Air quality modelling assessments can provide a robust and reliable indication of pollutant concentrations around a proposed development when they are prepared using appropriate data, recognised methodologies and professional judgement.
They should not, however, be viewed as perfectly precise predictions of the concentration that will occur at every location in the future.
Like any environmental model, an air quality model represents a complex real-world system using available information and assumptions. Traffic flows may change, weather varies from year to year and future vehicle emissions cannot be known with absolute certainty.
For planning purposes, the important question is therefore not whether a model can predict every future concentration exactly. It is whether the assessment is sufficiently robust to determine the likely air quality conditions, the scale of any effect associated with the proposed development and whether those effects could be significant.
UK guidance recognises this distinction. The established approach includes model verification, transparent reporting of inputs and assumptions, and consideration of uncertainty when interpreting results.
How does air quality modelling work for a development project?
Air quality modelling uses computer software to estimate how pollutants from particular sources disperse through the surrounding environment.
For many development-related assessments, road traffic is an important consideration.
The model may use information including traffic flows, vehicle composition, road locations, emission factors, background pollutant concentrations and meteorological data.
The consultant then defines representative receptor locations.
These might include existing homes close to roads affected by development traffic or proposed residential accommodation positioned close to an existing pollution source.
The assessment can then compare different scenarios.
Planning guidance produced by Environmental Protection UK and the Institute of Air Quality Management describes a typical assessment as considering the existing baseline, the future situation without the proposed development and the future situation with the development in place.
This comparison is particularly important.
For planning purposes, the aim is often not simply to predict the future concentration at a property. The consultant may also need to establish how much difference the proposed development itself is expected to make.
Is air quality modelling an exact prediction?
No, and it should not be presented as one.
An air quality model is a technical representation of real environmental processes.
It cannot reproduce every detail of the real world.
For example, the actual weather during a future year cannot be known in advance. Future traffic volumes may differ from forecasts. Buildings, vegetation and local street geometry can influence pollutant dispersion in ways that may be difficult to represent perfectly.
Vehicle emission calculations also rely on established datasets and assumptions.
These limitations do not make modelling unsuitable for planning.
Instead, they mean that the modelling process needs to be carried out carefully and the results interpreted with an understanding of uncertainty.
A competent air quality report should make the main methodology and assumptions clear rather than presenting modelled numbers as if they were exact measurements of the future.
What affects the accuracy of an air quality model?
The quality of the information entered into the model is fundamental.
A sophisticated model cannot compensate for poor or inappropriate input data.
Several factors can influence the reliability of an assessment, including the quality of traffic information, the suitability of background pollution data, the representation of relevant roads and sources, and the meteorological data used.
The location of receptors also matters.
If the assessment is intended to understand exposure at residential properties, receptors need to represent locations where relevant exposure could actually occur.
The modelling methodology should also be proportionate to the site.
A relatively simple setting may not require the same degree of complexity as a constrained urban location where buildings strongly influence the way pollutants disperse.
An experienced consultant therefore needs to understand the site as well as the modelling software.
How are modelled results checked against real measurements?
One of the most important quality-control processes is model verification.
For road traffic modelling, verification typically involves comparing modelled concentrations with suitable monitored concentrations in the study area.
If the model systematically underpredicts or overpredicts the measured road-related contribution, an appropriate adjustment can be considered.
EPUK and IAQM planning guidance states that model verification would normally be expected for modelling of road traffic emissions. It describes verification as a comparison between predicted and measured concentrations that can allow adjustment for systematic errors.
This matters because discrepancies can arise from several sources.
Traffic data may contain uncertainty. Emission factors may not perfectly represent the local vehicle fleet. Estimated background concentrations may differ slightly from actual local conditions.
Local features can also affect how pollution disperses.
Verification therefore provides an important reality check.
Rather than assuming the model is correct because it has produced a numerical answer, the consultant checks how well its outputs correspond with suitable measured information.
Does verification mean monitoring must be carried out on every development site?
No.
This is an important distinction.
Using monitoring information to verify an air quality model does not necessarily mean installing new monitoring equipment on the proposed development site.
Existing local authority monitoring data and other suitable datasets may provide sufficient information.
For much of Hawkins Environmental’s planning-related air quality work, new on-site monitoring is not required.
