Climatic–Environmental Effects of Aerosols and Their Sensitivity to Aerosol Mixing States in East Asia in Winter

Abstract

To establish the direct climatic and environmental effect of anthropogenic aerosols in East Asia in winter under external, internal, and partial internal mixing (EM, IM and PIM) states, a well-developed regional climate–chemical model RegCCMS is used by carrying out sensitive numerical simulations. Different aerosol mixing states yield different aerosol optical and radiative properties. The regional averaged EM aerosol single scattering albedo is approximately 1.4 times that of IM. The average aerosol effective radiative forcing in the atmosphere ranges from −0.35 to +1.40 W/m² with increasing internal mixed aerosols. Due to the absorption of black carbon aerosol, lower air temperatures are increased, which likely weakens the EAWM circulations and makes the atmospheric boundary more stable. Consequently, substantial accumulations of aerosols further appear in most regions of China. This type of interaction will be intensified when more aerosols are internally mixed. Overall, the aerosol mixing states may be important for regional air pollution and climate change assessments. The different aerosol mixing states in East Asia in winter will result in a variation from 0.04 to 0.11 K for the averaged lower air temperature anomaly and from approximately 0.45 to 2.98 μg/m³ for the aerosol loading anomaly, respectively, due to the different mixing aerosols.

1. Introduction

Aerosols have important impacts on global and regional climate change. In recent years, the concentration of aerosols has increased rapidly due to economic and social development, and it has become a nonnegligible factor affecting climate change, especially in East Asia. Anthropogenic aerosols in the atmosphere mainly include nitrates, sulfates, black carbon (BC) and organic carbon (OC) [1,2,3]. Sulfates are the main chemical component of anthropogenic aerosols in East Asia [4]. BC is mainly derived from the incomplete combustion of biomass fuels and fossil fuels [5,6,7,8], and is also crucial in the interaction of aerosols and the climate. Aerosols can affect the climate in many ways, such as the direct effect [9,10], indirect effect [11,12,13,14,15,16] and semi-direct effect [17].

East Asia is a major monsoon climate zone, and there are obvious seasonal differences in temperature, precipitation and atmospheric circulation. There are many studies on the interaction between aerosols and the East Asian monsoon (EAM) [18,19,20,21,22,23,24]. Wang et al. (2015) [25] showed that the direct and indirect radiative forcing of all aerosols and the BC direct radiative forcing at the top of the atmosphere (TOA) in summer in East Asia were −0.55 and +0.88 W/m², respectively. They further indicated that the total effects of aerosols would decrease the air temperature gradient between the land and ocean, total precipitation, and wind speeds at 850 hPa in the East Asian region in summer, which was also found in Ding et al. (2013) [26]. Zhuang et al. (2018) [27] indicated that BC could heat the atmosphere, resulting in an increase in lowest air temperature by 0.11–0.12 K in both winter and summer in East Asia, leading to anomalies in cyclones and southerly winds at 850 hPa. Huang et al. (2013) [28] showed that sulfates, BC and OC could intensify the East Asian winter monsoon (EAWM) in the tropical and subtropical regions in East Asia and reduce precipitation in southeastern China in winter using the RegCM4 model. Luo et al. (2020) [29] used the regional sea air coupled model (RIEMS_2.0-POM2K) and showed that the temperature and thermal stability of the atmosphere were increased, and the cloud cover was decreased due to the BC–radiation interaction. Liu et al. (2009) [30] concluded that BC and sulfate in China would weaken the EAM in both winter and summer. They further suggested that the climate effects of aerosols were determined by their optical properties. It is obvious that there is still considerable uncertainty in the impact of aerosols on the monsoon climate.

In addition to the effects of aerosols on regional climate change, aerosols are highly affected by variations in monsoon climates [31,32,33,34,35]. The East Asian monsoon can affect the transport and spatial distribution of aerosols and provide a background field of atmospheric circulation for the generation and development of haze caused by aerosols [36]. Liu et al. (2015) [37] indicated that the strong winter monsoon enhances the southward transport of aerosols in East Asia, and the increase in precipitation increases the moisture removal efficiency and subsequently reduces aerosol concentrations. Yuan et al. (2020) [38] showed that a higher BC concentration in the eastern Tibetan Plateau (TP) resulted from the weakening of westerlies north of the TP, the increase in westerly winds south of the TP, and the eastward movement of the East Asian trough. Wang et al. (2015) [25] showed that the aerosol-induced weakening of the EAM can further increase aerosol loadings. Lou et al. (2019) [39] suggested that BC could stabilize the lower atmosphere and weaken the EAWM, thereby increasing pollution in China.

