EUROPEAN COMMISSION

Brussels, 26.10.2022

SWD(2022) 545 final

COMMISSION STAFF WORKING DOCUMENT

IMPACT ASSESSMENT REPORT

Accompanying the document

Proposal for a Directive of the European Parliament and of the Council

on ambient air quality and cleaner air for Europe (recast)

{COM(2022) 542 final} - {SEC(2022) 542 final} - {SWD(2022) 345 final} - {SWD(2022) 542 final}

1Table of Contents

1Table of Contents

Annex 1: Procedural information

1.Lead DG, Decide Planning/CWP references

2.Consultation of the Regulatory Scrutiny Board (RSB)

3.Evidence, sources and quality

Annex 2: Stakeholder consultation (Synopsis report)

1.Aim of the consultation

1.1Consultation strategy

1.2Consultation activities

2Summary of consultation activities

2.1Open public consultation

2.2Targeted stakeholder consultation

2.3Stakeholder meetings

3Results of the stakeholder feedback

3.1Feedback by stakeholder groups – Policy Area 1

3.2Feedback by stakeholder groups – Policy Area 2

3.3Feedback by stakeholder groups – Policy Area 3

4Other consultation activities

4.1Ad-hoc contributions

4.2Third EU Clean Air Forum

4.3Inception Impact Assessment

4.4Fit for Future Platform opinion on the ambient air quality legislation

5Use of stakeholder feedback

Annex 3: Who is affected and how?

1.Practical implications of the initiative

2.Summary of costs and benefits

3.Relevant sustainable development goals

Annex 4: Analytical methods

1.Overview

2.Quantitative modelling of air pollutant standards

2.1Introduction of the modelling framework

2.2Application of GAINS

2.3Concentration modelling methodology

2.4Assessment of health impacts

2.5Monetisation of health and non-health impacts

3.Shortcomings of the modelling approach and sources of uncertainty

Annex 5: Baseline, Maximum Technically Feasible Reduction and policy Scenarios – modelling results

1.Description of the baseline scenario

2.Air pollutant emissions: baseline development and scope for MTFR reductions

3.Air pollutant concentration projections

Annex 1: Procedural information

1.Lead DG, Decide Planning/CWP references

The revision of the Ambient Air Quality Directives is led by the Directorate General for Environment. It was included as items PLAN/2020/8962 and PLAN/2020/8636 in the Agenda Planning.

This impact assessment started in December 2020.

An Interservice Group to steer the evaluation was set up in June 2020 with representatives from the Secretariat-General (SG); Legal Service (SJ); Directorates-General for Budget (BUDG); Economic and Financial Affairs (ECFIN); Internal Market, Industry, Entrepreneurship and SMEs (GROW); Defence Industry and Space (DEFIS); Competition (COMP); Employment, Social Affairs and Inclusion (EMPL); Agriculture and Rural Development (AGRI); Mobility and Transports (MOVE); Energy (ENER); Environment (ENV); Climate Action (CLIMA); Research and Innovation (RTD); Joint Research Centre (JRC); Maritime Affairs and Fisheries (MARE); Regional and Urban Policy (REGIO); Structural Reform Support (REFORM); Taxation and Customs Union (TAXUD); Health and Food Safety (SANTE) and Neighbourthood and Enlargement Negotiations (NEAR).

The Interservice Group met eight times during the impact assessment process.

Timeline

2.Consultation of the Regulatory Scrutiny Board (RSB)

An upstream meeting with the RSB took place on 10 June 2021.

After final discussion with the ISSG, a draft of the impact assessment was submitted to the RSB on 20 June 2022 and discussed at a meeting with the RSB on 19 July 2022.

In relation to this impact assessment, the Regulatory Scrutiny Board (RSB) delivered a positive opinion with reservations on 22 July 2022. The following table provides information on how the comments made have been addressed in this Staff Working Document:

3.Evidence, sources and quality

Support study

To support the analysis of different policy options, the European Commission awarded a specific support contract to external consultants on “Study to support the impact assessment for the revision of the EU Ambient Air Quality Directives”. The consortium comprised Trinomics (consortium lead), in collaboration with Ricardo, VITO, IIASA and MET Norway.

Two further support contracts provided input on specific aspects related to the revision of the Ambient Air Quality Directives:

·Support contract on “Strengthening of air quality monitoring, modelling and plans under the Ambient Air Quality Directives”. The consortium comprised Ricardo (consortium lead), NILU, VITO and Trinomics.

