Examination of Witnesses (Questions 6900
- 6919)
6900. One still gets an element of creep on
the minus 5 approach L90, but it is conspicuously different from
that if one is adopting L90+5.
(Mr Methold) Yes, it is considerably different.
6901. That I think reaches the position where
at that point we can pause, so to speak, because that deals with
your views of the approach on BS4142.
(Mr Methold) It does, yes. This is the first
part of the evidence, if you like, to deal with the design aim.
6902. There we see on our last slide, therefore,
the effect of the Promoter's approach. We turn in the slides to
LBH37, where we are concerned with the way in which the Promoter
has secured background noise levels.[35]
(Mr Methold) That is correct.
We have undertaken some fairly in-depth analysis of the background
noise level derivation process and we have revealed what we believe
to be some major flaws. I would like to take the Committee through
exactly what that is, because, as I have explained earlier, it
does have quite a bearing on how a 4142 assessment has been done
by the Promoter for the Environmental Statement.
6903. We can remind ourselves that that significance
is revealed by one of the earlier slides to which you drew particular
attention, that graphic representation on LBH11.[36]
Where one sees that the background noise level moves, then plainly
the effect is going to be very significant as far as the rating
levels there recorded.
(Mr Methold) Yes, that is correct
I gave an example. We now know that the Promoter wants to design
to a rating level of L90+5. Therefore, if the background noise
level that has been used for the assessment at this time is higher
than it should be, say, for example, the background noise level
should be at least 5 dB higher than has been assumed, then all
of a sudden we are into a situation where the equipment could
have been designed to something which we know will now give or
is likely to give rise to complaint. It is a very key element.
If we turn back to LBH37, there is fairly limited guidance provided
in 4142 on what background noise level we should use for assessments,
but I have listed what we do know, that it should be based upon
a typical quiet period of the day.[37]
That is a statement in the Planning Policy Guidance, note 24,
which is the national government policy on planning and noise.
It should be for the time of day when the new noise source will
be operating—clearly there is no point using a background
noise level in the middle of the night if your equipment is never
likely to be operating at that time. Monitoring must be undertaken
over a suitable period and day and night operations should be
assessed separately.
6904. Then you come to a place where the measurements
have taken place, and you have chosen 30 Hyde Park Gardens, LBH38.[38]
(Mr Methold) Yes. This graphic
shows a week's worth of measured background noise levels at 30
Hyde Park Gardens in Westminster. The graph runs from midnight
to midnight, and we have had each of the days overlaid on top
of each other in different colours. The first thing to note is
that, again, we had this drop-off in background noise levels around
about three or four o'clock in the morning mark. There are some
other interesting features. As you might expect, on Friday night
and Saturday night the noise levels are higher than a week-day
night, and during the day on Sunday the background noise levels
are lower than the rest of the days in the week, so there are
some intuitive features there. It shows that there is variability
in the lowest area of L90s there between three and four o'clock
in the morning, depending upon which day you look at Moving on
to LBH39 and how the Promoter has dealt with selecting an appropriate
background noise level.[39]
6905. The Promoter had to select a figure from
what is there revealed at 30 Hyde Park Gardens, or wherever else
that may be, and then you ask the question what has he done to
find that figure.
(Mr Methold) That is right, yes. He has used
what is called the statistical mode of all the measurements. This
is defined as the most frequently occurring value from a sample
of values. I have a table of six numbers in there, highlighting
fairly simply that the number 4 is the only one that has been
repeated and therefore it constitutes the mode of those six numbers.
6906. The mean is a different figure, of course,
and one can readily imagine further examples.
(Mr Methold) That is correct, but that, in
essence, is how the mode works . If we have a sample of several
hundred noise measurements, it is looking for the most frequently
occurring noise level. What does this mean? In the example we
have just looked at, LBH40 shows us that.[40]
6907. So we are back to 30 Hyde Park Gardens
and you are recording what the Promoter has done in respect of
that address to identify the mode.