The above studies showed that aerosols and the EAWM may interact with each other, but the uncertainty is still large. The sixth IPCC report suggested that the total aerosol ERF in 2019, relative to 1750, is −1.1 W/m², with a variation from −1.7 to −0.4 W/m². This uncertainty due to limited observations and conflicting evidence has not been addressed until now. The aerosol mixing state is one of the important reasons for this uncertainty [40,41,42]. Riemer et al. (2019) [42] introduced the concept of the aerosol mixing state to account for the differences in chemical composition across an aerosol and indicated that the aerosol mixing state is of great interest since cloud condensation nuclei (CCN), ice nucleating particles (INPs), and scatterers and absorbers of electromagnetic radiation all depend on properties such as per-particle composition and particle morphology. Jacobson (2001) [17] pointed out that aerosols of different mixing states would cause different IRFs because the aerosol optical properties were determined by the mixing state of aerosol particles [7]. The single scattering albedo (SSA) of the externally mixed aerosols was the largest, while that of the internally mixed aerosols was the smallest. The uncertainty of RF will result in uncertainty in the assessment of climate effects. Chandra et al. (2004) [43] indicated that the mixing states of BC might be the reason why the atmosphere absorbed more solar radiation. Zhuang et al. (2013) [44] showed that regional climate changes were also very sensitive to the mixing state of aerosols. The mixing state of secondary organic aerosol (SOA) determines the importance of SOA in future climate change [45]. Therefore, the aerosol mixing states may also have a significant influence on the interaction between aerosols and the EAWM. East Asia is still an important emission area of air pollutants such as particles and trace gases, and aerosols in this region may always be mixed with each other. To determine these influences, the climate effect of anthropogenic aerosols with different mixing states on the variations in the EAWM and aerosol loadings needs to be investigated.

In this study, the regional climate chemical coupled model RegCCMS is used to study the influence of anthropogenic aerosols on the EAWM and the feedback while considering the aerosol mixing states. Three aerosol mixtures are considered here, including external, internal and partial internal mixing states (EM, IM and PIM). Model descriptions and methods will be introduced in the second part of the article, the third part is the analysis and discussion of the experimental results, and the fourth part is the research conclusions and future research directions.

2. Methods and Materials

2.1. Model Description and Aerosol Optical Depth (AOD) Calculation in RegCCMS

The regional climate–chemical coupled model RegCCMS is an online simulation system that couples the regional climate model RegCM3 [46] and the tropospheric atmospheric chemistry model TACM [47,48]. It can realize the two-way feedback of aerosols and meteorological elements and includes the calculated schemes of the aerosol direct, semidirect, and indirect climate effects. TACM contains many complex atmospheric physical and chemical processes to address aerosols and trace gases. The physical processes contain the transport, dry and wet depositions, and so on. The chemical processes include the gas-phase chemistry model based on CBM4 and the thermodynamic equilibrium model ISORROPIA dealing with volatile inorganic aerosols. More information can be found in Wang et al. [47,48,49]. RegCM3 is a mature regional climate model that can effectively capture the characteristics of East Asian climates. It includes large-scale cloud and precipitation parameterization schemes that consider subgrid-scale cloud changes, new ocean surface flux parameterization schemes, cumulus convection schemes, and so on.

Large uncertainties of aerosol indirect effects still exist. Therefore, this study only considers anthropogenic aerosols’ (sulfate, nitrate, BC and POC) direct climate effects, which are determined by the aerosol-specific extinction coefficient, single scattering albedo and asymmetry factor. Given the type and mass of aerosols, the above aerosol optical properties can be calculated.

2.1.1. Externally Mixing Aerosols

Sulfate, nitrate, hydrophobic BC, hydrophilic BC, hydrophobic primary organic carbon (POC) and hydrophilic POC are separated from each other when aerosols are externally mixed, and the relevant optical depth is the sum of the AOD of each aerosol. The AOD of carbonaceous aerosols can be expressed as [50]:

$${\tau }_i\left(\lambda \right)=M_i{\delta }_{\lambda i}{\left(1-RH\right)}^{-K_i}$$

(1)

where i = hydrophilic BC, POC and hydrophobic BC, POC. For hydrophobic carbonaceous aerosols, K = 0, and for hydrophilic BC and POC, K = 0.25 and 0.2, respectively. $M_i$ is the mass of aerosols, $\lambda$ is the wavelength, and RH is the relative humidity. ${\delta }_{\lambda i}$ is the wavelength-dependent specific or mass extinction coefficient (m²·g⁻¹) of aerosol i at wavelength $\lambda$.

The AOD of sulfate and nitrate aerosols was as follows [51,52]:

$${\tau }_i\left(\lambda \right)=M_i{\delta }_{\lambda i,0}{e}^{({\delta }_{\lambda i,1}+\frac{{\delta }_{\lambda i,2}}{RH+{\delta }_{\lambda i,3}}+\frac{{\delta }_{\lambda i,4}}{RH+{\delta }_{\lambda i,5}})}$$

(2)

where betas within the e index are fitting coefficients of the specific extinctions.

