Ground-based remote sensing of aerosol climatology in China: Aerosol optical properties, direct radiative effect and its parameterization

Atmospheric Environment xxx (2015) 1e9

Contents lists available at ScienceDirect

Atmospheric Environment journal homepage: www.elsevier.com/locate/atmosenv

Ground-based remote sensing of aerosol climatology in China: Aerosol optical properties, direct radiative effect and its parameterization X. Xia a, b, *, H. Che c, **, J. Zhu a, H. Chen a, b, Z. Cong d, X. Deng e, X. Fan a, Y. Fu f, P. Goloub g, H. Jiang h, Q. Liu g, B. Mai f, P. Wang a, Y. Wu i, J. Zhang a, R. Zhang i, X. Zhang c a

LAGEO, Institute of Atmospheric Physics, CAS, Beijing, China Collaborative Innovation Center on Forecast and Evaluation of Meteorological Disasters, Nanjing University of Information Science & Technology, Nanjing, China c Chinese Academy of Meteorological Sciences, Beijing, China d Institute of Tibetan Plateau Research, CAS, Beijing, China e Guangzhou Institute of Tropical and Marine Meteorology, CMA, Guangzhou, China f College of Earth and Space Sciences, University of Science and Technology of China, Hefei, China g LOA, Universite Lille, Lille, France h International Research Center of Spatial-Ecology & Ecosystem Ecology, Zhejiang Forestry University, Hanzhou, China i RCE-TEA, Institute of Atmospheric Physics, CAS, Beijing, China b

h i g h l i g h t s
A high-quality database of aerosol optical properties in China was established.
An empirical relationship of AOD and SSA to ADRE has been introduced.
Spatio-temporal variation of aerosol optical properties and ADRE was revealed.

a r t i c l e i n f o

a b s t r a c t

Article history: Received 24 September 2014 Received in revised form 3 February 2015 Accepted 30 May 2015 Available online xxx

Spatio-temporal variation of aerosol optical properties and aerosol direct radiative effects (ADRE) are studied based on high quality aerosol data at 21 sunphotometer stations with at least 4-months worth of measurements in China mainland and Hong Kong. A parameterization is proposed to describe the relationship of ADREs to aerosol optical depth at 550 nm (AOD) and single scattering albedo at 550 nm (SSA). In the middle-east and south China, the maximum AOD is always observed in the burning season, indicating a significant contribution of biomass burning to AOD. Dust aerosols contribute to AOD significantly in spring and their influence decreases from the source regions to the downwind regions. The occurrence frequencies of background level AOD (AOD < 0.10) in the middle-east, south and northwest China are very limited (0.4%, 1.3% and 2.8%, respectively). However, it is 15.7% in north China. Atmosphere is pristine in the Tibetan Plateau where 92.0% of AODs are <0.10. Regional mean SSAs at 550 nm are 0.89e0.90, although SSAs show substantial site and season dependence. ADREs at the top and bottom of the atmosphere for solar zenith angle of 60 ± 5 are 16e37 W m2 and e66 e111 W m2, respectively. ADRE efficiency shows slight regional dependence. AOD and SSA together account for more than 94 and 87% of ADRE variability at the bottom and top of the atmosphere. The overall picture of ADRE in China is that aerosols cool the climate system, reduce surface solar radiation and heat the atmosphere. © 2015 Published by Elsevier Ltd.

Keywords: Aerosol Optical properties Aerosol direct radiative effect

1. Introduction

* Corresponding author. LAGEO, Institute of Atmospheric Physics, CAS, Beijing, China. ** Corresponding author. E-mail addresses: [email protected] (X. Xia), [email protected] (H. Che).

As the most populated and fastest developing country of the world, rapid economic growth in China leads to gradual drops in atmospheric environment (Li et al., 2007, 2011a). Recent studies suggested that increased aerosol loading might have significant

http://dx.doi.org/10.1016/j.atmosenv.2015.05.071 1352-2310/© 2015 Published by Elsevier Ltd.

Please cite this article in press as: Xia, X., et al., Ground-based remote sensing of aerosol climatology in China: Aerosol optical properties, direct radiative effect and its parameterization, Atmospheric Environment (2015), http://dx.doi.org/10.1016/j.atmosenv.2015.05.071

2

X. Xia et al. / Atmospheric Environment xxx (2015) 1e9

effects on weather and climate. For example, surface global and direct solar radiation decreased by about 8.6% and 4.6% per decade during 1961e2000 (Qian et al., 2006). Aerosols are likely the major driver of the cooling trend in the Sichuan Basin and central eastern China (Li et al., 1995). Aerosol effects on clouds could induce large changes in precipitation patterns, which in turn might change not only regional water resources, but also change the regional and global circulation systems (Qian et al., 2009; Li et al., 2011b). Aerosols, clouds and their interactions with climate are still the most uncertain areas of climate change and require multidisciplinary coordinated research efforts. Tropospheric aerosols are highly variable in time and space. Comprehensive understanding of aerosol optical properties is a foundation for further understanding of the aerosol-cloudradiation interactions and thereby an important base to reduce the uncertainty of aerosol's weather-climate effects (Huang et al., 2006a,b). Ground-based remote sensing of aerosol optical properties using sunphotometer is approved to be an important method to accurately characterize aerosol optical properties owing to its wide angular and spectral measurements of solar and sky radiation (Dubovik et al., 2002). The data are widely used to reveal spatiotemporal variation of aerosol optical properties, to evaluate satellite and model aerosol products, and to study aerosol-cloudradiation interactions (Holben et al., 2001). 2. Past research in ground based remote sensing of aerosol optical properties in China Sunphotomter has been used to observe aerosol since the beginning of 1980s in China (Mao et al., 2002). Aerosol optical depth (AOD) at 500 nm in Beijing during July 1980 to July 1981 ranged from 0.31 (September) to 0.65 (May) (Zhao et al., 1983). Lu et al. (1981) proposed to retrieve aerosol size distribution from simultaneous measurements of spectral extinction and forward scattering radiation. A field campaign was carried out in winter of 1981 to test this proposal, which showed that aerosol size distribution within 0.1e10 mm could be retrieved (Qiu et al., 1983). Qiu and Zhou (1986) further studied information content of sky radiance with regard to the aerosol size distribution, refractive index and surface reflectance. Aerosol refractive index was retrieved from sky radiance measurement in Beijing to be 1.517e0.034i and 1.533e0.016i during the heating and not-heating period, respectively (Li and Mao, 1989). Much progress has been made since the ending of 1990s. One year of AOD measurements at four stations were made (Zhang et al., 2002). Several SKYNET stations have been established since the beginning of 2000s as a result of an international cooperation (Shi et al., 2005). Extensive measurements of AOD at four wavelengths have been made with handheld sunphotometers at 24 stations starting since august 2004 (Xin et al., 2007). The data were used to reveal spatial-temporal variation of AOD and to classify aerosol types (Wang et al., 2011). Aerosol single scattering albedo (SSA) at 500 nm was retrieved from AOD and satellite measured reflected radiance at the top of the atmosphere (TOA). The nationwide means of SSA at 500 nm in 2005 was 0.89 ± 0.04 (Lee et al., 2007). In spring of 2001, four Aerosol Robotic Network (AERONET) stations have been established in North China for the first time. Aerosol optical properties in dust source region were compared with that in the downwind regions and potential variations of aerosol optical properties were studied (Cheng et al., 2006). Since spring of 2002, more than 30 AERONET stations have been established across China. Long-term measurements have been made at several AERONET stations, for example, at Beijing since April 2002, at Xianghe since September 2004 and at Taihu since September 2005 (Li et al., 2007), as well as at Lanzhou (SACOL) since July 2006

