Impacts of the high loadings of primary and secondary aerosols on light extinction at Delhi during wintertime

Atmospheric Environment 92 (2014) 60e68

Contents lists available at ScienceDirect

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

Impacts of the high loadings of primary and secondary aerosols on light extinction at Delhi during wintertime S. Tiwari a, A.K. Srivastava a, *, D.M. Chate b, P.D. Safai b, D.S. Bisht a, M.K. Srivastava c, G. Beig b a b c

Indian Institute of Tropical Meteorology (Branch), Prof Ramnath Vij Marg, New Delhi, India Indian Institute of Tropical Meteorology, Dr Homi Bhabha Road, Pashan, Pune, India Department of Geophysics, Banaras Hindu University, Varanasi, India

h i g h l i g h t s
First study on contribution of different species to extinction coefficient.
Scattering type aerosols were dominated by w76% than absorbing type aerosols.
The largest contribution was observed for organic carbon (w46%).
Lowest contribution was observed for ammonium nitrate (w4%).

a r t i c l e i n f o

a b s t r a c t

Article history: Received 7 November 2013 Received in revised form 27 March 2014 Accepted 29 March 2014 Available online 30 March 2014

High emissions of anthropogenic aerosols over Indo-Gangetic Plain (IGP) inspired continuous measurements of fine particles (PM2.5), carbonaceous aerosols (BC, OC and EC), oxides of nitrogen (NOx) and estimation of light extinction (bext) and absorption (babs) coefficients over Delhi during high pollution season in winter from December 2011 to March 2012. During study period, the mass concentrations of PM2.5, BC and NOx were 186.5  149.7 mg m3, 9.6  8.5 mg m3 and 23.8  16.1 ppb, respectively. The mass concentrations of OC and EC were studied by two different techniques (i) off-line (gravimetric method) and (ii) semi-continuous (optical method) and their mean mass concentrations were 51.1  15.2, 10.4  5.5 mg m3 and 33.8  27.7, 8.2  6.2 mg m3, respectively during the study period. The ratios of mass concentration of OC to EC in both cases were in between 4 and 5. The source contribution of carbonaceous aerosols in PM2.5 estimated over 24hrs, during day- and night-time where motor vehicles accounted for w69%, 90% and 61% whereas coal combustion accounted for w31%, 10% and 39%, respectively. The estimated mean values of bext and babs over the station were 700.0  268.6 and 71.7  54.6 Mm1, respectively. In day and night analysis, bext is w37% higher during night-time (863.4 Mm1) than in day-time (544.5 Mm1). Regression analysis between bext and visibility showed significant negative correlation (r ¼ 0.85). The largest contribution in the light extinction coefficients was found to be due to organic carbon (w46%), followed by elemental carbon (w24%), coarse mode particles (w18%), ammonium sulfate (w8%) and ammonium nitrate (w4%). The individual analysis of light extinction due to chemical species and coarse mode particles indicates that scattering type aerosols dominated by w76% over the absorbing type. The aforementioned results suggest that the policyinduced control measures at local administration level are needed to mitigate the excess emissions of carbonaceous aerosols over IGP region which ranks highest in India and elsewhere in worldwide. Crown Copyright Ó 2014 Published by Elsevier Ltd. All rights reserved.

Keywords: Particulate mass Carbonaceous aerosols Extinction coefficient Indo-Gangetic plain Meteorological effect

1. Introduction

* Corresponding author. E-mail address: [email protected] (A.K. Srivastava). http://dx.doi.org/10.1016/j.atmosenv.2014.03.064 1352-2310/Crown Copyright Ó 2014 Published by Elsevier Ltd. All rights reserved.

The Indo-Gangetic plain (IGP) is the most populous and one of the highly polluted regions in the world, surrounded by various anthropogenic sources such as burning of fossil fuels and agricultural residues, vehicular emissions etc. Its unique topography along

S. Tiwari et al. / Atmospheric Environment 92 (2014) 60e68

with atmospheric stability and calm winds during winter period are responsible for the higher level of pollution over the region during the period (Srivastava et al., 2012a,b; Tiwari et al., 2013a,b; Rehman et al., 2011). Delhi, the fourth most populous megacity with 19 million inhabitants is located in the western IGP region in northern part of India (http://www.indiaonlinepages-com/ population/delhi-population.html). Tremendous growth of population, urbanization, industrialization, human migration as well as meteorological conditions over the station increased the atmospheric pollution levels, especially during winter which causes visibility degradation, dense fog, smog and haze every year (Gautam et al., 2007; Singh et al., 2008; Lal et al., 2013). Carbonaceous aerosols are emitted from the combustion of fossil and biomass fuels and are more abundant component in fine particulate matter (i.e. PM2.5). They can be usually classified into three different categories: organic carbon (OC), elemental carbon (EC) and carbonate carbon (CC). In OC, carbon is associated with organic compounds emitted either directly (anthropogenic emissions) from the source, or by condensation due to atmospheric oxidation of volatile organic compounds (i.e. secondary OC) in the atmosphere. On the other hand, EC is formed during the combustion of hydrocarbons and is essentially non-volatile at ambient temperature (Jeong et al., 2004; Hussain et al., 2007; Rogge et al., 1998; Sheesley et al., 2003). Often, EC has inter-changeable relationship with black carbon (BC) (Turpin and Huntzicker, 1991) and are typically light absorbing (Jacobson, 2004, 2006). Mineral dust is one of the good sources for CC, which is not given much consideration due to low atmospheric mass concentration. In the urban areas, diesel automobiles, especially heavy duty trucks are the main source of EC in the atmosphere (Gray and Cass, 1998). In the Indian context, however, very few studies have been conducted for the measurement of total carbonaceous aerosols (Satsangi et al., 2012; Tiwari et al., 2013a; Srivastava et al., 2014). Venkataraman et al. (2005) reported the relative contributions of fossil fuel, open biomass and bio-fuel combustion in BC mass are w25%, 33% and 42%, respectively whereas w13%, 43% and 44% in OC mass, respectively. In another study, Stone et al. (2010) have estimated only 4% contribution of OC from diesel combustion whereas large contribution from biomass burning (21%) to OC over the Himalayas. In the present study, the mass of PM2.5 particles and its chemistry, BC, NOx, extinction coefficient (bext) along with the meteorological parameters were analyzed over Delhi during winter period (1st December 2011 to 30th March 2012). This is the first attempt to understand the impact of aerosol mass and its chemical composition on the light extinction coefficient over the study region. The main objectives of the present study are: (i) to examine the contribution of mass of atmospheric aerosols, carbonaceous fraction and water-soluble inorganic species in PM2.5; (ii) to estimate the contribution of bext along with absorption and scattering coefficients for their day-time and night-time variations, and (iii) to investigate the relationship between bext and visibility. 2. Sampling site and instrumentations The experimental site Delhi is located in the north western part of the IGP with Thar Desert in the west, which is the single largest contributor to the mineral dust aerosols in the region (Srivastava et al., 2011; Todd et al., 2007). During winter, the occurrence of fog, calm atmospheric conditions and low level boundary inversion are experienced over the station. Samples of aerosols and other pollutants were collected in the premises of Indian Institute of Tropical Meteorology (Branch), located in central part of New Delhi (28.38  N, 77.10  E and w235 m amsl), India. The population density of Delhi is w11,297 km2 in 2011, having approximately 19 million inhabitants spread over 1484 km2 area. It has a sub-tropical

