Mapping land suitability for informal, small-scale irrigation development using spatial modelling and machine learning in the Upper East Region, Ghana

Small-scale irrigation has gained momentum in recent years as one of the development priorities Sub-Saharan Africa. However, farmer-led is often informal with little support from extension services and a paucity data on land suitability for irrigation. To map spatial explicit dry season small-scale irrigation, we developed method using an ensemble boosted regression trees, random forest, maximum entropy machine learning models Upper East Region Ghana. Both biophysical predictors including surface groundwater availability, climate, topography soil properties, socio-economic which represent demography infrastructure such accessibility to cities proximity roads were considered. We assessed that 179,584 ± 49,853 ha suitable dry-season when only variables are considered, 158,470 27,222 included alongside predictors, representing 77-89% current rainfed-croplands. Travel time cities, small reservoirs, exchangeable sodium percentage, runoff can be potentially stored population density, roads, elevation percentile top suitability. These results suggested availability water alone not sufficient indicator area This calls strategic road improvement farmers market accessibility. The should put local context demographic indicators, development.