Revisiting the dynamic simulation model of land-use changes in the Sudano-sahelian countries of Africa (SALU): An open-access application to Senegal

This is the authors' abstract. We don't add key points for this paper.

In the Sahel, the sustainability of human land-use is threatened by climate change and demographic pressure. Research is urgently needed to provide governments with relevant leads to prevent land degradation and ensure the population access to resources such as food and fuelwood. Little observation data is available, and canonical land-use models have difficulty representing Sahelian dynamics. One of the few models developed to fit characteristics of the region is the dynamic simulation model of land-use changes in the Sudano-Sahelian countries of Africa (SALU) developed by Stephenne and Lambin (2001). This study aims at adapting this model to the current state of the art, providing explicit calculation steps, to reconstruct past land-use dynamics in Senegal from 1961 to 2020. An extensive bibliographic search was used to obtain parameter value ranges and the model’s sensitivity to parameter uncertainties was assessed through δ -indices calculation. Applying the model at national scale predicted trends coherent with available literature, with agricultural expansion leading to deforestation, and a switch to intensification in the mid-1990s affecting both livestock forage consumption and fallowing time. Calculating land-demands in the Groundnut basin, a subregion of Senegal, showed the limits of the model when downscaling, with current hypotheses insufficient to reconstruct land-uses in the region when faced with demands too high to be satisfied by the local production. This work opens perspectives to refine land-use modeling in the Sahel, including prospective scenarios. • We release a new version of the land-use model of Sudano-Sahelian Africa (SALU). • We assess uncertainties and identify the most influential parameters. • We reconstruct Senegal’s past land-uses from production and demand (1961–2020). • Good agreement with a national reference atlas with limitations subnationally. • Calculation steps, parameter values, and code are made fully available.