Composition of rain water in Agra city, a semi-arid area in India

(1)

Indian Journal of Radio & Space Physics Vol. 31, February 2002, pp. 49-55

Composition of rain water in Agra city, a semi-arid area in India

Richa Mudga\', Rajeev Upadhyal & Ajay Taneja I I Department of Chemistry, St. John's College, Agra 282 002

2U. P. Pollution Control Board. Aligarh 202 001

Received 26 Marcil 2001; revised 27 June 2001; accepted 30 October 2001

Rainwater samples were collected at St. John's crossing during the monsoon period of 2000. The cations (Ca2+, Mg2+, Na+. K+ and NH/ ) and anions (F. CI-, HCO.l-, NO)- and SO/-) along with pH were measured. The percentage contribution of soil components (Ca²+, Mg2+, Na+ and K+) and NH/ are observed to be higher than the acidic substances.

The ratio of F/Ca²+, Mg2+ICa2+, Na+ICa²+ and NO)-ICa2+ in rainwater samples indicates that local soil plays a significant role in precipitation chemistry. Hence. soil dust is responsible for neutralization of wet deposition. The data were also subjected to factor analysis based on principal component analysis using the SPSS software. Factor analysis also indicates that the soil as well as dust emission is the major contributor to rainwater composition.

1 Introduction

Wet deposition process is an efficient pathway for removing the gases and particles from the atmosphere to the biosphere. It also plays a significant role in controlling the concentration of these species.

Incorporation of Sand N oxides in wet deposition is particularly important, as they are precursors of major acidst.2 (H2S04 and HN0₃). Therefore, precipitation chemistry measurements play a critical role in defining both the level of acid deposition and the state of some important biogeochemical cycles of the earth atmospheric system³.4. Khemani et

ae

^took ^some

aeroplane observations and noticed that the rain drops, immediately coming out from the cloud, possess relatively low pH; but when they reach the earth's surface, the pH increases, Therefore, the environment through which the rain drops travel plays an important role in deciding the rain acidification at each location. Because of this concern, precipitation chemistry has been the subject of intense research in last two decades³-9

. Precipitation acidity is a function of its contents of both acids and bases and any attempt to understand the processes, causing acid precipitation, must deal with the potential acid neutralizing capacity of alkaline materials such as Ca, K, Mg and Na which are released mostly from soilS. It has been suggested that alkaline dust may react with and neutralize strong acid in the atmospheres. Several investigators⁶-9

have noticed the influence of calcareous particulate matter on rain chemistry.

Recently, Peenya industrial area of Bangalore city (in India) shows acidic pH (Ref. 10). At present, there are

more than 1000 stations conducting precIpitation chemistry measurements around the world. All these working stations may be categorized into Global, Regional and Local Networks. Global Networks comprising the Background Air Pollution Monitoring Network (BAPMoN) and Global Precipitation Chemistry Project (GPCP) sites. collect precipitation at remote areas and provide worldwide information on the background concentration of air pollutants and long range transport of trace substances in the atmospheret I.

The present paper deals with the analysis of rainwater collected during the monsoon period of 2000 at Agra, since Agra is recognized as a sensitive zone due to Taj Mahal. The International Monuments Association has also declared Agra as a heritage city. This paper presents the role of alkaline materials 111

preventing the extension of acid rain at Agra.

2 Experimental details

2.1 Study location

Agra (27° lO'N, 78° OS'E, 169 m amsl) is located in the north central part of India. It is bounded with the Thar desert of Rajasthan on its south-west and north- west peripheries and is, therefore, a semi-arid area.

Rainwater was collected at St. John's crossing, situated in the heart of the city. The traffic density is maximum at this crossing. The major industrial activities in Agra are foundries, rubber processing units, chemical and engineering units. These industries consume common fuels like hard coke, steam coal, wood and oil. Emission from these

(2)

50 ^INDIAN^JRADIO & SPACE PHYS. FEBRUARY 2002

industries includes CO, S02, NO.. gases and particulate matter^l2• In addition, the population (more than 12,00,000) and vehicular density per day (greater than 10⁵vehicles) in Agra city are also high.

