Commission Implementing Decision (EU) 2026/1208 of 9 June 2026 laying down rules for the application of Directive (EU) 2024/2881 of the European Parliament and of the Council as regards technical details for modelling applications and determining the spatial representativeness of sampling points

• CELEX: Implementing Decision (EU) 2026/1208
• Articles: 13

Article 1 Definitions

For the purposes of this Decision, the following definitions apply:

(1)

‘area of interest’ means a part of the territory of a Member State targeted for a specific assessment of air pollutant concentrations by a modelling application;

(2)

‘concentration interval’ means the range of the concentration levels of a given pollutant taking into account the maximum applicable tolerance;

(3)

‘central value’ means the modelled or measured concentration of a given pollutant around which the concentration interval is calculated.

Article 2 Criteria for the use of modelling applications

When using modelling applications for air quality assessment and for determining the spatial representativeness area of sampling points, Member States shall ensure that such a modelling application fulfils the following criteria:

(a)

the modelling application is able to reproduce results with the same averaging periods as the limit or target values set out in Annex I to Directive (EU) 2024/2881(‘limit or target values’). Percentile relations between annual means and short-term values may be used for the assessment of the hourly, daily eight-hour, and daily limit or target values provided their robustness is verified by measurements at representative locations;

(b)

the spatial resolution of the modelling application is such that the expected variability of pollutant concentrations in the atmosphere over the area of interest can be reproduced to the extent possible;

(c)

if the modelling application requires the use of input emissions data, such data:

(i)

are, where possible, spatially gridded or defined at a resolution that is in line with, or finer than, the spatial resolution of the modelling application;

(ii)

are, where possible, vertically distributed in accordance with the release height of the source;

(iii)

are, where possible, aligned with the temporal resolution of the modelling application;

(iv)

are, where possible, aligned with the requirements of the modelling application in terms of chemical speciation;

(v)

account for the relevant sources in the area of interest;

(d)

if the modelling application requires the use of meteorological data, such data is obtained either from observations or via a meteorological model which, where possible, has a similar spatial and temporal resolution to the modelling application and which ensures the representativeness of the modelled period;

(e)

if the modelling application requires the use of background concentrations, those are, to the extent possible, aligned in terms of temporal resolution, spatial resolution, spatial extent, and detail in the chemical speciation with the modelling application requirements over the area of interest;

(f)

the modelling application is, to the extent possible, capable of capturing the following elements specific to the area of interest:

(i)

site-specific dispersion characteristics;

(ii)

orographic boundary conditions;

(iii)

meteorological conditions;

(iv)

adverse climatic conditions;

(v)

transboundary contributions;

(g)

the modelling application is quality assured and validated in accordance with the criteria set out in Article 3.

Article 3 Quality assurance and validation of modelling applications

1.   Member States shall ensure modelling quality and validate modelling applications in accordance with the criteria laid down in this Article.

2.   Where possible, all input datasets shall be quality controlled and assured before being used as input for a modelling application. Member States shall perform those quality checks before executing each modelling application.

3.   Where possible, measurement data used to perform the validation of modelling applications shall have variability in terms of sampling-point locations to cover the relevant environments for a given pollutant over the area of interest.

4.   The modelling quality objective shall be verified at a sufficient number of sampling points in accordance with the requirements set out in Point A of Annex V to Directive (EU) 2024/2881.

5.   When evaluating modelling applications that incorporate the integrated use of models and measurements, only measurement data not used in the modelling application results and complying with the data quality objectives set out in Annex V to Directive (EU) 2024/2881 shall be used for the evaluation of whether the modelling quality objective is fulfilled. In such a case, Member States shall, where possible, evaluate modelling applications that incorporate the integrated use of models and measurements using the methodology set out in Annex II or any other methodology providing equivalent results.

6.   Where modelling applications are used for the determination of spatial representativeness areas of sampling points, such modelling applications shall fulfil the requirements set out in Point A of Annex V to Directive (EU) 2024/2881.

