Effect of Southwest Monsoon Withdrawal on Mass Loading and

Effect of Southwest Monsoon Withdrawal on Mass Loading and Chemical Characteristics of Aerosols in an Urban City over the Indo-Gangetic Basin. Sarwar ...
1 downloads 7 Views 1MB Size
Subscriber access provided by the Henry Madden Library | California State University, Fresno

Article

Effect of Southwest Monsoon Withdrawal on Mass Loading and Chemical Characteristics of Aerosols in an Urban City over the Indo-Gangetic Basin Sarwar Nizam, and Indra Sekhar Sen ACS Earth Space Chem., Just Accepted Manuscript • DOI: 10.1021/ acsearthspacechem.7b00140 • Publication Date (Web): 21 Feb 2018 Downloaded from http://pubs.acs.org on February 26, 2018

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

ACS Earth and Space Chemistry is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 25 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Earth and Space Chemistry

1 2

Effect of Southwest Monsoon Withdrawal on Mass Loading and Chemical Characteristics of Aerosols in an Urban City over the Indo-Gangetic Basin

3

Sarwar Nizam1 & Indra S. Sen1*

4 5

1

Department of Earth Sciences, Indian Institute of Technology Kanpur, Kanpur, UP 208016, India

6 7

*To whom correspondence should be addressed: IIT Kanpur, Department of Earth Sciences, WLE 201, phone: +91-(0512) 6796440, fax: +91-(0512) 6797436, E-mail: [email protected]

8 9



ABSTRACT

10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29

The southwest monsoon rainfall not only provides water and food security over the IndoGangetic Basin, it also plays an important role in reducing atmospheric pollution by removing ambient partciles via wet deposition processes. In addition to rainfall, aerosol loading and its removal from ambient air is also governed by other meteorological parameters such as temperature, humidity, wind speed, and wind direction. In order to understand the effect of southwest monsoon withdrawal on aerosol loading over the Indo-Gangetic basin, airborne particles (PM10 size fraction) and meteorological parameters including temperature, humidity, rainfall, wind speed, and wind direction data were collected between July and October 2015 at Kanpur, which is a large industrial city in the central part of the Indo-Gangetic Basin. The study shows that withdrawal of the southwest monsoon since July 2015 increased the aerosol loading in the ambient air by up to 28, 43 and 152% during August, September, and October respectively. The aerosol loading exceeded the ambient Indian National Air Quality Standard (NAAQS) limit of 100 µg/m-3 just within three months. In addition to increased aerosol mass loading, concentration of heavy metals (Cr, Ni, Cu and Cd) in the aerosols also increased with monsoon withdrawal. The only heavy metal that did not show an increasing trend was Pb, which indicates that Pb is either coming from local source(s) or that Pb was not efficiently scavenged by wet deposition processes. In general, Cd, Pb, and Cu concentrations were 10-1500 times higher when compared to upper continental crust and were mostly derived from coal burning products. The study shows that southwest monsoon strongly influence the physiochemical properties of aerosols over the Indo-Gangetic basin.

30 31

Keywords: (Southwest monsoon, PM10 mass loading, Aerosols, Indo-Gangetic Basin, Physiochemical properties of aerosols)

Page | 1 ACS Paragon Plus Environment

ACS Earth and Space Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

32

Page 2 of 25

1. INTRODUCTION

33

The southwest monsoon provides water and food security to a billion people in the Indo-

34

Gangetic Basin (IGB). In addition to food and water security, it also plays an important role in

35

reducing the concentrations of particulate matter suspended in the atmosphere.1,2 Concentrations

36

of suspended particulate matter (PM) or aerosols have a strong connection to Earth’s climate and

37

a direct connection to human health.3,4 For example, aerosols can directly affect Earth’s climate

38

by reflecting, scattering, and absorbing the solar radiation, as well as indirectly by controlling

39

cloud formation processes and perturbing the radiation budget.5,6,7 Similarly, aerosol-induced

40

forcing can also be one of the root causes of land-sea surface temperature variation, which in turn

41

is related to reduced precipitation or weakening of the Indian summer monsoon.8,9 These direct

