SCIENCE CHINA Earth Sciences • RESEARCH PAPER •
February 2017 Vol.60 No.2: 297–314 doi: 10.1007/s11430-016-0104-3
Aerosol properties over an urban site in central East China derived from ground sun-photometer measurements LIU Qi1*, DING WeiDong1,2, XIE Lei1, ZHANG JinQiang3, ZHU Jun3, XIA XiangAo3,4, LIU DongYang1, YUAN RenMin1 & FU YunFei1 1
3
School of Earth and Space Sciences, University of Science and Technology of China, Hefei 230026, China; 2 Anhui Meteorological Observatory, Hefei 230001, China; Key Laboratory of Middle Atmosphere and Global Environment Observation, Institute of Atmospheric Physics, Chinese Academy of Sciences, Beijing 100029, China; 4 Collaborative Innovation Center on Forecast and Evaluation of Meteorological Disasters, Nanjing University of Information Science & Technology, Nanjing 210044, China Received August 24, 2016; accepted October 11, 2016; published online December 7, 2016
Abstract Sun-photometer measurements at Hefei, an urban site located in central East China, were examined to investigate the variations of aerosol loading and optical properties. It is found that aerosol optical thickness (AOT) keeps higher in winter/spring and gets relatively lower in summer/autumn. The large AOT in winter is caused by anthropogenic sulfate/nitrate aerosols, while in spring dust particles elevate the background aerosol loading and the excessive fine-mode particles eventually lead to severe pollution. There is a dramatic decline of AOT during summer, with monthly averaged AOT reaching the maximum in June and soon the minimum in August. Meanwhile, aerosol size decreases consistently and single scattering albedo (SSA) reaches its minimum in July. During summertime large-sized particles play a key role to change the air from clean to mild-pollution situation, while the presence of massive small-sized particles makes the air being even more polluted. These complicated summer patterns are possibly related to the three key processes that are active in the high temperature/humidity environment concentrating on sulfate/nitrate aerosols, i.e., gas-to-particle transformation, hygroscopic growth, and wet scavenging. Regardless of season, the increase of SSA with increasing AOT occurs across the visible and near-infrared bands, suggesting the dominant negative/cooling effect with the elevated aerosol loading. The SSA spectra under varying AOT monotonically decrease with wavelength. The relatively large slope arises in summer, reinforcing the dominance of sulfate/nitrate aerosols that induce severe pollution in summer season around this city. Keywords Aerosol optical thickness, Single scattering albedo, Central East China, Sun-photometer Citation:
Liu Q, Ding W D, Xie L, Zhang J Q, Zhu J, Xia X A, Liu D Y, Yuan R M, Fu Y F. 2017. Aerosol properties over an urban site in central East China derived from ground sun-photometer measurements. Science China Earth Sciences, 60: 297–314, doi: 10.1007/s11430-016-0104-3
Introduction 1. Atmospheric aerosol, a major component in the earth-atmosphere system, is the primary contributor to atmospheric pollution and has potential impacts on human health. In a meteorological perspective, atmospheric aerosols affect the *Corresponding author (email:
[email protected]) © Science China Press and Springer-Verlag Berlin Heidelberg 2016
radiation budget of the earth-atmosphere system through a couple of ways, constituting the well-known aerosol direct, semi-direct, and indirect radiative effects (Twomey, 1977; Albrecht, 1989; Charlson et al., 1992; Hansen et al., 1997). The highly variable aerosols and the associated multiple radiative effects act as a potential internal forcing upon climate change (Rosenfeld, 2006). Unfortunately, so far it is one of the most important uncertainties in climate prediction (Leibensperger et al., 2012; Stocker et al., 2013). The obstaearth.scichina.com link.springer.com
298
Sci China Earth Sci February (2017) Vol. 60 No. 2
Liu Q, et al.
cle stems mainly from our limited understanding of complex interactions among aerosol, cloud/precipitation, radiation and dynamics. A great lot of endeavors are still needed to circumvent these challenges. But for the aerosol direct radiative effect (ADRE) that is not related to hydrological processes of cloud and precipitation, a promising perspective gradually emerges (Remer and Kaufman, 2006; Li et al., 2010). By virtue of many advanced measurement platforms that monitor the key optical and microphysical properties of aerosols, including in-situ samplings, ground-based and satellite remote sensing, many observation-based studies have been conducted to quantitatively explore the ADRE on a regional even global scale (Nakajima et al., 1996; Holben et al., 1998; Li et al., 2011). According to the basic radiative transfer equations that involve aerosol components, three most relevant optical properties are aerosol optical thickness (AOT), single scattering albedo (SSA), and scattering phase function (or asymmetry factor, AF). Although these bulk optical properties of aerosol volumes are essentially determined by their chemical compositions, concentration and microphysical characteristics such as phase, size, shape, etc., these three parameters, are sufficient to describe the aerosol-modulated radiative transfer processes. The above aerosol properties are therefore the main targets for most aerosol retrieval algorithms that are based on radiation measurements. In most cases, the derived AOT and SSA are used to describe the aerosol characteristics for a specific region. For general applications, the AOT is usually used to quantify the overall concentration of aerosols, while the SSA reflecting the radiative absorption capability, is used as a proxy to identify the dominant aerosol type. Commonly, Angstrom exponent (AE) instead of asymmetry factor is used as a proxy to denote the size feature of concerned aerosol particles. These aerosol properties could be employed to characterize the aerosol environment and in turn deduce the ADRE for a certain region (Eck et al., 2005; Xia et al., 2006; Liu et al., 2007; Xia et al., 2007; Wang et al., 2009; Che et al., 2014). As manifested in high-resolution satellite images and reported in many modeling studies, East China is the evident area that has rather heavy aerosol loadings. In contrast to other regions across the globe where aerosol loadings are temporally high, the excessive aerosols over East China are nearly permanent and attributed to both natural and anthropogenic sources (Nakajima et al., 2007; Xia et al., 2007; Pan et al., 2010; Han et al., 2012; Zhang et al., 2012). The massive local emissions from agriculture, industrial and traffic activities, which are often enhanced by favorable meteorological conditions, together with frequent long-range transport of dust particles from northern and western China, contribute much to the high aerosol loading therein. As a rapid developing region among the world and possessing a considerably large density of population, much attention has been
paid to the aerosol characteristics and their spatial and temporal variations over East China. This region has also served as a unique natural laboratory for exploring that to what extent anthropogenic aerosols may impose impacts on regional climate (Li et al., 2007; Han et al., 2012; Zhang et al., 2012) and, meteorological impacts on the aerosol distributions as it is located in the large monsoon domain (Zhang et al., 2010). Many observational studies have investigated the aerosol variations over East China based on satellite remote sensing such as Moderate Resolution Imaging Spectroradiometer (MODIS), Multi-angle Imaging Spectro Radiometer (MISR) and Ozone Monitoring Instrument (OMI), as well as measurements at ground remote sensing network such as Aerosol Robotic Network (AERONET) (Holben et al., 1998), Sky Radiometer Network (SKYNET) (Nakajima et al., 2003), Chinese Sun Hazemeter Network (CSHNET) (Xin et al., 2007), China Aerosol Remote Sensing Network (CARSNET) (Che et al., 2009). For the ground measurement-based studies concerning East China, due to uneven distribution of the measurement sites, they are mostly concentrated on Yangtze River Delta regions around the megalopolis, Shanghai, one of the most developed zones in the country (Bergin et al., 2001; Xu et al., 2002; Li et al., 2003; Xia et al., 2007; Yu et al., 2011; Chen et al., 2012; He et al., 2012; Liu et al., 2012). Since these sites are more or less close to the coastal regions of East China Sea, the aerosols in these locations are very likely interfered by marine environment, e.g., almost persistent contribution from sea salt aerosols. According to back trajectory analysis, Pan et al. (2010) revealed that more than 50% of air masses that modify the aerosols over Shanghai are from eastern marine area, while those faraway from coastlines get almost equivalent effects of diverse aerosol sources. Given the fact that a great part of areas in East China are far away from the economic core of Shanghai and suffer less from oceanic aerosols, dissimilar aerosol characteristics should be expected in these inland regions in East China. Hefei (117.28°E, 31.87°N) is the capital of Anhui province, located at the central location in East China as shown in Figure 1, west to Shanghai (121.48°E, 31.23°N) with a distance about four hundred kilometers. The central town of Hefei has an area of about 600 square kilometers and a population near 5.0 million. The majority surroundings of Hefei are cropland, woodland, and water body, which are very common for ordinary cities in this region. Most indices including the area, population, circumjacent environment, industrial enterprise amount, economic development, etc., indicate Hefei as a representative moderate city in East China. The aerosol variations derived from such an area should be meaningful and complementary to those obtained from Yangtze River Delta, supporting a comprehensive depiction of aerosol characteristics over the entire region of East China. According to the MODIS measurements displayed in Figure 1, the annual mean AOT at Hefei is around 0.7, comparable with most re-
Sci China Earth Sci February (2017) Vol. 60 No. 2
Liu Q, et al.
