Background sound level and why it governs

Background sound level is the A-weighted sound pressure level exceeded for 90 per cent of a given time interval, written as LA90,T. In a BS 4142 assessment it is the number everything else is measured against, because the standard works from the arithmetic difference between the rating level of the source and the background level at the same place.

Get the background wrong by 3 decibels and the conclusion moves by a whole category. That is why acousticians argue about it more than about the source.

What the measure describes

LA90 is a statistical descriptor. Over a measurement period, the sound level rises and falls as cars pass, doors close and birds sing. LA90 is the level that was exceeded for 90 per cent of that period, so it sits near the bottom of the distribution and approximates the underlying sound that is there when nothing in particular is happening.

That underlying sound is what BS 4142:2014+A1:2019 calls the residual sound, and the background level is the measure of it. The standard is explicit that background is not an average. LAeq, the energy-average level, is pulled upwards by short loud events and describes something different.

Use LA90 when you are asking how audible a new source will be. Use LAeq when you are asking about total exposure.

Why a statistical level rather than a mean

A new extract fan will be heard against the quiet moments, not against the average. A resident lying awake at two in the morning is comparing the fan to the sound of the street between vehicles.

The 90 per cent exceedance figure captures that quiet floor while still excluding the very quietest instants, which are unstable and easily contaminated by the measurement itself. It is a compromise, chosen because it behaves consistently across different environments, and it is one reason BS 4142 results are reproducible between consultants when the survey is done properly.

Choosing a representative period

The standard requires the background to be representative of the period during which the source will operate. That is a judgement, and it has to be justified in the report rather than asserted.

Reference time intervals are 1 hour during the day, taken as 07:00 to 23:00, and 15 minutes at night, taken as 23:00 to 07:00. A plant item running 24 hours needs both. A restaurant extract running until midnight needs a night-time background for the last hour.

• Survey long enough to see the pattern. A single evening tells you nothing about a Sunday.
• Avoid atypical conditions: heavy rain, wet road surfaces, wind above about 5 metres per second, school holidays, and any week when the neighbouring road is closed.
• Check what was happening. A works compressor running during the survey raises the background and flatters the assessment.
• Where levels vary through the week, report the range and explain which value has been taken forward and why.

Choosing the lowest single 15 minute LA90 of the week is over-cautious. Choosing the highest is indefensible. The usual practice is to identify the typical lowest levels that recur, and to test the assessment against a lower value as a sensitivity check.

Measurement position and height

Measure at the assessment location, which means at the nearest or most affected noise-sensitive receptor rather than at the site boundary.

Free field measurement is preferred: at least 3.5 metres from any reflecting surface other than the ground, with the microphone 1.2 to 1.5 metres above ground for a ground floor receptor. Where free field is impossible on a tight urban site, measure close to the facade and apply the correction the standard sets out, then say in the report that you have done so.

For an upper floor window, the position should represent that window, which often means a measurement at height from a neighbouring balcony or a modelled correction. Do not measure at ground level and quietly apply it to a third floor bedroom.

Record the position on a plan and photograph it. A reviewer who cannot see where the meter stood cannot check anything.

What the number does to the outcome

BS 4142 compares the rating level of the specific sound, which is the specific sound level plus corrections for tone, impulsivity, intermittency and other characteristics, with the background sound level.

The phrase "depending on context" is doing real work. The standard requires the initial estimate to be modified by context: the absolute level of the sound, the character of the area, the sensitivity of the receptor and the presence of other sources. A rating level 10 decibels above a background of 25 decibels in open countryside is a different problem from the same arithmetic in a city centre.

Character corrections are where the arithmetic gets contested. Tonality can add up to 6 decibels, impulsivity up to 9, and intermittency 3, and Annex C of the standard gives an objective method for tonality that is more defensible than a subjective judgement.

Errors that inflate the background

• Measuring with the existing plant running, so the source being replaced is inside the background.
• Taking a daytime background and applying it to night-time operation.
• Averaging LA90 values arithmetically across a night instead of selecting a representative interval.
• Measuring against a facade without applying the free field correction.
• Surveying in wet weather, when road noise is several decibels higher.
• Positioning the meter on the noisier side of the building from the receptor.

Each of these raises the background, narrows the difference and makes the proposal look better than it is. They are also the first things an environmental health officer looks for, which is why a proposal that fails on a re-survey is worse than one that never claimed compliance.

Illustrative example. A proposed air handling unit serves a new cafe in a suburban parade. The consultant measures a night-time background of 38 decibels at the rear of the parade, with the existing chiller running. Switched off, the background falls to 31. The rating level of the new unit is 40. On the first figure the difference is +2 and the impact reads as low. On the correct figure it is +9 and an acoustic enclosure is needed. The figures are illustrative.

Reporting it so someone else can check

State the instrument type and class, the calibration before and after with the drift recorded, the dates and times, the weather throughout, the positions with photographs and a plan, and the full time history of LA90, LAeq and LAmax.

Then set out a table of derived background levels per reference time interval with the value taken forward and the reasoning for it. Where the assessment is marginal, include the sensitivity test.

This level of detail belongs in a Noise Impact Assessment prepared by a suitably qualified acoustician, and professional review is strongly advised before submission because environmental health officers read these closely. Councils treat a well-evidenced background as the foundation of the whole case. A comparable example on the daylight side is the vertical sky component, where one derived figure carries the same weight in a decision.

Anyone scoping a scheme can check which technical reports apply before commissioning the survey, which usually saves a wasted visit in the wrong season.

This is general guidance, not legal advice. Check the position with your local planning authority or a qualified professional before relying on it.

Enter your postcode on PlanScope to see which reports your site is likely to need.

PlanScope provides AI-assisted draft reports for planning applications. Reports should be reviewed before submission. We do not guarantee planning approval.