The criterion of noise attenuation by hedges

Transcription

1 The criterion of noise attenuation by hedges C.-F. Fang Department of Landscape Design and Management, National Chin-Yi Institute of Technology, Taichung, Taiwan 411, R.O.C Abstract This investigation examined the noise attenuation effect of five hedges, and determined the quantitative criterion of noise attenuation by hedges. The amplifier was placed in front of the hedges, while the noise meter was placed at various heights and distances behind the hedges. Subtracting the sound pressure level at each measurement site on hedges from that on open ground obtains the net reduction effect of hedges, namely, relative attenuation. The parameters included visibility, height and width of hedge, received and noise source height, and the distance between noise source and receiver. The multiple regressive model demonstrated that the order of importance for the factors was tree visibility, width, height, received and noise source height, and distance between noise source and receiver. The five factors were simplified as three dimensionless parameters: V: receiver height / hedges height, H: distance between noise source and received / hedges height, and P: visibility / width. The relative attenuation was plotted on the coordinate axis of V, H and P, and its curve fitting provides a useful criterion map of noise attenuation by hedges. 1 Introduction Plants are the nature material for noise attenuation (Aylor [1]). Plants thus have advantages of economy and esthetics compared to other noise attenuation materials. Moreover, besides reducing noise plants also provide important psychological benefits to humans. Some works have demonstrated that the important factors influences on noise attenuation by plants included tree density, height and length, receiver height and the distance between noise source and receiver (Reethof [8]; Cook and Haverbeke [3]; Fricke [7]; Fang and Ling []). Most studies using sparse hedges that could reduce the noise by approximately 8 db A until its width exceeds 30 m. Such results were unsatisfied and unpractical

2 18 Design and Nature II to environmental designers, because the width of belts needed were too wide. The causes might be related with the using of plants that were too sparse to create the effect. Therefore, the use of more densely hedges may be better to help improving noise attenuation effect in small space. Some authors (Reethof [8]; Fang and Ling []) have noted that dense shrubs and hedges are good plants for noise attenuation because they have dense branches and foliage that can scatter and absorb acoustic waves at ear height. This investigation thus examines noise attenuation by the dense hedges which can provide the good noise attenuation within a finite space. Materials and methods This investigation assessed the noise energy behind five hedges. The parameters included visibility, height and width of hedges, receiver and noise source height (receiver height), and distance between noise source and receiver (distance). The order of various important factors was considered. Subsequently, the key factors were simplified into three dimensionless parameters, namely, horizontal parameter H, vertical parameter V, and penetrative parameter P. Eventually, the noise attenuation criterion map achieved by hedges from the three dimensionless parameters was established. The map will provide a quantity suggestion of noise attenuation by hedges. Five hedges (Table 1) that are common in Taiwan were chosen for conducting the noise measurement. The experiments were conducted on a flat area, and the ambient noise was maintained at 48 ± db A. To eliminate the effect of hedges length, the hedges length must exceed 50 m in this study because Fang and Ling [] have ascertained that the length parameter would insensitive when excess 50 m. Table 1: Characteristics of the hedges. Species V (1) H () W (3) MS (4) RSH (5) Casuarina nana 0.8 7, , 0.0, 0.80, 1.00, Sieber ex Spreng. 1.0, 1.40, 1.80 Casuarina , , 1.50, 3.00 equisetifolia Duranta repens , , 0.75, 1.50 Gloden leaves Ficus microcarpa L.f , , 0.9, 1.9 Golden Leaves Hibiscus rosasinensis , , 0.75, 1.50 (1) V: Visibility; () H: Height; (3) W: Width; (4) MS: Measuring site; (5) RSH: Receiver and noise source height.

