Better judgment or more elaborate subjectivity.
IAQM’s revised guidance is more mature and more demanding. Its central challenge is to make professional judgment transparent without allowing indicative numbers and matrices to acquire false precision.
IAQM’s Guidance on the assessment of odour for planning v2.0, published on 30 September 2026, is the first full revision of the document since 2014. It is unquestionably more mature than v1.1. It is also more demanding.
Its central message is that odour assessment cannot usually be reduced to a model contour, a sniff survey or a source-pathway-receptor matrix. A defensible conclusion must draw the evidence together and explain why each component deserves the weight assigned to it. That is sound advice. The question is whether the guidance will improve professional judgment or simply give practitioners more tables through which to express it.
The real change is evidential not numerical
The familiar architecture survives. V2.0 retains FIDOL, receptor sensitivity, the distinction between impacts and effects, qualitative risk assessment, dispersion modelling, sniff testing, complaints evidence and odour management plans. The established C98 modelling matrices for most offensive and moderately offensive odours also remain substantially intact.
The more important change is the expectation that these tools will be combined. V2.0 describes a multi-tool approach as best practice where practicable and expressly warns against using favourable modelling to prove the absence of an existing adverse effect when credible complaints, community evidence or sensory observations indicate otherwise.
That is a useful correction to the pseudo-objectivity that can surround odour modelling. A model result expressed to two decimal places is still only as good as its emission rates, operational scenarios and representation of intermittent releases. Conversely, a handful of sniff observations cannot represent an annual odour climate merely because someone stood downwind on three occasions.
V2.0 therefore moves the argument from which tool is right to what can this tool reliably tell us. That is progress. It makes assessment less mechanical, but also places more responsibility on the practitioner and technical reviewer.
The qualitative method has become much less qualitative
The largest visible revision is the expanded source-pathway-receptor framework. The former three-category approach has become a four-by-four structure with:
- source emissions bands below 2,000, 2,000 to 10,000, 10,000 to 50,000 and above 50,000 ouE/s;
- indicative distance bands above 400 m, 200 to 400 m, 20 to 200 m and below 20 m;
- wind-frequency bands below 3%, 3 to 10%, 10 to 20% and above 20%; and
- additional consideration of controls, containment, material handling, occasional activities, release characteristics, buildings, terrain and calm conditions.
This is a considerable improvement on v1.1’s broad examples of small, medium and large sources and ineffective, moderately effective and highly effective pathways.
The benefit is transparency. Two assessors should now find it harder to assign different categories without revealing where their judgments diverge. It also gives reviewers better questions to ask. Which emissions were included? What happens during cleanout, storm conditions or door opening? Is the nominally contained building continuously extracted? Does a prevailing-wind statistic conceal several sustained episodes?
The danger is that indicative bands become de facto thresholds. A receptor at 401 m is not materially different from one at 399 m. An estimated emission rate of 49,000 ouE/s is not inherently more certain than one of 51,000 ouE/s. Yet once a number appears in professional guidance, it tends to migrate into scoping opinions, committee reports and cross-examination as if it represented a scientifically validated boundary.
IAQM repeatedly says that no single factor should determine the category and that the tables require weighting and justification. That caveat needs to survive actual practice. The new method should be treated as a structured reasoning framework, not a calculator.
Sniff testing is more credible but less easy to price
The sniff-testing revision is technically important. V2.0:
- changes intensity level 3 from the confusing distinct but barely recognisable to identifiable;
- typically uses 30 observations over three to five minutes;
- derives percentage odour time and mean intensity from observations at intensity 3 or above;
- strengthens requirements for survey strategy, meteorological records, operational information and quality assurance;
- says assessors are expected to have their odour acuity quantified; and
- introduces separate effects matrices for highly and moderately offensive odours.
At the same time, it removes v1.1’s nominal absolute minimum of three survey days covering Pasquill stability categories representing 70% of annual conditions. That old requirement appeared objective but was difficult to defend scientifically. Three visits cannot meaningfully establish annual representativeness, and Pasquill-category coverage says little about source cycles or whether the relevant wind sectors occurred.
