AI for climate resilience

As the impacts of climate change grow more severe, UNICEF is focused on designing systems that not only track and anticipate climate risks but also enable targeted interventions that safeguard children and strengthen their resilience.

• Early warning systems: Similie (Timor-Leste) integrates groundwater, precipitation and vegetation data to forecast water shortages, enabling early mitigation. MapAction (Mali) uses AI and geolocation to enable citizen reporting of floods, illegal waste and other environmental hazards.  Equinoct (India) combines community-sourced data with hydrological models for flood forecasting, supporting local disaster management. Mozn (Libya) processes global forecasts and relays real-time meteorological data to first responders, local authorities and citizens.
• Disaster risk reduction: Map&Rank (Cameroon) uses machine learning, drone imagery, and community data to predict climate hazards and inform preparedness. In Malawi, UNICEF has tested drones for flood modeling.
• Anticipatory action: predictive models can trigger early interventions before disasters strike, reducing response costs and protecting child wellbeing. UNICEF’s Frontier Data Network has developed Ahead of the Storm, an initiative to predict child-specific impacts of natural disasters and better enable proactive humanitarian response.
• Parametric insurance: eSusFarm (Uganda) uses blockchain and AI to automate payouts to smallholder farmers after climate disasters.
• Clean energy: Grinplus (Uruguay) aggregates small solar generators to enable them to access renewable energy credits and financing.
• Groundwater modelling: We are exploring the potential of AI to make predictions about groundwater levels and drought, inform borehole drilling locations, and accelerate the impact of UNICEF’s More Water More Life initiative, helping provide vital access to water in drought-prone areas.