Railway vibration and the thresholds that apply

A railway vibration assessment for planning has to deal with two different effects that share a cause and nothing else. Groundborne vibration is felt, and it is assessed against the dose-based criteria in BS 6472-1. Groundborne noise, often called reradiated noise, is the low rumble radiated by a floor or wall set in motion by the same train, and it is assessed as an internal noise level. Mitigation that fixes one can make the other worse, and a report that treats them as a single topic will be sent back.

The other thing to know at the outset is that this work sits outside PlanScope's 66 report types. PlanScope does not produce a vibration assessment, and the measurement and modelling have to be commissioned from an acoustics and vibration consultant. The airborne noise element of a railway-side scheme is a different matter and can be drafted as a Noise Impact Assessment, where councils usually expect a suitably qualified acoustician and professional review before submission.

Vibration and reradiated noise as separate effects

Groundborne vibration from rail travels through soil and foundations and arrives as motion in the building. People perceive it in the frequency range of a few hertz to around eighty, and floors can amplify it at their own resonant frequencies, so an upper floor can be worse than a ground floor even though it is further from the source.

Reradiated noise is the audible consequence of the same motion. Surfaces vibrating at frequencies in the tens of hertz push air, and the occupant hears a rumble with no obvious source. It is unaffected by glazing, because it does not arrive through the window, which is why acoustic glazing specified for airborne train noise does nothing for it.

The two effects also peak at different times. Passenger services dominate the daytime picture, while heavy freight, which produces the highest vibration levels on many lines, often runs at night. A survey that misses freight movements will understate the problem entirely.

Measurement positions and durations

Measure on the site, at the proposed building positions, coupled properly to the ground. Triaxial transducers at ground level, and where possible at foundation depth in a trial pit, give the input data for the modelling.

Duration matters more than instrumentation. A representative survey covers a full week where possible, and at minimum several consecutive 24-hour periods, so that the range of train types, speeds and loadings is captured. Record a train log against the measurement so each event can be attributed, and obtain timetable and freight path information rather than assuming.

Two site-specific factors change the answer significantly. Ground conditions govern propagation, so the soil profile, stiffness and water table from the ground investigation feed directly into the prediction, which is why a geotechnical assessment and the vibration work should share the same ground model. And track condition, including rail joints, switches and crossings, produces local peaks that a survey at the wrong chainage will miss.

Dose based thresholds and probability of complaint

BS 6472-1 assesses human response using the vibration dose value, which accumulates weighted acceleration over a period. It gives ranges rather than pass or fail limits, expressed as the probability of adverse comment in residential buildings, with separate 16-hour daytime and 8-hour night-time values.

Consult the current standard for the full table and its qualifications. The point for a planning submission is that a dose value sitting in the "possible" band is a matter of judgement rather than a breach, and the report should say what it recommends and why rather than leaving the officer to interpret a number.

For reradiated noise there is no equivalent British Standard limit, and criteria are set case by case. Figures around 35 dB LASmax in bedrooms are commonly adopted on rail schemes, and the appropriate value should be agreed with the environmental health officer at scoping rather than asserted in the final report.

Prediction where measurement is impossible

Sometimes you cannot measure the relevant thing: the building does not exist, the line is not yet operational, or access to the site is refused during the survey window.

Prediction then proceeds in stages. Establish the source term from measurements elsewhere on comparable track and rolling stock. Apply a propagation model from source to the foundation position using the site ground model. Apply a coupling loss for the transition from ground into the building, which depends on foundation type and building mass. Then apply floor response factors for the specific structure, including amplification at floor resonances.

Each step carries uncertainty, and the honest approach is to state the assumed value and the range at each stage rather than presenting a single predicted figure. Where the prediction lands close to a threshold, propose a condition requiring measurement and detailed design once access is available, with a mitigation scheme to be approved.

Mitigation at source, path and receiver

The options are unevenly available.

At source, resilient rail fastenings, under-sleeper pads, ballast mats and floating slab track all reduce transmission substantially. None of them is in a developer's gift, because the track belongs to the infrastructure operator, and changes to it are a matter for that operator rather than for a planning condition.

At the path, the news is worse. Trenches, barriers and rows of piles have limited effect on the low frequencies that carry rail vibration, because the wavelengths are long. Distance works, which is why the planning answer is frequently to move the building rather than to engineer it.

At the receiver, the measures that work are structural. Isolated floating floors within the building for localised protection. Box-in-box construction for critical rooms such as recording studios or theatres. And full base isolation, where the whole superstructure sits on elastomeric or spring bearings above the substructure.

Base isolation and its cost consequence

Base isolation is effective and it is not a late addition. Every service entering the building has to cross the isolation plane with a flexible connection. Lift shafts, stairs, drainage, ductwork and structural movement joints all need detailing around it. The bearings need to be accessible for inspection and replacement over the building's life, which affects the substructure layout.

That makes it a concept design decision. Retrofitting the idea at technical design stage means redrawing the structure and the services, and on a scheme where the vibration survey was commissioned after planning permission it is the change that most often makes a site unviable.

So the sequence matters more here than almost anywhere. Survey, predict, decide on isolation, then design. Where a scheme is marginal, an isolation allowance in the cost plan with a decision point after detailed measurement is a reasonable commercial position, and one worth explaining to the authority rather than concealing.

What to submit and what to condition

A proportionate submission for a railway-side residential scheme contains:

1. A measurement report with instrument positions, durations, the train log and the raw data summary.
2. Calculated vibration dose values at representative positions, day and night, against the BS 6472-1 response categories.
3. A predicted internal reradiated noise level for the worst affected rooms, with the criterion used and its basis.
4. An outline mitigation strategy proportionate to the results, with the design implications stated.
5. Where relevant, evidence of engagement with the infrastructure operator about works near the railway boundary.

Then accept conditions that are checkable. A condition requiring a detailed vibration and reradiated noise design to be approved in writing before development begins works well. So does a post-completion verification requirement: measurement in a sample of completed dwellings, submitted within a stated period after first occupation, with remedial works if the criteria are not met.

And be realistic about the residual position. Some sites beside heavily used freight routes cannot be made to reach a low probability of adverse comment at reasonable cost, and saying so, with the mitigation you propose and the residual effect quantified, is better than a report that quietly rounds a number down. Specialists who can carry out this work and review the outputs can be found through the professional directory, and the airborne side is covered in our note on road traffic noise assessment for a new housing site.

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.