Simulating present-day and future air quality as climate changes: Model evaluation

The global-regional climate-air pollution modeling system (GRE-CAPS) has been developed, coupling an existing general circulation model/chemical transport model (GCM/CTM), a regional meteorological model, and a regional chemical transport model. This system is intended to enable studies of the effects of changes in climate, intercontinental transport, and emissions on regional and urban air quality. The GRE-CAPS system consists of the GISS II′ GCM/CTM, the MM5 regional meteorological model, and the PMCAMx regional CTM. The modeling system is evaluated for the present day, with comparisons between model-predicted, measured ozone, and speciated PM2.5 concentrations. The ability of the model to predict present-day concentrations of ozone and PM2.5 is compared to that of PMCAMx when used for retrospective modeling. Comparisons between model-predicted temperatures and precipitation are also made. The model was used to simulate five present-day Januaries and six present-day Julys. The biases and errors in GRE-CAPS-predicted ozone concentrations were similar to those of PMCAMx when used for standard retrospective modeling. The fractional biases in mean daily peak ozone concentration and mean daily maximum 8-h average ozone concentration are both <10%. The model-predicted distribution of peak hourly and daily maximum 8-h average values agreed rather well with the measured distribution. There is less agreement between the model and measurements in the number of hours with ozone mixing ratios >70 or 80 ppb, though this is also the case with standard PMCAMx modeling. The predictions of PM2.5 concentrations by GRE-CAPS were also of similar quality to those of PMCAMx driven by historical meteorology. The fractional biases in the predictions of total PM2.5, sulfate, ammonium, and nitrate were all <25% in both January and July. The model agrees well with organic PM2.5 measurements from the IMPROVE network, though there is less agreement with measurements from the STN network. The GRE-CAPS system is shown to reproduce ozone and PM2.5 concentrations for the present day rather well, with model performance similar to that of PMCAMx for standard retrospective episode modeling with historical meteorology. GRE-CAPS will be used in future studies to examine the effects of changes in climate, global emissions, and intercontinental transport on regional air quality.