2.1.2. Internally Mixing Aerosols

Hydrophilic BC is assumed to be the core, and other hydrophilic aerosols (sulfate, nitrate and hydrophilic POC) are assumed to be shells to wrap BC in the internal mixture. The optical depth of the internally mixing aerosols is the sum of the AOD of internally mixing aerosols and the AOD of hydrophobic carbonaceous aerosols. The AOD of internally mixed aerosols was calculated as follows [40,53]:

$$\tau {\left(\lambda \right)}_{int}={\delta }_{\lambda }\left(fvo{l}_{sulfate+nitrate},fvo{l}_{HL BC},fvo{l}_{HL OC},fvo{l}_{dust}\right){{\displaystyle \sum }}_i{M}_i$$

(3)

where ${\delta }_{\lambda }$ is the specific extinction coefficient (m²/g) of the internally mixed particles at wavelength $\lambda$. $fvol$ is a volume function of each aerosol in a mixed particle. More information can be seen in Zhuang et al. (2013) [44].

2.1.3. Partially Internally Mixing Aerosols

When aerosols are partially internally mixed, some of the hydrophilic aerosols are internally mixed, and the rest are externally mixed. Only when the BC particles are completely converted into hydrophilic aerosols can they act as cloud condensation nuclei [54]. The conversion time from fresh BC to aged BC is approximately 1.6 days [55]. Therefore, the content of hydrophilic carbonaceous aerosols in the atmosphere may change with time. However, due to a lack of observations, we assume that 32.2%, 32.2%, 35.5% and 48.5% of sulfate, nitrate, BC and OC are internally mixed according to Kim et al. (2008) [56] when aerosols are partially internally mixed.

2.2. Simulation Schemes

The simulated domain covered most regions of Asian countries, centering around 34.5°N, 106.8°E, with a horizontal resolution of 50 × 50 km and with 18 layers in the vertical direction from the surface to 50 hPa. The study site was mainly focused on East Asia within the region 100°−140°E, 15°−55°N, which can be seen in Figure 1. RegCCMS was run for four months from November to February in the following year during the period from 1986 to 2006 due to the calculation limitation. The first integrated month (November) of each year was a spin-up time. The weekly mean sea surface temperature (SST), which was obtained from NOAA Optimum Interpolation SST-V2 [57], was used here and fixed when estimating aerosol climate effects of different mixtures. NCEP reanalysis data (NNRP2) were implemented for initial and boundary conditions to drive the model. The emission inventory of air pollution was derived from the Intercontinental Chemical Transport Experiment-Phase B [6]. The inter-annual variations of the aerosol emissions were not taken into account in this study. To validate the ability of RegCCMS to simulate the air pollutants and meteorological fields in East Asia, satellite data, ground based observational data, and reanalysis data were employed here. Satellite data were used to verify aerosol optical properties, mainly by comparing seasonal average values. Ground-based observations were used to verify the average aerosol concentration in winter. Reanalysis data were devoted to validating the simulated dynamic and thermal fields, including multi-year seasonally averaged temperatures and winds. Details can be found in the next section.

To study the influence of mixing states on the interactions between aerosols and the EAWM, four simulation experiments were carried out in this study, as shown in

Table 1. Numerical simulation experiments performed in this research.

Numerical Simulation Experiments Done in This Research
Experiment	Experiment content
CLR	Does not consider the effects of aerosols.
EM	Aerosols are independent of each other and mixed externally.
IM	Hydrophilic BC is assumed to be the core, and other hydrophilic aerosols are wrapped around the BC as a shell.
PIM	Partial hydrophilic aerosols (32.2%, 32.2%, 35.5% and 48.5% of sulfate, nitrate, black carbon and organic carbon) are internally mixed, and other aerosols are externally mixed.

: a control experiment, CLR, without considering the influence of aerosols—that is, the influence of aerosol concentration was not considered when calculating radiation; and sensitivity experiments EM, IM, and PIM, considering the direct climate effect of external mixing, internal mixing and partially internal mixing aerosols, respectively. Through the differences between the results of the three sensitivity experiments and the control experiment, the influence of aerosols in different mixing states on the East Asian winter monsoon was obtained. Only the direct climate effects of anthropogenic aerosols (sulfate, nitrate, BC and POC) were considered in this study. This study also performed Student’s t-tests to verify the statistical significance of the differences between sensitive experiments and control experiments.

3. Results

3.1. Model Validation

Previous studies [25,44,48,58,59] have proved that the RegCCMS can effectively reproduce the main characteristics of meteorological fields and pollutants in East Asia, so we simply verify the credibility of the model here.

3.1.1. Meteorological Element Field

The study compared the meteorological element fields at different altitudes from the simulation and NCEP reanalyzed data to verify the accuracy of the RegCCMS in simulating climate effects. As shown in Figure 1, the model simulation results are basically consistent with the reanalysis data at any altitude. The RegCCMS model can effectively reproduce the main characteristics of meteorological fields in winter in East Asia.

