Abstract

Urban air pollution is rapidly becoming an environmental problem of public concern worldwide. It can influence public health and local/regional weather and climate. In the present study, the washout effect of rain on surface air pollutants (PM_2.5, O₃, NO_x, NO₂, NO ,SO₂, and CO) has been investigated over three ambient air quality monitoring stations (Gurugram, Rohtak and Panchkula) in Haryana state in Northern India. The airborne particulate pollutants data were collected by the Haryana State Pollution Control Board (HSPCB) for a period of one year (January 01, 2016 to December 31, 2016) at Vikash Sadan location in Gurugram, MDU location in Rohtak and HSPCB Sector 6 location in Panchkula.. The rainfall data for the study period has been taken from rainfall Statistics in India-2016, India Meteorological Department. The regression correlation analysis has been performed to investigate the effect of rainfall on surface air pollutants. The concentration of air pollutants are found to decrease with increase of rainfall. It has been observed that, PM_2.5 is most effected and O₃ is least effected by rainfall. Most of the pollutants under study show statistically significant negative correlations between their concentrations and rainfall due to washout and convection but their sensitivity towards rain significantly varies. O₃ shows either positive correlation or weak negative correlation with rainfall depending on the meteorological season and observing site.

1. Introduction

Air pollution is one of the most challenging problems which our cities face today as their atmospheres are getting highly polluted due to discharge of gaseous pollutants and particulate matters from various household, industrial and vehicular sources. Because of the great impact of air pollutants on human health and ecological environments, it is important to better understand the removal mechanism of these pollutants from the atmosphere. The removal of air pollutants by rainfall remains of great interest to the scientific community and many theoretical and experimental studies carried out to understand the wet scavenging of air pollutants by rain fall. ^{1, 2, 3, 4, 5}. The washout mechanism of sulfur dioxide (SO₂) and Nitrogen dioxide (NO₂) by rainfall has been a main global and regional concern, since these compounds play an important role in producing acid precipitation. Carbon monoxide (CO) and ozone (O₃) are much less soluble in water than NO_x (NO₂ + NO) & SO₂ ⁶. O₃ concentration was observed to increase under rainy conditions ⁷. It may be due to vertical mixing of the stratospheric and tropospheric O₃ concentrations during convective rain activity and thunderstorm. A significant negative correlation between monthly averaged aerosol concentration and rain intensity has been reported over some region in Russia ⁸. Influence of rainfall on concentration of PM₁₀ and NO_x has been studied at Gurgaon city ² and reported negative correlation between rainfall and air pollutants PM₁₀ and NO_x. The inter correlation of air pollutants in different meteorological seasons has been studied in Haryana state ⁹ and observed significant variation in correlation in monsoon and other seasons. Since it is very difficult to separate the washout effect from other gas – phase processes such as dry deposition, atmospheric mixing, and chemical transformation and so the correlation analysis between various air pollutants and rainfall could be helpful for investigating their interactions and long term trends ¹⁰.

In the present study, the washout effect of rain on surface air pollutants (PM_2.5, O₃, NO_x, NO₂, NO, SO₂, and CO) has been investigated over three rapidly developing cities (Gurugram, Rohtak and Panchkula) during the year 2016. The airborne particulate pollutants data were collected by the Haryana State Pollution Control Board for a period of one year (January 01, 2016 to December 31, 2016). The rainfall data for the study period has been taken from rainfall Statistics in India-2016, India Meteorological Department ¹¹. The regression correlation analysis has been performed to investigate the effect of rainfall on surface air pollutants.

2. Description of the Study Area

Haryana is the 20^th state of India that came into being on 1^st November 1966. It is situated in the North Western region surrounded by Himachal Pradesh from North, Uttrakhand from North East, Rajasthan from the South, U.P and Delhi from East and Punjab from North West. Three districts, Gurgaon, Rohtak and Panchkula of Haryana state have been selected for the present study. The brief information’s about sampling sites are given in Table 1. The detail information’s about the sampling sites are given in our previous publication ⁹.

