Increased Nonstationarity of Stormflow Threshold Behaviors in a Forested Watershed Due to Abrupt Earthquake Disturbance

Abstract

Abstract. Extreme earthquake disturbances to local and regional landscape vegetation could rapidly impair original hydrologic functioning, significantly increasing the hydrologic nonstationarity and complexity in threshold behaviors of rainfall-runoff processes. It is unclear how alternating catchment behaviors under an ongoing large earthquake disruption are mediated by long-term interactions of landslides and vegetation evolutions. In a famous Wenchuan earthquake-affected watershed, China, the presence and form of three-linear stormflow threshold behaviors are examined, and both thresholds are identified as a diagnostic tool to characterize variations in hydrologic emergent patterns pre- and post-earthquake. It was revealed that lower rising threshold (Tr) value (210.48) in post-earthquake landslide regions exhibited faster stormflow responses, possibly triggering huge flood disasters. An integrated watershed average (IWA) index for both thresholds (generation threshold Tg-IWA and Tr-IWA) at the watershed scale was proposed based on long-term vegetation dynamics and threshold-based hydrological theory. The interannual variations of both hydrologic thresholds were assessed to detect the nonstationarity in hydrologic extremes and nonlinear runoff response pre- and post-earthquake. 2011 was a tipping point of the unsteady recovery process, as post-earthquake landslides evolutions reached a state of extreme heterogeneity in space. At that moment, the Tr-IWA value at the watershed scale decreased by ~ 9 mm compared to the pre-earthquake level, and the fast expansion of landslides generally led to a larger extension of variable source area from channel to neighboring hillslopes and faster subsurface stormflow contributing to flash floods. Additionally, we present a conceptual model interpreting how the short- and long-term interactions of earthquake-induced landslides and vegetation affect flood hydrographs at event timescale that generated an increased nonstationary hydrologic behavior. This study expands our knowledge about the threshold-based hydrological behavior and the nonstationary stormflow threshold behaviors in response to abrupt earthquake disturbance for the prediction of future flood regimes.