3.1.2. Aerosol Concentration and AOD

Figure 2 shows the spatial distribution of the seasonal mean surface concentration of all anthropogenic aerosols in winter. The concentrations are higher in the southwestern, central and eastern regions of China (100°−120°E, 20°−40°N), which have high air pollutant emissions [60]. The aerosol concentration reaches a maximum near the Sichuan Basin (100°−110°E, 25–34°N) because the meteorological diffusion conditions in the basin areas are also poor. Han et al. (2011) [61] also drew the same conclusion that the maximum mass burden appeared over the Sichuan Basin in January due to the weaker convection strength. The regional average surface concentrations of nitrate, sulfate, BC, and POC aerosols in East Asia in winter (

Table 2. Regionally averaged surface concentration (μg/m³) for 100°–130°E, 20°–50°N and column burden (mg/m²) of different types of anthropogenic aerosols in East Asia in winter.

Average Aerosol Concentration in East Asia in Winter
Aerosol Type	Surface (μg/m3)	Column (mg/m2)
Total	18.23	39.32
Nitrate	4.93	11.34
Sulfate	8.14	19.90
Black Carbon	2.23	3.46
Primary Organic Carbon	2.93	4.63

) are 4.93, 8.14, 2.23, and 2.93 μg/m³, respectively, and the column burdens are 11.34, 19.90, 3.46, and 4.63 mg/m², respectively. Sulfate aerosols have the highest concentration, with a maximum column burden of 120 mg/m², accounting for approximately half of the total anthropogenic aerosols. BC has the lowest concentration, accounting for only approximately one-tenth of the total anthropogenic aerosols. In order to verify the ability of RegCCMS to simulate aerosols, the simulated aerosol loadings will be compared with observations. The seasonal mean surface concentrations of total simulated aerosols over 14 stations (

Table 3. Information of 14 observation sites of aerosol loadings.

Site	Latitude (°N)	Longtitude (°E)
Chengdu	30.65	104.04
Dalian	38.9	121.63
Gucheng	39.13	115.8
Gaolanshan	36.0	105.85
Jinsha	29.63	114.2
LinAn	30.3	119.73
Longfengshan	44.73	127.6
Lhasa	29.67	91.13
Nanning	22.82	108.35
Panyu	23.0	113.35
Zhengzhou	34.78	113.68
Dunhuang	40.15	94.68
Taiyangshan	29.17	111.71
XiAn	34.4	108.8

) in China in winter of 2005 and 2006 are selected and compared with ground-based observations, because country-wide observations of aerosol surface concentrations were carried out during these periods [60,62]. Comparisons between simulations and observations are also presented in Figure 2, where NSP represents the sum of nitrate, sulfate and primary organic carbon aerosols. RegCCMS basically described the distributions and levels of anthropogenic aerosols over China, but the simulated values were slightly smaller than observations, possibly due to biomass burning and other sources of aerosols being excluded.

The distribution of aerosol optical depth as shown in Figure 3 is consistent with the aerosol loadings. The AOD is larger in the southwestern, central and eastern regions of China (100°–120°E, 20°–40°N) in winter and is the largest near the Sichuan Basin. Satellite data are additionally used to further verify the simulation capability of the model. The 550 nm AOD data from satellite products (OMI) in East Asia during the winter of 2004–2006 are used. The seasonal mean AOD from OMI is also present in Figure 3. The simulated AOD in East Asia is similar to that of the OMI satellite product. The AOD in northwest regions was very small due to excluding the natural aerosols. Xin et al. (2007) [63] also obtained similar conclusions based on the observation results of CSHNET from 2004 to 2005, further validating the simulated abilities of RegCCMS to anthropogenic aerosols over East Asia.

3.1.3. Single Scattering Albedo (SSA)

Figure 4 shows the seasonal mean aerosol SSA, which has obvious differences in East Asia in winter under different mixing states. When aerosols are mixed internally, the SSA is much smaller. The maximum value is approximately 0.76, indicating that internal mixing aerosols have a strong absorption of solar radiation, which may further affect the formation and lifetime of clouds [64]. When the aerosols are mixed externally, the SSA is the largest, and the maximum value reaches approximately 0.99, which is 1.3 times the internally mixed SSA. This means that externally mixed aerosols have a relatively weaker (stronger) ability to absorb (reflect) solar radiation. The seasonal average SSAs in winter are 0.97, 0.69, and 0.96 for EM, IM and PIM aerosols, respectively, over the region (100°–130°E, 20–50°E). The regional mean SSA in PIM is the closest to the satellite products (OMI, 0.94) in East Asia in winter. SSA is relatively small in land areas, especially in the North China Plain, because BC in these areas accounts for a higher proportion of total anthropogenic aerosols.

3.2. Aerosol Radiative Forcing (RF)

The results discussed below are the difference between the sensitivity experiments (EM, IM, PIM, 1986–2006 averaged) affected by aerosols in different mixing states and the control experiment (CLR) that does not consider the aerosol effects.

