A dynamic land use/land cover input helps in picturing the Sahelian paradox: assessing variability and attribution of changes in surface runoff in a Sahelian watershed

Abstract In Sahelian landscapes, land use/land cover (LULC) dynamics and climate variability are already known to affect the water cycle. In its current practice however, hydrological modelling does not account for LULC changes. This issue pertains to rapidly evolving watersheds and might result in critical inaccuracies in the simulated processes. In this study, the Soil and Water Assessment Tool (SWAT) model was used to simulate surface runoff in the small Sahelian watershed of Tougou, which underwent significant LULC changes between 1952 and 2017. Based on rainfall/runoff data acquired from 2004 to 2018, the SWAT model was calibrated under two scenarios : a static land use scenario (SLU) using a single LULC map (in 1999) and a dynamic land use scenario (DLU) integrating 3 LULC maps (1999, 2009 and 2017). The DLU scenario estimated with higher accuracy surface runoff, deep aquifer infiltration and actual evapotranspiration processes. Based on the calibrated parameters, surface runoff was simulated during the historical period 1952-2003 under four scenarios with static LULC maps (in 1952, 1973, 1986 and 1999) opposed to a fifth scenario integrating these LULC maps dynamically. The DLU scenario was found to be more effective at picturing the so-called Sahelian paradox (i.e the increase in surface runoff despite the decrease in rainfall), reported in the literature for small watersheds in the Sahel. The analysis of variability revealed that fluctuations in surface runoff were both influenced by rainfall and LULC changes. Furthermore, the isolated contributions of climate variability and LULC changes on surface runoff showed that LULC conditions played a dominant role (ηlulc = +393.1%) in the runoff increase over climate (ηcl = -297%) during the historical period. These results highlight the importance of accounting for LULC dynamics in hydrological modelling and advocate the development of integrated modelling frameworks for hydrologists and water resource managers.