61

climate with extremely hot temperatures during summer (Aprile MayeJune) and moderately cold temperature in winter (DecembereJanuaryeFebruaryeMarch). Samples of PM2.5 were collected by medium volume air sampler (gravimetric: off-line; APM 550, Envirotech Pvt. Ltd., India) once in a week during day-time (1000e1800 h) and night-time (1900e 0700 h). Details of the instrumentation and the chemical analysis technique are given elsewhere (Tiwari et al., 2009; Srivastava et al., 2012a). Further, aerosol mass concentrations (PM2.5 and PM10) were also measured by online samplers (beta attenuation: optical method) from Thermo Andersen, USA, Inc., Series FH 62 C14 (C14 BETA) during the study period. Thermal-optical transmittance based semi-continuous OC-EC instrument (Sunset Lab, USA: Model-4L) was used for the measurement of organic and elemental carbon (Tiwari et al., 2013a). Black carbon, was measured continuously using an Aethalometer (Model AE-31, Magee Scientific Company, Berkeley, CA, USA) with high temporal resolution (5min interval) using quartz fiber filter tape transmission at an 880 nm wavelength (Tiwari et al., 2013b). Chemiluminescence NOe NO2eNOx analyzer (Model 42i Thermo electron Corporation, US) was used for monitoring of NOx. Meteorological parameters (including visibility) during study period were obtained from India Meteorological Department, New Delhi which were recorded near the sampling site (w500 m) in the campus of Indian Agricultural Research Institute; however, rainfall data was collected inside the premises of the institute. The boundary layer mixing depths for Delhi was obtained from the stability time series data available on the NOAA Air Resources Laboratory web server (http://www.arl. noaa.gov/ready.html). The general equation is adopted for conversion of ppb to mg m3 for NOx is as below:

mg m3 ¼ ðppbÞ*ð12:187Þ*ðMÞ=ð273:15 þ CÞ

(1)

Where, M is molecular weight of NOx. An atmospheric pressure of 1 atm is assumed. 3. Results and discussion 3.1. Status of primary aerosols over Delhi during winter Day to day variability of mass concentrations of PM2.5, BC and NO during study period over Delhi are shown in Fig. 1. The monthly mean concentrations of PM2.5, BC and NO measured from different techniques are depicted in Table 1. The seasonal mean mass concentration of PM2.5 was found to be 186.5  149.7 mg m3 with online sampling and 151.7  30.9 mg m3 with off-line sampling during the study period. The mean mass concentration of on-line PM2.5 mass was w23% higher than that of off-line PM2.5, which might be due to (i) difference in sampling protocol for on-line and off-line and (ii) chocking of the pores of the filter media in off-line sampling due to high humidity, especially during foggy nights. Monthly variability analysis shows on-line PM2.5 mass fraction is relatively higher in December (108%) and January (14%) as compared to off-line PM2.5 mass. On the other hand, during February and March, on-line PM2.5 mass fraction is found to be relatively lower by about 36% and 16%, respectively as compared to off-line PM2.5 mass. Results are attributed due to the influence of meteorological parameters, especially boundary layer condition and wind speed. During stable atmospheric condition, the pollutants released by different sources are trapped in lower atmosphere resulting in enhancement of the pollution level, whereas during the unstable conditions, strong horizontal winds disperse the trapped atmospheric pollution levels due to air mass transportation (Dutkiewicz et al., 2009; Srivastava et al., 2012a). In recent studies, Tiwari et al. (2013b) have also reported the impact of boundary

62

S. Tiwari et al. / Atmospheric Environment 92 (2014) 60e68

Fig. 1. Day to day variability of BC (A), NO (B) and PM2.5 (C) during winter period from December 2011 to March 2012.

morning hours as a result of its photolysis. After the sunset, NO2 levels start rising which peak during 01:00 am in the night-time. Badarinath et al. (2009) reported higher column NO2 concentrations during winter, pre-monsoon and post-monsoon seasons and suggested that the agriculture crop residue burning may be one of the strong contributors from the IGP region. In another study over Punjab, Sharma et al. (2010) have also reported very high values of NO2 during October 2007 and suggested that these enhancements were associated with agriculture crop residue burning. Ghude et al. (2008, 2011) have presented major emissions of NO2 over different regions of India which are mostly from the major urban and industrial sectors. Singh and Peshin (2014) reported very high concentrations of gaseous pollutants such as CO and NOx over Delhi due to increased traffic emissions by diesel trucks even at nighttime. Mass concentration of BC varied from 0.4 to 49.6 mg m3 during winter period with an average of 9.6  8.5 mg m3. Higher mass concentrations of BC were also reported by earlier studies over IGP region (Tiwari et al., 2013b; Tripathi et al., 2005, 2006; Safai et al., 2008; Praveen et al., 2012) and suggested the major sources from fossil fuel combustion as well as biomass burning, especially during winter. Reddy and Venkataraman (2002) reported that fossil/ biomass fuels, wood and crop wastes were the primary contributors to BC aerosols over northern part of India, which exhibited high amount of biomass consumption sources. Under the Aerosol Radiative Forcing in India (ARFI) network, a national program by ISROGBP, measurements on mass concentrations of BC were conducted at 33 locations over different regions in India. This study

