Observed Trends in Global Indicators of Mean and Extreme Streamflow

Abstract

This study investigates global changes in indicators of mean and extreme streamflow. The assessment is based on the Global Streamflow Indices and Metadata archive and focuses on time series of the annual minimum, the 10th, 50th, and 90th percentiles, the annual mean, and the annual maximum of daily streamflow. Trends are estimated using the Sen‐Theil slope, and the significance of mean regional trends is established through bootstrapping. Changes in the indices are often regionally consistent, showing that the entire flow distribution is moving either upward or downward. In addition, the analysis confirms the complex nature of hydrological change where drying in some regions (e.g., in the Mediterranean) is contrasted by wetting in other regions (e.g., North Asia). Observed changes are discussed in the context of previous results and with respect to model estimates of the impacts of anthropogenic climate change and human water management. Plain Language Summary Studies of trends in streamflow data from across the globe are essential for understanding patterns and changes in water availability (e.g., regions of deficit and abundance) and evaluating the fidelity of global water availability models. This study evaluates historical trends in streamflow data, using a new data set of observations from over 30,000 sites around the world. The study is comprehensive, looking at changes in low flows (defined as the lowest day of flow in each year), average flows, and high flows (the highest day of flow in each year). An interesting outcome is that where trends are present in a region, the direction of the trend is often consistent across all indices for that region (consistently drier or wetter), as distinct from the possibility of stronger extremes (wetter maximums and drier minimums).