Controls on magnitude and timing of peak runoff response to rainfall across the continental US

Abstract Climate, topography, and subsurface properties control how rainfall becomes streamflow. However, the interactions between these factors are poorly understood and remain largely unquantified. Here, we use ensemble rainfall–runoff analysis to quantify streamflow response to rainfall at hourly time scales in 516 catchments across the continental United States. Our results reveal patterns in runoff response that reflect differences in climatic conditions and landscape characteristics, as demonstrated through mapping and multiple regression analysis. More than three-fourths (77%) of the studied catchments exhibit streamflow responses that are too rapid to be captured by conventional analyses of daily rainfall–runoff data. Mean precipitation and potential evapotranspiration (PET) strongly affect peak height per unit rainfall (but not peak lag), whereas channel slope strongly affects peak lag (but not peak height), indicating a decoupling between controls on runoff timing and magnitude. Runoff peaks are quicker and higher in smaller basins with lower soil conductivity and water storage capacity, reflecting shorter flow paths, reduced infiltration, and less dispersion of flow peaks. In addition, higher mean precipitation and lower PET (implying wetter antecedent conditions) are associated with higher flow peaks per unit rainfall, and steeper channels (implying higher flow celerity) are associated with shorter peak lags. These mechanistically intuitive relationships are quantified here using long-term, high-frequency data at continental scale. Our results highlight the distinct yet interrelated roles of climate, topography, and subsurface properties in shaping rainfall–runoff response, and may provide insights for flood mitigation strategies in vulnerable basins under a changing climate.