Development of a Relation Between Traffic Variables and Environmental Noise Descriptors for Four-Lane National Highways Ashish Singh 1 Centre for Transportation Systems Indian Institute of Technology Roorkee E.Rajasekar 2 Joint faculty Centre for Transportation Systems Indian Institute of Technology Roorkee Manoranjan Parida 3 Joint faculty Centre for Transportation Systems Indian Institute of Technology Roorkee

ABSTRACT In India, transportation sector is quickly expanding. The heterogeneity of traffic plays a vital role in shaping noise ambience in vicinity of highways. This study aims to evaluate the factors that contribute to traffic noise along four-lane divided National Highways. The data was collected at 4 locations with free-flowing traffic on the National Highway 334 in India. The parameters included traffic noise levels, classified traffic volumes, vehicular speed and geometry of highway cross section. Traffic volume and composition was captured using videography, while traffic speed was measured using radar gun. Single octave spectral characteristics of the traffic noise was measured using multi-channel noise spectrum analyzer. The data was sampled at 15 minutes’ interval and measurements carried out for 3-hour duration during peak and off-peak periods weekdays and weekends. Noise descriptors such as equivalent noise level (L eq ), 10 percentile sound level (L 10 ), 50 percentile sound level (L 50 ), 90 percentile sound level (L 90 ) were estimated from the measured data. Effect of traffic characteristics on the traffic noise levels was established through regression analysis. The study concludes that traffic volume, percentage of heavy vehicles and traffic speed are the major factors influencing the noise levels.

1. INTRODUCTION

Excessive noise is one of the primary sources of disturbance in urban areas. Public transport in India is mostly used and it is the second largest and busiest transportation in the world it transports 8.225 billion passengers annually.

1 asingh@ts.iitr.ac.in 2 raj@ar.iitr.ac.in 3 m.parida@ce.iitr.ac.in

The Rapid growth in population and urbanization since 1980 and also rapid growth of the country’s economy has increased the higher vehicle ownership among the household with the growth of population and urbanization, transportation demand has increased almost eight times in India[1].Human impact due to terrestrial Noise is bound to increase in the future as a result of urbanization. Devos & Van beek reports that all over the world, an estimated 2 billion citizens are subjected to road traffic Lden of Over 55 dB[2]. Because of its negative effects on human life and the environment, road traffic noise has arisen as a major source of worry in India and throughout the world. According to a research done in Curitiba City,73% of respondents identified traffic noise as the primary source of noise disturbance on their streets.[3]. In most Indian cities, noise levels have exceeded the stipulated regulatory limit[4].In many studies found that residential areas are more exposed to noise pollution in other regions. The level of highway traffic noise is primarily determined by traffic volume, composition of traffic mix, traffic speed, and road geometry characteristics[5].Noise descriptors Such as L eq ,L 10 ,L 50 ,L 90 used to determine the noise level at different locations[6].Road traffic noise is not constant throughout the whole road network, and it varies with the space[7]; because of the rapid development, it is essential to study highway noise concerning various causative factors[6].Most places near the roadways have greater noise levels than the permissible limit (Central Pollution Control Board (CPCB), 2001). This fluctuation is assessed as a result of noise variations impacting parameters. Many factors influence the noise level at observers' positions, including traffic flow rates, composition, proportion of heavy vehicles, Speed, metrological parameters, geometrical layout, and pavement qualities[8]. While the previously listed criteria such as traffic volume, Speed, and heavy vehicle proportion significantly influence traffic noise [9]. The major source of highway road traffic noise is Noise caused by a vehicle's engine, exhaust, tire-pavement contact between moving cars and pavement, road quality, traffic speed, and traffic management[9]. The Speed of a vehicle has a considerable impact on urban traffic noise and also the speed effect is more substantial than any other contributing component[10]. The proportion of heavy vehicles is an essential component that influences traffic noise on the roads. Traffic volume is also the most influencing factor to increase the noise level on the highways there is a relationship between traffic flow and sound level, which is stated as Leq=C logq. Where C is a constant and q is the hourly traffic flow[11]. Therefore, this Study aims to develop a relation between traffic variables and noise descriptors for four lane national highways at Indian Conditions and elaborate the characteristics of traffic at selected highways. 2. METHODOLOGY: This methodology helps to understand the which traffic variable is more influencing factor to contribute the highway noise figure 1 shows the methodology adopted.