299
The Figure 1 geographic location of Hefei within the East Asia (a) and the two sites therein (b). The circle near the center of (a) indicates the location of Hefei, associated with background color indicating the annual mean AOT across the East Asia in 2012. In the (b) of Hefei map, the yellow pentacle indicates the location of USTC site used in this study, while the orange pentacle indicates KXD site. The distance between these two sites exceeds 10 km and their surrounding environment is absolutely different.
gions in East China except for those near Shanghai. By using more than one year aerosol data from a sun photometer (CIMEL CE318), Wang et al. (2012) documented the optical characteristics of aerosols in Hefei. A mild seasonality of AOT was found with a relatively large value in spring and comparable values in other seasons. Besides the combination of dust in spring, contribution from fine-mode aerosols consisting of urban pollution and smoke keeps nearly invariant during the whole year. More recently, Wang et al. (2014) conducted a more comprehensive analysis about aerosols in Hefei based on multi-year data from a sky radiometer (PREDE POM02) and gave a bit different results. Though dust aerosols were also reported to be dominant in spring, a summer maximum of AOT was found and attributed to accumulation of pollution aerosols under special weather conditions in this specific region during the warm season. Such a result is more consistent with relevant studies that examined measurements from other sites in East China (Pan
et al., 2010; Yu et al., 2011; He et al., 2012). In addition, Fan et al. (2010) conducted in situ measurements of aerosol properties using facilities under aegis of Atmospheric Radiation Measurement (ARM) program at Shouxian in Anhui province, a rural site approximately 100 km northwest to Hefei, and distinct aerosol variations were reported. In order to improve our understanding of aerosol properties at Hefei, a CE318 sun photometer was deployed in Hefei in December 2011. Compared to the instruments used in Wang et al. (2012) and Wang et al. (2014) that are both located at Kexuedao (KXD), a suburban area northwest to the urban center, close to a great reservoir and surrounded by massive vegetations, this newly-installed sun photometer is situated in the campus of University of Science and Technology of China (USTC), which is definitely in the downtown Hefei as labeled in Figure 1. The USTC site is away from the previous two instruments at KXD site by about 13 km. The surroundings of this new site are roadways, commercial and residential build-
300
Sci China Earth Sci February (2017) Vol. 60 No. 2
Liu Q, et al.
ings. More representative characteristics for urban aerosols are thus expected from the observations at this central town site. The measurements have been routinely collected since then and the retrieved data provide an opportunity to further clarify the aerosol characteristics over Hefei, a representative moderate city in East China. Both the aerosol loading and aerosol type were considered, and three key aerosol parameters (i.e., AOT, AE, and SSA) were investigated in this study.
Data 2. and methodology CE318 2.1 instrument The CE318 sun photometer was installed on the roof of a building with 16 floors inside the USTC campus. Since the elevation of the instrument exceeds 50 meters, the routinely sun and sky scanning observations would not be disturbed even for relatively low elevation angle. The instrument has been well maintained and calibrated. In most time it is in automatic mode along the scheduled sequences and has retained normal operation. Besides data vacancy during severe weathers, mainly raining or snowing events, the approximately 15-minute record series are continuous. Artificial gaps of measurements occur periodically in the last months of each year due to routine calibration works. This kind of photometer measures radiance across the ultraviolet, visible and near-infrared bands. Direct solar irradiance is received at eight wavelengths (340, 380, 440, 500, 670, 870, 936, 1020 nm) while diffuse sky radiance is received at four wavelengths (440, 670, 870, 1020 nm). Ex-
cept for 936 nm that is a strong absorption channel caused by water vapor, the rest seven channels are well atmospheric window channels and are all used to conduct spectral AOT retrieval relying on direct solar radiation measurements. The sky radiance measurements are used to implement retrievals to derive SSA and particle size distribution according to the prescribed procedure. Two basic scanning geometry modes (solar almucantar and principal plane) are employed to obtain aureole and sky radiances at a large range of scattering angles relative to the solar direction. The field of view of both the sun and directional sky target is 1.2°, with collimator accuracy of about 0.1°. In this study, a two-year record from January 2012 to November 2013 was analyzed. This duration represents a nearly steady measuring period without any instrument malfunction. The intermit in November and December is due to routine calibration while the zero-record in August 2012 and May 2013 is probably due to bad weather. Although it does not cover all the 24 months, this collection of data includes considerable samples. As displayed in Figure 2, there are totally 941 valid records, covering 276 days. Combining the two years together, the sample volumes are basically comparable in each month except for November and December. Retrieval 2.2 algorithm The retrieval scheme is based on the AERONET standard aerosol algorithm, which was used to make a full set of aerosol parameter retrieval (Dubovik and King, 2000; Dubovik et al., 2006). As suggested in Dubovik et al. (2000),
Sample Figure 2 volumes (a) and the covering days (b) in each month during the two-year period.
Sci China Earth Sci February (2017) Vol. 60 No. 2
Liu Q, et al.
the uncertainty of retrieved AOT and SSA, is about 0.01 and 0.03, respectively. In addition, the uncertainty of SSA would increase when AOT decreases to below 0.5 or 0.2 depending on aerosol types. Two separated modules are contained in the algorithm to actualize the aerosol parameter retrieval. One is the forward module that simulates the measurable quantities taking illuminative, geometry, environmental and aerosol parameters as input to radiative transfer model. In this part, aerosol particles are deemed as either spheres or spheroid, ensuring that the scattering simulation for non-spherical dust particulate is also reasonable. In addition, plane parallel atmosphere is assumed in the radiative transfer calculations. The other is the inversion module that achieves the optimal estimation of aerosol parameters through rigid comparison between simulation results with the actual measurements. The kernel equations relating the aerosol parameters to radiative measurements are as follows. For the direct solar radiation measurements, the total optical thickness is directly related to the ratio of measured radiance (I ) to that at the top of the atmosphere (I 0 ), while AOT ( aer ) is just the residue from the total optical thickness ( ) by subtracting the gaseous extinction contribution ( gas). In the following equations, m 0 = 1 / cos 0 is air mass, and 0 is the solar zenith angle. I = I 0 exp( m 0 ),
=
aer
+
aer
,
gas
.
aer
,P
(1b)
aer
( )].
(2)
Since aer and P aer are both determined by the complex index of refraction and aerosol particle size parameter, by using measurements at multiple scattering angles ( i = 1, N ), several modes of aerosol size distribution in analytic form and the corresponding complex index of refraction could be retrieved via appropriate methods. Thus AOT and SSA at each wavelength would be deduced, together with aerosol size parameters. aer
( )=
.
AE greater than 2 indicates small particles that are mostly associated with combustion processes, while AE smaller than 1 indicates large particles that are most associated with mechanical processes, like sea salt and dust (Kaufman, 1993; Schuster et al., 2006). Actually AE varies with wavelength and its value depends on the selection of two or more reference wavelengths. In the present study, AE derived from linear regression of AOT at four wavelengths, 440, 500, 670, and 870 nm, is used and termed as AE-440/870 consistent with the nomination in Eck et al. (2005). AOT and SSA are two optical parameters depending strongly on wavelengths, requiring that a reference wavelength is employed for indicating their specific values. In this study, 550 nm, the central wavelength in the visible band, is used as the reference wavelength consistent with relevant works (Fan et al., 2010; Liu et al., 2011; He et al., 2012). The AOT at 550 nm is derived based on the above eq. (3) using the reference AOT at 500 nm and AE-440/870, while SSA at 550 nm is derived just via linear interpolation using the retrieved SSA at 440 and 670 nm. For typical aerosols, the dependence of SSA on wavelength within the visible band is mostly monotonic and linear (Remer et al., 2005; Li et al., 2015). It is therefore acceptable to utilize a linear approximation to derive SSA at 550 nm.
