Taking a longer-term view

Measuring noise, part 6

In my blog of 18 Jun I looked at how the noise nuisance resulting from a single noise event can be assessed. The example that I used was a train passing an observer, but the methodology would equally apply to an aircraft taking off from an airport. The characteristic of such a noise event is that it has a start point, where the noise level exceeds a specified threshold, and an end point, where the noise level sinks back below that threshold.

In many situations a noise nuisance will be virtually continuous and possibly fluctuating. Examples of this would be noise from a busy road or motorway or from a factory. This situation is illustrated in the trace below, which is taken from a useful FAQ website provided by the Civil Aviation Authority (here).

Equivalent continuous noise level (source: CAA)

One useful way of characterising the nuisance from this type of waveform is the parameter Leq, which is marked on the trace as a constant noise level. This is identified on the figure as “average sound energy”; the purists would have a problem with this description and it is more correctly described as “equivalent continuous noise level”. It is simply the constant level of noise that has the same total noise energy content as the varying noise signal that we are characterising.

The parameter Leq is important because it is the one that is used most often to determine the nuisance level of a noise source and is being employed by HS2 Ltd in its work. So how do we work it out?

The method is very similar to working out SEL, as described in my blog of 18 Jun, except that the total noise energy is calculated over a very much longer time period, T. Typically T will be a minimum of one hour, but longer periods are also common; HS2 Ltd is using T=18 hours to cover the whole of its operating day (virtually).

As was the case with SEL, it is necessary to work with a linear vertical scale with noise expressed as a pure weighted power rather than in dBA. The area under the noise trace is worked out over the period T and a rectangle of equal area constructed just as is done for SEL; however this time the base of the rectangle is not one second, but T. If this is done then the height of the rectangle will be the constant noise power equivalent of the fluctuating noise trace. Convert this power to decibels and the result is the equivalent continuous noise level parameter dBA Leq,T (which may also be written LAeq, T). The “T” term is normally omitted, but if included the unit is written in the form dBA Leq,1 hr; in this case this indicates that a time period of one hour has been employed.

The parameter Leq is also used to define non-continuous noise sources, comprising a series of separate noise events, such as aircraft noise near an airport and high speed train noise near the track. For such events the noise energy rectangle can be constructed from the individual one-second rectangles that were used to work out the SEL for each of the separate noise events. The area of this noise energy rectangle will be the sum of the areas of all of the one-second rectangles added together. The remaining steps in obtaining dBA Leq,T are the same as for the continuous noise example.

I have doubts about the validity of using Leq to assess the nuisance caused by a non-continuous noise source. Since SEL is used as the basis of the calculation, the reservation that I expressed towards the end of my blog of 18 Jun about the efficacy of SEL is also valid for Leq of a non-continuous noise source.

I am also concerned about the effect of the quiet periods between noise events on the value of Leq; these will not contribute to Leq and so will dilute the value of Leq relative to Lmax. So the single events could each be quite severe, but if they are separated by long intervals this severity will not necessarily be reflected in the value of Leq.

I am not alone in having doubts about the use of Leq when the noise source is a series of separate noise events. There is a campaigning organisation called Airport Watch, which opposes the expansion of aviation and airports likely to damage the human or natural environment. This is an umbrella organisation for a number of national bodies with serious environmental credentials and so it should know what it is talking about. In a briefing sheet How Aircraft Noise is Measured … and how it is flawed (available here) it too raises concerns about the averaging of noise inherent in the use of the Leq parameter.

Airport Watch cites an interesting result thrown up by averaging measurements at Heathrow. It says that four hours of Boeing 757 flights, one every two minutes, caused the same noise nuisance according to official measurements as one Concorde flight followed by 3 hours and fifty-eight minutes of quiet. Not very reassuring!

No less an authority than the World Health Organisation has also expressed reservations about the unquestioned use of equivalent continuous noise level to characterise noise irrespective of its nature. In its publication Guidelines for Community Noise (here) it says (in section 2.1.5 on page 23):

LAeq,T should be used to measure continuing sounds such as road traffic noise, many types of industrial noises and noise from ventilation systems in buildings. When there are distinct events to the noise such as with aircraft or railway noise, measures of the individual events should be obtained (using, for example, LAmax or SEL), in addition to LAeq,T measurements.


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: