Facing a brick wall, part 1

Originally published on HS2 and the Environment:

The middle of a large field is probably a good place to take measurements of ambient noise with a sound level meter, although I notice that, for some reason, HS2 Ltd acousticians prefer to set their meters up beside busy roads. The reason why the field is to be preferred is that it should be fairly devoid of reflecting surfaces that might cause local sound wave reflections that could distort the measurement. Measurements that are made in the absence of such local reflections are termed “free-field”, and all of the noise figures in the tables in the sound, noise and vibration section of Volume 5 of the HS2 Phase 1 Environmental Statement (ES), whether they be empirical or predicted, are expressed as free-field values (see footnote 1).

Whilst people do sometimes sleep in fields, either due to misfortune or for recreation, acousticians assume that, for the purposes of assessing the sleep disturbance potential of noise, sleep takes place in a bedroom. Whilst this seems a reasonable assumption, the parallel supposition that people sleep only at night is probably far less justifiable given the demands of modern living and shift work in particular. But I will pass over this particular objection, as it is the assumption about the bedroom upon which I wish to concentrate in this posting.

Bedrooms may be thought of as a volume enclosed by walls, a floor and a ceiling, and it is the noise that enters into this enclosed space from a noise source outside of the bedroom that gives rise to sleep disturbance effects. It is most convenient for this purpose to think of the bedroom as having one external wall facing the noise source and for the noise to be entering the room through this wall, which is referred to as the “most exposed façade”. A commonly-encountered way of assessing the noise level that will be present within the bedroom is to predict the noise pressure that will be incident upon the exterior of the most exposed façade and reduce it by the degree of sound insulation afforded by the external wall that forms the façade, expressed in decibels. This implies that it is the external incident sound level that is important, and this dictates that a so-called “façade” measurement is required; this is achieved in practice by setting up the sound level meter just in front of the façade of the external wall, typically at a distance of one metre from the wall.

HS2 Ltd has followed this convention in High Speed Two Information Paper E20 in defining the Significant Observed Adverse Effect Level (SOAEL) and Lowest Observed Adverse Effect Level (LOAEL) at night in terms of the maximum level of a single train pass-by, L_pAFMax, in that this level has been specified as a façade measurement. For reasons that I can’t begin to understand, but which I am sure could be satisfactorily explained by an acoustician, when LOAEL and SOAEL have been defined in terms of the day and night continuous equivalent sound levels, L_pAeq,16hr and L_pAeq,8hr, free-space levels have been specified (see footnote 2).

What is important to recognise is that a sound level meter will, when measuring sound levels emanating from the same source, tend to register a different level when used in free-space configuration to what it will record when set up to make a façade measurement. The reason for this is perhaps best understood by returning to the sound level meter that I described at the start of this posting, which is sitting in the middle of a field happily registering the free-field sound level. If we were able instantly to construct a wall just behind the sound level meter, with respect to the source of the noise, we would likely find that the measurement will change. The reason for this is that the wall will act as a sound reflector and send sound energy back towards the sound level meter.

The impact that the façade will have upon sound level meter readings can only confidently be predicted by complex analysis of each individual case, and depends upon a number of factors including the size of the wall, the angle that it makes with the line connecting it to the noise source and the distances involved; it will also vary across the frequency spectrum. However, the literature encourages us to regard the effect as one of increasing the level measured and to employ a fixed value to convert from façade to free-field and vice versa (see footnote 3).

However, what we appear to be faced with when consulting the HS2 Ltd documentation is a total failure to take the façade effect into account when it comes to assessing impacts, or even, for the most part, to acknowledge that the effect exists, and this omission has been repeated countless times in front of the HS2 Select Committee when the Promoter’s legal team have quoted predicted and measured sound levels to petitioners. Without exception, all of the levels that I have heard quoted have been free-field and, in the case of the train pass-by thresholds, experts, lawyers, petitioners and Committee Members alike have, almost to a man/woman, failed to comment that the LOAEL and SOAEL thresholds to which they are comparing them are expressed in coin of a different currency.

It all seems a bit of a mess to me and one which is – surprise, surprise – serving to understate the impact of the train pass-by noise. Since, by the time that I was given a slot before the HS2 Select Committee in January this year, nobody had mentioned this issue, I determined that it would be one of the matters that I would raise, but for a report on that you will have to wait until part 2 of this blog.

(To be concluded …)

Footnotes:

1. In the case of the operational noise assessment, that all of the figures in the ES are expressed as free-field values is confirmed in paragraph 1.3.7 in Annex B to Appendix SV-001-000 of the ES (for baseline data) and paragraph 1.1.15 in Annex D2 of that appendix (for the predictions of operational noise).
2. That the L_pAFMax thresholds for SOAEL and LOAEL are expressed as façade levels is explicitly stated in paragraphs 15 and 18 of Appendix A to Information Paper E20 and in Table 1 in Appendix B. That the L_pAeq thresholds are all free-field may be inferred from footnote 5 to Appendix A and footnote 9 to Appendix B.
3. Take, for example, the guidance in paragraph 9 in Annex A of Planning Policy Guidance 24(PPG24), which I understand is still used by local authorities despite being superseded by the Planning Practice Guidelines. This paragraph advises that “façade levels should be assumed to be 3 dB(A) higher than levels measured away from any buildings [i.e. free-field levels], unless a more accurate figure is available”.