Checklist: noise survey design for a mixed use scheme

A noise survey for a planning mixed use scheme has one job: to produce a defensible baseline that someone else could repeat and get the same answer. Everything in the checklist below exists because a reviewer, an objector's acoustician or an inspector has at some point asked a question the original survey could not answer. Get the positions, the duration and the records right and the assessment that follows is straightforward. Get them wrong and no amount of modelling rescues it.

Mixed use makes this harder than a residential scheme, because the site contains both the receptors and the sources. New flats above a restaurant need a baseline for road traffic at the facade, a characterisation of the commercial plant that will sit on the same building, and an understanding of the existing businesses whose operations the new residents will hear.

Choosing measurement positions

Positions have to serve two separate assessments. For the incoming residential use, measure at the proposed facade locations, at a height representative of the upper floors rather than at ground level, because traffic noise at the fourth floor is several decibels different from the pavement. For the outgoing plant and activity noise, measure at the nearest existing noise-sensitive receptors, as close to the affected facade as access allows.

Record for every position:

• Grid reference or latitude and longitude, plus a site plan with the positions marked.
• Microphone height above local ground, and whether the measurement is free-field or facade-mounted.
• A photograph in each direction, with the nearest reflecting surfaces visible.
• The reason the position was chosen, including any position you wanted and could not get.

Facade measurements and free-field measurements are not interchangeable. A facade-mounted microphone reads higher because of the reflected component, and the convention is a correction of about 3 dB when converting between them. Say in the report which you did, rather than leaving the reviewer to infer it from a photograph.

Duration and the periods to cover

The minimum that survives scrutiny on a mixed use site is a continuous unattended measurement over at least one full weekday 24-hour period, plus coverage of a weekend period where the surrounding uses behave differently at weekends, which near a high street they always do.

Within that, the parameters to log differ by assessment. For internal ambient noise levels under BS 8233 the relevant figures are the daytime 16-hour and night-time 8-hour equivalent continuous levels, plus the maximum levels during the night for the sleep disturbance question. For plant noise under BS 4142 the relevant figure is the representative background level, measured as the 90th percentile over reference periods of one hour during the day and 15 minutes at night.

Choosing the representative background is the single most contested number in any plant assessment. Do not use the arithmetic mean of all the night-time 15-minute values, because that overstates the background and flatters the proposal. Look at the distribution, identify the typical lowest values that recur rather than the single lowest, and show the data so the reader can see your reasoning.

Weather, calibration and instrumentation

A laboratory calibration within the standard two-year interval and a field drift under 0.5 dB are the usual acceptance criteria. If the post-measurement check shows more drift than that, say so and explain what you did about it, because a reviewer who finds an unexplained drift discounts the whole dataset.

Wind above about 5 m/s at the microphone and any period of rainfall should be excluded from the analysis and the exclusions listed. And the met data has to come from somewhere credible, either an on-site station or the nearest official observing station, with the distance stated.

Sources to characterise separately

Lumping everything into an ambient figure wastes the survey. On a mixed use site the sources that need separating are road traffic, any rail or tram corridor, existing commercial plant on neighbouring roofs, deliveries and refuse collections, licensed premises and their patrons, and the proposed plant itself once specified.

Attended measurements are the only way to do this properly. Spend time on site during the evening dispersal period outside a pub, during a morning delivery, and during a refuse collection, and record short samples of each with notes. Those samples are what let you apply an acoustic feature correction honestly later, and they are what answer the agent of change question when an existing business objects.

Where the scheme includes a kitchen, the extract and its termination need characterising at the same time, which brings the acoustic and the odour work together. The design responsibility for that system is set out in the guide to the Ventilation and Extraction Statement, and the two assessments should share the same equipment schedule.

Avoiding atypical periods

Survey timing changes the answer, so document the decision. Periods to avoid unless the assessment is specifically about them: school holidays and half terms near a school or a residential area with families, the fortnight either side of Christmas, bank holiday weekends, university vacations in a student area, and any week when roadworks or a diversion is in place on the nearest classified road.

Check the local highway authority's roadworks register for the survey week before you book the equipment. A survey carried out while the main road is closed produces a baseline that is unusable and a cost that is unrecoverable.

Term time in a university city is a particular trap. The evening ambient level on a street of student housing in June bears no relation to the level in November, and an assessment that uses the quieter baseline will be challenged.

Processing the data and handling outliers

Keep the raw data. Present the time history as a chart alongside the summary tables, because the shape of the curve shows the reviewer what kind of place this is. Mark the excluded periods on the chart rather than silently deleting them.

Outliers need a reason, not a rule. A single 15-minute interval 20 dB above its neighbours at 03:00 is probably a vehicle alarm or a dog, and excluding it is reasonable if you say so. Excluding the four quietest intervals because they produce an inconvenient background is not.

Where the proposal relies on openable windows for ventilation, the internal levels have to be assessed with windows open and the overheating strategy checked against the same assumption, which is where a noise assessment and an Overheating Risk Assessment can contradict each other if they are commissioned separately.

Reporting so the survey can be repeated

The test is simple. Could a competent acoustician with your report, and no access to you, go to the site next month and repeat what you did? That means positions with coordinates, heights, dates and times to the minute, equipment serial numbers, calibration certificates in an appendix, met data, the exclusion log, and the raw logged data supplied on request.

Name the author and their professional membership. Where a draft has been prepared from site information and then finished by the acoustician, say which parts were measured and which were modelled. The professional directory sets out the review route for drafts that need a named competent author before submission.

This is general guidance, not legal advice. Survey conventions, standards and local validation requirements change, so confirm the scope with the environmental health officer before you commission anything. For specialist technical reports, qualified professional input may be required.

PlanScope will draft the structure and policy sections of a Noise Impact Assessment from your scheme details, leaving the measurement and analysis to the acoustician who signs it.

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