CFD improves the accuracy of a wind assessment by simulating airflow in full three-dimensional detail, across any number of wind directions, at any scale — from a single building to an entire city district. Unlike a physical wind tunnel, CFD is not limited by model size or test time, which means results reflect real-world complexity more completely. The sections below walk through exactly how that works and what it means for your project.
What makes CFD more accurate than traditional wind tunnel testing?
CFD produces more detailed spatial data than a wind tunnel because it calculates airflow at every point in a three-dimensional computational domain, not just at discrete measurement locations. A wind tunnel gives you readings where you place sensors; CFD gives you the full picture across the entire site, at any height, in any direction.
That said, accuracy is not a binary choice between CFD and wind tunnels. Both methods are reliable when applied correctly. The difference is in what each method does well. Wind tunnel testing excels at capturing complex pressure distributions on facade surfaces, which is why it remains the preferred approach for detailed wind loading studies on individual buildings. CFD, on the other hand, handles large and complex urban geometries far more efficiently. A physical wind tunnel model of an entire city district is impractical; a CFD model of one is not.
For pedestrian-level wind engineering studies, CFD also allows you to test multiple design variants quickly. Changing a building’s setback, adding a canopy, or rotating the tower orientation can each be re-run computationally without rebuilding a physical model. That flexibility directly supports better design decisions.
How does CFD simulate real-world wind conditions around buildings?
CFD simulates real-world wind conditions by solving the fundamental equations of fluid motion across a detailed three-dimensional mesh that represents the buildings, terrain, and surrounding environment. The simulation calculates wind speed, direction, pressure, and turbulence at every cell in that mesh, for each wind direction studied.
The process starts with a 3D model built from geodata — topographic datasets, building footprints, and height information. Existing and permitted buildings are included; small details below roughly one metre are typically omitted because they have a negligible effect on results at pedestrian level. A boundary ring surrounds the core study area, usually several hundred metres wide, to ensure wind enters the model realistically.
The air volume above and between buildings is then divided into a computational mesh. In large-scale studies, this mesh can contain hundreds of millions of cells. Cell sizes are smallest near surfaces where accuracy matters most — down to 0.25 m near walls — and increase toward the model boundary. Near wall surfaces, several prismatic cell layers capture the velocity gradient in the atmospheric boundary layer accurately.
Meteorological inputs drive the simulation. In the Netherlands, wind data from KNMI datasets — either the NPR6097 dataset required by NEN 8100 or the more recent DOWA dataset — define the wind rose. The simulation runs for each wind direction, typically 12 directions, and results are then weighted by wind frequency to produce statistically meaningful comfort and safety classifications.
What inputs does a CFD wind assessment need to produce reliable results?
A reliable CFD wind assessment needs four categories of input: an accurate 3D model of the site and surroundings, representative meteorological wind data, a clearly defined study area and scope, and specification of the applicable assessment standard.
- 3D geometry: Building footprints, heights, and terrain data. Public geodata sources such as 3DBAG and AHN work well for Dutch projects. Planned buildings should be included as permitted volumes.
- Meteorological data: A site-specific wind rose showing wind speed frequency by direction. For NEN 8100 compliance, the NPR6097 dataset is legally required. The DOWA dataset offers higher resolution and can be split by season.
- Study scope: The radius of the core study area, the assessment height (typically 1.75 m for pedestrian comfort), and which locations or zones need to be evaluated.
- Assessment standard: NEN 8100 for projects in the Netherlands; Lawson criteria for international projects, including the UK and Gibraltar.
The quality of the 3D model has the greatest effect on result reliability. Errors or omissions in building geometry — particularly for tall or irregular structures — propagate through the simulation. Starting with clean, complete geometry is not a formality; it is the foundation of a defensible result.
How does CFD accuracy affect compliance with NEN 8100 and Lawson criteria?
CFD accuracy directly determines whether your wind assessment will hold up during permit review. NEN 8100 and the Lawson criteria both require that wind conditions at pedestrian level are assessed against defined thresholds, and the quality of those results depends entirely on how well the simulation represents actual conditions.
