Atmospheric Environment 38 (2004) 1705–1714

ARTICLE IN PRESS

AE International – Asia

*Correspond

E-mail addr

1352-2310/$ - se

doi:10.1016/j.at

Chemical composition of precipitation during 1984–2002at Pune, India

P.D. Safai*, P.S.P. Rao, G.A. Momin, K. Ali, D.M. Chate, P.S. Praveen

Indian Institute of Tropical Meteorology, Dr. Homo Bhabha Road Pashan, Pune 411 008, India

Received 26 September 2003; accepted 26 December 2003

Abstract

Since 1984, bulk precipitation samples have been collected at Pune, a tropical urban location in India, in the

monsoon season. Data from these studies are used to analyze the long-term trends in the major chemical constituents of

precipitation. Significantly increasing trends were observed for SO4 and NO3 which could be attributed to the rise in

industrial and vehicular activities during this period. Also Ca, the chief neutralizing constituent, showed decreasing

trend, mainly due to the rapid urbanization that reduced the availability of open land which is the major source for Ca.

This has resulted in the overall decrease in trend of pH. However, the average pH value is still in the alkaline range due

to the dominance of neutralizing potential of precipitation over the acidic potential.

r 2004 Elsevier Ltd. All rights reserved.

Keywords: Ionic ratio; Fractional acidity; Original acidity; Neutralization factors; Enrichment factors; Frequency distributions

1. Introduction

A number of studies on time trend analysis in

precipitation chemistry have been reported from Central

and Western Europe, North America and recently from

South East Asia. Puxbaum et al. (1998) have taken a

comprehensive account of such studies from USA,

Netherlands, Denmark, Spain, Canada and Germany.

Most of the studies have reported decrease in SO4

concentrations along with increase in pH that could be

directly related with the decrease in the emissions of

SO2. Nilles and Conley (2001) also reported rainfall

composition data for a period of 1981–1998, from about

144 sites of National Atmospheric Deposition Program

of USA. About 35% decrease in SO4 has been reported.

No significant trends were observed for NO3, NH4 and

Ca, although about 64 sites showed decreasing trend for

Ca and 30 sites showed increasing trend for NH4. Fujita

et al. (2001) have reported trends in chemical composi-

tion of precipitation at six rural stations in western

ing author.

ess: pdsafai@tropmet.res.in (P.D. Safai).

e front matter r 2004 Elsevier Ltd. All rights reserve

mosenv.2003.12.016

Japan during 1987–1996. There was no significant

change in the concentrations of non-seasalt fraction of

SO4, i.e., nssSO4 and non-seasalt fraction of Ca, i.e.,

nssCa whereas, concentrations of NO3 and NH4 showed

increasing (45%) trends. The ratio of neutralizing

potential (NP) to acidic potential (AP) showed 24%

increase. Lee et al. (2001) have reported changes in

chemical composition of precipitation at four sites in

South Korea during 1993–1998. Concentrations of

NssSO4, NH4 and Ca showed decreasing trends at

statistically significant level whereas, NO3 did not show

much variation. Overall, pH did not show any

significant trend.

Acid precipitation has been a growing problem in

China, especially in its southern parts. About 73% of its

energy is produced by coal burning, resulting in

substantial increase in sulfur emissions. Natural alkaline

dust neutralizes much of the acidity in north but in south

the problem of acidity is more severe. pH values

reported are between 4 and 5 with sometimes well below

4 also (Zhao et al., 1988; Qin and Huang, 2001).

However, due to some recent policies in fuel changes and

other restrictions, growth of sulfur emissions has been

d.

ARTICLE IN PRESS

Fig. 1. Location of Pune and other sites under GAW network in India along with pH as reported by Shende (2000).

P.D. Safai et al. / Atmospheric Environment 38 (2004) 1705–17141706

decelerated, but the increase in NOx emissions is still

substantial (Streets et al., 2001). According to Hedin

et al. (1994), the lowering of pH in precipitation may

take place in Asia in future due to the decrease in

alkaline dust owing to mobilization and urbanization.

Studies on long-term monitoring of precipitation

chemistry, especially from the Asian region are very

few. Due to the rapid economic growth in this region,

emissions and consequently atmospheric concentrations

of many pollutants have increased enormously. For any

comprehensive and realistic assessment of the changes in

atmospheric composition, it is very much essential to

have data on the chemical constituents of precipitation.

In India, even though the chemical composition of

precipitation has been reported in many studies, most of

them are limited to a short period of time (Handa, 1973;

Mukherjee, 1964; Mahadevan et al., 1989; Khemani

et al., 1985, 1987). Under the Global Atmospheric

Watch (GAW) network, previously known as Back-

ground Atmospheric Pollution Monitoring Network

program of WMO, studies relating to precipitation

composition have been carried out at 10 different

stations in India since last three decades. Pune is one

of these stations. Results on this data have been reported

by many studies (Varma, 1989; Mukhopadhyay et al.,

1992; Shende, 2000; Soni and Kannan, 2001). Recently a

comprehensive review on precipitation chemistry studies

in India has been reported by Kulshrestha et al. (2003).