Article 4 Methods for assessing the spatial representativeness areas of sampling points

1.   In all zones classified as above the assessment thresholds pursuant to Article 7 of Directive (EU) 2024/2881, Member States shall use either measurements in accordance with Article 6 of this Decision or modelling applications in accordance with Article 7 of this Decision or a combination of the two, to assess the spatial representativeness areas of sampling points in those zones.

2.   In all zones classified as below the assessment thresholds pursuant to Article 7 of Directive (EU) 2024/2881, where sampling points are used for the assessment of air quality, Member States may assess the spatial representativeness areas of those sampling points. Where such an assessment is carried out, it may be underpinned by measurements, modelling applications, expert analysis, or a combination of these methods.

Article 5 Criteria for assessing the spatial representativeness areas of sampling points

1.   Member States shall assess the spatial representativeness area of sampling points in accordance with the provisions laid down in this Article.

2.   Member States shall identify locations or areas with similar concentrations to those at a sampling point for a given pollutant using tolerance levels in accordance with Article 6(3) for measurements and tolerance levels in accordance with Article 7(2) for modelling applications.

For the purpose of the first subparagraph, the central value shall be selected as follows:

(a)

where measurements are used, the annual mean of the concentration measured at the sampling point is used as the central value to identify locations with similar concentrations;

(b)

where modelling applications are used, the annual mean of the concentration modelled for the sampling point location is used as the central value to identify areas with similar concentrations.

Member States may use other metrics than the annual mean, such as relevant percentiles or seasonal averages, including hourly, eight-hour, daily, AOT40, and winter averages for pollutants with limit or target values established for metrics other than the annual mean.

3.   Member States shall refine the areas identified pursuant to paragraph 2 in accordance with the rules laid down in this paragraph.

The geographical area may include non-contiguous domains but shall, in principle, be limited in its extension by the borders of the zone under consideration. For sampling points intended to be representative of areas broader than a given air quality zone, Member States may apply alternative geographical limitations.

Locations at or near heavily trafficked roads in suburban or rural areas may be excluded from the areas identified for urban background sampling points even if concentrations in those areas fall within the tolerance interval calculated for the sampling point.

If the area identified for a sampling point covers a wider area than the city or urban area in which the sampling point is located, the area identified may further be limited to the city or the urban area.

Locations where air quality assessment is not required pursuant to point A.2 of Annex IV to Directive (EU) 2024/2881 may be excluded from the identified area.

4.   Member States shall consider source-related information to exclude locations from the areas identified and refined in accordance with paragraphs 2 and 3 that have significantly different pollution profiles or dispersion conditions. Member States shall exercise expert analysis to identify such locations.

Article 6 Use of measurements to assess the spatial representativeness areas of sampling points

1.   Both fixed and indicative measurements may be used to assess the spatial representativeness areas of sampling points, as relevant.

2.   When using indicative measurements to assess the spatial representativeness area of a sampling point, Member States shall ensure that the measurements are evenly distributed over the calendar year, and over the April-to-September period for ozone, and that certain periods of the year are not over- or under-represented.

When using only indicative measurements to assess the spatial representativeness area of a sampling point, Member States shall ensure that the number and spatial distribution of indicative measurements is such that the expected variability of pollutant concentrations can be reproduced to the extent possible.

The uncertainty of the indicative measurement for a specific pollutant may be the uncertainty established for modelling applications in point A of Annex V to Directive (EU) 2024/2881 for the pollutant in question.

3.   Member States shall identify locations with similar concentrations to the central value, as set out in Article 5(2), point (a), for a given pollutant in accordance with the methodology set out in this paragraph.

A tolerance level of 15 % of the central value or a minimum tolerance level as set out in Annex I, whichever is greater, shall be applied above and below the central value to calculate a concentration interval.

Values may be rounded to two significant digits when calculating the concentration interval referred to in the second subparagraph.

