Gladiatorial games, part 14

(… continued from Gladiatorial games, part 13, posted on 10 Feb 2016).

I ended last time on the claim by Rick Methold, the expert acoustics witness put up by Chiltern District Council for its petition hearing in front of the HS2 Select Committee, that the annoyance response relied upon by HS2 Ltd to support its decision to specify 50dB LpAeq,16hr as the daytime LOAEL (lowest observed adverse effect level) threshold was “a highly annoyed curve”. An appreciation of what he meant may be gained by considering the imaginary experiment that I described in part 11 of this blog series. If we had wanted to gather data from this experiment to construct the highly annoyed bell curve, all that we would have needed to have done was to instruct the subjects to stay in the room until they found the noise level not just annoying, but highly annoying. It follows that the subjects would, on the whole, stay longer in the room, and that the mean of the highly annoyed bell curve would be at a higher sound level than the curve for the annoyed response.

Mr Methold’s exhibit A1571(47) serves to illustrate and support his claim; it is, in fact, a straight reproduction of a graph used by the Promoter’s acoustics expert, Rupert Thornely-Taylor, when he made a general presentation on acoustics to the Select Committee in July 2014 (see footnote 1). The title that Mr Thornely-Taylor has given to his graph indicates the source as a 2001 research paper from the Netherlands (see footnote 2). Although this source paper considers the distributions for little annoyed, annoyed and highly annoyed responses, Mr Thornely-Taylor’s title makes it clear that only the highly annoyed response has been used for his graph.

A1571(47) is an example of a “cumulative distribution function” (CDF), which is an alternative – and generally more common – way of presenting the data from which the bell curve that I introduced in part 11 is derived. You will recall that the height of the bell curve at any point along the horizontal axis represents the percentage of subjects that leave the room when the noise level reaches that point, and that the area under the curve to the left of that point represents the total percentage of subjects that are outside of the room at that point. Well, the CDF differs in that it plots that total percentage outside the room against the noise level in the room. So, for example, the 50 per cent line extended from the vertical axis will cut the CDF at the point on the horizontal axis corresponding to the peak of the bell curve (i.e. at the mean noise level).

Exhibit A1571(47) displays three cumulative distribution functions: one for aircraft noise, one for road traffic noise, and one for conventional (but not high speed) rail. As Mr Thornely-Taylor pointed out, “the percentage highly annoyed is greater for aircraft than it is for road which, in turn, is greater than rail”, but conceded that “maybe annoyance to aircraft has actually gone up and is higher than [is shown on the graph]”. Just where a CDF for high speed rail would lie on the graph appears to be a contentions point. Mr Thornely-Taylor claimed that “the response to high speed rail is not materially different from the response to conventional rail”. Mr Methold, on the other hand, ventured that annoyance levels for high speed rail would be higher than for conventional rail, and provided the Select Committee with a number of supporting references from international sources (see footnote 3).

Mr Thornely-Taylor pointed out to the Select Committee when he first showed them the CDFs in July 2014 that the curves demonstrate “a progressive increase as you raise the noise level” and that there is “no sudden cliff edge”. The absence of any trigger point where annoyance suddenly increases means, however, that there is no obvious level that can be identified as LOAEL.

One further feature of A1571(47) that I should point out is that the noise levels are expressed in the day-evening-night equivalent continuous sound level, which is given the symbol Lden (see footnote 4). It is hardly perverse of the Dutch authors to employ this parameter, since it is the “common noise indicator” specified in the EU Environmental Noise Directive to represent the “noise indicator for overall annoyance” (see footnote 5), but it does mean that the noise levels along the horizontal axis are not the same as are being used by HS2 Ltd – this appears to be another example of the UK refusing to march in step with our partners across the water. Fortunately, Mr Methold was able to furnish a simple conversion rule; he told the Committee to add 1.5dB to the levels on the graph to convert “into High Speed 2 talk”, So, according to Mr Methold (see footnote 6):

LpAeq,16hr = Lden + 1.5

Not to be outdone by Mr Thornely-Taylor, Mr Methold produced some cumulative distribution functions of annoyance responses of his own, but that will have to wait until the next posting.

