Skill in forecasting extreme ozone pollution episodes with a global atmospheric chemistry model

摘要

From the ensemble of stations that monitor surface air quality over theUnited States and Europe, we identify extreme ozone pollution events and findthat they occur predominantly in clustered, multiday episodes with spatialextents of more than 1000 km. Such scales are amenable to forecasting withcurrent global atmospheric chemistry models. We develop an objective mappingalgorithm that uses the heterogeneous observations of the individual surfacesites to calculate surface ozone averaged over 1° by 1° gridcells, matching the resolution of a global model. Air quality extreme (AQX)events are identified locally as statistical extremes of the ozoneclimatology and not as air quality exceedances. With the University ofCalifornia, Irvine chemistry-transport model (UCI CTM) we find there is skillin hindcasting these extreme episodes, and thus identify a new diagnosticusing global chemistry–climate models (CCMs) to identify changes in thecharacteristics of extreme pollution episodes in a warming climate.