42

and indirect radiative effects of aerosols are a function of aerosol concentration, size, and

43

chemical composition.10 Additionally, aerosols are considered to be one of the most important

44

atmospheric pollutants. On an average, approximately 3% of cardiopulmonary and 25% of lung

45

cancer deaths are attributed to PM pollution globally.11 In the mega-cities (populations >10

46

million) of the IGB such as Delhi and Kolkata, heavy metal associated PM pollution is also an

47

emerging major concern,12,13 as elevated concentrations of atmospheric particulate matter were

48

linked to increased morbidity and mortality rates in the region.14,15

49

Monsoonal rainfall provides relief from PM pollution as ambient particles are removed

50

from the atmosphere via several wet deposition processes.16,17 The scavenging of atmospheric

51

aerosols by precipitation is one of the most important mechanisms that results in a drastic

52

reduction in ambient atmospheric aerosols.18,19 The effect of the monsoon on aerosol loading has

53

been previously recorded. For example, Zhang et al.20 applied a global three dimensional

54

Goddard Earth Observing System (GEOS) chemical transport model (GEOS-Chem) propelled by

55

NASA/GEOS-4 assimilated meteorological data and quantified the impacts of East Asian

56

summer monsoon on seasonal and inter-annual variations of aerosols over eastern China. Their

57

analysis showed that the Asian summer monsoon reduced PM2.5 levels up to 50-70% largely due

58

to wet deposition processes. Similarly, Hyvarinen et al.2 investigated the effect of the summer

59

monsoon on the particulate matter concentrations at two measurement stations: Gaul Pahari and

60

Mukteshwar in Northern India. They found that the summer monsoon was able to reduce 55-70%

61

of the aerosol loading at both sites largely due to wet deposition processes. Since monsoonal

62

rainfall provides relief from PM pollution, it is important to understand the causal relationship

63

between aerosols and the southwest monsoon over the IGB, as well as its impact on aerosol

64

chemistry. In addition to rainfall, aerosol removal and its loading in the atmosphere is also

Page | 2 ACS Paragon Plus Environment

Page 3 of 25 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Earth and Space Chemistry

65

governed by other meterological parameters such as temperature, humidity, windspeed, and wind

66

direction. In this study, we report on the concentration of particulate matter having an

67

aerodynamic diameter less than 10µm (PM10) and their heavy metal concentrations collected at

68

Kanpur, a large industrial city in the central part of the IGB. The main objective of the study is to

69

understand the effect of southwest monsoon withdrawal from the Indo-Gangetic Basin on the

70

physiochemical properties of PM10. The other objective is to investigate the magnitude and

71

source(s) of heavy metal concentrations in Kanpur, which is likely to be significantly

72

contaminated by anthropogenic activities.

73

2. MATERIALS AND METHODS

74

2.1. Site and Sampling Details

75

Sampling was carried out in Kanpur (latitude: 26°30’47.69” N; longitude: 80°13’56.39”

76

E) using high-volume (1000 L/min) atmospheric aerosol samplers (Envirotech PM10 sampler,

77

model APM 460 DXNL) between July 2015 and October 2015. The aerosol sampler was placed

78

on the roof of a 15 m tall Environmental Science and Engineering building inside the Indian

79

Institute of Technology Kanpur (IITK) campus. The IITK is located ~16 km west of the Kanpur

80

city center and ~5 km north of a coal-fired thermal power plant (Panki Thermal Power Station).

81

The campus itself has very limited construction, commercial, and industrial activities. The site is

82

located about 1 km west of a national highway (NH-91), with a total traffic volume of

83

approximately 800 vehicles per hour. Samples were collected once a week and the sampler was

84

opereated for 24 hours during the waning phase of monsoon period (July to September). Whereas

85

samples were collected twice a week and the sampler was opereated in daytime for 12 hours

86

during the post-monsoon period (October). We have reduced the sampling duration in October

87

because the aerosol loading in October was much higher than the monsoon period, and as a result,

88

our instrument was malfuncting since the filter papers were getting clogged. Therefore

89

uncertainties related to different sampling duration should be kept in mind while interpreting our

90

data. A total of 21 samples were collected and the dates of the samples (Figure 1a). It is

91

noteworthy to mention that all the samples were collected during non-rainy days.

