- Email: [email protected]
Aerosol scattering properties in northern China
ARTICLE IN PRESS
Atmospheric Environment 41 (2007) 6916–6922 www.elsevier.com/locate/atmosenv
Short communication
Aerosol scattering properties in northern China Hao Yan Division of Remote Sensing Applications, National Meteorological Center, China Meteorological Administration, Zhongguancun South Street 46#, Haidian District, Beijing 100081, China Received 1 February 2007; received in revised form 25 April 2007; accepted 25 April 2007
Abstract The aerosol scattering properties were investigated at two continental sites in northern China in 2004. Aerosol light scattering coefficient (ssp) at 525 nm, PM10, and aerosol mass scattering efficiencies (a) at Dunhuang had a mean value of 165.17148.8 M m1, 157.67270.0 mg m3, and 2.3073.41 m2 g1, respectively, while these values at Dongsheng were, respectively, 180.27151.9 M m1, 119.07112.9 mg m3, and 1.8771.41 m2 g1. There existed a seasonal variability of aerosol scattering properties. In spring, at Dunhuang PM10, ssp, and a were 184.17211.548 mg m3, 126.3789.6 M m1, and 1.0570.97 m2 g1, respectively, and these values at Dongsheng were 146.47142.1 mg m3, 183.4781.7 M m1, and 1.9871.52 m2 g1, respectively. However, in winter at Dunhuang PM10, ssp, and a were 158.17261.4 mg m3, 303.37 165.2 M m1, and 3.1771.93 m2 g1, respectively, and these values at Dongsheng were 155.77170.1 mg m3, 304.47 158.1 M m1, and 2.9071.72 m2 g1, respectively. ssp and a in winter were higher than that in spring at both the sites, which coincides with the characteristics of dust aerosol and pollution aerosol. Overall, the dominant aerosol types in spring and winter at both sites in northern China are dust aerosol and pollution aerosol, respectively. r 2007 Elsevier Ltd. All rights reserved. Keywords: Dust and pollution aerosol; Integrating nephelometer; Scattering coefficient; PM10
1. Introduction The effect of aerosols on the climate has drawn much attention in recent years due to their largely uncertain climate forcing (The Intergovernmental Panel on Climate Change (IPCC), 2001). Through scattering and absorption of solar radiation, aerosols directly affect visibility and climate through the modification of the Earth’s energy balance. Thus, the aerosol scattering properties are required in evaluating the radiative forcing of aerosols (Vrekoussis et al., 2005; Kim et al., 2005; Xu et al., 2002). Tel.: +86 10 68407470; fax: +86 10 62172982.
E-mail address: [email protected] 1352-2310/$ - see front matter r 2007 Elsevier Ltd. All rights reserved. doi:10.1016/j.atmosenv.2007.04.052
However, the distribution of aerosols is quite patchy and it depends on the presence of aerosol sources (Vrekoussis et al., 2005). The regional climate effects of these aerosols are predicted to increase in the near future (Takemura et al., 2001). More specifically, in Asia, in the context of global warming, the desert and desertification may become much drier than ever before and emit a larger amount of dust aerosol from dust storms. In addition, China is the world’s fastest growing economy, which may lead to an increase in pollution aerosols from industrial sources. Asian aerosols have drawn much attention during the last 10 years. Some observations of Asian aerosols have been carried out during recent years. The Asian Pacific
ARTICLE IN PRESS H. Yan / Atmospheric Environment 41 (2007) 6916–6922
6917
Regional Aerosol Characterization Experiment (ACEAsia) shows that the dust from East Asia mixed with pollutants, resulting in a multi-component aerosol of great complexity and variability (Clarke et al., 2004), and dust aerosols had complex chemical and optical properties (Anderson et al., 2003). However, pollution aerosol was poorly investigated in ACE-Asia. Some measurement conducted in China mainly focused on chemical, physical, and radiative properties of pollution aerosols in urban regions, such as Beijing (Bergin et al., 2001), Wuhan (Waldman et al., 1991), Datong (Salmon et al., 1994), and Yangtze delta region (Xu et al., 2002) for a short period. Besides, the aerosol characters and mass scattering efficiency of different sources transported southeastward from Asian continent to Taiwan were investigated in Taiwan during East Asian winter monsoon (Chang et al., 2006). Similarly, Kim et al. (2005) analyzed the aerosol optical properties for two cases, viz. a heavy dust episode in April and a regionalscale pollution event in November in Korea. It can be concluded that there exist two types of aerosols, namely dust aerosol and pollution aerosol. However, the measurement in the above research was conducted far from their sources or just for a few cases. This study focuses on the seasonal variability of dominant aerosol type on the basis of the continuous measurement of aerosol light scattering coefficient and mass concentration for a whole year of 2004 at two sites in northern China.
(94.61E, 40.11N, 1139 m), in Gansu Province, which is near the Takelamagan Desert. The second one is located in Dongsheng city (109.91E, 39.81N, 1460 m) in inner-Mongolian Province which is near the Kubuqi Desert. They belong to arid and semiarid regions in northern China characterized by dry conditions. Spring refers to March, April, and May and winter refers to January, February, and December. Aerosol light scattering coefficient (ssp) was measured at these locations with a standard integrating nephelometer (Ecotech-M9003). This instrument measures ssp at 525 nm at 5 min intervals. The optical and electrical background noise is sufficiently low to allow measures of ssp (for particles) from o10% of air Rayleigh (sspo0.3 M m1) to 42000 M m1. PM10 was measured at these locations with an ambient particulate monitor (Rupperecht and Patashnick (R&P) model TEOM Series 1400a), continuously. This instrument equipped with an inlet having a 10 mm cut-point measures the ambient particulate mass concentration of PM10 in real time at 5 min intervals. All PM10 and ssp data in the following were averaged from the smallest 5 min intervals.
2. Experiment
The scatter plot of daily mean PM10 and ssp for Dunhuang and Dongsheng observation stations are presented. At Dunhuang, the PM10 daily mean values ranged between 4.2 and 1795.6 mg m3 with a
Sampling was conducted at two remote sites in 2004. The first one is located in Dunhuang city
3. Results and discussion 3.1. Comparison of PM10 vs ssp
Fig. 1. Scatter plots of daily mean values of PM10 and aerosol light scattering coefficient (ssp) at Dunhuang (a) and Dongsheng (b) in 2004.
ARTICLE IN PRESS 6918
H. Yan / Atmospheric Environment 41 (2007) 6916–6922
mean value of 157.67270.0 mg m3, and the ssp daily mean values ranged between 22.2 and 880.5 M m1 with a mean value of 165.17148.8 M m1 (Fig. 1(a)). At Dongsheng, the PM10 daily mean values ranged between 8.0 and 886.6 mg m3 with a mean value of 119.07112.9 mg m3, and the ssp daily mean values ranged between 19.2 and 1511.7 M m1 with a mean value of 180.27 151.9 M m1 (Fig. 1(b)). In order to show the seasonal variability of ssp clearly, daily scatter plots of 4 seasons with a regression equation at Dunhuang are provided, respectively (Fig. 2). At Dunhuang, several heavy dust events occurred in spring and summer with higher values of PM10 (1600 mg m3) and ssp (700 M m1) (Figs. 2(a and b)). Similarly, at Dongsheng, several heavy dust events occurred in
spring and winter with high values of PM10 (500 mg m3) and ssp (300 M m1) (Fig. 1). In all, high values of PM10 and ssp corresponded with dust aerosol checked with surface observation. There are some methods for discriminating between dust aerosol and pollution aerosol. For example, one can analyze the ratio of PM2.5 to PM10, the column A˚ngstro¨m exponent, chemical compositions, and air mass trajectories (Xu et al., 2002; Chang et al., 2006; Kim et al., 2005). However, the dust particles and pollutants are also likely to be mixed with each other even during dust episode or pollution episode (Kim et al., 2005). Thus, heavy dust events were just checked with surface meteorological observation in this paper, and heavy aerosol with values of PM10 (200 mg m3) and ssp (400 M m1) in winter and fall
Fig. 2. Scatter plots of daily mean values of PM10 and aerosol light scattering coefficient (ssp) at Dunhuang in spring (a), summer (b), fall (c), and winter (d) for 2004.
ARTICLE IN PRESS H. Yan / Atmospheric Environment 41 (2007) 6916–6922
was expected to be pollution aerosol due to an accepted fact that pollution aerosol in cold seasons prevails in urban area of northern China. This point was further verified with aerosol mass scattering efficiency data, because dust and pollution aerosols have different mass scattering efficiencies (Xu et al., 2002; Chang et al., 2006; Kim et al., 2005; Vrekoussis et al., 2005). In fall and winter, low values of PM10 (200 mg m3) and higher values of ssp (400 M m1) (Figs. 2(c and d)) were found at Dunhuang. Similarly, low values of PM10 (200 mg m3) and high ssp (700 M m1) (Fig. 1) were also found at Dongsheng in winter. In all, it is our inference that higher values of ssp and lower values of PM10 correspond with pollution aerosol. To our knowledge, few researches report such scatter plots of two typical aerosols with heavy mass concentration at one location. Similarly, an observation of anthropogenic pollution with PM10 ranging from 57 to 77 mg m3 and dust aerosol with PM10 ranging from 103 to 132 mg m3 measured at Taipei basin was reported (Chang et al., 2006). Kim et al. (2005) stated another measurement of pollution with values for PM10 and ssp of 56.9 mg m3 and 244.2 M m1 and dust with values for PM10 and ssp of 107.9 mg m3 and 239.8 M m1 during Asian dust and pollution periods at Gosan, Korea. As these aerosols were transported far from Asian continent, PM10 and ssp are much lower than the values measured in this paper. 3.2. PM10 The monthly and seasonal variability of PM10 for Dunhuang and Dongsheng observation stations is presented in Fig. 3. At Dunhuang, higher values of PM10 were found all the year except September and October (Fig. 3(a)). Higher values of PM10 were always related with dust storm. A lot of dust storm occurred in spring, winter, and even in summer at Dunhuang. At Dongsheng, higher values of PM10 were found in February and March (Fig. 3(b)), which was caused by dust storm. However, a lower value of annual PM10 could be found in Dongsheng compared with Dunhuang. PM10 presents a distinct seasonal pattern (Fig. 3(c)). Higher values were found in spring and winter, and lower values were encountered in fall at both stations. However, PM10 also shows higher values in summer at Dunhuang due to dust events.
