INTRODUCTION:

Man is making rapid strides in science and technology. His eagerness to increase his comforts by utilising the natural resources leads to rapid industrialisation and abnormal urbanisation, which consequently witness excessive air pollution. To impose proper control measures, regular air quality monitoring is essential, which is a time consuming and tedious procedure involving heavy utilisation of money and manpower. The current concern on environmental pollution and sustainable development makes it imperative to have air quality models which serve as a tool for providing objective answers to many questions regularly faced by decision makers in the fields of air pollution control, transportation planning and land use planning.

MONITORING METHODOLOGY FOR SAMPLE COLLECTION:

The entire area of the city is divided into five zones as shown in Figure 1., and a survey station was selected in each of these areas. The growth pattern of Coimbatore city does not follow any zonal pattern. Hence, all the survey areas chosen are categorised as industrial and mixed. Based on a questionnaire survey on ambient air quality from the general public, it was understood that the main source of air pollution was the dust arising from roads and vehicular emissions. Hence, the major air pollutants considered were SPM, SO2 and NOx, which were observed using High Volume Samplers (ENVIROTECH APM 410-411 MODEL). The analytical procedures for determination of pollutants were as per Bureau of Indian standards given by Methods for measurement of air quality. The concentrations of the air pollutants were measured once a week on a particular day in each of these stations. The ambient air quality standards prescribed by Central Pollution Control Board, India for the major air pollutants are given in Table 1. The daily 8 hours concentrations observed during January to March 2000 in each of these stations are given in Table 2.

TABLE 1: AMBIENT AIR QUALITY STANDARDS BY CENTRAL POLLUTION CONTROL BOARD, INDIA

MODELING METHODOLOGY FOR THE ANALYSIS OF MONITORED DATA IDEAL MODEL

Strictly speaking, an ideal model should have the following qualities:

1. Physically realistic and accurate.
2. Universally applicable with respect to spatial and temporal scale, meteorology and topography.
3. Supported by easily available input data.
4. User friendly.
5. Properly validated against observations.
6. Verified against real database.

However, ideal models rarely exist. A compromise should be made between simplicity and sophistication, accuracy and practicality. The accuracy of the model depends on its input data. Depending on the availability of input data, decisions on the model type, its degree of sophistication, range of applicability and other important features can be decided.

CLASSIFICATION OF AIR QUALITY MODELS:

Air quality models can be categorised into four generic classes. Analytical models are based on the analytical explicit solutions of diffusion equations, which require a large database. Numerical models basically deal with different types of numerical approximations in the solutions of partial differential equations representing the physics of atmospheric dispersion phenomena. Physical models are small-scale representations of atmospheric dispersion. They include wind tunnels, smog chambers, etc. Statistical models are based on the time series (or any other trend) analysis of meteorological and air quality data. These models are useful for real time analysis and short term forecasting. Among these modelling techniques, statistical modelling is adopted for analysis of the monitored data since the observed data are only for a short period.

REGRESSION ANALYSIS:

This analytical tool performs linear regression analysis by using the "least squares" method to fit a line through a set of observations. Using this, one can analyze how a single dependent variable is affected by the values of one or more independent variables. In this case study, regression analysis is done to know how a pollutant concentration in this city is affected by such factors as wind temperature, pressure, direction and speed. One can apportion shares in the concentration measure to each of these three factors based on a set of monitored data, and then use the results to predict the pollutant concentration in the future. The monitored data at the five selected stations in Coimbatore City are analyzed by multi regression and the evolved models for prediction of each of the pollutants as a function of the chosen meteorological parameter are given below with respective regression coefficients.

SPM= -82.0703 T -80.5704 P -0.76381 WD -2.03456 WV +64531.68; R = 0.5

SO2= 2.397 T -1.1481 P +0.016 WD +1.173 WV +831.5413; R = 0.2

NOx=5.728 T +3.2582 P -0.0636 WD +2.1923 WD +2.192 WV -2601.85; R=0.36

Where T- temperature, P - pressure, WD - wind direction and WV - wind velocity. Back forecasting is done with the evolved models and the validation is done with the plot between the observed and predicted pollutant concentration, which is given in Fig 2 (Table 3).

CONCLUSION:

The comprehensive survey of air quality at five selected stations is compared with ambient air quality standards of Central Pollution Control Board.
1. The SPM concentration in the ambient air in Coimbatore is extremely high than the threshold limit.
2. The NOx and SO2 concentrations are within the prescribed limits but show-increasing trend.
3. The wind condition is observed to be calm in Gandhi Park and Gandhipuram resulting in high level of SPM in these localities.
4. Traffic congestion and high rise buildings are the main cause for increase in pollutant concentrations.
5. The wind direction, in all the survey stations is predominantly in the western direction during the sampling periods.
6. The evolved models can be used for any city having similar meteorological conditions.
7. The lower regression coefficients are due to less number of observations, which can be increased by having more number of observations.

ACKNOWLEDGEMENT:

The first author expresses her sincere thanks to All India Council for Technical Education for providing funds to carry out this project.

FIGURE-1: COIMBATOTE CITY MAP

TABLE 2: AMBIENT AIR QUALITY DATA MEASURED DURING JANUARY 2000 TO MARCH 2000

TABLE 3: PREDICTED AMBIENT AIR QUALITY FOR THE PERIOD OF MONITORING BY BACK FORECASTING

SPM	SO2	NOx
792.2148	40.30383	35.75872
679.0027	53.19073	51.76784
1033.936	53.79536	56.00047
748.6561	55.3019	50.38444
705.2172	53.11992	44.42507
683.5028	52.17048	45.06546
673.7244	56.43866	46.95907
715.7571	57.3609	47.98224
665.1273	53.80944	44.99014
818.1573	52.25684	45.10428
537.7846	56.45623	62.08221
651.0301	53.36596	45.74696
651.1127	52.43524	48.11174
584.6219	53.48629	51.11254
510.8096	56.29097	52.70571
527.6341	55.7971	56.27942
648.8821	56.86734	56.30452
812.1404	54.35388	63.99119
522.1728	53.81278	56.12057
79.69658	70.36588	85.52403
562.0611	50.12992	48.70276
459.5031	56.01232	61.84726
399.5711	56.42617	63.57014
439.4394	54.103	58.65507
685.5655	51.65555	60.38168
766.6853	57.41448	61.52153
750.16	55.0552	56.771
660.8009	60.97387	50.7606
1017.738	46.08646	27.14158
685.9746	52.52975	53.15251
534.2126	58.72785	57.48875
583.1972	57.88739	55.1563
679.7742	51.87525	45.94797
631.7892	48.31716	44.92162
606.304	52.90737	49.02587
622.2998	52.87585	49.38477
511.9324	57.21381	59.13863
583.4044	53.0627	55.10496
511.9324	57.21381	59.13863
403.6244	53.43234	57.32173

FIGURE2.VALIDATION OF THE EVOLVED MODELS

Address:

1.) Lecturer in Civil Engg.,
Coimbatore Institute of Technology,
Coimbatore -641 014,
Tamilnadu.
2.) Asst.Professor in Civil Engg. Coimbatore Institute of Technology,
Coimbatore -641 014,
Tamilnadu