Assessment of Traffic Noise at Kolar Gold Fields, India

MOHIT KUMAR*+, SURENDRA ROY** AND PIYUSH GUPTA*** Traffic noise produces environmental problems on surroundings. At Kolar Gold Fields (KGF), traffic load at different locations can cause noise pollution. Therefore, a detailed study was carried out at KGF in commercial, residential and silence zone. Different monitoring locations were identified in these zones and A-weighted equivalent sound pressure levels (Leq) were measured. The study was conducted on different weekdays and weekend for different periods of the day and compared with ambient noise standards. Since many variables were noted during noise monitoring, therefore, a multiple regression analysis was also carried out to develop a regression model for prediction of Leq. Stepwise regression procedure was followed for the selection of most influencing variables. Model adequacy was checked using normal distribution with zero mean and constant variance, difference between F-observed and F-critical, and correlation between measured and predicted noise levels. Key words: Traffic noise, weekday, weekend, Kolar Gold Fields (KGF), regression analysis Introduction Transport noise originates from the road traffic, aircraft and rail traffic etc. Traffic noise creates more disturbance to people than other noise sources since the road network is usually laid through the very core of a rural or an urban habitat, while the rail networks and the airports are generally situated on the peripheries of the cities and villages. All vehicles produce noise from the gear box and exhaust system. Heavy vehicles also produce rattles, squeaks and vibrations during movement on the roads according to degree of loading and age. Noise levels vary with the traffic density and the time of the day1. With a majority of the Indian population preferring to stay in urban areas rather than the rural areas, a great deal of noise pollution can be expected in these regions. Apparently, many studies have been conducted in various cities of India like Lucknow, Vishakhapatnam, Jaipur, Kolkata etc2-6. A study, carried out by the Central Pollution Control Board has shown that the noise level at most of the places exceeded the permissible limit in Delhi. Mumbai too suffers from high levels of noise pollution7. However, noise pollution in the developing areas of the country has been overlooked. Ravichandran et al. (2000)8 measured noise levels at Pudukkottai, Tamilnadu and found that vehicular traffic and pressure hours are the main cause of noise pollution in the city as reported by Datta et al. (2006)9. Noise causes interference in conversation, disturbance in sleep, annoyance, mental disorders, hearing loss, and adverse physiological and psychological impacts10-11. *Under Graduate Student, Department of Mechanical Engineering, National Institute of Technology Calicut, NIT Campus, Kozhikode-673601, Kerala, India ** Scientist E1 & Head, Environmental Engineering Department, National Institute of Rock Mechanics, Kolar Gold Fields-563117, Karnataka, India *** Scientist B, Environmental Engineering Department, National Institute of Rock Mechanics, Kolar Gold Fields-563117, Karnataka, India + Corresponding author: Telephone: +91 9567100674; e-mail:


etc. commercial areas. and to develop regression model for the prediction of noise levels. Movement of vehicles on the roads may affect these areas. 1. The noise levels for different zones were 2 . The instrument was held at arm’s length while making the measurement. Materials and Methods At KGF. The monitoring locations are shown in Fig. In order to protect the health of the people. Fig. a detailed study at KGF was conducted to assess the ambient noise for different categories like commercial. Considering this. courts. This standard limits the sound level for different times of the day.2 m from the ground level. The measurement range of the instrument was set from 50 dB to 120 dB. Therefore. schools. 1st Cross Robertsonpet and 6th Cross Geetha Road) and silence zones (Government hospital. the traffic noise assessment was carried out for different zones like commercial areas (Anupama Plaza.In this study. the town of Kolar Gold Fields (KGF) was selected for traffic noise assessment. 2000 of the Environmental (Protection) Act. hospitals. 1986. In this township. are located at different distances from the roads. residential and silence zone during different times of weekday and weekend. residential areas (8th Cross Swarnanagar. the noise generated by traffic movement needs to be monitored in terms of ambient noise to determine whether the noise criteria specified in the regulations are complied with or not. The instrument was kept at an elevation of 1. residential areas. Bhagwan Mahavir Jain College and District Court). near traffic light and 1st Cross Mini Ibrahim Road). the Government of India has made ambient noise standards for different categories of the areas under the Noise Pollution (Regulation and Control) Rules. 1: Noise monitoring locations at KGF The noise levels were measured using a Bruel & Kjaer model 2239A Integrating Sound Level Meter.