(Huang et al., 2008a,b). Furthermore, China Aerosol Research Network (CARSNET) with more than 20 stations was established in 2002 by the Chinese Meteorological Bureau (Che et al., 2009). Climatology of aerosol optical properties and aerosol direct radiative effects (ADRE) were studied on the basis of these measurements. For example, a distinct seasonal variation of AOD and SSA was observed in Beijing, i.e., higher values in summer and lower in winter, which was quite different from surface aerosol concentration (Xia et al., 2006). Spring maximum AOD in the northwestern China was observed in spring (Bi et al., 2010). Dust aerosol's absorption and atmospheric heating by dust absorption in the Taklimakan Desert was estimated (Ge et al., 2011; Huang et al., 2009). The objective of this study is to present the spatio-temporal variation of aerosol optical properties based on aerosol optical data at 21 sunphotometer stations with at least four months’ worth of measurement. This is the first attempt to combine ground-based remote sensing data as many as possible to reveal spatio-temporal variability of aerosol optical properties in China. Distinct seasonal variation of aerosol optical properties is firstly revealed. Analysis of aerosol direct radiative effect at the bottom (BOA) and top of the atmosphere is then performed. Finally, a parameterization of the relationship of ADRE to AOD and SSA at 550 nm (AOD and SSA hereafter if not specified) is established. The paper is organized as follows. Site and data are described in following section. Section 3 presents seasonal variation of aerosol optical properties in different regions of China. A parameterization is proposed in Section 4 to reveal how ADRE varies with AOD and SSA. Section 5 summarizes the research and the major results. 3. Site and data Aerosol optical data at forty-two sunphotomter stations with at least one month of measurements are available. In order to present climatological aspect of spatio-temporal variation of aerosol optical properties, 21 stations with more than four months of measurements are used. Taking the predominant aerosol types and the proximity to source areas into account, the stations have been grouped into 5 regions, i.e., R1: south and southwest China; R2: the middle-east China; R3: the north and northeast China; R4: The northwest region; R5: The Tibetan Plateau (Fig. 1). Table 1 summarized information on all stations.

Fig. 1. Spatial distribution of sunphotometer stations. The stations are grouped into five regions, i.e., R1: south and southwest China; R2: the middle-east China; R3: the north and northeast China; R4: the northwest China; and R5: the Tibetan Plateau.

Please cite this article in press as: Xia, X., et al., Ground-based remote sensing of aerosol climatology in China: Aerosol optical properties, direct radiative effect and its parameterization, Atmospheric Environment (2015), http://dx.doi.org/10.1016/j.atmosenv.2015.05.071

X. Xia et al. / Atmospheric Environment xxx (2015) 1e9

3

Table 1 Site information of 21 stations used in the analysis. Region

Site information

R1

Kunming, 25.01 N, 102.65 E, 1889 m a.s.l., urban, 265 days, 2012.04e2013.08 Panyu, 23.00 N,113.35 E, 182 m a.s.l., urban, 140 days, 2011.01e2013.12 Hong_Kong_PolyU, 22.30 N, 114.18 E, 30 m a.s.l., urban, 741 days, 2005.04e2011.10 Hong_Kong_Hok_Tsui, 22.21 N, 114.26 E, 80 m a.s.l., urban, 186 days 2007.11e2010.07 Hong_Kong_Sheung, 22.48 N, 114.12 E, 40 m a.s.l., urban, 97 days, 2012.02e2013.07 Hefei, 31.98 N, 116.38 E, 92 m a.s.l., urban, 271 days, 2011.01e2013.11 Taihu, 31.42 N, 120.22 E, 20 m a.s.l., urban, 936 days, 2005.09e2012.10 Shouxian, 32.56 N,116.78 E, 22 m a.s.l., Suburban, 90 days, 2008.05e2008.12 Hanzhou_ZFU, 30.26 N, 119.73 E, 14 m a.s.l., urban, 50 days, 2007.08e2008.11 Beijing, 39.98 N, 116.38 E, 92 m a.s.l., urban, 2431 days, 2001.03e2012.08 Xianghe, 39.75 N, 116.96 E, 36 m a.s.l., suburban, 1849 days, 2001.03e2012.05 Xinglong, 40.40 N, 117.58 E, 970 m a.s.l., Mountain, 1090 days, 2001.03e2012.05 Tongyu, 44.42 N, 122.92 E, 182 m a.s.l., rural, 834 days, 2010.01e2013.12 Dunhuang, 40.09 N, 94.41 E, 1140 m a.s.l., Oasis, 506 days, 2011.11e2013.12 Yulin, 38.28 N, 109.72 E, 1080 m a.s.l., urban, 213 days, 2001.04e2002.10 Lanzhou, 36.03 N, 103.53 E, 1517 m a.s.l., urban, 336 days, 2012.06e2013.12 SACOL, 35.95 N, 104.14 E, 1965 m a.s.l., rural, 1296 days, 2006.01e2012.08 Lhasa, 29.50 N, 91.13 E, 3648 m a.s.l., urban station over the Tibetan Plateau, 333 days, 2011.12e2013.12 NAM_CO, 30.77 N, 90.96 E, 4740 m a.s.l., Mountain, 626 days, 2006.08e2011.01 QOMS_CAS, 28.36 N, 86.95 E, 4276 m a.s.l., Mountain, 660 days, 2010.09e2012.12 Mt_WLG, 36.28 N, 100.90 E, 3816 m a.s.l., Mountain, 309 days, 2009.09e2012.12

