Random Forest classification of land cover from Sentinel satellite images in 2021 for the Agro2Ecos project area, Maradi, Niger

This dataset was acquired using the combination of Sentinel-1 radar and Sentinel-2 optical image time series with indices (BI2, NDVI, MSAVI2, NDII). A total of 112 sentinel images were acquired in 2021 allowing a coherent and reliable mapping of land use in relation to the cropping systems around Maradi, Niger. Rainfed cropping systems (millet, cowpea, sorghum, groundnut and nutsedge) are well classified and occupy more than 94% of the cultivated area. In contrast, irrigated crops occupy only 1% of the total area. This work was conducted during the Agro2Ecos research project, funded by the McKnight Foundation, the overall objective of which was the evaluation of organic waste reuse as a means to improve soil quality and food production in West Africa. The dataset highlights the dominance of rainfed cropping systems and points towards the need to further investigate the agro valorization of organic waste in these cultivated areas.