Figure 1 Flowchart of methodology

2.1 Site Selection The sites are located at National Highway 334 from Haridwar to Delhi formerly it was NH-58. Sites were selected near the settlements in the vicinity of highway. Road mainly constructed the bituminous pavement moreover highway carried out the mix traffic conditions. The selected sites were fully accomplished with the study area selection criteria. The flow traffic was free flow and the Sites were located away from the other sources of Noise like airport, railway stations. The selected site was dry and free from noise barriers. The settlements were on both sides of the highway. Site was chosen in that manner a good data sets of all the vehicle types could collect. Map of the selected study area shown in figure 2.

Figure 2 Map Of Study Area (Source:Google Map) 2.2 Data Collection Data was collected at four sections of the National Highway 334 pavement surface was bituminous pavement. At each section, the noise descriptors was measured L eq ,L 10 ,L 50 ,L 90 the noise level meter was set at 7.5m from centreline of the road and the microphone was set towards traffic at a height of 1.2m from the ground level. Parameters included in the data collection was types of vehicle, Pavement type, metrological parameter, highway geometrical dimensions were also measured in terms of number of lanes, lane width, median width, shoulder width and right of way width. The Speed of the vehicle was measured simultaneously by using a radar gun. Traffic volume was recorded with the video-camera along with the vehicle classifications. The recorded data was extracted manually with the classified vehicle count, Car, Two-wheeler, Bus, Heavy-Truck, Light- Truck, LCV, Three-Wheeler, Tractor. Metrological parameters such as temperature, wind speed, humidity were noted down at each section. Data was collected Morning, After-noon off peak hours of the day. Total 4 locations were considered for noise measurement after identification of flow of traffic in the study area. Measurements were performed in August and October 2021. Speed range of each vehicle type was considered 45-140 km/h. Noise measurement was done with Delta ohm class 1 integrating sound level meter. 3. RESULTS This Study is utilized to develop a relation between various noise descriptors and traffic variables. The data was statistically analysed collected from multiple section of the highway to establish a correlation matrix between noise descriptors and traffic variables. The evaluated variables were total traffic volume, percentage of heavy vehicles, and Speed of the vehicles. Correlation coefficients for measured noise levels and evaluated variables are shown in table 2. In the subsections influencing noise parameters and traffic characteristics are discussed.

3.1 Traffic Characteristics of the highway The total number of vehicles flowing every hour is referred to as traffic volume. As traffic volume grows, so the noise level increases. The vehicles are classified into eight groups as shown in table 1.

Table 1 Classification of Vehicle

S.No. Types of vehicle Description

1 Cars All four-wheelers

2 Motorcycles All two-wheelers

3 Heavy Truck More than 10 wheels

4 Light Truck All 6 Wheelers

5 Bus All buses

6 LCV Vehicle weight 3.5t

7 Tempos All 3-Wheelers

8 Tractors All Tractors

Several such samples have to be collected in morning afternoon peak hours on the selected highway cars and two-wheelers are the dominant vehicle as shown in figure 3.

0 200 400 600 800 1000 1200 1400 1600 1800 2000CARTwo- WheelerBusHeavy truckLight truckThree WheelerLCVTractor

Traffic Volume

Morning Afternoon Evening

Figure3: Vehicle classification peak and non-peak hours on the highway 3.2 Impact of Traffic Volume In this Study, traffic volume is the major factor influencing traffic noise on the highways. As shown in the scatter figure 4, traffic volume and equivalent noise level Leq is highly correlated with traffic volume. In the correlation matrix table 2 it was found that the correlation coefficient is 0.610 it shows that as the traffic volume increases noise level increases. The highest noise level was observed near about 82 dB(A) at a traffic volume of 546 in 15 minutes. In the matrix table we can also see that the correlation between traffic volume and other noise descriptors are not as much strong correlation between L10 & L50 with the traffic volume is 0.186 and 0.131.