3.2.1. Instantaneous Direct Radiative Forcing (IRF) and Effective Radiative Forcing (ERF) of Aerosol

In this subsection, IRF refers to the instantaneous direct radiative forcing—that is, the difference in the shortwave radiation flux in the atmosphere with or without aerosol influence. The calculation formula [44] can be expressed as

$$\mathrm{RF}=(R{F}_{withaerosols}^{\downarrow }-R{F}_{withaerosols}^{\uparrow })-(R{F}_{noaerosols}^{\downarrow }-R{F}_{noaerosols}^{\uparrow })$$

where $R{F}^{\uparrow }$ and $R{F}^{\downarrow }$ represent the upward and downward shortwave radiation, respectively.

Figure 5 (upper) shows the IRF in the atmosphere under different aerosol mixing states in all sky conditions in winter, and all regions are statistically significant (not shown in the figure). The IRF is positive because BC absorbs the solar radiation reaching the surface and heats the atmosphere. IRF is strong in southwestern, central and eastern China, where the aerosol concentration is high (100°–120°E, 20°–40°N). When aerosols are mixed internally, IRF is the strongest, and the maximum value appears in the Sichuan Basin, at nearly 25 W/m². In the EM, IM, and PIM experiments, the average IRFs in the atmosphere in East Asia in winter are 3.70, 5.07, and 4.14 W/m², respectively.

The IRFs are all negative at surface (not shown) for EM, IM and PIM aerosols, which is strongest near the Sichuan Basin in EM experiment, indicating that the shortwave radiation reaching the surface is reduced under the three aerosol mixing states. Thus, aerosols cool the surface by absorbing and scattering solar shortwave radiation.

IPCC AR5 proposes the concept of effective radiative forcing (ERF), which refers to the change in net radiation flux after keeping the globally averaged lower air temperature or parts of the surface conditions constant, and sea temperature and sea ice density are fixed, allowing adjustments to atmospheric stratosphere temperature, water vapor, and clouds [65]. The calculation of ERF is much more complicated than that of IRF, but it can better represent the response of climate elements to radiative forcing. Figure 5 (lower) shows the ERF in the atmosphere in winter under the three aerosol mixing states. The ERF is relatively strong in the high aerosol concentration areas (100°–120°E, 20°–40°N). In the atmosphere, the area around the Sichuan Basin (105°–110°E, 25°–35°N) shows an obviously positive ERF, and reaches the strongest value under the internal mixing state, with the maximum exceeding 13 W/m². The absorption of solar radiation by BC leads to an increase in ERF in the atmosphere. In the EM, IM and PIM experiments, the average ERFs in East Asia are −0.35, 1.4 and 0.2 W/m², respectively. Aerosols have cooling effects at the surface and warming effects in the atmosphere, which is consistent with previous research results [66,67,68,69]. Bi et al. (2014) [69] showed that the RF of aerosols had a significant impact on the thermal structure of the atmosphere, and nearly half of the solar radiation was absorbed by aerosols, resulting in abnormal atmospheric heating and surface cooling effects.

The radiative forcing of aerosols in China has a great influence on atmospheric circulation, surface pressure, temperature, humidity and fog [70]. The RF is very sensitive to aerosol mixing states, so the climate effects of aerosols with different mixing states will also have obvious differences.

3.2.2. Shortwave Heating Rate

Aerosols change the net radiation flux of the Earth system by affecting the shortwave heating rate (QRS) in the atmosphere [71]. Figure 6 shows the QRS in winter under different aerosol mixing states. The upper row presents the average QRS below 850 hPa and the lower row presents the average QRS in the region of 25°–35°N in the vertical direction. The QRS increases significantly due to the absorption of solar radiation by BC, and the largest change occurs near 105°E, 27°N, coinciding with the high value area of aerosol concentration. Positive shortwave heating rates lead to positive radiative forcing in the atmosphere. The QRS changes are mainly distributed in the lower atmosphere below 800 hPa and gradually decrease with increasing altitude because aerosols mainly accumulate in the lower troposphere. When the aerosols are mixed internally, the QRS increases the most, and the maximum change reaches 1.71 K/day, followed by those of PIM and EM aerosols. Under the three mixing states (EM, IM, PIM), the average QRS changes are 0.18, 0.24, and 0.20 K/day in the region of 100°–130°E and 20°–50°N in winter, respectively (

Table 4. Regional (100°–130°E, 20°–50°N) averaged changes in near-surface solar heating rate (K/day), sensible heat flux (W/m²), 2 m air temperature (K), 850 hPa wind velocity (m/s), precipitation (mm/day), and 850 hPa humidity (×10⁻³ g/kg) under different aerosol mixing states in winter.

Regional Averaged Weather Field Changes in Winter
Mixing State	Near-Surface Solar Heating Rate (K/day)	Sensible Heat Flux (W/m2)	2 m Air Temperature (K)	850 hPa Wind Velocity (m/s)	Precipitation (mm/day)	850 hPa Humidity (×10−3 g/Kg)
EM	0.18	−1.87	−0.11	0.06	−0.02	−9.47
IM	0.24	−2.29	−0.06	0.08	−0.01	4.62
PIM	0.20	−1.95	−0.10	0.06	−0.02	−8.37

).