(Mr Methold) That is correct. We have here
a period that has been selected from the previous measurements.
Between nine o'clock in the evening and seven o'clock in the morning
is the period that the Promoter in this particular instance has
said is the most sensitive for the operation of ventilation shafts.
The other point to note here is that the Promoter is saying that
between 1.30 and 5.30 in the morning he will not be operating
the ventilation shafts, and, as such, the background noise levels
for that period are not included in this assessment.
6908. That is why you have put the little box
with the minimum, mean and mode in that area of the page, with
the dashed lines running perpendicularly so that one can see that
that area is not counted.
(Mr Methold) That is correct. The coloured
lines are not shown through that zone. The horizontal red lines
are the interesting results from this analysis. The bottom dashed
line represents the minimum measured L90 from all that data
6909. It is just below 47.
(Mr Methold) That is right, just over 46. The
next dashed line up represents the arithmetic mean of all of these
samples of noise measurements, and then our solid red line across
the top is our modal value of all the measurements. In this situation,
understandably, the line is pitched around where all the squiggly
lines are the most flat (that is, where those noise levels will
be most frequently reoccurring). In this situation we have a modal
value up at 57, which is some 11 dB higher than the most sensitive
period measured during that period.
6910. And also higher than the arithmetical
mean, which some people might describe as the average.
(Mr Methold) That is right; just to show that
it is erring towards the higher noise levels in this example.
6911. Erring towards the higher means that the
whole exercise is ratcheted up. Is that right?
(Mr Methold) Exactly—the point we were
making earlier on. We have a difference here of, say, 11 dB, and
you can imagine what that means in terms of a 4142 assessment.
It is the difference, from the Promoter's point of view, designed
to L90+5, of being of "marginal significance" and of
complaints being "very likely".
6912. We can pass from LBH40, having seen what
that does as far as the mode, mean and minimum figures, and go
to LBH41 and why it is wrong.[41]
(Mr Methold) This is a summary
of the points we have just discussed, that the mode only relies
on the repetition of a particular value, which is a random coincidence,
in my view, rather than any systematic approach. It ignores the
typically quiet periods of the day, which is the guidance we are
given by national policy. As we have seen in the example before,
it is biased towards higher noise levels. I believe this is a
major flaw in the Promoter's assessment methodology and importantly
could conceal some impacts that should have been identified in
the environmental statement. We know of at least one situation
where that is the case. Of course the importance of the Environmental
Statement is to identify those impacts and provide additional
mitigation. The use of this mode is inconsistent with other recent
projects. I have never seen this approach used before, which is
why I have paid particular attention to the effects of it, and
all the other railway projects that we know about have used a
minimum value of L90 instead of this modal value Quite clearly
we do not believe this can be permitted for Crossrail on that
basis. LBH42 is a summary of the other railway projects.[42]
In relation to the Jubilee Line Extension, the extract that I
was provided with by the Promoter on request has indicated that
the reported background noise levels for fixed plant design were
based on the minimum one hour L90. In relation to Crossrail, back
in 1991, there is a report indicating that its previous assessments
were done to the minimum five minute L90. The Channel Tunnel Rail
Link has used a typical minimum five minute L90 and Thameslink
2000, we know, uses the lowest measured L90.
6913. Then you say "much lower assumptions
used compared to Crossrail". Bearing in mind what others
have done, can you help us by reference to the illustration from
30 Hyde Park Gardens at LBH40 as to which lines the others would
have used.
(Mr Methold) The measurements we see on LBH40
are 15 minute measurements, so they are recorded every 15 minutes.
The other projects have varied that slightly, in that some projects
have gone to a higher resolution of five minutes and have applied
a form of smoothing technique to come up with a representative
background noise level. But all of them are in the low region,
so they are essentially represented by the lower dashed line on
that particular graphic.
6914. The minimum line?
(Mr Methold) The minimum.