R2

R3

R4

R5

A CE-318 sun/sky radiometer is used in these stations. The radiometer measures direct solar spectral radiation at wavelengths from 340 nm to 1020 nm and the angular distribution of sky radiance at 440, 675, 870, and 1020 nm. AOD accuracy was estimated to be 0.01 to 0.02 (Eck et al., 1999). The modeled error of AOD owing to the effect of aerosol forward scattering is lower than AOD calibration uncertainty for 99.53% of AERONET level 2.0 data € m exponent (AE) is calculated (Sinyuk et al., 2012). The Ångstro from AOD at 440 and 870 nm. The inversion algorithm retrieves

aerosol optical properties from almucantar scans of radiances and AODs. The aerosol inversion algorithm was developed by Dubovik and King (2000) and was then further improved to consider aerosol's non-sphericity (Dubovik et al., 2006). SSA uncertainty is estimated to be less than 0.03 for AOD at 440 nm > 0.4 (Dubovik et al., 2000). The inversion algorithm also calculates broadband solar radiation (Garcia et al., 2012). Solar radiation products under different aerosol environments at the BOA showed an excellent agreement with surface measurements (Garcia et al., 2008; Li et al.,

Fig. 2. Monthly mean and one standard deviation of aerosol optical depth at 550 nm (upper panel), Angstrom exponent (440e870 nm) (middle panel) and single scattering albedo at 550 nm (bottom panel) at each station in five regions.

Please cite this article in press as: Xia, X., et al., Ground-based remote sensing of aerosol climatology in China: Aerosol optical properties, direct radiative effect and its parameterization, Atmospheric Environment (2015), http://dx.doi.org/10.1016/j.atmosenv.2015.05.071

4

X. Xia et al. / Atmospheric Environment xxx (2015) 1e9

Table 2 Annual mean aerosol optical depth at 550 nm, Angstrom exponent and single scattering albedo at 550 nm. Region

Station

AOD550nm

R1

Kunming Panyu Hong_Kong_PolyU Hong_Kong_Hok_Tsui Hong_Kong_Sheung Hefei Taihu Hanzhou_ZFU Shouxian Beijing Xianghe Xinglong Tongyu Dunhuang Yulin Lanzhou SACOL Lhasa NAM_CO QOMS_CAS Mt_WLG

0.28 0.55 0.46 0.39 0.37 0.63 0.69 0.68 0.74 0.62 0.63 0.31 0.25 0.36 0.40 0.66 0.37 0.10 0.04 0.05 0.12

R2

R3

R4

R5

± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ±

0.15 0.23 0.24 0.18 0.18 0.31 0.39 0.39 0.43 0.59 0.56 0.28 0.27 0.35 0.24 0.28 0.22 0.08 0.02 0.29 0.11

AE440e870nm 1.30 1.27 1.30 1.28 1.09 1.23 1.22 1.35 1.24 1.02 1.11 1.10 1.11 0.39 0.76 0.82 0.89 0.67 0.56 0.94 0.79

± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ±

0.26 0.19 0.20 0.20 0.21 0.26 0.23 0.20 0.20 0.29 0.29 0.31 0.42 0.22 0.28 0.26 0.27 0.30 0.31 0.44 0.44

SSA550nm 0.87 0.90 0.89 0.93 0.88 0.90 0.92 0.87 0.92 0.90 0.91 0.94 0.91 0.90 0.87 0.83 0.92 N/A N/A N/A N/A

± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ±

0.03 0.02 0.04 0.02 0.06 0.04 0.03 0.05 0.01 0.04 0.03 0.02 0.04 0.03 0.04 0.04 0.03

N/A represents retrievals are not available.

2010a,b). ADRE is derived as the difference in the net solar radiation at the TOA and BOA between model calculations in the present and absence of aerosols. 4. Spatio-temporal variation of aerosol optical properties Aerosols are classified into different types based on aerosol size and absorption. AOD versus AE or fine mode fraction of AOD (FMF) versus SSA are generally used in the literature (Wang et al., 2011; Lee et al., 2010). Here we use both methods to classify aerosol types. For the former method, cases with AOD <0.10 were defined

as the background level of aerosol loading (Holben et al., 2001). Cases with AOD >0.10 were classified into three groups, i.e., dust with AE < 0.20, polluted dust with AE within 0.20e0.60 (Eck et al., 2010) and fine mode dominated aerosols with AE > 0.60. Analysis of FMF and AE data showed that these FMF threshold values (0.30 and 0.50) were equivalent to AE values of 0.20 and 0.60, respectively. Therefore, dust and polluted dust are defined to be aerosols with FMF <0.30 and FMF within 0.30e0.50 for the latter method. Fine mode dominated aerosols (FMF > 0.50) were classified into four groups according to their SSA values, i.e., highly absorbing with SSA < 0.85, moderately absorbing with SSA of 0.85e0.90, slightly absorbing with SSA of 0.90e0.95, and very weakly absorbing with SSA > 0.95. The occurrence frequency of dust and polluted dust types derived from AOD versus AE is somewhat larger than that from FMF versus SSA because a portion of data is classified to the background level of aerosols by the former. Fig. 2 presents monthly mean AOD and AE at each station in five regions; furthermore, monthly mean SSA in all regions except R5 is also presented. Annual mean AOD, AE and SSA at each station are presented in Table 2. Fig. 3 presents the relative number density plot of AOD and AE (upper panel) and FMF and SSA (bottom panel). There are five urban stations in R1 region where anthropogenic emissions are dominant and seasonal biomass burning emissions also contribute to a lot. The winter period is the peak months of fire activity as a result of burning sugar-cane and rice residues (Korontzi et al., 2006; R. Zhang et al., 2013), which is supported by the elevated levels of potassium in October and November (Andreae et al., 2008). Peak values of AOD in March and October in this region are likely related to biomass burning (Fig. 2). Relatively lower AOD and AE were generally observed in rainy season (MayeSeptember), which is likely owing to strong wet deposition of precipitation (lower AOD) and aerosol hydroscopic growth (lower AE) under humid environment. A great majority of AE is larger than 1.0, indicating that fine mode aerosols are dominant in this region. The occurrence frequency of AOD <0.1 is 1.3%, indicating nearly always

