FAIRMODE WG5 Bias-Correction Intercomparison Dataset - Phase 1

This dataset supports the FAIRMODE Working Group 5 intercomparison exercise on bias correction for air quality projections. It is based on a synthetic modelling experiment in which participants estimate a bias correction by comparing a perturbed base-case simulation with point-based synthetic observations, and then apply the correction to a perturbed emission scenario. The dataset contains annual mean concentrations of PM2.5, PM10 and O3. PM2.5 and PM10 are expressed in µg m⁻³, while O3 is expressed in ppb. Data are provided as gridded model fields and as point values extracted from the reference base case. The archives are clearly divided between data available during the blind exercise and synthetic truth data released afterward. The blind dataset includes the perturbed gridded base case, point-based synthetic observations and the perturbed gridded scenario. The truth dataset includes the complete gridded reference base case and reference scenario, enabling post-exercise evaluation of the bias-correction methods.