82.0

80.0

78.0

76.0

Leq

74.0

72.0

70.0

68.0

0 200 400 600

Total Volume

Figure 4: Influence of Traffic Volume on the Equivalent Noise Level(Leq)

Table 2 : Correlation matrix table between traffic variables and Noise descriptors

Correlations

Leq L10 L50 L90 Q V %Hv

Pearson Correlation 1 .454 .392 .269 .610 .582 .592

Leq

Sig. (2-tailed) 0.000 0.000 0.010 0.000 0.000 0.000

N 92 92 92 92 92 92 92

Pearson Correlation .454 1 .934 .839 0.188 .237 .225

L10

Sig. (2-tailed) 0.000 0.000 0.000 0.072 0.023 0.031

N 92 92 92 92 92 92 92

Pearson Correlation .392 .934 1 .957 0.130 0.200 0.074

L50

Sig. (2-tailed) 0.000 0.000 0.000 0.218 0.056 0.485

N 92 92 92 92 92 92 92

Pearson Correlation .269 .839 .957 1 0.073 0.198 -0.028

L90

Sig. (2-tailed) 0.010 0.000 0.000 0.489 0.059 0.788

N 92 92 92 92 92 92 92

Pearson Correlation .610 0.188 0.130 0.073 1 .564 .413

Q

Sig. (2-tailed) 0.000 0.072 0.218 0.489 0.000 0.000

N 92 92 92 92 92 92 92

Pearson Correlation .582 .237 0.200 0.198 .564 1 .426

V

Sig. (2-tailed) 0.000 0.023 0.056 0.059 0.000 0.000

N 92 92 92 92 92 92 92

Pearson Correlation .592 .225 0.074 -0.028 .413 .426 1

%Hv

Sig. (2-tailed) 0.000 0.031 0.485 0.788 0.000 0.000 N 92 92 92 92 92 92 92

3.3 Impact of Speed Found results of the correlation between Speed of the vehicles and equivalent noise levels is significant the correlation coefficient is 0.582 as shown in table 2 and in the scatter plot diagram of speed and equivalent noise level (Leq) we can see that the highest value near about 80 dB(A) at Speed of 80km/hr. The minimum Speed observed of the vehicles on the highways is 45 km/hr and at that Speed the (Leq) ranged 67-72 db(A). In the figure 5 it shows the variation of (leq) when the speed increases. It was observed that the correlation between Speed of vehicles with other noise descriptors is not so strong. The correlation of V speed with the L10 and L50 is 0.237 and 0.201 but also less correlate with the L90 (0.197).

82.0

80.0

78.0

76.0

Leq

74.0

72.0

70.0

68.0

0 20 40 60 80

Speed

Figure 5: Influence of Traffic Speed on the Equivalent Noise Level(Leq) 3.4 Impact of Percentage of Heavy Vehicles: In this Study, the equivalent noise level is highly affected by the percentage of heavy vehicles on the highways. This is due to the heavy vehicles having a large cubic capacity and exhaustive engine power . In the scatter figure 6 between equivalent noise level and percentage of heavy vehicles, the correlation is significant, with a correlation coefficient value of 0.59 shown in the table. The correlation between heavy vehicles and other noise descriptors is not up to the mark L10 ,l50 and L90 is the less affective having correlation coefficient value is 0.225 and 0.074. In many studies, it is found that the measured Leq has a significant effect as the percentage of heavy vehicles increases on the roads[12].

68.0 70.0 72.0 74.0 76.0 78.0 80.0 82.0

Leq

0.0 5.0 10.0 15.0

%Hv

Figure 6: Influence of Traffic Speed on the Equivalent Noise Level(Leq) According to the statistical analysis of the data, the results shown above show that the equivalent noise level influenced by road traffic noise is highly affected. Consequently, heavy vehicles and the Speed of the vehicles are also the main influencing factor introducing the highway traffic noise.