3. Methods and Materials

For the present study, we used ambient air quality data collected by the Haryana State Pollution Control Board (HSPCB) for a period of one year from January 01, 2016 to December 31, 2016. The PM_2.5, O₃, NO_x, NO₂, NO, SO₂, and CO data were collected by the HSPCB using their respective analyzers having sampling duration 24 hrs as accepted by the Environmental Protection Agency (EPA) of the U.S.A. and the Central Pollution Control Board (CPCB) of India. Oxides of nitrogen analyzer use proven chemi-luminescence technology to measure NO, NO₂ and NOx in ambient air quality. The CO analyzer works on the principle of non-dispersive absorption and SO2 analyzer operates on the principle of light absorption, where the SO2 molecules are excited by absorbing light at one wavelength and later decay to a lower energy state by emitting UV light at a different wavelength which is proportional to SO2 concentration. The O₃ analyzer also works on the absorption principle i.e. O3 molecules absorbs UV light at 254nm wavelength. The degree of absorption is directly related to O₃ concentration as described by Beer-Lambert law ^{12, 13, 14}. O₃ measurements are automatically corrected for gas temperature/pressure changes and can be displayed in units of ppm, µg/m³ or mg/m³. The particulate pollutants samples were collected using commercially available and calibrated Respirable Dust Sampler (APM460BL) using appropriate filters ¹⁵. In this study the linear regression correlation analysis, which establish functional relationship called Regression equation between study variables and explanatory variables and determine coefficient of correlation and coefficient of determination, has been performed between rainfall & PM_2.5, O₃, NO_x, NO₂, NO, SO₂, and CO for different meteorological seasons to investigate the relationships between them. This analysis will give an idea about washout effect of rainfall on PM_2.5, O₃, NO_x, NO₂, NO, SO₂, and CO over the sampling sites and also how rainfall affects the PM_2.5, O₃, NO_x, NO₂, NO, SO₂, and CO concentration under different meteorological conditions.

4. Results and Discussion

In this study, we have analyzed and correlated the surface air pollutants (PM_2.5, O₃, NO_x, NO₂, NO, SO₂, and CO) data with Rainfall at Gurugram, Rohtak and Panchkula in Haryana state Northern India for the period from January 01, 2016 to December 31, 2016. The effects of rainfall on concentration of PM_2.5, O₃, NO_x, NO₂, NO, SO₂, and CO in ambient air were analyzed using regression analysis. The graphs are presented in Figure 2 - Figure 25 and results of regression analysis are presented in Table 2 - Table 4.

In Gurugram, the monthly average rainfall ranges between 0.5 mm and 186.8 mm. Average rainfalls varies between 0 – 16.2mm in spring, 0.5mm – 38.8mm in pre-monsoon, 35.7mm-186.8mm in monsoon and 0-15.4mm in post monsoon season. Minimum rainfall has been recorded in the month of April and maximum in the month of July. No rainfall recorded in the months of January, February, November and December. The average rainfall in Gurugram from January 2016 to December 2016 is depicted graphically in Figure 2. The season wise variations of rainfall and its influence on the concentration of PM_2.5, O₃, NO_x, NO₂, NO SO₂ and CO are presented in Figure 3 – Figure 9. The results of regression analysis study are presented in Table 2 – Table 4.

Regression analysis reveals that, in Gurugram, PM_2.5, O₃, NO_x, NO₂, NO SO₂ and CO are negatively correlated with rainfall, but their degree of correlation significantly varies. PM2.5 is found to be most affected while O₃ is found to be least affected by rainfall. The degree of correlation based on the coefficient of determination is found as PM_2.5 (r² = 0.7937) > CO (.r² = 3671) > SO₂ (r²= O.2362) > NO₂ (r² = 0.2311) > NO_x (r² = 0.2170) > NO (r² = 0.1138) > O₃ (r² = 0.0837).

In Rohtak, the monthly average rainfall ranges between 0.2 mm and 122.7 mm. Average rainfalls varies between 0 – 15.3mm in spring, 16.5mm – 21.5mm in pre-monsoon and 8.0mm – 122.5 mm in monsoon. Minimum rainfall has been recorded in the month of February and maximum in the month of August. No rainfall recorded in the months of January, April, October, November and December. The average rainfall in Rohtak from January 2016 to December 2016 is depicted graphically in Figure 10. The season wise variations of rainfall and its influence on the concentration of PM_2.5, O₃, NO_x, NO₂, NO SO₂ and CO are presented in Figure 11 - Figure 17. The results of regression analysis study are presented in Table 2 - Table 4.

Regression analysis reveals that, in Rohtak, PM_2.5, NO_x, NO₂, NO SO₂ and CO are negatively correlated with rainfall, but their degree of correlation significantly varies. The behavior of O₃ is found to be different from other pollutants and is positively correlated with rainfall (r² = 0.7338). The degree of correlation based on the coefficient of determination is found as PM_2.5 (r² = 0.5183) > CO (.r² = 0.4543) > NO (r² = 0.3445)> SO₂ (r²= O.2853) > NO_x (r² = 0.1635) > NO₂ (r² = 0.0763).