(To be continued …)


  1. Mr Thornely-Taylor’s original slide is P5(24) in his bundle of exhibits. The comments that he made to this slide are recorded in paragraph 47 of the transcript for the morning session of the HS2 Select Committee that was held on Tuesday 8thJuly 2014.
  2. The source paper is Miedema, H M E and Oudshoorn, C G M, Annoyance from Transportation Noise: Relationships with Exposure Metrics DNL and DENL and Their Confidence Intervals, Environmental Health Perspectives, Vol 109, No 4, April 2001, pp409-416.
  3. See paragraphs 331 to 342 of the transcript of the morning session of the HS2 Select Committee that was held on Wednesday 4thNovember 2015. The supporting slides are A1571(48) to A1571(54) in Mr Methold’s bundle of exhibits.
  4. For an explanation of this parameter refer to my blog How annoying is that? (posted 18 Oct 2012).
  5. See paragraph 9 in the preamble and Annex 1 of Directive 2002/49/EC of the European Parliament and of the Council of 25thJune 2002, Official Journal of the European Communities, 18th July 2002, pp L189/12 to L189/25.
  6. See paragraph 354 of the transcript of the morning session of the HS2 Select Committee that was held on Wednesday 4th November 2015. I hesitate to take issue with Mr Methold, but feel that it seems more likely that Lden will have a higher numerical value than LpAeq,16hr. Accordingly, I suggest that the equation should be LpAeq,16hr=Lden – 1.5, instead of LpAeq,16hr=Lden + 1.5 as suggested by Mr Methold.


I wish to thank Michael Woodhouse for his suggestions and comments, which I have found invaluable in preparing this series of blogs.

Exhibit A1571(47) has been extracted from the bundle of evidence submitted to the HS2 Select Committee by Rick Methold and published on the website of the HS2 Select Committee.

Important Note: The record of the proceedings of the HS2 Select Committee from which the quotes reproduced in this blog have been taken are uncorrected transcripts of evidence, which are not yet an approved formal record. Neither witnesses nor Members have had the opportunity to correct the record in such instances, and it may therefore be subject to changes being made in the light of any such corrections being requested.


4 responses to this post.

  1. Posted by Mike on February 16, 2016 at 12:41 am

    I suggest that High Speed Rail noise follows Airplane noise, and that peak values, rather than average values, are used.
    Here in currently tranquil Toton, despite being near a railway, motorway & airport, we have no noise from road, rail or air. Even on the footbridge over the freight line with a train chugging slowly underneath. But if the flightpaths are changed to allow one or two overhead planes a month, the result is highly annoying. More so than a continuous rumble.


    • This has been a very live issue with campaigners against HS2 Mike, and the question of LAMax versus LAeq often crops up in the literature, where the consensus is that LAMax has particular relevance to sleep disturbance. The reason for this is that sleep disturbance is likely with a single noise event, which can be characterised fairly well by LAMax, but annoyance is more of a function of the cumulative effect of a number of noise events. Where LAMax falls down when looking at the annoyance response is that it takes no account of the frequency of events; it seems common sense to me that a railway with 30 train pass-bys an hour is going to be more annoying than one with a single pass-by each day.
      As to the application of LAMax to HS2, then the noise policy (“High Speed Two Information paper E20”) already follows your suggestion and LAMax values have been calculated and reproduced in the tables in Appendix 5 of the Environmental Statement. For more on this please refer to my blogs “I’m sorry what did you say” ( and the upcoming “Gladiatorial games, part 18”.


      • Posted by Mike on February 16, 2016 at 1:59 pm

        Yes, include frequency. My concern is that mitigation comparable to average noise levels will have negligible effect on High Speed Rail, with peak levels being so much higher.

      • Yes, but how would you “include frequency” Mike? The LAeq metric reflects very well the impact of higher, or lower, traffic frequencies. For example, the difference between LAeq,day and LAeq,night is directly related to the lower number of trains running during the night hours. This is why the World Health Organisation’s Guidelines recommend that a single train pass-by measure, such as LAMax, is used in addition to, not as a replacement for, LAeq and HS2 Ltd has, albeit as a late conversion, seen fit to follow this advice.
        Regarding your reference to “average noise levels”, may I be so bold as to suggest that a review of Rupert Thornely-Taylor’s evidence to the Select Committee on the morning of 8th July 2014 (paragraphs 39 to 42 in the transcript at might convince you that LAeq is not simply a crude average.

Leave a Reply

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

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

Google+ photo

You are commenting using your Google+ 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 )


Connecting to %s

%d bloggers like this: