Wind tunnel testing or computational modelling for a scheme
Both methods report pedestrian wind conditions against the Lawson criteria. Which one an authority will accept depends on height, exposure and what it has asked for in writing.
Choosing between wind tunnel or CFD planning evidence usually comes down to building height, the sensitivity of the surrounding streets and what the authority has asked for in writing. Both methods produce the same output: predicted wind conditions at pedestrian level, expressed against comfort and safety criteria. They get there differently, they cost differently, and authorities do not treat them as equivalent in every case.
The short version. Computational modelling is faster, lower in cost and better at iterating design options. Boundary layer wind tunnel testing is the accepted reference method for tall and complex buildings, and in parts of central London it is effectively expected. Many schemes use both, with modelling during design and a tunnel test to confirm the final form.
What each method produces
A pedestrian wind assessment reports conditions at a grid of receptor points around and within the site: entrances, footways, crossings, amenity terraces, play space and cycle parking. Results are classified against the Lawson comfort criteria, which band conditions by the activity they suit, from sitting through standing and strolling to business walking, with a separate distress or safety threshold.
Wind tunnel testing puts a scale model of the scheme and its surroundings on a turntable in a boundary layer tunnel, with the approaching flow profile matched to the site's terrain. Measurements are taken at each receptor for a set of wind directions, then combined with long-term meteorological data for the nearest representative station.
Computational fluid dynamics solves the flow numerically over a meshed domain. The output looks similar, and the same criteria are applied, but the result depends on the turbulence model, the mesh resolution near the ground and the treatment of the boundary conditions.
| Wind tunnel | Computational modelling | |
|---|---|---|
| Physical basis | Measured flow over a scale model | Numerical solution of the flow equations |
| Typical lead time | Longer, driven by model fabrication | Shorter once the geometry exists |
| Design iteration | Slow and costly per change | Fast, several options in a week |
| Accepted for tall buildings | Widely, and often expected | Varies by authority |
| Sensitivity to operator choices | Lower | Higher, especially turbulence model and mesh |
| Seasonal and directional detail | Strong | Strong, subject to the directions run |
Cost, programme and model preparation
The cost driver in a tunnel test is the model. A scheme and its surroundings have to be built at scale, usually by additive manufacture from the massing model, and the surrounding context extends well beyond the site boundary because upwind buildings shape the flow. Changing the scheme means changing the model.
Computational work has a different cost shape. The geometry preparation is quick if the architect's model is clean, and slow if it is not. Watertight massing, simplified but correct, is what the analyst needs. Sending a fully detailed model with mullions and balustrade spindles makes the job harder.
Programme is where developers get caught. A tunnel test booked late becomes the critical path to submission, and slots at established facilities are not always available at short notice. Book the slot before the design is frozen, not after.
Where computational modelling is accepted
For mid-rise schemes in suburban and low-rise contexts, computational modelling is routinely accepted as the primary evidence. The flow field is less complex, the consequences of error are smaller, and the authority's interest is usually in amenity space and entrances rather than in safety thresholds.
It is also the right tool early. Comparing four massing options at concept stage is a computational exercise, and doing it in a tunnel would be an expensive way to answer a question the architect has not settled yet.
Modelling is stronger than its reputation suggests when the analyst reports honestly: states the turbulence model, shows the mesh near ground level, runs enough wind directions, and presents sensitivity testing. It is weaker than its reputation when a glossy contour plot arrives with none of that.
Where a physical model is expected
Tall buildings, clusters of tall buildings and schemes creating new public space at the base of a tower are the cases where authorities press for tunnel testing. Downdraught from a tall slab and corner acceleration around a sharp edge are precisely the phenomena where numerical results are most sensitive to method.
Some authorities in central London set this out in their own published guidance, and the requirement can be tied to a height threshold or to the presence of a safety exceedance in a preliminary study. Confirm the local position in writing at pre-application, because retrofitting a tunnel test into a determination period is difficult.
