A weighty matter

Measuring noise, part 7

Whilst the primary aim of this series of blogs on measuring noise has been to provide the reader with the basic knowledge required to understand what the acoustics experts are saying about the noise nuisance caused by HS2, I have not been slow to point out where I feel that the standard methodologies employed by these experts are open to question. Now I am afraid I have yet another spanner to lob into the works.

In my blog of 10 Jun I introduced the application of A-weighting to a noise measurement to reflect the decreased sensitivity of the human ear to low frequency or high frequency sounds. I said then that A-weighting was the most commonly used weighting standard. This weighting standard first came into being in the 1930s when it was adopted as a US standard following research carried out into the sensitivity of human hearing to single tones. Its use became widespread in the 1950s, when it was adopted by the International Standards Organisation (ISO) and was warmly embraced by the acoustic engineering community, largely because it was simple to implement and employ. Since that time the use of A-weighting has been largely unquestioned; largely unquestioned, but not by everyone. There is a growing body of opinion that the use of A-weighting has severe drawbacks in certain applications.

There is an excellent review of the growing body of evidence against A-weighting in the paper The Impact of A-weighting Sound Pressure Level Measurements during the Evaluation of Noise Exposure by St Pierre and Maguire (here). Commenting on the evolution of the A-weighting standard, the authors of this paper say that it tells us two very important things:

First is that it is only representative of human ear response at low sound levels, mainly below 60 dB. Numerous studies have shown that the correlation between dBA measurements and loudness erodes as the sound pressure level is increased. Secondly, the contours were designed using single tones and therefore are mainly applicable to single tone sounds. For example, random noise is generally perceived as louder than a single tone at the same sound pressure level, regardless of the weighting.

They also state:

The sharp rolloff at low frequencies minimizes their effect on the overall dBA reading, and in some instances, large low frequency tonal components can have no effect on the actual dBA measurement. Low frequency noise, however, has been established as an important factor in subjective assessment of loudness and annoyance.

So what do we know about the characteristics of the noise generated by the aerodynamic effects caused by a train travelling at speeds above 300 kph?

  • It is very loud (certainly above 60 dB in many locations)
  • It is a complex mixture of random noise effects (not a single tone)
  • It is dominated by low-frequency sounds (refer to section 2.2.3 of the US FRA manual)

So if we are to believe St Pierre and Maguire, using A-weighted measurements is a fairly rubbish way of assessing the noise nuisance of HS2. Just how rubbish can be assessed by the following additional comment made by our authors:

… dBA measurements can underestimate loudness by as much as 14 dB when the noise primarily consists of low frequency components (below 400 Hz).

That this underestimation is important can be assessed by another comment made by St Pierreand Maguire:

Low frequency noise has been shown to increase cortisol values, which is an indicator of stress. Other physical effects attributed to low frequency noise include peripheral vasoconstriction, elevated blood pressure and greater risk of cardiovascular disease.

In summary, St Pierre and Maguire say:

… there is a large amount of evidence that measuring A-weighted sound pressure level is not necessarily indicative of the loudness of noises. This is especially true when the noise is complex and/or composed of low frequency components.

And they conclude:

… until the acoustic community begins to seriously question the use of A-weighted measurements, more accurate measurements will continue to be ignored by both engineers and manufacturers.

Our two authors are not just two lonely voices in the wilderness. The World Health Organisation its publication Guidelines for Community Noise (which may be downloaded here) says (in section 3.9 on page 53):

The evidence on low-frequency noise is sufficiently strong to warrant immediate concern. Various industrial sources emit continuous low-frequency noise (compressors, pumps, diesel engines, fans, public works); and large aircraft, heavy-duty vehicles and railway traffic produce intermittent low-frequency noise. Low-frequency noise may also produce vibrations and rattles as secondary effects. Health effects due to low-frequency components in noise are estimated to be more severe than for community noises in general (Berglund et al. 1996). Since A-weighting underestimates the sound pressure level of noise with low-frequency components, a better assessment of health effects would be to use C-weighting.


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: