ADMS-Urban is fit-for-purpose

Under Directive (EU) 2024/2881 Member States can choose appropriate model(s) to use for modelling applications, provided they are fit-for-purpose. The EU Air quality modelling for air quality policy document does not formulate any legal provisions, rather it outlines current good practices in the European modelling community and the ambition level for air quality modelling in Europe for applications in relation to air quality policy. Here, we review this information in the context of ADMS-Urban in order to assess this model’s suitability for use in modelling applications as part of the latest Directive.

The tables below summarise the compatibility of ADMS-Urban model features and modelling approaches with the revised Directive (EU) 2024/2881 based on the EU Air quality modelling for air quality policy document. The tables contain, for each model capability item, the associated category, which is either ‘recommendation’ or ‘good practice’. The associated reference to the EU Air quality modelling for air quality policy document is given.

The information in the tables demonstrates that ADMS-Urban, alongside associated tools such as CERC’s Model Evaluation Toolkit, are in line with the majority of recommendations and good practice guidance.

Compatibility of ADMS-Urban model features with European good-practice standards/recommendations

Category	Type	Model capability	ADMS-Urban compatibility	Reference1
Fitness-for-purpose	Recommendation	The model is specifically designed to meet the requirements of particular planning application	YES	1.5, 5.4.1
Appropriate spatial and temporal resolution	Recommendation	The model has the appropriate spatial and temporal resolution	YES	1.4.1, 1.4.2, 5.4.1
	Recommendation	The model is capable of reproducing long-term and short-term indicators	YES	1.4.1, 3.3
	Recommendation	The model is capable of addressing hotspots, such as street canyons, and broader exceedances	YES	5.4.1
Quality of emissions inputs and scenario design	Recommendation	The model can be provided with emission data that are spatially gridded, vertically distributed, and temporally resolved	YES	1.4.3
	Recommendation	The model can utilise detailed source data and emissions factors for mitigation impacts	YES	1.4.3, 5.2.4
	Recommendation	The model can produce a baseline (business-as-usual) scenario	YES	5.2.4
Meteorological data considerations	Recommendation	The model utilises meteorological data covering the modelled period and consider inter-annual variability	YES	1.4.4, 5.3.3
	Recommendation	The model utilises WMO-standard data	YES with data pre-processing	1.4.4
	Good practice	The model is capable of assessing representativeness if using single-year meteorological data	YES	5.3.3
Boundary conditions and background contributions	Recommendation	The model can make use of boundary conditions that reflect future trends, integrating regional policies	YES with data pre-processing	1.4.5, 5.3.2
	Recommendation	The model can use harmonised EU-wide data (e.g. CAMS, EMEP) for background trends	YES with data pre-processing	5.3.2
	Recommendation	The model utilises robust chains to integrate projections and avoid double-counting	YES	1.4.7, 5.3.2
Source apportionment integration	Recommendation	Source apportionment (SA) should assess the impact of mitigation measures	YES	4.1, 5.2.2
	Recommendation	For planning, methods providing "potential impacts" (concentration changes from emission changes) are suitable. Linearity of reactive compounds must be checked.	YES	4.2.1, 4.3.1

¹ References to the EU Air quality modelling for air quality policy document

Compatibility of the ADMS-Urban modelling approach with European good-practice standards/recommendations

Category	Type	Model capability	ADMS-Urban compatibility	Reference1
Quality of emissions inputs and scenario design	Good practice	The model can produce Maximum Feasible Reduction2 scenarios to identify highest achievable reductions	YES with emissions data processing	5.2.4
	Good practice	The model can use Emission Reductions for Attainment3 scenarios for compliance projections	YES with emissions data processing	5.2.4
	Recommendation	If needed, sensitivity tests for best/worst-case scenarios, related to uncertainties in efficiency of emission abatement technology, implementation effectiveness, adoption, exemptions, and enforcements of measures can be performed	YES with additional data processing	5.2.4, 5.4.2
Model validation and bias adjustment for future scenarios	Recommendation	The model must be validated and fulfil the Modelling Quality Objective (MQO) for the reference situation	YES with data post-processing	1.5, 3.3, 3.4, 5.4.2
	Recommendation	For future scenarios, the model can be adjusted for bias observed in the reference case, applying methods (e.g. absolute or relative differences) based on understanding the bias's cause	YES with data post-processing	1.4.6, 5.3.5, 5.4.2
	Recommendation	If data fusion or assimilation is used, the model can use independent observations for validation	YES with data post-processing	1.4.6, 3.4.4
Dynamic sensitivity and evaluation	Recommendation	Evaluate the model’s sensitivity to emissions changes, especially for non-linear chemistry	YES with emissions data processing	5.4.3
	Good practice	Reproduce historical trends to validate model and emissions	YES	5.4.3
	Good practice	Use tools like DELTA for dynamic sensitivity evaluation	YES	3.4.5, 5.4.3
Comprehensive spatial coverage	Recommendation	Projections should cover the entire zone, using Exceedance Situation Indicators	YES	5.3.1
Accounting for natural sources and winter-sanding/salting	Recommendation	The model is capable of subtracting the contributions from natural sources and winter-sanding/salting from exceedances if appropriately quantified and reasonable measures to lower concentrations are demonstrated	YES with data pre- and post-processing	3.7, 5.3.4
Quality Assurance/Quality Control (QA/QC) and documentation	Recommendation	A rigorous QA/QC process is essential, involving thorough checks of all input data, model implementation, and output	YES	1.5, 5.4.2
	Recommendation	The modelling system and its validation must be well documented	YES	1.5
	Recommendation	Information on methodology, assumptions, projection data, and, where possible, uncertainty margins and sensitivity scenarios must be described in air quality plans and roadmaps	YES with analysis reporting	5.1.2, 5.4.2

¹ References to the EU Air quality modelling for air quality policy document

² Defined in the Air quality modelling for air quality policy as "An emission scenario that reflects what is achievable within a specific time period and includes all known technically feasible measures, irrespective of their cost, as well as structural or behavioural changes, that are realistically achievable"

³ The Air quality modelling for air quality policy describes this scenario as a simpler case where exceedances are due to just one major and local source, where it can be useful to first estimate the emission reductions required to achieve compliance. From here, measures must be found that can attain this emission reduction. Once the required emission reductions are determined, an air quality calculation still needs to be made to confirm whether compliance is attainable.