92

The airborne particulate matter was collected on 20 x 25 cm quartz fiber filter sheets

93

(Whatman™ EPM 2000). Quartz fibre filters were selected because of their chemical inertness

94

and low affinity for moisture.21 The filters were precombusted at 550°C to remove absorbed

95

moisture from the filter before they were loaded in the aerosol sampler. After collection, the filter

96

papers were postcombusted at 550°C before they were weighed gravimetrically. The final mass

Page | 3 ACS Paragon Plus Environment

ACS Earth and Space Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 25

97

of the aerosol-laden filter was recorded only when repeated post-conditioning weights were

98

identical. Following gravimetric analysis, the filters were pulverized using an agate grinding set

99

in a vibrating mill at IITK.

100

2.2. Trace Metal Analysis

101

Metal concentration analyses were performed at the Department of Earth Sciences of

102

IITK. Briefly, approximately 0.3 g of sample powder was leached in pre-cleaned PTFE vials at 95

103

± 5°C using HNO3 and H2O2 mixtures following the U.S. EPA protocol 3050B.22 This leaching

104

procedure mostly releases metals from silicate particulates. We did not carry out a full digestion

105

procedure using hydrofluoric acid, since the main focus of the study was to determine the metals

106

that are adsorbed on the silicate particulates i.e. the portion that can be washed out during rainfall

107

events. Three blank filters were also digested following the same procedures. The blank filters

108

were analyzed to quantify the total procedural blank, whereas a reference material, Brush Creek

109

Shale (SBC-1) from United States Geological Survey (USGS) was analyzed to assess the

110

accuracy of the analysis. Since a reference material with a matrix identical to that of our aerosol

111

samples was unavailable, all the samples and standards were spiked with 5 ppb In solution as an

112

internal standard.

113

The trace element analyses were performed on an inductively-coupled plasma mass

114

spectrometer (Thermo Scientific™ iCAP Q ICP-MS). Concentrations were determined using a

115

multi-element standard solution (High Purity Standards, Fluka) diluted to appropriate

116

concentrations to generate a 6-point calibration curve, and the instrument was run both in

117

standard and He Kinetic Energy Discrimination mode to optimize the separation of measured

118

isotopes from polyatomic interferences and improve detection limits. The final concentrations

119

were blank-corrected using the average blank filter concentrations and matrix effects were

120

corrected through In normalization. Average blank corrections were less than 5% for Cd and Pb

121

and less than 10% for Cr, Ni, and Cu. Blank corrections for REE concentrations were on the

122

order of 25-45%. The higher blank corrections for REE were expected since REE predominantly

123

reside in the silicate minerals that were not quantitatively dissolved in our method. Blank-

124

corrected concentrations of each element were converted to mass m-3 of air (ngm-3) by dividing

125

the total mass collected on the filter by the total volume of air pumped through the filters. The

126

measured heavy metal and REE concentration of SBC-1 agrees well with the certified values

127

(Supporting Information, Table S1 and Table S2).

128

2.3. Meteorological Data Acquisition

Page | 4 ACS Paragon Plus Environment

Page 5 of 25 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Earth and Space Chemistry

129

Meteorological data was obtained from a weather station located inside the IITK campus.

130

The meteorological data that were included were temperature, humidity, wind speed, wind

131

direction, total rainfall, and number of rain events (Figure 1 and Table 1). The meteorological

132

data presented here is a part of the INCOMPASS (Interaction of Convective Organization and

133

Monsoon Precipitation, Atmosphere, Surface, and Sea) project that is jointly funded by Ministry

134

of Earth Sciences (MoES) and National Environmental Research Council (NERC), UK. Prof.

135

S.N.Tripathi shared the data on personal communication. Table 1 reports the variability of

136

temperature, humidity, rainfall, wind speed, wind direction, and number of rainy days in July,

137

August, Spetember, and October corresponding to 2015.