6919
3.3. Aerosol light scattering coefficient (ssp) At Dunhuang, higher values of ssp were found in November, December, and even in July, while lower values were found in August and September (Fig. 3(d)). However, ssp at Dongsheng shows a distinct seasonal pattern (Fig. 3(e)). Maximum values were discovered in January and December, and minimum values appeared in September and October. ssp shows a different seasonal pattern at both stations compared with PM10. Maximum values of ssp were often found in winter rather than in spring (Fig. 3(f)). Although dust storm resulted in higher PM10 and higher ssp, dust aerosol could not explain this discrepancy in seasonal patterns for PM10 and ssp. The discrepancy was largely due to the pollution aerosol dominant in winter with higher values of ssp. 3.4. Mass scattering efficiencies (a) The ratio of ssp to PM10 mass concentration is the aerosol mass scattering efficiency (a). The a at both the sites in this paper presents a distinct seasonal pattern. At Dunhuang, maximum values were found in December, whereas minimum values from March to June (Fig. 3(g)). At Dongsheng, maximum values were also found in December and January, whereas minimum values in September and October (Fig. 3(h)). Obviously, the dust events could affect the PM10 and ssp at both the sites (Fig. 1). So in order to calculate the mass scattering efficiency for dust (adust), only significant dust observations (above 300 mg m3) with high ssp levels above 100 M m1 were analyzed in the following text. The regression between PM10 and ssp indicates that there exists a significant correlation with a slope of 0.43 (r ¼ 0.9, samples ¼ 23) at Dunhuang and a slope of 0.29 (r ¼ 0.75, samples ¼ 13) at Dongsheng. Further, the regression between all PM10 and ssp in spring also gives a slope of 0.34 (r ¼ 0.79, samples ¼ 52) at Dunhuang (Fig. 3(g). In addition, the adust for Dunhuang and Dongsheng coincide with the value of 0.52 m2 g1 measured at Negev Desert in Israel (Andreae et al., 2002) and that of 0.21–0.96 m2 g1 measured in the Eastern Mediterranean (Vrekoussis et al., 2005) for dust aerosol. Similarly, only significant pollution observations (below 300 mg m3) with high ssp levels above 200 M m1 were selected for calculation of the mass
ARTICLE IN PRESS 6920
H. Yan / Atmospheric Environment 41 (2007) 6916–6922
Fig. 3. Monthly and seasonal mean values and standard deviations of PM10 (upper panels), ssp (middle panels), and a (bottom panels) at Dunhuang and Dongsheng in 2004.
scattering efficiency for pollution aerosol (apollutant) in this paper. The regression between PM10 and ssp shows that there exists a correlation with a slope of 1.75 (r ¼ 0.54, samples ¼ 21) at Dunhuang and a slope of 2.62 (r ¼ 0.52, samples ¼ 43) at Dongsheng. Further, the regression between PM10 and ssp in winter gives a slope of 1.05 (r ¼ 0.31, samples ¼ 29) at Dunhuang (Fig. 2(d)) regardless of the two dust events (PM104400 mg m3) checked with surface observation. In addition, the apollutant for Dunhuang and Dongsheng coincide with the value of 1.4–2.3 m2 g1 measured at Taipei basin (Chang et al., 2006) and that of 1.65–2.0 m2 g1 measured at Yulin in China (Alfaro et al., 2003) for pollution aerosol.
Although higher values of adust and apollutant were also found in other measurements compared with measurements in this paper. adust had higher values of 0.89–1.15 m2 g1 (Alfaro et al., 2003), 2.2 m2 g1 (Kim et al., 2005), and 3.3–4.4 m2 g1 (Kim et al., 2004). apollutant had higher values of 4.13 m2 g1 (Kim et al., 2001), 4.5 m2 g1 (Kim et al., 2005), and 6.7 m2 g1 (Kim et al., 2004). The higher mass scattering efficiencies measured in Korea (Kim et al., 2004, 2005) is likely due to truncation corrections in the nephelometer and secondary growth of the aerosol. The dust and pollution aerosol measured in Korea passed over the highly polluted eastern China and was far from the source region and had time to accumulate more sulfate and nitrate
ARTICLE IN PRESS H. Yan / Atmospheric Environment 41 (2007) 6916–6922
than aerosol close to the source region. The truncation correction of near-forward scattered light for TSI nephelometer measurements has been described by Anderson and Ogren (1998). This sizedependent truncation correction method has been widely applied based on the A¨ngstro¨m exponent (Anderson and Ogren, 1998; Sheridan et al., 2001; Anderson et al., 2003; Kim et al., 2005). However, this correction is large, especially for larger dust particles, quite uncertain, and related with instrument and aerosol size distribution (Anderson et al., 2003; Kus et al., 2004), so for the present study, we make no truncation correction. The values of adust and apollutant measured at the same site through the same measure method were compared, and the comparison shows that adust is always lower than apollutant even at different sites. Kim et al. (2005) reported values for adust and apollutant, namely 2.2 and 4.5 m2 g1, respectively, in Korea. Chang et al. (2006) presented values for adust and apollutant, viz. 0.7–1.0 and 1.3–1.6 m2 g1 in Taiwan, respectively. The values for adust and apollutant (0.29–0.43 and 1.75–2.62 m2 g1, respectively) were found in northern China. This is largely attributed to the large particle size, non-spherical shape, and lower refractive index of the dust aerosol compared with the pollution aerosol. Obviously, a was bigger in winter than that in spring at both sites (Fig. 3(i)), which suggests that the dominant aerosol in spring is dust aerosol and the dominant aerosol in winter is pollution aerosol. During the cold seasons, especially in winter, pollution aerosol often becomes the dominant type of aerosol due to coal burning (Akimoto et al., 2006), weak surface winds, and stable atmospheric conditions. The discrepancy in PM10 and ssp seasonal patterns (Figs. 3(c and f)) are largely due to the difference of dominant aerosol types in different seasons. 4. Conclusion Aerosol scattering efficiency, PM10, and mass scattering efficiency measured at two continental sites in northern China in 2004 were analyzed in this paper. In spring, at Dunhuang PM10, ssp, and a were 184.17 211.548 mg m3, 126.3789.6 M m1, and 1.0570.97 m2 g1, respectively, and at Dongsheng, these values were 146.47142.1 mg m3, 183.4781.7 M m1, and 1.9871.52 m2 g1, respectively. However, in winter at Dunhuang PM10, ssp, and a were 158.17261.4 mg m3, 303.37165.2 M m1, and 3.1771.93 m2 g1,
