The model employed in this research was the Regional Climate Model version 3 (RegCM3) developed at the Abdus Salam International Centre for Theoretical Physics (ICTP) (Pal et al., 2007). Radiative transfer processes were computed from the Community Climate Model (CCM3) from the National Center for Atmospheric Research (NCAR) package (Kiehl et al., 1996). Land surface processes were described by the Biosphere-Atmosphere Transfer Scheme (BATS; Dickinson et al., 1993). The nonlocal planetary boundary layer scheme byHoltslag et al. (1990) was used to depict boundary layer processes. The Grell cumulus parameterization scheme (Grell, 1993) was used for describing convective precipitation and the subgrid explicit moisture scheme (SUBEX) for nonconvective precipitation. An important aspect of our work was that a desert dust module (Zakey et al., 2006) and simplified anthropogenic aerosol modules (Solmon et al., 2006) were implemented within the RegCM3. In the two modules, scattering and absorption of solar and infrared radiations by aerosols were parameterized along with optical properties of aerosols; transport, diffusion, dry deposition, and wet removal processes of aerosols were also taken into consideration. Aerosol module of the RegCM3 included sulfate, black carbon (BC; hydrophilic and hydrophobic), organic carbon (OC; hydrophilic and hydrophobic), and dust (Qian and Giorgi, 1999; Solmon et al., 2006; Zakey et al., 2006). The RegCM3 was used to simulate effects of aerosols on regional climate in different areas across the world (e.g.,Zhang et al., 2009; Ji et al., 2011; Ju and Han, 2011; Nair et al., 2012; Zanis et al., 2012).

The simulation domain covered East Asia with 144× 90 grid points, with the center being at 33° N, 105° E. A 60-km horizontal resolution and 18 nonuniform vertical layers were prescribed. The RegCM3 was driven by initial and 6-hr updating lateral boundary conditions from the National Centers for Environmental Prediction (NCAR/NCEP) reanalysis data (Kalnay et al., 1996).

In this study, dust emission was calculated online in RegCM3 using the soil texture dataset. The anthropogenic emission inventory of SO₂, BC, and OC was obtained from the David Streets’ 2006 emission inventory (http://www.cgrer.uiowa.edu/EMISSION_DATA_new/data/intex-b_emissions/) over the area of 0-53.5° N, 60-157.5° E and from the Emission Database for Global Atmospheric Research (EDGAR) database for the year 2000 (Solmon et al., 2006) covering other areas. Monthly mean data of biomass burning emissions were taken from the study results ofLiousse et al. (1996). The chemical conversion of SO₂ to SO₄^2- follows two oxidation pathways: one in the gas phase and the other in the aqueous phase.

Two continuous integrations were carried out with RegCM3: one from November 2002 to December 2003 and the other from November 2005 to December 2006. In case of both the simulations, the first month was considered as spin-up. In this article, only the situations occurring in summer (June-July-August, JJA) were analyzed.

Figure 1 shows the simulated clear-sky direct aerosols radiative forcing in two summers and their differences at the top of the atmosphere (TOA). The aerosols (dust, sulfate, BC, and OC) were assumed to mix externally, producing negative forcing at the TOA over most areas of East Asia, except for a small area in the north fringe of the Tibetan Plateau where positive effect was observed. Two dominant negative centers, one covering Northwest China and one spreading from Southwest China to the northeast of East Asia, were observed during the study period. The northwest dominant center (not entirely shown in this figure) was mainly induced by dust and it’ s out of the domain controlled by the EASM. This article focused only on the other key negative EASM-dominated areas in East Asia where the main sources of aerosols were human activities. In our study, we found that aerosols could induce stronger negative radiative forcing in the relatively weak EASM circulations in 2003 from Southwest China to Northeast East Asia, especially over the Sichuan Basin and North China, than in the relatively strong circulations in 2006. In the Sichuan Basin, the largest negative forcing was up to -30 W m^-2 in 2003, but only -12 to -15 W m^-2 in 2006. In North China, the negative forcing in 2003 was generally stronger by 3-6 W m^-2 than that in 2006.

Figure 1

Figure 1 Simulated aerosol-induced radiative forcing at the TOA in summer (JJA) for (a) 2003, (b) 2006, and (c) their difference (former minus latter) (units: W m-2).

At the surface (figure not shown), aerosols induced negative forcing around the whole research areas. Spatial distributions of the radiative forcing at the surface were similar to those at the TOA both in 2003 and 2006, but the negative forcing effects at the surface were stronger. In the EASM-dominated areas, the maximum negative forcing could exceed -40 W m^-2. In the Sichuan Basin, the difference in aerosol-induced surface forcing between the two summers could reach 15 W m^-2. In North China, negative radiative forcing was also stronger in 2003.