2.2 Sampling and analysis

Rainwater samples were collected on the roof of faculty building at St. John's College. Agra. It is about 8 m from the ground level and I m above the floor of the roof to prevent contamination by splashes from the ground. The general set-up and sampling procedure were opted as described by Mahadevan et a/.¹³ Manual sample collectors were made by attaching a funnel with an internal diameter of 14 cm via polyethylene bottle. The collectors were deployed as soon as rain began and retrieved immediately after the rain stopped. The efficiency of collection was 92%. A total of 36 samples were collected during the monsoon of 2000. The pH of the samples was determined immediately within one hour after the collection of samples. The samples were filtered through Whatman-41 filter paper into two clean polyethylene bottles. One part was stored under refrigeration and used for anion analysis, and the other was acidified with HN03 and used for cation analysis.

The anions F, CI, N03, S04 were analyzed by Dionex DX 500 ion chromatography using AS4A-SC column with self-regenerating suppresser (SRS).

Mixture of Na2C03 (1.8 mM) and NaHC0₃(1.7 mM) was used as the eluent. The major cations Ca, Mg, Na, K were analyzed by atomic absorption technique using Perkin-Elmer atomic absorption spectroscopy (AAS) (2380 model). The NH4 value was determined colorimetrically by the indophenol blue method using UV -VIS spectrophotometer^l4. The calibration curves for different chemical components have been obtained by preparing standard solutions using grade chemicals. These calibrations of different chemical constituents have been determined with reference to blank solutions. The analytical errors are nominal and vary within

±

10 %.

3 Results and discussion

3.1 Chemical composition

Precipitation weighted mean and standard deviation of ions in rainwater samples at St. John's Crossing, Agra. is shown in Table 1. Precipitation weighted means (PWM ) and standard deviation of precipitation weighted means (SDpw ) have been calculated as,

N

L P [X]

PWM [X] = ;=1 N' ,

LP

;=1 '

112

N N

N L p2[Xf - L[P{X.

If

i=1 I I i=1 1 I

The concentration of ions follows a general pattern as Ca²+ > HC03- > Mg2+>Na+> SO/->

cr

^{> N0}3->

NH4 + > K+ > F . The acidity of water depends on the concentration of anionic as well as cationic species.

Acidic pH reveals the excess acids in rainwater, while neutral or alkaline pH values indicate neutralization of acids by ammonia and soil dust (predominantly CaC03 and MgC03). The pH of rainwater varies between 6.6 and 7.7 indicating an alkaline nature as compared to the reference level of 5.6 (Ref. 15). In India, pH between 6.0 and 7.5 have been reported in urban areas^l6and between 5.22 and 7.65 in forested areas 17. In contrast, in temperate countries much lower pH values have been reported; e.g. 5.1 at Amsterdam Island4

(remote marine site), 4.96 at Chile, Pacific Ocean^l8and 4.03-5.6 at Vennezuelan Savannah 19. In Table 2, the concentrations have been compared with available data from (i) Chembur- Trombay, one of the highly industrialized area of the metropolitan city of Mumbai, (ii) Dayalbagh (Agra), the suburban .area located about 200 km south-west of New Delhi, (iii) Gopalpura which is a rural area

Table I-Precipitation weighted (PW) concentration (jleqll) and standard deviation (SD) for rainwater samples

Species P Wmean SD

Ca²+ 145 39

HCO)- 87 62

Mg2+ 65 14

Na+ 59 17

S042- 47 18

CI- 38 II

NO)- 33 13

NH/ 12 10

K+ 9 4

F 4 2

Note: LF+CI+NO)+S04 = 122 LCa+Mg+Na+K+NH4

=

290 Inclusion of HCO)-

=

209 Number of samples

=

36

(3)

MUDGAL el al.: RAIN WATER COMPOSITION AT AGRA 51

Table 2 -Composition of rainwater (Jleq/l) at different sites in India

Concentration of sl2ecies Reference

Ca¹+ Mgz+

Site CI- S042- NO] NH/

Chembur 164.5 70.4 29.5 41.1

(Urban)

Dayalbagh 23.7 28.4 17.6 12.7

(Suburban)

Gopalpura 31.9 16.7 27.5 44.2

(Rural)

Silent Valley 43.0 20.0 21.0 3.0

The concentration of acidic components, mainly SO/

-, is 1.6, 3.0

and

2.3

times higher than the values reported in Dayalbagh, Gopalpura and Si

lent Vall

ey, respectivel y, whi

le 1.5

times

lower

than the value reported in Chembur-Trombay. The concentration of N0

₃-

is

I.l, 1.8, 1.5

times higher than the values reported at Chembur, Dayalbagh and Silent Valley

.