Fig. 3. Relative number density plots for the aerosol optical depth at 550 nm and Angstrom exponent (440e870 nm) in five regions (upper panel) and for the fine mode fraction of aerosol optical depth at 550 nm and single scattering albedo at 550 nm in four regions (bottom panel). The color scale represents the relative density of points, where orange to red colors (levels ~45e64) indicate the highest number density. The data points are classified into four aerosol types and the percentage of each type is presented. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

Please cite this article in press as: Xia, X., et al., Ground-based remote sensing of aerosol climatology in China: Aerosol optical properties, direct radiative effect and its parameterization, Atmospheric Environment (2015), http://dx.doi.org/10.1016/j.atmosenv.2015.05.071

X. Xia et al. / Atmospheric Environment xxx (2015) 1e9

presence of haze. Aerosol retrieval products are limited in summer, the rainy season. SSAs in fall and winter show a moderate variation and SSA difference among five stations is remarkable, indicating potential different aerosol emissions at the local scale. Aerosols with moderate and slight absorption (SSA within 0.85e0.95) account for 67% of fine mode dominated aerosols (FMF > 0.5). Measurements were made at four urban stations in R2. Monthly mean AOD is generally close to or larger than 0.5, indicating a regional pollution in this region. This statement is supported by the fact that a consistent seasonal variation of aerosol optical properties at all stations. The maximum AOD (ranging from about 1.0 to 1.5 depending on station that is e two times larger than those in the following months) occurs in June that is likely ascribed to crop residue burning. Fire season generally spans from May to July and June is the peak month of fire activity as a result of harvest of winter wheat. Cheng et al. (2014) found that biomass open burning contributed 37% of PM2.5, 70% of organic carbon and 61% of elemental carbon in MayeJune 2011. Xia et al. (2013) revealed that crop fires enhanced AOD, nitrogen dioxide and carbon monoxide near the fire source regions and in downwind regions by ~15e60% in June 2012. The largest standard deviation of AOD is also observed in June. This fact is likely accounted for by the occasional occurrence of huge smoke layers. However, it should also be noted that measurements in summer are very limited as a result of cloud contamination that may contribute to this fact, for example, there are only 7 days measurements in June at Shouxian. Aerosol loading in spring is occasionally affected by mineral dust transport from North China, which is supported by 3.6% of measurements characterized by AOD > 0.1 and AE < 0.6. Analysis of FMF versus SSA suggests that 3.5% of retrievals are classified into dust and polluted

5

dust, showing relatively larger contribution of long-range transport of dust aerosols here than that in R1. Very few occurrences of AOD < 0.1 (only 0.4%) are observed here, which is likely because of the lack of a strong air pollution dispersion mechanism. SSA varies around 0.90 and shows slight seasonal variation. About 70% of SSAs of fine mode dominated aerosols (FMF > 0.5) are within 0.85e0.95. The data in R3 are obtained from two urban and two rural stations. 15.7% of AOD < 0.1, much larger than that in R1 and R2. This is mainly because air pollution is dramatically dispersed by synoptic processes that occasionally occur in winter months (Li et al., 2007). A consistent seasonality of AOD and AE is observed. Higher AOD occurs in spring and summer but lower in fall and winter. The lowest AE in spring suggests substantial impact of dust events on aerosol size. Analysis of AOD versus AE shows that 6.8% of measurements are associated with dust and polluted dust cases. The percentage decreases to 3.0% based on FMF versus SSA, indicating that dust and polluted dust account for a portion of background samples. A distinct and consistent seasonal variation of SSA is observed. Higher SSAs in summer months is likely due to larger amount of production of secondary particles with weak absorption. Zhang et al. (2012, 2013) found that high humidity accelerated the conversion rate of SO2 to sulfate and photochemistry was enhanced to convert NO2 to ammonium. High humidity also favors hydroscopic growth of aerosols (Eck et al., 2010). Relatively smaller SSA in winter is likely associated with large emission of absorbing aerosols as a result of using coal for heating (Zhang et al., 2013). SSA at Xinglong is generally larger than that at Beijing and Xianghe by 0.03e0.07, indicating strong local emission of absorbing aerosols in industrial regions (Zhang et al., 2012). SSAs at Xianglong and Tongyu, although both are rural sites, are quite different, indicating

Fig. 4. The scatter-plot of aerosol optical depth at 550 nm and aerosol direct radiative effects at the top and bottom of the atmosphere as well as in the atmosphere. The analysis is limited at solar zenith angle of 60 ± 5 . The solid line represents the result of linear regression analysis of aerosol and its radiative effects. The values in the brackets are the standard deviations.

Please cite this article in press as: Xia, X., et al., Ground-based remote sensing of aerosol climatology in China: Aerosol optical properties, direct radiative effect and its parameterization, Atmospheric Environment (2015), http://dx.doi.org/10.1016/j.atmosenv.2015.05.071

6

X. Xia et al. / Atmospheric Environment xxx (2015) 1e9

substantial different emissions of absorbing aerosols between North China Plain (Xinglong) and Northeast China (Tongyu). The largest spread of SSA is observed in this region that is partly ascribed to substantial spatial and seasonal variation of SSA. 47% of SSAs for fine mode dominated aerosols (FMF > 0.5) are 0.85e0.95, being slightly less than that in other regions. The maximum percentage (36%) of fine mode dominated aerosols with high and moderate absorption (SSA < 0.85) is observed, on the other hand, this region is also characterized by the largest concurrency frequency (16%) of very weak aerosols (SSA > 0.95). There are four representative stations in R4 region. Maximum AOD and minimum AE is observed in spring as a result of dust events. Winter AOD in Lanzhou city is >0.50, which is likely associated with strong emission of anthropogenic aerosols (AE > 1.00) trapped by the strong inversion layer in winter. This region is the most vulnerable area impacted by dust among five regions because it is close to the dust source regions (Zhang et al., 2003). The occurrence frequency of dust and polluted dust is 22.6% based on AOD/AE that is more than three times larger than that in R3. The percentage decreases to 17.2% based on FMF and SSA analysis. Substantial spatial variation of SSA is observed (0.1 in some cases). SSA at urban site (Lanzhou) is generally lower, implying strong local emission of absorbing aerosols in the urban area. Furthermore, the different mixing state of dust and pollution could substantially modulate SSA in this region (Li et al., 2010a,b). About 67% of