6921
respectively, and at Dongsheng these values were 155.77170.1 mg m3, 304.47158.1 M m1, and 2.907 1.72 m2 g1, respectively. ssp and a in winter had higher values than that in spring at both the sites, which coincide with the characteristics of dust aerosol and pollution aerosol discovered by other researches (Kim et al., 2005; Chang et al., 2006; Alfaro et al., 2003). Thus, it can be inferred that the dominant aerosol type in spring and winter at both sites is dust aerosol and pollution aerosol, respectively. References Akimoto, H., Ohara, T., Kurokawa, J.I., Horii, N., 2006. Verification of energy consumption in China during 1996–2003 by using satellite observational data. Atmospheric Environment 40, 7663–7667. Alfaro, S.C., Gomes, L., Rajor, J.L., Lafon, S., Gaudichet, A., Chatenet, B., Maille, M., Gautenet, G., Lasserre, F., Cachier, H., Zhang, X.Y., 2003. Chemical and optical characterization of aerosols measured in spring 2002 at the ACE-Asia supersite, Zhenbeitai, China. Journal of Geophysical Research 108 (D23), 8641. Anderson, T.L., Ogren, J.A., 1998. Determining aerosol radiative properties using the TSI 3563 integrating nephelometer. Aerosol Science and Technology 29, 57–69. Anderson, T.L., Masonis, S.J., Covert, D.S., Ahlquist, N.C., Howell, S.G., Clarke, A.D., McNaughton, C.S., 2003. Variability of aerosol optical properties derived from in situ aircraft measurements during ACE-Asia. Journal of Geophysical Research 108 (D23), 8647. Andreae, T.W., Andreae, M.O., Ichoku, C., Maenhaut, W., Cafmeyer, J., Karnieli, A., Orlovsky, L., 2002. Light scattering by dust and anthropogenic aerosol at a remote site in the Negev Desert, Israel. Journal of Geophysical Research 107 (D2), 4008. Bergin, M.H., Cass, G.R., Xu, J., Fang, C., Zeng, L.M., Yu, T., Salmon, L.G., Kiang, C.S., Tang, X.Y., Zhang, Y.H., Chameides, W.L., 2001. Aerosol radiative, physical, and chemical properties in Beijing during June 1999. Journal of Geophysical Research 106 (D16), 17969–17980. Chang, S.Y., Fang, G.C., Chou, C., Chen, W.N., 2006. Chemical compositions and radiative properties of dust and anthropogenic air masses study in Taipei Basin, Taiwan, during spring of 2004. Atmospheric Environment 40, 7796–7809. Clarke, A.D., et al., 2004. Size distribution and mixture of dust and black carbon aerosol in Asian outflow: physiochemistry and optical properties. Journal of Geophysical Research 109, D15S09. IPCC (The Intergovernmental Panel on Climate Change), 2001. Climate Change 2001: The Scientific Basis. Cambridge University Press, Cambridge, p. 896. Kim, K.W., Kim, Y.J., Oh, S.J., 2001. Visibility impairment during Yellow Sand periods in the urban atmosphere of Kwangju, Korea. Atmospheric Environment 35, 5157–5167. Kim, K.W., He, Z., Kim, Y.J., 2004. Physicochemical characteristics and radiative properties of Asian dust particles observed at Kwangju, Korea, during the 2001 ACE-Asia intensive
ARTICLE IN PRESS 6922
H. Yan / Atmospheric Environment 41 (2007) 6916–6922
observation period. Journal of Geophysical Research 109, D19S02. Kim, S.W., Yoon, S.C., Jefferson, A., Ogren, J.A., Dutton, E.G., Won, J.G., Ghim, Y.S., Lee, B.I., Han, J.S., 2005. Aerosol optical, chemical and physical properties at Gosan, Korea during Asian dust and pollution episodes in 2001. Atmospheric Environment 39, 39–50. Kus, P., Carrico, C.M., Rood, M.J., Williams, A., 2004. Measured and modeled light scattering values for dry and hydrated laboratory aerosols. Journal of Atmospheric and Oceanic Technology 21, 981–994. Salmon, L.G., Christoforou, C.S., Cass, G.R., 1994. Air pollutants in the buddhist cave temples at the Yungang Grottoes, China. Environmental Science and Technology 28, 805–811. Sheridan, P.J., Delene, D.J., Ogren, J.A., 2001. Four years of continuous surface aerosol measurements from the Department of Energy’s Atmospheric Radiation Measurement Program Southern Great Plains Cloud and Radiation Testbed
site. Journal of Geophysical Research 106 (D18), 20735–20748. Takemura, T., Nakajima, T., Nozawa, T., Aoki, K., 2001. Simulation of future aerosol distribution, radiative forcing, and long-range transportation in East Asia. Journal of the Meteorological Society of Japan 79, 1139–1155. Vrekoussis, M., Liakakou, E., Koc- ak, M., Kubilay, N., Oikonomou, K., Sciare, J., Mihalopoulos, N., 2005. Seasonal variability of optical properties of aerosols in the Eastern Mediterranean. Atmospheric Environment 39, 7083–7094. Waldman, J.M., Lioy, P.J., Zelenka, M., Jing, L., Lin, Y.N., He, Q.C., Qian, Z.M., Chapman, R., Wilson, W.E., 1991. Winterime measurements of aerosol acidity and trace elements in Wuhan, a city in central China. Atmospheric Environment 25B, 113–120. Xu, J., Bergin, M.H., Yu, X., Liu, G., Zhao, J., Carrico, C.M., Baumann, K., 2002. Measurement of aerosol chemical, physical and radiative properties in the Yangtze delta region of China. Atmospheric Environment 36, 161–173.
