11:00 AM - 11:15 AM
[AAS12-20] Evaluation of performance of simulated secondary air pollutants by using air quality models for the Kanto area in summer 2011
Keywords:Air quality model, secondary pollutants, Urban Air quality
The urban air quality model inter comparison study in Japan (UMICS) was started to improve performances of air quality models (eg. Chatani et al., 2014, Shimadera et al., 2014). UMICS showed some critical problems immanent in the air quality models. For example, the models tended to overestimate NO3- but to underestimate OA, although simulated PM2.5 concentrations were reasonable with comparing to observations at Kanto area (Shimadera et al., 2014). In terms of O3, the models reproduced well the diurnal and inter-diurnal variations in the O3 concentrations at most observational stations in Kanto area but tended to overestimate nighttime O3 and to underestimate daytime O3 at several observational stations (Morino et al., 2010), therefore these models might have a risk failing to predict some of high pollution events.
In order to find the causes of discrepancies between the simulated and observed concentrations of secondary pollutants, constituents of PM2.5 and O3, in this study, air quality simulations were performed using the Weather Research and Forecasting (WRF) model for a meteorological model and the Community Multi-scale Air Quality (CMAQ) model system for a chemical transport model under the following different model settings. Meteorological analysis data (FNL/NCEP and MSM/JMA) with both different temporal and spatial resolutions were used as for input data of meteorological simulations by the WRF model, respectively. Additionally, the updated JEI-DB (JATOP Emission Inventory Data Base) was used for input emission information for the CMAQ model. Performances of these models under different settings were evaluated by comparing with observed concentrations (O3, PM2.5, and constituents of PM2.5) of secondary pollutants at Kanto area, which were provided by UMICS and MOE. These results are also compared with the previous studies (eg. Shimadera et al., 2014).