Pune is about 100 km inland on west coast of India,

on the leeward side of Western Ghats (Fig. 1). At the

Indian Institute of Tropical Meteorology (IITM), Pune,

monitoring of precipitation chemistry has been carried

out since 1970s. Results relating to these studies have

been reported earlier (Khemani, 1989; Naik et al., 1988;

Rao, 1997). Khemani (1985) has reported a decadal

(1974–1983) variation of ionic constituents of rain water.

However, then the observational site was on the western

outskirts, about 3 km from the city center and latter,

after 1984 it was shifted to Pashan, situated to the

northwest, on the outskirts of the city, about 10 km from

city center. Since then, continuous monitoring of rain-

water has been carried out at Pashan. Results relating to

the data on precipitation composition for a period of 19

years from 1984 to 2002 are discussed.

2. Sampling and analysis techniques

Sampling was carried out on the terrace of the IITM

building, about 10m above the surface. The funnel and

collector bottles were daily cleaned with triple distilled

water to avoid dry deposition of gaseous and particulate

pollutants. Samples were immediately analyzed for pH

and stored in refrigerator at about 4�C until the

completion of chemical analysis. Spectrophotometric

methods were used to measure Cl, SO4, NO3 and NH4

ARTICLE IN PRESSP.D. Safai et al. / Atmospheric Environment 38 (2004) 1705–1714 1707

and Atomic Absorption Spectrophotometric technique

was used to measure Na, K, Ca and Mg. Concentrations

of HCO3 were computed from pH and atmospheric CO2

partial pressure, assuming gas–liquid equilibrium (Brim-

blecombe, 1986). Details about analysis have been

mentioned elsewhere (Safai, 1999). Since 2000, analysis

of Cl, NO3 and SO4 is carried out by Ion Chromato-

graphy technique (Dionex-100, USA) using analytical

columns Ion Pac-AS4A-SC 4mm, anion micromem-

brane suppressor ASRS-1, 1.8mM Sodium Carbonate/

1.7mM Bicarbonate as eluent and triple distilled water

as regenerant.

3. Quality control methods

In the present study, data relating to the analysis of

bulk rainwater samples have been considered as wet-

only sampling was carried out during 1992–1997 only

and then continued after 2000 onwards and data related

to it have been reported elsewhere (Pillai et al., 2001).

Also, samples collected during sequential shower have

been omitted because they are better suitable for studies

relating to washout coefficients for different chemical

components of precipitation. Samples contaminated

with dust or bird droppings were removed. A simple

statistical data reduction technique (Rodhe and Granat,

1984) was adopted. Data points outside the range

(m � 3s; m þ 3s) were excluded, where m=average

and s=standard deviation. In this method, the filtering

process converges to a stable value of m and s after a

few iterations. It serves the purpose of excluding extreme

outliers and at the same time allows the real (sometimes

episodic) data to be retained.

4. Results and discussions

4.1. Variations of pH and ionic constituents of rain water

Figs. 2(a) and (b) show the temporal variations of pH

and different ionic constituents of rain water at Pune

during 1984–2002. pH showed significantly decreasing

trend with time (r ¼ �0:75 at o1% level of significance)

and its variation closely resembled that of Ca, which

also showed decreasing trend (r ¼ �0:42 at 7% level).

Concentrations of major acidifying components, i.e. SO4

and NO3 have shown increasing trend. Increasing trend

of SO4 was more significant (r ¼ 0:60; at o1% level)

than that of NO3 (r ¼ 0:36 at 10% level). Another

alkaline component NH4 has shown slightly increasing

trend but it was not significant (r ¼ 0:24 at 30% level).

Therefore, Ca was the major alkaline component

neutralizing the acidity generated by SO4 and NO3.

Similar results have been reported by many studies

(Noguchi et al., 1997; Al Momani et al., 1995). Other

ionic constituents, i.e., Na, Cl and Mg have shown

overall decreasing trend. However, the decreasing trend

was more significant for Mg (r ¼ �0:60 at 1% level)

than those for Na (r ¼ �0:35 at 10% level) and Cl

(r ¼ �0:23 at 30% level). K showed slightly increasing

trend (r ¼ 0:30; at 20% level) whereas, HCO3 showed a

significantly decreasing trend (r ¼ �0:81 at 1% level).

Average pH during 1984–2003 was 6.60. However, at

the beginning in 1984, it was 7.10 and remained around

6.90 until 1987. From 1988 onwards, it reduced down

below 6.80 with the lowest (6.36) recorded in 1999.