Hence, at this site the concentration of ionic species may be said to reflect the chemistry of precipitation influenced by local anthropogenic sources .

3.2 Ion balance

The equivalent concentrations of Lanions and Lcations are presented in Table 1. The sum of major anions (F, CI-, N0

_3-,

SO/- ) is

122 /Leq/l,

while the sum of major cations (Ca 2 +, Mg2+, Na+, K+ and NH/ ) is

290

/Leq/l. From these val

ues a cation excess of 168 /Leq/l

is observed. Th e significant anion deficiency in rainwater samples may be due to the exc

lusion of

some anion s. The major anions, which may cau se the imbalance, are bicarbonate, N0

2,

P0

4

and Br. The bicarbonate ion is one of the major anions in rainwater in India during monsoon period

^'⁶.

The HC0

3 -

ion exists in precipitation through dissolution

Na+ K+

168.2 6.9 89.5 36.5 Khemani el al?O

54.8 7.7 83.9 26.4 Saxena el al²¹ 17.1 2.9 107.0 72.3 Satsangi el al.²² 46.0 4.0 43.0 14.0 Rao el al.¹⁷ 59.0 9.0 145.0 65.0 Present data

of atmospheric CO 2 and other particulate carbonate species in cloud and rain droplets. The concentration of HC0

_3-

ions would be significant at the alkaline pH values encountered in these rain sa mples. Since no direct measurements of HC0

3-

were avail able, the concentration was estimated from the th eoreti cal relati on between pH and HC0

₃-

when the pH is above

5.6

and the sample is in equilibrium with atmospheric carbon dioxide pressure as

with HCO

_J-

concentration in molli.

The mean concentration of HC0

₃-

is

87.0

/L eg/I.

Inclusion of concentration of HC0

₃-

to the ani on-sum increases it to

209

/Leg/I. Inclusion of HCO

_J-

in the anion

-sum

reduces the anion deficit to

81.0

/L eg/I.

Cation excess ranging from 29 to 197 /L eg/I in rainwater samples ha s been reported earli er at Alibagh, Colaba, Pune and Kalyan

⁵.

Previous studies on precipitation in India have also revealed an excess of cations 5 . 9.2

\.23 .

Mg2 + , Na+, K+ , F, cr and N0

₃-

in soi

l

extract were

determined as done in case of rain water samples. The

concentrations of ions and ratios of various ion s with

respect to Ca

²

+ (F/Ca

²

+ ,

N0₃-ICa²+,

Mg

2+/ci+,

(4)

52 ^INDIAN^{J RADIO}^&^SPACEPHYS. FEBRUARY 2002

Na+ /Ca

2+

and K+ /Ca

2

+) in th e soil are shown in Table 3. This shows that loca l soil can potenti ally affect the composition of precipitation. In India, the particulate

. k d " d 16

matter

IS

nown to pre om111 ate 111 coarse mo e

.

Coarse mode particulates origin ate fro m natural

sources,

whil e fine mode particulates are know n to have

anth

ro poge ni c sources. Kulshrestha

et al.²⁴

have tri

ed to separate

the anthropogenic and natural components in precipitati on based on data on mass size distribution of aerosols at Agra. The calculated co nce ntration and percentage contribution of coarse and fine fractions are listed in Table

4.

The spec ies Ca

2+

, Mg

2

+ and K+ are largely so il derived

(52-68 %),

whereas

58 %

SO/- is contributed from industrial so urces and

42 %

from natural sources.

3.4 Neutralization of acidity

Th e correlati on between N0

₃-

and SO/- with Ca2+, Mg2+and NH/ suggests that

'acidity

in rainwater is neutrali zed by the se alkaline species. Sample-to- sampl e variation of

pH

and of the ions Ca

2

+, Mg2 +, NH/, N0

₃-

and SO/- are shown in Fig.

I.

Hi gh

pH

valu es are observed to be acco mpani ed by high Ca2+

,

Mg2+ and NH / concentration. Vari ation of NO

_{J -}

and SO/

-

also follows a similar trend . These observations suggest that:

(i) Ca

2

+, Mg2+ and NH 4 + act as a ne utralizing agent!

buffer.