Figure 2 shows the simulated aerosol optical depth (AOD) at 350-640 nm. The simulated AOD by RegCM3 was in the same range as that of satellite data retrieved using instruments such as the Moderate Resolution Imaging Spectroradiometer (MODIS) and the Multiangle Imaging Spectroradiometer (MISR) (figures not shown). Like aerosols direct radiative forcing, spatial distributions of AOD also exhibited local differences. The simulated AOD was in the range of 0.2-1.5 in both summers. The main difference between the two summers was that a center with higher-magnitude AODs (a maximum of up to 1.5) occurred in 2003 in the Sichuan Basin, but no center occurred in 2006. In North China, AODs with larger values generated over a wider area and their magnitudes were larger by about 0.2-0.4 in 2003 than those in 2006.

Figure 2

Figure 2 Simulated aerosol-induced AOD in JJA for (a) 2003, (b) 2006, and (c) their difference (former minus latter).

Because the emission inventory of aerosols used in the two simulations was the same, the differences in radiative forcing in the EASM-dominated areas in the two summers were related to the monsoon circulations. From the difference of wind velocity at a height of 10 m (Fig. 3), we could realize preliminarily that the transport and diffusion of aerosols by wind played a key role in that difference of radiative forcing. Compared with 2003, the wind velocity from the Sichuan Basin to the middle and lower reaches of the Yellow River were stronger by 0.5-2 m s^-1 in 2006, which were conducive to the downward transport of the aerosols by the southwest wind guided by EASM. Increased transport of aerosols led to a decrease in their concentrations in the source areas and further to the weakening of the aerosol-induced radiative forcing.

Figure 4 shows aerosol-induced changes in surface temperatures, as simulated by the model. In general, the dimming effects of aerosols cooled the surface, but the spatial distributions of the changes in surface temperature were nonuniform. In 2003, surface temperature decreased by 0-1.5° C in most parts of the EASM-dominated areas, and a small increase of 0-0.5° C occurred around the southeast coast areas and the southern part of Shandong Peninsula. The maximum cooling dominated the middle reaches of the Yellow River, just west to the negative center of the radiative forcing in North China. Compared with 2003, the pattern of surface temperature change was complex in 2006. The cooling effects extended from the Sichuan Basin to the middle reaches of the Yellow River, and the maximum magnitude of cooling reached up to -1.5° C. However, there were two areas that warmed by up to 1.5° C: One was located northeast of the Sichuan Basin along the Yangtze River and the other in the lower reaches of the Yellow River. In certain areas, changes in temperature resulted from combined effects of aerosol radiative forcing and other important factors, such as water vapor or cloud change. Aerosol forcing may induce changes in the atmospheric structure, consequently affecting circulation, transport of water vapor, and cloud distribution, which in turn affect incident solar flux and surface temperature. We found that, during strong EASM (2006), the cloud cover (figure not shown) in the above areas decreased due to changes in the circulation, resulting in more solar radiation reaching the surface and an increase in temperature, whereas during weak EASM (2003), changes in the cloud cover were little and hence the temperature was controlled by aerosols’ radiative cooling effect. Therefore, changes in temperature were governed by a complex interplay among radiation, circulation, and cloud.

Figure 3

Figure 3 Difference of wind velocity at a height of 10 m between summers of 2003 and 2006 (former minus latter) (units: m s-1).

Figure 4

Figure 4 Simulated aerosol-induced change of surface temperature in JJA for (a) 2003, (b) 2006, and (c) their difference (former minus latter) (units: ° C).

Compared with the impacts of aerosols on surface temperature, aerosol-induced changes in precipitation were found to be more complicated and uncertain (Fig. 5). In general, spatial distributions of aerosol-induced pre-cipitation change were very different in the two summers.

Figure 5

Figure 5 Simulated aerosol-induced change of precipitation in JJA for (a) 2003, (b) 2006, and (c) their difference (former minus latter) (units: mm d-1).

In the Sichuan Basin, middle reaches of Yangtze River, and lower reaches of Yellow River, the signs of precipitation were nearly opposite in these two periods, leading to a 2-6 mm d^-1 increase in precipitation in 2003 than in 2006 over these regions. On the other hand, precipitation decreased by about 2-4 mm d^-1 in the southeast regions of China in 2003, compared to that in 2006. We also noticed an interesting phenomenon in 2003 that spatial distributions of the surface cooling along the middle and lower reaches of the Yangtze River were nearly consistent with the increase in precipitation. A possible explanation might be that the dimming effect of aerosols caused land surface cooling, which reduced the temperature difference between the land and the ocean, further weakening the EASM and increasing precipitation along the Yangtze River (Zhang et al., 2003).