Khemani (1985) has reported decreasing trends in pH

and other ionic constituents of rainfall at Pune during

1974–1983, except NO3, Ca and Na, which did not show

any significant trend. Pillai et al. (2001) have reported

average pH 6.10 for rain water using wet-only collector

at the same site and also have shown a decreasing trend

with lowest value (5.71) in 1998.

Shende (2000) has reported trends in pH values at 10

locations in India inclusive of Pune, under the GAW

network of WMO, for the period 1985–1994. The

average values of pH at all these locations have been

indicated in Fig. 1, along with that obtained in the

present study. However, the average pH for Pune for

1985–1994 was 6.25 which is less as compared to the

present value of 6.60. This could be due to the difference

in observational site. The site under GAW network is in

close proximity of city center where vehicular and other

urban activities are more as compared to those at the site

under present study, which is away from city center.

Soni and Kannan (2001) have also reported the trends in

pH and ionic constituents at these 10 sites in India for

the period 1981–1998. They have reported that the pH

of precipitation in India varies between 3.60 and 8.50

with the maximum frequency observed in the range of

6.00–7.50. Also, there was a decreasing trend of pH at all

the 10 stations. Nagpur, Allahabad and Mohanbari

showed significant increasing trends of SO4 and NO3,

whereas, Minicoy, Port Blair and Visakhapatanam

showed significant increasing trend of NO3. Varma

(1989) has reported highest frequency of pH in the range

of 6.5–7.0 for rain water in India.

As seen from Fig. 3, about 53% of average ionic

composition is formed by cations whereas; remaining

47% is formed by anions. Ca shows the maximum

contribution (27%), followed by HCO3 (17%). The

minimum contribution is from K (1%). Seasalt, i.e., Na

and Cl contributed about 30% whereas, SO4 and NO3

together contributed about 14%. Significantly high

concentrations of Ca have been reported in precipitation

as well as aerosol samples over Pune, which has been

attributed to the transport of soil dust from the Arabian

Peninsula to the west coast of India (Khemani, 1989;

Rao, 1997; Naik, 2001).

The average value of the ratio between total anions to

total cations was 0.91, with a minimum of 0.60 and a

ARTICLE IN PRESS

1984 1988 1992 1996 20000

102030405060

10

20

30

40

05

1015202530

60

80

100

120

140

1984 1988 1992 2004

6.26.46.66.87.07.2

Con

cent

ratio

ns in

µ e

q / l

Ca

pH

SO4 NO

3

Year

NH4

Year20001996

2004 1984 1988 1992 1996 2000 2004

1984 1988 1992 1996 2000 20041984 1988 1992 1996 2000 2004

020406080

100120

10

20

30

40

50

0

3

6

9

12

15

020406080

100120

020406080

100120

Mg

K Na

Cl

HCO3

Year

Year

1984 1988 1992 1996 2000 2004 1984 1988 1992 1996 2000 2004

1984 1988 1992 1996 2000 20041984 1988 1992 1996 2000 2004

1984 1988 1992 1996 2000 2004

Con

cent

ratio

ns in

µ e

q / l

(a)

(b)

Fig. 2. (a) Temporal variations of pH, Ca, SO4, NO3 and NH4 in rain water at Pune during 1984–2002. (b) Temporal variations of

HCO3, K, Mg, Na and Cl in rain water at Pune during 1984–2002.

P.D. Safai et al. / Atmospheric Environment 38 (2004) 1705–17141708

maximum of 1.20. Thus, ionic balance was nearly

complete, mainly due to the inclusion of HCO3,

computed from H+ ion. Earlier, Khemani (1985) has

reported an average ionic ratio of 0.60 at Pune during

1974–1983, without including HCO3 ion which is very

essential, especially when average pH is above 5.60.

Noguchi and Hara (2002) have also reported ionic

imbalance until HCO3 was included.

4.2. Correlation analysis

Correlation analysis was carried out to detect the

possible common sources of ionic constituents. As seen

from Table 1, pH was mainly related with HCO3, Ca,

Mg and seasalt (Na, Cl), with r ¼ 0:85; 0.66, 0.63, and0.50, respectively. It did not show significant correla-

tions with both SO4 and NO3. Ca showed good

ARTICLE IN PRESSP.D. Safai et al. / Atmospheric Environment 38 (2004) 1705–1714 1709

correlations with Mg (r ¼ 0:75), Na and Cl (r ¼ 0:70 for

both). Also, both Ca and Mg showed good correlations

(rX0:60) with HCO3, indicating towards formation of

calcium and magnesium carbonates, possibly originated

from fine clay size particles fractured from calcite and

dolomite. Similar results were reported by Saxena et al.

