In this study, the Soil and Water Assessment Tool (SWAT) model is used to assess changes in surface runoff between the baseline (1995–2014) and future (2031–2050) periods in the Tougou watershed ( 37 km 2 ) in Burkina Faso. The study uses a combination of land use maps (for current and future periods) and a bias-corrected ensemble of 9 CMIP6 climate models, under two warming scenarios. An increase in rainfall (13.7% to 18.8%) is projected, which is the major contributor to the increase in surface runoff (24.2% to 34.3%). The land use change narrative (i.e. conversion of bare areas to croplands) is expected to decrease in surface runoff, albeit minor in comparison to the effect of future climate change. Similar findings are observed for annual maximum surface runoff. This study sheds light on the need to consider simultaneously future climate and land use in framing water management policies.

Appropriate methods and tools accessibility for bi-dimensional flow simulation leads to their weak use for floods assessment and forecasting in West African countries, particularly in urban areas where huge losses of life and property are recorded. To mitigate flood risks or to elaborate flood adaptation strategies, there is a need for scientific information on flood events. This paper focuses on a numerical tool developed for urban inundation extent simulation due to extreme tropical rainfall in Ouagadougou city. Two-dimensional (2D) shallow-water equations are solved using a finite volume method with a Harten, Lax, Van Leer (HLL) numerical fluxes approach. The Digital Elevation Model provided by NASA’s Shuttle Radar Topography Mission (SRTM) was used as the main input of the model. The results have shown the capability of the numerical tool developed to simulate flow depths in natural watercourses. The sensitivity of the model to rainfall intensity and soil roughness coefficient was highlighted through flood spatial extent and water depth at the outlet of the watershed. The performance of the model was assessed through the simulation of two flood events, with satisfactory values of the Nash–Sutcliffe criterion of 0.61 and 0.69. The study is expected to be useful for flood managers and decision makers in assessing flood hazard and vulnerability.

Soil erosion varies in space and time. As the contributing surface area increases, heterogeneity effects are amplified, inducing scale effects. In the present study, soil erosion processes as affected by the observation scale and the soil surface conditions are assessed. An experimental field scale setup of 18 plots (1–150 m²) with different soil surface conditions (bare and degraded, cultivated) and slopes (0.75–4.2%) are used to monitor soil losses between 2010 to 2018 under natural rainfall. The results showed that soil loss rates range between 2.5 and 19.5 t.ha⁻¹ under cultivated plots and increase to 12–45 t.ha⁻¹ on bare and degraded soils, which outlines the control of soil surface conditions on soil erosion. At a larger scale (38 km²), soil losses are estimated at 2.2–4.5 t.ha⁻¹, highlighting the major contribution of scale. The scale effect is likely caused by the redistribution of sediments in the drainage network. These findings outline the nature and contribution of the emerging and dominant soil erosion processes at larger scales. At the plot scale, however, diffuse erosion remains dominant, since surface runoff is laminar and sediment transport capacity is limited, resulting in lower soil erosion rates.