The measured concentration of the pollutant in question at a given location shall be compared to the concentration interval calculated in accordance with the second subparagraph.

When the measured concentration at a given location falls within the concentration interval calculated in accordance with the second subparagraph, that location shall be included in the area identified in accordance with Article 5(2).

4.   Member States may use modelling applications in accordance with Article 7 to also include areas in the vicinity of the locations identified in accordance with paragraph 3 of this Article in the spatial representativeness area of a sampling point.

Expert analysis, including but not limited to the assessment of location and source-related information in accordance with Article 5(3) and (4), may be applied to include also areas in the vicinity of the locations identified in accordance with paragraph 3 of this Article in the spatial representativeness area of a sampling point. Expert analysis may also be underpinned by measurement results from non-reference or non-equivalent measurement methods.

Article 7 Use of modelling applications to assess the spatial representativeness areas of sampling points

1.   When using modelling applications to assess the spatial representativeness area of a sampling point, Member States shall prepare a map of the modelled concentrations of a given pollutant in the area of interest where the sampling point is located.

Member States shall ensure that modelling applications used to prepare the map of modelled concentrations fulfil the criteria set out in Articles 2 and 3 of this Decision.

2.   Member States shall identify areas with modelled concentrations similar to the central value, as set out in Article 5(2), point (b), for a given pollutant in accordance with the methodology set out in this paragraph.

A tolerance level of 15 % of the central value or a minimum tolerance level as specified in Annex I, whichever is greater, shall be applied above and below the central value to calculate a concentration interval.

Values may be rounded to two significant digits when calculating the concentration interval referred to in the second subparagraph.

The modelled concentration at all areas shall be compared to the concentration interval calculated in accordance with the second subparagraph.

When the modelled concentration for an area falls within the concentration interval calculated in accordance with the second subparagraph, that area shall be included in the area identified in accordance with Article 5(2).

3.   Member States may use expert analysis, including but not limited to the assessment of location and source-related information in accordance with Article 5(3) and (4) to assess the similarity of an area to a given sampling point.

Article 8 Methodology for preparing a spatial representativeness area map

1.   Without prejudice to Article 4, Member States shall prepare a map that delineates the spatial representativeness areas of all sampling points for a given pollutant within a given zone or alternative geographical area in accordance with the methodology laid down in this Article.

2.   When, following the methodology laid down in Articles 5 to 7, a location or an area is identified for the spatial representativeness area of more than one sampling point, Member States shall either keep the location or the entire overlapping area in the spatial representativeness areas of all sampling points in question, or include it only in the spatial representativeness area of the sampling point which is most similar to the location or the area in question in accordance with paragraph 3 of this Article.

3.   The similarity between sampling points and between a sampling point and a modelled area referred to in paragraph 2 shall be assessed in terms of sampling point type, location characteristics, relevant emission sources, and concentration levels.

4.   Once the spatial representativeness area of all sampling points for a given pollutant is assessed in accordance with the provisions laid down in Articles 4 to 8, Member States shall prepare the map of spatial representativeness areas of the sampling points referred to in paragraph 1 of this Article.

The map of spatial representativeness areas shall be prepared using data from one individual year within the last five years. Member States may use a suitable method, including the use of data from multiple years, to account for inter-annual variability, provided that the reliability of the method is demonstrated.

Member States shall review the spatial representativeness area map, at least following each review performed in accordance with point D.9 of Annex IV to Directive (EU) 2024/2881, and also when there are relevant changes to the monitoring network or relevant substantial changes in the pollution sources or in the meteorology.

Article 9 Use of indicative measurements for air quality assessment

1.   Member States shall use indicative measurement results for air quality assessment and for providing information on the spatial distribution of air pollutants when they show an exceedance of a limit or target value.

2.   Member States may use indicative measurement results for air quality assessment and for providing information on the spatial distribution of air pollutants when there is no exceedance of a limit or target value.

3.   When compliance with the hourly, daily eight-hour, and daily limit or target values is assessed via indicative measurements, Member States may use the corresponding percentile for the calculation of exceedance statistics.