layer condition at Delhi in a year long study. The average mass concentrations of PM2.5 measured with both on-line and off-line methods was w3e4 times higher than the threshold limit for annual mean mass of PM2.5 level (40 mg m3) stipulated by the Indian National Ambient Air Quality Standards (NAAQS; http:// www.cpcb.nic.in). The value is also much higher in comparison to the annual mean standard of PM2.5 (15 mg m3) stipulated by United State Environmental Protection Agency Standards (USEPA; http:// www.epa.gov/ttn/naaqs/standards) and that by European Union Standards (25 mg m3) (EUS; http://ec.europa.eu/environment/air/ quality/standards.htm). High loading of PM2.5 particles have been reported earlier at other urban locations in the IGP, including Delhi (Tiwari et al., 2009; Begum et al., 2004; Chowbury et al., 2007; Kulshrestha et al., 2009; Ram et al., 2012). The above studies reported that the source of PM2.5 particles over urban locations in IGP is mainly due to man-made activities, biomass burning and transportation of fine particles. Simultaneously, observation of real time concentrations of primary pollutant NOx [nitrous oxide (NO) and nitrogen dioxide (NO2)], which is mostly produced by combustion, show their mean concentrations as 23.8  16.1 ppb [14.4  13.2 and 9.5  5.2 ppb, respectively for NO and NO2]. NO2 is an important pollutant, which plays a crucial role in the photochemical production of ozone in the atmosphere. During the study period, NO2 varied from 2.3 to 45.9 ppb whereas NO varied from 4.8 to 105.3 ppb. The mean NO concentration was 52% higher than that of NO2. The maximum values of NOx were found in the month of December similar to that for PM2.5. During study period, it was observed that NO2 concentration starts to decline during early

Table 1 Monthly mean concentrations of PM2.5 (offline and online), OC, EC (gravimetric and near real time), BC and NO (real time) over Delhi. Months

3

Grav. (mg m December, 11 January, 12 February, 12 March, 12 Mean

157.1 167.3 158.9 146.9 151.7

Gravimetrica

PM2.5

PM2.5 )

3

Opt. (mg m 327.3 190.1 116.7 126.1 186.5

Grav. ¼ gravimetric technique; Opt. ¼ optical technique. a Off-line monitoring (gravimetric technique). b Semi-continuous (optical technique: near real time).

)

3

OC (mg m 63.7 34.0 25.1 39.9 51.1

)

Near real timeb 3

EC (mg m 13.0 8.9 4.6 10.4 10.4

)

OC (mg m 68.6 32.9 24.7 14.8 33.8

3

)

Real time 3

EC (mg m 15.1 7.6 5.4 6.4 8.2

)

BC (mg m3)

NO (ppb)

17.6 8.3 5.1 7.8 9.57

13.7 8.1 7.6 18.2 14.4

S. Tiwari et al. / Atmospheric Environment 92 (2014) 60e68

documented the mass concentrations of BC in different seasons over IGP region and reported much higher BC mass concentrations (>10 mg m3) over IGP as compared to other parts of India. Recently, a report by National Carbonaceous Aerosol Program (NCAP) by the Ministry of Environment & Forests, Government of India has compiled the studies on BC aerosols over India (http://moef.nic.in/ downloads/public-information/Black Carbon Research Initiative). In another study, Nair et al. (2007) have also reported very high BC mass concentrations (20e30 mg m3) during a land campaign conducted over IGP during winter 2004. In the present study, regression analysis showed that BC and NO2 were significantly correlated to each other (r ¼ 0.68), indicating the common combustion sources such as fossil fuel/biomass burning. In Fig. 1, it is seen that in the first week of January 2012, concentrations of all species suddenly dropped. To substantiate this, we have analyzed various meteorological parameters and found occurrences of rain on 6th (1.4 mm), 7th (7.4 mm) and 16th (6 mm) January 2012. The mass concentrations of BC were separated before and after the rainy days. BC mass concentration was found to be 10.4 mg m3 before rain (on 5th January 2012), which was dropped by w54% (4.9 mg m3) after rain (on 6th January 2012). Next three days, it was around 5.5 mg m3 and thereafter it reached its normal level of w12.7 mg m3 on 10th January 2012. In another rainy event on 16th January 2012, similar pattern was seen when the average mass concentration of BC was dropped by w44% (4.8 mg m3) as compared to that on a non-rainy day on 15th January 2012 (8.6 mg m3) and then again reached to its normal condition after three days on 19th January 2012 (10.1 mg m3). These analyses suggest that the BC may recoup its normal values after 3 days interval. The concentrations of organic carbon (OC) and elemental carbon (EC) were also studied during the same period by two techniques (i) off-line (Gravimetric method) and (ii) semi-continuous (near real time: optical method). The monthly mean concentrations of OC and EC measured from different techniques are depicted in Table 1. Their seasonal mean mass concentrations were found to be 51.1  15.2 and 10.4  5.5 mg m3; and 33.8  27.7 and 8.2  6.2 mg m3, respectively during the study period. The ratios of OC to EC in both cases were found in between 4 and 5. Higher ratio indicates the emissions from biomass burning (Andreae and Merlet, 2001); however, relatively lower ratio suggests the dominance of fossil fuel emission. In a recent study, Ram et al. (2012) have reported lower ratios over IGP region with dominance of absorbing type aerosols (EC) over the scattering aerosols (OC). Recent studies over Delhi suggested that the carbonaceous aerosols are, by and large, deficient in OC and abundant in BC (or EC) content, and are mainly originated from the fossil fuel burning (Soni et al., 2010; Tiwari et al., 2013a; Srivastava et al., 2012c). In another study, Ram et al. (2012) have reported relatively higher OC/ EC ratios (8.7  2.3) at Kanpur and suggested that the elevated ratios are mainly derived from the primary emission sources. Further, in another study, Ram et al. (2010) have also reported relatively higher ratio (7.7  3.4) at Manora Peak- a high-altitude station in the foothills of Himalayas and suggested transportation of BC from biomass burning from IGP region. Satsangi et al. (2012) reported OC and EC at Dayalbagh (Agra), India during winter in 2009 with an annual average of OC about 37.4  23.4 mg m3 while EC about 6.3  4.7 mg m3 with the average OC/EC ratio of 5.9. Pavuluri et al. (2011) reported the OC and EC mass concentrations of 9.1 and 6.5 mg m3, respectively at Chennai in southern part of India during winter. Duan et al. (2007) reported that secondary OC concentrations were 10.2e12.8 mg m3, accounting for 36e42% of OC in winter over Guangzhou, China. In another study, Duan et al. (2005) reported higher OC/EC ratio (w5) in Beijing, China during winter, which mostly associated with the secondary organic carbon (SOC) formation; however the lower values of OC/EC ratio (2.8) are

