OEMC project use case: Air quality assessment at continental scale
Description
Air pollution is a health risk to millions of people in Europe. Heavier pollution occurs in densely populated or industrial areas where we can observe more combustion of fossil fuels. Relevant indicators for air quality are the concentrations of particles with sizes of ~10 µm (PM10) and ~2.5 µm (PM2.5), ozone (O3), and nitrogen-dioxide (NO2).
The EEA's European Topic Center for Human Health and the Environment (ETC HE) has developed a complex approach for mapping these indicators, the Regression-Interpolation-Merging-Mapping (RIMM) method. It combines reliable in situ measurements collected by the EEA member states with atmospheric transport model outputs (ERA5, CAMS), satellite observations (Sentinel-5P TROPOMI), and anthropogenic factors (land cover, population density, traffic) to produce Europe-wide annual maps at 1 km spatial resolution.
We implement this method as open source software and apply it to produce air quality maps for all four indicators at annual, monthly, and daily frequency. Minor activities explore possible improvements of this method with regard to prediction errors or computational cost.
Stakeholder needs:
- Open-source tools for data pre-processing and predictive modelling of air quality indicators, including uncertainty estimation
- Annual, monthly, and daily air quality maps for 2015-2023
Planned implementation:
- Develop and document R code implementing the RIMM method
- Produce and publish annual, monthly, and daily maps of PM10, PM2.5, O3, and NO2 and respective prediction errors as cloud-optimized GeoTIFFs