aerosols have SSA within 0.85e0.95 for the fine mode dominated aerosols, which is close to that in R1 and R2. The lowest AOD is observed in the Tibetan Plateau since there is little trace of human habitation and few anthropogenic emissions. Its high altitude (>3 km) also prevents frequent long-range transport of aerosols from outside. A distinct seasonal AOD variation is evident. Lower AODs (<0.05) are observed in fall and winter. The maximum seasonal AOD is observed in spring (NAM_CO) or in summer (Lhasa). AOD at Lhasa is generally double than that at NAM_CO and QOMS_CAS, indicating notable local anthropogenic emissions at Lhasa, one of the highest cities in the world. A few AODs >0.10 are resulted from dust events (AE < 0.60, 5.2%) or occasional anthropogenic pollution events (AE > 0.60, 2.8%). Longrange transport of fine particles during the dry season from South Asia occasionally impacts this region, which is characterized by AOD > 0.2 and AE > 1.0 (Xia et al., 2011). Cases with AE < 0.20 and AOD > 0.20 are likely attributable to transport of dust aerosols from the desert (Huang et al., 2007; Chen et al., 2013) or local dust events (Zhang et al., 2000), which is supported by the individual particles analysis (Cong et al., 2009). Monthly AOD at MT_WLG is approximately 0.20 and 0.29 in March and April that is accompanied by the minimum AE values, which is even double than that at Lhasa. This is because MT_WLG is located at the transport path of dust aerosols from the Taklimakan Desert (Che et al., 2011).

Fig. 5. The scatter plot of aerosol optical depth at 550 nm and aerosol direct radiative effect at the bottom of the atmosphere for 20 narrow range of the solar zenith angle. Varying SSAs are represented by difference colors. The solid line represents the regression result with fixed SSA.

Please cite this article in press as: Xia, X., et al., Ground-based remote sensing of aerosol climatology in China: Aerosol optical properties, direct radiative effect and its parameterization, Atmospheric Environment (2015), http://dx.doi.org/10.1016/j.atmosenv.2015.05.071

X. Xia et al. / Atmospheric Environment xxx (2015) 1e9

5. Aerosol direct radiative effects Fig. 4 presents the scatter-plot of ADRE and AOD in four regions. The regional mean AOD and SSA are also presented in the figure. In order to compare results reported in the literature (Garcia et al., 2012), ADRE results have been limited at the solar zenith angle (SZA) of 60 ± 5 . The mean and one standard deviation of ADREBOA are 87 ± 42 (AOD ¼ 0.49 ± 0.29), e111 ± 43 (AOD ¼ 0.64 ± 0.35), e86 ± 62 (AOD ¼ 0.50 ± 0.48) and 66 ± 34 W m2 (AOD ¼ 0.34 ± 0.22) in regions from R1 to R4. Large standard deviation of ADREBOA is related to variations of SZA, aerosol optical properties and surface albedo. ADREBOA is close to the value derived in the Asiatic urbanindustrial region (80 ± 52 W m2) (Garcia et al., 2012). The regional ADREBOA efficiencies derived from the linear analysis vary from 148 in R3 to 181 W m2 in R4. The efficiency is slightly larger than those observed in the American regions (135 W m2) but close to those in the European urban-industrial region (165 W m2) (Garcia et al., 2012). ADRETOA is 31 ± 15, e37 ± 18, e26 ± 24 and 16 ± 17 W m2 in four regions, indicating that aerosols cool the climate system, which is not consistent with the result of Li et al. (2010a,b) who reported that ADRETOA was marginal. A couple of cases with positive ADRETOA occur in winter because of SSA <0.80 and snow cover (Che et al., 2014).ADRETOA efficiencies vary from 48 W m2 in R4 to 59 W m2 in R1. On the basis of ADRETOA and ADREBOA , the overall picture of ADRE can be summarized as follows. Aerosols

7

result in a significant reduction of surface solar radiation, about 12%e23% of which is reflected back to the TOA and 77%e82% of which is consumed to heat the atmosphere. Thus one would expect a significant aerosol radiative effect on the atmospheric stability. 6. Parameterization of ADRE to AOD and SSA A large scatter of ADRE to AOD is obvious as shown in Fig. 4. This scatter is likely associated with variation of SZA, aerosol optical properties except AOD and surface albedo. In order to better understand how aerosol optical properties determine ADRE, a parameterization of ADRE to AOD and SSA is established. The relationship of ADRE to AOD and SSA at a narrow range of SZA is firstly studied to minimize the effect of SZA on ADRE. The scatterplot of AOD and ADREBOA and ADRETOA at twenty SZAs is presented in Figs. 5 and 6. SSAs for AOD at 440 nm < 0.4 are from Level 1.5 retrieval products. The ADRE rate decreases as AOD increases as shown in Figs. 5 and 6. This fact is due to the increase multiple scattering effects and attenuation of the transmitted radiation as AOD increases (Garcia et al., 2012). Hence, a power law equation is used to describe this relationship: ADRESZA ¼ a  AODb550nm . SSA moderately modulates ADRE for the same AOD and SZA. We found that a varied linearly with SSA, but b not, so we derived the following equation to parameterize the relationship of ADRE to AOD and SSA: ADRESZA ¼ ða1  SSA þ a2 Þ  AODb550nm . The relative standard deviations of the regression analysis using this equation are generally <5%. Notable contribution of SSA to ADRE can be

Fig. 6. Similar as Figure 5 but for aerosol direct radiative effect at the top of the atmosphere.

Please cite this article in press as: Xia, X., et al., Ground-based remote sensing of aerosol climatology in China: Aerosol optical properties, direct radiative effect and its parameterization, Atmospheric Environment (2015), http://dx.doi.org/10.1016/j.atmosenv.2015.05.071