(1a)
For the diffuse sky radiation measurements, in particular those received at solar almucantar, they are correlated to aerosol SSA ( aer ) and scattering phase function (P aer ) with relatively simple mathematic expressions, given prescribed gaseous absorption, molecular scattering, and surface albedo. I( )=I[
301
(3)
As mentioned previously, AE is a parameter quantifying the spectral dependence of AOT as defined in the eq. (3) and marked as . This parameter is commonly used for qualitatively identifying the aerosol particle size (Angstrom, 1964; Schuster et al., 2006). For a given aerosol type (with definite complex index of refraction) the AOT decreases more quickly with wavelength when composed of small particles than large ones. A large AE thus indicates small particles being dominant in the aerosol sample and vice versa. Generally,
Results 3. Comparison 3.1 of aerosol properties between the two sites The AOT and SSA derived from the CIMEL sun photometer at USTC site were compared with those from PREDE sky radiometer at KXD site. The matched period for these two sets of measurements is the first half year in 2012 and the amount of matched pairs (temporal difference within 5 minutes) is 107. These two instruments give quite consistent AOT as shown in Figure 3, with bias falling within the uncertainty of AERONET AOT retrieval (0.01–0.02, Holben et al., 1998). The correlation coefficients are higher than 0.95 at all the four reference wavelengths. It is suggested that these two AOT data are both reliable. On the other hand, it is also indicated by such high consistency that for the sky-scale optical measuring regime of radiometer, specific differences in aerosol loading at locations away from each other by ~10 km cannot be resolved. Since AE is just a straightforward parameter derived from multi-wavelength AOT and reflects its spectral variation, it is apparent the AE retrievals should also be reasonably consistent between these two kinds of measurements. As shown in Figure 4, AE derived from CIMEL measurements agrees well with those from PREDE measurements as expected, especially for the one derived from the two wavelengths within the visible spectrum, with correlation coefficient exceeding 0.9 and bias lower than 0.05. The aerosol
302
Sci China Earth Sci February (2017) Vol. 60 No. 2
Liu Q, et al.
Comparison Figure 3 of AOT at four wavelengths between the PREDE/KXD and CIMEL/USTC measurements.
Comparison Figure 4 of AE500/670 (a) and AE870/1020 (b) between the PREDE/KXD and CIMEL/USTC measurements.
size information implied by the present AE data is thus also reliable. But for SSA as shown Figure 5, great differences emerge with very large biases and low correlations. PREDE gives too small dynamic range of SSA, which is mostly limited between 0.95 and 1.0. In contrast, SSA derived from CIMEL ranges mainly from 0.80 to 0.95, more consistent with those derived in previous studies (He et al., 2012; Li et al., 2015). Such a small range of SSA leads to an apparent overestimation by PREDE measurements and give a false implication of weakly-absorbing aerosols dominating this region. In
fact the overestimated SSA by PREDE was also reported in other studies (Che et al., 2008; Estellés et al., 2012). Both measuring strategy and the SKYRAD algorithm employed by PREDE may contribute to its overestimation of SSA, which may also affect other retrievals by PREDE, such as the size distribution. As suggested in these studies (Che et al., 2008; Estellés et al., 2012), the aerosol products derived from AERONET algorithm based on CIMEL photometer measurements provide more consistent estimation of aerosol multiple properties. Therefore the measurements at USTC site with fairly reliable information of AOT and SSA, as well
Sci China Earth Sci February (2017) Vol. 60 No. 2
Liu Q, et al.
303
Comparison Figure 5 of SSA at four standard wavelengths between the PREDE/KXD and CIMEL/USTC measurements.
as aerosol size parameters, are quite appropriate for an overall investigation about the aerosol variations on their loadings and components at this urban site. Elementary 3.2 features of aerosol variations The frequency distributions of aerosol properties are calculated to get the primary knowledge about aerosol variations at Hefei. Herein the three concerned parameters are calculated, i.e., AOT, AE, and SSA, as displayed in Figure 6. For AOT and SSA, they are represented by the corresponding values on the reference wavelength of 550 nm, while AE-440/870 defined in the previous section is used to represent AE for denoting the overall characteristics of aerosol size. It should be pointed out that the quantities of valid measurements from sun photometer are rather uneven. Some days may have adequate records, but many others may have too few records or even being vacant. Therefore the daily average values are used in order to avoid statistical biases from the uneven sampling. Since the aerosol properties on rainy and overcast days can never be retrieved with nowadays techniques, it is believed that daily-average based statistics provide unbiased results, although there are a good many days without any measurement. As shown in Figure 6a, AOT ranges extensively from 0.2 to 2.0, peaking approximately at
0.4. Most AOT values are between 0.2 and 1.0, accounting for more than 80% of the total samples. The days with AOT lower than 0.5 cover about 50% with the minimum AOT as low as 0.17. The critical value for isolating the top 5% of AOT is around 1.5, suggesting that severe pollution situations at Hefei could be defined as daily average of AOT larger than 1.5, which is thus employed in the later part of this study. Such an AOT frequency distribution determines a mean value of 0.69. This AOT mean is ~18% lower than the annual mean of 0.84 reported in Wang et al. (2014). Since the effective data used by Wang et al. (2014) are mostly before 2012, this result indicates that 2012/2013 should be a relatively clean period compared with the previous several years. But for revealing the strict long-term variation pattern of AOT at Hefei, more data are required in particular the continuous satellite data, since the currently available ground-based aerosol measurements hardly have steady temporal extension longer than a decade. Although AE cannot be straightforwardly related to aerosol size, their values are largely determined by the specific size distribution of aerosols. Aerosol samples dominated by small-sized aerosols lead to large AE, and vice versa. Especially, AE could be readily derived under multi-wavelength measuring strategies for both ground- and satellite-based instrument, which has been extensively used to semi-quanti-
304
Sci China Earth Sci February (2017) Vol. 60 No. 2
Liu Q, et al.
Frequency Figure 6 distribution of AOT (a), AE (b), and SSA (c) as histograms with scale on the left side, and cumulative frequency as solid lines with scale on the right side.
tatively characterizing aerosol size. It is shown in Figure 6b that daily value of AE at Hefei does not exceed 2.0, with the maximum of 1.8. This upper limit of AE means that large and moderate-sized aerosols always exist in the air around this city. There is hardly situation that aerosols consist exclusively of fine-mode aerosols, such as carbonaceous aerosols from combustion and those sulfate/nitrate aerosols from gasto-particle transformation without experiencing hygroscopic growth. According to the cumulative frequency, aerosols with AE less than 1.0 contribute about 30%, which should be dust-like coarse-mode aerosols. The rest ones with AE ranging between 1.0 and 2.0 are moderate-sized aerosols. In contrast to AOT, the distribution pattern of AE resembles the normal distribution. The mode frequency location, mean, and median value are all around 1.2, meanwhile the frequency approximately has a monotonically decreasing trend with increasing and decreasing AE. The AE average is close to the AE derived in Wang et al. (2014), and also close to the report derived at Taihu site (Pan et al., 2010; Yu et al., 2011), but a bit
lower than that in Shanghai (Pan et al., 2010; He et al., 2012). Despite that AE is not a straightforward quantitative parameter, this contrast suggests relatively large size of aerosol ensemble occurring at Hefei compared with that of Shanghai. In other words, the large and moderate-sized aerosols have a considerable fraction at Hefei, while it is slightly lower in Shanghai. It is thus indicated that Hefei is more likely to be contaminated by dust particles, which may be caused by either local strong-wind driving or remote transportation from arid areas in North and Northwest China. The distribution pattern shown in Figure 6c gives large dynamic range of SSA, with a mean/median near 0.91/0.92, almost identical with the value derived from Shouxian (Fan et al., 2010), smaller than that of Shanghai (~0.94, He et al., 2012) and a bit larger than that of Taihu (~0.90, Xia et al., 2007) and Beijing (~0.89, Eck et al., 2005; Xia et al., 2006). As the SSA annual mean is close to 0.90, the aerosol ensemble in Hefei should be classified into moderately absorbing ones (Levy et al., 2007) if the seasonality is omitted. Ap-
Sci China Earth Sci February (2017) Vol. 60 No. 2
Liu Q, et al.
