Rwanda launched Geo-Hub, a machine-learning platform combining satellite imagery and field data to map crop area and climate stress, piloted in four districts for rice, maize, potatoes and beans, with the FAIR Forward-backed system currently restricted to institutional users.
Details
Maturity
Pilot
Promoter
National Institute of Statistics of Rwanda (NISR), with Rwanda Space Agency (RSA) and Ministry of Agriculture and Animal Resources (MINAGRI)
Period
2025–ongoing
Keywords
agriculture, earth observation, official statistics, climate resilience
Context
Rwanda's National Institute of Statistics (NISR) launched Geo-Hub in December 2025, a geospatial platform built with the Rwanda Space Agency (RSA) and the Ministry of Agriculture and Animal Resources (MINAGRI) that combines satellite imagery with field survey data and machine-learning models to monitor farmland in near real time. The system is being piloted across four districts, tracking rice, maize, potatoes and beans, and produces maps of crop area, growth condition and climate risk (such as drought or flood exposure) to support planting and harvest planning. The government's stated roadmap is to expand crop-area estimates to all districts and to add seven-day disease-risk forecasts during 2026. Technical and financial support comes from Germany's FAIR Forward initiative (funded by the Federal Ministry for Economic Cooperation and Development, BMZ) and the Alliance of Bioversity International and CIAT.
Implementation
Indicative cost
Medium (€50k–€500k) — No budget figure is published; classified as medium cost given international technical-partner backing (FAIR Forward/BMZ, Alliance of Bioversity International and CIAT) for a still-limited four-district pilot.
Time to results
Short (< 1 year) — Launched December 2025 as a four-district pilot, with a roadmap to national coverage and disease-risk forecasting during 2026.
Staffing & skills
National Institute of Statistics of Rwanda (NISR), Rwanda Space Agency (RSA), Ministry of Agriculture and Animal Resources (MINAGRI), Technical support from FAIR Forward (BMZ) and the Alliance of Bioversity International and CIAT
Conditions for success
Multi-institutional collaboration spanning statistics, space and agriculture agencies plus international technical partners
Combining satellite imagery with field survey data to offset satellite-only classification limits
Common failure modes
Intercropped fields remain harder for satellite-based models to classify accurately
Access is currently restricted to government and research institutions, not a public portal
No independent accuracy, yield-improvement, or cost figures have been published as of the pilot's first months
Do you run this practice?
Claim it —
verified implementers get a public contact pathway and can propose corrections.
Data sources
Where this practice's information was retrieved from, and when.
A $12.7m Korean-funded project has installed an AI-assisted flood forecasting system and warning centre for the Pasig-Marikina-Tullahan river basins, intended …