Table 3-Composition of soil and ratios of various ions with respect to Ca2+

Species Concentration in Ratio Soil Rain water (j.teq/g)

Ca²+ 11.9 F/Ca2+ 0.07 0.03

Mg2+ 17.2 NO)-/Ca2+ 0.21 0.22

NO)- 2.5 Mg2+/Ca²+ 1.40 0.45

Na+ 6.7 Na+/Ca²+ 0.56 0.40

cr

^10.1 ^K+^/Ca2+ ^0.23 ^0.06

K+ 2.8

F 0.86

(ii) ci+, Mg2

+

and NH4 + react with H

2

S0

1

and HN0

₃

to form their salts of NO]- and SO/

-.

The relative neutralization ac ti ons of NH/, Ca2+ and Mg2+ were examined by calcul

ating

the neutralization factor

2

1. The neutralization factors for Ca

2+

, Mg2+ and NH/ we re 1.75, 0.82 and 0. 1 8, respectively

.

As show n by the va lues of neu trali zat ion factor Ca and Mg pl ay an important role in neutralizing the ac idit y.

Similar results have been observed at Goa and Pun e25 . The study of aeroso ls at Agra reveals that Ca

2+ and

Mg2 + are predominantly coarse mod e parti cles and NH/ ex ists in fine mode

24

.

g-

:1.

o Z

r: < _ct:

f-;z:

Lll U ;z:

o u

o

2 4 6

^g

10

12:

14 16 ;8 20 2 2 2 4 26

~8

J On--rtJ637

SAMPLE NUMBER

Fig. I- Variation in concentration of different ions in rainwater samples

Table 4-Estimation of different ions (mg/l) from different sources in rainwater

Species % Composition of aerosol Rainwater lonnss Industrial Soil Sea

Fine Coarse concentration

(mg/I)

Ca²+ 47 53 2.14 2.13 1.00 1.13 0.01

Mg2+ 40 60 0.87 0.82 0.33 0.49 0.05

S04²- 58 42 0.80 0.70 0.41 0.29 0.10

CI- 36 64 1.13 0.43 0.15 0.28 0.70

K+ 69 31 0.11 0.10 0.07 0.03 0.01

Source: Kulshrestha el al,z4

(5)

MUDGAL el al.: RAIN WATER COMPOSITION AT AGRA 53

3.5 Ionic correlation

Correlation matrix is a common way of hypothesizin

g

potential precursors of ions in rainwater sam ples. Corre lation between ions suggests the likely sources of po llutants and also indicates the gaseo us reaction s occurring in the atmosphere.

Loga rithmic

correlation coeffici

ents between ion pairs are shown in Table 5. The hi ghest correlation appears fo r the ion pairs N0

₃-

and SO/

- (r

=

0.94)

indicating their origin from simil ar source areas. It is possibl e that a fraction of N0

₃-

0.63),

Mg2+ and SO/

Potassium is another terrigenous species but

is

not found to correlate with the other soil components Ca2+ and

Mg2+. Th e poor corre

lation

is presumab

ly due

to the fact that a greater contribution of K+ is from biomass combustion rather than soil

28

.

Biomass combustion

+ . I d²⁹

sources of K are ex tensive y reporte .

3.6 Factor analysis

Factor extraction ha s bee n calculated by princi pal component meth od

³⁰

by using the Stati sti cal Package for Social Sciences (SPSS ) programme. Factor analy sis was carried out on th e data in an attempt to determine the factors underlying the

inter-correl

ation between the measured species. According to thi s th e grouping of ions in each factor could be attributed to chemica

l, physical as

well as common sources. These resul ts have been presented in Table

6.

Factor

loadings

above

0.5

are deemed to be stati sti cally significane

^l.

This analysis identified four factors that contributed

88.7 %

of the variance to the data set.

Factor 1 explains about

47.0 %

of the total variance 2+ ?+ + + CI- NO

-

SO

2-

F- HCO

-

of Ca ,Mg-, Na , K , ,

3 ,

4"

3

and NH/. The high loading of Ca

²

+ , Mg2+ and Na

+

clearl y indicates the influences of

local

sources like soil. Soil around Agra is ca

lcareous

and is swept into the atmosphere by the wind adding to the total suspended particulate matter

³².

