Studies on the Climate Effects of Black Carbon Aerosols in China and Their Sensitivity to Particle Size and Optical Parameters

In this paper, based on the principle of Mie scattering, we calculated the optical parameters of BC aerosols at different scales and then applied the new optical parameters to simulate the BC aerosols concentration distribution, radiative forcing, and their climate effects.We also compared the results of optical parameters of BC aerosols with homogeneous scales and analyzed the effect on climate. Compared with the conventional uniform-scheme optical parameterization, the concentrations of the first mode of BC aerosols simulated with the optical parameters that were recalculated based on the particle size are significantly higher, while the concentrations of the other modes and the total of BC aerosols are lower. In the respective of statistics, the changes of column burdens of BC in four modes are 0.085, −0.095, −0.089, −0.054mg/m. ,e clear-sky TRF of BC are weakened in the value of 0.03W/m averaged over the domain, while the all-sky TRF of BC are enhanced of 0.06W/m in general.,e warming effect of BC becomes weaker when using the new scheme by −0.04K to −0.24K. When using the new optical parameters scheme, the regional average surface concentrations of BC in four modes are 0.372, 0.264, 0.055 and 0.004 μg/m, respectively. Especially, the first and the second mode account for as large as 53% and 38%. ,e surface concentration and column burden of total BC are 0.69 μg/m and 0.28mg/m can be dropped. ,e regional average direct RFs of BC at the top of the atmosphere are 0.49W/m under clear-sky and 0.36W/m under all-sky averaged over the domain. Over most areas of central China, North China, and East China, BC may increase the temperature in a range of 0.05∼0.15K, while over South China, BC shows cooling effect. In average, the precipitation variations caused by BC over East China, North China, South China, and Northeast China are −0.83, −0.05, −0.11, and −0.13mm/d, respectively. As a whole, the variations of circulation, pressure, and temperature show a good correspondence.


Introduction
Atmospheric aerosols can affect the earth's climate through both direct and indirect effects and play important roles in atmospheric radiation and climate change. Among all of the anthropogenic aerosols, black carbon (BC) aerosols can effectively absorb solar radiation in the visible and infrared bands to heat the atmosphere and affect the climate and air quality [1,2] and therefore play a unique and important role in the climate systems.
e Intergovernmental Panel on Climate Change (IPCC) report [3] indicates that the average direct radiative forcing of global BC aerosols is 0.4 W·m −2 . BC aerosols make an important contribution to global warming [4] and have become the second major warming factor after CO 2 [2].
East Asia is an important source region of global aerosol emissions [5]. e anthropogenic aerosols in East Asia have increased, which has had a nonnegligible influence on the regional climate. e global atmosphere model CAM3 was used to study the influence of sulfate aerosols and BC aerosols on the East Asian summer monsoon, and the results indicate that sulfate aerosols reduce the temperature in most areas of China, decreasing precipitation in some regions and weakening the East Asian summer monsoon [6]. BC aerosols also weaken the East Asian summer monsoon, but their influence is more complicated. e relationship between BC aerosols and regional precipitation has been studied, and researchers suggested that the BC aerosol emissions in China are related to the droughts in southern China and the floods in northern China over the past several decades [7]. BC aerosols have a significant influence on the weakening of the wind velocity [8]. BC aerosols have a relatively heterogeneous spatiotemporal distribution, which makes it difficult to evaluate their influence on the regional climate. Coal combustion is the main anthropogenic source of BC aerosols [9]. Coal is the main component of the energy structure in China; therefore, several researchers have concluded that one quarter of the BC aerosols generated by global anthropogenic emissions comes from China [10,11]. With the accelerated development in China in recent years, the emissions of BC aerosols have increased significantly, and BC aerosols in China have drawn broad international attention. China is located in the Asian monsoon area. e response of the climate in this region to the atmospheric compositions could be different from those in other regions, so it is necessary to use a regional model to study the radiative forcing and climate effect of BC aerosols in this region.
Aerosols affect atmospheric radiation through scattering and absorption, and their optical parameters are important for the calculation of radiation effects. In numerical simulations, the optical parameters of BC aerosols directly affect the calculated radiation effects. e radiative forcing effect of BC aerosols has received wide attention, and the analysis of their optical properties is especially important [12]. Many analyses of aerosol optical parameters have focused on dust and sulfates, and calculations of the optical properties BC aerosols are relatively limited. No analyses of the light extinction coefficient, absorption coefficient, and scattering phase function of BC aerosols have been conducted.
e climate forcing of carbon aerosols varies significantly with their size distribution [13,14]. e average indirect radiative forcing effect of carbon aerosols at the top of the atmosphere ranges from −0.34 to 1.08 W/m 2 , and its variation strongly depends on the assumed particle size distribution of the aerosols in the model [14]. e sensitive of BC aerosols climate effects on the particle size distribution is due to the two reasons: different assumptions of the particle size distribution will cause differences in the calculation of the BC aerosols' concentration distribution and then result in changes in the optical parameters of the BC aerosols. In many previous studies, the particle size distribution of BC aerosols adopted the homogeneous scale (median radius) method, and the optical parameters only varied with the wavelength. is method can cause relatively large errors and increase the uncertainty of the climate effects of BC aerosols.
Because the concentration of BC aerosols and the climate effect are relatively sensitive to the particle size, in previous studies, we adopted a size-resolved model of BC aerosols and studied the climate effect of BC aerosols based on the measured particle size distribution [15]. However, the optical parameters did not vary with the particle size, which may cause several errors. In this paper, we calculate the optical parameters of BC aerosols with different scales mainly through the principle of Mie scattering and further use the calculated optical parameterization scheme to simulate the size-resolved climate effects of BC aerosols. e conventional scale division scheme for the BC aerosols was used [15]. We compare the differences in the simulation results caused by the differences in the optical parameterizations of BC aerosols.