3 Design and Nature II 187 The method of producing noise was the same as that used by Fang and Ling []. First, the traffic noise on the artery was recorded, and was termed the unedited source. A stable sound pressure level (SPL) of approximately 10 seconds was selected from the unedited source and recorded repeatedly on a tap for around 30 min, and was designated as the edited source. The edited source was served as the noise source in this investigation because it is portable, has low SPL variation, and is repeatable. Accordingly, stable SPL can be obtained quickly by using edited noise. The range of the SPL of the edited noise was 73~77 db A at 5 m from the source. Fang and Ling [] provided further details of the noise source. Lines 1 and were located perpendicular to the length of the hedges (Fig. 1). The intervals of two lines was 5 m apart. Furthermore, the noise source was located m in front of the hedges on the transecting lines. The first measuring site was located 7 m behind the hedges on the transecting line from the source, the second was located 1 m behind the hedge, third was located at 17 m and so on. The measuring sites totalled 4~1 according to site condition. The SPL of the measuring site on the two transecting lines must be averaged to obtain the mean SPL of that measuring site. An equivalent experiment was also conducted on the open ground near the hedge. Subtracting the mean SPL of the measuring site on the open ground from that in the hedges yielded the relative attenuation, which represents the net noise attenuation at the measuring site due to the hedge. line 1 line 5 m measuring site 5 m 5 m hedges width 5 m m noise source Figure 1: The experiment design. The various climate conditions will influent the noise attenuation in the air (Beranke and Vèr []). Therefore Embleton [4], Cook and Haverbake [3] and Fang and Ling [] conducted the noise measurements in the similar climate, so that the molecular absorption would be slight and the atmosphere effect would be neglected. In order to minimize the atmosphere effect, this study conducted the noise measurement during sunny day and at wind velocities of less than m sec -1.

4 188 Design and Nature II The noise source (AIWA amplifier, 50W) was positioned at each transecting line m in front of the hedges. The height of the noise source was 0.3~3 m depending on the hedges condition (Table 1). The noise source was 74.8 db A at 5 m from the source. The noise meter ( NIST MODEL1900) must faced the source and measured the noise source at each measuring site. The height of the receiver (noise meter) was the same as that of the source height (0.3~3 m). However, before the formula acoustic measurement, the noise meter should be calibrated using the NIST QS1900 acoustic calibrator. To ensure a stable acoustic measurement, the meter and amplifier should be operated for over one minute before regular measurement. Following apparatus calibration, the noise was measured by a noise meter 10 times, with each measurement lasting 30 sec and being conducted at each measuring site (Fang and Ling []). The noise meter was set up as follow: A- weight signal level (db A) and fast characteristic time response. Some investigations have used the visibility to calculate the denseness of hedge, that is, the distance that an object is obscured by the plant (Eyring [5]; Embleton [4]; Fang and Ling []). The visibility in this investigation was revised from Fang and Ling []. Researcher A stood near the edge of the hedges and the researcher B stretched a white gloved hand into the hedges in increments of 10 cm until the researcher A became unable to see the white glove. Then the distance between researcher A and white gloved hand was termed visibility. The average distance measured twice on each belt, was deemed the unit of visibility (unit: m). First, the attenuation effect of the relative attenuation behind the hedges was displayed. Next, the multiple regressive model was established for determining the relative importance of the hedges noise attenuation factors. In this model, the dependent factor was relative attenuation and the independent factors were visibility, height, width, receiver height and distance. Subsequently, the five important parameters sought from the regressive model should be dimensionless by logical deduction. The unit of the combinative parameters was nullified; therefore they become the dimensionless parameters. Three dimensionless parameters can be obtained: V (receiver height / height), H (distance / height) and P (visibility / width). The trend of relative attenuation is discussed for various H under various V( ) values. Therefore the relation between the factor of receiver height, tree height, distance and relative attenuation can be clarified. Afterward, the H, P and V were established using the coordinate axis X, Y and Z. The relative attenuation was plotted on the coordinate axis and the noise attenuation criterion map of the hedges was established based on the curve fitting. 3 Results 3.1 Multiple regressive model The Standardized coefficient (Beta) in the multiple regressive model (Table ) indicated that relative importance of the various factors followed the sequence

