Evaluation of Chemical Lateral Boundary Conditions from the GEOS-5 Global Chemical Transport Model to the Performance of Operational NAQFC

As a step to improve the National Air Quality Forecasting Capability (NAQFC) that is housed in NOAA to provide the operational air quality forecast over the continuous United States using the Community Multi-scale Air Quality model (CMAQ) v5.0.2, we have developed the framework to implement the Goddard Earth Observing System Model, version 5 (GEOS-5) to furnish both gaseous and aerosol lateral boundary conditions (LBCs) to CMAQ. A table that maps GEOS-5 chemical species to CMAQ species has also been developed in the process. We test-ran CMAQ with the GEOS-5’s monthly-mean chemical LBCs and 3-hourly varying LBCs for June and July of 2014. The results show that the use of the GEOS-5 chemical LBCs captured the transport of fire smoke originating from Canada, which was otherwise impossible with the static climatology chemical LBCs implemented in the current NAQFC. As a result, the bias of the modeled CO reduced by 10% while the bias of the simulated PM2.5 decreased by more than 70% as compared to the baseline simulation that used the climatology LBCs. Meanwhile the statistics of root mean square error (RMSE), correlation coefficient (r), and agreement index (IOA) for CO and PM2.5 all improved to some degrees. The modeled surface ozone carried reduced RMSE, better r and IOA, but did bear an increased positive bias. The 3-hourly varying GEOS-5 chemical LBCs generally yielded better correlation than the monthly mean GEOS-5 LBCs.