138

2.4. Enrichment Factor and Aerosol Loading Calculations

139

Enrichment Factor (EF) were calculated with respect to the upper continental crust, and

140

as a proxy for upper continental crust Al, Si, Ca, Fe, Mn, as well as La and Sc concentration of

141

crust can be used. Here we have focused on Sc concentration of upper continental crust23 since Sc

142

is non-volatile and has virtually no commercial or industrial uses and therefore can be used as a

143

representative of a crustal element.23,24,25,26 To crosscheck the results, EF were also calculated

144

using La as a normalizing reference element. EF was calculated using the following equation: EFX=

145 146 147 148 149

‫܆‬/‫܋܁‬aerosol ‫܆‬/‫܋܁‬crust

Where X/Sc is the concentration ratio of an element to Sc. To quantify the effect of monsoon withdrawal on aerosol loading, the rate of increase of aerosol loading was calculated using the equation of Zhang et al.20: Aerosol increase proportion =

[୔୑భబ ]ౣ౟ ି[୔୑భబ ]౨ౣ [୔୑భబ ]౨ౣ

× 100

150

Where [PM10]rm represents the monthly mean concentration of PM10 for the reference month

151

(July), while [PM10]mi represents the monthly mean concentration of PM10 (for the month of

152

interest, i.e. August, September, or October). The month of July was chosen as the reference

153

month, as July had the maximum number of rain days or rain events and received the maxium

154

amount of rainfall as well. Therefore, percent increase in aerosol loading in the troposphere was

155

calculated using aerosol loading in July as the baseline.

156

3. RESULTS AND DISCUSSION

157

3.1. Meteorological Characteristics and PM10 concentration

Page | 5 ACS Paragon Plus Environment

ACS Earth and Space Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

158

Page 6 of 25

The concentration of airborne particulate matter in the ambient air is partly governed by

159

meteorological parameters.27

160

rainfall, and wind direction data set were plotted against PM10 concentrations to understand how

161

micro-meteorology would affect the ambient aerosol loading (Figure 1). In general, a substantial

162

(22-35°C) temeparature variability was observed during the sampling period between July and

163

October 2015. Wind speed, wind direction, and humidity on the other hand did not show much

164

variations. For example, the average humidity for the months of July, August, September, and

165

October was 75 ± 6%, 78 ± 6%, 74 ± 6%, and 73 ± 6%, respectively (Table 1). To evaluate the

166

qualitative effect of meteorological parameters on PM10 concentration, Pearson’s coefficient of

167

correlation was applied to study the relationship between meterological parameters and PM10

168

concentrations. Pearson’s coefficient of correlation test was conducted between daily average

169

meteorological parameters and PM10 concentrations using statistical software SPSS (Statistical

170

Package for Social Sciences) of version 19.0. The test revealed that temperature, humidity and

171

wind speed has negative correlations with PM10 concentrations, though the strength of

172

correlations were not high (Table 2). Negetive correlation between temperature and PM10

173

concentrations indicate that high temperature does not favor the resuspension of fine particulate

174

matter, whereas humidity has congregated airborne particles to obtain mass and further settle

175

down. It is noteworthy to mention that rain fall data was not incorporated in statistical analysis as

176

sampling was done on non rainy days. We would like to mention that the individual effects of

177

temperature, humidity, and wind speed on aerosol loading could not be quantiatvely assessed in

178

this study.

Therefore, daily average temperature, humidity, windspeed,

179

The diurnal and monthly average PM10 aerosol concentrations were plotted in Figure 1a.

180

The average PM10 concentration at Kanpur during monsoon and post-monsoon period were 49 ±

181

14 (n = 12, 1 S.D.) and 99 ± 20 (n = 9, 1 S.D.) µg m-3 respectively (Figure 1a). A significant

182

increase in PM10 concentration was observed as the monsoon withdraws from the IGB. Figure 1a

183

shows that the PM10 loading was minimal during the month of July, which was warmest, most

184

humid, had higher wind speed and received maximum rainfall in 2015 when compared to post-

185

monsoon period (Figure 1a,b). The average monthly PM10 mass concentrations during July,

186

August, September and October were 39 ± 14 (n = 4, 1 S.D.), 50 ± 8 (n = 5, 1 S.D.), 59 ± 19 (n =

