FORECAST-AIR aimed to innovate health early warning systems by integrating air quality forecasts with health risk modelling, addressing a key gap in current systems that rely solely on pollutant thresholds and ignore population vulnerability. The project advanced technically and scientifically in the following areas:
Database generation and formatting: FORECAST-AIR exploited the EARLY-ADAPT database -a harmonised, continental-scale dataset combining cause-specific mortality data with environmental and socioeconomic variables. This database enabled fine-scale, high-resolution analyses across Europe and laid the foundation for pan-European epidemiological modelling.
Epidemiological modelling: We developed and validated models that quantified the short-term effects of air pollutants (PM2.₅, NO2, O₃, etc.) on mortality, stratified by demographic and socioeconomic groups and causes of death.
Health-based forecasting: We transformed air quality forecasts into health impact estimates using our epidemiological models. This represents a shift from conventional physical forecast systems to health-driven early warnings that better reflect population-level impacts.
Predictability and system design: We assessed the forecast window for both pollutant concentrations and predicted health outcomes, identifying lead times sufficient for decision-making. This work supports trust and usability by public health authorities and enables evidence-based alert thresholds.
Online platform: We expanded the early warning system platform that was originally created in a previous ERC PoC project (HHS-EWS) to assess the health impacts of environmental temperatures (Forecaster.Health
https://forecaster.health/(öffnet in neuem Fenster)). Within FORECAST-AIR, we added and integrated the counterpart air pollution component, making Forecaster.Health a more complete, multi-exposure tool that now also captures the health effects of air quality. The health predictions that appear in the early warning system platform are automatically updated and uploaded every day by a series of protocols and scripts that we wrote, and that are synchronised in real time with the time of release of the daily air quality forecasts in the CAMS data store.
Scientific outputs and impact: FORECAST-AIR led to publications in Nature Medicine, Nature Communications, Nature Climate Change, The Lancet Planetary Health, and Science Advances.
FORECAST-AIR provides a robust, health-centred methodology for early warnings of air pollution impacts, supporting more targeted and equitable public health interventions. By combining environmental epidemiology with real-time forecasting and vulnerability mapping, the project offers a breakthrough framework in climate services for health, aligning with EU goals on air quality, health equity, and early action.