Calculation of the Optical Coefficient of Black Carbon Aerosols
In numerical models, the optical coefficient of aerosols directly affects the calculated radiation effect, and the radiation transfer scheme usually requires three input parameters: single scattering albedo (SSA), dissymmetry factor (DF), and mass absorption coefficient (MAC). e acquisition of these three optical parameters must be based on Mie scattering theory. e SSA (ω 0 ) is the ratio between the aerosol scattering coefficient and the light extinction coefficient, and it measures the absorption strength of aerosols. It is an important optical property of aerosols. e MAC (K) measures the absorption properties of a unit aerosol mass, and the DF (g) describes the asymmetry of the forward and backward scatterings of aerosols. In numerical models, these parameters play important roles in calculating the radiative forcing of aerosols. e three conditions for the numerical calculation are (1) the complex refractive index of BC aerosols in different wave bands, (2) the spectral distribution of the BC aerosols, and (3) the program of Mie scattering theory.

Complex Refractive Index.
e real and imaginary parts of the complex refractive index of aerosol particles represent the characteristics of scattering and absorption, respectively, and the magnitudes of the absolute values determine the strengths of the scattering and absorption properties of aerosols, which have a direct influence on the radiation of aerosols. For a given scale spectrum of particles, the variation in the radiation characteristics of aerosols with the wavelength is determined by its complex refractive index. Figure 1 shows the values of complex refractive index of BC aerosols. ese values of the complex refractive index used by Shettie are selected in this paper [16]. is data source has also been used to study the optical properties of BC aerosols by others [17]. e variation of the imaginary part of the complex refractive index of BC aerosols with the wavelength is relatively small, but its absolute value is usually greater than 0.4; therefore, BC aerosols mainly absorb the long radiation.

Particle Size Distribution.
e particle size of the aerosols also has an important effect on the final radiation. In previous studies, the RIEMS2.0 model divided BC aerosols into four bins, which somewhat improves the simulation performance of the model. We use the same division and assign different optical parameters to BC aerosols of every particle size bin according to the Mie scattering theory.
Using the program of the Mie scattering model, we calculate the SSA, DF, and MAC of the BC aerosols in different bands.
Because the radiation parameters of BC aerosols have an important relationship with the particle size distribution, the spectral distribution of the particles is very important to numerical calculations of the optical parameters. Winchester et al. [18] indicated that a lognormal distribution can accurately describe the spectral distribution of particles. Other study also adopted a lognormal distribution to describe the spectral distribution of aerosols [19]. In this chapter, we adopt this distribution for the calculations: Based on the division of BC aerosol particle sizes, we consider BC with four bins (i.e., BC1, BC2, BC3, and BC4) with radii of 0.075, 0.205, 0.48, and 2.7 μm, respectively. In Equation (1), the values of r g and N 0 are from the results of Levy et al. [4]. Based on the conversion between the median radius and the distribution radius of the number concentrations, we obtain the radius distribution of the number concentrations for four modes, r g : In Equation (1), N 0 represents the particle number density of BC aerosols, and we can calculate di erent values of N 0 from r g under the di erent modes. V 0 is the amplitude of the volume distribution. σ represents the standard deviation of the radius:

Mie Scattering Model.
In using the Mie scattering theory to calculate the solution, we assume that the particles are spheres, and we can obtain the relevant optical characteristics (optical coe cient) based on the radius and complex refractive index of the particles. e Mie scattering is based on the Maxwell equation, and it assumes that the vector wave equation has separable solutions and is derived in the spherical coordinate system. Many studies have focused on numerical algorithms for Mie scattering, and the complete process is described by Liou, who used the Legendre function and Bessel function [20]. Figure 2 shows the numerical calculation results of the optical coe cients of BC in four particle sizes bins. BC1, BC2, BC3, and BC4 represent BC particles with a radius of 0.075, 0.205, 0.48 and 2.7 μm, respectively. BC5 represents the BC optical parameters in the conventional scheme.
As shown in the gure, BC aerosols in di erent modes show di erent absorption ability. BC5 in the conventional scheme shows the highest absorption while BC1, BC2, BC3, and BC4 show lower absorption in order. e optical parameters of the BC aerosols used in the conventional model are taken from database of the University Corporation for Atmospheric Research (UCAR) (http://www. ccsm.ucar.edu/models/atm-cam/download/) and are not di erentiated based on the particle size. BC5 represents the conventional settings of the optical parameters. A comparison of the results shows that the calculated optical parameters for the rst mode of BC aerosols (particle size of 0.075 μm) are most similar to those of the conventional settings, and the calculated MAC is smaller than that of the conventional settings.