(1991), Noguchi and Hara (2002) and Mukhopadhyay

et al. (1992). Significantly good correlation was seen

between Na and Cl (r ¼ 0:92) and also, both showed

good correlations with Mg (r > 0:70), indicating towards

a common marine source. Good correlations between

these marine components and Ca indicate towards the

existence of deposited seasalt particles on the local soil.

SO4 and NO3 did not show significant correlation

among themselves (r ¼ 0:21). In fact, SO4 was correlated

with Ca and seasalt (Na and Cl) rather than with NO3

Fig. 3. Average chemical composition of rain water at Pune

during 1984–2002.

Table 1

Correlation coefficients for the pH and ionic constituents of precipita

Cl SO4 NO3 HCO3 NH4 N

Cl 1 0.41 0.10 0.44 �0.12

SO4 1 0.21 0.15 0.04

NO3 1 0.15 0.17

HCO3 1 �0.06

NH4 1 �Na

K

Ca

Mg

H

pH

and NH4. Thus, the major portion of SO4 was in the

form of salts with soil/sea originated constituents. Jain

et al. (2000) have reported significant fractions of nssSO4

in rain water at Delhi, contributed by soil. SO4 in rain

water has been reported due to the resuspension of

gypsiferous soil in India (Jacks et al., 1994). Naik et al.

(1988) have reported similar results for rain water at

Pune and surrounding rural areas. Khemani et al. (1982)

have shown a bimodal distribution for SO4 aerosols at

Pune during the monsoon season, with a major portion

of mass in coarse fraction. It was suggested that

atmospheric SO2 may be adsorbed on particulate matter

and thereby react with either soil-derived components

like Ca, Mg and K or with marine component like Na.

NO3 showed some correlation with K and Ca (r ¼ 0:30)indicating towards the formation of its salts with the

crustal components. Similar results have been reported

earlier (Mukhopadhyay et al., 1992; Naik et al., 1988).

4.3. Quantification of acidifying and neutralizing

potentials

Over large areas of Northern Hemisphere, acidity of

rain is controlled mainly by the strong acids, e.g., H2SO4

and HNO3 (Overrain et al., 1981). Considering both SO4

and NO3 as the main acidifying components of rain,

Fractional Acidity (FA) was computed as, FA=[H+]/

([nssSO42�]+[NO3

�]) (Balasubramanian et al., 2001). If

this ratio is one, it is considered that the acidity

generated by these strong acids is not neutralized at

all. Fig. 4 shows variation of FA in rain water at Pune

during 1984–2003. There is slow but gradual increase in

FA indicating some decrease in neutralization of

acidification. Table 2 shows the annual variation of

FA. The average FA value was 0.013, indicating that

about 99% acidity was neutralized by alkaline constitu-

ents. Relative contribution of NO3 to the acidification

was computed using a ratio [NO3�]/([NO3

�]+[nssSO42�])

and its average value was 0.54, showing that about 54%

of acidity of rain was due to NO3 and 46% due to SO4.

tion at Pune (significant at 1% level)

a K Ca Mg H pH

0.92 0.45 0.68 0.71 �0.02 0.50

0.43 0.31 0.47 0.31 �0.18 0.19

0.10 0.29 0.27 0.24 0.06 0.17

0.46 0.30 0.63 0.61 0.13 0.85

0.13 0.07 �0.05 �0.06 �0.06 �0.03

1 0.47 0.69 0.74 0.04 0.51

1 0.51 0.48 0.02 0.35

1 0.75 0.00 0.66

1 0.13 0.63

1 0.11

1

ARTICLE IN PRESS

1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002 20040.000

0.008

0.016

0.024

0.032

0.040

[H+]

/ [N

O3+

] + [N

ss S

O42-

]

Year

Fig. 4. Variation of FA in rain water at Pune during

1984–2002.

Table 2

Annual variation of fractional acidity, original acidity and ratio

of neutralization potential to AP in the rain water at Pune

during 1984–2002

Year Fractional

acidity (meq l�1)

Original acidity

(meq l�1)

NP/AP

1984 0.005 128.25 2.39

1985 0.022 122.20 3.40

1986 0.015 152.70 3.30

1987 0.006 107.31 0.69

1988 0.013 127.01 1.44

1989 0.025 81.19 2.01

1990 0.004 98.54 0.75

1991 0.009 113.30 1.01

1992 0.020 130.63 2.39

1993 0.009 81.12 1.21

1994 0.007 136.32 0.73

1995 0.012 79.80 0.68

1996 0.010 139.32 0.87

1997 0.008 86.89 0.71

1998 0.020 110.72 1.15

1999 0.010 87.82 0.48

2000 0.030 84.34 0.84

2001 0.007 99.24 0.58

2002 0.009 125.42 0.86

Average 0.013 110.14 1.34

1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002 20040.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

4.0

Nf Ca Nf NH4 Nf Mg Nf K

Neu

tral

izat

ion

Fac

tors

1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002 2004

0

1

2

3

4

NP

/ AP

Year

(a)

(b)

Fig. 5. (a) Variation of NFs for Ca, NH4, Mg and K for rain

water at Pune during 1984–2002. (b) Variation of NP/AP in

water at Pune during 1984–2002.

