Evaluating the Ubiquity of Thresholds in Rainfall-Runoff Response Across Contrasting Environments

Thresholds in precipitation-runoff relationships have been observed in numerous studies using scatter plots comparing meteorological factors and hydrologic response metrics. Most thresholds reported in the literature have been identified from relationships between meteorological factors that quantify volumes or depths of water (e.g., total event rainfall) and metrics capturing hydrologic response magnitude (e.g., runoff ratio), with a strong emphasis on hillslopes and catchments in temperate humid environments. Knowledge gaps, however, remain regarding the ubiquity of hydrologic thresholds across different climatic environments and different meteorological factors that affect different response metrics. This study therefore aimed to evaluate relationships for a wide range of meteorological factors and response metrics derived from event-scale rainfall-runoff analysis for 21 sites spanning seven contrasting geographic areas. Specifically, meteorological factors quantifying rainfall depth, rainfall intensity, and hydrologic abstractions related to evapotranspiration were considered, along with response metrics that describe response timing and response magnitude, leading to 4,557 relationships being evaluated. While rainfall depth thresholds were observed for most sites, rainfall intensity thresholds were also observed. Additionally, threshold behavior was shown to be sensitive to antecedent conditions over specific durations of time preceding a rainfall-runoff event. The large number of relationships evaluated in this study allowed for the development of a typology of threshold dynamics and the formulation of hypotheses about dominant hydrological processes. This typology may not only promote standardized threshold descriptions but also make intersite comparisons of nonlinear rainfall-runoff behavior easier.

Automated summary

Thresholds in precipitation-runoff relationships have been observed at multiple sites across different geographic areas, showing sensitivity to antecedent conditions and different meteorological factors. This study evaluated a wide range of relationships between meteorological factors and response metrics, leading to the development of a typology of threshold dynamics and hypotheses about dominant hydrological processes. The findings not only promote standardized threshold descriptions but also facilitate intersite comparisons of nonlinear rainfall-runoff behavior.