Mapping and Forecasting
The center has expertise in air quality forecasting for ozone and PM2.5 using a series of statistical models including meteorological pattern classification, factor analysis, and multiple linear regression. In addition, the center incorporates artificial neural network, enhanced model calibration and evaluation procedure to improve prediction accuracy and model performance. These forecasts are directly linked to ozone and PM2.5 mapping system to visualize regional air quality forecasts in real time.