Atmospheric Environment 41 (2007) 6916–6922 www.elsevier.com/locate/atmosenv
Short communication
Aerosol scattering properties in northern China Hao Yan Division of Remote Sensing Applications, National Meteorological Center, China Meteorological Administration, Zhongguancun South Street 46#, Haidian District, Beijing 100081, China Received 1 February 2007; received in revised form 25 April 2007; accepted 25 April 2007
Abstract The aerosol scattering properties were investigated at two continental sites in northern China in 2004. Aerosol light scattering coefficient (ssp) at 525 nm, PM10, and aerosol mass scattering efficiencies (a) at Dunhuang had a mean value of 165.17148.8 M m1, 157.67270.0 mg m3, and 2.3073.41 m2 g1, respectively, while these values at Dongsheng were, respectively, 180.27151.9 M m1, 119.07112.9 mg m3, and 1.8771.41 m2 g1. There existed a seasonal variability of aerosol scattering properties. In spring, at Dunhuang PM10, ssp, and a were 184.17211.548 mg m3, 126.3789.6 M m1, and 1.0570.97 m2 g1, respectively, and these values at Dongsheng were 146.47142.1 mg m3, 183.4781.7 M m1, and 1.9871.52 m2 g1, respectively. However, in winter at Dunhuang PM10, ssp, and a were 158.17261.4 mg m3, 303.37 165.2 M m1, and 3.1771.93 m2 g1, respectively, and these values at Dongsheng were 155.77170.1 mg m3, 304.47 158.1 M m1, and 2.9071.72 m2 g1, respectively. ssp and a in winter were higher than that in spring at both the sites, which coincides with the characteristics of dust aerosol and pollution aerosol. Overall, the dominant aerosol types in spring and winter at both sites in northern China are dust aerosol and pollution aerosol, respectively. r 2007 Elsevier Ltd. All rights reserved. Keywords: Dust and pollution aerosol; Integrating nephelometer; Scattering coefficient; PM10
1. Introduction The effect of aerosols on the climate has drawn much attention in recent years due to their largely uncertain climate forcing (The Intergovernmental Panel on Climate Change (IPCC), 2001). Through scattering and absorption of solar radiation, aerosols directly affect visibility and climate through the modification of the Earth’s energy balance. Thus, the aerosol scattering properties are required in evaluating the radiative forcing of aerosols (Vrekoussis et al., 2005; Kim et al., 2005; Xu et al., 2002). Tel.: +86 10 68407470; fax: +86 10 62172982.
E-mail address: [email protected] 1352-2310/$ - see front matter r 2007 Elsevier Ltd. All rights reserved. doi:10.1016/j.atmosenv.2007.04.052
However, the distribution of aerosols is quite patchy and it depends on the presence of aerosol sources (Vrekoussis et al., 2005). The regional climate effects of these aerosols are predicted to increase in the near future (Takemura et al., 2001). More specifically, in Asia, in the context of global warming, the desert and desertification may become much drier than ever before and emit a larger amount of dust aerosol from dust storms. In addition, China is the world’s fastest growing economy, which may lead to an increase in pollution aerosols from industrial sources. Asian aerosols have drawn much attention during the last 10 years. Some observations of Asian aerosols have been carried out during recent years. The Asian Pacific
ARTICLE IN PRESS H. Yan / Atmospheric Environment 41 (2007) 6916–6922
6917
Regional Aerosol Characterization Experiment (ACEAsia) shows that the dust from East Asia mixed with pollutants, resulting in a multi-component aerosol of great complexity and variability (Clarke et al., 2004), and dust aerosols had complex chemical and optical properties (Anderson et al., 2003). However, pollution aerosol was poorly investigated in ACE-Asia. Some measurement conducted in China mainly focused on chemical, physical, and radiative properties of pollution aerosols in urban regions, such as Beijing (Bergin et al., 2001), Wuhan (Waldman et al., 1991), Datong (Salmon et al., 1994), and Yangtze delta region (Xu et al., 2002) for a short period. Besides, the aerosol characters and mass scattering efficiency of different sources transported southeastward from Asian continent to Taiwan were investigated in Taiwan during East Asian winter monsoon (Chang et al., 2006). Similarly, Kim et al. (2005) analyzed the aerosol optical properties for two cases, viz. a heavy dust episode in April and a regionalscale pollution event in November in Korea. It can be concluded that there exist two types of aerosols, namely dust aerosol and pollution aerosol. However, the measurement in the above research was conducted far from their sources or just for a few cases. This study focuses on the seasonal variability of dominant aerosol type on the basis of the continuous measurement of aerosol light scattering coefficient and mass concentration for a whole year of 2004 at two sites in northern China.
(94.61E, 40.11N, 1139 m), in Gansu Province, which is near the Takelamagan Desert. The second one is located in Dongsheng city (109.91E, 39.81N, 1460 m) in inner-Mongolian Province which is near the Kubuqi Desert. They belong to arid and semiarid regions in northern China characterized by dry conditions. Spring refers to March, April, and May and winter refers to January, February, and December. Aerosol light scattering coefficient (ssp) was measured at these locations with a standard integrating nephelometer (Ecotech-M9003). This instrument measures ssp at 525 nm at 5 min intervals. The optical and electrical background noise is sufficiently low to allow measures of ssp (for particles) from o10% of air Rayleigh (sspo0.3 M m1) to 42000 M m1. PM10 was measured at these locations with an ambient particulate monitor (Rupperecht and Patashnick (R&P) model TEOM Series 1400a), continuously. This instrument equipped with an inlet having a 10 mm cut-point measures the ambient particulate mass concentration of PM10 in real time at 5 min intervals. All PM10 and ssp data in the following were averaged from the smallest 5 min intervals.