parently, the SSA value around 0.9 results from a combination of non-, moderately- and strongly-absorbing aerosols with SSA mostly ranging from 0.8 to 1.0. In contrast to the pattern of AOT and AE, two significant modes arise at 0.89 and 0.93 separately, which may be caused by external mixture of two major aerosol types, such as low-SSA burning carbonaceous aerosols and high-SSA industrial sulfate/nitrate aerosols. This is different with SSA variations in Shanghai, which is consistently above 0.90, implying very little contamination from carbonaceous aerosols. It could be thus concluded that the aerosol sources affecting Hefei are more complicated. Besides sulfate emitted mainly from industrial activities and vehicles, dust and carbonaceous aerosol loadings in this city are considerable, which lead to the larger dynamic range and lower the SSA mean value. Figure 7 gives monthly average variations of the aerosol properties. In addition to AOT, AE, and SSA abovementioned, AAOT was also examined, which is readily derived from the product of AOT and co-albedo (1-SSA), indicating the extinction magnitude solely caused by absorption. The synchronous analysis on the temporal series of all these parameters should give clues about variations of both aerosol types and their loadings at local scale. For AOT as shown in Figure 7a, the result is in a good agreement with previous studies (e.g., He et al., 2012; Wang et al., 2014), where AOT reaches the maximum in June and declines to very low values in autumn. The AOT peak in June is likely related to crop residues burning. It was revealed that a wide spread
305
of crop residues burning occurred in June as a result of harvest of winter wheat that resulted in a significant increase of aerosol loading (Xia et al., 2013). It is noteworthy that there is a very intense decline of AOT during summer, with the monthly minimum almost arising in August. The compromise of two situations may lead to the significant changing of AOT in summer. The first should be the strengthening of secondary aerosol production and hygroscopic growth in the favorable environment with high temperature and high moisture in late-spring and early summer. At this stage within the East Asia summer monsoon cycle, the relative humidity gets rather high while precipitation events remain still infrequent. The other is the onset of rainy season in mid-summer. The frequent wet scavenging caused by abundant rainfall cleanses the atmosphere and leads to decreasing AOT. Such an AOT variation pattern during summer is nearly consistent among the various regions in East China (Pan et al., 2010) and even the entire East Asia (Eck et al., 2005), implying that the meteorological condition determines to a large extent the aerosol loadings in this region given the approximately steady aerosol emissions from natural processes and anthropogenic activities. The fact that AOT dramatically decreases from June to July also agrees well with the GEOS-Chem simulations reported by Zhang et al. (2010), which used non-seasonal emissions and found very low AOT in July over eastern China regions. In addition to the June peak, AOT reaches another minor peak in March. This should be owing to the interfusion of dust aerosols emitted from arid/semiarid areas in central
The Figure 7 monthly variation of AOT (a), AE (b), SSA (c) and AAOT (d).
306
Sci China Earth Sci February (2017) Vol. 60 No. 2
Liu Q, et al.
Asia, which is carried by the cold air mass that frequently intrudes East China during early spring, leading to the significant long-range transportation of dust aerosols. The dominance of anthropogenic fine aerosols in summer and natural coarse aerosols in early spring is also evidently presented in Figure 7b, where AE gets lower than 1.0 in March and April, and becomes consistently increasing during the summer season. In particular, the AE is around 1.1 in June, a bit smaller than in May, which is probably due to the hygroscopic growth that generally augments the aerosol size. As the frequency of cloud formation and precipitation generation increases along summer months, only those very small aerosols that are hard to be activated as being cloud condensation nuclei (CCN) remain in the atmosphere, resulting in large AE. When the wet deposition regime gets diminishes and large-sized aerosols can survive for a while in the autumn, AE comebacks to the moderate value and then deceases gradually throughout winter and early spring. Given the annual cycle of AE and the minimum of monthly AE in spring exceeding 0.9, it is explicit that fine-mode urban pollution always dominates the aerosol ensemble over Hefei, despite occasional dust intrusion. This is consistent with the finding in Eck et al. (2005) that used measurements from Beijing site located in North China, which revealed that fine mode pollution aerosols contribute more to aerosol burden than dust particles throughout the year with AE never below 0.8. As shown in Figure 7c, SSA has an annual cycle similar to that of AOT, with a local peak at June, but gets its maximum ~0.94 in November and remains large during winter months. It is clear that aerosol properties in June always exhibit special features compared with adjacent months. The high relative humidity induced hygroscopic growth of hydrophilic aerosols from urban emission, consisting mainly of sulfate and nitrate, leads to increased aerosol size and thus increased scattering sections, which eventually increases the SSA of aerosols. On the other hand, the aerosols with large scattering section are more prone to be scavenged by precipitation. Since the small-sized aerosols tend to have low SSA values, SSA gets decreased when rainy weather begins to be dominant in midsummer. In particular, the variability of SSA in July is significantly large, which is possibly attributed to alternant sampling between moderate-sized aerosols in high-humidity weather before precipitation and small-sized aerosols after precipitation. It is very likely that precipitation occurrences and the associated wet scavenging play a critical role in determining the aerosol loading and their optical properties especially within summer. As a contrast, in the dry season at Hefei, the variability of SSA is mild and remains almost larger than 0.92, demonstrating lower absorption compared to wet season. The above deductions are further reinforced by the monthly variations of AAOT. As synthetic information from AOT
and SSA, AAOT gives the extent of absorptive extinction, which increases dramatically in late spring and decreases in late summer. In contrast to the AAOT variations in North China, generally associated with a winter peak, there is not AAOT accretion during winter at Hefei, suggesting that aerosols emitted from biomass burning contribute few to the aerosol ensemble in this urban area. It should be pointed out that such a summer low SSA and high AAOT is firstly reported in the East China regions, since a summer high SSA generally occurs in the limited studies (Fan et al., 2010; He et al., 2012). As mentioned previously, precipitation induced sampling differences could impose considerable effects on the results derived. Thereby further investigations based on strictly valid measurements accompanied by in situ measurements are still required to further clarify the annual cycle of aerosol properties in this region. Seasonality 3.3 of aerosol variation over Hefei As revealed in Figure 7, all these aerosol parameters experience considerable oscillation throughout the year, exhibiting notable seasonality. Since the local emission is mainly from anthropogenic activities and has little seasonal dependence, such seasonal variations should be largely attributed to meteorological conditions. The factors include temperature and humidity that facilitate/restrain gas-particle transformation, along with large-scale wind field that possibly expedite dilution of local pollution or import external aerosol burden. In this section, the seasonality of aerosol properties is investigated in depth to acquire the underlying cause that drives aerosol variations in each season at this urban site. The statistical parameters for individual aerosol property in the four seasons are summarized in Table 1. In order to get more quantitative descriptions about the aerosol size, the volume concentration and mean radius, for total, fine-mode, and coarse mode particles, respectively, retrieved from the AERONET standard algorithm, were also considered. The boundary size separating fine-mode and coarse-mode particles is around 0.5–1.0 μm (Dubovik et al., 2006). According to the statistics, the aerosol loading gets its maximum in winter and spring, with AOT near 0.75, approximately 10% higher than the annual mean. In spring, the total/coarse-mode volume concentration reaches 0.28/0.17 μm3 μm−2, greatly larger than those in the other three seasons. Since there is little chance of sea-salt aerosols being transported to this inland site, it is indicated that dusts contribute much and lead to remarkably large AOT in spring. The frequent passage of strong cold front during late winter and spring, which often picks up mineral particles at the upwind arid/semiarid and desert regions, should be the main factor causing high AOT at Hefei in spring season. The cause of high AOT in winter, however, is somewhat different, since the total and coarsemode volume concentration both remain moderate values.
Sci China Earth Sci February (2017) Vol. 60 No. 2
307
Liu Q, et al.