63

associated with the primary sources from coal burning. In view of the above, we have observed a large variability in the mass concentrations of carbonaceous aerosols as well as fine particle during winter period over Delhi. Due to large variability in mass concentrations as well as meteorological parameters during day-time and night-time, the diurnal variability of OC and EC (semi-continuous near real time: optical method) was studied over Delhi during the study period (Fig. 2). The ratio between OC and EC was 4.3, which is nearly similar to that observed with gravimetric technique (4.9). An interesting result was seen in diurnal variability in OC and EC, showing lower concentrations (OC: 22.3 mg m3 and EC: 4.4 mg m3) during 1300e1800 h (when mixing height was w735 m), whereas higher concentrations (OC: 39 mg m3 and EC: 9.6 mg m3) during night-time (1900e0700 h) when mixing height was very low (<100 m). Apart from this, another peak was seen for both OC and EC during the morning between 0900 and 1000 h, which could be due to morning rush-hours (vehicular emissions) and local domestic burning activities. The boundary layer usually become deeper in day-time due to strong turbulent eddies as compared to night-time (Zhang et al., 2012). Boundary layer heights were also plotted in Fig. 2, which clearly indicated the impact of boundary layer during day- and night-time on variation of pollutants. In night-time, the boundary layer was observed below 100 m (mean) whereas in day-time, it was around 500 m. The higher concentrations of OC and EC during night-time are directly influenced by variations in mixing heights. Apart from this, biomass burning during night-time in winter also enhances the concentrations of OC and EC in this region. Present results are consistent with the study reported by Rehman et al. (2011) over IGP which showed biomass burning (burning of wood, crop residue, cow dung etc) as the major source of carbonaceous aerosols over the region. The higher ratio value of OC to EC (5.3  1.9) suggested by Ram and Sarin (2010) is due to biomass burning and lower value (<2) for fossil fuels. These results are consistent with the observed concentration of OC and EC in year 2010e2011 by using optical method at the same location (Tiwari et al., 2013a). Since combustion is the major source of EC, it is often used as tracer of primary OC (Turpin and Huntzicker, 1991). Hence, the relation between OC and EC can be used for qualitative estimation of the origin of carbonaceous particles (Turpin and Huntzicker, 1995). In the present study, significant correlation (r ¼ 0.94) between optical OC and EC is attributed to the dominance of primary sources (vehicular emissions, biomass burning and coal combustion). If the primary sources are dominant, the relative emission rates of EC and OC would be proportional to each other; therefore, the correlation between them will be higher. Also, correlation analysis between OC and EC (off-line) in PM2.5 during winter in day-time and night-time (Fig. 3) were studied and significant correlations were observed, with relatively higher value during night-time (r ¼ 0.91) as compared to day-time (r ¼ 0.74). These results imply that the primary aerosols were higher during night-time whereas the formation of secondary aerosols was higher during day-time. In night-time, large amount of organic compounds are emitted (as primary particulates) from various combustion sources in heavily industrialized region over Delhi; however, during day-time, this may be explained by the impact of motor vehicles exhaust. Day and night variations in mass concentrations of PM2.5, OC and EC (off-line) over Delhi during the study period was also separated and depicted in Fig. 4. Overall, the mass of PM2.5, OC and EC was found to be higher by w26%, 44% and 97%, respectively during day-time as compared to night-time. Monthly analysis showed large variation in all the above components. It is observed from the analysis that the concentrations of PM2.5, OC and EC in the month of January were higher by w23%, 46% and 133%, respectively during night-time as

64

S. Tiwari et al. / Atmospheric Environment 92 (2014) 60e68

Fig. 2. Diurnal variation in mass concentrations of OC and EC as well as mixing height (MH) over Delhi in winter.

Fig. 3. Relationship between OC and EC concentration during winter in Delhi.

compared to day-time, which is mainly due to the impact of local emissions and meteorology. On the other hand, in the month of March, their concentrations were w23% (PM2.5), 67% (OC) and 89% (EC) higher during night-time as compared to day-time. These variation during March may be due to long range transport of pollutants during crops burning episodes in the surrounding Punjab, Haryana regions (Awasthi et al., 2011; Badarinath et al., 2009). 3.2. Estimation of relative contribution to sources of carbonaceous aerosols Coal combustion is a major source of atmospheric particles in urban environment in India and elsewhere due to its low cost.

Moreover, emissions from mobile sources have been growing along with increasing number of vehicles. For example, the number of vehicles in Delhi now exceeded more than 3.1 million. Vehicular exhaust is an important source for ambient particles in Delhi during winter (SAFAR, 2010). In the present study, we tried to investigate the contribution of two major sources of carbonaceous aerosols using OC/EC ratio with the speculation that carbonaceous aerosols are from primary emission over Delhi. The ratios (OC/EC) reached at a lowest value (1.1) at 0060 h on 10 March 2012; which is due to dominant emissions of primary aerosols during morning time and was used as the OC/EC ratio from motor vehicles. The OC/EC ratio reached the highest value (12.1) at 1400 h on 24 December 2011; when few motor vehicles were playing on road giving rise to less

S. Tiwari et al. / Atmospheric Environment 92 (2014) 60e68

65

Fig. 4. Day and night variations in mass concentration of PM2.5, OC and EC over Delhi during winter.

Fig. 5. Source contribution of carbonaceous aerosol in Delhi.

fossil fuel combustion. A value of 12.1 was used as the OC/EC ratio from coal combustion. Further, the source apportionment equation (Zhang et al., 2007) was used in estimation of their contribution, which is given below:

1:1ή þ 12:1ð1  ήÞ ¼ Cavg

(2)

where ή is the fraction of carbonaceous aerosol of motor vehicle origin and (1  ή) is fraction of carbonaceous aerosol of coal

66

S. Tiwari et al. / Atmospheric Environment 92 (2014) 60e68

combustion origin, Cavg is averaged OC/EC ratio in day-time, nighttime, and overall (24 h). The average OC/EC in overall observation period, day-time and night-time were 4.4, 4.9 and 4.0, respectively. We calculated the contribution of carbonaceous aerosols by Equation (2) from coal combustion and motor vehicle (Fig. 5). During winter period over Delhi, the contribution of carbonaceous aerosol in PM2.5 from motor vehicle during 24 h, day-time and night-time sources were accounted for w69%, 90% and 61% respectively, while contribution from coal combustion accounted for only w31%, 10% and 39%, respectively. 3.3. Extinction effects of chemical compositions In order to estimate total light extinction coefficient (bext), an empirical Equation (3), which was developed by Interagency Monitoring of Protected Visual Environment (IMPROVE) by multiplying the concentrations of each major measured chemical constituents as well as mass of atmospheric aerosols by compositionspecific light extinction efficiency and sum of all major ionic species (IMPROVE, 2006, 2011). The equation is given here under:

bext ¼ 3f ðRHÞfAmmonium Sulfate þ Ammonium Nitrateg

Fig. 7. Relative contribution of chemical components to light extinction coefficients in the atmosphere during winter period over Delhi.