2. Experiment
The scatter plot of daily mean PM10 and ssp for Dunhuang and Dongsheng observation stations are presented. At Dunhuang, the PM10 daily mean values ranged between 4.2 and 1795.6 mg m3 with a
Sampling was conducted at two remote sites in 2004. The first one is located in Dunhuang city
3. Results and discussion 3.1. Comparison of PM10 vs ssp
Fig. 1. Scatter plots of daily mean values of PM10 and aerosol light scattering coefficient (ssp) at Dunhuang (a) and Dongsheng (b) in 2004.
ARTICLE IN PRESS 6918
H. Yan / Atmospheric Environment 41 (2007) 6916–6922
mean value of 157.67270.0 mg m3, and the ssp daily mean values ranged between 22.2 and 880.5 M m1 with a mean value of 165.17148.8 M m1 (Fig. 1(a)). At Dongsheng, the PM10 daily mean values ranged between 8.0 and 886.6 mg m3 with a mean value of 119.07112.9 mg m3, and the ssp daily mean values ranged between 19.2 and 1511.7 M m1 with a mean value of 180.27 151.9 M m1 (Fig. 1(b)). In order to show the seasonal variability of ssp clearly, daily scatter plots of 4 seasons with a regression equation at Dunhuang are provided, respectively (Fig. 2). At Dunhuang, several heavy dust events occurred in spring and summer with higher values of PM10 (1600 mg m3) and ssp (700 M m1) (Figs. 2(a and b)). Similarly, at Dongsheng, several heavy dust events occurred in
spring and winter with high values of PM10 (500 mg m3) and ssp (300 M m1) (Fig. 1). In all, high values of PM10 and ssp corresponded with dust aerosol checked with surface observation. There are some methods for discriminating between dust aerosol and pollution aerosol. For example, one can analyze the ratio of PM2.5 to PM10, the column A˚ngstro¨m exponent, chemical compositions, and air mass trajectories (Xu et al., 2002; Chang et al., 2006; Kim et al., 2005). However, the dust particles and pollutants are also likely to be mixed with each other even during dust episode or pollution episode (Kim et al., 2005). Thus, heavy dust events were just checked with surface meteorological observation in this paper, and heavy aerosol with values of PM10 (200 mg m3) and ssp (400 M m1) in winter and fall
Fig. 2. Scatter plots of daily mean values of PM10 and aerosol light scattering coefficient (ssp) at Dunhuang in spring (a), summer (b), fall (c), and winter (d) for 2004.
ARTICLE IN PRESS H. Yan / Atmospheric Environment 41 (2007) 6916–6922
was expected to be pollution aerosol due to an accepted fact that pollution aerosol in cold seasons prevails in urban area of northern China. This point was further verified with aerosol mass scattering efficiency data, because dust and pollution aerosols have different mass scattering efficiencies (Xu et al., 2002; Chang et al., 2006; Kim et al., 2005; Vrekoussis et al., 2005). In fall and winter, low values of PM10 (200 mg m3) and higher values of ssp (400 M m1) (Figs. 2(c and d)) were found at Dunhuang. Similarly, low values of PM10 (200 mg m3) and high ssp (700 M m1) (Fig. 1) were also found at Dongsheng in winter. In all, it is our inference that higher values of ssp and lower values of PM10 correspond with pollution aerosol. To our knowledge, few researches report such scatter plots of two typical aerosols with heavy mass concentration at one location. Similarly, an observation of anthropogenic pollution with PM10 ranging from 57 to 77 mg m3 and dust aerosol with PM10 ranging from 103 to 132 mg m3 measured at Taipei basin was reported (Chang et al., 2006). Kim et al. (2005) stated another measurement of pollution with values for PM10 and ssp of 56.9 mg m3 and 244.2 M m1 and dust with values for PM10 and ssp of 107.9 mg m3 and 239.8 M m1 during Asian dust and pollution periods at Gosan, Korea. As these aerosols were transported far from Asian continent, PM10 and ssp are much lower than the values measured in this paper. 3.2. PM10 The monthly and seasonal variability of PM10 for Dunhuang and Dongsheng observation stations is presented in Fig. 3. At Dunhuang, higher values of PM10 were found all the year except September and October (Fig. 3(a)). Higher values of PM10 were always related with dust storm. A lot of dust storm occurred in spring, winter, and even in summer at Dunhuang. At Dongsheng, higher values of PM10 were found in February and March (Fig. 3(b)), which was caused by dust storm. However, a lower value of annual PM10 could be found in Dongsheng compared with Dunhuang. PM10 presents a distinct seasonal pattern (Fig. 3(c)). Higher values were found in spring and winter, and lower values were encountered in fall at both stations. However, PM10 also shows higher values in summer at Dunhuang due to dust events.
