Modelling of Traffic Noise Pollution at Signalized Intersection: A Case Study of Baneshwor Intersection

The growth of motor vehicles and human activities in major cities has led to noise pollution problems due to traffic movement and other motorized activities in these confined areas. Road traffic noise decreases human efficiency and productivity, and also lowers the quality of living standard. The people residing near such regions are worried about the environmental effects of traffic noise and its impact on human health. Assessing traffic-generated noise problems and their effects requires planning actions, such as traffic noise modeling, to develop a preventive and mitigatory plan that provides a framework for reducing noise levels. The study was conducted which presents the status and modelling of noise pollution at signalized intersections in New Baneshwor, Kathmandu, Nepal. Noise level data collection was conducted through a sound level meter (Lutron Model: SL-4023SD) whereas simultaneously classified traffic volume, composition, and speed data was collected using a videographic survey. A video camera was used to record the movement of traffic vehicles to determine traffic volume, traffic speed, and traffic composition during peak hours. The sound meter was used simultaneously with a video camera to measure the sound level at that time interval. Simultaneous measurement of the video graphics survey and sound level survey was done with the help of the time setting function in the sound meter with that clock time exactly as seen in the video graphics survey. At each station, the location was marked and noted through GPS devices on smartphones, and the sound level meter (SLM) was used for sound measurement at that station. For the modeling of noise induced by road traffic, statistical analysis was carried out using SPSS software whereas Arc GIS was used for mapping urban road noise. The findings show that noise levels at the selected intersection where we had carried out our study, exceeded the allowable noise level limit which was set by the Ministry of Environment Science and Technology, Government of Nepal. A statistical model was developed based on multiple linear regression (MLR) analysis, which was carried out between traffic noise as the dependent variable and traffic volume, traffic composition, and traffic speed as independent variables that help to predict noise levels. In this study, for the determination of road traffic noise, A-weighted equivalent continuous sound level (LAeq) is used as the dependent variable, while the independent variables include the percentage of heavy vehicles, traffic volume, and speed. The paired t-test was employed to assess the goodness-of-fit of the developed model. The Durbin-Watson test was conducted to check for the independence of the statistical data observations. An ANOVA test was used to examine homoscedasticity, and the results were found to be satisfactory. A histogram with an overlaid normal curve was created to verify whether the residuals of the variables followed a normal distribution. Also, the modeled noise level was compared with measured noise levels obtained from the sound level meter and the validity of the prediction model was confirmed using statistical tools. The noise levels induced by traffic were positively correlated with the volume of moving vehicles and composition of mixed traffic, i.e., the presence of heavier vehicles, whereas it was negatively correlated with the speed of vehicle. With the rise in traffic volume, there was an increase in noise level. The distribution of traffic has also an impact on the noise level produced on roads. Motorcycles have dominated the traffic composition, which become a key factor in influencing the traffic noise level. At intersections, the speed of vehicles was found to be negatively correlated with traffic noise level due to having a lower speed value rather than at free road sections due to congestion. At low speeds, the impact of vehicles on noise measurement was greater, and noise level was found to be higher.