In Panchkula, the monthly average rainfall ranges between 0.3 mm and 165.6 mm. Average rainfalls varies between 0.3mm – 21.6mm in spring, 5.2mm – 62.5mm in pre-monsoon, 37.8mm – 165.6mm in monsoon and 2.3mm – 17.6mm in post monsoon season. Minimum rainfall has been recorded in the month of April and maximum in the month of July. No rainfall recorded in the month of November. The average rainfall in Panchkula from January 2016 to December 2016 is depicted graphically in Figure 18. The season wise variations of rainfall and its influence on the concentration of PM_2.5, O₃, NO_x, NO₂, NO SO₂ and CO are presented in Figure 19 - Figure 25. The results of regression analysis study are presented in Table 2 - Table 4.

Regression analysis reveals that, in Panchkula, PM_2.5, O₃, NO_x, NO₂ and SO₂ are negatively correlated with rainfall, but their degree of correlation significantly varies. CO (r² = 0.1443) and NO (r² = 0.1821) are positively correlated with rainfall but their correlation is observed to be very week. The degree of correlation based on the coefficient of determination is found as PM_2.5 (r² = 0.3213) > SO₂ (r²= O.2524) > NO₂ (r² = 0.1557) > NO_x (r² = 0.1026) > O₃ (r² = 0.0933) ⁵.

It was noted that, the degree of correlation significantly varies from pollutant to pollutant with same level of rainfall. Also, statistical correlation depends on the sources of production and other removal mechanism of pollutants at the the monitoring site. The negative correlation between PM_2.5 and NO_x in Gurugram confirm the findings of previous study ². The significant negative correlation between CO and rainfall at Gurugram and Rohtak is interesting in view of low water solubility of CO ^{16, 17}. It may be due to other meteorological processes in addition to the wash out effect ⁸. One of these processes is the active convection that under rainy conditions mixes air from aloft with air nearer the surface and this dilute compound with emission sources near the Earth surface. Weak negative correlation between O₃ and rainfall in Gurgugram and in Panchkula and positive correlation in Rohtak may be due to strong convection during heavy rain in monsoon which leads low level O₃ that vertically mixes upper level O₃ in the upper atmosphere down to the troposphere ¹⁸. The washout effect of rainfall on NO_x, NO₂ and NO is observed to be negative in all three monitoring sites except for NO in Panchkula but their degree of correlation are of significance.

5. Conclusion

The statistical investigation of influences of rainfall on concentration of PM_2.5, O₃, NO_x, NO₂, NO ,SO₂, and CO were evaluated for the year 2016 at three monitoring sites of Haryana State , using regression analysis and the findings are stated below.

(i) In Gurugram, PM_2.5, O₃, NO_x, NO₂, NO SO₂ and CO are negatively correlated with rainfall, but their degree of correlation significantly varies. The degree of correlation based on the coefficient of determination is found as PM_2.5 (r² = 0.7937) > CO (.r² = 3671) > SO₂ (r²= O.2362) > NO₂ (r² = 0.2311) > NO_x (r² = 0.2170) > NO (r² = 0.1138) > O₃ (r² = 0.0837).

(ii) In Rohtak, PM_2.5, NO_x, NO₂, NO SO₂ and CO are negatively correlated with rainfall, but their degree of correlation significantly varies. The behavior of O3 is found to be different from other pollutants and is positively correlated with rainfall (0.7338). The degree of correlation based on the coefficient of determination is found as PM_2.5 (r² = 0.5183) > CO (.r² = 0.4543) > NO (r² = 0.3445)> SO₂ (r²= O.2853) > NO_x (r² = 0.1635) > NO₂ (r² = 0.0763).

(iii) In Panchkula, PM_2.5, O₃, NO_x, NO₂ and SO₂ are negatively correlated with rainfall, but their degree of correlation significantly varies. CO (r² = 0.1443) and NO (r² = 0.1821) are positively correlated with rainfall but their correlation is very week. The degree of correlation based on the coefficient of determination is found as PM_2.5 (r² = 0.3213) > SO₂ (r²= O.2524) > NO₂ (r² = 0.1557) > NO_x (r² = 0.1026) > O₃ (r² = 0.0933).

(iv) It has been observed that, PM_2.5 is most effected and O₃ is least effected by rainfall.

The result of this study will be useful in further investigation in air pollution research to analyze the relative influence of the washout effect on the air pollutants and the impact of precipitation on air quality.

Acknowledgements

The authors thank the Haryana State Pollution Control Board for providing the O3, NOx, NO₂, NO, CO, SO₂ and PM2.5 data for the measuring sites Gurugram (NCR), Rohtak and Panchkula & Indian Meteorological Department for providing Rainfall data. The author Dr. Ram Chhavi Sharma is grateful to Shree Guru Gobind Singh Tricentenary (SGT) University Gurugram for providing an excellent research environment, infrastructure facilities and encouragement during the course of this work.

References