P.D. Safai et al. / Atmospheric Environment 38 (2004) 1705–17141710

Nitric acid is found to be of common occurrence in rain

water in India (Mukhopadhyay et al., 1992).

According to Galloway et al. (1987), the measured

hydrogen ion concentration [H+] reflects the acidity

of precipitation after neutralization by atmospheric

bases especially, NH4 and Ca. Therefore, the original

acidity should be estimated by calculating [H+]+

[NH4+]+[Ca2+] to compare with the measured acidity

[H+]. Year-wise variation of the measured acidity and

calculated original acidity for rainfall at Pune is shown

in Table 2. It can be seen that the average pH of rainfall

at Pune was 3.96 in the absence of Ca and NH4 (average

original acidity=110.14meq l�1). Momin (1990) has

reported about two orders of magnitude more original

acidity than the measured one for rainfall at Delhi. He

has suggested that the neutralization reactions are more

rapid and extensive during the early stages of rainfall

when the airborne dust is plentiful and pH of precipita-

tion is low than during the latter stages when the pH is

higher and amount of dust is lower.

Varma (1989) formulated an empirical relationship

between total suspended particulates and pH for

Jodhpur and Srinagar, two north Indian locations

which receive high incursions of dust from adjoining

arid regions. The soil particles are rich in alkaline

ions like Ca, Mg, and K which neutralize the acidity

of rain.

Neutralization factors (NF) were computed for

different alkaline constituents using the formulae

suggested by Parashar et al. (1996). Fig. 5(a) shows the

variation of NF values for Ca, Mg, K and NH4. Ca

showed maximum value among all the cations. Average

NF values were 1.21, 0.18, 0.19 and 0.09 for Ca, NH4,

ARTICLE IN PRESSP.D. Safai et al. / Atmospheric Environment 38 (2004) 1705–1714 1711

Mg and K, respectively. To assess the balance between

acidity and alkalinity, ratio of NP to AP was compu-

ted as, NP/AP=[nssCa2+]+[NH3+]/[nssSO4

2�]+[NO3+].

Table 2 reveals the variation of this ratio for each year.

The average value was 1.33, indicating overall dom-

inance of alkaline constituents that prevented acidifica-

tion of rain water. However, as seen in Fig. 5(b), the

decreasing trend of this ratio showed that the neutraliz-

ing capacity is declining.

4.4. Frequency distributions of pH and ionic constituents

Figs. 6(a) and (b) show the frequency distributions of

pH and ionic constituents of rain water at Pune during

0

20

40

60

80

100

140120100806040200

Per

cent

age

Fre

quen

cyP

erce

ntag

e F

requ

ency

SO4

0

20

40

60

80

0

20

40

60

80

100

1201007550250

NH4

0

20

40

60

80

100

2180120600

Concentrations in µ eq / l

100

6040200

0

20

40

60

80

360300240180120600

HCO3

0

20

40

60

80

1280400

0

20

40

60

80

2520151050

K

0

20

40

60

80

1280400

100

100100

100

Concentrations in µ eq / l

(a)

(b)

Fig. 6. (a) Frequency distribution of SO4, NO3, pH, NH4 and Ca in

HCO3, Na, Mg, K and Cl in water at Pune during 1984–2002.

1984–2002. About 71% samples showed pH in the range

of 6.5–7.5 while 17% showed in the range of 6.0–6.5 and

10% in the highly alkaline range of 7.5–8.0. Less than

2% samples showed pHp5.

In case of Ca, about 30% samples showed concentra-

tions below 50meq l�1 whereas, more than 90% for NH4

and NO3, 85% for Mg, 78% for SO4, 60% for Na and

Cl and 56% for HCO3 showed concentration below

50 meq l�1. Concentrations above 100 meq l�1 were rarely

found in case of SO4, NO3 (2% each), Mg (1%) and

NH4 (nil). For K, about 98% samples showed concen-

trations below 15 meq l�1. Only Ca showed maximum

concentrations with some samples (about 2%) reaching

above 300meq l�1.

