Development of Source Object-Based Model for Emissions (SOME) for Multiscale Anthropogenic Emissions in Urban Environments

The needs of high resolution air quality modeling are increasing in urban environments for assessment of human health impacts. Computational Fluid Dynamics (CFD) and Large-Eddy Simulations (LES) with chemistry are applied for building-scale (O(10 m)) simulations and local air quality models with full chemical processes are applied for inner city-scale (O(100 m)), for which a flexible emission processing tool is required to support multi-scale air quality models. This study presents the Source Object-based Model for Emissions (SOME) that is a new novel emission processing model developed for support of multi-scale air quality models. The SOME develops anthropogenic emissions of all buildings and roads/vehicles by top-down and/or bottom-up estimation approaches. The new emission processing model was applied for the Seoul metropolitan area (40×30 km²) to produce gridded anthropogenic emissions for local- and micro-scale air quality modeling of the Weather Research and Forecasting-Chemistry (WRF-Chem). The SOME top-down emissions compared well with the SOME bottom-up emissions with differences of -11.0% in CO, -1.1% in NO_x, 0.3% in SO_x, and -4.8% in NH₃ at a building sector and 28.5% in CO, -23.2% in NO_x, and 8.2% in VOCs at a road sector in commercial/residential area. The SOME model represented more realistic spatio-temporal variations than a traditional method. In order to investigate fidelity of the emissions, the air quality modeling has been conducted using the WRF-Chem model, which was configured over the East Asia domain (DX = 32.4 km) and nested down to the Seoul metropolitan area domain (DX = 0.133 km) using a one-way nesting technique. The finest domain are simulated using a Large-Eddy Simulations (LES) mode and the simulated ozone, nitrogen dioxide, and fine particulate matters (PMs) are compared against intensive in-situ field measurements. Details of model structure of the SOME model and its potential capability will be presented.