Application of WRF/Chem over East Asia: Part II. Model improvement and sensitivity simulations. (January 2016)

• Record Type: Journal Article
• Authors: Zhang, Yang; Zhang, Xin; Wang, Kai; Zhang, Qiang; Duan, Fengkui; He, Kebin
• Is Part Of: Atmospheric environment. Volume 124:Part B(2016)
• Page Start: 301
• Page End: 320
• Publication Date: 2016-01
• Subjects: WRF/Chem -- Model improvement -- Sensitivity simulation -- Aerosol direct effects -- Aerosol indirect effects -- East Asia; Air -- Pollution -- Periodicals; Air -- Pollution -- Meteorological aspects -- Periodicals; 551.51
• DOI: 10.1016/j.atmosenv.2015.07.023
• Languages: English
• ISSNs: 1352-2310

Abstract

Abstract: To address the problems and limitations identified through a comprehensive evaluation in Part I paper, several modifications are made in model inputs, treatments, and configurations and sensitivity simulations with improved model inputs and treatments are performed in this Part II paper. The use of reinitialization of meteorological variables reduces the biases and increases the spatial correlations in simulated temperature at 2-m (T2), specific humidity at 2-m (Q2), wind speed at 10-m (WS10), and precipitation (Precip). The use of a revised surface drag parameterization further reduces the biases in simulated WS10. The adjustment of only the magnitudes of anthropogenic emissions in the surface layer does not help improve overall model performance, whereas the adjustment of both the magnitudes and vertical distributions of anthropogenic emissions shows moderate to large improvement in simulated surface concentrations and column mass abundances of species in terms of domain mean performance statistics, hourly and monthly mean concentrations, and vertical profiles of concentrations at individual sites. The revised and more advanced dust emission schemes can help improve PM predictions. Using revised upper boundary conditions for O3 significantly improves the column O3 abundances. Using a simple SOA formation module further improves the predictions of organic carbon and PM2.5 . The sensitivity simulation that combines all above model improvements greatly improves the overall model performance. For example, the sensitivity simulation gives the normalized mean biases (NMBs) of −6.1% to 23.8% for T2, 2.7–13.8% for Q2, 22.5–47.6% for WS10, and −9.1% to 15.6% for Precip, comparing to −9.8% to 75.6% for T2, 0.4–23.4% for Q2, 66.5–101.0% for WS10, and 11.4%–92.7% for Precip from the original simulation without those improvements. It also gives the NMBs for surface predictions of −68.2% to −3.7% for SO2, −73.8% to −20.6% for NO2, −8.8%–128.7% for O3, −61.4% to −26.5% for PM2.5, and −64.0% to 7.2% for PM10, comparing to −84.2% to −44.5% for SO2, −88.1% to −44.0% for NO2, −11.0%–160.3% for O3, −63.9% to −25.2% for PM2.5, and −68.9%–33.3% for PM10 from the original simulation. The improved WRF/Chem is applied to estimate the impact of anthropogenic aerosols on regional climate and air quality in East Asia. Anthropogenic aerosols can increase cloud condensation nuclei, aerosol optical depth, cloud droplet number concentrations, and cloud optical depth. They can decrease surface net radiation, temperature at 2-m, wind speed at 10-m, planetary boundary layer height, and precipitation through various direct and indirect effects. These changes in turn lead to changes in chemical predictions in a variety of ways. Highlights: WRF/Chem performance is significantly improved with improved inputs and treatments. Adjusting emissions and their vertical distributions lead to most improvement. WRF/Chem v3.3.1 with improved inputs and treatments outperforms MM5/CMAQ. Asian anthropogenic aerosols play an important role in regional climate & air quality.