Design of the Numerical Scheme
In this paper, we use the RIEMS2.0 model, and the emission source data are taken from the MEIC-2010 emissions inventory, which has a spatial resolution of 0.25°× 0.25°. e time period of the numerical experiments is from January 1, 2010 to December 31, 2010.
We input the optical coe cients of the BC aerosols calculated for the four particle sizes bins into the RIEMS2.0 model and simulate the concentration, direct radiation, and the climate e ect of the BC aerosols. In addition, we compare and analyze the results with those obtained by the conventional simulation scheme of BC aerosols with uniformscale-optical parameters (hereinafter referred to as Scheme 1) [15]. In the following analysis, the scheme used in this paper is called the size-resolved scheme of optical parameters (referred to as Scheme 2), and the results show the di erences between Schemes 2 and 1 simulation results. In this paper, we consider the direct radiative forcing e ect of BC aerosols and its climate e ect and compare them with the previously calculated direct radiative forcing and climate e ects. e method used for the regional division is the same as those in Reference [15].

Difference between the Two Schemes of Optical
Parameters on the Simulation of BC Aerosols. We use Schemes 1 and 2 to simulate the BC aerosols in China and find that there are differences in the concentrations, optical thicknesses, direct radiative forcing, and climate effects of the BC aerosols simulated using the two optical parameterization schemes. e differences between the ground surface concentrations of BC aerosols simulated by the two schemes of optical parameters ( Figure 3) show that the BC aerosol concentration of the first mode simulated by Scheme 2 is significantly higher, and the maximum difference is 3. e distributions of the variations in the column concentrations of BC aerosols are similar to that of the ground surface concentrations (not shown). e differences in the BC aerosol concentrations occur due to the different optical parameters, which has the complicated feedback effect in the aerosol-radiation-meteorological field could generate different meteorological fields, and further affect the transport and diffusion of aerosols. e optical thicknesses simulated by Scheme 2 are lower than those simulated by Scheme 1 and the differences range from −0.008 to −0.0064 (figure omitted). In the region of high BC aerosol optical thicknesses, the differences between the simulations are relatively large. For example, in the region near the Sichuan Basin and the North China Plain, the difference is approximately −0.0064; in the area with lower optical thicknesses, the differences are smaller. e regional  Advances in Meteorology average di erence in the BC aerosol optical depth (AOD) is −0.0019. e causes of the di erences in optical thickness include the di erences in the concentrations and the di erences in the optical parameters. Figure 4 shows the distributions of the di erences in climate e ects of BC aerosols simulated with two optical parameterization schemes. e direct radiative forcings under all-sky conditions simulated by Scheme 2 are higher than those from Scheme 1 with the range of 0.1-0.7 W/m 2 in most regions of central and eastern China but are slightly lower near the Sichuan Basin. e di erences in clear-sky top-of-atmosphere radiative forcing (TRF) are negative in most areas. In the North China Plain and South China, the TRF changes by −0.2 to −0.8 W/m 2 ; near the Sichuan Basin, there is a region of strongly negative variations of TRF with a maximum value of −0.94 W/m 2 , which is consistent with the variations in the BC aerosol optical depth. Over south of the Sichuan Basin, the clearsky TRF of BC is also intensi ed, but the range of enhancement is smaller than that of the all-sky TRF, which is approximately 0.2-0.4 W/m 2 . e regional average clearsky TRF of BC aerosols is 0.034 W/m 2 lower, whereas the all-sky TRF is 0.06 W/m 2 higher. In general, in comparison with the change of the all-sky TRF, the variation in the clear-sky direct TRF of BC aerosols is more similar to the distributions of the variations in the concentration. Using the size-resolved optical parameters, the total concentration of BC aerosols, the simulated AOD, and TRF are lower. Because the calculation of the all-sky TRF considers the complicated in uence of clouds, there are several di erences between the changes in the all-sky TRF of BC aerosols and the changes in the other variables. Some research results also show that the e ects of BC on clouds vary between regions and present much uncertainty [21]. e use of the principle of Mie scattering to calculate the size-resolved optical parameters causes a change in the direct climate e ect of BC aerosols. e ground surface temperatures simulated by the two schemes increase in many regions, and there are several similarities. However, the extents of the warming regions simulated by Scheme 2 are smaller, and the magnitudes of the warming are lower in some regions, which is consistent with the weakening of the clear-sky TRF of BC in these regions. In Northeast China, the warming e ect of BC is enhanced by 0.04 to 0.16 K. Figure 4(b) shows that this region also has signi cant increase in the clear-sky TRF of BC aerosols. In addition, in the Beijing-Tianjin-Hebei area and the Shandong Peninsula, the warming e ect of BC is intensi ed by 0 to 0.16 K. BC simulated by the two schemes mostly have cooling e ects in the arid and semiarid areas of western China.
e cooling e ects simulated using Scheme 2 are greater in some regions, and the cooling ranges from −0.04 to −0.24 K. e simulation of BC using Scheme 2 results in increased precipitation in South China, and the variations range from 0. 4   e climate e ects of aerosols show much uncertainty due to many factors and so far the mechanism has not been formed [22]. Compared to the e ects on temperature, the e ects of BC on precipitation are more unclear by the research [23]. In general, BC aerosols show a complicated in uence on precipitation, and the sensitivity of optical parameterization schemes makes it more di cult to analyze, which merits further studies.

