Maximum noise events and sleep disturbance at night

An average sound level says nothing about being woken at 03:40 by a single lorry. That is why night time assessment for residential development works with two sets of numbers: the equivalent continuous level across the night period, and the individual maximum events within it. The lafmax night noise events criterion most commonly applied in English planning practice is an internal maximum of 45 dB LAFmax in bedrooms, not exceeded more than about ten times a night, drawn from BS 8233:2014 and the ProPG guidance on planning and noise, with World Health Organisation night noise advice behind both.

Miss that part of the assessment and the scheme can meet every average target and still generate complaints in its first winter.

Why an average level misses the problem

The eight hour equivalent level flattens a night into a single number. Two sites with the same figure can behave completely differently: a constant road hum at a steady level, and a quiet lane crossed by six freight trains and two refuse rounds.

Sleep research underlying the WHO guidance points to individual events as the driver of awakenings and of shifts in sleep stage, rather than the average alone. In planning terms, that means a site next to a railway, a bus depot, a hospital entrance, a service yard or a road with night HGV movements needs the event analysis whatever its average level shows.

The corollary is worth stating plainly. A site can fail on events while passing on averages, and the mitigation for the two is not the same.

Metrics used for individual events

Three measures matter, and mixing them up is the commonest error in a submitted report:

• LAFmax, the maximum A-weighted level using fast time weighting, which captures the peak of a single pass-by
• the number of events exceeding a stated LAFmax threshold within the night period, usually 23:00 to 07:00
• the sound exposure level, which describes the energy of a single event and helps compare events of different durations

Fast time weighting matters. A maximum measured with slow time weighting reads lower on a short event, and a report that does not state the weighting cannot be relied upon. State the metric, the weighting, the threshold and the counting period every time a number appears.

Internal criteria for bedrooms

The internal targets come from BS 8233:2014, which gives guideline values for steady noise in bedrooms at night, and from the maximum event criterion described above. The ProPG approach builds on both, asking for a systematic assessment of the acoustic environment rather than a single pass or fail test.

Bedrooms are the room type that governs, and the assessment should be done window by window, not as a site average. A flat with one bedroom facing the railway needs a different treatment from the flat above it whose bedroom faces the courtyard.

Where a scheme cannot meet the criterion with windows open, that has to be stated and reasoned, because it drives the ventilation strategy and the overheating position at once. Our note on railway noise and vibration for sites beside a line covers the source characteristics that make rail sites particularly event driven.

Counting events and what is acceptable

The criterion has two halves: the level and the frequency. Ten exceedances of 45 dB LAFmax across a night is the figure in common use, and it comes with two practical questions.

First, which nights. Freight timetables differ between weekdays and weekends, and a Sunday night survey on a rail site can miss most of the movements. Second, what counts as an event. A goods train pass-by is one event. A shunting movement lasting four minutes with three peaks is a matter of judgement, and the report should explain the convention used.

Where the count exceeds the criterion, the assessment should say by how much and at which facades, and then set out what the mitigation achieves. A statement that the criterion is exceeded at three of twenty eight bedroom windows, with the facade treatment reducing the exceedance to two, is a professional answer. A statement that levels are broadly acceptable is not.

Facade attenuation for peak events

Glazing performance is quoted as a weighted sound reduction index, and the weighting used matters more for event noise than for steady noise. Events with strong low frequency content, such as a freight pass-by or a heavy vehicle on a poor road surface, are attenuated less than the single figure rating suggests, which is why the low frequency correction term is part of the specification.

The package that usually results has four parts: glazing specified with the correction term stated, acoustic trickle or mechanical background ventilation so windows can stay closed, attention to the whole facade rather than the glass alone, and consideration of the window type since an openable light in a frame performs differently from a fixed pane.

Closed windows then raise the overheating question, assessed under Approved Document O and, where a planning requirement applies, through an overheating risk assessment using the CIBSE methodology. Solving noise by sealing the facade and then solving overheating by opening it is the circle these two assessments have to square together.

Measuring so events are captured properly

Survey design decides whether the data is usable:

1. Log the full time history rather than summary statistics, so events can be identified and counted afterwards.
2. Use fast time weighting and a short sample period, typically one second or shorter, for the maximum levels.
3. Cover several nights, including a weekend night and a weekday night, and state which nights and why.
4. Place the microphone at a height representative of the affected bedroom windows, and note the free field or facade condition.
5. Record the weather, and discard periods affected by rain or high wind.
6. Keep an observer log or an audio recording so that events can be attributed to a source rather than counted blind.

Attribution is what separates a useful report from a table of numbers. A 62 dB LAFmax at 02:15 attributed to a passing freight train tells the officer something. The same figure with no source given tells them nothing, and it might have been a fox.

Reporting alongside average levels

Put both sets of results in the same table, facade by facade, with the criterion stated next to the result, and a clear statement of compliance or exceedance for each. Follow that with the mitigation and the residual position after mitigation.

Then say what the scheme does about the rooms that remain marginal. Options include moving bedrooms to a quieter facade, single aspect units reoriented, a wintergarden or a screened balcony, or a layout change that puts circulation on the exposed side. Design solutions carry more weight with environmental health officers than glazing specifications alone, because they do not depend on the occupier's behaviour.

The noise impact assessment page sets out the wider scope of the report, and the same event logic applies to road sources, as our piece on road traffic noise for a new housing site describes.

This is general guidance, not legal advice; agree the assessment method with your local planning authority's environmental health team, and have the work carried out by a suitably qualified acoustician. PlanScope provides AI-assisted draft reports for planning applications. Reports should be reviewed before submission. We do not guarantee planning approval.