þ4½Organic Mass þ 10½Elemental Carbon þ ½Fine Soil þ0:6½Coarse Mass þ 0:161½Nitrogen Dioxide þ 10

(3)

where f (RH) is relative humidity which depends on adjustment factor that illustrates the relationships between RH and scattering efficiencies for nitrate and sulfates. The unit of bext for atmospheric aerosol is Mm1 (1 Mm1 ¼ 106 m1); chemical composition concentrations are in mg m3 and the hygroscopic growth terms [f (RH)] is unit less. In this study, we have estimated the extinction effect of measured chemical species in atmospheric particles, which are causing visibility degradation (Fig. 6). We excluded the contribution of fine soil particles (being small fraction of aerosols) in the estimation of bext. Also, the coarse mode particle was included during estimation of bext due to high mass loading over IGP region (Tiwari et al., 2012). Malm and Day (2001) have presented different f (RH) values in selected relative humidity ranges, which were

Fig. 6. Temporal variations in extinction coefficients of chemical components in the atmosphere and visibility during study period over Delhi. (Amm. Sul. ¼ ammonium sulfate; Amm. Nit. ¼ ammonium nitrate and Visi. ¼ visibility).

incorporated in calculating bext. The mean bext over the station during study period was found to be about 700.0  268.6 Mm1, varied from 313.0 to 1342.4 Mm1. A large variability was seen in bext of measured chemical constituents as well as in mass concentration of aerosols in different months. Higher bext was observed in the months of January (847.1 Mm1) followed by December (822.5 Mm1), February (616.4 Mm1) and March (514.4 Mm1). This large variability is due to metrological effects. In day and night analysis, a significant variability was also seen where bext is w37% higher in night-time (863.4 Mm1) than in day-time (544.5 Mm1), which confirms our assumptions of meteorological impact. The corresponding visibility data was also plotted in Fig. 6, which showed opposite variation and touched bottom during foggy period mostly in the last week of December 2011. Regression analysis between bext and visibility showed significant anticorrelation (r ¼ 0.85). Hence, the higher values of bext represent poor visibility (<0.5 km). The poorest visibility (<0.22 km) was recorded on 29 December 2011 during night-time and the corresponding bext was 1119.8 Mm1 with higher RH (96%). On the basis of estimated bext of chemical constituents as well as coarse mode particles, the contributions of individual components were plotted in Fig. 7. The largest contribution to the light extinction coefficients in the atmosphere was due to OC (46%), followed by EC (24%), coarse mode particles (18%), ammonium sulfate (8%), and ammonium nitrate (4%). As very less contribution of NO2 to the bext was observed (<0.2%), it was not included in contribution analysis. In another study, Jung et al. (2009) have reported the contributions of ammonium sulfate, ammonium nitrate, organic mass by carbon, elemental carbon, and sea-salt to total PM2.5 mass of w36.5%, 5.7%, 27.1%, 7.8%, and 3.7%, respectively at an urban site in the metropolitan area of Guangzhou, China in 2006. On the basis of IMPROVE equation to estimate chemical extinction (bext) in PM2.5, Cao et al. (2012) reported ammonium sulfate [(NH4)2SO4] to be the largest contributor, accounted w40% of bext, followed by organic matter (w24%), ammonium nitrate (NH4NO3: w23%), and elemental carbon (w9%), with minor contributions from soil dust (w3%), and NO2 (w1%). They also suggested that the high sec ondary inorganic aerosols (i.e. SO2 4 and NO3 ) are the main contributors for visual range <5 km. Further, Xu et al. (2012) reported that the major chemical species (OC þ EC þ major ions) at three sampling sites in China, were ranging from w80 to 110 mg m3.

S. Tiwari et al. / Atmospheric Environment 92 (2014) 60e68

Recently, Zhang et al. (2013) have studied the contribution of light extinction coefficient during winter period of 2010 and 2011, which was 39.3% for organics followed by 19.9% for elemental carbon, 16% for ammonium sulfate, 13% for coarse mass and 11.8% for ammonium nitrate. Out of all three, the OC was dominant chemical species accounting for w50% of PM2.5 mass. On the basis of individual study of light extinction due to chemical species and mass of coarse mode particles over Delhi during winter period, it is clearly indicated that the scattering type aerosols are dominating (w76%) than the absorbing type (w24%). To substantiate the above, the absorption coefficient (sabs) and scattering coefficient (sscat) were estimated from the measured BC and bext, respectively. The absorption coefficient was determined from BC data, which were measured by Aethalometer at 880 nm during the same period. Uncertainty in the measured BC data due to multiple scattering in the unloaded fiber filter and particles embedded on the filters is well known (Coen et al., 2010). We have used the procedure suggested by Weingartner et al. (2003) using a constant value for C (2.14) for correction of measured attenuation coefficient by Aethalometer. The mean value of sabs was 71.7  54.6 Mm1 during the study period, with a decreasing order of December 2011 (114.6 Mm1) > January 2012 (76.7 Mm1) > March 2012 (59.2 Mm1) > February 2012 (41.5 Mm1). The estimated sabs values were subtracted from the calculated extinction coefficient (bext) values and the remaining values were considered as scattering coefficient (sscat). The mean value of sscat was 628.4  28.1 Mm1, which were in order of January 2012 (770.4 Mm1) > December 2011 (707.9 Mm1) > February 2012 (574.9 Mm1) > March 2012 (455.2 Mm1). These results indicate that the scattering type aerosols dominated by w89% over the absorbing type aerosols (w11%) at Delhi. The contribution of bext due to absorbing aerosols was lower as compared to the measured chemical constituents, which could be due to coating of absorbing aerosols with scattering types of particles. In another study, Soni et al. (2010) reported nearly similar scattering coefficient (565.6  274.6 Mm1) over Delhi during winter of 2008e2009 at 550 nm. They also estimated the absorption coefficient of 189.7  85.9 Mm1, with lower SSA (0.74). 4. Summary and conclusions In the present study, we have used gravimetric (off-line) and optical (on-line) techniques to carry out mass concentrations of fine particles (PM2.5) and real time measurement of BC and NOx as well as concentrations of OC and EC during the winter season of 2011e 12 over Delhi, India. During study period, the concentrations of PM2.5, BC and NOx were 186.5  149.7 mg m3, 9.6  8.5 mg m3 and 23.8  16.1 ppb respectively. BC mass concentration dropped by w54% due to rain, which took three days time to regain its normal level. The concentrations of OC and EC were found to be 51.1  15.2, 10.4  5.5 mg m3; and 33.8  27.7, 8.2  6.2 mg m3 based on offline and semi-continuous methods, respectively. The ratios of OC to EC in both cases were in between 4 and 5. Correlation analysis between OC and EC (off-line) in PM2.5 during day-time and nighttime were studied and a significant relationship was observed during night-time (r ¼ 0.91) than during day-time (r ¼ 0.74). The source contribution of carbonaceous aerosol in PM2.5 from motor vehicle during 24hrs, day-time and night-time sources were about 69%, 90% and 61%, while contribution from coal combustion accounted for about 31%, 10% and 39%, respectively. The estimated mean values of bext and sabs were 700.03  268.6 and 71.7  54.6 Mm1, respectively during the study period. In day and night-time analysis, a significant variability was observed in bext, which was about 37% higher in night-time (863.4 Mm1) than in