8

X. Xia et al. / Atmospheric Environment xxx (2015) 1e9

easily estimated from this parameterization. It was shown that variation of SSA by 0.01 could result in changes of ADRE efficiency at the BOA and TOA by about 2 and 5 W m2, respectively, indicating that accurate estimation of SSA is absolutely necessary for better understanding of ADRE, especially in polluted regions. AOD and SSA determine >94% and >87% of ADRE variability at the BOA and TOA, respectively. As a comparison, the standard deviations of the linear regression are 2e4 times larger, showing the better performance of the power law equation. The remaining variation is likely associated with variation of aerosol size and surface albedo, which needs further study. The advantage of this parameterization is that it can be used to derive instantaneous ADRE from observations of AOD and SSA, for example, by satellite retrievals. 7. Discussion and conclusions We attempt to further our understanding of spatio-temporal variations of aerosol optical properties based on the most complete aerosol optical data with high quality. Potential regional and intra-regional differences in aerosol optical properties are explored. Furthermore, ADREs and their relationships to AOD and SSA are studied in detail. Distinct seasonal variation of AOD is revealed. Higher AOD is observed in early spring and middle fall in South and southwest China as a result of significant biomass burning contribution. Biomass burning in June accounts for the maximum AOD in the middle-east China. Dust events in spring result in higher AOD in the northwest China and in the Tibetan Plateau. Spring and summer AOD in the north and northeast China is larger than that in fall and winter. The occurrence frequencies of background level AOD (AOD < 0.10) in five regions are substantially different. Only 0.4% of AODs are <0.10 in the middle-east China. The percentage is 1.3% and 2.8% in the south China and the northwest China, respectively. The percentage increases to 15.7% in north and northeast China. Atmosphere is pristine in the Tibetan Plateau where 92.0% of AODs are <0.10. For the fine mode dominated aerosols, the majority of SSAs are within 0.85e0.95. SSAs at the same region show substantial site and season dependence, suggesting remarkable difference in emission strength of absorbing aerosols or in the mixing state between absorbing and scattering components. However, the regional mean SSA is 0.89e0.90 and shows little regional dependence. A simple but effective relationship between ADRE and AOD as well as SSA has been established, on the basis of which >94% and >87% of ADRE variability at the BOA and TOA can be explained by AOD and SSA. ADREBOA and ADRETOA ranges from 66e111 W m2 and from 16e37 W m2 for the SZA of 60 ± 5 . The general function of aerosols is to cool the climate system, cool the surface but heat the atmosphere. Acknowledgments We thank Prof. Janet Elizabeth Nichol and Prof. Jianping Huang and their staff for establishing and maintaining the Hong Kong sites and SACOL site used in this investigation. This research is funded by the Strategic Priority Research Program of the Chinese Academy of Sciences (XDA05100301) and National Natural Science Foundation of China (41175031). References Andreae, M., Schmid, O., Yang, H., Chand, D., Yu, J., Zeng, L., Zhang, Y., 2008. Optical properties and chemical composition of the atmospheric aerosol in urban Guangzhou, China. Atmos. Environ. 42, 6335e6350.

Bi, J., Huang, J., Qiang, F., Wang, X., Shi, J., Zhang, W., Huang, Z., Zhang, B., 2010. Toward characterization of the aerosol optical properties over Loess Plateau of Northwestern China. J. Quant. Spectrosc. Rad. Transf. 112, D00K17. http:// dx.doi.org/10.1029/2009JD013372. Che, H., Zhang, X., Chen, H., Damiri, B., Goloub, P., Li, Z., Zhang, X., Wei, Y., Zhou, H., Dong, F., Li, D., Zhou, T., 2009. Instrument calibration and aerosol optical depth validation of the China aerosol remote sensing network. J. Geophys. Res. 114, D03206. http://dx.doi.org/10.1029/2008JD011030. Che, H., Wang, Y., Sun, J., 2011. Aerosol optical properties at Mt. Waliguan observatory, China. Atmos. Environ. 45, 6004e6009. Che, H., Xia, X., Zhu, J., Li, Z., Dubovik, O., Holben, B., Goloub, P., Chen, H., Estelles, V., Cuevas-Agullo, E., Blarel, L., Wang, H., Zhao, H., Zhang, X., Wang, Y., Sun, J., Tao, R., Zhang, X., Shi, G., 2014. Column aerosol optical properties and aerosol radiative forcing during a serious haze-fog month over North China Plain in 2013 based on ground-based sunphotometer measurements. Atmos. Chem. Phys. 14, 2125e2138. http://dx.doi.org/10.5194/acp-14-2125-2014. Chen, S., Huang, J., Zhao, C., Qian, Y., Leung, R., Yang, B., 2013. Modeling the transport and radiative forcing of Taklimakan dust over the Tibetan Plateau: a case study in the summer of 2006. J. Geophys. Res. 118 http://dx.doi.org/10.1002/ jgrd.50122. Cheng, T., Wang, H., Xu, Y., Li, H., Tian, L., 2006. Climatology of aerosol optical properties in northern China. Atmos. Environ. 40, 1495e1509. Cheng, Z., Wang, S., Fu, X., Watson, J., Jiang, J., Fu, Q., Chen, C., Xu, B., Yu, J., Chow, J., Hao, J., 2014. Impact of biomass burning on haze pollution in the Yangtze River delta, China: a case study in summer 2011. Atmos. Chem. Phys. 14, 4573e4585. Cong, Z., Kang, S., Dong, S., Zhang, Y., 2009. Individual particle analysis of atmospheric aerosols at Nam Co, Tibetan Plateau. Aer. Air Qua. Res. 9, 323e331. , D., Dubovik, O., Holben, B., Eck, T., Smirnov, A., Kaufman, Y., King, M., Tanre Slutsker, I., 2002. Variability of absorption and optical properties of key aerosol types observed in worldwide locations. J. Atmos. Sci. 59, 590e608. Dubovik, O., King, M., 2000. A flexible inversion algorithm for retrieval of aerosol optical properties from sun and sky radiance measurements. J. Geophys. Res. 105, 20673e20696. Dubovik, O., Smirnov, A., Holben, B., King, M., Kaufman, Y., Eck, T., Slutsker, I., 2000. Accuracy assessments of aerosol optical properties retrieved from Aerosol Robotic Network (AERONET) Sun and sky radiance measurements. J. Geophys. Res. 105, 9791e9806. Dubovik, O., Sinyuk, A., Lapyonok, T., Holben, B., Mishchenko, M., Yang, P., Eck, T., Volten, H., Munoz, O., Veihelmann, B., van der Zande, W., Leon, L., Sorokin, M., Slutsker, I., 2006. Application of light scattering by spheroids for accounting for particle non-sphericity in remote sensing of desert dust. J. Geophys. Res. 111, D11208. http://dx.doi.org/10.1029/2005JD006619. Eck, T., Holben, B., Reid, J., Dubovik, O., Smirnov, A., O'Neill, N., Slutsker, I., Kinne, S., 1999. Wavelength dependence of the optical depth of biomass burning, urban, and desert dust aerosols. J. Geophys. Res. 104, 31333e31349. Eck, T., Holben, B., Sinyuk, A., Pinker, R., Goloub, P., Chen, H., Chatenet, B., Li, Z., Singh, R., Tripathi, S., Reid, J., Giles, D., Dubovik, O., O'Neill, N., Smirnov, A., Wang, P., Xia, X., 2010. Climatological aspects of the optical properties of fine/ coarse mode aerosol mixtures. J. Geophys. Res. 115 (D19205) http://dx.doi.org/ 10.1029/2010JD014002. Ge, J., Huang, J., Su, J., Bi, J., Fu, Q., 2011. Shortwave radiative closure experiment and direct forcing of dust aerosol over northwestern China. Geophys. Res. Lett. 38, L24803. http://dx.doi.org/10.1029/2011GL049571. Garcia, O., Diaz, J., Exposito, F., Diaz, A., Dubovik, O., Derimian, Y., Dubuisson, P., Roger, J., 2012. Shortwave radiative forcing and efficiency of key aerosol types using AERONET data. Atmos. Chem. Phys. 12, 5129e5145. Garcia, O., Diaz, A., Exposito, F., Diaz, J., Dubovik, O., Dubuisson, P., Roger, J., Eck, T., Sinyuk, A., Derimian, Y., Dutton, E., Schafer, J., Holben, B., Garcia, C., 2008. Validation of AERONET estimates of atmospheric solar fluxes and aerosol radiative forcing by ground-based broadband measurements. J. Geophys. Res. 113, D21207. http://dx.doi.org/10.1029/2008JD010211. Holben, B., Tanre, D., Smirnov, A., Eck, T., Slutsker, I., Abuhassan, N., Newcomb, W., Schafer, J., Chatenet, B., Lavenue, F., Kaufman, Y., Castle, J., Setzer, A., Markham, B., Clark, D., Frouin, R., Halthore, R., Karnieli, A., O'Neill, N., Pietras, C., Pinker, R., Voss, K., Zibordi, G., 2001. An emerging ground-based aerosol climatology: aerosol optical depth from AERONET. J. Geophys. Res. 106, 12067e12097. Huang, J., Fu, Q., Su, J., Tang, Q., Minnis, P., Hu, Y., Yi, Y., Zhao, Q., 2009. Taklimakan dust aerosol radiative heating derived from CALIPSO observations using the FuLiou radiation model with CERES constraints. Atmos. Chem. Phys. 9, 4011e4021. Huang, J., Lin, B., Minnis, P., Wang, T., Wang, X., Hu, Y., Yi, Y., Ayers, J., 2006a. Satellite-based assessment of possible dust aerosols semi-direct effect on cloud water path over East Asia. Geophys. Res. Lett. 33 http://dx.doi.org/10.1029/ 2006GL026561. Huang, J., Minnis, P., Lin, B., Wang, T., Yi, Y., Hu, Y., Sun-Mack, S., Ayers, K., 2006b. Possible influences of Asian dust aerosols on cloud properties and radiative forcing observed from MODIS and CERES. Geophys. Res. Lett. 33, L06824. http:// dx.doi.org/10.1029/2005GL024724. Huang, J., Zhang, W., Zuo, J., Bi, J., Shi, J., Wang, X., Chang, Z., Huang, Z., Yang, S., Zhang, B., Wang, G., Feng, G., Yuan, J., Zhang, L., Zuo, H., Wang, S., Fu, C., Chou, J., 2008a. An overview of the semi-arid climate and environment research observatory over the loess plateau. Adv. Atmos. Sci. 25 (6), 1e16. Huang, J., Minnis, P., Chen, B., Huang, Z., Liu, Z., Zhao, Q., Yi, Y., Ayers, J., 2008b. Longrange transport and vertical structure of Asian dust from CALIPSO and surface measurements during PACDEX. J. Geophys. Res. 113, D23212. http://dx.doi.org/

