amplitudes, and diffuse sources controlled P dynamics during storms. The C-Qquick-slow model has the potential to improve nutrient load estimations because of the good predictability of appropriate C-Q archetypes and the possibility to interpolate low frequency concentration data to a daily frequency.