e Climate E ect of BC Aerosols with the Size-Resolved
Scheme of Optical Parameters. e results described above indicate that the radiative forcing and climate e ect of BC aerosols are relatively sensitive to the optical parameters. If the dependence of the BC aerosols' optical parameters on the particle size is not considered in simulations of the climate e ect of BC aerosols, errors may occur. In this section, we analyze the distribution of BC aerosol concentrations and its climate e ect for the distribution scheme that considers the dependence of the aerosols' optical parameters on the particle size (Scheme 2). Figure 5 shows the spatial distributions of the annual average ground surface concentrations for the four modes and the total of BC aerosols simulated using Scheme 2. Figures 5(a)-5(d) show the results for the modes of BC aerosols with sizes of 0.075, 0.205, 0.48, and 2.7 μm, respectively, and Figure 5(e) shows the results for the total BC aerosols. e spatial distributions of BC aerosols with di erent particle sizes are similar, and the ground surface concentrations decrease from east to west, which is generally consistent with the emissions distribution of BC aerosols. In the North China Plain, Yangtze River Delta, and Sichuan Basin, which have dense industries and relatively high population densities, the BC aerosol concentrations are relatively high. Although the BC aerosol distributions of the four modes are similar, there are di erences between the modes. e regional average ground surface concentrations of the four modes of BC aerosols are 0.372, 0.265, 0.055, and 0.004 μg/m 3 , respectively. Before we adjust the optical properties of BC aerosols based on the particle size, the simulated proportions of the four modes are essentially consistent with their initial emissions proportions, whereas after recalculating the optical parameters according to Mie scattering, there are obvious di erences in the proportions of the four modes. e proportions of the rst and second modes, which have relatively small aerosol radii, are 53% and 38%, respectively, and the proportions of the third and fourth modes are very small, especially the fourth mode. e range of the total ground surface concentrations of BC aerosols in North China and East China is 1-6 μg/m 3 ( Figure 5(e)). In the lower reaches of the Yellow River and the Yangtze River Basin, as well as the Sichuan Basin, there are regions of high BC aerosol ground surface concentrations, and the maximum concentration reaches 9.83 μg/m 3 . e regional average total ground surface concentration of BC aerosols is 0.70 μg/m 3 . e distribution of column concentrations of BC aerosols is similar to the distribution of ground surface concentrations (not shown). e regional average column concentration of the rst mode of BC aerosols is the highest (0.15 mg/m 2 ) and that of the second mode is slightly lower (0.10 mg/m 2 ). e regional average total BC aerosol column concentration is 0.28 mg/m 2 , and the maximum is 2.05 mg/m 2 . In most other regions except for the arid and semiarid areas in western China, the range of BC aerosol column concentration is 0.20-2.0 mg/m 2 .