6919
3.3. Aerosol light scattering coefficient (ssp) At Dunhuang, higher values of ssp were found in November, December, and even in July, while lower values were found in August and September (Fig. 3(d)). However, ssp at Dongsheng shows a distinct seasonal pattern (Fig. 3(e)). Maximum values were discovered in January and December, and minimum values appeared in September and October. ssp shows a different seasonal pattern at both stations compared with PM10. Maximum values of ssp were often found in winter rather than in spring (Fig. 3(f)). Although dust storm resulted in higher PM10 and higher ssp, dust aerosol could not explain this discrepancy in seasonal patterns for PM10 and ssp. The discrepancy was largely due to the pollution aerosol dominant in winter with higher values of ssp. 3.4. Mass scattering efficiencies (a) The ratio of ssp to PM10 mass concentration is the aerosol mass scattering efficiency (a). The a at both the sites in this paper presents a distinct seasonal pattern. At Dunhuang, maximum values were found in December, whereas minimum values from March to June (Fig. 3(g)). At Dongsheng, maximum values were also found in December and January, whereas minimum values in September and October (Fig. 3(h)). Obviously, the dust events could affect the PM10 and ssp at both the sites (Fig. 1). So in order to calculate the mass scattering efficiency for dust (adust), only significant dust observations (above 300 mg m3) with high ssp levels above 100 M m1 were analyzed in the following text. The regression between PM10 and ssp indicates that there exists a significant correlation with a slope of 0.43 (r ¼ 0.9, samples ¼ 23) at Dunhuang and a slope of 0.29 (r ¼ 0.75, samples ¼ 13) at Dongsheng. Further, the regression between all PM10 and ssp in spring also gives a slope of 0.34 (r ¼ 0.79, samples ¼ 52) at Dunhuang (Fig. 3(g). In addition, the adust for Dunhuang and Dongsheng coincide with the value of 0.52 m2 g1 measured at Negev Desert in Israel (Andreae et al., 2002) and that of 0.21–0.96 m2 g1 measured in the Eastern Mediterranean (Vrekoussis et al., 2005) for dust aerosol. Similarly, only significant pollution observations (below 300 mg m3) with high ssp levels above 200 M m1 were selected for calculation of the mass
ARTICLE IN PRESS 6920
H. Yan / Atmospheric Environment 41 (2007) 6916–6922
Fig. 3. Monthly and seasonal mean values and standard deviations of PM10 (upper panels), ssp (middle panels), and a (bottom panels) at Dunhuang and Dongsheng in 2004.
scattering efficiency for pollution aerosol (apollutant) in this paper. The regression between PM10 and ssp shows that there exists a correlation with a slope of 1.75 (r ¼ 0.54, samples ¼ 21) at Dunhuang and a slope of 2.62 (r ¼ 0.52, samples ¼ 43) at Dongsheng. Further, the regression between PM10 and ssp in winter gives a slope of 1.05 (r ¼ 0.31, samples ¼ 29) at Dunhuang (Fig. 2(d)) regardless of the two dust events (PM104400 mg m3) checked with surface observation. In addition, the apollutant for Dunhuang and Dongsheng coincide with the value of 1.4–2.3 m2 g1 measured at Taipei basin (Chang et al., 2006) and that of 1.65–2.0 m2 g1 measured at Yulin in China (Alfaro et al., 2003) for pollution aerosol.
Although higher values of adust and apollutant were also found in other measurements compared with measurements in this paper. adust had higher values of 0.89–1.15 m2 g1 (Alfaro et al., 2003), 2.2 m2 g1 (Kim et al., 2005), and 3.3–4.4 m2 g1 (Kim et al., 2004). apollutant had higher values of 4.13 m2 g1 (Kim et al., 2001), 4.5 m2 g1 (Kim et al., 2005), and 6.7 m2 g1 (Kim et al., 2004). The higher mass scattering efficiencies measured in Korea (Kim et al., 2004, 2005) is likely due to truncation corrections in the nephelometer and secondary growth of the aerosol. The dust and pollution aerosol measured in Korea passed over the highly polluted eastern China and was far from the source region and had time to accumulate more sulfate and nitrate
ARTICLE IN PRESS H. Yan / Atmospheric Environment 41 (2007) 6916–6922
than aerosol close to the source region. The truncation correction of near-forward scattered light for TSI nephelometer measurements has been described by Anderson and Ogren (1998). This sizedependent truncation correction method has been widely applied based on the A¨ngstro¨m exponent (Anderson and Ogren, 1998; Sheridan et al., 2001; Anderson et al., 2003; Kim et al., 2005). However, this correction is large, especially for larger dust particles, quite uncertain, and related with instrument and aerosol size distribution (Anderson et al., 2003; Kus et al., 2004), so for the present study, we make no truncation correction. The values of adust and apollutant measured at the same site through the same measure method were compared, and the comparison shows that adust is always lower than apollutant even at different sites. Kim et al. (2005) reported values for adust and apollutant, namely 2.2 and 4.5 m2 g1, respectively, in Korea. Chang et al. (2006) presented values for adust and apollutant, viz. 0.7–1.0 and 1.3–1.6 m2 g1 in Taiwan, respectively. The values for adust and apollutant (0.29–0.43 and 1.75–2.62 m2 g1, respectively) were found in northern China. This is largely attributed to the large particle size, non-spherical shape, and lower refractive index of the dust aerosol compared with the pollution aerosol. Obviously, a was bigger in winter than that in spring at both sites (Fig. 3(i)), which suggests that the dominant aerosol in spring is dust aerosol and the dominant aerosol in winter is pollution aerosol. During the cold seasons, especially in winter, pollution aerosol often becomes the dominant type of aerosol due to coal burning (Akimoto et al., 2006), weak surface winds, and stable atmospheric conditions. The discrepancy in PM10 and ssp seasonal patterns (Figs. 3(c and f)) are largely due to the difference of dominant aerosol types in different seasons. 4. Conclusion Aerosol scattering efficiency, PM10, and mass scattering efficiency measured at two continental sites in northern China in 2004 were analyzed in this paper. In spring, at Dunhuang PM10, ssp, and a were 184.17 211.548 mg m3, 126.3789.6 M m1, and 1.0570.97 m2 g1, respectively, and at Dongsheng, these values were 146.47142.1 mg m3, 183.4781.7 M m1, and 1.9871.52 m2 g1, respectively. However, in winter at Dunhuang PM10, ssp, and a were 158.17261.4 mg m3, 303.37165.2 M m1, and 3.1771.93 m2 g1,