3.3. Climate Effects of Aerosols

3.3.1. Cloud Cover

Figure 7 shows the change in average cloud cover in the near-surface layer (below 850 hPa) under three aerosol mixing states. The cloud cover in southwestern China, central China, and the North China Plain, where the aerosol burden is high, reduced to a certain degree. Many other studies also show that cloud cover has decreased in most areas of China [72,73,74]. This is mostly caused by the semidirect effects of aerosols on clouds [75,76,77]. Additionally, absorptive aerosols absorb solar radiation and heat the atmosphere; thus, the atmosphere is more stable, and clouds struggle to form [78]. When aerosols are mixed internally, cloud cover changes the most, and the maximum change is close to −3%. When aerosols are mixed externally, cloud cover changes the least because of the weaker absorbability of aerosols in the EM. In the EM, IM, and PIM simulation experiments, the average cloud cover changes in the regions (100°–130°E, 20°–50°N) are −0.37%, −0.42%, and −0.40%, respectively. The decrease in cloud cover increases the solar radiation on the surface and in the atmosphere.

3.3.2. Sensible Heat Flux and Near-Surface Air Temperature

Figure 8 shows the change in surface sensible heat flux under the three aerosol mixing states. The sensible heat flux is negative over most of East Asia and stronger in areas with high aerosol concentrations, indicating that the energy in the atmosphere is transported downward to heat the surface, consistent with the results of shortwave heating rates and radiative forcing. In the three simulation experiments, EM, IM, and PIM, the average sensible heat flux in East Asia in winter is −1.87, −2.29 and −1.95 W/m², respectively (Table 4), which is the strongest in the IM experiment, indicating that the heating effect of BC is the most obvious when aerosols are internally mixed.

The changes in the above factors such as the radiation flux, cloud cover, and shortwave heating rate of the Earth–atmospheric system represent all of the ways that aerosols affect the air temperature in different mixing states. There are many studies about the influence of aerosols on temperature [79,80,81,82].

Figure 9 shows the air temperature changes at the height of 2 m and in the vertical direction under different aerosol mixing states. The 2 m air temperature decreases in the East Asian continent, where aerosol distribution and surface negative RF are more obvious, especially in southwestern and central China. The IRF and ERF on the surface are both negative, implying that the solar radiation reaching the ground is reduced, resulting in a potential decrease in 2 m temperature. Thus, the direct effect of anthropogenic aerosols has a cooling effect on the land surface of East Asia. Li et al. (2016) [83] had similar findings. The temperature changes are significantly different among the three aerosol mixing states. When aerosols are mixed externally, the temperature decreases the most, and the maximum change exceeds −0.7 K, while in the IM experiment, the maximum temperature change is much smaller. Internally mixed BC absorbs more solar radiation, which neutralizes the climate cooling effect of scattering aerosols. The 2 m air temperature in the East China Sea rises, possibly because of the positive QRS, cloud cover change and weak negative RF. The temperature increases the most in the IM case and the least in the EM case. The change in surface air temperature here is more consistent with the conclusion of Lou et al. (2019) [39]. The average 2 m air temperature changes in the area of 100°–130°E, 20°–45°N are −0.11 K, −0.06 K, and −0.10 K when the aerosols are externally, internally and partially mixed, respectively.

The temperature change in the vertical direction is shown in Figure 9 (lower). In the middle and lower latitudes of the East Asian continent, air temperature near the surface decreases as described before, and increases around 950 hPa to 700 hPa, especially when aerosols are mixed internally. The maximum temperature change exceeds 0.45 K, around the height of 880 hPa, which is mainly due to the warming effect of BC. BC absorbs solar radiation, causing positive QRS and RF change and negative cloud cover change in the atmosphere, which in turn heats the atmosphere. The maximum value of atmospheric warming caused by internally mixing aerosols is 1.77 times that of externally mixing.

Based on the above analysis, the air temperature decreases more significantly at the surface than at the lower atmosphere. At the same time, the near-surface layer heating is more pronounced than the surface heating, consistent with the energy transfer represented by the sensible heat flux in Figure 8, which may improve the stability of the low-level atmosphere and reduce the height of the boundary layer, as suggested in [84,85,86]. When aerosols are mixed internally, the temperature rises the most, and when aerosols are mixed externally, the temperature decreases the most. Thus, the air temperature is also very sensitive to the mixing state of aerosols.