Concentration Distribution of BC Aerosols.
e distributions of BC aerosol ground surface concentrations in this study are similar to the results of previous studies, but the concentrations in this study are smaller, which could be caused by the size-resolved scheme [24,25]. Because many previous studies adopted a particle size distribution scheme with a constant scale and simply assumed the radius of BC aerosols to be approximately 0.1 μm [26][27][28]. ey did not accurately consider the presence of BC aerosols with larger particle sizes, which caused the simulated BC aerosol concentrations to be higher than they actually are.
e wind eld at the 850 hPa height during the spring ( gure omitted) shows that the eastern part of the Indian Peninsula is mainly a ected by westerly air ow, and the BC aerosols accumulate. In China, regions of high concentrations of BC aerosols are located in the Sichuan Basin and North China Plain. ere are many mountains and hills in the Sichuan Basin, and the topography makes it di cult for pollutants to di use. e westerly air ow also causes the accumulation of BC aerosols in this region, which causes the concentration of BC aerosols to be very high. e high-concentration region in the North China Plain is mainly caused by industrial and anthropogenic emissions. In the summer, southern China and the Indian Peninsula are both a ected by southwesterly air ows, and long narrow areas of high BC aerosol concentrations form in the Sichuan Basin and eastern China. Compared with the spring, the climate characteristics of northward and eastward transportation are obvious in southern China, and the column concentrations of BC aerosols increase in eastern China. Northwesterly air ows prevail in most regions north of the Yangtze River; therefore, the highconcentration area of BC aerosols in the North China Plain weakens. In the fall, the distribution of column BC aerosol concentrations is similar to those in the other seasons, but the concentrations are lower; southerly winds prevail in most of central and eastern China. In the winter, northwesterly winds prevail to the north of the Yangtze River,  and easterly winds prevail to the south of the Yangtze River. e concentrations of BC aerosols in North China are higher than that in the fall.
To validate the effect of models on the simulation of BC aerosol concentrations, we compare the simulated BC aerosol concentrations with observations. e observed values in different regions are obtained from several studies [29][30][31][32][33][34][35][36].
e simulated and observed annual average concentrations of BC aerosols in different regions of China and the linear relationship between the data have been studied (figure omitted). e straight black line is the fit between the observed values and simulated values, and the different symbols represent different regions.
e results show that the model can simulate the ground surface concentrations of BC aerosols relatively well, and the correlation coefficient with the observations is 0.78. e linear relationship is given by the equation y � 0.78x + 0.29. In general, the simulated values are lower than the observed values, which is likely because the emission source we used does not consider straw combustion. e simulated BC aerosol concentrations of Zhuang et al. [37] are also lower than the observed values. In the areas of Longfengshan, Jinsha, Taiyangshan, Nanning, and Chengdu, the simulation results and observations are relatively consistent. In the winter, in Jinsha and Zhengzhou stations, the model-simulated BC aerosol concentrations are lower than the observations. At most stations, the observations are low in the summer and high in the winter. Except for the stations in Gucheng, Dalian, and Gaolanshan, the model simulates the monthly variations in the BC aerosol concentrations relatively well. In Xi'an, Dunhuang, and Lhasa, the simulated concentrations of BC aerosols are significantly lower than the observations, which is likely because the anthropogenic emission sources in western areas are difficult to estimate.
is problem was also found in a similar study [25].

Optical ickness Distribution and Radiative Forcing of BC Aerosols.
Similar to the distribution of column concentrations, the regions of high BC aerosol optical thicknesses are located in North China and the Sichuan Basin (figure omitted). In most areas of North China and East China, the BC aerosol optical thicknesses range from 0.003 to 0.006. In the Sichuan Basin, the middle and lower reaches of the Yellow River and the Yangtze River Basin, and the Yangtze River Delta, the optical thicknesses of the BC aerosols are greater. In these regions, the industry is more well-developed, economic development has occurred more rapidly, and the population density is higher, which leading to higher emissions and therefore inducing greater BC aerosol concentrations. e regional average BC aerosol optical thickness is 0.0013, and the maximum is 0.0073. e model can simulate the spatial distribution of the BC aerosol optical thicknesses relatively well, and the distribution is similar to the results of previous studies [38,39]. However, the simulated optical thicknesses of BC aerosols in this study are slightly lower than those in some early studies [40]. Figure 7 shows the distributions of the all-sky TRF and clear-sky TRF of BC aerosols. e distributions of the TRF are similar to the distributions of the BC aerosol emissions and concentrations. However, the radiative forcing is affected by the aerosol mass as well as by the vertical structure of the aerosols, optical properties, ground surface albedo, and clouds [41]. erefore, there is a difference between the distribution of the BC aerosol radiative forcing and the concentrations. e all-sky TRFs of BC aerosols have a range of 0-2.0 W/m 2 . e high values are located near the Sichuan Basin and in Northeast China, and the maximum TRF is 2.11 W/m 2 . In the region between the Yangtze River and the Yellow River Basin, the all-sky TRFs of BC aerosols are relatively strong (1.4-1.8 W/m 2 ). e TRFs of BC aerosols under the clear-sky conditions are lower (0-1.8 W/m 2 ); the high values are located north of the Yellow River, and the maximum value is 1.98 W/m 2 , which is similar to the distribution of the BC aerosol optical thicknesses. In comparison with the all-sky TRFs, the distribution of the clear-sky TRFs is more similar to the distribution of the optical thicknesses. is is because the all-sky TRF is also affected by the mass and cloud water content of the aerosols, so it is more complicated. e regional average clear-sky and all-sky TRFs of BC aerosols are 0.49 and 0.36 W/m 2 , respectively, which are similar to the global average BC aerosol TRF (0.4 W/m 2 ) in the IPCC report [3]. In the same class of research areas, the regional average of the BC aerosol direct TRF estimated by Chang and Park [42] is 0.5 W/m 2 , and the BC aerosol TRFs calculated by Wu et al. [43] range from 0.5 to 4.1 W/m 2 . e regional average TRF of BC aerosols estimated by Li et al. [25] is 1.22 W/m 2 , and the maximum is 5-6 W/m 2 , whereas the average and maximum BC aerosol TRFs simulated by Zhuang et al. in 2010 [37] are 0.75 and 5.5 W/m 2 , respectively. In Chang et al. [44], the regional average TRF of BC aerosols is 0.58 W/m 2 . e results of this study are lower than those in these previous studies.
is is because the size-resolved scheme assumes that a proportion of the BC aerosols is larger and has weaker optical properties. In the numerical model, the radiation effect of BC aerosols is weakened by a feedback effect between the concentration and other physical quantities. If we do not consider the particle size distribution of BC aerosols and their optical parameters, the climate effect of BC aerosols will likely be overestimated.

