When to run baseline noise surveys and for how long

Baseline noise survey timing decides whether the assessment that follows is worth anything. Get the duration and the periods wrong and the background sound level is unrepresentative, the rating level calculation inherits the error, and the environmental protection officer asks for the survey again. The general rule is simple enough to state: the survey has to capture the quietest conditions that will exist while the noise source is operating, at the position where the sensitive use is, in the same seasonal and operational conditions.

For most architects the practical question is how long to book the acoustician for. The answer depends on what the noise source does and when, and it should be agreed with the council before the equipment goes out.

Matching the survey to the noise source

The survey is not measuring the development. It is measuring what is there now, so that the proposed sound can be judged against it. So the survey design follows from the operating pattern of the proposal.

A plant item running 24 hours needs night-time background data, because the night background is lower and it is the night figure that constrains the design. A retail unit open 09:00 to 18:00 does not need a night survey for its plant if the plant genuinely switches off, though officers will ask you to prove that with a condition. A scheme introducing dwellings next to an existing industrial estate needs the reverse: a survey of the estate's noise, over a period covering its busiest operations.

Where the proposal is residential development exposed to existing noise, the assessment is about internal and external levels for future occupiers, and the relevant guidance shifts towards BS 8233:2014 and ProPG. Where the proposal introduces a new source near existing dwellings, BS 4142:2014+A1:2019 governs and the background sound level becomes the reference. Two different questions, two different survey designs, one common mistake of running a single generic survey and hoping it serves both.

Weekday and weekend coverage

Background sound varies across the week, and the variation is not small in mixed areas. A weekday daytime survey in a district with commercial activity records a background that a Sunday morning will not reproduce.

The defensible minimum for most schemes is a period covering at least one full weekday and one weekend day, with the night periods included where the source operates at night. For sources operating every day of the week, five to seven consecutive days of unattended logging is common practice and removes the argument about representativeness.

Where the proposal is a venue or a use whose own pattern is weekly, match the survey to it. A place of worship with a Friday or Sunday peak, a sports facility with weekend fixtures, a nightclub: in each case the assessment day is the busy day, and a Tuesday survey tells you nothing about it.

Night-time periods and the hours measured

The night period in BS 4142 terms runs 23:00 to 07:00, and the reference time interval at night is 15 minutes, against one hour during the day. That matters because a 15 minute interval catches short quiet troughs that an hourly average smooths away.

The background sound level used is a statistical measure, the level exceeded for 90 per cent of the measurement interval, and the lowest representative value across the relevant periods is what a careful assessment uses. Picking the mean of the night intervals is a common way of producing a flattering answer, and reviewers look for exactly that.

One practical point about the small hours. In many residential areas the quietest period is roughly 01:00 to 04:00, and a survey that stops at midnight has missed it. If the plant runs all night, the survey has to be there at 3am.

Weather conditions that invalidate data

Rain on a microphone produces spurious high levels. Wind above roughly 5 metres per second produces both microphone noise and changed propagation. Both make data unusable, and both have to be logged.

Deploy with a weather station or use nearby official observations, record the conditions, and exclude affected intervals with the reason stated rather than silently dropping them. A survey report that presents a clean dataset with no gaps over a British winter week invites the question of what was removed.

Frozen or waterlogged ground, snow cover and unusually still anticyclonic conditions all change measured levels. Note them. The point is not perfectionism, it is that the officer needs to know which intervals they can rely on.

Seasonal sources and the calendar

Season changes what is there to measure. Open windows in summer raise the background in residential streets. Outdoor venues, beer gardens and sports pitches are audible in summer and silent in February. Agricultural operations and grain driers are seasonal. Leaf cover on a tree belt alters propagation slightly, though far less than people assume.

School term matters, and it cuts both ways. A survey in half term understates the background near a school and understates traffic on the local network, which is why the same caution applies here as in noise surveys, school holidays and background levels. Avoid the Christmas period entirely for the same reason traffic counts avoid it.

If the proposal's own noise is seasonal, the assessment needs the season that matters, and that can mean waiting. Say so early in the programme rather than at the point the application was meant to be submitted, and treat it the same way you would treat any other seasonal survey constraint on the project timeline.

Attended and unattended measurements

Unattended logging gives you duration and statistics. Attended measurement gives you the acoustician's ears, which is how sources get identified, how acoustic features such as tonality and impulsivity are recognised, and how spurious events get explained.

Most competent surveys combine them: several days of unattended logging with attended spot measurements during the critical periods. For a small scheme with a single plant item, a shorter attended survey at the quietest expected time may be proportionate, and there is no benefit in a week of logging on a garden office.

• Unattended logging: representativeness, statistics, the lowest background value
• Attended measurement: source identification, acoustic features, subjective description
• Weather logging: interval validity
• Photographic record of the measurement position, height and distances

The measurement position is worth its own sentence. Record height above ground, distance from the facade, distance from reflecting surfaces and the exact location on a plan, because a repeat survey in three years needs to be at the same place.

Agreeing the survey with environmental health

The single most useful thing you can do is agree the survey design in writing before it happens. Most environmental protection teams will confirm the measurement positions, the periods, the duration and the guidance to be applied if you ask them, and some publish a standing note setting out what they expect.

1. Write to the environmental protection team with the site, the proposal and the proposed survey design.
2. Include a plan showing the proposed measurement positions and the nearest sensitive receptors.
3. State the periods, the duration and the standards you intend to apply.
4. Ask whether any local guidance or supplementary planning document applies.
5. Keep the reply and reference it in the Noise Impact Assessment.

Where that agreement exists, an objection to the survey method rarely succeeds. Where it does not, you are arguing about method when you should be arguing about mitigation. Programming the survey alongside the other seasonally constrained work on the project, including any protected species survey windows, keeps the whole evidence base moving together rather than in series.

Noise assessments are normally expected to be authored by a suitably qualified acoustician, and professional review is strongly advised. This is general guidance, not legal advice. Check the position with your local planning authority or a qualified professional before relying on it. PlanScope provides AI-assisted draft reports for planning applications. Reports should be reviewed before submission. We do not guarantee planning approval.