At Agra, the total suspended particulate matter ranges between

50

and

3 . h . d3? V ?3 h

600

fJ..g/ m dunng t e monsoon peno -. arm a- as confirmed that, in this part of India, a high

load of

soi

l-oriented particles is present in

the atmosphere,

Table 5-Correlation coefficient among measured parameters in rainwater

Species F CI

F 1.00

CI- 0.13 1.00

NO)- 0.29 0.37

SO/- 0.05 0.62**

HCO)- 0.26 0.26

Na+ 0.06 0.68**

Ca2+ 0.13 0.57**

K+ 0.26 0.23

Mg2+ 0.21 0.58**

NH/ 0.32 0.59**

Note: I tailed signal: p = 0.0 I *,0.001 **

*p- 0.01, ** p -0.001

NO) S042

1.00

0.94** 1.00 0.44* 0.42 0.72** 0.62**

0.57** 0.62**

0.59** 0.59**

0.63** 0.66**

0.56** 0.44*

Species

Ca2+ Mg2+ NH/

HCO) Na+ K+

1.00

0.38 1.00

0.42 0.72** 1.00

0.10 0.41 0.22 1.00

0.32 0.32 0.81** 0.32 1.00

0.50* 0.65** 0.22 0.47* 0.37 1.00

(6)

54

^INDIAN^JRADIO & SPACE PHYS. FEBRUARY 2002

Table 6-Factor analysis of concentration of chemical species in precipitation

Species Factor I Factor 2 Factor 3

F 0.67 0.16 0.16

C1- 0.83 -0.05 0.33

NO]- 0.86 0.42 0.11

S042- 0.88 0.45 0.Q7

HCO]- 0.90 -0.20 0.12

Na+ 0.85 0.27 0.29

Ca2+ 0.91 0.21 0.16

K+ 0.27 0.15 0.60

Mg2+ 0.68 0.31 0.10

NH/ 0.27 0.34 0.80

% Variance 47.0 29.2 12.5

Predicted Soil/ dust D.G.*sets/ Biogenic contributor emission Incomplete emission/

combustion combustion of fuel

*Diesel generator sets

which are predominantly alkaline in nature and consist mainly of carbonates and bicarbonates of Ca2+, Mg2+, Na+and K+. Thus, the grouping of soil derived species with the anions like CI-, N0

_3-

and SO/

-

indicates both direct emissions of these species from soil or by neutralization reactions in the atmospherel

⁷^•

Factor 2, being loaded on SO/- and N0

3

-, indicates an incomplete combustion of fuels due to vehicu

lar

exhaust as well as diesel generator (D

.G.)

sets. Agra city, whi ch has been always in the grip of power cuts due to local faults, uses D.G sets whose existing number}3 is approximately

30,000.

Factor 3 shows a strong dependence on NH/ and K+ and may be attributed to the widespread use of firewood as a domestic fuel in this region and the contribution of NH4 + from the decomposition of animal waste

.

4 Conclusion

Finall y on the basis of the chemical characteristics of wet deposition, it may be concluded that:

(i) Depositions are alkaline and dominated by Ca2+, HC0 3-, Mg2+, Na+. SO/- and cr.

(ii) Higher concentration of Ca 2

+ seems to be much more important than NH4 + for

neutralizing.

(iii)The ratios of F/Ca

²

+, Mg2+/Ca

²+,

Na+/Ca

²

+ and NO}

-/Ca²

+ in rainwater samples indicate that local soil plays a significant role in precipitation chemistry.

(iv) Factor analysis proved to be a powerful technique for relating the elemental concentrations of a large number of atmospheric pollutants to their sources.

Acknowledgements

The authors are extremely thankful to Dr S S Srivastava, Department of Chemistry, Dayalbagh Educational Institute, Agra, for his guidance and unstinted support throughout the study. They also thank Dr Ashok Kumar, Head, Chemistry Department, St. John's College, Agra, for providing the necessary facilities.

References

Dikaiakos J G. Tsitouris C G, Siskos P A, Melissos D A &

Nastos P, Almas Environ ( UK), 24B (1990) 171.

