The purpose of this project is to develop and maintain forecasting systems to facilitate daily forecast of air quality over the Taiwan area. The main accomplishments include: (1) Maintain the “Air Quality Forecast Auxiliary System” to provide the information for daily forecast and develop probability forecasting for various classes of air quality levels; (2) Maintain and operate the “East Asia Mineral Dust Model” to provide results for 120 hours forecast, as well as develope dynamic numerical models (CMAQ、WRF-CHEM) coupled with mineral dust deflation scheme to decrease the run-time efficiently; (3) Maintain and operation the “Taiwan local Dust Model” to provide simulations for 72-hours forecast, and improve the dust deflation coefficients for the main river beds in central and southern Taiwan; (4) Maintain and continue to upgrade air quality numerical models (biomass burning included), to provide simulations of local and long range transport of air pollutants, and to assist routine operation for 72-hour forecasts; Evaluate inverse modeling for the adjustment of emission inventories over Taiwan and and East Asia so as to improve future simulation of pollutant concentration ; (5) Assist the porting and testing of all the above numerical systems to the High Performance Computation facility of the Central Weather Bureau (CWB) under the agreement between CWB and EPA; (6) Develop the MODIS AOD retrieval system with a resolution of 2 km to improve aerosol concentration estimate and transfer the technology to CWB under the agreement between CWB and EPA; (7) Analyze the relationship between 10-km AOD and PM2.5 concentration for application in aerosol forecasting: (8) Conduct detailed analyses of selected air pollution episodes so as to improve air quality forecasts; (9) Provide special duty manpower in EPA for daily forecasts of air pollution. On-going works for the system sustainable development will focus on: elevating the air-quality forecasting capability by integrating the statistical model and numerical model, developing different numerical forecast models to offer diverse information for a better objective forecast, improvement of emission inventory for air quality numerical modeling, continuing development of MODIS AOD retrieval system, as well establishing a more accurate land-use type classification for a local-area mineral dust model, so that these quality daily forecast systems can be better suited for the Taiwan area that has complex terrain and geographical features.

air-quality forecast, mineral dust model, East Asia mineral dust, local mineral dust, MOIDIS, AOD