V2.0 substitutes project-specific justification. Scientifically, that is preferable. Commercially, it creates uncertainty. The consultant must now determine how many visits are necessary to support the conclusion, often before knowing whether the source will operate normally, whether appropriate weather will occur or whether land access will be available.
A robust proposal should therefore distinguish between:
- an initial survey allowance;
- weather-dependent attendance and aborted visits;
- additional surveys triggered by unrepresentative conditions;
- operational liaison and source verification;
- assessor-acuity testing; and
- additional work needed to demonstrate absence rather than merely confirm presence.
The guidance rightly observes that one well-timed observation may establish an adverse episode, while substantially more evidence is normally needed to demonstrate its absence. That asymmetry should be explicit in both scopes and technical reviews.
Planning and permitting are moving apart
V2.0 is also a quiet assertion of planning’s independent role. Current Environment Agency permitting guidance does not generally require dispersion modelling or routine off-site sniff testing. IAQM nevertheless continues to support both as potentially valuable planning tools, particularly where a proposed source does not yet exist or housing is encroaching on an existing operation.
This distinction matters. A permit regulates an operator and the measures it must take. Planning determines whether the relationship between land uses is acceptable. A permit is not a certificate that neighbouring housing will experience no odour, nor does the prospect of future permitting remove the need to understand likely amenity effects.
The guidance could perhaps have gone further here. The consultation exposed real concern about the divergence between regulatory and planning practice, but the working group’s essential answer was that the document addresses planning, not permitting. That is correct, but it leaves applicants and decision-makers managing two evidential systems that may look inconsistent. Reports will need to explain the distinction rather than cite one regime as authority against the other.
Professional judgment now requires an audit trail
V2.0 repeatedly invokes professional judgment. That should not be understood as discretion to choose the most convenient answer. Good judgment is visible and reproducible even when the conclusion itself is not mathematical.
An adequate assessment should now show:
- what evidence was available and what was missing;
- why each tool was selected;
- what spatial, temporal and operational conditions it represents;
- how uncertainty might change the conclusion;
- how conflicting evidence was reconciled; and
- why the resulting effect and significance judgments are proportionate.
For technical reviewers, the key warning signs will be assessments that:
- use the qualitative matrix as the sole evidence for a material or contentious proposal;
- treat the new numerical bands as fixed thresholds;
- consider prevailing wind direction but not calm conditions or sustained episodes;
- omit abnormal, intermittent or seasonal releases;
- rely on absence of complaints without considering the propensity and ability to complain;
- use modelling to dismiss credible observational evidence;
- rely on an odour management plan without demonstrating that the underlying controls are technically achievable; or
- treat permit compliance as proof of planning acceptability.
The commercial consequence is more honest scoping
V2.0 will tend to increase the apparent cost of odour assessments because it makes previously hidden tasks harder to ignore. Baseline investigation, operational liaison, complaints analysis, representative sniff testing, dispersion modelling, sensitivity analysis and reconciliation of results are separate pieces of work.
That does not mean every assessment requires all of them. The guidance expressly permits a single, potentially qualitative tool for a demonstrably low-risk case. But a proposal offering a fixed, minimal assessment for a high-risk encroachment site should now attract more suspicion.
The better response is not automatically to include every tool. It is to price a staged assessment:
- initial evidence review and qualitative screening;
- methodology agreement;
- proportionate observations;
- escalation to sampling or modelling if screening identifies material uncertainty; and
- defined allowances for additional visits, operator data deficiencies and sensitivity work.
This is likely to produce more defensible fees and fewer disputes over whether extra work was foreseeable.
A stronger guidance document with one unresolved tension
Overall, v2.0 is a substantive improvement. It makes odour assessment more evidence-led, improves the qualitative framework and gives sniff testing a more coherent structure. Its emphasis on agent-of-change cases and weight of evidence is particularly timely.
Its unresolved tension is between consistency and false precision. The guidance wants practitioners to exercise judgment, yet supplies increasingly detailed numerical categories and matrices. Those devices can expose judgment, but they can also conceal it behind a coloured cell.
The success of v2.0 will therefore depend less on whether everyone uses its tables and more on whether practitioners and reviewers preserve the reasoning around them. The guidance’s most important instruction is not a distance, emission rate or model benchmark. It is that an odour assessment must explain why its evidence deserves to be believed.
Sources