A-weighted equivalent continuous sound pressure level (Leq) measurements were carried out at different distances from the centre of the road. 2) indicating that more people used to come for shopping in the evening. residential and silence zone are 65 dBA. residential and silence zone. noise levels during both the periods were higher than the standards set for the residential areas.00 noon to 2.e. Shopping and return of people from their offices could be a few possible reasons. The monitoring was done during weekdays and weekend at each location. In commercial areas.00 PM to 8. the noise levels during the midday and the evening were almost the same. At the 1st Cross Robertsonpet. the noise levels were higher during the evening than the midday. the noise levels during the evening were higher than the midday because the number of passing vehicles was higher (Fig. The numbers and types of vehicles passed during monitoring period were also noted. therefore. which varied from one location to other. 3 . To compare the noise levels with the standards. number of vehicles passed along the road and distance from the location is also given in this Table. which could have negative impacts on people. traffic noise study was conducted for these two periods considering one as less sound generating period and other as high sound generating period. Therefore. traffic load during the evening is more compared to the midday. Fig. and control measures are as under: Traffic noise on weekdays The noise levels monitored at different locations during midday and evening on weekdays are shown in Table 1. the average of the value was evaluated for each location. Results and discussion Ambient noise standards for day time as per the Indian Regulation for commercial.00 PM (Evening) for duration of 10 min. since the court functioned only during the day time. Since the study was being carried out to assess the noise. This was due to the fact that the total number of two-wheelers and autos passing during the midday and the evening period were almost the same. However. Each zone represents the values of noise level for three locations. 55 dBA and 50 dBA respectively12. Generally. The monitoring locations. therefore. 2 shows the variation in noise level for commercial. indicating noise as a serious threat to the health of the people in this region. people come in the evening for shopping. The noise levels exceeded the limit during both the periods for commercial areas. The analysis of traffic noise on weekdays and weekend.monitored during 12.00 PM (Midday) and 6. i. no readings were taken during the evening period at the court. The other two residential locations produced similar trends as observed in the commercial areas.

2 73.6 66.9 69.96 4.1 75. areas Robertsonpet Day 1 Day 2 Day 3 6th Cross.3 74. Swarnanagar Day 1 Day 2 Day 3 Silence zones Government Hospital Day 1 Day 2 Day 3 Bhagwan Mahavir Jain College Day 1 Day 2 Day 3 District Court Day 1 Day 2 Day 3 4 .67 6.9 74. Mini Ibrahim Road Day 1 Day 2 Day 3 Residential 1st Cross.7 73.9 68.4 70.0 66.2 75.6 75.6 68.6 74.3 75.6 22.93 6.4 67.96 11.5 72.15 34.7 73.5 74.78 6.88 Distance (m) Auto 115 61 100 38 30 38 28 39 33 5 5 6 31 28 46 26 23 33 115 110 124 19 28 32 45 44 72 Car 19 15 14 8 7 10 9 3 2 1 2 7 9 4 11 7 13 18 6 8 9 8 13 14 15 Bus 3 12 7 4 6 5 1 1 2 2 2 7 5 5 3 1 1 - Truck 2 2 4 4 6 2 2 1 5 4 6 9 9 8 3 4 1 6 5 5 - Auto 117 94 82 35 43 37 33 36 32 1 6 8 14 24 30 29 26 25 100 67 100 19 14 21 - Car 20 14 17 18 11 13 2 1 6 1 9 4 4 5 7 7 10 17 4 8 7 14 - Bus 4 8 9 4 2 5 1 2 2 1 1 1 6 8 6 1 1 - Truck 3 1 2 3 4 2 5 2 3 7 3 4 4 6 3 2 - Commercial Traffic Light areas Day 1 Day 2 Day 3 Anupama Plaza Day 1 Day 2 Day 3 1st Cross.6 67.4 72.2 74.5 76. Geetha Road Day 1 Day 2 Day 3 8th Cross.9 78.9 75.7 75.7 66.4 75.6 72.4 76.9 74.6 75.4 73.95 8.42 3.0 78.0 76.4 71.6 77.6 69.0 77.2 74.6 68.4 Twowheeler 179 139 156 118 115 170 63 85 78 63 35 50 104 119 129 84 102 83 144 136 131 96 100 99 6:00 PM to 8:00 PM Leq dB(A) 75.7 73.7 74.3 76.6 74.4 73.0 71.6 76.5 68.Table 1: Noise levels monitored at different locations for different types of vehicles on weekdays 12 Noon to 2:00 PM Categories Locations Days Twowheeler 159 129 158 143 106 123 45 66 51 51 41 57 102 110 84 85 76 68 126 151 133 134 103 92 82 103 136 Leq dB(A) 73.