·Support contract on “Systematic assessment of monitoring of other air pollutants not covered under Directives 2004/107/EC and 2008/50/EC (with a focus on ultrafine particles, black carbon and ammonia)”. The consortium comprised IIASA (consortium lead), Umweltbundesamt, EMISIA and RIVM.

Consultation strategy

Guided by the consultation strategy, 9 a broad range of stakeholders was consulted for the revision of the Ambient Air Quality Directives, including:

·Public authorities – i.e. EU Member States and their public authorities, at different governance levels (national, regional, local) and other institutions;

·Civil society & NGOs – i.e. non-governmental organisations (NGOs) and civil society representatives;

·Industry & businesses – i.e. private economic sector operators such as business associations, organisations, trade unions, companies;

·Academia & research – i.e. research community, academia, medical professionals, and patient organisations;

·EU citizens – i.e. citizens not directly affiliated with the above stakeholder groups, but with a keen interest in the topic of air pollution.

Consultation activities included an open public consultation, a targeted stakeholder survey, stakeholder meetings, interviews and further outreach, such as through the third EU Clean Air Forum. Stakeholders also provided ad hoc contributions. A detailed overview is presented in Annex 2.

Bespoke modelling

Quantitative modelling has been conducted, focusing in particular on the impacts of different air quality standards, with a state-of-the-art modelling framework including: the Greenhouse gas – Air pollution Interactions and Synergies (GAINS) model and MET Norway’s chemical transport model (EMEP CTM) with the uEMEP downscaling extension for fine resolution. This modelling assesses a number of effects, in particular: air pollutant emissions, concentrations, ecosystem impacts, feasibility to attain particular air quality targets as well as respective measures and their costs. A detailed overview of the modelling framework is included in Annex 4.

Evidence from air quality monitoring and reporting

Under the two Ambient Air Quality Directives, Member States make available the information they use for reporting and reciprocal exchange of information via and air quality data repository ( http://www.eionet.europa.eu/aqportal ), including:

·monitoring and assessment regimes, including assessment methods: http://aidec.apps.eea.europa.eu and http://aided.apps.eea.europa.eu  

·attainment of environmental objectives, including information on exceedance situations: http://aideg.apps.eea.europa.eu  

·air quality plans and programmes, as well as air quality measures: http://aideh.apps.eea.europa.eu and http://aidek.apps.eea.europa.eu  

·information on source apportionment in zones and agglomerations: http://aidei.apps.eea.europa.eu  

·information on air data and aggregated validated assessment data as summarised in the annual air quality reports published by the European Environment Agency

·online EEA indicators, such as:

oExceedance of air quality standards in urban areas: https://www.eea.europa.eu/ims/exceedance-of-air-quality-standards  

Evidence from selected studies and policy documents

·COM(2005)446. ‘Thematic Strategy on air pollution’

·COM(2013)918. ‘A Clean Air Programme for Europe’, including, in particular:

SWD(2013)531. ‘Clean Air Programme for Europe Impact Assessment’

·COM(2017)312. ‘Actions to Streamline Environmental Reporting’

·COM(2018)446. ‘The First Clean Air Outlook’

·COM(2018)330. ‘A Europe that protects: Clean air for all’

·COM(2019)149. ‘Environmental Implementation Review 2019’

·SWD(2019)427. ‘Fitness Check of the Ambient Air Quality Directives’

·COM(2021)3. ‘The Second Clean Air Outlook’

·EEA Annual Air Quality Reports and briefings published from 2011 to 2022, including

o https://www.eea.europa.eu/publications/status-of-air-quality-in-Europe-2022  

o https://www.eea.europa.eu/publications/air-quality-in-europe-2021

o https://www.eea.europa.eu/publications/air-quality-in-europe-2020-report

o https://www.eea.europa.eu/publications/air-quality-in-europe-2019  

o https://www.eea.europa.eu/publications/air-quality-in-europe-2018  

·EEA Briefing 9/2018. ‘Improving Europe’s air quality — measures reported by countries’

·EEA Report 11/2014. ‘Effects of air pollution on European ecosystems’

·EEA Report 6/2018. ‘European Union emission inventory report 1990-2016’

·EEA Report 22/2018. ‘Unequal exposure and unequal impacts’