Where suitable representative data already exist, they can often be used alongside modelling and other established information.
Project-specific monitoring may occasionally be necessary where appropriate data are unavailable or a particular site requires further evidence.
Where meaningful air quality monitoring is genuinely needed, it is not normally a short exercise. Hawkins Environmental’s client guidance indicates that a representative monitoring period would generally require at least three to six months rather than only a few weeks.
This is one reason developers should identify air quality requirements early.
What is meant by uncertainty in an air quality assessment?
Uncertainty simply means recognising that inputs and predictions are not infinitely precise.
This is normal in environmental assessment.
Traffic forecasts, emissions data, background concentrations and meteorological conditions all contain some degree of variability.
The purpose of good modelling practice is to manage that uncertainty rather than pretend it does not exist.
One way of doing this is through verification.
Another is through sensitivity or professional judgement where results are close to an important threshold.
The interpretation of the modelling is therefore just as important as the calculations themselves.
For example, a modelled concentration comfortably below a relevant objective may lead to a different level of concern than one predicted very close to it.
The latter result may require more careful consideration of assumptions and uncertainty before conclusions are drawn.
How close do modelled and monitored results need to be?
There is not a useful universal rule that every predicted result must precisely match every monitoring result.
The consultant needs to examine model performance in the context of the assessment.
Defra’s Local Air Quality Management technical guidance provides methodologies for assessing and verifying air quality information, and its current LAQM.TG22 guidance was updated in May 2025.
Large infrastructure assessments provide examples of how this is applied in practice.
For example, a published model verification exercise for a major development examined modelled results against monitoring and considered statistical measures as well as whether results were reasonably conservative. In many locations, modelled values were reported as being within approximately 25% of measured concentrations, consistent with the verification approach being used for that assessment.
This should not be interpreted as a simple rule that every planning assessment must fall within exactly the same percentage.
The appropriate verification process depends on the project, available monitoring locations and modelled pollutants.
The broader principle is that model performance should be examined rather than assumed.
Can air quality models account for future conditions?
Yes, and this is one of the main reasons modelling is useful for development planning.
Monitoring can tell consultants what concentrations were measured during a particular period.
Modelling allows them to assess scenarios that do not yet exist.
For example, an assessment can consider an anticipated opening year and compare predicted conditions with and without the development.
Forecast background concentrations and changes in vehicle emissions can be incorporated using recognised datasets and methodologies.
However, future conditions naturally involve additional uncertainty.
No consultant can know exactly how many vehicles will use every road several years into the future or what the precise weather conditions will be.
The value of modelling lies in applying consistent, evidence-based assumptions so that likely conditions can be compared on a reasonable basis.
Why does traffic data matter so much?
Traffic-related air quality models depend heavily on understanding how vehicles use the surrounding road network.
Relevant information may include the number of vehicles, proportion of heavy-duty vehicles and average speeds.
If the proposed development is expected to alter traffic patterns, the modelling needs to represent those changes appropriately.
Inaccurate traffic inputs can therefore affect the predicted pollution concentrations.
This is why air quality consultants commonly work with traffic information supplied by transport specialists or other reliable sources associated with the planning project.
Consistency between the transport assessment and air quality assessment is important.
If the two reports use materially different assumptions about development traffic, the planning submission may become difficult to interpret.
How important are background pollution concentrations?
Road traffic is not the only contributor to the pollutant concentration predicted at a receptor.
There is already pollution present in the wider environment from numerous sources.
This is generally represented through background concentrations.
Defra provides mapped air quality information that can support detailed modelling studies, and its Local Air Quality Management resources are regularly updated.
Selecting suitable background information is important because the final predicted concentration may combine the modelled local contribution with the wider background level.
Consultants therefore need to ensure that the data are appropriate for the assessment location and year being considered.
Can buildings affect modelling accuracy?
Yes.
Urban form can influence how pollutants disperse.
On an open road, emissions may disperse relatively freely.
A road closely enclosed by buildings can behave differently because structures influence airflow and may reduce dispersion.
This is one reason site context matters.
The consultant needs to consider whether the selected modelling approach adequately represents relevant local conditions.
It would be inappropriate to assume that every road environment behaves in exactly the same way.
For more constrained or complicated sites, the assessment may require additional consideration of local geometry or other factors.