3.3.3. Atmospheric Circulation

Thermal disturbances caused by aerosols can lead to anomalies of the wind field and circulation [87]. In winter, northerly winds affect most areas of East Asia [77], and the climatological wind at 850 hPa in small parts of southwestern China is southerly [88]. Figure 10 shows the changes in wind fields at 850 hPa under different aerosol mixing states in winter. The greatest wind field change appears in southern and southwestern China (105°–115°E, 20°–35°N). Abnormal southerly winds appear in the 105°–120°E area and abnormal northerly winds appear in the 100°–105°E area, opposite of the winter monsoon indicating that aerosols have weakened the climatological winds in the area in winter. The results are consistent with those of Niu et al. (2010) [70]. There is an abnormal southeasterly wind in the eastern coastal area of China, which weakens the westerly wind of the EAWM. The wind field changes the most when the aerosols are mixed internally, with an average wind speed change of 0.08 m s⁻¹ in East Asia. When the aerosols are externally mixed and partially internally mixed, the wind field changes are much weaker. This study shows similar results to Zhang et al. (2012), Wang et al. (2014), Ding et al. (2020), Xu et al. (2006) and Jacobson and Kaufman (2006) [89,90,91,92,93], and further emphasizes the importance of aerosol mixing states.

The radiative forcing of aerosols changes the thermal-dynamic properties of the atmosphere, which further affects the changes in clouds, precipitation and humidity in East Asia (not shown) [77]. Investigations show that the direct influence of anthropogenic aerosols on humidity and precipitation is not obvious in winter (Table 4) because the air is dry in wintertime. Overall, the direct effects of aerosols decrease the total precipitation in East Asia in winter.

3.4. Environmental Effects of Aerosols

Aerosols have significant impacts on the East Asian monsoon. Monsoon responses will also change the concentration and spatial distribution of aerosols [94,95,96,97,98,99]. Figure 11 (upper) shows the influence of winter monsoon anomalies on the total anthropogenic aerosol surface concentration under different mixing states. Aerosols are likely to accumulate in the low layer atmosphere because the transportation and scavenging efficiencies of air pollutants are decreased. As shown in Figure 10 and Table 4, the EAWM is weakened and the precipitation in most parts of East Asia is reduced. Additionally, the more stable atmospheric boundary layer induced by aerosols also favors aerosol accumulation. Figure 11 (lower) shows the change in the boundary layer height under the three aerosol mixing states. Anthropogenic aerosols lower the height of the boundary layer in southwestern, central and eastern China, making the atmosphere more stable and inhibiting the diffusion of aerosols. Therefore, the aerosol concentration increased significantly in southwestern China. Similar to the regional climate changes, the aerosol concentration increases the most in IM, with the maximum change exceeding 30 mg/m².

Li et al. (2016) [100] found that the interannual variability in winter haze days in central and eastern China was closely related to the EAWM. Winter haze days increased (less) when the EAWM was weak (strong). The observation results showed that a large volume of aerosols could reduce incident solar radiation and enhance the stability of the regional atmosphere, thus facilitating the accumulation, retention and growth of aerosols [68,101]. The results here are consistent with some of these results and further reflect the importance of aerosol mixtures in regional air pollution issues.

4. Brief Discussion

Overall, the results here suggest that the direct effect of anthropogenic aerosols has a cooling effect on the land surface while having a warming effect in low layers of East Asia, which is likely weakens the East Asian winter monsoon and decreases the total precipitation. These findings are consistent with Bi et al. (2014) [69], Li et al. (2016) [83] and so on. Li et al. (2016) [83] believed that the cooling trend in central and eastern China was related to the long-term effects of the rapid growth of pollutants on the surface solar radiation. Zhang et al. (2012a) [89] and Wang et al. (2014) [90] showed that the increase in aerosols weakened EAM and suppressed convection near 30°N in China, thereby enhancing atmospheric stability. Ding et al. (2020), Xu et al. (2006) and Jacobson and Kaufman (2006) [91,92,93] also showed that aerosols reduced wind speed. Previous studies have also shown that aerosols reduced light rain and increased heavy rain [102,103,104,105]. These changes are caused by many factors, including complex aerosol–cloud interactions [106,107]. Ding et al. (2015) [108] analyzed long-term meteorological records and showed that the EAM has been weakened in precipitation and circulation since the late 1970s, which is closely related to environmental changes, especially aerosols. This study here shows similar results and further emphasizes the importance of aerosol mixing states.

The above results and discussions all indicate that aerosols have important impacts on the climate and environment on East Asia in winter, and the effects are relatively complex. In addition, the monsoon interannual variations, which exclude the influence of the aerosol effect, also have visible impacts on aerosol loadings and regional climates. Therefore, a brief discussion on the aerosol–East Asian winter monsoon interactions in strong and weak monsoon years is further carried out here.