0.1

1

10

8.07.57.06.56.05.55.0

pHNO3

42036030040

Ca

100

14012010080

2402001600

Na

0

20

40

60

80

1201007550250

Mg

2402001600

Cl

100

water at Pune during 1984–2002. (b) Frequency distribution of

ARTICLE IN PRESSP.D. Safai et al. / Atmospheric Environment 38 (2004) 1705–17141712

4.5. Source assessment for different ionic constituents

The geographical location of Pune is such that it

receives marine as well as continental airmasses during

different seasons. Generally, during June–September,

southwesterly and westerly winds, rich with marine

airmasses from the Indian Ocean and Arabian Sea,

prevail over this region. Westerlies tend to be weak and

occasionally turn easterlies in winter during December–

February. There is a major industrial belt in the north

west of Pune about 200 km, near Mumbai and another

nearby industrial zone to the north is situated at Pimpri,

Chinchwad, about 20 km from the city. In addition,

there are various other sources in the surroundings, e.g.,

several small-scale industries, vehicles (more than

13,00,000 of which, about 10,00,000 are two wheelers),

brick kilns, rapidly growing construction activity in

various parts of the city, etc. The population of the city

is nearly 25,00,000. Thus, the city receives inputs from

both natural as well as man-made sources. As such,

there is a broad diversity in the origin of airmasses

associated with rainfall at Pune. Also, soil in this part

has been reported to be alkaline with dominance of

calcarious components (Khemani et al., 1989).

Using Ca as reference element for soil and Na as

reference element for marine source, enrichment factors

(EF) were computed as

EF for soil ¼ ½X=Ca�sample=½X=Ca�crust and

EF for marine ¼ ½X=Na�sample=½X=Na�sea;

where, X is the desired constituent, X/Ca ratio from

crustal composition (Horn and Adams, 1966) and X/Na

ratio from marine composition (Goldberg et al., 1971).

Generally, contribution for NO3 and NH4 from marine

source is very meagre; therefore, EF for marine source

were computed for Cl, Mg, K, and SO4. Also, EF for

soil source were computed for K and Mg. Average EF

values (marine source) for Cl, Mg, K and SO4 were 1.2,

3.6, 8.5 and 14.1, respectively whereas, average EF

values (soil source) for K and Mg were 0.13 and 0.30,

respectively. Thus, Cl was from marine source whereas,

K was mostly from soil but the contribution of marine

source was also considerable. In case of Mg, both the

sources were equally contributing. Resuspension of

seasalt, deposited on local soil is reported as source for

Na, and Cl (Norman et al., 2001). Considering Na as

reference element for marine origin, seasalt contribution

for both K and Mg was computed with the help of

method used by Khemani (1989), and was found to be

29% for K and 50% for Mg.

SO4 was highly enriched for soil source (EF ¼ 30), but

contribution from marine source was also significant

(EF ¼ 14). Computation of non-seasalt fraction showed

about 21% SO4 from sea. An attempt has been made to

quantify the share of natural (marine and soil) and man-

made sources for both NO3 and SO4. Considering no

concentration from marine source for NO3, contribution

from anthropogenic sources was obtained by removing

soil contribution from the total NO3. Contribution from

soil was computed using (NO3/Ca) ratio for crustal

composition (Horn and Adams, 1966). About 99% NO3

was contributed from anthropogenic sources.

While assessing sources for SO4, contribution from

soil was removed from that of non-seasalt fraction, and

the remaining portion was considered due to the

anthropogenic sources. Soil contribution was computed

using SO4/Ca ratio for crustal composition (Horn and

Adams, 1966). About 71% of total SO4 was contributed

by anthropogenic sources while 21% was contributed

from marine and 8% from soil source. Norman et al.

(2001) have reported the possibility of soil source for

both SO4 and Ca in the rainfall over the western coast of

India due to long-range transport of Ca and SO4 from

Africa or Arabia as some wind trajectories reaching to

the west coast of India pass through these regions.

Major source of NH4 was probably the human and

animal excretions. Use of fertilizers could not be the

major source as agricultural activities are not predomi-

nant in and around the city. Emissions from both brick

kilns and vehicles could be other sources.

5. Conclusions

An assessment of the chemical composition of

precipitation at Pune during 1984–2002 revealed that

the pH of precipitation though in alkaline range, showed

gradual decreasing trend, mainly due to the increasing

acidic constituents, NO3 and SO4, and simultaneously

decreasing alkaline constituents, especially, Ca. This

feature is attributed to the influence of industrial,

vehicular and commercial activities that have risen

enormously during the past few decades in and around

the city. NO3 contributed more (55%) than SO4 towards

acidification of rain, indicating the impact of vehicular

emissions. About 71% SO4 originated from man-made

sources followed by sea (21%) and soil (8%). Alkaline

constituents neutralized about 99% of the acidity caused

by SO4 and NO3. The major neutralizing constituent, Ca

was either in the form of CaCO3 or CaSO4. In the

absence of Ca and NH4, pH of precipitation could have

been p4.00. Thus, the chemistry of precipitation at

Pune is largely affected by variations in the concentra-

tions of Ca, NO3 and SO4. Studies regarding the long-

term trends in chemical constituents of precipitation are

very few, especially from the Indian subcontinent,

mainly due to the paucity of continuous data record.