187

3, 1 S.D.) and 99 ± 20 (n = 9, 1 S.D.) µg m-3 respectively, and aerosol loading gradually

188

increased with monsoon withdrawal over the IGB. It is noteworthy to mention that July had the

189

highest number of rainy days (12) when compared to other months (Table 1), and the last 2

190

samples of July had the lowest PM10 levels. The consistent increase in PM10 levels were only

Page | 6 ACS Paragon Plus Environment

Page 7 of 25 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Earth and Space Chemistry

191

observed after mid-September, when there were no significant rain events, and a stable

192

atmospheric condition with relatively low humidity and temperature was acheived. In general, we

193

can conclude that July showed the lowest aerosol loadings (LALDs), followed by moderate

194

aerosol loadings (MALDs) in August and September, and October was characterized by the

195

highest aerosol loadings (HALDs) (Figure 1a).

196

The reported aerosol mass concentration ranges are similar to previously reported aerosol

197

loading over the IGB. For example, average PM10 concentration during October (99 ± 20 µg m-3)

198

was similar to the previous values of 80 and 101 µg m-3, reported by Sharma and Maloo28 and

199

Ghosh et al.29 It is noteworthy that such direct comparison should be used with caution, since the

200

sampling periods in previous studies were different from ours. We further compared our weekly

201

and bi-weekly ambient PM10 level with daily PM10 data from Uttar Pradesh Pollution Control

202

Board (UPPCB).30 The average monthly PM10 concentration in July, August, September, and

203

October were 39, 50, 59 and 99 µg m-3 respectively in this study, while UPCCB data were 31,

204

29, 35 and 108 µg m-3 respectively. From our data it is also apparant that monsoon rainfall and

205

changes in the meterological parameters reduced the PM10 mass concentration levels in Kanpur

206

air below the Indian National Air Quality Standard (NAAQS) limit of 100 µg/m3.

207

3.2 Trace Element Systematics

208

The trace metal content of the aerosol-laden filter papers are plotted in Figure 2. The

209

absolute heavy metal concentrations (Cr, Ni, Pb, Cu and Cd) in Kanpur vary by a factor of 3 or

210

less. The Cr, Ni, Pb, Cu and Cd concentrations during monsoon season vary in the range of 1.9-

211

13.9, 5-28.5, 48.2-2965.6, 8.9-190.3 and 1.4-29.1 ng m-3 respectively. The Cr, Ni, Pb, Cu and Cd

212

concentrations during post-monsoon season vary between 5.8-27.5, 16.4-55.8, 71.5-1476.6, 37.8-

213

77.9, and 4.5-106.5 ng m-3 respectively. (Supplementry Section, Table S1). Therefore, during the

214

post-monsoon season the Cr, Ni, Cu and Cd concentrations increased significantly. The only

215

heavy metal that showed consistently high concentration throughout the study period was Pb.

216

This indicates that Pb may be derive from local coal emission sources, as previously indicated by

217

Pb isotopic studies of aerosols collected over Kanpur.32 As a result, even if monsoon rainfall

218

removes Pb from the atmosphere, it is quickly replaced by emission from local sources such as

219

the coal-fired thermal power plant (Panki Thermal Power Station) that is located ~5 km south of

220

the sampling site and Figure 1c shows that air masses are coming from southernly direction. It is

221

also likely that Pb is not removed by the monsoon rains. Based on our available data, the behavior

222

of aerosol Pb cannot be definitely ascertained. The Rare Earth Eelement (REE) concentrations

223

vary between 0.57-3.56, 1.04-6.5, 0.44-2.73, 0.12-0.79, 0.06-0.42, 0.12-0.79, 0.05-0.35, 0.09-

Page | 7 ACS Paragon Plus Environment

ACS Earth and Space Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 25

224

0.61, 0.04-0.35, 0.06-0.46, 0.04-0.31, 0.06-0.42 and 0.04-0.31 ng m-3 for La, Ce, Nd, Sm, Eu, Gd,

225

Tb, Dy, Ho, Er, Tm, Yb and Lu respectively (Supplementery Section, Table S2). Overall REE