5 Design and Nature II 189 visibility, width, tree height, receiver height and distance. The visibility, receiver height and distance are negatively correlated with relative attenuation. Meanwhile, height and width are positively correlated with relative attenuation. Table : Multiple regressive model of hedges and their noise reduction effects. Variable Unstandardized Standardized t-value coefficient (B) coefficient (Beta) Constant Visibility *** Width *** Height *** Receiver *** height Distance *** R =0.7 F=77.4 *** P Traditional distribution The noise attenuation was good when the receiver was low and was small when receiver was high (Fig. ). The trend of relative attenuation behind each hedges exhibited a turning point. In front of this turning point the relative attenuation was better and more stable than outside the turning point. Moreover, the relative attenuation worsened and decreased with distance behind the turning point. Relative attenuation (db A) Turning point Distance Reciever height 0.38 m 0.75 m 1.5 m Figure : The noise attenuation of Hibiscus rosa-sinensis. 3.3 The relation between V and H Figure 3 shows illustrate the relationship between H and relative attenuation for various V. The results indicate that the relative attenuation was large when V = 0.15 and was then concentrated between 7 and 3 db A. It was moderate when V

6 190 Design and Nature II = 0.3 and was then concentrated between 5 and 3 db A. The relative attenuation was small when V = 0. and the relative attenuation was less than 4 db A. Turning point existed approximately at H = 8 when V = 0.15 and 0.3. The relative attenuation were stable when H 8, and decreased rapidly when H > 8. However, the turning point was not obvious when V = 0. and the relative attenuation decreased with H. 7 Casuarina nana Casuarina equisetifolia Relative Attenuation (db A Ficus microcarpa Hibiscus rosa-sinensis Duranta repens Legend H H Figure 3: The relation between H and V of each hedge. 3.4 The criterion of the noise attenuation Figure 4 illustrates the curve fitting of relative attenuation at various H, V and P. The map represented that the relative attenuation was significant when V, H and P was small. Restated, low receiver height was associated with high hedges, close distance, large width, low visibility and effective noise attenuation 4 Discussion 4.1 Noise attenuation effect When the noise meet up barrier the noise shadow zone was appeared behind the barrier. The noise attenuation was improved within shadow zone and deteriorated

7 Design and Nature II 191 without shadow zone. Moreover, the range of shadow zone increased with barrier height and length (Beranek and Vèr []). The turning point (Figs. ) represented the edge of the noise shadow zone. Therefore, the noise attenuation was greater and more stable in front of the turning point and less stable, decaying with distance, behind it. V dba dba 3 dba H dba 5 dba P 0.7 Figure 4: The criterion map of noise attenuation from hedges. 4. The reduction parameters in multiple regressive model The multiple regressive model (Table ) implied that the main determinants of noise attenuation by hedges include visibility, width and height, receiver height and distance. The visibility provided an accurate index of the plant thickness. Lower visibility implied the existence of more foliage and branches to scatter noise (Aylor [1]; Cook and Haverbeke [3]; Fang and Ling []). Consequently, the relative attenuation displayed a negative relationship with visibility. Similar results were also found in Eyring [5] and Fang and Ling []. The production of scattering, diffraction and absorption effect increased with hedges width and height (Cook and Haverbeke [3]; Fang and Ling []). Consequently, relative attenuation thus was positive with width and tree height. Theoretically, the SPL of noise will be attenuated when the propagating noise wave meet a barrier. Because the barrier can cause diffraction and then lengthen the pathway of the noise. However, the effect of attenuation can be various and dependent on the height of hedges tested, the height of noise sources and receiver set in the field study site. If the ratio between hedges height and sources (receiver) height was higher, the attenuation effects might be larger. This is

8 19 Design and Nature II attributed to the effect of diffraction and then is affected by the increase of the path length of propagating wave. In other word, the hedges can do better on the noises near the ground than that near the treetop. Therefore, the relative attenuation has the negative relationship with receiver height. The Beta value in the multiple regressive model (Table ) indicates that the distance parameter was less important than the other parameters, because of the confounding effect from the shadow zone. The noise attenuation was stable within the shadow zone and declined with distance away from the shadow zone; therefore, the distance parameter was slightly negatively correlated with relative attenuation. 4.3 The V and H The noise attenuation was good when H 8 if V = 0.15 or 0.3 (Fig. 3). The above indicates that when the ratio of the receiver height to tree height was from 1:. to 1: 3.3, that of distance from receiver to noise source to height was less then 8: 1, so that the hedges can provide the effective reduction. Restated, if the height of the hedges exceeds that of the source (receiver) by between five and three times, then the receiver must be placed at a distance from noise source no greater than eight times the height of the hedge. Accordingly, this setup optimizes the noise attenuation effect of the hedges. The reduction effect was low and decreased as H when V = 0., implying that the receiver should be placed near the source only if the height of hedges is at least double that of the receiver twice. 4.4 The criterion of noise attenuation by hedges The criterion map (Fig. 4) simplified five parameters into three dimensionless parameters and presented the reduction effect of each parameter. The map presented that the reduction effect was great when V and P was small, and H was 8. These results imply that the hedges were very effective when the receiver height and hedges visibility are low, the hedges were high and wide, and the distance was less than eight times the height. The term V represents the range of the shadow zone; H specifies whether the receiver is located in the shadow zone and P is an accurate index of the scattering of the noise from the hedge. The map can predict the noise attenuation effect achievable by hedges and provide useful information for environmental planners. For example, possible methods of reduce 5 db A noise using hedges include the condition P = 0.4, H = 9 and V =0.1, or P = 0.5, H =1 and V= 0.3. Under the former conditions, when the receiver was 0. m height and the space of hedges width was m, the hedges must be planted with the structure by 0.8 m visibility, m height, and receiver was placed at 54 m from the source. Under the latter condition, the hedges must be planted with the structure by 1 m visibility, m height, and receiver was placed at 4 m from the source. In conclusion, this work provided a practical and flexible method for reducing noise using a hedge. In this work, V and H were used to evaluate the diffraction of noise by hedges and P was determined to elucidate the porosity of hedge. Usually, the noise

9 Design and Nature II 193 attenuation by the barrier is evaluated in terms of Fresnel number (Beranek and Vèr []), which is also a dimensionless parameter. The reasons of why this study had not used Fresnel number was that it predicted the diffraction effect of a wall with solid content only and was inapplicable to predict the effect of the hedges which have many gaps. Consequently, by combining with the above three dimensionless parameters, the figure 4 can more effectively predict the reduction of noise by hedges References [1] Aylor, D.E., Noise attenuation by vegetation and ground. J. Acoust. Soc. Am., 51: , 197. [] Beranke, L.L., Vèr, I.L., Noise and vibration control engineering. Wiley- Inter science Publication, New York, 199. [3] Cook, D.I., Haverbeke, D.F.V., Trees and shrubs for noise abatement. Univ. of. Neb Col of Agr Exp Sta Bull., RB4, [4] Embleton, T.F.W., Sound propagation in homogeneous deciduous and evergreen woods. J. Acoust. Soc. Am., 35: , 193. [5] Eyring, C.F.,. Jungle acoustics. J. Acoust. Soc. Am., 18(): 57-70, 194. [] Fang, C.F., Ling D.L. Investigation of the noise attenuation provided by hedges. Landscape and Urban Planning, 3(4): , 003. [7] Fricke. F., Sound attenuation in forests. J. Sound and Vibration, 9(1), , [8] Reethof, G., Effect of plantings on radiation of highway noise. Air Pollut. Control Aso., 3(3): , 1973.