Rainfall–stream flow responses in a mixed-land-use and municipal watershed of the central USA

Abstract

Competing influences of climate, watershed characteristics, and human activities can confound classic hydrologic process assumptions and, therefore, management efforts in mixed-land-use watersheds. Seven water years (WY2009–2015) of sub-hourly rainfall and stream flow data were collected from five nested gauging sites in a mixed-land-use watershed of the central USA. Pasture/cropland use decreased by 18% while urban land use increased by 22% from the headwaters to the watershed outlet. A flow separation algorithm was used to assess rainfall–stream flow response variables (i.e. stream flow, event flow and base flow) at each gauging site and at multiple time steps (7-year study, annual, monthly, event based). Annual total rainfall ranged from 678 to 1610 mm during the study period. Urban land use explained 61–93% of the variance in the slope of the best fit line between total rainfall and stream flow, event flow, and base flow confirming significant (CI = 95%; p < 0.006; n = 5) impacts of urban land use on rainfall–stream flow response relationships. Total stream flow, event flow and base flow increased by 101, 95, and 110%, respectively, at the rural/urban interface of the watershed during a year of extreme drought. Base flow index values were positively correlated with urban land use and negatively correlated to total rainfall (R2 = 36.98; p < 0.023; n = 111). Stream flow alterations were greatest during drier years, and during smaller flow events at gauging sites with greater than 6% urban land use. Results advance quantitative understanding of rainfall–stream flow relationships for urbanizing mixed-land-use watersheds.