Can Machine Learning Forecast Rice Yields in Data-Constrained Settings? Satellite Climate Data, National Crop Statistics, and Lessons from Sierra Leone
arXiv:2606.13959v1 Announce Type: new Abstract: Sierra Leone's agriculture operates with almost no data-driven decision support, and no published machine learning study has examined the country's crop yields. We ask whether rice yield can be forecast from data Sierra Leone currently has. Using 25 years of FAOSTAT production data (2000-2024) for nine major crops, we train XGBoost, Gradient Boosting, and Random Forest under a strict anti-leakage protocol with expanding-window walk-forward evaluation across seven held-out years, benchmarked against naive persistence. No model trained on crop stat
The proliferation of open-source machine learning models and readily available satellite climate data makes previously data-poor regions viable for AI-driven agricultural solutions.
This demonstrates concrete applications of AI in developing economies to address fundamental issues like food security, potentially improving agricultural output and stability without relying on extensive local data infrastructure.
The ability to forecast crop yields using external data sources and sophisticated ML methods opens new pathways for agricultural planning and resource allocation in data-constrained nations.
- · Developing nations with limited agricultural data
- · Food security initiatives
- · Agricultural technology sector
- · Farmers in Sierra Leone
- · Traditional agricultural forecasting methods
- · Regions overly reliant on manual data collection for agriculture
Improved resource allocation and famine early warning systems in regions like Sierra Leone.
Increased adoption of satellite and AI-driven agricultural tools across other data-poor nations, fostering broader digital transformation in agriculture.
Reduced food price volatility and enhanced regional stability as food security improves through data-driven planning.
This signal links to a primary source. Continuum Brief monitors and indexes it as part of the live intelligence stream — we do not republish source content.
Read at arXiv cs.LG