Fig. Since the noise made by the autos and buses are more compared to two-wheelers. which might be the reason for lower noise values at the court. The main reason for the increased number of autos and buses during the midday could be the fact that more people preferred to use the public vehicles during the midday period.2: Variation in noise levels at different locations during weekdays For the silence zones. This indicates that in weekend. different trends were observed. 3 shows the variation in noise levels during the midday and the evening. In case of Government hospital. The noise levels recorded during the midday were higher than the evening. more people went to the market than the weekdays. No significant differences in traffic load were observed in between Jain college and District court. the noise during midday was higher than the evening. Fig. the noise levels obtained were higher. where it is slightly high. The noise levels measured at the court was lower than the other locations of silence zone during midday. it is clear from Table 1. noise level during the midday is lower than the noise level during the evening except near the traffic light. that the number of autos and buses were higher while the number of twowheelers was lower during midday than the evening. while during the evenings. 3). number of vehicles passed and distance from the roads for weekends are given in Table 2. Distance also influenced the noise levels. most of the classes are over by 1:00 PM. Traffic noise on weekend Monitored noise levels for different periods. The noise levels obtained during the midday and the evening for commercial areas were higher on weekend than weekday (Fig. Therefore. locations. In case of commercial zone. therefore. more number of people used their private vehicles. The noise levels did not meet the norms for silence zone at any location. In case of Bhagwan Mahavir Jain College. The court was located at higher distance than college. The same 5 . 2 and Fig.

However.8 68.6 77.4 74. Mini Ibrahim Road Residential areas 1sCross. Table 2: Noise levels monitored at different locations for different types of vehicles on weekend 12 noon to 2:00 PM Categories Locations Twowheeler Commercial areas Traffic Light Anupama Plaza 1st Cross.96 4. Geetha Road 8th Cross.78 Fig.9 75.15 22.93 34.trend was also seen at other two locations of the residential areas.0 67.3 75.1 75.96 6. Robertsonpet 6 Cross. in the case of Geetha road.8 73.3 Distance (m) 11.3: Variation in noise levels at different locations during weekend 6 .88 8.67 6. Little variation at Government hospital might be due to variation in number of vehicles. Swarnanagar Silence zones Government Hospital Jain College Court th 6:00 PM to 8:00 PM Truck 3 Leq dB(A) 75.3 75.6 - 3.8 77.4 66.4 75.2 75.1 TwoWheeler 184 144 73 Auto 121 50 34 Car 20 12 2 Bus 3 3 Truck 2 1 1 Leq dB(A) 75. Maintenance of vehicles also influences the noise levels13.0 44 113 121 141 119 - 6 16 35 98 20 - 12 15 20 10 - 1 1 13 - 3 2 - 68. the noise level during the midday was higher.95 Auto 108 41 40 Car 24 17 3 Bus 8 5 1 126 98 80 48 111 78 138 104 125 6 36 37 120 38 67 1 5 7 15 6 8 1 4 1 1 2 4 2 2 67.42 6.

