Document Type


Publication Date




In this study, we investigate changes in future streamflows in California using bias-corrected and routed streamflows derived from global climate model (GCM) simulations under representative concentration pathways (RCPs): RCP4.5 and RCP8.5. Unlike previous studies that have focused mainly on the mean streamflow, annual maxima or seasonality, we focus on projected changes across the distribution of streamflow and the underlying causes. We report opposing trends in the two tails of the future streamflow simulations: lower low flows and higher high flows with no change in the overall mean of future flows relative to the historical baseline (statistically significant at 0.05 level). Furthermore, results show that streamflow is projected to increase during most of the rainy season (December to March) while it is expected to decrease in the rest of the year (i.e., wetter rainy seasons, and drier dry seasons). We argue that the projected changes to streamflow in California are driven by the expected changes to snow patterns and precipitation extremes in a warming climate. Changes to future low flows and extreme high flows can have significant implications for water resource planning, drought management, and infrastructure design and risk assessment.

Copyright Statement

This is an author-produced, peer-reviewed version of this article. © 2018, Elsevier. Licensed under the Creative Commons Attribution-NonCommercial-No Derivatives 4.0 license. The final, definitive version of this document can be found online at Journal of Hydrology, doi: 10.1016/j.jhydrol.2018.10.023