As the fine-mode volume concentration gets the maximum in winter, and SSA also reaches its maximum, the dominant aerosols in winter should be sulfate/nitrate aerosols from various industrial emissions, in contrast to coal-burning induced carbonaceous aerosols dominant during wintertime over regions in North China (Xia et al., 2006). It is noteworthy that SSA reaches the minimum in summer, in accordance with the occurrence of the maximum AAOT. Taking into account the monthly variation pattern of aerosol parameters during summer, such low SSA is subtle because it is actually from a mixture of moderate SSA in June/August and very low SSA in July. Since the volume of valid samples in summer is rather small, such a summer pattern may be related to the very limited samples, which cannot be completely excluded at present. Besides, a reasonable explanation for the summer low SSA should be that abundant rainfall in the rainy season result in effective wet deposition of airborne particles, especially those hydrophilic and water-soluble ones that should be mostly sulfate/nitrate aerosols, which grow violently in highly moist environment. They are apt to form cloud droplet embryo and then be eliminated from the atmosphere via precipitation. Although soot and black carbon aerosols also have chance to be eliminated by precipitation events through mechanisms other than CCN, they are more likely to be remained in the air than sulfate/nitrate aerosols. In this way, summertime aerosols in Hefei are characterized by relatively low concentration but strong absorption. This result is just opposite to that of Shanghai (He et al., 2012), which shows the SSA maximum in summer, and consistently increasing SSA from June to August. Since Shanghai is a coastal city, the extremely high humidity, frequent import of coarse-sized sea salt particles, as well as less impact from biomass burning processes, among others, should in part account for its summer maximum of SSA. Another feature of summer aerosols is its markedly small
size. Since the mean radius of both fine-mode and coarsemode particles during summer are not small compared to the other seasons, such a small averaged-size arises from a large contrast between fine-mode and coarse-mode particle concentration. In other words, the predominance of fine-mode aerosols in summer is much stronger than in the other seasons. As for volume concentration shown in Table 1, the quantity of total, fine-mode, or coarse-mode aerosols in autumn is nearly equal to the counterpart in summer, but the mean size of total aerosols is larger in autumn than in summer, implying a lower number concentration and thus smaller cross section that is effective for light extinction. This explains the even lower AOT in autumn, the minimum in the annual cycle. Especially, if considering the severe pollution event as daily AOT greater than 1.5, there is no such situation in autumn. But in the other seasons, 4.7% (spring), 8.3% (summer), and 7.0% (winter) of the total sampling days encounter severe pollution. It is again manifested that during summer, very clean days and highly polluted days arise alternately, which is probably attributed to the competition between secondary aerosol production and hygroscopic growth under high temperature/moisture condition and powerful wet deposition due to precipitation. Consequently AOT and SSA in the rainy season actually experience intense fluctuations and thus have larger standard deviation than those in the other seasons. As suggested in the above analysis, the seasonally averaged quantities are not sufficient to characterize the specific feature of aerosol variations. Especially in the wet season, the aerosol loading and optical properties both have very large variances. In order to obtain knowledge about the potential alteration of aerosol types as the magnitude of aerosol loading changes, AE and SSA are calculated for varying AOT. By using an AOT bin of 0.2, the corresponding results for each season are provided in Figure 8. The variation pattern of both AE and
Table 1 Seasonal statistics of aerosol parameters at Hefei derived from two-year measurements
Sample days
Spring
Summer
Autumn
Winter
106
60
67
43
AOT-550
0.75±0.37
0.65±0.40
0.59±0.31
0.74±0.36
AE-440/870
0.95±0.25
1.25±0.24
1.36±0.21
1.23±0.27
SSA-550
0.91±0.04
0.87±0.07
0.92±0.03
0.93±0.05
AAOT-550
0.06±0.02
0.07±0.03
0.04±0.02
0.05±0.03
Days with AOT ≥ 1.0
25 (23.6)
11 (18.3%)
7 (10.4%)
10 (23.3%)
Days with AOT ≥ 1.5
5 (4.7%)
5 (8.3%)
0
3 (7.0%)
Total
0.28±0.10
0.19±0.09
0.20±0.07
0.23±0.08
Fine
0.11±0.07
0.12±0.06
0.12±0.06
0.13±0.06
Coarse
0.17±0.09
0.07±0.05
0.08±0.03
0.10±0.07
Total
0.93±0.38
0.58±0.31
0.65±0.25
0.68±0.37
Fine
0.18±0.05
0.20±0.05
0.18±0.04
0.21±0.06
Coarse
2.37±0.42
3.13±0.53
2.77±0.31
2.74±0.38
Volume concentration (μm3 μm−2)
Mean radius (μm)
308
Sci China Earth Sci February (2017) Vol. 60 No. 2
Liu Q, et al.
Variation Figure 8 of AE (red lines with scale on the left side) and SSA (blue lines with scale on the right side) depending on increasing AOT in each season.
SSA are found to be comparable between autumn and winter. In these two seasons, AE monotonically decreases and SSA monotonically increases with increasing AOT, indicating that pollutions are mainly caused by coarse-mode weakly-absorbing particles. Such particles are very likely to be dust aerosols for such an inland region, which could significantly reduce AE and raise SSA given a considerable increment of loading. The severe pollution situation with AOT near 1.5 has AE down to 1.0 and SSA up to 0.96. In spring, as AOT increases, both AE and SSA increase steadily until AOT reaching 1.3, and then get almost saturated when AOT increases any more. For polluted situation with AOT approaching 1.5, SSA keeps 0.94 and AE gets around 1.1, suggesting that pollution occurring in spring is largely caused by excessive fine-mode aerosols. It is instead the relatively clean days that are dominated by coarse-mode aerosols with AE less than 1.0 during spring season. This is completely distinct with the other seasons, where clean days are contributed fairly from fine-mode aerosols and coarse-mode aerosols, resulting in AE around 1.5. It is indicated that the strong northwesterly and/or northerly prevailing in spring can clean out to a large extent the artificial aerosols from local urban emissions, but on the other hand this northerly or northwesterly wind would probably bring dust aerosols from upwind inland regions. These especial regimes jointly lead to the ubiquitous coarse-mode aerosols, which even constitute the background loading of particles in the air during spring-
time at Hefei site. As a contrast, in autumn/winter, fine-mode particles have a large fraction when the aerosol loading is very low, and the contribution from coarse-mode particles gets considerable when AOT increases gradually. The summer pattern is a bit different from that of autumn/winter and spring. Although SSA seems to keep increasing with AOT in the entire valid range, the dependence of AE on AOT is not monotonic. It is demonstrated in Figure 8c that in summer AE decreases to near 1.0 at moderate AOT value (~1.1), and then increases with AOT up to 1.2. Such a pattern suggests that it is the addition of large-sized particles that modifies the air from clean to moderate pollution situation, while small-sized particles contribute much to make the air being even more polluted. So at Hefei site the severe pollution event in summer is not caused by dust aerosols as in autumn/winter, but caused by fine-mode particles that are probably sulfate/nitrate aerosols. It is the rising fraction of sulfate/nitrate in the aerosol ensemble that results in the increasing SSA, which goes up eventually to 0.93 for the near 1.5 AOT. Apparently, the very high AOT in summer is possibly accompanied by highly excessive sulfate/nitrate aerosols. It is speculated that the gas-particle transformation gets vigorous and the resulting fine-mode aerosols accumulate under some favorable conditions, leading to the magnitude of the generated fine-mode particles become quite high. Given relatively high humidity in the rainy season, these hydrophilic particles would grow
Sci China Earth Sci February (2017) Vol. 60 No. 2
Liu Q, et al.
up dramatically by coalescing abundant water. The sufficient hygroscopic growth then result in the systematic increase of fine-mode particle size. But for the averaged size of the aerosol ensemble consisting of fine-mode and coarse-mode particles, the increase of fine-mode fraction should lower the mean size and thus increase AE. It is again manifested that the aerosol variation in summer is more complicated than in the other seasons. The variable meteorological conditions within summer modify the aerosol loading, composition and the relevant optical properties to a very large extent. In order to further clarify the alteration of aerosol types as AOT varies, the wavelength dependence of SSA is examined following the strategy employed by Russell et al. (2010) and Li et al. (2015), which revealed the effectivity of SSA spectra in discerning the chemical composition of dominant aerosols. As shown in Figure 9, it is almost the same in four seasons that SSA increases at every wavelength with increasing AOT, although the variational amplitude at each wavelength is somewhat different. This suggests that as AOT increases, the increase of SSA not only occurs at the reference wavelength of 550 nm, but also consistently arises at the whole visible and near-infrared spectrum. Such a coherent increase of SSA with AOT is of great importance since the spectrally averaged SSA within the visible and near-infrared band plays a critical role in determining the overall radiative effects of aerosols. It could be deduced that the aerosol ensemble has more scattering capability and less absorption capability as the aerosol loading increases notably. In other words, it is
309
only in very clean days that the radiative effect of aerosols over Hefei is likely to be positive. As the air quality gets bad and even worse, cooling effects would eventually overwhelm. Although black carbon and other carbonaceous aerosols generated from various combustion processes are an important ingredient of aerosols in Hefei, these industrial aerosols seem not to be the leading contributor to severe pollution events around this city. Also note that all the SSA curves in four seasons are monotonic, and decrease approximately linearly with wavelength. Since there is not a SSA peak around 670 nm as found in Li et al. (2015) for aerosols mixed with considerable dusts, it is indicated that aerosols in Hefei are always dominated by artificial particles, which are mainly non-absorbing sulfate/nitrate and highly-absorbing carbonaceous aerosols, both being fine-mode particles. So dust aerosols transported from remote desert areas have very limited contribution to the local pollution. Even in spring, SSA consistently decreases with wavelength for every AOT range, showing the absorption characteristics of non-dust aerosols. In autumn and winter, only when the air is severely polluted, the SSA variation becomes wavelength-independent, which resembles to some extent the SSA spectral characteristics of dusts. This agrees well with the previous analysis that the amount of coarse-mode aerosols increases evidently as AOT increases in autumn/winter. Just on the contrary, the SSA curves in summer have relatively large slope, much lower contribution from absorption at long wavelength than short wavelength, which flects the pre-
Spectral Figure 9 dependence of SSA for different AOT ranges in each season.