Influence on Temperature and Precipitation.
BC aerosols change the energy equilibrium in the atmosphere through the absorption of solar radiation and cause changes in the dynamic-thermodynamic circulation. Figure 8 shows the effects of BC aerosols on the annual     Figure 9(a) shows that BC aerosols cause changes in the ground surface temperature in many regions of western China. is is because the large-scale transfer process of BC aerosols can change the properties of clouds in relatively distant regions and thus a ect the climate, and it can also change the climate at short distances through its in uences on the circulation, water vapor transportation, and cloud distribution [45]. It is worth noting that the distribution of strong BC aerosol radiative forcing is roughly consistent with the high value region of ground surface warming caused by BC aerosols. However, there are also several regions in which strong BC aerosol radiation causes decreases in the ground surface temperature. is phenomenon is possibly due to the inuences of complicated regional cloud systems and precipitation, which also re ect the complicated BC aerosolcloud feedback mechanism [40]. In several previous studies, the ranges of temperature changes caused by BC aerosols in East China and the eastern region of East Asia are −0.6 to 0.3 K and −0.4 to 0.1 K, respectively, which are greater than the climate e ects of BC aerosols simulated in this study [46,47]. Considering the particle size distribution of the optical parameters of BC aerosols causes the climate e ect of BC aerosols to weaken. e change in precipitation caused by BC aerosols is more complicated and has a greater regional uncertainty (Figure 9(b)). In parts to the south of the Yangtze River, the southeast coastal area, and north of the Yellow River, the BC aerosols increase the precipitation, varying from 0.5 to 2.5 mm/d. However, in some regions to the north of the Yangtze River and central China, the BC aerosols cause decreases in precipitation, and the range of the decreases is 0.5-3.0 mm/d. At a regional scale (Table 1), the BC aerosols cause decreases in precipitation in central China, North China, South China, and Northeast China, and the variations are −0.83, −0.05, −0.11, and 0.13 mm/d, respectively. In central China, the BC aerosols cause the precipitation to increase by 0.01 mm/d; for the entire simulation domain, the average decrease in precipitation caused by BC aerosols is 0.02 mm/d. BC aerosol emissions and precipitation in China are correlated over the past several years [7]. BC aerosols can cause increases in precipitation in southern China and decreases in precipitation in northern China.
is study found that the precipitation in the northern regions decreases. However, the regional average precipitation in the southern regions decreases, which is likely associated with the di erences between the models and di erent regional divisions. Due to the complexity of the in uences of BC aerosols and precipitation in the climate system, there is still signi cant uncertainty in the relationship between BC aerosols and precipitation. Many studies have analyzed the in uence of BC aerosols on precipitation in China. Several studies have indicated that BC aerosols cause decreases in precipitation in eastern China with a regional average reduction in precipitation of 0.09 mm/d [48], whereas other studies have indicated that BC aerosols can increase the precipitation and that the precipitation increased by 0.07 mm/d from 1950 to 2000 [44]. e in uences of BC aerosols on the temperatures at di erent heights have been researched ( gure omitted). At the 500 hPa height, BC aerosols cause warming in most regions of China. e range of warming is 0.01 to 0.05 K, which is signi cant in the region near the middle and lower reaches of the Yangtze River. On the other hand, in parts of Northeast China, BC aerosols cause cooling. e simulation at the height of 850 hPa indicates that BC aerosols cause increased temperatures near the Sichuan Basin and in the North China Plain as well as decreases in some regions of western China. A comparison of the in uences of BC aerosols on the temperatures at these two heights shows that in the north of the Yellow River and near the Sichuan Basin, the warming caused by BC aerosols in the lower troposphere is more signi cant than that in the middle and upper troposphere. erefore, the atmospheric instability in these regions increases, which may cause convective precipitation. Figure 8(b) shows that the precipitation in these regions also increases. A comparison of Figures 10(a) and 8(a) shows that at the 850 hPa height and at the ground surface, the distributions of the variations in the annual average temperatures caused by BC aerosols are similar.