a row of trees ranks among the frequently desired protective measures14. In addition to this. For a model.118.e. The study of residuals (or error) is very important in deciding the adequacy of the statistical model. Therefore. Table 3 shows the model summary of Leq.01. In outdoor noise control. rugs and draperies can reduce noise levels. For the regression model. Grivas and Chaloulakou (2006)18 and Papanastasiou et al. 7 . stepwise multiple regression procedure is commonly used to produce a parsimonious model that maximizes accuracy with an optionally reduced number of predictor variables.3. According to different researchers16-18. it is not good to retain negligible variables. Stepwise regression algorithm was followed for the selection of most influencing variables.Traffic noise control measures Barriers reduce the overall sound pressure level. Interior treatment of rooms with sound absorbing materials like tiles. multiple regression analysis was carried out with the help of SPSS software version 13. using 68 sets of data. It reveals that observed value is many times higher than critical value. If the error shows any kind of pattern. For this. VIF lower than 10 do not imply problems with multicollinearity whereas higher values cause poor prediction equations. Multiple linear regression analysis of data Since many variables affected the noise levels at different locations. therefore.231*Truck (Nos. they should follow the normal distribution with zero mean and constant variance19-20. Traffic noise from highways can be shielded by a barrier. Variance inflation factor (VIF) for the input variables is lower than 10 indicating that there is no multicollinearity. which is as under: Leq = 70. variables with zero coefficients or the coefficients less than their corresponding standard errors19. these variables were used for the development of noise prediction model. (2003)19.) + 0.055*Two-wheeler (Nos.0. According to Montgomery et al. The derived regression coefficients are neither zero nor less than the standard error.516 for Leq whereas critical value of F0. that is.) (1) Analysis of variance (ANOVA) for the model indicated that observed value of F was 33. adjusted R2 increases if the addition of the variable reduces the residual mean square. to be useful as a predictor. then it is considered that the model is not taking care for all the systematic information. (2003)19. The adjusted R2 value is the highest and the residual mean square is the lowest for model 3. residuals should be random i.368. (2007)21 used correlation coefficient (R) between measured and predicted values for the evaluation of model performance.0. For the best performance of the model. regression coefficients for the predictors of model 3 were used to derive the equation for Leq. 64 was 4.304*Distance (m) + 0 . Various models were developed by the software for Leq prediction. observed F ratio must be at least four or five times greater than critical value of F as reported in Montgomery et al. Cunniff (1977)15 stated that barriers like solid thin wall attenuate the noise levels.

Fig. 4: Standardized residual analysis of Leq Fig. which is significant in statistical sense at 1% level of significance. All the regression coefficients of predictors are also statistically significant (Table 3). 4 indicates histograms of the residuals of Leq model.1 is 0. 5: Correlations between predicted and measured values of Leq 8 . Fig.78. Fig. The residuals analysis shows that the residuals are distributed normally with zero mean and constant variance. The correlation coefficient (R) for Eq.1 can be used to estimate Leq for the study area. Considering the adequacy of the model. 5 shows the plots of predicted and measured values of Leq. Eq.

especially for silence zones. therefore. C. distance.304 0.058 70.000 2 (Constant) Distance Two-wheelers 3 (Constant) Distance Two-wheelers Trucks Conclusions During weekdays in commercial areas.368 -.000 0. at different locations of the monitoring areas.611 . On weekend.307 0. the noise levels were higher in the evening than the midday indicating that more numbers of vehicles passed during the evening.532 Significance Variance Inflation Factor (VIF) 1 (Constant) Distance 0.312 0. The developed multiple regression model can be used for the prediction of Leq for the study area.033 0. The noise levels trend for the silence zones were different compared to others indicating that vehicles activities in the evening are lower than midday. 9 .629 0.974 -0.138 -0.586 0.000 0. It is difficult to control the noise generated by different vehicles while moving on the roads. of two-wheelers and trucks were found as most influencing variables on noise levels.000 Standard error 0. noise levels obtained were almost higher than weekdays indicating that more people came for shopping on weekend.348 1. This work can be further studied.837 . etc.009 .055 . acoustic tiles.032 0. None of the three zones met the limits as prescribed in the Indian Regulation. Director.593 4.000 0. Residential locations also showed similar trends as in commercial areas. National Institute of Rock Mechanics (NIRM) for his encouragement and providing facilities throughout the study. In addition to numbers and types of vehicles. for the control of noise levels.046 1. Nawani.348 1. etc.847 0.391 1.197 70.231 .746 .000 1. rugs. P.461 0.000 0. it is suggested to provide barriers.Table 3: Model summary of Leq Model Predictors R square Adjusted R square Residual mean square Regression coefficients Coefficient 75.000 0. considering an attenuation of noise with barriers.000 0.036 0.350 1.573 4. maintenance of vehicle also causes fluctuations in noise levels.035 0.009 . Acknowledgement The principal author is thankful to Dr. Based on the stepwise regression procedure.114 0. no.302 8. trees.

