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
