Hydrological modeling of spatial and temporal variations in streamflow due to multiple climate change scenarios in northwestern Morocco

Abstract

Climate change is one of the most important factors impacting hydrological regimes. In this paper, climate change impact on streamflow of Loukkos basin (northwestern Morocco) is evaluated using SWAT model for three future periods: near (2021–2040), mid (2041–2070), and far (2071–2100), compared to baseline 1981–2020. A set of bias-corrected climate models was used: five regional climate models (EURO-CORDEX), four global climate models (CMIP6) and their ensemble mean, under two representative concentration pathways respectively (RCP 4.5; RCP 8.5) and (SSP2-4.5; SSP5-8.5). Furthermore, SUFI-2 algorithm in SWAT-CUP was performed to calibrate (1981–1997), validate (1998–2015), and analyze uncertainty for each dataset at ten hydrological stations. In most stations, statistical performance indicated a good simulation, with a Nash–Sutcliffe efficiency (NSE) greater than 0.77 and percent bias (PBIAS) within ±10% on a monthly basis. Overall, 82% of models indicated that future climate could decline streamflow. The largest decrease would be for 2071–2100 and under RCP 8.5/SSP5-8.5. Our findings could help planners and policymakers in developing reasonable water management policies and climate change adaptation measures.