Are conservative assumptions used in air quality assessments?
They can be.
Environmental assessments often use assumptions intended to avoid understating potential impacts where genuine uncertainty exists.
However, “conservative” should not mean deliberately producing unrealistic worst-case results.
The aim is to create a robust assessment that is appropriately cautious while remaining representative of the development.
This balance is important.
An assessment that systematically understates concentrations could fail to identify a genuine planning issue.
An unrealistically pessimistic assessment could suggest a problem that is unlikely to occur.
Professional judgement is therefore required when selecting and interpreting assumptions.
What happens if predicted concentrations are close to an air quality objective?
Results close to an important objective usually warrant particularly careful consideration.
The consultant may review the modelling inputs, verification process and uncertainties before reaching a conclusion.
The significance of any change caused by the development must also be considered.
A predicted concentration should not normally be interpreted in isolation from the wider assessment.
EPUK and IAQM planning guidance emphasises the need for assessments to explain the basis for determining the significance of air quality effects, as well as the methodology, input information and assumptions used.
This transparency allows planning officers and environmental health professionals to understand how the conclusions have been reached.
Is monitoring more accurate than modelling?
Monitoring and modelling perform different functions, so one should not simply be described as “more accurate” than the other.
Monitoring measures conditions at a particular location during a particular period.
That makes it extremely useful for understanding actual concentrations.
However, monitoring cannot directly measure what pollution will be after a development that has not yet been built.
It also cannot by itself separate the future contribution of a proposed development from other sources.
Modelling can evaluate those future scenarios.
For development planning, the strongest approach is therefore often to use appropriate existing measured information alongside modelling rather than viewing the two methods as competitors.
This combination helps ground the model in real-world evidence while retaining the ability to assess future conditions.
How can developers improve the reliability of an assessment?
Developers can help by providing consultants with reliable and up-to-date project information.
Where traffic data are required, the air quality and transport work should use consistent assumptions.
The consultant should also be informed if the project changes substantially.
A revised access arrangement, increase in development size or change in intended use could affect the assessment.
Early appointment can also be helpful.
If modelling reveals that additional information or monitoring is required, there is more time to obtain it without creating avoidable pressure on the planning programme.
Developers can find more information about Hawkins Environmental’s relevant planning work through its environmental services.
Frequently Asked Questions
Can an air quality model predict future pollution exactly?
No. Air quality models estimate future conditions using established scientific methods and available data. They are designed to provide a robust basis for assessment rather than an exact prediction of every future concentration.
Why are air quality models verified?
Verification compares modelling results with suitable measured concentrations. This helps identify systematic differences between the model and real-world data and, where appropriate, allows adjustments to improve the robustness of the assessment.
Is new on-site air quality monitoring always required before modelling?
No. Many development-related assessments can use existing monitoring information, recognised background datasets and modelling without requiring new on-site monitoring. Project-specific monitoring is only needed where it is justified by the circumstances of the assessment.
Air quality modelling is reliable when the process is robust
Air quality modelling assessments can provide reliable evidence for development planning, but their accuracy depends on much more than the software being used.
The quality of traffic information, background pollution data, meteorological inputs, receptor selection and representation of local conditions all influence the assessment.
Verification against suitable measured information is particularly important for road traffic modelling because it allows model performance to be checked rather than simply assumed.
Uncertainty will always remain when predicting future environmental conditions. That is not a flaw unique to air quality modelling.
The important point is that uncertainty is understood, managed and taken into account when results are interpreted.
A well-prepared air quality assessment should therefore be transparent about its methodology and assumptions, apply recognised guidance and provide conclusions that are proportionate to the quality and limitations of the available evidence.
For developers, the result is not a claim to know future pollution concentrations with perfect precision. It is a robust technical basis for determining whether air quality is likely to represent a material planning issue and whether the proposed development could have a significant effect.
Hawkins Environmental provides air quality assessments and reports for development and planning projects, with modelling used where appropriate rather than assuming that every project requires on-site testing. If you need advice on the appropriate approach for a proposed development, contact Hawkins Environmental to discuss the site and likely assessment requirements.
Phone: 01256 522332
Email: enquiry@hawkinsenvironmental.co.uk
Find out more: hawkinsenvironmental.co.uk