Article 10 Use of modelling results for air quality assessment

1.   Member States shall use modelled concentrations for the purpose of assessing air quality with respect to the limit or target values when both the modelled concentration at a location covered by the spatial representativeness area of a sampling point and the sampling point’s measurement show an exceedance of a limit or target value.

2.   Member States shall not use modelled concentrations for the purpose of assessing air quality with respect to the limit or target values at a given location when they fail to show an exceedance of a limit or target value at this location and at the same time the location is covered by the spatial representativeness area of a sampling point showing an exceedance of the given limit or target value.

When the situation described in the first subparagraph occurs, Member States shall assess whether a review of their modelling application is necessary. That review shall assess whether the modelling application is fit for purpose and shall be completed no later than one year after the modelling application that triggered the review was carried out.

3.   Member States shall use modelled concentrations for the purpose of assessing air quality with respect to the assessment thresholds at a given location when the modelled concentration at this location is above the assessment threshold and below the limit or target value, and the location is not covered by the spatial representativeness area of a fixed sampling point.

In such a case, Member States may choose not to use modelled concentrations for the purpose of assessing air quality with respect to the assessment thresholds if they perform at least one additional measurement, the spatial representativeness area of which covers the modelled location. Those measurements shall be established within two calendar years after the modelling application was carried out if they are fixed measurements, and within one calendar year after the modelling application was carried out if they are indicative measurements. The additional measurements shall cover at least one calendar year in accordance with the minimum data coverage requirements set out in point B of Annex V to Directive (EU) 2024/2881.

4.   In any situation not covered by paragraphs 1 to 3, and without prejudice to Article 8(6) of Directive (EU) 2024/2881, Member States may use modelled concentrations for the purpose of assessing air quality with respect to the limit or target values and with respect to the assessment thresholds.

5.   Member States may use modelling results to provide information on the spatial distribution of air pollutants concentrations, to identify potential hotspots in the zone, and, if applicable, to provide information on the area of exceedance calculated by the modelling application.

Article 11 Entry into force and application

This Decision shall enter into force on the twentieth day following that of its publication in the Official Journal of the European Union .

This Decision shall apply from 30 June 2028.

Schedules & Appendices

ANNEX I Minimum tolerance levels

ANNEX I

Minimum tolerance levels

Air pollutant

Minimum tolerance level

PM 10 , NO 2 , O 3 , SO 2

± 2,0 μg/m 3

PM 2,5

± 1,0 μg/m 3

CO

± 0,025 mg/m 3

B(a)P

± 0,1 ng/m 3

C 6 H 6

± 0,34 μg/m 3

Pb

± 0,05 μg/m 3

As

± 0,6 ng/m 3

Cd

± 0,5 ng/m 3

Ni

± 2,0 ng/m 3

ANNEX II ‘Leaving one out’ cross-validation methodology

ANNEX II

‘Leaving one out’ cross-validation methodology

The ‘leaving one out’ cross-validation strategy is a methodology used for the evaluation of modelling applications that incorporate the integrated use of modelling and measurements. It consists of the following steps:

(a)

select a sampling point to be left out in the data fusion or data assimilation step (i.e. Step (b)).

Data fusion in the context of this decision refers to combining observational datasets and models to obtain the best estimate of a certain variable.

Data assimilation in the context of this decision refers to coupling observations with dynamic numerical models to optimise model states and forecasts;

(b)

derive a fused or assimilated concentration field with the remaining sampling points;

(c)

extract a time series at the location of the sampling point that is left out in the fusion or assimilation process (i.e. the station selected in Step (a));

(d)

calculate a modelling quality indicator value based on the independent measurement values and the fused or assimilated result of Step (c);

(e)

repeat the process from Step (a) with another sampling point until all points have been left out in one iteration of this methodology each;

(f)

evaluate the modelling quality objective with the set of independently derived modelling quality indicators at all sampling point locations.