Background sound level determination and the typical period

The background sound level is the reference point the whole BS 4142 assessment hangs on, and it is the number most often chosen rather than measured. Two assessors with the same dataset can report background levels differing by 6 dB simply by taking different views on which periods are representative, and a 6 dB shift moves a rating excess from a marginal indication of adverse impact to a clear one.

The standard describes the background sound level as the LA90 over the reference time interval. What it does not do is tell you which of the many LA90 values in a multi-day dataset to adopt.

What the background level stands for

The LA90 is the level exceeded for 90 per cent of the measurement period. It describes the underlying sound that remains when intermittent events have passed, which is why it is used as the baseline rather than the LAeq.

Reference time intervals are fixed. One hour during the day, taken as 07:00 to 23:00, and 15 minutes at night, taken as 23:00 to 07:00. Every LA90 used in the assessment must be derived over the applicable interval, and a night-time background quoted over an hour is not a BS 4142 night-time background.

The purpose of the figure is comparative. It sets the level against which the rating level of the specific sound is compared, so the question to ask of any candidate value is whether it describes the conditions under which the specific source will be heard.

Choosing a representative period

Representative does not mean average and it does not mean lowest. It means the conditions that will prevail when the source operates.

Start from the operating pattern. A plant running continuously across the whole week needs a background characterising every part of that week, including quiet periods. A delivery operation running between 06:00 and 08:00 on weekdays needs the background during those two hours on weekdays, and the rest of the dataset is context rather than the answer.

Where the source operates at night, the night-time background usually governs, because it is lowest. Where the source operates only during the working day, adopting a night-time background inflates the assessment and will be criticised.

Two further judgements sit inside the choice. Whether weekends differ enough to need separate treatment, which they often do in areas dominated by road traffic or commercial activity. And whether the seasonal position matters, which it does where the background is driven by vegetation, agricultural activity or seasonal traffic.

Day and night reference intervals in practice

The shoulder periods cause more disputes than the main ones. Early morning deliveries, shift changes and plant start-up frequently fall in the last hour of the night-time period, when the background is at its lowest and the assessment is at its most sensitive. Assessing that activity against a daytime background is the single most common methodological objection an environmental health officer raises, and assessing night deliveries and servicing needs the measurement periods matched to the activity rather than to convenience.

Excluding atypical data

Some exclusions are uncontroversial and some are contested. State every one and the reason, with the excluded intervals visible in the dataset rather than deleted from it.

Generally accepted exclusions:

• Rainfall, which raises measured levels and affects the microphone
• Wind speeds above the range stated in the methodology, commonly taken as 5 metres per second at microphone height
• Periods when the specific source under assessment was operating, since it would otherwise be counted in its own baseline
• Instrument faults, calibration drift beyond tolerance and obvious anomalies

Contested exclusions include the dawn chorus, individual events such as an aircraft overflight or a passing emergency vehicle, and periods of unusual activity such as roadworks or a local event. The defensible approach is to present the data both with and without the exclusion, so the effect of the judgement is visible. A dataset silently stripped of the periods that would have raised the background is the criticism that does most damage on cross-examination.

Where the specific source cannot be shut down for the survey, measure the background at a location with comparable acoustic character where the source is inaudible, and justify the equivalence explicitly.

How many days of data are enough

There is no fixed number, and the honest answer is that it depends on variability.

A suburban location with a steady road traffic background can be characterised in a few days including one weekend. A rural location with a background driven by wind in trees, agricultural activity and occasional traffic may need a week or more before the pattern stabilises. A location adjacent to an intermittent industrial neighbour may need longer still.

The test to apply is whether adding another day changes the answer. Plot the accumulating dataset, and if the distribution is still moving after five days, keep measuring. That is a defensible basis for stopping, and it is more persuasive than citing a convention.

Unattended monitoring should be supported by at least one attended period, so that somebody has actually listened to the acoustic environment and can describe the sources present. An assessment written entirely from logged data, with no observer notes, misses tonal and impulsive character that changes the rating.

Presenting the statistical spread

Reporting a single figure hides the judgement. Report the distribution and then state the value adopted and why.

A complete presentation includes a time history chart showing LA90 against time for the whole survey, a table of LA90 values by reference interval for each day and night period, a cumulative distribution or histogram of the night-time values, the modal and median values alongside the minimum, and a clear statement of the value adopted for the assessment with the reasoning.

The reasoning sentence is what a reviewer reads first. Something in the form of: the night-time background adopted is the modal 15 minute LA90 across five nights, excluding two intervals affected by rainfall, on the basis that the source operates continuously through the night. That is checkable. An unqualified statement that the background sound level is a given value is not.

Uncertainty deserves a paragraph of its own. BS 4142 expects an estimate covering instrument accuracy, the representativeness of the period and the measurement position, and a competent estimate strengthens rather than weakens the assessment. Councils normally expect a suitably qualified author for acoustic work, and professional review of any draft is strongly advised before it forms part of a Noise Impact Assessment.

Where a low background drives the outcome

In quiet rural locations the night-time background can fall low enough that almost any plant produces a rating level well above it, and the initial estimate of impact points to a significant adverse effect however modest the source.

That is where the context provisions of the standard matter. The assessment does not end with the arithmetic difference. The absolute level of the specific sound, the character of the residual sound, the sensitivity of the receptors and the degree to which the specific sound is masked all form part of the final judgement, and the standard expects them to be addressed.

Design responses follow the same logic. Selecting quieter plant at source, enclosing or attenuating it, relocating it away from the sensitive elevation and controlling its operating hours are all more productive than arguing about the baseline, and what a barrier actually achieves is worth checking before it is relied on in the design.

The last consideration outlives the application. A background that was understated at planning stage tends to resurface as a complaint after occupation, and nuisance allegations after permission is granted are decided under a different test from the one the assessment was written to.

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

This is general guidance, not legal advice. Check the position with your local planning authority or a qualified professional before relying on it. For specialist technical reports, qualified professional input may be required. PlanScope provides AI-assisted draft reports for planning applications. Reports should be reviewed before submission. We do not guarantee planning approval.