Investigation of the noise reduction provided by tree belts
Introduction
Vegetation has been proposed as a natural material to reduce noise energy outdoors (Aylor, 1972). There are a few qualitative recommendations regarding principles for design of plantings to reduce noise (Reethof, 1973, Cook and Haverbeke, 1974, Herrington, 1976, Reethof and Heisler, 1976). However, few quantitative data have been reported on the significance of height, density, width and length of tree belts for noise reduction. A model of noise reduction by tree belts was therefore examined in this study. Most earlier studies on noise reduction deal with deciduous and coniferous trees (Embleton, 1963, Aylor, 1972, Cook and Haverbeke, 1974, Kragh, 1979, Kragh, 1981). Evergreen trees of the subtropics might have somewhat different effects. Furthermore, many earlier studies have only examined tree belts of particular species (Embleton, 1963, Kragh, 1979, Kragh, 1981), but failed to discuss noise reduction effect in relation to the form of the tree, or the density, height, length and width of the tree belt. Therefore, the noise reduction effect of various broadleaf tree belts was examined in this study. This work was performed in plantations where many complicated variances could be controlled. The tree belt parameters examined included visibility, height, width and length. Subsequently, a noise reduction model was developed, and the results were summarized in a map showing the relationship between noise reduction and both visibility and width.
Section snippets
Materials
Thirty-five large plantations of single species with a uniform density and height, and an ambient noise maintained at 48±2 dB A were selected for acoustic measurement. There were 19 species of evergreen trees and shrubs on the plantations. The species and the characteristics of the plantations are given in Table 1. The relative attenuation data obtained from 35 plantations were used for regression analysis. Eighteen plantations selected on the basis of height and length (refer to Section 2.7 for
Excess attenuation of all tree belts
The excess attenuation of the 35 tree belts is illustrated in Fig. 5. Three groups are apparent.
- •
Group 1: Effective reduction region. Excess attenuation exceeded 6 dB A. All tree belts were comprised of large shrubs with a visibility of less than 5 m.
- •
Group 2: Sub-reduction region. Excess attenuation was 3–5.9 dB A. This group included trees and shrubs whose visibility ranged between 6 and 19 m.
- •
Group 3: Invalid reduction region. Excess attenuation was less than 2.9 dB A. This group included sparsely
Discussion
Sound reduction occurs via normal attenuation and excess attenuation (Herrington, 1976, Harris, 1979). Normal attenuation is due to spherical divergence (Wilson, 1994) and friction between atmospheric molecules when sound progresses (Herrington, 1976). This has been termed the distance effect; noise attenuation increases with distance. Furthermore, reflection, refraction, scattering and absorption effects due to any obstruction between a noise source and a receiver results in excess attenuation
References (13)
Pilot study on railway noise attenuation by belts of trees
J. Sound Vibration
(1979)Road traffic noise attenuation by belts of trees
J. Sound Vibration
(1981)Noise reduction by vegetation and ground
J. Acoust. Soc. Am.
(1972)- Beranke, L.L., Vèr, I.L., 1992. Noise and Vibration Control Engineering. Wiley/Interscience, New...
- Cook, D.I., Haverbeke, D.F.V., 1974. Trees and shrubs for noise abatement. University of Nebraska College of...
Sound propagation in homogeneous deciduous and evergreen woods
J. Acoust. Soc. Am.
(1963)