Please cite this article in press as: Xia, X., et al., Ground-based remote sensing of aerosol climatology in China: Aerosol optical properties, direct radiative effect and its parameterization, Atmospheric Environment (2015), http://dx.doi.org/10.1016/j.atmosenv.2015.05.071

X. Xia et al. / Atmospheric Environment xxx (2015) 1e9 10.1029/2008JD010620. Huang, J., Minnis, P., Yi, Y., Tang, Q., Wang, X., Hu, Y., Liu, Z., Ayers, K., Trepte, C., Winker, D., 2007. Summer dust aerosols detected from CALIPSO over the Tibetan Plateau. Geophys. Res. Lett. 34, L18805. http://dx.doi.org/10.1029/ 2007GL029938. Korontzi, S., McCarty, J., Loboda, T., Kumar, S., Justice, C., 2006. Global distribution of agricultural fires in croplands from 3 years of Moderate Resolution Imaging Spectroradiometer (MODIS) data. Glob. Biogeochem. Cycles 20, GB2021. http:// dx.doi.org/10.1029/2005GB002529. Lee, K., Li, Z., Wong, M., Xin, J., Wang, Y., Hao, W., Zhao, F., 2007. Aerosol single scattering albedo estimated across China from a combination of ground and satellite measurements. J. Geophys. Res. 112, D22S15. http://dx.doi.org/10.1029/ 2007JD009077. Lee, J., Kim, J., Song, C., Kim, S., Chun, Y., Sohn, B., Holben, B., 2010. Characteristics of aerosol types from AERONET sunphtometer measurements. Atmos. Environ. 2010 (44), 3110e3117. Li, C., Tsay, S., Fu, J., Dickerson, R., Ji, Q., Bell, S., Gao, Y., Zhang, W., Huang, J., Li, Z., Chen, H., 2010a. Anthropogenic air pollution observed near dust source regions in northwestern China during springtime 2008. J. Geophys. Res. 114, D00K22. http://dx.doi.org/10.1029/2009JD013659. Li, J., Mao, J., 1989. Optical remote sensing of aerosol properties. ACTA Meteo. Sin. 47, 450e456. Li, X., Zhou, X., Li, W., 1995. The cooling of Sichuan province in recent 40 years and its probable mechanism. Acta Meteorol. Sin. 9, 257e268. Li, Z., Xia, X., Cribb, M., Mi, W., Holben, B., Wang, P., Chen, H., Tsay, S., Eck, T., Zhao, F., Dutton, E., Dickerson, R., 2007. Aerosol optical properties and their radiative effects in northern China. J. Geophys. Res. 112 (D22) http://dx.doi.org/10.1029/ 2006JD007382. Li, Z., Lee, K., Wang, Y., Xin, J., Hao, W., 2010b. First observation-based estimates of cloud-free aerosol radiative forcing across China. J. Geophys. Res. 115, D00K18. http://dx.doi.org/10.1029/2009JD013306. Li, Z., Li, C., Chen, H., Tsay, S.C., Holben, B., Huang, J., Li, B., Maring, H., Qian, Y., Shi, G., Xia, X., Yin, Y., Zheng, y., Zhuang, G., 2011a. East Asian studies of tropospheric aerosols and their impact on regional climate (EAST-AIRC): an overview. J. Geophys. Res. 116 (D7) http://dx.doi.org/10.1029/2010JD015257. Li, Z., Niu, F., Fan, J., Liu, Y., Rosenfeld, D., Ding, Y., 2011b. Long-term impacts of aerosols on the vertical development of clouds and precipitation. NatureGeoscience. http://dx.doi.org/10.1038/NGEO1313. Lu, D., Zhou, X., Qiu, J., 1981. The principle and numerical experiment of remote sensing of atmospheric aerosol size distribution by combined solar extinction and forward scattering method. Sci. China 12, 1516e1524. Mao, J., Zhang, J., Wang, M., 2002. Summary comment on research of atmospheric aerosol in China. ACTA Meteo. Sin. 60, 625e634. Qian, Y., Kaiser, D., Leung, L., Xu, M., 2006. More frequent cloud-free sky and less surface solar radiation in China from 1955 to 2000. Geophys. Res. Lett. 33 (1), 1e4. Qian, Y., Gong, D., Fan, J., Leung, L., Bennartz, R., Chen, D., Wang, W., 2009. Heavy pollution suppresses light rain in China: observations and modeling. J. Geophys. Res. 114, D00K02. http://dx.doi.org/10.1029/2008JD011575. Qiu, J., Wang, H., Zhou, X., Lu, D., 1983. Experimental study of remote sensing of