67

day-time (544.5 Mm1). Regression analysis between bext and visibility showed significant anti-correlation (r ¼ 0.85) between them, hence, the higher values of bext represent poor visibility (<0.5 km). The poorest visibility (<0.22 km) was recorded on 29 December 2011 during night-time and the corresponding bext was 1119.8 Mm1 with higher RH value (96%). On the basis of estimated bext of chemical constituents as well as coarse mode particles, the largest contribution was due to organic carbon (46%) of the light extinction coefficients in the atmosphere, followed by elemental carbon (24%), coarse mode particles (18%), ammonium sulfate (8%), and ammonium nitrate (4%). Based on individual study of light extinction due to chemical species and mass of coarse mode particle, scattering type of aerosols were found to be dominant (w76%) than the absorbing type aerosols (w24%) over Delhi. Acknowledgments The authors gratefully thank Prof B. N. Goswami, Director, IITM, Pune and Ministry of Earth Sciences, Government of India for their encouragement for the study and financial supports, respectively. Authors are grateful to the anonymous reviewers for their constructive comments and suggestions to improve the manuscript. References Andreae, M.O., Merlet, P., 2001. Emission of trace gases and aerosols from biomass burning. Global Biogeochemical Cycles 15 (4), 955e966. Awasthi, A., Agarwal, R., Mittal, S.K., Singh, N., Singh, K., Gupta, P.K., 2011. Study of size and mass distribution of particulate matter due to crop residue burning with seasonal variation in rural area of Punjab, India. Journal of Environmental Monitoring 13, 1073e1081. Badarinath, K.V.S., Sharma, A.R., Kharol, S.K., Prasad, V.K., 2009. Variations in CO, O3 and black carbon aerosol mass concentrations associated with planetary boundary layer (PBL) over tropical urban environment in India. Journal of Atmospheric Chemistry 62, 73e86. Begum, B.A., Kim, E., Biswas, S.K., Hopke, P.K., 2004. Investigation of atmospheric aerosol at urban and semi-urban areas in Bangladesh. Atmospheric Environment 38, 3025e3038. Cao, J.J., Wang, Q.Y., Chow, J.C., Watson, J.G., Tie, X.X., Shen, Z.X., Wang, P., An, Z.S., 2012. Impacts of aerosol compositions on visibility impairment in Xi’an, China. Atmospheric Environment 59, 559e656. Chowbury, Z., Zheng, M., Schauer, J.J., Sheesley, R.J., Salmon, L.G., Cass, G.R., Russell, A.G., 2007. Speciation of ambient fine organic carbon particles and source apportionment of PM2.5 in Indian cities. Journal of Geophysical Research 112. http://dx.doi.org/10.1029/2007JD008386. Coen, M.C., Weingartner, E., Apituley, A., Ceburnis, D., Fierz-Schmidhauser, R., Flentje, H., Henzing, J.S., Jennings, S.G., Moerman, M., Petzold, A., Schmid, O., Baltensperger, U., 2010. Minimizing light absorption measurement artifacts of the Aethalometer: evaluation of five correction algorithms. Atmospheric Measurement Techniques 3, 457e474. Duan, F., He, K., Ma, Y., Jia, Y., Yang, F., Lei, Y., Tanaka, S., Okuta, T., 2005. Characteristics of carbonaceous aerosols in Beijing, China. Chemosphere 60, 355e364. Duan, J., Tan, J., Cheng, D., Bi, X., Deng, W., Sheng, G., Fu, J., Wong, M.H., 2007. Sources and characteristics of carbonaceous aerosol in two largest cities in Pearl River Delta Region, China. Atmospheric Environment 41, 2895e2903. Dutkiewicz, V.A., Alvi, S., Ghauri, B.M., Choudhary, M.I., Husain, L., 2009. Black carbon aerosols in urban air in South Asia. Atmospheric Environment 43, 1737e 1744. Gautam, R., Hsu, N.C., Kafatos, M., Tsay, S.-C., 2007. Influences of winter haze on fog/ low cloud over the Indo-Gangetic plains. Journal of Geophysical Research 112, D05207. http://dx.doi.org/10.1029/2005JD007036. Ghude, S.D., Fadnavis, S., Beig, G., Polade, S.D., Van der, A.R.J., 2008. Detection of surface emission hot spots, trends, and seasonal cycle from satellite-retrieved NO2 over India. Journal of Geophysical Research 113 (D20305), 1. Ghude, S.D., Kulkarni, S.H., Kulkarni, P.S., Kanawade, V.P., Fadnavis, S., Pokhrel, S., Jena, C., Beig, G., Bortoli, D., 2011. Anomalous low tropospheric column ozone over Eastern India during severe drought event of monsoon 2002: a case study. Environmental Science and Pollution Research 18, 1442e1455. Gray, H.A., Cass, G.R., 1998. Source contributions to atmospheric fine carbon particle concentrations. Atmospheric Environment 32 (22), 3805e3825. Hussain, L., Dutkiewicz, V.A., Khan, A.J., Ghauri, B.M., 2007. Characterization of carbonaceous aerosols in urban air. Atmospheric Environment 41, 6872e6883. IMPROVE, 2006. Spatial and Seasonal Patterns and Temporal Variability of Haze and Its Constituents in the United States, Report V. IMPROVE, 2011. Spatial and Seasonal Patterns and Temporal Variability of Haze and Its Constituents in the United States, Report IV.