Safety is the deciding factor. Where a preliminary assessment shows conditions approaching the distress threshold at any receptor, expect the authority and its reviewer to want measured evidence rather than modelled evidence before granting permission.
Validation and reporting expectations
Whichever method is used, the report should state the meteorological dataset and station, the period of record, the directional resolution, the receptor locations on a plan, and the criteria applied. Missing any of those makes the results unreviewable.
Baseline, existing and cumulative scenarios matter as much as the proposed case. Assessing the proposed scheme alone tells the authority nothing about whether conditions get better or worse, and consented neighbouring schemes have to be included in the cumulative case.
The Wind Microclimate Assessment sits alongside the other environmental evidence for a tall building, and reviewers read it against the Daylight and Sunlight Assessment and the Townscape and Visual Impact Assessment. Consistency between them is checked. A terrace shown as amenity space in one document and excluded from the wind receptors in another is the kind of discrepancy that generates a further information request.
Testing mitigation options iteratively
Mitigation is a design problem before it is a landscape problem. Massing changes, setbacks, podium levels and canopy positions do more than planting, and they cost less to test computationally than in a tunnel.
The usual hierarchy runs: change the form, then add built mitigation such as canopies, screens and colonnades, then add planting, then accept restricted use of the affected space. Trees are the weakest option because their effect is seasonal and depends on maturity, and reviewers know it.
Where a tunnel test is required, run the computational iterations first and take a settled scheme with one or two mitigation variants into the tunnel. That sequence keeps the model count down.
Choosing the method at the right stage
Make the decision at three points. At concept, model computationally to compare massing. Keep the receptor set stable between stages so the results can be compared as the design moves. At planning application stage, match the method to the authority's stated requirement and the building's height and exposure. Before determination, be ready to test mitigation if the officer's reviewer identifies an exceedance.
And put the question to the authority early. An email at pre-application asking which method it expects, and which criteria it applies, costs nothing and removes the worst outcome, which is a completed assessment the authority declines to accept.
Enter your postcode on PlanScope to see which environmental reports a tall building proposal 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. PlanScope provides AI-assisted draft reports for planning applications. Reports should be reviewed before submission. We do not guarantee planning approval. For specialist technical reports, qualified professional input may be required.
Frequently asked questions
Is CFD acceptable instead of a wind tunnel test?
For mid-rise schemes in less exposed contexts, computational modelling is routinely accepted as the primary evidence. For tall buildings, clusters and new public space at the base of a tower, many authorities expect boundary layer wind tunnel testing. Ask the authority in writing at pre-application stage rather than assuming, because the answer varies by borough.
What are the Lawson criteria?
A set of pedestrian wind comfort bands that classify conditions by the activity they suit, running from sitting through standing and strolling to business walking, with a separate distress or safety threshold. Results at each receptor are compared against the criterion appropriate to how that space is intended to be used.
How long does a wind tunnel test take?
The lead time is dominated by model fabrication and by availability of a tunnel slot, so it is measured in weeks rather than days. Design changes mean model changes. Book the slot before the massing is frozen, because a late test can become the critical path to submission.
What should a wind assessment report contain?
The meteorological dataset and station, the period of record, the wind directions modelled, a plan showing receptor locations, the criteria applied, and results for existing, proposed and cumulative scenarios including consented neighbouring schemes. Without those, a reviewer cannot check the work and will usually ask for more information.
Does planting count as wind mitigation?
It helps, but it is the weakest option. Tree effects are seasonal, depend on species and maturity, and take years to arrive. Reviewers give far more weight to changes in massing, setbacks, podium levels, canopies and colonnades, which work from day one and can be secured by condition.
Sources and further reading
This article is general guidance about the planning system in England and, where stated, the rest of the UK. It is not legal or professional advice. Check your local planning authority’s validation requirements and take professional advice for your specific site.