310
Sci China Earth Sci February (2017) Vol. 60 No. 2
Liu Q, et al.
dominance of sulfate/nitrate aerosols and/or carbonaceous aerosols. If taking into account the wholly elevated SSA, carbonaceous aerosols could be ruled out and the marked increase of sulfate/nitrate aerosols should be the primary cause for inducing severe pollution during summertime at Hefei. The above deduction is also supported by the results of analysis on the AOT-dependent aerosol concentration and particle size, as shown in Figures 10 and 11, respectively. The volume concentration is employed to characterize the aerosol concentration, while the mean radius is used to quantify the particle size. Total aerosols, fine-mode, and coarse-mode aerosols are calculated separately. As expected, the volume concentration of total aerosols increases dramatically as AOT increases. For AOT changing from 0.5 to about 1.5, the total concentration generally becomes double, while the mean radius of total aerosols keeps nearly invariant or has a very limited variation. Apparently AOT is largely determined by the change of aerosol volume concentration rather than the alteration of mean size of the aerosol ensemble. The concentration variation of fine-mode aerosols is approximately proportional to the pattern of total ones, but the concentration variation of coarse-mode aerosols show evident seasonal difference. In autumn and winter, although the concentration of coarse-mode aerosols remains always lower than that of fine-mode aerosols, it experiences a considerable increase with AOT, approaching the amplitude of fine-mode ones. It
is because the simultaneous concentration increase of both fine-mode and coarse-mode aerosols that the overall size of total aerosols keeps nearly invariant with varying AOT during autumn and winter. It is evident that in these two seasons, in addition to the common increase of fine-mode artificial aerosols, coarse-mode dust aerosols contribute much to the pollution at Hefei. The contrast between volume concentration of fine-mode and coarse-mode aerosols is unique in spring. The concentration of coarse-mode aerosols predominate when the air is clean, since the concentration of fine-mode aerosols in clean spring days gets as low as below 0.1 μm3 μm−2 while that of coarse-mode aerosols is around 0.2 μm3 μm−2. But the concentration of fine-mode aerosols increases rapidly as AOT increases, which is much higher than that of coarse-mode ones. When AOT increase to near 1.1, these two concentration become equal to each other, and the concentration of fine-mode aerosols exceeds coarse-mode ones for further increase of AOT. As pointed out previously, in spring the concentration of dust aerosols remains a high level but has little variability. The pollution event in spring is instead caused by artificial aerosols, which generally experiences very large amplitude of raise and radically determines the increase of overall aerosol concentration. Such a notable increase of fine-mode particles even reduces the mean size of the total aerosols from larger than 1.0 μm in clean situation to about 0.7 μm in severe pollution situation, which is the maximum change of mean size
Variation Figure 10 of aerosol volume concentration with increasing AOT in each season. In each panel, black, blue and red lines indicate total, fine-mode and coarse-mode aerosols, respectively.
Sci China Earth Sci February (2017) Vol. 60 No. 2
Liu Q, et al.
311
Variation Figure 11 of aerosol mean radius with increasing AOT in each season. In each panel, black, blue and red lines indicate total, fine-mode and coarse-mode aerosols, respectively.
among the four seasons. In summer, the concentration of coarse-mode aerosols remains also almost invariant but it is at a very low level. In addition, the coarse-mode concentration keeps always lower than the fine-mode concentration with varying AOT. The mean size of fine-mode/coarse-mode aerosols increases/decreases to some extent while the mean size of total aerosols increases and then decrease with further increasing AOT, accordant with the foregoing conclusions from the examination on AOT-dependent AE variation. It is thus the increase of fine-mode aerosols that leads to the increase of overall aerosol concentration and so acts as the main factor inducing the increasing AOT. Consistent with the previous analysis on AE and SSA, severe pollution occurring in summer is primarily caused by excessive sulfate/nitrate aerosols that are fine-mode/non-absorbing aerosols, rather than coarse-mode/weakly-absorbing dusts or fine-mode/highly-absorbing carbonaceous substance. In order to ascertain the leading contributor causing severe pollution events at Hefei, the 13 days with daily mean AOT greater than 1.5 during the concerned period are especially examined as given in Table 2. Although the statistics on seasonal scale gives some variational features of aerosol type and the corresponding optical properties with varying aerosol loading, the results of these individual daily cases are diverse and they are not always consistent with the previous findings. In particular, both the minimum (0.898) and the maximum (0.992) of SSA occur in summer pollution cases.
This reflects to some extent the SSA contrast between June and July as shown in the monthly variations, and again manifests the subtle average of aerosol properties in summertime, which is strongly affected by meteorological conditions. It is found that in winter, the severely polluted days are generally induced by aerosols that have high SSA and very low AAOT. Fine-mode particles always have a very large fraction in the aerosol concentration and constitute the major type in the aerosol ensemble at this site. But in spring, coarse-mode can play a key role in some cases to significantly raise the total aerosol concentration. In the case that coarse-mode concentration gets the maximum of about 0.46 μm3 μm−2, approximately three times that of fine-mode ones, the mean size of total aerosols also reaches its maximum of about 0.86 μm, which is definitely a dust pollution case. Since the seasonal statistics gives a reduced aerosol size for increasing AOT, it is suggested that dust-only induced pollution is not frequent even in spring. It is true that there is generally a high level of coarse-mode aerosol concentration in this season, but most severe pollution events are still caused by excessive fine-mode aerosols that are from industrial emissions around this city.
Conclusions 4. East China is a typical region that has a high level of aerosol burden from diverse emission sources and often experiences
312
Sci China Earth Sci February (2017) Vol. 60 No. 2
Liu Q, et al.
