Assessing the effect of stationarity on annual peak streamflow in conterminous USA

Abstract:

Hydrologic design involves the use of annual peak streamflow data from gauging stations, which are often assumed to be stationary. While the assumption of stationarity is now being challenged at many locations, it is necessary to know the role of anthropogenic as well as climate factors play in causing non-stationarity in the time series data. The goal of this study is to assess the trends in past annual peak streamflow to understand the effect of anthropogenic and climatic factors on stationarity in streamflow annual maximum time series at different USGS stations throughout conterminous USA. Stationarity is checked by using the Augmented Dickey Fuller (ADF) test, the Kwiatkowski-Phillips-Schmidt-Shin (KPSS) test, and the Modified Mann-Kendall test. Additionally, the effect of stationarity on the 100-year design flow is investigated by computing the 100-year design flow by using a moving window and accumulation method analysis. The trends in the data, the presence of non-stationarity and the change in 100-year design flow are then related to land use and climate change. The findings from this study can be used understand the change in flood risk at the study sites.