226

concentration vary by a factor of 3 or less

227

correlations not shown) but they do not correlate with the heavy metals. The Cr, Ni, Pb, Cu and

228

Cd concentrations in Kanpur are similar to the reported values for other urban centers in Asia,

229

and are almost an order of magnitude higher when compared to European and most of the Asian

230

urban centers (Table 3). It is noteworthy to mention that the previously reported heavy metal

231

levels at Kanpur26,29 were significantly higher than the present study (Table 3). This difference

232

can be explained by the differences in the sampling period and overall methodology. For

233

example, samples in the previous study29 were collected during summer to early monsoon period,

234

whereas in our study, samples were collected between monsoon and post-monsoon period.

and they correlate with each other (R2>0.9,

235

The ambient trace metal in the atmosphere can be originated from various sources. For

236

example, Cu may be sourced from diesel burnig and vehicular brake abrasion.33,34 Pb and Cd are

237

mainly derived from coal combustion products, vehicular exhaust and industrial emission.35,36 Ni

238

and Cr are mainly contributed from industrial emission.37,38,39 The source of the metal in this

239

study will be further discussed under section 3.3 and 3.5.

240

3.3. Natural vs. Anthropogenic Signature

241

The high Cr, Ni, Pb, Cu and Cd concentrations in Kanpur can be attributed to

242

anthropogenic sources since the heavy metal ratios of aerosol-laden filter paper are higher than

243

those in the average upper continental crust, which can be considered as a natural source of

244

airborne dust. For example, the average Pb/Cu ratio of airborne particles sourced from eroding

245

upper continental crust is 1.7750, while the aerosols in Kanpur have much higher Pb/Cu ratios

246

(6.68 ± 5.78, n = 21, 1 S.D.). Similarly, the average Cd/Ni and Cd/Cu ratios of eroding

247

continental crust range between 0.002-0.005 and 0.004-0.007, respectively23,51, whereas the

248

aerosols collected at Kanpur showed much higher values (0.83 ± 0.82 and 0.42 ± 0.39

249

respectively, n = 21, 1 S.D.). In contrast, the REE ratios in Kanpur were similar to the upper

250

continental crust signature. For example, the Sm/Nd ratio of aeolian dust is ~0.1752, whereas the

251

average Sm/Nd ratios of Kanpur aerosols were 0.25 ± 0.11 (n = 21, 1 S.D.). Thus, based on the

252

above data, we conclude that the REEs are predominantly derived from a natural (crustal) source,

253

whereas most of the heavy metals at are anthropogenically derived. To understand the extent of

254

heavy metal pollution, Enrichment Factors (EF) were calculated. The result shows (Figure 3) that

255

the Cd, Pb and Cu were 10-1500 times higher in Kanpur aerosols when compared to upper

Page | 8 ACS Paragon Plus Environment

Page 9 of 25 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Earth and Space Chemistry

256

continental crust.

257

3.4. Effect of the Monsoon Withdrawal on Aerosol Loading and Heavy Metals

258

The 2015 Monsoon over India was one of the weakest of the last few decades.31 In 2015,

259

the average annual rainfall over the IGB, as well as in Kanpur, was highest in the month of July

260

(peak monsoon), and thereafter the monsoon was on the wane (Figure 1a). For example, Kanpur

261

and its surrounding region recorded 153, 40, 0.25 and 0 mm of rain in July, August, September

262

and October, respectively (Table 1). Figure 4 clearly shows that rainfall decreased with the

263

waning phase of the monsoon, whereas aerosol loading in the air increased. Rain Day Frequency

264

(RDF) was calculated to better understand how rainfall frequency is related to the aerosol loding.

265

RDF was calculated by dividing the total number of rainy day with the total number of days in

266

that particular month.. The data showed that July experienced 12 rainy days, and the RDF was

267

~39% (Figure 4). Similary, RDFs for August and septemeber were 12 and 1%, respectively.

