EFFECT OF TRAFFIC VOLUME ON AMBIENT AIR QUALITY ON MULTILANE DIVIDED URBAN ROADS AND SIGNALIZED INTERSECTIONS

Abstract

Road traffic emissions are one of the major sources of air pollution in urban areas. Various hazardous gases, including Carbon Monoxide (CO), Nitrogen Oxides (NOx), Hydrocarbons (HC), Particulate Matter (PM2.5 and PM10), Sulphur Dioxide (SO2), Carbon Dioxide (CO2), Formaldehyde (HCHO), Volatile Organic Compounds (VOC), Lead (Pb), Ammonia (NH3), and others, are present in the air across any urban roadway network. The concentration of air pollutants close to the main traffic routes is much higher than the regional background levels, putting the local people in danger and exposing them disproportionately to air pollution caused by traffic. The efficient investigation of the status of air quality in urban areas on both road mid-block sections and signalized intersections needs to be performed for better planning and designing of roadway network considering traffic volume growth. The present study deals with measuring, analysing, and modelling of concentrations of different pollutants observed in three medium scaled cities namely Warangal, Tirupathi and Vijayawada. The present study measures the concentration of six different pollutants namely Carbon Monoxide (CO), Carbon Dioxide (CO2), Formaldehyde (HCHO), Total Volatile Compounds (TVOC), Particulate Matter (PM2.5), and Particulate Matter (PM10) using portable equipment. The study also obtains field data related to traffic volume, temperature, signals, queuing, proportional share of different types of vehicles and Air Quality Index (AQI) at road mid block sections and signalized intersections. The study carries out a detailed analysis on pollution data, volume data and AQI data obtained from different locations. The study also analyses the variation of concentration of different pollutants and AQI with respect to traffic volume and time of the day. The present study proposed concentration model with respect to different pollutants (CO, CO2, HCHO, TVOC, PM2.5, and PM10) considering traffic flow observed at road mid-block sections and signalised intersections. The other variables such as proportion of vehicle types, temperature and signal control parameters are also included in the model development for predicting test estimate of concentration and AQI index. The present study also developed an emission model to estimate CO, CO2 and O2 emissions using age of the vehicle, emission norms and vehicle type as independent variables. The study also implemented different machine learning tools like Support Vector Regression (SVR) and Artificial Neural Networks i (ANN) to develop air concentration and AQI models. The validation of models was performed successfully with different set of data collected in the field indicate their suitability under mixed traffic condition. The sensitivity analysis is performed with different combinations of model input variables namely traffic volume and percentage of 3W. The status of air quality is defined based on AQI values obtained when there is incremental change in the traffic volume from 0 to 7000 PCU/hr and percentage of 3W from 0 to 30%. The concentration models developed in the study indicate that the variation in variables such as traffic volume, proportional share of vehicle types and signal control parameters would have a significant change in the concentration of pollutants. The study also identified from the emission model that age of the vehicle, emission norms and vehicle types are responsible factors for emissions. The study also signifies that the percentage of three wheelers could interpret the change in AQI. The findings of the study also illustrate the status of air quality is in moderate level in some regions and in satisfactory condition in some locations of the study areas according to Indian Air Quality Index (IND-AQI) standards. The study also infers that ANN models performed better in predicting the concentration of pollutants and AQI compared to MLR and SVM models. The results of the study provide insights to develop empirical guide for urban transportation planners and policy makers and traffic system designers to predict the level of pollution and AQI at various busy traffic facilities in the cities with similar characteristics.