In uences on the Circulation, Pressure, Temperature, and Precipitation in the Winter and Summer.
e absorption of solar radiation by BC aerosols causes heating of the middle and lower atmosphere, and the air ow rises. e research indicates that BC aerosols can a ect the wind changes in China [49]. Figure 11 shows the variations of the 850 hPa heights and the wind elds during the summer and winter. At the 850 hPa height, the decrease (increase) of the 850 hPa height eld during the winter and summer caused by BC aerosols corresponds to a convergence (divergence) of the wind eld. In the summer, the eastern region (near 110°E, 40°N) and central region (near 110°E, 30°N) of China correspond to low values and high values of height change, respectively. is region is located near the region of high BC aerosol emissions. In the winter, there are two dispersed regions of large height changes in central and eastern China. Correspondingly, BC aerosols cause the enhancement of the southwesterly air ow in the midlatitude region during the summer and the intensi cation of the southerly air ow in the high-latitude region. e change of the wind eld is the opposite during the winter, and the strength in the winter is stronger than that in the summer; it causes a clear divergence of the air ow in the region north of the Yangtze River and enhances the northerly wind during the winter. ese changes in wind and height eld caused by BC are similar to the results of other studies [37,49]. Figures 12 and 8 show the in uences of BC aerosols on the ground surface temperature and pressure eld during the summer and winter. BC aerosols may cause the temperature increase or decrease between di erent areas [40]. Seen in the gure, in summer, except for some regions north of the Yellow River and a few areas along the eastern coast, the BC aerosols generally cause the ground surface temperature to decrease; at the 500 hPa height, however, the temperatures in most regions increase. It is assumed that the cooling e ect caused by BC aerosols in South China and East China during the summer is induced by the complicated feedback and changes in the amount of clouds in the model, which merits further study [2,[50][51][52]. e region where the ground surface temperature decreases corresponds to an increase of the sea surface pressure. In most regions of South China and East China, the BC aerosols cause increases in the sea surface pressure, and the area of maximum pressure increase is located near the Sichuan Basin. is is because the cooling of the land over a large area causes the accumulation of the air mass, which results in increasing sea surface pressures  Advances in Meteorology 11 in most regions of central and southern China. e change of the ground surface temperature caused by BC aerosols is di erent in the winter than in the summer, and the temperatures increase in most regions of central and eastern China. In the region between the Yangtze River and the Yellow River Basin, the increases in ground surface temperature are signi cant, and the maximum warming is 0.42 K. In the Yangtze River Delta area, the ground surface temperatures decrease. e temperature decrease in winter can also be seen in Figure 10. In general, the changes of ground surface temperature corresponds relatively well to the changes of sea surface pressure in the winter (an increase in the ground surface temperature corresponds to a decrease in the sea surface pressure), which is coincident with other studies [37]. In the summer, the BC aerosols increase the temperature in some coastal regions in eastern China and intensify the di erence in the thermodynamic properties of both the sea and mainland. In the winter, however, warming occurs in some coastal regions in eastern China, and the di erence in the thermodynamic properties of the sea and mainland decreases. In general, the change in temperature at the 500 hPa height is smaller than that at the ground surface. Figure 10 shows the vertical distribution of the zonal average temperature changes caused by BC aerosols during the summer and winter. During the summer, in the region from 21°N to 31°N, the BC aerosols cause cooling of the boundary layer, and the maximum temperature decrease is −0.05 K. Also as seen in Figure 12(a), the surface temperature decreases due to BC in summer. In the region from 33°N to 41°N, the warming caused by the BC aerosols reaches approximately the 300 hPa height, and the maximum warming is 0.02 K. At heights from 700 hPa to 300 hPa, the BC aerosols generally cause the temperatures to increase. At heights above 200 hPa, the BC aerosols cause weak cooling. In the winter, BC aerosols cause temperature increases in most regions at heights below 500 hPa, and the temperature changes in the middle and high atmosphere are relatively weak.
is is coincident with the temperature changes shown in Figure 12. Warming centers are located near 26°N and 36°N, and the maximum warming exceeds 0.07 K. Figure 13 shows the in uence of BC aerosols on the precipitation in China during the summer and winter. Not the same as the temperature changes, the precipitation changes caused by BC show more uncertainty which make the distribution more complicated [21]. During the summer, the BC aerosols cause increases of precipitation in southern China and some regions north of the Yellow River. On the other hand, in the Yangtze River Basin and some regions of

Advances in Meteorology
complicated, such as circulation, radiation, and water vapor. erefore, the distribution of the influence of BC aerosols on the precipitation is random and irregular. At the regional scale, BC aerosols decrease the precipitation in South China, Northeast China, North China, and East China by 0.11, 0.13, 0.05, and 0.83 mm/d, respectively. In central China, BC aerosols increase the precipitation by 0.01 mm/d. Over the entire domain, the regional average reduction of precipitation caused by BC aerosols is 0.02 mm/d.