Table 2 Aerosol parameters during 13 polluted days at Hefeia)
Winter
Spring
Summer
Date
AOT
SSA
AAOT
20120122
1.52
0.99
20120216
1.53
20130227
1.84
Volume concentration (μm3 μm−2)
Mean radius (μm)
Total
Fine
Coarse
Total
Fine
Coarse
0.02
0.33
0.27
0.06
0.45
0.33
2.15
0.99
0.02
0.35
0.27
0.09
0.47
0.28
2.46
0.99
0.03
0.38
0.34
0.04
0.50
0.42
1.84
20120306
1.52
0.94
0.08
0.37
0.25
0.12
0.48
0.24
2.02
20120311
1.88
0.96
0.08
0.57
0.27
0.29
0.83
0.24
2.38
20120318
1.56
0.95
0.08
0.59
0.14
0.46
0.86
0.19
1.35
20120411
1.77
0.96
0.06
0.42
0.30
0.11
0.57
0.30
2.75
20130312
1.94
0.98
0.05
0.46
0.40
0.06
0.34
0.25
2.10
20120612
1.65
0.91
0.15
0.39
0.23
0.16
0.73
0.26
2.71
20120615
1.50
0.90
0.13
0.42
0.24
0.19
0.71
0.25
2.75
20120716
1.68
0.92
0.13
0.33
0.28
0.05
0.49
0.33
4.02
20120717
1.67
0.97
0.05
0.33
0.29
0.04
0.44
0.34
2.57
20130623
1.56
0.99
0.01
0.35
0.27
0.08
0.48
0.31
2.06
a) daily averaged AOT>1.5
intensive aerosol variations due to monsoon-dominated meteorological conditions. Although many features of aerosol variation in this region have been reported by using ground measurements, they are mostly concentrated around Shanghai, a megalopolis located near the coast, which has been proved to be inappropriate for representing aerosol situation in the major inland part of East China. A two-year sunphotometer measurement derived from a downtown site at Hefei, a moderate city located in central East China, was examined in this study to clarify the aerosol variations over this specific area. Relying on the retrievals of the standard AERONET algorithm, the key aerosol parameters that are related to aerosol concentration and chemical composition, i.e., AOT, AE, and SSA, as well as volume concentration and mean size of aerosols were jointly analyzed and the major findings are as follows. (1) The AOT at Hefei is higher in winter/spring and gets relatively lower in summer/autumn, with the annual mean at 0.69, close to that of Shanghai. The annual mean of SSA (0.91) is a bit lower than that of Shanghai (0.94), but the daily SSA values at Hefei have a much larger dynamic range, resulting from the mixture of light-absorbing soot/dust aerosols and non-absorbing sulfate/nitrate aerosols, in contrast to Shanghai that is predominated by daily SSA above 0.9. The annual mean of AE is also lower than that of Shanghai, indicating a larger fraction of coarse-mode aerosols at Hefei, and correspondingly more fine-mode particles suspended in the air at Shanghai. The increase of dust aerosols is notable in spring, with the seasonal mean of AE getting lower than 1.0 and the volume concentration of coarse-mode particles greatly higher than the other months. The prominent aerosol loading in winter is found to be related to industrial sulfate/nitrate aerosols rather than biomass burning aerosols
that are remarkable at wintertime in North China. (2) Although summer AOT is moderate among the four seasons, it is in June that AOT reaches the maximum of the year, which is the unique monthly AOT exceeding 0.9. There is a dramatic decline of AOT in summer, reaching the minimum of the year in August. The competition of two distinct situations causes the intense fluctuation of aerosol loading during summer. The first is the strengthening of secondary aerosol production and hygroscopic growth in late-spring and early summer, when the weather is characterized by high temperature/moisture but infrequent precipitation. The second is the effective wet scavenging caused by abundant rainfall since the onset of rainy season in mid-summer. Along with the rapid decrease of AOT from June to August, the overall size of aerosols decreases. Meanwhile, AAOT decreases monotonically but SSA reaches its minimum in July, suggesting the possibility that considerable carbonaceous aerosols emerge around mid-summer. It is speculated that precipitation occurrences play a critical role in modifying the aerosol loading, while the different efficiencies of wet scavenging on carbonaceous and sulfate/nitrate aerosols largely determine the aerosol optical properties in summer. (3) Artificial fine-mode particles are found to be the major component in the aerosol ensemble at Hefei, which in most time have much higher volume concentration that its coarse-mode counterpart. In autumn/winter, fine-mode particles dominate the aerosol loading when AOT is low, and the contribution from coarse-mode particles gets increasingly considerable when AOT gradually increases. But in spring the pollution is very likely caused by excessive fine-mode aerosols. It is instead the dust particles that constitute the background aerosol loading in relatively clean days during springtime. In summer the increase of large-sized particles
Sci China Earth Sci February (2017) Vol. 60 No. 2
Liu Q, et al.
seems to modify the air from clean to mild-pollution situation, while small-sized particles contribute much to make the air being even more polluted. The three key processes that are active in the high temperature/humidity environment concentrating on sulfate/nitrate aerosols, i.e., gas-particle transformation, hygroscopic growth, and precipitation associated wet scavenging, should be the radical reason for the complicated aerosol variations occurring in summer. (4) The wavelength dependence of SSA was also examined for varying aerosol loading in each season. It is identical in the four seasons that as AOT increases, the increase of SSA not only occurs at the reference wavelength of 550 nm, but also across the entire visible and near-infrared spectrum. The aerosol ensemble at Hefei would have more scattering and less absorption capability as the aerosol loading increases notably. It is only in very clean days that the radiative effect of aerosols around this city is likely to be positive/heating. As the air quality gets bad and even worse, the negative/cooling effect would eventually overwhelm, which exclude soot and black carbon as the main contributor to severe pollution at this city. Moreover, all the SSA spectra in the four seasons were found to consistently decrease with wavelength. Such monotonic patterns indicate that dust aerosols hardly act as the dominant part to the local pollution even in spring. It is only under severely polluted situation in autumn/winter, the SSA variation becomes nearly wavelength-independent, representing considerable contribution from dusts. The SSA spectra in summer have relatively large slope, reinforcing the fact that marked increase of sulfate/nitrate aerosols induces the severe pollution occurring in summer at Hefei. Acknowledgements The authors greatly acknowledge Wang Zhenzhu and Liu Dong for their support on PREDE data and the two reviewers for their valuable suggestions. This work was supported by the National Natural Science Foundation of China (Grant Nos. 41175032 & 41575019).
References Albrecht B A. 1989. Aerosols, cloud microphysics, and fractional cloudiness. Science, 245: 1227–1230 Angstrom A. 1964. The parameters of atmospheric turbidity. Tellus, 16: 64–75 Bergin M H, Greenwald R, Xu J, Berta Y, Chameides W L. 2001. Influence of aerosol dry deposition on photosynthetically active radiation available to plants: A case study in the Yangtze Delta Region of China. Geophys Res Lett, 28: 3605–3608 Charlson R J, Schwartz S E, Hales J M, Cess R D, Coakley Jr. J A, Hansen J E, Hofmann D J. 1992. Climate forcing by anthropogenic aerosols. Science, 255: 423–430 Che H Z, Shi G, Uchiyama A, Yamazaki A, Chen H, Goloub P, Zhang X. 2008. Intercomparison between aerosol optical properties by a PREDE skyradiometer and CIMEL sunphotometer over Beijing, China. Atmos Chem Phys, 8: 3199–3214 Che H Z, Xia X A, Zhu J, Li Z, Dubovik O, Holben B, Goloub P, Chen H, Estellés V, Cuevas-Agulló E, Blarel L, Wang H, Zhao H, Zhang X, Wang Y, Sun J, Tao R, Zhang X, Shi G. 2014. Column aerosol optical
313
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: 2125–2138 Che H Z, Zhang X Y, Chen H B, 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 Chen J, Jiang H, Wang B, Xiao Z, Jiang Z, Zhou G, Yu S. 2012. Aerosol optical properties from sun photometric measurements in Hangzhou district, China. Int J Remote Sens, 33: 2451–2461 Dubovik O, King M D. 2000. A flexible inversion algorithm for retrieval of aerosol optical properties from Sun and sky radiance measurements. J Geophys Res, 105: 20673–20696 Dubovik O, Smirnov A, Holben B N, King M D, Kaufman Y J, Eck T F, Slutsker I. 2000. Accuracy assessments of aerosol optical properties retrieved from Aerosol Robotic Network (AERONET) Sun and sky radiance measurements. J Geophys Res, 105: 9791–9806 Dubovik O, Sinyuk A, Lapyonok T, Holben B N, Mishchenko M, Yang P, Eck T F, Volten H, Muñoz O, Veihelmann B, van der Zande W J, Leon J F, Sorokin M, Slutsker I. 2006. Application of spheroid models to account for aerosol particle nonsphericity in remote sensing of desert dust. J Geophys Res, 111: D11208 Eck T F, Holben B N, Dubovik O, Smirnov A, Goloub P, Chen H B, Chatenet B, Gomes L, Zhang X Y, Tsay S C, Ji Q, Giles D, Slutsker I. 2005. Columnar aerosol optical properties at AERONET sites in central eastern Asia and aerosol transport to the tropical mid-Pacific. J Geophys Res, 110: D06202 Estellés V, Campanelli M, Utrillas M P, Expósito F, Martínez-Lozano J A. 2012. Comparison of AERONET and SKYRAD4.2 inversion products retrieved from a Cimel CE318 sunphotometer. Atmos Meas Tech, 5: 569–579 Fan X, Chen H, Xia X A, Li Z, Cribb M. 2010. Aerosol optical properties from the atmospheric radiation measurement mobile facility at Shouxian, China. J Geophys Res, 115: D00K33 Han Z W, Li J W, Xia X A, Zhang R. 2012. Investigation of direct radiative effects of aerosols in dust storm season over East Asia with an online coupled regional climate-chemistry-aerosol model. Atmos Environ, 54: 688–699 Hansen J, Sato M, Ruedy R. 1997. Radiative forcing and climate response. J Geophys Res, 102: 6831–6864 He Q, Li C, Geng F, Yang H, Li P, Li T, Liu D, Pei Z. 2012. Aerosol optical properties retrieved from Sun photometer measurements over Shanghai, China. J Geophys Res, 117: D16204 Holben B N, Eck T F, Slutsker I, Tanré D, Buis J P, Setzer A, Vermote E, Reagan J A, Kaufman Y J, Nakajima T, Lavenu F, Jankowiak I, Smirnov A. 1998. AERONET—A federated instrument network and data archive for aerosol characterization. Remote Sens Environ, 66: 1–16 Kaufman Y J. 1993. Aerosol optical thickness and atmospheric path radiance. J Geophys Res, 98: 2677–2692 Leibensperger E M, Mickley L J, Jacob D J, Chen W T, Seinfeld J H, Nenes A, Adams P J, Streets D G, Kumar N, Rind D. 2012. Climatic effects of 1950–2050 changes in US anthropogenic aerosols-Part 1: Aerosol trends and radiative forcing. Atmos Chem Phys, 12: 3333–3348 Levy R C, Remer L A, Dubovik O. 2007. Global aerosol optical properties and application to Moderate Resolution Imaging Spectroradiometer aerosol retrieval over land. J Geophys Res, 112: D13210 Li C C, Mao J T, Lau K H A, Chen J C, Yuan Z, Liu X Y, Zhu A H, Liu G Q. 2003. Characteristics of distribution and seasonal variation of aerosol optical depth in eastern China with MODIS products. Chin Sci Bull, 48: 2488–2495 Li J, Carlson B E, Lacis A A. 2015. Using single-scattering albedo spectral curvature to characterize East Asian aerosol mixtures. J Geophys Res Atmos, 120: 2037–2052 Li Z Q, Chen H, Cribb M, Dickerson R, Holben B, Li C, Lu D, Luo Y, Maring
314
Sci China Earth Sci February (2017) Vol. 60 No. 2
Liu Q, et al.