The present work could serve as a bench-mark study for

this region regarding the future assessment of chemical

nature of precipitation.

ARTICLE IN PRESSP.D. Safai et al. / Atmospheric Environment 38 (2004) 1705–1714 1713

References

Al Momani, I.F., Ataman, O.Y., Anwari, M.A., Tuncel, S.,

Kose, C., Tuncel, G., 1995. Chemical composition of

precipitation near an industrial area at Izmir, Turkey.

Atmospheric Environment 29, 1131–1143.

Balasubramanian, R., Victor, T., Chun, N., 2001. Chemical and

statistical analysis of precipitation in Singapore. Water, Air

and Soil Pollution 130, 451–456.

Brimblecombe, P., 1986. Air Composition and Chemistry,

Cambridge Environmental Chemistry Series. Cambridge

University Press, Cambridge.

Fujita, S.I., Takahashi, A., Hayami, H., Sakurai, T., 2001.

Long-term trends in the chemical composition of precipita-

tion over western Japan. Water, Air and Soil Pollution 130,

415–420.

Galloway, J.N., Diawu, Z., Jiling, X., Likens, G.E., 1987. Acid

rain: China, United States and a remote area. Science 236,

1559–1562.

Goldberg, E.D., Broecker, W.S., Gross, M.G., Turekian, K.K.,

1971. Marine chemistry. In: Radioactivity in the Marine

Environment. National Academy of Sciences, Washington,

DC, pp. 137–146.

Handa, B.K., 1973. Chemical composition of monsoon rain

water over Chandigarh in 1971. Indian Geohydrology 9,

31–37.

Hedin, L.O., Granat, L., Likens, G.E., Buishand, T.A.,

Galloway, J.N., Butler, T.J., Rodhe, H., 1994. Steep

declines in atmospheric base cations in regions of Europe

and North America. Nature 367, 351–354.

Horn, M.K., Adams, J.A.S., 1966. Computer-derived geochem-

ical balances and element abundances. Geochimica et

Cosmochimica Acta 30, 279–297.

Jacks, G., Sharma, V.P., Torssander, P., Oberg, G., 1994.

Origin of sulphur in soil and water in a precambrian terrain.

Geochemical Journal 28, 351–358.

Jain, M., Kulshrestha, U.C., Sarkar, A K., Parashar, D.C.,

2000. Influence of crustal aerosols on wet deposition at

urban and rural sites in India. Atmospheric Environment

34, 5129–5137.

Khemani, L.T., 1985. Characteristics of atmospheric gaseous

and particulate pollutants and their influence on cloud

microphysics and rain formation. Ph.D. Thesis, University

of Poona, Poona.

Khemani, L.T., 1989. Physical and chemical characteristics of

atmospheric aerosols. In: Cheremisinoff, P.N. (Ed.),

Air Pollution Control, Encyclopedia of Environmental

Control Technology, Vol. 2. Gulf Publishing Co., USA,

pp. 401–452.

Khemani, L.T., Momin, G.A., Naik, M S., Vijayakumar, R.,

Ramana Murty, Bh.V., 1982. Chemical composition of size

distribution of atmospheric aerosols over the Deccan

Plateau, India. Tellus 34, 151–158.

Khemani, L.T., Momin, G.A, Naik, M.S., Prakasa Rao, P.S.,

Kumar, R., Ramana Murty, Bh.V., 1985. Impact of alkaline

particulates on pH of rain water in India. Water, Air and

Soil Pollution 24, 365–376.

Khemani, L.T., Momin, G.A., Naik, M.S., Prakasa Rao, P.S.,

Safai, P.D., Murty, A.S.R., 1987. Influence of alkaline

particulates on pH of cloud and rain water in India.

Atmospheric Environment 21, 1137–1145.

Khemani, L.T., Momin, G.A., Rao, P.S.P., Safai, P.D., Singh,

G., Kapoor, R.K., 1989. Spread of acid rain over India.

Atmospheric Environment 23, 757–762.

Kulshrestha, U.S., Granant, L., Rodhe, H., 2003. Precipitation

chemistry studies in India, a search for regional patterns.

Department of Meteorology, Stockholm University, Inter-

national Meteorological Institute in Stockholm, Report

CM-99.

Lee, B.K., Lee, D.S., Kim, M.G., 2001. Rapid time variations

in chemical composition of precipitation in South Korea.

Water, Air and Soil Pollution 130, 421–426.

Mahadevan, T.N., Negi, B.S., Meenakshy, V., 1989. Measure-

ments of elemental composition of aerosol matter and

precipitation from a remote background site in India.

Atmospheric Environment 23, 869–874.

Momin, G.A., 1990. The role of alkaline particulates on

pH of rain water in India. M.Sc. Thesis, University of Pune,

Pune.