39-46 (2009) 6. S K. Sharma K. 54(2). 64-78 (2006) 14. Comrie A C. John Wiley & Sons. 431-434 (2000) 9. Anon.References 1. Tamil Nadu Poll Res. Status of road traffic noise in Calcutta metropolis. 3-8 (1981) 8. Profile of noise pollution in Lucknow city and its impact on environment. New York (1977) 16. Noise Control Eng J. Scavanger. New York (1998) 12.. Sagar T V & Rao G N. Gupta S. Study of vehicular delay and noise pollution at signalized intersection of Agartala City. J Air Waste Manage Assoc. Assessment of heavy earth moving machinery noise vis-à-vis routine maintenance. Ravichandran C. 6(2). The noise festivals: can we not change?. Agarwal S & Swami B L. Barman S C. Assessment of noise level in Burdwan town. Chakrabarty D. Universal Law Publishing Company Pvt. Cunnif P F. International Journal of Engineering Studies. 48(2). Saha R. 653-663 (1997) 10 . The Environmental (Protection) Act. 101(2). John Wiley & Sons. 10th (Ed.). Kolkata. J Acoust Soc Am. Sadhu S. J Environ Science & Engg. Magrab E B. Sarkar D & Saha P. New York (1975) 15. India. Environmental Noise Control. Miller G T. Kisku G C. Handbook of Noise Assessment. Mishra D & Bhargava. Noise pollution levels in Vishakhapatnam city (India). Comparing neural networks and regression models for Ozone forecasting. Mondal N K & Mukhopadhyay B. West Bengal. Sekaran G E C & Kumar M V.. 609-612 (2006) 10. J Environ Biol. Santra S C. 943-949 (1997) 4. Karmakar N C & Rao Y V. 19(3). Pal M. Kapoor R K & Mahadevan T N. India. Wadsworth. Int J Appl Eng Res. Environmental Science. 47. Shetye R P. Vardhan H. Santra S C & Mukherjee A. 1(1). 27(2). Datta J K. Kidwai M M. 139-142 (2006) 5. Khan A H. May D N. 409-416 (2006) 3. Van Tostrand Reinfold Company. Delhi. New Central Book Agency (P) Ltd. Singh R. Noise pollution assessment in Pudukkottai. London (1978) 11. 243 (2004) 2. Living in the Environment. Environmental Noise Pollution. India (2002) 13. Road traffic noise annoyance in Jaipur city. 161-170 (2011) 7. Ltd. J Environ Biol. 27(3).

Montgomery D C. 40(11). Inc. Papanastasiou D K.17. 182. Goyal P. Development and assessment of neural network and multiple regression models in order to predict PM10 levels in a mediumsized Mediterranean city. 325-334 (2007) 11 . Atmos Environ. 2068-2077 (2006) 21. 53(10). J Air Waste Manage Assoc. Artificial neural network models for prediction of PM10 hourly concentrations. Chan A T & Jaiswal N. 40(7). New York (2003) 20. Grivas G & Chaloulakou A. Grivas G & Spyrellis N. 1183-1190 (2003) 18. in the Greater area of Athens. 1216-1229 (2006) 19. Atmos Environ. Water Air Soil Pollut. Chaloulakou A. Neural network and multiple regression models for PM10 prediction in Athens: A comparative assessment. Introduction to Linear Regression Analysis. Melas D & Kioutsioukis I.. Peck E A & Vining G G. John Wiley & Sons. Greece. Statistical models for the prediction of respirable suspended particulate matter in urban cities.