68

S. Tiwari et al. / Atmospheric Environment 92 (2014) 60e68

Jacobson, M.Z., 2004. Climate response of fossil fuel and biofuel soot, accounting for soot’s feedback to snow and sea ice albedo and emissivity. Journal of Geophysical Research 109, D21201. http://dx.doi.org/10.1029/2004JD004945. Jacobson, M.Z., 2006. Effects of externally-through-internally-mixed soot inclusions within clouds and precipitation on global climate. Journal of Physical Chemistry 110, 6860e6873. Jeong, C.H., Hopke, P.K., Kim, E., Lee, D.W., 2004. The comparison between thermale optical transmittance elemental carbon and Aethalometer black carbon measured at multiple monitoring sites. Atmospheric Environment 38, 5193e 5204. Jung, J., Lee, H., Kim, Y.J., Liu, X., Zhang, Y., Gu, J., Fan, S., 2009. Aerosol chemistry and the effect of aerosol water content on visibility impairment and radiative forcing in Guangzhou during the 2006 Pearl River Delta campaign. Journal of Environmental Management 90 (11), 3231e3244. Kulshrestha, A., Satsangi, P.G., Masih, J., Taneja, A., 2009. Metal concentration in PM (2.5) and PM (10) particles and seasonal variations in urban and rural environment of Agra, India. Science of the Total Environment 407, 6196e 6204. Lal, D.M., Patil, S.D., Singh, H.N., Ghude, S.D., Tiwari, S., Srivastava, M.K., 2013. Influence of aerosol on clouds over the Indo-Gangetic Plain, India. Climate Dynamics 41, 601e612. Malm, W.C., Day, D.E., 2001. Estimates of aerosol species scattering characteristics as a function of relative humidity. Atmospheric Environment 35, 2845e2860. Nair, V.S., Moorthy, K.K., Alappattu, D.P., Kunhikrishnan, P.K., George, S.K., Nair, P.R., Babu, S.S., Abish, B., Satheesh, S.K., Tripathi, S.N., Niranjan, K., Madhavan, B.L., Srikant, V., Dutt, C.B.S., Badarinath, K.V.S., Reddy, R.R., 2007. Wintertime aerosol characteristics over the Indo-Gangetic Plain (IGP): impacts of local boundary layer processes and long-range Transport. Journal of Geophysical Research 112, D13205. http://dx.doi.org/10.1029/2006JD008099. Pavuluri, C.M., Kawamura, K., Aggarwal, S.G., Swaminathan, T., 2011. Characteristics, seasonality and sources of carbonaceous and ionic components in the tropical aerosols from Indian region. Atmospheric Chemistry and Physics 11, 8215e 8230. Praveen, P.S., Ahmed, T., Kar, A., Rehman, I.H., Ramanathan, V., 2012. Link between local scale BC emissions in the Indo-Gangetic Plains and large scale atmospheric solar absorption. Atmospheric Chemistry and Physics 12, 1173e1187. Ram, K., Sarin, M.M., 2010. Spatio-temporal variability in atmospheric abundances of EC, OC and WSOC over northern India. Journal of Aerosol Science 41, 88e98. Ram, K., Sarin, M.M., Hegde, P., 2010. Long-term record of aerosol optical properties and chemical composition from a high-altitude site (Manora Peak) in Central Himalaya. Atmospheric Chemistry and Physics 10 (23), 11791e11803. Ram, K., Sarin, M.M., Tripathi, S.N., 2012. Temporal trends in atmospheric PM2.5, PM10, elemental carbon, organic carbon, water-soluble organic carbon, and optical properties: impact of biomass burning emissions in the Indo-Gangetic Plain. Environmental Science & Technology 46, 686e695. Reddy, M.S., Venkataraman, C., 2002. Inventory of aerosol and sulphur dioxide emissions from India. Part II-biomass combustion. Atmospheric Environment 36, 699e712. Rehman, H., Ahmed, T., Praveen, P.S., Kar, A., Ramanathan, V., 2011. Black carbon emissions from biomass and fossil fuels in rural India. Atmospheric Chemistry and Physics 11, 7289e7299. Rogge, W.F., Hildemann, L.M., Mazurek, M.A., Cass, G.R., Simoneit, B.R.T., 1998. Sources of fine organic aerosol, Pine, oak and synthetic log combustion in residential fireplaces. Environmental Science & Technology 32 (1), 13e22. Safai, P.D., Kewat, S., Pandithurai, G., Praveen, P.S., Ali, K., Tiwari, S., Rao, P.S.P., Budhawant, K.B., Saha, S.K., Devara, P.C.S., 2008. Aerosol characteristics during winter fog at Agra, North India. Journal of Atmospheric Chemistry 61, 101e118. SAFAR, 2010. (System for Air Quality forecasting & Research): a Special Report Emission Inventory for National Capital Region Delhi Ministry of Earth Sciences. Government of India. http://safar.tropmet.res.in. Satsangi, A., Pachauri, P., Singla, V., Lakhani, A., Kumari, K.M., 2012. Organic and elemental carbon aerosols at a suburban site. Atmospheric Research 113, 13e21. Sharma, A.R., Kharol, S.K., Badarinath, K.V.S., Singh, D., 2010. Impact of agriculture crop residue burning on atmospheric aerosol loading e a study over Punjab State, India. Annals of Geophysics 28, 367e379. Sheesley, R.J., Schauer, J.J., Chowdhury, Z., Cass, G.R., Simoneit, B.R.T., 2003. Characterization of organic aerosols emitted from the combustion of biomass indigenous to South Asia. Journal of Geophysical Research 108 (D9), 4285. http://dx.doi.org/10.1029/2002JD002981. Singh, T., Khillare, P.S., Shridhar, V., Agarwal, T., 2008. Visibility impairing aerosols in the urban atmosphere of Delhi. Environmental Monitoring and Assessment 141, 67e77. Singh, S., Peshin, S.K., 2014. Air pollution scenario over Delhi city. Environment and Sustainable Development, 77e85. Soni, K., Singh, S., Bano, T., Tanwar, R.S., Nath, S., Arya, B.C., 2010. Variations in single scattering albedo and Angstrom absorption exponent during different seasons at Delhi, India. Atmospheric Environment 44, 4355e4363.