6921
respectively, and at Dongsheng these values were 155.77170.1 mg m3, 304.47158.1 M m1, and 2.907 1.72 m2 g1, respectively. ssp and a in winter had higher values than that in spring at both the sites, which coincide with the characteristics of dust aerosol and pollution aerosol discovered by other researches (Kim et al., 2005; Chang et al., 2006; Alfaro et al., 2003). Thus, it can be inferred that the dominant aerosol type in spring and winter at both sites is dust aerosol and pollution aerosol, respectively. References Akimoto, H., Ohara, T., Kurokawa, J.I., Horii, N., 2006. Verification of energy consumption in China during 1996–2003 by using satellite observational data. Atmospheric Environment 40, 7663–7667. Alfaro, S.C., Gomes, L., Rajor, J.L., Lafon, S., Gaudichet, A., Chatenet, B., Maille, M., Gautenet, G., Lasserre, F., Cachier, H., Zhang, X.Y., 2003. Chemical and optical characterization of aerosols measured in spring 2002 at the ACE-Asia supersite, Zhenbeitai, China. Journal of Geophysical Research 108 (D23), 8641. Anderson, T.L., Ogren, J.A., 1998. Determining aerosol radiative properties using the TSI 3563 integrating nephelometer. Aerosol Science and Technology 29, 57–69. Anderson, T.L., Masonis, S.J., Covert, D.S., Ahlquist, N.C., Howell, S.G., Clarke, A.D., McNaughton, C.S., 2003. Variability of aerosol optical properties derived from in situ aircraft measurements during ACE-Asia. Journal of Geophysical Research 108 (D23), 8647. Andreae, T.W., Andreae, M.O., Ichoku, C., Maenhaut, W., Cafmeyer, J., Karnieli, A., Orlovsky, L., 2002. Light scattering by dust and anthropogenic aerosol at a remote site in the Negev Desert, Israel. Journal of Geophysical Research 107 (D2), 4008. Bergin, M.H., Cass, G.R., Xu, J., Fang, C., Zeng, L.M., Yu, T., Salmon, L.G., Kiang, C.S., Tang, X.Y., Zhang, Y.H., Chameides, W.L., 2001. Aerosol radiative, physical, and chemical properties in Beijing during June 1999. Journal of Geophysical Research 106 (D16), 17969–17980. Chang, S.Y., Fang, G.C., Chou, C., Chen, W.N., 2006. Chemical compositions and radiative properties of dust and anthropogenic air masses study in Taipei Basin, Taiwan, during spring of 2004. Atmospheric Environment 40, 7796–7809. Clarke, A.D., et al., 2004. Size distribution and mixture of dust and black carbon aerosol in Asian outflow: physiochemistry and optical properties. Journal of Geophysical Research 109, D15S09. IPCC (The Intergovernmental Panel on Climate Change), 2001. Climate Change 2001: The Scientific Basis. Cambridge University Press, Cambridge, p. 896. Kim, K.W., Kim, Y.J., Oh, S.J., 2001. Visibility impairment during Yellow Sand periods in the urban atmosphere of Kwangju, Korea. Atmospheric Environment 35, 5157–5167. Kim, K.W., He, Z., Kim, Y.J., 2004. Physicochemical characteristics and radiative properties of Asian dust particles observed at Kwangju, Korea, during the 2001 ACE-Asia intensive
ARTICLE IN PRESS 6922
H. Yan / Atmospheric Environment 41 (2007) 6916–6922
observation period. Journal of Geophysical Research 109, D19S02. Kim, S.W., Yoon, S.C., Jefferson, A., Ogren, J.A., Dutton, E.G., Won, J.G., Ghim, Y.S., Lee, B.I., Han, J.S., 2005. Aerosol optical, chemical and physical properties at Gosan, Korea during Asian dust and pollution episodes in 2001. Atmospheric Environment 39, 39–50. Kus, P., Carrico, C.M., Rood, M.J., Williams, A., 2004. Measured and modeled light scattering values for dry and hydrated laboratory aerosols. Journal of Atmospheric and Oceanic Technology 21, 981–994. Salmon, L.G., Christoforou, C.S., Cass, G.R., 1994. Air pollutants in the buddhist cave temples at the Yungang Grottoes, China. Environmental Science and Technology 28, 805–811. Sheridan, P.J., Delene, D.J., Ogren, J.A., 2001. Four years of continuous surface aerosol measurements from the Department of Energy’s Atmospheric Radiation Measurement Program Southern Great Plains Cloud and Radiation Testbed
site. Journal of Geophysical Research 106 (D18), 20735–20748. Takemura, T., Nakajima, T., Nozawa, T., Aoki, K., 2001. Simulation of future aerosol distribution, radiative forcing, and long-range transportation in East Asia. Journal of the Meteorological Society of Japan 79, 1139–1155. Vrekoussis, M., Liakakou, E., Koc- ak, M., Kubilay, N., Oikonomou, K., Sciare, J., Mihalopoulos, N., 2005. Seasonal variability of optical properties of aerosols in the Eastern Mediterranean. Atmospheric Environment 39, 7083–7094. Waldman, J.M., Lioy, P.J., Zelenka, M., Jing, L., Lin, Y.N., He, Q.C., Qian, Z.M., Chapman, R., Wilson, W.E., 1991. Winterime measurements of aerosol acidity and trace elements in Wuhan, a city in central China. Atmospheric Environment 25B, 113–120. Xu, J., Bergin, M.H., Yu, X., Liu, G., Zhao, J., Carrico, C.M., Baumann, K., 2002. Measurement of aerosol chemical, physical and radiative properties in the Yangtze delta region of China. Atmospheric Environment 36, 161–173.