Our new NSF-funded project seeks to understand the origins of spatial vs temporal variability in water quality patterns, answering the following specific questions: What is the relative importance of stream solute spatial vs. temporal variability? What are the drivers of stream solute spatial and temporal variability?
Interested in water quality data science? Interested in applying to work on this project? Apply here.
We focus on three contrasting solute groups (geogenic, biogenic, anthropogenic) with varying links to geographical, anthropogenic, hydrological, and climatological drivers. We test a series of linked predictions about controls on patterns of space-time variation using a continental-scale data inventory across the eastern United States, spanning large gradients in latitude and hydroclimate zones, lithology, wetland and forest cover, and anthropogenic land use intensity. First, we seek to explain the substantial spatial variation in mean concentrations observed for all three solute groups, based on the coupled solute-specific influences of ecoregion, network position, and solute source area. Mean concentrations are expected to vary with aggregated landscape properties, such as wetland coverage or agricultural land use intensity, and mean climate conditions, such as aridity and temperature. Second, we expect observed spatial variability to be reduced with aggregation schemes carefully informed by those factors. Third, we predict that temporal variability around the mean can be understood and explained in terms of key drivers such as flow variation, seasonality, and network position, albeit with solute-specific sensitivities.