Comparative evaluation of the ground reflection algorithm in FHWA Traffic Noise Model (TNM 2.5)

This paper investigates the algorithm of ground reflection regarding highway noise propagation, and evaluates the accuracy of predicted absolute noise levels without barriers of FHWA TNM 2.5. To achieve this purpose, new highway noise prediction program, Highway Noise Predictor (HNP 1.0) with the improved ground reflection algorithm, is compared with TNM 2.5 based on the field measurement data of a point source by Parkin and Scholes and the field measurement on roadway 111 in Indiana State. The findings of the study show that TNM 2.5 over-predicts the absolute noise levels by an average error of 2.0 dB(A) compared with field measurements with soft ground, while HNP predicts more accurate absolute noise levels with an average error of 0.5 dB(A). This improved performance is due to the improved reflection algorithm deployed in HNP. These improvements in noise prediction accuracy of absolute noise levels projects major implications on highway noise planning.