4. CONCLUSION This Study Concludes that to develop the relation between different noise descriptors with the different traffic variables, in this research the data of Noise was calculated at four different locations of peak and non-peak hours with the traffic diversity of each location. The vehicles was classified into eight categories. According to the linear regression analysis the total traffic volume is the main influencing factor of the road traffic noise other traffic variables are also playing vital role to introduce the Noise. To develop a correlation between noise descriptors and traffic variables regression analysis was conducted. ACKNOWLEDGEMENTS The author is very thankful to the Indian Institute of Technology Roorkee for giving valuable support to provide data collection instruments. The author would also like to thank his supervisors, Prof.E.Rajasekar & Prof. Manoranjan Parida, for their guidance and support. REFERENCES:

[1] S. Agarwal, B. L. Swami, and A. B. Gupta, “Development of a noise prediction model under interrupted traffic flow conditions: A case study for Jaipur city,” Noise Heal. , vol. 11, no. 45, pp. 189–193, 2009, doi: 10.4103/1463-1741.56211.

[2] P. De Vos and A. Van Beek, “Environmental Noise,” Encycl. Environ. Heal. , pp. 476–488, 2011, doi: 10.1016/B978-0-444-52272-6.00252-X.

[3] P. H. T. Zannin, A. Calixto, F. B. Diniz, and J. A. C. Ferreira, “A survey of urban noise annoyance in a large Brazilian city: The importance of a subjective analysis in conjunction with an objective analysis,” Environ. Impact Assess. Rev. , vol. 23, no. 2, pp. 245–255, 2003, doi: 10.1016/S0195-9255(02)00092-6.

[4] K. Kumar, V. K. Katiyar, M. Parida, and K. Rawat, “Mathematical Modeling of Road Traffic Noise Prediction,” Int. J. Appl. Math Mech , vol. 7, no. 4, pp. 21–28, 2011.

[5] S. A. Abo-qudais, “traffic noise levels,” no. March 2005, 2019.

[6] P. Kumar, S. P. Nigam, and N. Kumar, “Vehicular traffic noise modeling using artificial neural network approach,” Transp. Res. Part C Emerg. Technol. , vol. 40, pp. 111–122, 2014, doi: 10.1016/j.trc.2014.01.006.

[7] Z. Lan, C. He, and M. Cai, “Urban road traffic noise spatiotemporal distribution mapping using multisource data,” Transp. Res. Part D Transp. Environ. , vol. 82, no. March, p. 102323, 2020, doi: 10.1016/j.trd.2020.102323.

[8] S. Afandizadeh and H. Gharehdaghli, “A new steady-state traffic noise model for estimating L10(h) on free flow roads using Reference Energy Mean Emission Levels,” Build. Environ. , vol. 196, no. November 2020, p. 107685, 2021, doi: 10.1016/j.buildenv.2021.107685.

[9] H. Halim and R. Abdullah, “Equivalent noise level response to number of vehicles: A comparison between a high traffic flow and low traffic flow highway in Klang Valley, Malaysia,” Front. Environ. Sci. , vol. 2, no. JUN, pp. 1–4, 2014, doi: 10.3389/fenvs.2014.00013.

[10] D. Skarlatos and T. Zakinthinos, “A simplified model for urban traffic noise prediction,”

Noise Control Eng. J. , vol. 55, no. 2, pp. 266–274, 2007, doi: 10.3397/1.2711615.

[11] B. Li, S. Tao, and R. W. Dawson, “Evaluation and analysis of traffic noise from the main urban roads in Beijing,” Appl. Acoust. , vol. 63, no. 10, pp. 1137–1142, 2002, doi: 10.1016/S0003-682X(02)00024-5.

[12] P. R. Rao and M. G. S. Rao, “Prediction of LA10T traffic noise levels in the city of Visakhapatnam, India,” Appl. Acoust. , vol. 34, no. 2, pp. 101–110, 1991, doi: 10.1016/0003- 682X(91)90025-A.