Under NEN 8100, the relevant threshold for wind nuisance is the probability that wind speed exceeds 5 m/s at eye height (1.75 m). Results are classified into five classes — A through E — based on exceedance frequency. For wind danger, the threshold is 15 m/s. An exceedance probability above 0.30% is considered unacceptable. These classifications feed directly into permit decisions, so an inaccurate simulation can produce a false pass or flag a problem that does not exist.
The Lawson criteria used in the UK and internationally follow a similar logic but use different activity categories and threshold values. Both frameworks require that the CFD methodology is documented and defensible — turbulence model, mesh resolution, boundary conditions, and meteorological inputs must all be specified and justified.
A CFD assessment built on validated algorithms, appropriate mesh resolution, and the correct meteorological dataset will produce results that permit authorities and planning consultants can scrutinise without issue. Cutting corners on any of these inputs risks producing a report that gets challenged or rejected at exactly the wrong moment in your project timeline. You can find a broader overview of assessment approaches on the Actiflow website.
When should CFD be used instead of — or alongside — a wind tunnel?
CFD is the better choice when the study area is large, when multiple design variants need to be tested, or when a physical model is impractical. Wind tunnel testing is preferable when detailed facade pressure data is needed for structural or cladding design, or when a regulatory body specifically requires physical testing.
For most pedestrian wind comfort assessments in urban environments, CFD is now the standard method. It handles complex geometries efficiently, produces full spatial coverage, and integrates directly with permit workflows. For large-scale area studies — city districts, masterplans, or airport environments — a wind tunnel is physically unsuitable; CFD is the only practical option.
The two methods complement each other in some situations. A CFD assessment can identify which areas or facades are most exposed, and a targeted wind tunnel test can then validate pressure coefficients for those specific zones. This combined approach makes sense when both pedestrian comfort and structural loading are in scope for the same project.
How early in a design process can CFD wind results influence building layout?
CFD can meaningfully inform building layout from the earliest schematic design stage, as soon as basic massing volumes are defined. You do not need a fully detailed design to run a useful simulation — block models with approximate heights and footprints are enough to identify wind risks and test mitigation strategies before any design decisions become fixed.
Early involvement matters because the most effective interventions are at the urban planning level. Adjusting building orientation, varying block heights, introducing setbacks, or repositioning towers relative to each other can all but eliminate pedestrian wind problems — but only if those options are still on the table. Once a design is committed, the remaining tools (canopies, screens, vegetation) are less effective and more expensive to implement.
The hierarchy of mitigation measures reflects this clearly:
- Building massing and orientation — most effective, only available early
- Facade features such as setbacks and recesses — effective, possible during design development
- Aerodynamic shaping — useful but limited in scope
- Public realm measures such as screens and planting — least effective, used when design is fixed
A quick scan CFD study on simple massing volumes takes a fraction of the time of a full assessment, and the findings can directly shape which design direction gets developed further. Discovering a wind hazard after planning approval is far more disruptive than catching it during concept design.
How Actiflow helps with CFD wind assessments
We specialise in CFD simulations for wind assessments across the full range of project types and scales — from a single high-rise to city-wide masterplan studies like the one we carried out for the municipality of Rotterdam. Our team works with both NEN 8100 (for the Netherlands) and Lawson criteria (for international projects including the UK and Gibraltar), so we can advise you on which standard applies and what the assessment needs to include.
Here is what working with us looks like in practice:
- We advise on the right type of study for your project stage — from a quick scan on early massing to a full permit-ready CFD assessment
- We build validated 3D models from geodata and run simulations using internally developed, validated CFD algorithms on our own HPC cluster
- We deliver colour-coded maps and clear visual output that you can present directly to clients, planners, or permit authorities
- We stay close to your project from intake to final report, with one point of contact throughout
- For regular clients, we can start the next day — speed and flexibility are part of how we work
Curious how we can help with your wind assessment? Contact us and we will be happy to discuss your project and help you find the right approach. You can also learn more about us and our background as a spin-off from Delft University of Technology.