Srivastava, A.K., Tiwari, S., Devara, P.C.S., Bisht, D.S., Srivastava, M.K., Tripathi, S.N., Goloub, P., Holben, B.N., 2011. Pre-monsoon aerosol characteristics over the Indo-Gangetic Basin: implications to climatic impact. Annals of Geophysics 29, 789e804. Srivastava, A.,K., Singh, S., Tiwari, S., Kanawade, V.P., Bisht, D.S., 2012a. Variation between near-surface and columnar aerosol characteristics during winters and summers at a station in the Indo-Gangetic Basin. Journal of Atmospheric and Solar-Terrestrial Physics 77, 57e66. Srivastava, A.K., Singh, S., Tiwari, S., Bisht, D.S., 2012b. Contribution of anthropogenic aerosols in direct radiative forcing and atmospheric heating rate over Delhi in the Indo-Gangetic Basin. Environmental Science and Pollution Research 19, 1144e1158. Srivastava, A.K., Tripathi, S.N., Dey, S., Kanawade, V.P., Tiwari, S., 2012c. Inferring aerosol types over the Indo-Gangetic Basin from ground based sunphotometer measurements. Atmospheric Research 109-110, 64e75. Srivastava, A.K., Bisht, D.S., Ram, K., Tiwari, S., Srivastava, M.K., 2014. Characterization of carbonaceous aerosols over Delhi in Ganga basin: Seasonal variability and possible sources. Environmental Science and Pollution Research (in press). Stone, E.A., Schauer, J.J., Pradhan, B.B., Dangol, P.M., Habib, G., Venkataraman, C., Ramanathan, V., 2010. Characterization of emissions from South Asian biofuels and application to source apportionment of carbonaceous aerosol in the Himalayas. Journal of Geophysical Research 115, D06301. http://dx.doi.org/ 10.1029/2009JD011881. Tiwari, S., Srivastava, A.K., Bisht, D.S., Safai, P.D., 2013a. Assessment of carbonaceous aerosol over Delhi in the Indo-Gangetic Basin: characterization, sources and temporal variability. Natural Hazards 65, 1745e1764. Tiwari, S., Srivastava, A.K., Bisht, D.S., Parmita, P., Srivastava, M.K., Attri, S.D., 2013b. Diurnal and seasonal variations of black carbon and PM2.5 over New Delhi, India: influence of meteorology. Atmospheric Research 125-126, 50e62. Tiwari, S., Chate, D.M., Srivastava, M.K., Safai, P.D., Srivastava, A.K., Bisht, D.S., Padmanabhamurty, B., 2012. Statistical evaluation of PM10 and distribution of PM1, PM2.5, and PM10 in ambient air due to extreme fireworks episodes (Diwali festivals) in megacity Delhi. Natural Hazards 61 (2), 521e531. Tiwari, S., Srivastava, A.K., Bisht, D.S., Bano, T., Singh, S., Behura, S., Srivastava, M.K., Chate, D.M., Padmanabhamurty, P., 2009. Black carbon and chemical characteristics of PM10 and PM2.5 at an urban site of North India. Journal of Atmospheric Chemistry 62, 193e209. Todd, M.C., Washington, R., Martins, J.V., Dubovik, O., Lizcano, G., M’Bainayel, S., Engelstaedter, S., 2007. Mineral dust emission from the Bodélé Depression, northern Chad, during BoDEx 2005. Journal of Geophysical Research 112, D06207. http://dx.doi.org/10.1029/2006JD007170. Tripathi, S.N., Dey, S., Tare, V., Satheesh, S.K., 2005. Aerosol black carbon radiative forcing at an industrial city in northern India. Geophysical Research Letters 32, L08802. Tripathi, S.N., Tare, V., Chinnam, N., Srivastava, A.K., Dey, S., Agarwal, A., Kishore, S., Lal, R.B., Manar, M., Kanawade, V.P., Chauhan, S.S.S., Sharma, M., Reddy, R.R., Gopal, K.R., Narasimhulu, K., Reddy, L.S.S., Gupta, S., Lal, S., 2006. Measurements of atmospheric parameters during Indian Space Research Organization Geosphere Biosphere Programme Land Campaign II at a typical location in the Ganga Basin: 1. Physical and optical properties. Journal of Geophysical Research 111, D23209. http://dx.doi.org/10.1029/2006JD007278. Turpin, B.J., Huntzicker, J.J., 1991. Secondary formation of organic aerosol in the Los Angeles basin: a descriptive analysis of organic and elemental carbon concentrations. Atmospheric Environment 25A, 207e213. Turpin, B.J., Huntzicker, J.J., 1995. Identification of secondary organic aerosol episodes and quantification of primary and secondary organic aerosol concentrations during SCAQS. Atmospheric Environment 29, 3527e3544. Venkataraman, C.G., Habib, A., Eiguren, F., Miguel, A.H., Friedlander, S.K., 2005. Residential biofuels in south Asia: carbonaceous aerosol emissions and climate impacts. Science 307, 1454e1456. Weingartner, E., Saathoff, H., Schnaiter, M., Streit, N., Bitnar, B., Baltensperger, U., 2003. Absorption of light by soot particles: determination of the absorption coefficient by means of aethalometers. Journal of Aerosol Science 34 (10), 1445e1463. Xu, L., Chen, X., Chen, J., Zhang, F., He, C., Zhao, J., Yin, L., 2012. Seasonal variations and chemical compositions of PM2.5 aerosol in the urban area of Fuzhou, China. Atmospheric Research 104-105, 264e272. Zhang, F., Chen, J., Qiu, T., Yin, L., Chen, X., Yu, J., 2013. Pollution characteristics of PM2.5 during a typical haze episode in Xiamen, China. Atmospheric and Climate Sciences 3, 427e439. Zhang, R., Cao, J., Lee, S., Shen, Z., Ho, K., 2007. Carbonaceous aerosols in PM10 and pollution gases in winter in Beijing. Journal of Environmental Sciences 19, 564e 571. Zhang, R., Tao, Jun, Ho, K.F., Shen, Zhenxing, Gehui, W., Junji, C., Suixin, Liu, Zhang, L., Lee, S.C., 2012. Characterization of atmospheric organic and elemental carbon of PM2.5 in a typical semi-arid area of Northeastern China. Aerosol and Air Quality Research 12, 792e802.