IDENTIFICATION OF POTENTIAL PLANT SPECIES FOR TRAFFIC  NOISE ATTENUATION

Rapid urbanization in Sri Lanka has led to increased vehicle density and traffic noise, posing health risks to humans and animals. While many countries use physical barriers,  vegetation-based noise barriers are gaining attention as sustainable and cost-effective alternatives. However, research on the acoustic properties of Sri Lankan plant species is still  limited. The main objective of this study was to identify locally available plant species for traffic noise attenuation by following in situ measurements, laboratory-based experiments,  and theoretical modelling. Matara City, where some road developments were recently completed, was selected as the primary study location. Traffic noise levels, vehicle speeds,  vehicle densities, and other related parameters were collected and analyzed. Noise propagation maps were generated using GIS tools, and practical recommendations for traffic  noise mitigation beyond conventional methods were proposed. The construction of a small-scale reverberation room, the first of this kind, developed in Sri Lanka, to assess the sound absorption characteristics of various plant species under controlled conditions, is an innovation of this study Two adjoining rooms at the Department of Physics, University of Ruhuna, were acoustically modified to serve as a reverberation room and a receiving room, respectively, satisfying international standards.  More than 50 Plant species, including grasses, shrubs, and household varieties, were tested using the reverberation room, alongside in-situ measurements from naturally growing vegetation barriers in residential landscapes. Acoustic parameters were analyzed using SPSS, revealing that the reverberation time (RT₃₀) of the species studied followed a  logarithmic relationship with frequency, while sound absorption coefficients (SAC) exhibited a quadratic dependency. These relationships were expressed as: 𝑅𝑇₃₀ = 𝑎 𝑙𝑛(𝑓) + 𝑏  and 𝑆𝐴𝐶 = 𝑐 𝑓² +𝑑𝑓+𝑒, where a, b, c, d and e are constants, and f is the frequency.   Among the grass species, Stenotaphrum dimidiatum (SAC = 0.6) and Eleusine indica (L.) Gaertn (SAC = 0.4) exhibited the highest sound absorption coefficients. Among the  closely growing and shady plant species, Terminalia catappa (SAC = 0.12) and Anacardium occidentale (SAC = 0.11) demonstrated notable attenuation in the higher frequency range (2000–3150 Hz), while Ficus drupacea var. pubescens (SAC = 0.10) was more effective in  the mid-frequency range (1250–2000 Hz). Furthermore, household plant species such as  Ficus benjamina ‘Yellow’ (SAC = 0.14), Dillenia triquetra (SAC = 0.13), and Ochlandra stridula (SAC = 0.08) maintained consistently high absorption values across a broader  frequency spectrum.  In-situ measurements were conducted to evaluate the noise reduction capacity of  man-made green hedges. A uniform vegetation barrier (Barrier 01), composed of Ficus benjamina (Green) plants, reduced noise by approximately 1.7 dB within 4 m from the source to the receiver. Analysis revealed that the compound vegetated garden (Barrier 02) reduced noise by 36 dB across a 12 m distance. The findings suggest that such vegetated areas are highly effective for sites at least 10 m from roadways, with strong potential to  achieve more comfortable sound levels for urban environments. Additionally, a theoretical formula was derived to estimate the sound reduction index  (Rₙ) of a multi-layered vegetation barrier as: 𝑅_𝑛 = −10 𝑙𝑛₁₀ ∏ _i (1 − 𝛼_𝑖),  where αⱼ represents the absorption coefficient of each plant layer. This formula can be used to model landscaping for the best noise reduction. This study contributes valuable insights to bioacoustics and establishes a scientific  foundation for incorporating vegetation into urban noise mitigation strategies. The findings and experimental methods developed through this research can support the future design of eco -friendly sound barriers in Sri Lanka and other tropical urban environments. 

Keywords: Traffic noise, Sound barriers, Sound absorption coefficient, Reverberation time, 
Reverberation Room