9

atmospheric aerosol size distribution by combined solar extinction and forward scattering method. Sci. Atmos. Sin. 7, 33e41. Qiu, J., Zhou, X., 1986. Simultaneous determination of aerosol size distribution and refractive index and surface albedo from radiance e part I: theory. Adv. Atmos. Sci. 3, 162e171. Shi, G., Wang, H., Wang, B., Gong, S., Zhao, T., Li, W., Aoki, T., 2005. Sensitivity experiments on the effect of optical properties of dust aerosols on their radiative forcing under clear sky condition. J. Meteo. Soc. Jpn. 83A, 333e346. Sinyuk, A., Holben, B., Smirnov, A., Eck, T., Slutsker, I., Schafer, J., Giles, D., Sorokin, M., 2012. Assessment of error in aerosol optical depth measured by AERONET due to aerosol forward scattering. Geophys. Res. Lett. 39, l23806,. http://dx.doi.org/10.1029/2012GL053894. Wang, Y., Xin, J., Li, Z., Wang, S., Wang, P., Hao, W., Nordgren, B., Chen, H., Wang, L., Sun, Y., 2011. Seasonal variations in aerosol optical properties over China. J. Geophys. Res. 116, D18209. http://dx.doi.org/10.1029/2010JD015376. Xia, X., Chen, H., Wang, P., Zhang, W., Goloub, P., Chatenet, B., Eck, T., Holben, B., 2006. Variation of column-integrated aerosol properties in a Chinese urban region. J. Geophys. Res. 111, D05204. http://dx.doi.org/10.1029/2005JD006203. Xia, X., Zong, X., Cong, Z., Chen, H., Kang, S., Wang, P., 2011. Baseline continental aerosol over the central Tibetan plateau and a case study of aerosol transport from South Asia. Atmos. Environ. 45, 7370e7378. Xia, X., Zong, X., Sun, L., 2013. Exceptionally active agricultural fire season in mideastern China in June 2012 and its impact on the atmospheric environment. J. Geophys. Res. Atmos. 118 http://dx.doi.org/10.1002/jgrd.50770. Xin, J., Wang, Y., Li, Z., Wang, P., Hao, W., Nordgren, B., Wang, S., Liu, G., Wang, L., Wen, T., Sun, Y., Hu, B., 2007. Aerosol optical depth (AOD) and Angstrom exponent of aerosols observed by the Chinese Sun Hazemeter Network from August 2004 to September 2005. J. Geophys. Res. 112, D05203. http://dx.doi.org/ 10.1029/2006JD007075. Zhang, J., Mao, J., Wang, M., 2002. Analysis of aerosol extinction characteristics in different areas of China. Adv. Atmos. Sci. 19, 136e152. Zhang, R., Jing, J., Tao, J., Hsu, S., Wang, G., Cao, J., Lee, C., Zhu, L., Chen, Z., Zhao, Y., Shen, Z., 2013. Chemical characterization and source apportionment of PM2.5 in Beijing: seasonal perspective. Atmos. Chem. Phys. 13, 7053e7074. Zhang, X., Arimoto, R., Cao, J., An, Z., Wang, D., 2000. Atmospheric dust aerosol over the Tibetan Planteau. J. Geophys. Res. 106, 18,471e18,476. Zhang, X., Gong, S., Shen, Z., Mei, F., Xi, X., Liu, L., Zhou, Z., Wang, D., Wang, Y., Cheng, Y., 2003. Characterization of soil dust aerosol in China and its transport and distribution during 2001 ACE-Asia: 1. Network observations. J. Geophys. Res. 108, 4261. http://dx.doi.org/10.1029/2002JD002632. Zhang, X., Wang, Y., Niu, T., Zhang, X., Gong, S., Zhang, Y., Sun, J., 2012. Atmospheric aerosol compositions in China: spatial/temporal variability, chemical signature, regional haze distribution and comparisons with global aerosols. Atmos. Chem. Phys. 12, 779e799. Zhang, Y., Shao, M., Lin, Y., Luan, S., Mao, N., Chen, W., Wang, M., 2013. Emission inventory of carbonaceous pollutants from biomass burning in the Pearl River Delta Region China. Atmos. Environ. http://dx.doi.org/10.1016/ j.atmosenv.2012.05.055. Zhao, B., Wang, Q., Mao, J., Qin, Y., 1983. Optical remote sensing atmospheric aerosol and water vapor. China Sci. D. 10, 951e962.

Please cite this article in press as: Xia, X., et al., Ground-based remote sensing of aerosol climatology in China: Aerosol optical properties, direct radiative effect and its parameterization, Atmospheric Environment (2015), http://dx.doi.org/10.1016/j.atmosenv.2015.05.071