The impact of climate change on land productivity and agricultural production and thus, food security is increasing in Uganda. Annual and inter-seasonal climate variations in the Lake Kyoga basin have been documented to affect land productivity and rainfed crop production systems in the basin. Smallholder farmers have learned to cope with climate-induced yield losses through an array of strategies. However, the community adaptive capacity being a function of exposure and sensitivity to climate shocks is likely to vary across landscapes. Understanding of these farm-based land management practices with their associated impacts on crop yields along land productivity levels is limited. Yet this information is vital for appropriately scaling out the best combination of farm-based land management practices for improved crop production. Therefore, Identifying and quantifying the relationships between crops yields and land productivity, climate change and land management practices, allows for better ways to close the yield gaps, increase yield potentials and improve predictive capability of crop yield for climate change adaptation. Machine Learning offers an important opportunity as a decision support tool for crop yield prediction, including supporting decisions on what crops to grow and what to do during the growing season of the crops under changing climate and land management practices. Unfortunately, in Uganda, paucity of research work exists that deals with the issues of agricultural crop yield prediction under changing climate and different land management practices using artificial intelligence approach. Most of the research is based on predictions based on classical statistical models and biophysical models. Therefore, this project seeks to contribute to improved understanding of climate change adaptation and crop production through assessment of land productivity and crop production of different farm-based land management under changing climate in Kyoga basin and build capacity for Makerere University students in crop productivity and climate change adaptation assessment using artificial intelligence. Specifically, to, i) Assess the soil and land productivity of standing biomass in the Lake Kyoga basin; ii) Determine sustainable farm-based land management practices for climate change adaptation in the Kyoga basin; and iii) Determine the current and future spatio-temporal responses of rainfed maize and sorghum yields to climate change and farm-based land management practices using machine learning –based modeling in the Lake Kyoga basin. The study will be conducted in three agroecological zones of Lake Kyoga basin of Uganda, i.e. the semiarid rangelands; highland ranges and the Kyoga plains. The study will employ both extensive field surveys, Geospatial analysis and machine learning-based modeling approaches. The study will provide opportunity to train two Master of Science (MSc) students in crop productivity, climate change adaptation and Artificial intelligence studies. The study outputs include; two (2) MSc. Students trained as part of human capital development in climate change adaptation, land productivity and application of artificial intelligence for crop production prediction; An inventory of Land productivity indices and farm-based sustainable land management practices; two research theses; two peer reviewed journal articles published; a catchment machine learning based model developed for predicting crop yields; and one policy brief.