6915. In other words, a difference of 11 or
so.
(Mr Methold) In this example, yes.
6916. So LBH42 is what others have done. Then
you ask the question at LBH43: What do the Petitioners want?[43]
(Mr Methold) Clearly we want Crossrail
to abandon this approach. We do not believe it is appropriate.
We think it is concealing the results of their Environmental Statement.
We want them to reassess those impacts using a minimum background
noise level, as we would have expected them to. We also want to
ensure that we do not see this approach re-emerge during detailed
design stage
6917. That is the method of calculation. At
LBH44 you come on to Thameslink 2000 and an extract from the Environmental
Statement.[44]
What is the purpose of this reproduction?
(Mr Methold) These two paragraphs
are quite clearly stating that Thameslink has acknowledged the
emergence of new guidance; that it has had to take on board and
review and change its assessment criteria. You will remember that
Thameslink has had several incarnations. The most recent, in 2004,
reviewed its criteria and changed its criteria as a result. There
is particular reference in these paragraphs to the National Ambient
Noise Strategy and the Mayor's London Ambient Noise Strategy.
It is making particular reference to the concern about already
high noise levels and alluding to the issue of creeping noise
levels and absolute noise levels.
6918. That is LBH44, where you say that greater
emphasis has been given to environments already exposed to high
noise levels as well as the requirements of the European Directive
Would you go to LBH45, please.[45]
(Mr Methold) LBH45 is the agreed
wording for the fixed noise sources for Thameslink 2000. I will
not read this out, but the summary at the bottom is essentially
concluding that the rating level is equal to L90-5. The important
aspect with regard to Thameslink 2000 is that they developed and
agreed these criteria with the Inner London local authorities,
and we believe that the Thameslink 2000 approach is consistent
with the Petitioner's arguments.
6919. There is a consistency there between those
two. The question might be asked, therefore: Why is it said that
the Promoter cannot do this? We can see the answer to that at
LBH46.[46]
(Mr Methold) We have asked the
question directly to the Promoter and their answer has been that
they do not feel that they can accommodate the Thameslink 2000
design aim specifically in the case of ventilation shafts. The
implication of this, of course, is that, because this design aim
is being rolled out across all of the fixed noise sources, it
means that anything that is not a vent shaft is allowed to design
up to that less stringent design aim, which is in major conflict
with the Thameslink 2000 approach. The other problem we have with
that is that it is disregarding what we believe is the most recent
review of all guidance and strategies on this matter conducted
by another contemporary major railway infrastructure project.
35 Committee Ref: A81, Selection of background noise
levels (HAVGLB-14705-037). Back
36
Committee Ref: A81, British Standard 4142-Graphical Representation
(HAVGLB-14705-011). Back
37
Committee Ref: A81, Selection of background noise levels (HAVGLB-14705-037). Back
38
Committee Ref: A81, Background noise levels measured by Promoter
over one week at 30 Hyde Park Gardens (HAVGLB-14705-038). Back
39
Committee Ref: A81, What has the Promoter done? (HAVGLB-14705-039). Back
40
Committee Ref: A81, Analysis of Modal Value-30 Hyde Park Gardens
(HAVGLB-14705-040). Back
41
Committee Ref: A81, Why is this wrong? (HAVGLB-14705-041). Back
42
Committee Ref: A81, What others have done? (HAVGLB-14705-042). Back
43
Committee Ref: A81, What do the Petitioners want? (HAVGLB-14705-043). Back
44
Committee Ref: A81, Thameslink 2000 Extract from Environmental
Statement-Scoping and Methodology Report June 2004 (HAVGLB-14705-044). Back
45
Committee Ref: A81, Thameslink 2000 Extract from Environmental
Statement-Scoping and Methodology Report June 2004 (HAVGLB-14705-045). Back
46
Committee Ref: A81, Thameslink 2000 Design Aim (HAVGLB-14705-046). Back
| 
