Effectiveness of noise barriers at residential areas in Klang Valley, Malaysia

• Main Author: Halim, Herni
• Format: Thesis
• Language: English
• Published: 2016
• Online Access: http://psasir.upm.edu.my/id/eprint/66339/1/FPAS%202016%202IR.pdf; http://psasir.upm.edu.my/id/eprint/66339/

The use of environmental noise barrier has increased as concerns of side effects of noise pollution in the environment grow. However, a lot of complaints from the residents occur after the installation of the noise barrier regarding its effectiveness. Thus, this study is carried out to determine the effectiveness of existing noise barriers such as vegetation, concrete hollow block, and panel concrete. This study is designed to evaluate the influence of source of noise (traffic composition) as well as meteorological condition towards level of noise at study sites with and without the presence of noise barrier. The results of the study reveals that LAeq in all study sites throughout all measurement periods recorded 61.9 dB(A) to 74.8 dB(A) during daytime (morning, afternoon, and evening) and 57.6 dB(A) to 73.2 dB(A) during night time. These results exceed the noise limit in Malaysia guidelines (daytime = 60 dB(A) and night time = 50 dB(A)). Results of PCA show the percentages of the total variance up to 95% by variables of traffic composition such as car, motorcycles, and light lorries. Apart from that, panel concrete noise barrier meet the minimum value of effective noise barrier insertion loss. The concrete hollow block partially meet the minimum value of effective noise barrier with computed insertion loss, while vegetation does not provide enough insertion loss of effective noise barrier. Therefore, the finding indicates that panel concrete provides consistent insertion loss followed by concrete hollow block and vegetation. The MLR model based on the traffic characteristics group exhibited optimal performance in term of coefficient of determination for with and without noise barrier location with values of 0.858 and 0.839, respectively.