268

Figure 4 shows that withdrawal of the monsoon

269

concentrations in the atmosphere by 28, 43 and 152% with respect to July during August,

270

September, and October, respectively. This increased aerosol loading was negatively correlated

271

with RDF. Monsoon rain not only caused a drastic reduction in ambient PM loading by wet

272

removal process, it also prevented further aerosol loading in the atmosphere for the next couple of

273

days.

in Kanpur increased the PM10 mass

274

Withdrawal of the southwest monsoon increased the heavy metal loading in the air by a

275

factor of up to 2-3. This increase of heavy metal concentration can either be explained by lower

276

washout rates as rainfall decreased or changing emission sources or an increased magnitude of

277

emissions from a constant emission source(s). However, emission inventories over Kanpur are

278

more or less constant throughout the year.54 In order to further understand whether long distance

279

emission sources are involved, air mass trajectories were computed 5 days back in time at an

280

elevation (height above ground level-AGL) of 50m, 1000m and 1500m during the sampling

281

period using the HYSPLIT (Hybrid Single Particle Lagrangian Integrated Trajectory) data from

282

the U.S. National Oceanic and Atmospheric Administration’s (NOAA) website.55 The archived

283

data file GDAS1 was used for computations. The air mass back trajectory models reveal that the

284

air masses are coming from within the Indo-Gangetic Basin and from south, as further indiacted

285

by the wind direction data (Figure 1c). So it is presumed that the air mass came over similar

286

anthropogenic sources. Therefore, the significant increase in heavy metal loading most likely be

287

attributed to the aerosol build-up over the IGB after monsoon withdrawal.

Page | 9 ACS Paragon Plus Environment

ACS Earth and Space Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

288

Page 10 of 25

3. 5. Sources of Heavy Metals Aerosols in the environment can be derived from natural and anthropogenic sources

289 56,57,58,59

290

.

291

burning and industrial emissions.60 For example, one of the largest pollution hazes in the world

292

called the “brown cloud” covers an area of more than 10 million km2 over Southern Asia and is

293

mostly attributed to biomass and fossil fuel burning sources.61,62 The source of natural particulate

294

matter in the atmosphere is deflation of soil, which is a weathered product of upper continental

295

crustal rocks.

The dominant anthropogenic aerosol sources in the IGB are biomass burning, fossil fuel

296

In order to understand the sources of heavy metals in aerosol-laden filter papers we

297

investigated the trace metal systematics, since anthropogenic and natural end members should

298

have distinct trace metal signatures. The plot of Cu/Cd versus Ni/Cd ratios of the aerosols (Figure

299

5) reveals that the heavy metals in Kanpur aerosols are mostly derived from combustion sources,

300

primarily coal burning products with lesser contribution from vehicular and industrial emission.

301

However for better heavy metal source apportionment studies, isotopic composition of the heavy

302

metals will be required.

303

4. Conclusion

304

In the present study, the relationship between the changes in meterological parameters during

305

waning phase of monsoon and PM10 loading was investigated over the Central Indo-Gangetic

306

Basin. An urban site (Kanpur) over the Indo-Gangetic Basin was selected for the study. The study

307

demonstrates that withdrawal of the monsoon since July (peak monsoon month) increased the

308

aerosol laoding in the atmosphere by 28, 43 and 152% during August, September and October,

309

respectively. PM10 mass concentration in Kanpur reached >100 µg/m3, which is the upper limit of

310

Indian National Air Quality Standard (NAAQS) within just 3 months following the peak

311

monsoon in 2015. The study further shows that monsoon withdrawal increased the Cr, Ni, Cu and

312

Cd loading in the atmosphere. Only Pb showed a consistently high concentration throughout the

313

study period, which indicates that Pb may be derive from local emission sources or it is not

314

efficiently scavenged by wet deposition processes. Based on our available data, the behavior of

315

Pb cannot be definitely ascertained. Enrichment Factors further show that Cd, Pb and Cu in

316

aerosols were 10-1500 times higher than upper continental crust, while other heavy metals had

317

EF values 10 indicates anthropogenic origin, while EF ≤10 indicates crustal sources 53

352

Cd, Pb and Cu concentrations in our aerosol samples were 10-1500 times higher than upper continental crust, while the remaining elements had

353

EF values