Summary
In numerical simulations, the optical parameters of aerosols can directly affect the radiation properties. e optical parameters of BC aerosols are related to the particle size and the complex refractive index. erefore, based on the size-resolved scheme used in previous studies, we use the principle of Mie scattering to calculate the optical parameters of BC aerosols of different scales and apply the new optical parameters to simulate the concentration distribution, radiative forcing, and climate effects of BC aerosols. We also compare the results with simulation results that use the conventional scheme and analyze the effects of the optical parameters on the simulation of the climate effects of BC aerosols.
In comparison with the conventional homogeneous parameterization scheme, the concentrations of the first mode of BC aerosols simulated with the optical parameters recalculated according to the particle size were higher, and the maximum increase was 3.30 μg/m 3 . e concentrations of the other modes of BC aerosols were lower, the total BC aerosol concentration was also lower, and the range of the reductions was −0.2 to −2.8 μg/m 3 . e regional average changes of the BC aerosol column concentrations for the different modes were 0.085, −0.095, −0.089, and −0.054 mg/m 2 . e BC aerosol optical thicknesses varied by −0.008 to −0.0064.
In general, the changes in the clear-sky direct radiative forcing of BC aerosols are similar to the distribution of the change in its optical thickness and concentration changes. e regional average clear-sky radiative forcing of BC aerosols decreases by 0.034 W/m 2 . e distributions of the influence of BC aerosols on the ground surface temperatures simulated with the two optical parameterization schemes are similar in many regions, but the warming region and the magnitude of warming simulated with the new scheme are smaller. However, the simulated cooling effect is enhanced in some regions, and the variation of the cooling is from −0.04 to −0.24 K. e simulation of BC aerosols using Scheme 2 enhances the precipitation in South China even more, and the variation is from 0.4 to 2.4 mm/d. In some regions in the middle and lower reaches of the Yangtze River Basin, the BC aerosols cause greater reductions in precipitation (−0.4 to −2.4 mm/d). e influence of BC aerosols on precipitation is complicated and variable, and the difference in the optical parameterization scheme also causes various factors to influence the precipitation, which merits further study.
Comparisons with observations show that Scheme 2 can relatively accurately simulate the ground surface concentrations of BC aerosols. e simulation results indicate that the regional average ground surface concentrations for BC aerosols with the four particle sizes are 0.372, 0.264, 0.055, and 0.004 μg/m 3 . e proportions of the first and second modes, which have relatively small radii, are 53% and 38%, respectively, and the proportions of the third and fourth modes are very small. e regional average total ground surface BC aerosol concentration is 0.69 μg/m 3 . e distribution of the column concentrations is similar to that of the ground surface concentrations. e regional average total BC aerosol column concentration is 0.28 mg/m 2 , and the maximum is 2.06 mg/m 2 . Regions of high BC aerosol concentrations are located in the Sichuan Basin and the North China Plain and are related to the spatial distribution of their emissions and wind fields. e concentrations of BC aerosols in the spring and summer are higher than those in the other seasons. e regional average BC aerosol optical thickness is 0.0013, and the maximum is 0.0073. e maximum direct radiative forcing of BC aerosols under allsky conditions is 2.11 W/m 2 . e region of high values of the clear-sky radiative forcing is located to the north of the Yellow River, and the maximum value is 1.98 W/m 2 . e regional average clear-sky and all-sky TRF values of BC aerosols are 0.49 and 0.36 W/m 2 , respectively.
In most regions of East China, North China, and central China, BC aerosols cause increases in the annual average ground surface temperature, which ranges from 0.05 to 0.15 K. In South China and some regions of Northwest China, BC aerosols cause decreases in temperature, and the cooling ranges from 0.05 to 0.15 K. At the regional scale, BC aerosols cause decreases in the annual average precipitation in East China, North China, South China, and Northeast China of −0.83, −0.05, −0.11, and −0.13 mm/d, respectively. In central China, BC aerosols cause the annual average precipitation to increase by 0.01 mm/d. In the winter and summer, the region where the 850 hPa height field decreases corresponds to a convergence of the wind field. e change in airflow caused by BC aerosols in the winter is greater than that in the summer. In the winter, BC aerosols cause a greater increase in the ground surface temperature. In the region between the Yangtze River and the Yellow River Basin, the increase in the ground surface temperature is significant, and the maximum warming is 0.42 K. During the summer, BC aerosols increase the precipitation in southern China and some regions to the north of the Yellow River. However, in the Yangtze River Basin and some regions of Northeast China, BC aerosols decrease the precipitation.
ere are good correlations between the changes of circulation, pressure, and temperature induced by BC aerosols. e climate effects of BC are related to its size distributions, and when using other size distributions, the results will change. is may be discussed in more detail in later studies. e results of this paper are related to the mixing sate of BC, and we just considered the simple situation. Taking the mixing state of BC and other aerosols into consideration may show different results.

Data Availability
e data used to support the findings of this study are available from the corresponding author upon request. 14 Advances in Meteorology