Mukherjee, A.K., 1964. Acidity of monsoon rain water. Indian

Journal of Meteorology and Geophysics 5, 267–270.

Mukhopadhyay, B., Datar, S.V., Srivastava, H.N., 1992.

Precipitation chemistry over the Indian region. Mausam

43, 249–258.

Naik, M.S., 2001. Study of ionic constituents in rain water in

polluted and non-polluted regions of India. Ph.D. Thesis,

University of Pune, Pune.

Naik, M.S., Khemani, L.T., Momin, G.A., Prakasa Rao, P.S.,

1988. Measurement of pH and chemical analysis of rain

water in rural area of India. Acta Meteorologica Sinica 2,

91–100.

Nilles, M.A., Conley, B.E., 2001. Changes in the chemistry of

precipitation in the United States, 1981–1998. Water, Air

and Soil Pollution 130, 409–414.

Noguchi, I., Hara, H., 2002. Precipitation hydrogen carbonate

and its significance in East Asia. In: Peng, L.C., Bala, R.,

Rodhe, H., Hicks, K. (Eds.), Proceedings of First Work-

shop on Composition of Asian Deposition, Malaysia, 21–23

November 2002, pp. 67–75.

Noguchi, I., Kato, T., Akiyama, M., Otsuka, H., Matsumoto,

Y., 1997. The precipitation chemistry in northern Japan—

effect of alkaline dust. In: Murano, K. (Ed.), Proceedings of

International Workshop on Unification of Monitoring

Protocol of Acid Deposition and Standardization of

Emission Inventory, Japan, March 1997, pp. 76–83.

Norman, M., Das, S.N., Pillai, A.G., Granat, L., Rodhe, H.,

2001. Influence of mass trajectories on the chemical

composition of precipitation in India. Atmospheric Envir-

onment 35, 4223–4235.

Overrain, L.N., Seip, H.M., Tollan, A., 1981. Acid precipita-

tion—effects on forest and fish. SNSF project final report

175.

Parashar, D.C., Granat, L., Kulshrestha, U.C., Pillai, A.G.,

Naik, M.S., Momin, G.A., Prakasa Rao, P.S., Safai, P.D.,

Khemani, L.T., Naqvi, S.W.A., Narvekar, P.V., Thapa,

K.B., Rodhe, H., 1996. Chemical composition of precipita-

tion in India and Nepal. A Preliminary Report on an Indo-

Swedish Project on Atmospheric Chemistry, Department of

Meteorology, Stockholm University, International Meteor-

ological Institute, Stockholm, CM-90.

Pillai, A.G., Naik, M.S., Momin, G.A., Rao, P.S.P., Safai,

P.D., Ali, K., Rodhe, H., Granat, L., 2001. Studies of wet

ARTICLE IN PRESSP.D. Safai et al. / Atmospheric Environment 38 (2004) 1705–17141714

deposition and dust fall at Pune, India. Water, Air and Soil

Pollution 130, 475–480.

Puxbaum, H., Simeonov, V., Kalina, M.F., 1998. Ten years

trends (1984–1993) in the precipitation chemistry in central

Austria. Atmospheric Environment 32, 193–202.

Qin, G., Huang, M., 2001. A study on rain acidification

processes in ten cities of China. Water, Air and Soil

Pollution 130, 163–174.

Rao, P.S.P., 1997. Some studies on the deposition of atmo-

spheric pollutants in different environments in India. Ph.D.

Thesis, University of Pune, Pune.

Rodhe, H., Granat, L., 1984. An evaluation of SO4 in

European precipitation 1955–1982. Atmospheric Environ-

ment 18, 2627–2639.

Safai, P.D., 1999. A study of the air pollutants in the

environment of the Nilgiri Biosphere Reserve, South India.

Ph.D. Thesis, University of Pune, Pune.

Saxena, A., Sharma, S., Kulshrestha, U.C., Srivastava, S.S.,

1991. Factors affecting alkaline nature of rain water in Agra

(India). Environmental Pollution 74, 129–138.

Shende, P.R., 2000. Changes in trends of pH over BAPMON

stations in India. Vayu Mandal 30, 36–43.

Soni, V.K., Kannan, P.S., 2001. Background acidic precipita-

tion trends in India, Proceedings of National Symposium

Tropmet 2001 on Meteorology for Sustainable Develop-

ment, Mumbai, pp. 616–621.

Streets, D.G., Tsai, N.Y., Akimoto, H., Oka, K., 2001. Trends

in emissions of acidifying species in Asia, 1985–1997. Water,

Air and Soil Pollution 130, 187–192.

Varma, G.S., 1989. Background trends of pH of precipitation

over India. Atmospheric Environment 23, 747–751.

Zhao, D., Xiong, J., Xu, Y., Chan, W.H., 1988. Acid rain

in southwestern China. Atmospheric Environment 22,

349–358.