H, Shi G, Tsay S C, Wang P, Wang Y, Xia X, Zheng Y, Yuan T, Zhao F. 2007. Preface to special section on East Asian Studies of tropospheric aerosols: An international regional experiment (EAST-AIRE). J Geophys Res, 112: D22S00 Li Z Q, Lee K H, Wang Y, Xin J, Hao W M. 2010. First observation-based estimates of cloud-free aerosol radiative forcing across China. J Geophys Res, 115: D00K18 Li Z Q, 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. 2011. East Asian studies of tropospheric aerosols and their impact on regional climate (EASTAIRC): An overview. J Geophys Res, 116: D00K34 Liu J J, Xia X A, Wang P, Li Z, Zheng Y, Cribb M, Chen H. 2007. Significant aerosol direct radiative effects during a pollution episode in northern China. Geophys Res Lett, 34: L23808 Liu J J, Zheng Y, Li Z, Flynn C, Cribb M. 2012. Seasonal variations of aerosol optical properties, vertical distribution and associated radiative effects in the Yangtze Delta region of China. J Geophys Res, 117: D00K38 Liu Q, Ding W D, Fu Y F. 2011. The seasonal variations of aerosols over East Asia as jointly inferred from MODIS and OMI. Atmos Ocean Sci Lett, 4: 330–337 Nakajima T, Tonna G, Rao R, Boi P, Kaufman Y, Holben B. 1996. Use of sky brightness measurements from ground for remote sensing of particulate polydispersions. Appl Opt, 35: 2672–2686 Nakajima T, Sekiguchi M, Takemura T, Uno I, Higurashi A, Kim D, Sohn B J, Oh S N, Nakajima T Y, Ohta S, Okada I, Takamura T, Kawamoto K. 2003. Significance of direct and indirect radiative forcings of aerosols in the East China Sea region. J Geophys Res, 108: 8658 Nakajima T, Yoon S C, Ramanathan V, Shi G Y, Takemura T, Higurashi A, Takamura T, Aoki K, Sohn B J, Kim S W, Tsuruta H, Sugimoto N, Shimizu A, Tanimoto H, Sawa Y, Lin N H, Lee C T, Goto D, Schutgens N. 2007. Overview of the atmospheric brown cloud east asian regional experiment 2005 and a study of the aerosol direct radiative forcing in east Asia. J Geophys Res, 112: D24S91 Pan L, Che H, Geng F, Xia X, Wang Y, Zhu C, Chen M, Gao W, Guo J. 2010. Aerosol optical properties based on ground measurements over the Chinese Yangtze Delta Region. Atmos Environ, 44: 2587–2596 Remer L A, Kaufman Y J. 2006. Aerosol direct radiative effect at the top of the atmosphere over cloud free ocean derived from four years of MODIS data. Atmos Chem Phys, 6: 237–253 Remer L A, Kaufman Y J, Tanré D, Mattoo S, Chu D A, Martins J V, Li R R, Ichoku C, Levy R C, Kleidman R G, Eck T F, Vermote E, Holben B N. 2005. The MODIS aerosol algorithm, products, and validation. J Atmos Sci, 62: 947–973 Rosenfeld D. 2006. Aerosols, clouds, and climate. Science, 312: 1323–1324 Russell P B, Bergstrom R W, Shinozuka Y, Clarke A D, DeCarlo P F, Jimenez J L, Livingston J M, Redemann J, Dubovik O, Strawa A. 2010. Absorption angstrom exponent in AERONET and related data as an indicator of
aerosol composition. Atmos Chem Phys, 10: 1155–1169 Schuster G L, Dubovik O, Holben B N. 2006. Angstrom exponent and bimodal aerosol size distributions. J Geophys Res, 111: D07207 Stocker T F, Qin D H, Plattner G K, Tignor M, Allen S K, Boschung J, Nauels A, Xia Y, Bex V, Midgley P M. 2013. IPCC, 2013. Climate change 2013. The Physical Science Basis. Contribution of Working Group I to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge: Cambridge University Press. 1535 Twomey S. 1977. The influence of pollution on the shortwave albedo of clouds. J Atmos Sci, 34: 1149–1152 Wang J, Yang S, Zhao Q, Cui S. 2012. Analyses of aerosol properties in Hefei based on polarization measurements of a sun photometer. Int J Remote Sens, 33: 69–80 Wang Y, Che H, Ma J, Wang Q, Shi G, Chen H, Goloub P, Hao X. 2009. Aerosol radiative forcing under clear, hazy, foggy, and dusty weather conditions over Beijing, China. Geophys Res Lett, 36: L06804 Wang Z Z, Liu D, Wang Z E, Wang Y J, Khatri P, Zhou J, Takamura T, Shi G Y. 2014. Seasonal characteristics of aerosol optical properties at the SKYNET Hefei site (31.90°N, 117.17°E) from 2007 to 2013. J Geophys Res Atmos, 119: 6128–6139 Xia X A, Chen H B, Wang P C, Zhang W X, Goloub P, Chatenet B, Eck T F, Holben B N. 2006. Variation of column-integrated aerosol properties in a Chinese urban region. J Geophys Res, 111: D05204 Xia X A, Li Z Q, Holben B, Wang P C, Eck T, Chen H B, Cribb M, Zhao Y X. 2007. Aerosol optical properties and radiative effects in the Yangtze Delta region of China. J Geophys Res, 112: D22S12 Xia X A, Zong X, Sun L. 2013. Exceptionally active agricultural fire season in mid-eastern China in June 2012 and its impact on the atmospheric environment. J Geophys Res Atmos, 118: 9889–9900 Xin J Y, Wang Y S, Li Z Q, Wang P C, Hao W M, Nordgren B L, Wang S G, Liu G R, Wang L L, Wen T X, Sun Y, Hu B. 2007. Aerosol optical depth (AOD) and Ångström exponent of aerosols observed by the Chinese Sun Hazemeter Network from August 2004 to September 2005. J Geophys Res, 112: D05203 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. Atmos Environ, 36: 161–173 Yu X N, Zhu B, Yin Y, Fan S X, Chen A J. 2011. Seasonal variation of columnar aerosol optical properties in Yangtze River Delta in China. Adv Atmos Sci, 28: 1326–1335 Zhang H, Wang Z L, Wang Z Z, Liu Q X, Gong S L, Zhang X Y, Shen Z P, Lu P, Wei X D, Che H Z, Li L. 2012. Simulation of direct radiative forcing of aerosols and their effects on East Asian climate using an interactive AGCM-aerosol coupled system. Clim Dyn, 38: 1675–1693 Zhang L, Liao H, Li J. 2010. Impacts of Asian summer monsoon on seasonal and interannual variations of aerosols over eastern China. J Geophys Res, 115: D00K05