2 Minoura H & 1wasaka Y J, AlnlOS Enviroll (UK). 24 (1996) 39.

3 Galloway J N & Gaudry A. Almas Environ (UK). 18 (1984) 2649.

4 Moody J L, Pszenny A A P, Gaudry A, Keene W C.

Galloway J N & Polian G, J Geophys Res (USA), 96 (DIl), 20 (191) 769.

5 Khemani L T, Momin G A, Naik M S, Rao P S P, Safai P D

& Murty A S R, Almos Environ (UK), 21(5) (1987) 1137.

6 Lewis W M, Grant M C & Saunders J F, WaleI' Resow' Res (USA), 20 (1984) 1691.

7 Wagner G H & Steele K F, Almas Environ (UK), 21 (1987) 1353.

8 Zhao D, Xiong J, Xu Y & Chan W H, Almas Environ (UK), 22 (1988) 349.

9 Satsangi G S, Lakhani A, Khare P, Singh S P, Kumari K M

& Srivastava S S, AliI/OS Enviro/1 (UK), 32 (1998) 3783.

10 Shivashankara G P, Ranga K & Ramalingaiah Rao M, Indian J Environ Heallh. 41 (1999) 229.

11 WMO, World Melerological Organisalion. FOCI sheel No.5.

1989.

12 Conlrol of Urban Pollution Series (CUPS), CUPS 7: Cenlra!

Pol/ution Conlrol Board. New Delhi, India, 1987.

13 Mahadevan TN, Meenakshy V & Sathe A P, Mausam (India), 35 (1984) 87.

14 Weatherburn M W, Anal Chem (USA). 39 (1967) 971.

15 Charlson R J & Rodhe H, Nature (UK), 295 (I 982) 683.

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

Kapoor R K,Armos Environ (UK), 23 (1989) 757.

17 Rao P S P, Momin G A, Safai P D, Pillai A G & Khemani L T, Almas Environ (UK). 29 (1995) 2025.

18 Galloway J N, Keene W C & Likkens G E, J Geophys Res (USA), 10 I (1996) 6883.

19 Sanhueza E, Elbert W, Rodon A, Arias M C & Hermoso M, Almas Environ (UK), 41 B (1989) 170.

20 Khemani L T, Momin G A. Rao P S P. Safai PD. Pillai A G, Mohan K & Rao M G, AlfI10S Environ (UK), 28 (1994) 3145.

21 Saxena A, Kulshrestha U C, Kumar N. Kumari K M &

Srivastava S S, Almas Environ (UK), 30 ( 996) 3405.

22 Satsangi G S, Lakhani A. Khare P. Singh S P, Kumari K M

& Srivastava S S, Almas Environ (UK), 32 (1998) 3783.

23 Varma G S, Almas Environ (UK), 23 (1989) 2723.

(7)

MUDGAL et al.: RAIN WATER COMPOSITION AT AGRA

55

24 Kulshrestha U C, Saxena A. Kumar N. Kumari K M &

Srivastava S S, Indian J Radio & Space Phys, 24 (1995) 178.

25 Parashar D C. Granat L. Kulshrestha U C, Pillai A G, Naik M S. Momin G A. Rao P S P, Safai P D. Khemani L T, Naqvi S W A, Narverkar P V, Thapa K B & Rodhe H, A preliminary report on an Indo-Swedish project all atll/ospheric chell/istty. Stockhoill/ University. Report No.

CM-90, 1996.

26 Applin K R & Jcrsak J M. AIIIlOS Environ (UK), 20 (1986) 965.

27 Munger J W, Atll/os Environ ( UK). 16 (1982) 1633.

28 WolffGT,Atll/os Environ (UK). 18(1984)977.

29 Lacaux J P. Delmas R, Cros B & Andreae M O. J Geophys Res (USA), 97 (1992) 6195.

30 SPSS Jar Windows, release 7.5.1, Standard version (Copyright (c) SPSS Inc., 1989-1996. also available on Internet site [http://www.SPSS.com/.]). 20 Dec. 1996.

31 Comrey, A first coarse in Jactor analysis (Academic Press.

New York), 1973, p. 109.

32 Mudgal R, Sharma B, Upadhyay R & Taneja A. Indian J Radio & Space Phys. 29 (2000) 127.

33 Rajalakshmi T K. Agra '.I' Dilell/II/a Frontline. October. Vol.

8, 1999, pp. 42-43.