Land use changes are mainly driven by human activities and climate change. Uncontrolled land use can alter surface runoff and affect several sectors of activity such as agriculture, hydropower, drinking water supply, sanitation, etc. The main objective of this study is to assess the influence of changes in land surface conditions on runoff in the Haut Bandama catchment. For this purpose, LANDSAT satellite images from 1990, 2006 and 2020 were used. These images were processed by the maximum likelihood supervised classification method under ENVI to establish the land use maps. Based on the past land use trends, a future prediction (2035 and 2050) is made using the Land Change Modeler. The diachronic analysis of the land use maps for the period 1990-2020 showed a decline in the area of forest and savannah patches, followed by that of water bodies. Conversely, the area of built-up and bare land as well as crops and fallow land is increasing. The same dynamics are predicted for the 2035 and 2050 horizons, where the savannah class dominates most of the net changes. The dynamics of the land surface condition has led to an increase in the runoff coefficient from 1990 to 2020 from 18.5% to 30.7% respectively. This evolution continues until 2050 with 34.5% in 2035 and 36.4% in 2050.