The Uncertainty Associated with Short-Term Noise Measurements of Passenger and Freight Trains

Abstract

When assessing the potential impacts from passenger or freight train operations, it is necessary to undertake noise measurements at representative locations. It is important to recognise that the level of uncertainty associated with these measurements is significant and should be included in the assessment. Rail noise emissions vary each day due to many parameters such as speed, rolling stock type, wagon type, track and wheel condition, minor timetable changes and weather conditions. Measurement data also varies as a result of the type of sound level meter used, measurement locations used and other operator factors. Limited guidance is available which identifies the number of train passby events of each type that are required to obtain a reliable estimate of long-term overall noise levels. Guidance in relation to the uncertainty associated with the measurement of L_Aeq(period) and L_Amax noise levels from rail traffic is provided in Draft International Standard ISO/DIS 1996-2. However, in the literature this guidance has not been reviewed in the context of Australian conditions and assessment criteria. This paper expands on the information provided in the draft standard via the analysis of noise measurements at a variety of independent sites across the rail network in Sydney Australia. Nine sites are based on unattended measurements undertaken over a continuous period of one week. Two sites are based on unattended measurements undertaken over a continuous period of up to six months. At each site, the standard uncertainty associated with the short-term measurement of five, ten, twenty and sixty train passbys of each type are compared with the longer-term measurement results. The uncertainty results are compared with the default values in the draft standard. Guidance is provided in relation to the reliability of undertaking short-term measurements and the associated uncertainty in utilising this information to calculate long-term noise levels. The paper provides guidance on how the uncertainty associated with varying operating conditions and measurement locations can be accounted for as part of the noise compliance stage (where attended noise measurements are utilised to assess compliance with noise criteria or predictions).