In this study, we select 1986, 1988, 1996, 2000, 2001, and 2006 as strong monsoon years and 1989, 1990, 1993, 1997, 1998, and 2002 as weak monsoon years according to the monsoon index in Mao et al. (2017) [109]. In Mao et al. (2017), the monsoon index was calculated through adding the zonal sea level pressure differences between 110°E and 160°E over 20°N and 70°N using the reanalyzed NCEP/NCAR datasets. Previous studies [110] suggested that East Asian monsoon variations would affect aerosol loadings and distributions. Here, anthropogenic aerosols are divided into two categories to analyze the effects of EAM variations on aerosols according to their physical and chemical properties. Category 1 is for primary aerosols such as BC and POC aerosols. Category 2 is for secondary inorganic aerosols such as sulfate and nitrate aerosols. Figure 12 shows the percentage of aerosol concentration difference between strong and weak monsoon years in the CLR experiments. In strong monsoon years, the concentration of secondary inorganic aerosols is lower than that in weak monsoon years in most regions of East Asia, especially in areas with high temperatures, potentially because the higher temperature is not conducive to the transformation of aerosols, and accompanied by an abnormal westerly wind that affects the downstream areas, resulting in lower concentrations of secondary inorganic aerosols in most areas. The concentration of primary carbonaceous aerosol is higher in central, eastern and coastal areas of China. The annual precipitation of strong monsoons is less than that of weak monsoons, and the regions with lower boundary layer heights are more consistent with regions with higher aerosol concentrations, which are all reasons for the higher concentrations of BC and POC in strong monsoons. Different aerosol concentrations and optical properties between strong and weak monsoon years result in different responses of the EAWM.

5. Conclusions

In this study, the well-developed regional climate–chemical coupling model RegCCMS is used to investigate the interactions between aerosols and the East Asian winter monsoon under different aerosol mixing states.

The aerosol concentration is relatively high in southwestern, central and eastern China (100°–120°E, 20°–40°N) in winter, especially around the Sichuan Basin. BC accounts for approximately 10% of the total anthropogenic aerosols, showing that the aerosol mixing states are very important. The average SSA of externally mixing aerosols in East Asia is approximately 1.4 times that of internally mixing aerosols.

The surface IRFs are all negative under the three aerosol mixing states, and IRF is the strongest in the EM experiment. In the atmosphere, there are obviously positive ERFs, especially near the Sichuan Basin, due to the absorption of BC. The ERF is the strongest when aerosols internally mix. The average ERFs in the East Asian atmosphere are −0.35, 1.40, and 0.20 W/m² for EM, IM, PIM aerosols, respectively. The absorption of aerosols results in substantial increases in the QRS of the lower atmosphere below 800 hPa, especially for IM aerosols. The average QRS changes in the three mixing states (EM, IM, PIM) are 0.18, 0.24 and 0.20 K/day, respectively. Therefore, cloud cover in the lower atmosphere decreased in the regions from southwestern to central-eastern China.

Anthropogenic aerosols have a cooling effect in continental Asian regions in winter. The cooling effect is the most obvious when aerosols are mixed externally. The vertical air temperature shows an obvious increase from 950 hPa to 700 hPa in most parts of East Asia. This warming effect is the most (least) obvious due to IM (EM) aerosols. The average near-surface air temperature (below 850 hPa) varies by 0.08 K in East Asia owing to the aerosol mixing states. Consequently, the atmospheric boundary will become more stable. Thermal disturbances caused by aerosols further result in changes in wind fields and circulations and subsequently in hydrologic cycles. Aerosols weaken the EAWM, and the weakness is the greatest when aerosols are mixed internally. The regional changes in the lower atmospheric temperature, wind field and precipitation in the PIM are 0.06 K, 0.06 m/s and −0.02 mm/day, respectively.

The concentration of aerosols in East Asia has increased significantly due to the EAWM responses. The weakening of the EAWM circulation, the decrease in precipitation, and the increase in atmospheric stability all favor the accumulation of aerosols. The loading of aerosols increases the most when they are internally mixed, with an increase in the aerosol concentration of approximately 2.98 μg/m³ in southwestern China in winter.

In addition to EAWM responses, the interannual variations in the EAWM also have a considerable influence on aerosol loadings. Overall, the secondary inorganic aerosols are relatively lower in concentration in most parts of East Asia in the strongest monsoon years than in the weakest monsoon years, which is the opposite to the primary aerosols. Therefore, aerosols are more absorptive in the strongest monsoon years, which leads to the effect of EM, IM and PIM aerosols on the EAWM being all more significant in these years compared with that in the weakest monsoon years, especially for IM aerosols.

This study further indicates that the aerosol mixing states have a significant influence on the aerosols’ optical and radiative properties, as well as interactions between the aerosols and EAWM. The results here may help us better understand regional air pollution and climate change issues and better assess aerosols’ climate and environmental effects in the future. This study mainly considers the direct climate and environmental effects of anthropogenic aerosols in East Asia in winter. The characteristics in other seasons and the indirect effects of aerosols will continue to be considered in subsequent studies. Additionally, due to the lack of systemic observational studies, the aerosol internal mixing ratio in East Asia is still unclear in current knowledge, which may also have a significant influence on assessments of aerosols’ direct and indirect climate effects. This is also a major limitation for this study. In the future, the aerosol internal mixing ratios should be quantified based on multi-platform observations, including using ground-based and remote sensing data, to further improve the aerosol chemical and optical processes of the model. Subsequently, the aerosol radiative forcing and regional climate effects would be assessed more accurately.