Google Flood Hub — Bangladesh Water Development Board's AI-Powered River Flood Alert System
Bangladesh
Since 2020, Google Research's machine-learning flood model — which forecasts river levels even in ungauged basins — has combined with …
Nigeria · Abuja · See the Nigeria profile
Nigeria's Hydrological Services Agency says it deployed deep-learning models trained on 200+ years of flood data for its 2026 Annual Flood Outlook, citing ~88% precision and flagging 14,118 high-risk communities — widely reported but not independently verified.
NIHSA is Nigeria's federal hydrological forecasting agency, responsible for the country's Annual Flood Outlook (AFO), which classifies local government areas (LGAs) into high, moderate and low flood-risk categories each year. Nigeria has suffered severe flooding in recent years, including major disasters in 2022 and 2024, driving pressure to modernise forecasting beyond traditional hydrological methods.
In press remarks reported in May 2026, NIHSA Director-General Umar Mohammed said the agency has deployed artificial intelligence and deep-learning models trained on more than 200 years of historical flood data, and that forecasting precision has risen to about 88%, which he said surpasses World Meteorological Organization standards. The 2026 Annual Flood Outlook, drawing in part on this tooling and NIHSA's 250+ monitoring stations, identified 14,118 communities across 266 LGAs in 33 states and the Federal Capital Territory as high flood risk, and a further 15,597 communities across 405 LGAs as moderate risk. NIHSA says this sits alongside a new community-based flood reporting system and a flood micro-insurance scheme.
The claim was reported by Nigeria's national news agency and multiple domestic outlets (Guardian Nigeria, News Agency of Nigeria, EnviroNews Nigeria), giving it distribution across several independent news domains.
Every source located traces back to the same NIHSA press briefing; no independent technical evaluation, peer-reviewed study, or third-party audit of the 88% precision figure, its methodology, or what "precision" is measured against was found. Unlike other AI forecasting cases, this entry rests on a self-reported government figure amplified by domestic media rather than externally verified evidence, and should be read accordingly.
Read the full analysis: https://guardian.ng/news/30000-communities-face-flood-threat-as-fg-deploys-ai-forecast-eyes-water-economy/
Implementation detail (cost, timeline, staffing, conditions for success) is not yet available for this practice.
Where this practice's information was retrieved from, and when.
Bangladesh
Since 2020, Google Research's machine-learning flood model — which forecasts river levels even in ungauged basins — has combined with …
Kenya
Researchers at the University of Sussex and Kenya's RCMRD built machine-learning models forecasting NDMA's official VCI3M drought indicator up to …
Switzerland
SPSP integrates AI-powered genomic analysis to detect pathogen variants across Switzerland. Mandated by the Federal Office of Public Health since …
Singapore
Singapore's National Environment Agency runs a peer-reviewed machine-learning system that forecasts dengue cases up to three months ahead. In the …
Open full copilot Grounded in cited practices — always check the sources.