
This research focuses on improving crop yield estimation in smallholder farming systems, particularly in regions like Sub-Saharan Africa where fragmented fields and limited data present challenges. The study proposes an integrated approach that combines Earth Observation (EO) data, Artificial Intelligence (AI), and process-based Crop Modelling (CM) to overcome these limitations and enhance yield prediction accuracy in data-scarce environments. The primary goals include assessing the value of high-resolution satellite imagery for understanding variations in smallholder farms, refining crop models with local agricultural and climate data, and developing AI models for scalable yield prediction across different regions. A key aspect involves fusing data from multiple sources, such as satellite imagery, weather records, and ground observations, to build robust models. The research will be validated using field yield data from various agroecological zones in Kenya to ensure the models are reliable and transferable. Ultimately, this research aims to create a validated and scalable framework for estimating crop yields. This framework will support climate-resilient agricultural planning, improve early warning systems for food security, and contribute to digital advisory services tailored for smallholder farmers. By advancing data-driven tools for crop productivity monitoring in under-resourced regions, the study contributes to the broader goal of enhancing food security.
This research aims to develop advanced tools and methodologies applicable to agricultural planning, food security initiatives, and the development of digital advisory services for farmers. Graduates with expertise in integrating Earth Observation, AI, and Crop Modelling can pursue roles in agricultural research, international development agencies, governmental agricultural bodies, and private sector companies focused on agritech and sustainable farming.
Fee information is not available for this research topic.