Determining background sound level for a BS 4142 assessment

The background sound level in a BS 4142 assessment is the A-weighted level exceeded for 90 per cent of the measurement interval, and it is the figure that decides the outcome. Everything else in the method is arithmetic applied to it. Get the background wrong by 3 dB and an assessment that showed a low impact shows an adverse one, or the reverse, which is why reviewers read the survey section before they read the conclusion.

The standard asks for a level that represents the background during the periods when the source operates. That word, representative, does most of the work and causes most of the arguments.

What the standard asks the background level to represent

BS 4142:2014+A1:2019 compares a rating level, derived from the specific sound of the source plus corrections for acoustic character, against the background sound level at the assessment position. The difference drives the conclusion: a difference around +10 dB or more is an indication of a significant adverse impact, around +5 dB an indication of an adverse impact, and a difference at or below zero an indication of low impact, all read in context.

The background must therefore describe the acoustic environment the source will sit within, at the place where people are affected, during the times the source runs. It is not the quietest moment recorded, and it is not an average of everything measured across a week.

Two decisions follow from that. Which periods to measure, and how to reduce the measured data to a value you can defend. Both belong in the report with the reasoning visible.

Choosing measurement positions and heights

Measure at the assessment location, which is where people are exposed: the garden, the window, the balcony of the nearest sensitive receptor.

Free-field measurements should be taken away from reflecting surfaces, conventionally at least 3.5 metres from any facade, and facade measurements need the standard correction applied and stated. Height should reflect the receptor. A ground-floor garden position does not represent a third-floor bedroom window with a different view of the source and a different screening situation, and on a scheme with upper-floor receptors you need either a raised position or a justified correction.

Where access to the receptor is refused, a surrogate position is acceptable if the report explains why it is representative and what the differences are. Say where the microphone actually was, with a photograph and a marked plan. Reviewers check that first, because a position chosen for convenience next to a busy road inflates the background and flatters the assessment.

Survey duration and the representative period

Match the survey to the operating pattern of the source. Plant running continuously through the night needs night-time data across the relevant reference interval. A delivery yard operating early morning needs data covering those hours specifically.

Reference time intervals in the standard are an hour during the day and fifteen minutes at night, which sets the resolution of the data rather than the length of the survey. In practice a defensible survey runs for several consecutive days, covering weekdays and weekend where the environment differs, and enough nights to show whether the quietest period is consistent or variable.

Avoid school holidays where the acoustic environment depends on term-time traffic, and note any unusual conditions such as roadworks, a closed neighbouring business or an event. Record weather throughout, and discard periods with rain or wind above the limits the instrument manufacturer and the standard indicate.

Excluding extraneous events without distorting the data

Extraneous sound is anything not representative of the normal acoustic environment: a dog barking at the microphone, a neighbour's lawnmower, an aircraft on an unusual routing, the survey team talking.

Exclusion needs evidence. The defensible approach is continuous logging with audio recording, then time-history analysis to identify and remove the affected intervals, with the number of intervals removed reported. An unattended survey with no audio and a sentence saying extraneous events were excluded cannot be checked and will be challenged.

Be careful about what counts as extraneous. Traffic on the road that has always been there is part of the background, however much it raises the figure. A seasonal source such as a nearby sports ground is part of the environment during its season. Removing everything that inconveniences the assessment leaves a background level that describes nowhere.

Reducing the data to a typical value

The standard does not prescribe a single statistic, which is where practitioners diverge. Common approaches include the modal value of the LA90 values across the relevant periods, the arithmetic or energy average of those values, or the lowest level that occurs with reasonable regularity.

Whichever you choose, state it, show the distribution, and justify the choice against the operating pattern of the source. A histogram or a time-history plot of the LA90 values does more for the credibility of the report than a paragraph of prose.

Using the single lowest measured LA90 is the most common criticism from reviewers and from environmental health officers. It is defensible only where the source operates precisely during that quietest period and the value is not an outlier. The mirror error is using an average across a whole night when the plant runs only in the small hours.

A quiet rural baseline

Low background levels compress the margin for error and inflate the calculated difference. A rating level of 35 dB against a background of 25 dB gives a difference of +10 dB, even though the absolute level is modest.

The standard deals with this through context. The assessment should discuss the absolute levels, the character of the sound, the existing use of the area, and the likely response, rather than resting on the arithmetic alone. Where the difference indicates an adverse impact but the absolute levels are low, say so and explain the reasoning. Where the background is so low that measurement approaches the noise floor of the instrument, report that too.

Context works in the other direction as well, and an assessment claiming low impact purely because an urban background is high should set out what the source adds in absolute terms.

Uncertainty and instrumentation

Report the uncertainty. The standard expects an estimate covering the measurement position, the instrument, the duration and the operating conditions, and an assessment that presents figures to one decimal place with no uncertainty statement reads as inexperienced.

Use a Class 1 instrument, calibrate before and after each survey period, and report the drift. Keep the calibration certificates current and reference them. These are basic points, and reviewers still find reports that omit them.

What reviewers find most often

• A single short attended survey used to characterise a night-time background.
• Background taken at a convenient position rather than at the receptor.
• No audio recording, with extraneous exclusions asserted rather than evidenced.
• The lowest measured LA90 adopted as the typical value with no distribution shown.
• Facade measurements used without the correction, or with it applied twice.
• Character corrections applied to the source with no justification, which changes the rating level and therefore the difference.

Where plant serves a kitchen or a commercial unit, the noise assessment needs to be read alongside the Ventilation and Extraction Statement, because the fan selection that solves an odour problem often creates the sound source being assessed.

PlanScope drafts a Noise Impact Assessment structured to the sections councils expect, with survey data, positions and measured levels flagged as facts that must be supplied rather than assumed. Councils expect a suitably qualified author for this work and professional review is strongly advised: the professional directory is the route to an acoustician who can review and sign off the draft. PlanScope provides AI-assisted draft reports for planning applications. Reports should be reviewed before submission. We do not guarantee planning approval.

This is general guidance, not legal advice, and it summarises rather than reproduces the standard. Work from the current published version of BS 4142 and confirm the approach with your local authority's environmental health team.