·EEA Report 24/2018. ‘Europe’s urban air quality’

·EEA Briefing 19/2021. ‘Managing air quality in Europe’

·ETC/ACC Technical paper 2010/1. ‘The state of the air quality in 2008’

·ETC/ACM Technical paper 2011/20. ‘Co-benefits of climate and air pollution regulations’

·European Commission (2013). Flash Eurobarometer 360: ‘Attitudes of Europeans towards air quality’

·European Commission (2017). Special Eurobarometer 468: ‘Attitudes of European citizens towards the environment’

·European Commission (2019). Special Eurobarometer 497: ‘Attitudes of Europeans towards Air Quality’

·European Court of Auditors Special Report 05/2018 on Renewable Energy

·European Court of Auditors Special Report 23/2018 on Air Pollution

·European Parliament (2017). ‘Report on the inquiry into emission measurements in the automotive sector’

·European Parliament (2019). ‘Sampling points for air quality: Representativeness and comparability of measurements in accordance with Directive 2008/50/EC on ambient air quality and cleaner air in Europe’ (study requested by the ENVI Committee)

·EUROSAI (2019). ‘Joint report on air quality by the European Organisation of Supreme Audit Institutions’

·IIASA (2014). ‘Complementary Impact Assessment on interactions between EU air quality policy and climate and energy policy’

·IIASA (2017). ‘Costs, benefits and economic impacts of the EU Clean Air Strategy and their implications on innovation and competitiveness’

·IIASA (2018). ‘Progress towards the achievement of the EU’s air quality and emissions objectives’

·JRC (2013). ‘Assessment on siting criteria, classification and representativeness of air quality monitoring stations’

·JRC (2017). ‘Urban PM2.5 Atlas: Air Quality in European Cities’

·JRC (2017). ‘Global Energy and Climate Outlook 2017: How climate policies improve air quality’

·JRC (2019). ‘Urban NO2 Atlas’

·JRC (2021). ‘Urban PM2.5 Atlas: Air Quality in European Cities’

·Nationale Akademie der Wissenschaften Leopoldina (2019). ‘Saubere Luft. Stickstoffoxide und Feinstaub in der Atemluft: Grundlagen und Empfehlungen’

·OECD (2016). ‘The Economic Consequences of Outdoor Air Pollution’

·OECD (2020). ‘The economic cost of air pollution – Evidence from Europe’

·World Health Organization (2006). ‘Air quality guidelines – global update 2005’

·World Health Organization (2013). ‘Review of evidence on health aspects of air pollution’

·World Health Organization (2021). ‘WHO global air quality guidelines: particulate matter (‎PM2.5 and PM10)‎, ozone, nitrogen dioxide, sulfur dioxide and carbon monoxide.’

Additional sources of evidence, including relevant academic literature and scientific articles, reports and conference papers, online and data sources, as well as further policy documents and guidelines, are listed in the respective Annex 4, and also in the support study informing this impact assessment or cited as footnotes where referred to.

Annex 2: Stakeholder consultation (Synopsis report)

1.Aim of the consultation

This annex summarises the results of the stakeholder consultation activities undertaken as part of the revision of the Ambient Air Quality Directives. The stakeholder consultation aimed to collect supporting information, data and knowledge in order to provide input for the different policy options for the revision of the Ambient Air Quality Directives, with a view to fill any potential information/data gaps and gather views of stakeholders on the different policy options and the feasibility of their implementation. The thorough stakeholder consultation ensures that the view from different stakeholder groups are duly represented and considered in the impact assessment.

1.1Consultation strategy 10

The consultation focused on gathering stakeholders’ responses on the following:

·extent and feasibility of a closer alignment of EU air quality standards with the latest recommendations of the World Health Organization (policy area 1); 

·ways to improve legislative provisions and their coherence, including in relation to penalties, public information and air quality assessments (policy area 2);

·ways of strengthening of air quality monitoring, modelling and air quality plans (policy area 3).

A broad range of stakeholders was consulted for the revision of the Ambient Air Quality Directives, including:

·Public authorities – i.e. EU Member States and their public authorities, at different governance levels (national, regional, local) and other institutions;

·Civil society & NGOs – i.e. non-governmental organisations (NGOs) and civil society representatives;

·Industry & businesses – i.e. private economic sector operators such as business associations, organisations, trade unions, companies;

·Academia & research – i.e. research community, academia, medical professionals, and patient organisations;

·EU citizens – i.e. citizens not directly affiliated with the above stakeholder groups, but with a keen interest in the topic of air pollution.

1.2Consultation activities

As outlined in the consultation strategy, the following activities were applied as complementary activities that formed the core of the stakeholder consultation:

·Open public consultation allowing the interested public and stakeholders to express their views (see section 2.1);

·Targeted stakeholder consultation addressing selected stakeholders in all Member States and at EU level via a targeted survey and interviews (see section 2.2);

·Stakeholder meetings aimed at assisting in the identification and confirmation of the policy measures and at receiving feedback that would support its completion (see section 2.3).

2Summary of consultation activities

2.1Open public consultation 11

The open public consultation ran for 12 weeks, from 23 September 2021 to 16 December 2021. 12 The online questionnaire contained 13 introductory and 31 specific questions and was hosted on the EU Survey tool. The questionnaire aimed to confirm the issues identified for the impact assessment and gather initial views on the ambition level and potential impacts of certain options for the revision of the Ambient Air Quality Directives.

A total of 934 responses were received, and 116 position papers were submitted. In the general part of the questionnaire, respondents could choose whether they wished to respond further to a targeted section. The targeted section received a total of 555 responses. On average, open questions received 124 individual responses, with a minimum of 11 and a maximum of 406 individual responses. 23 Member States were represented in the responses.

The stakeholder types of respondents, their country of origin and other information about their profile was collected for the analysis. The distribution of stakeholder groups and countries is presented below in Box A2.1.

On policy area 1, stakeholder feedback pointed to a desire to opt for a high level of ambition. 72% (n=673) of all respondents expressed a preference to align with WHO Air Quality Guidelines. In particular, a large majority of civil society & NGOs (93%, n=98) and EU citizens (79%, n=485) indicated that EU air quality standards should be fully aligned with the latest WHO recommendations. This view was only shared by a minority of public authorities (36%, n=19), with a majority of public authorities (62%, n=32) here calling for partial alignment. Furthermore, civil society & NGOs, academia & research and EU citizens were largely in favour of meeting the current EU air quality standards as soon possible, while public authorities and industry & business representatives were less in favour of meeting current EU air quality standards as soon as possible. Nevertheless, there was a general agreement across the stakeholder groups that meeting current air quality standards is the most feasible and the most important policy measure. Regarding the extent of applicability of air quality standards, civil society & NGOs and EU citizens thought these should apply everywhere while some public authorities were also of the opinion that these should apply only at selected locations.

On policy area 2, the majority of civil society & NGOs and EU citizens were of the opinion that legislative changes in regards to air quality should include a provision ensuring access to justice for citizens as well as a provision ensuring compensation for health damage caused by air pollution. To expand requirements for action by national / regional / local authorities in case of exceedances as deemed as highly important by 65% (n=607) of respondents. Regarding the availability of information on air quality, public authorities and industry & businesses felt the most informed while civil society & NGOs thought there was a room for improvement. In addition, this measure was thought to be highly feasible to implement by 58 % (n=541) of respondents. In general, regarding the feasibility and importance of new policy measures, there was quite a divide between the stakeholder groups, though there was a general agreement that making it easier for EU air quality standards to update to scientific progress would be an important measure.

On policy area 3, regarding monitoring the majority of civil society & NGOs, academia & research and EU citizen representatives were of the opinion that there is a need for additional monitoring everywhere, while there was a lesser support for this additional monitoring from public authorities and an even lesser support from industry & business representatives. To establish more detailed rules on the location of sampling points was seen to be highly important by 60% (n= 560) of respondents as well as the expansion of monitoring requirements to a broader set of harmful air pollutants (60%, n=560). There was a general agreement across most of the stakeholder groups that the clarity of air quality plans needs to be improved, especially when addressing specific sources and origins of air pollution. Additionally, regarding the clarity of air quality plans, public authorities believed that responsibility across different tiers of governance and stakeholders need to better clarified. Addressing specifically the proposed policy measures, the option considered most important and feasible across all stakeholder group was to establish more detailed rules on the location of sampling points.

2.2Targeted stakeholder consultation 

Targeted survey

The targeted survey was published on EU survey in two parts (i.e. part 1 on policy area 1 on 13 December 2021, and part 2 on policy area 2 and 3 on 13 January 2022), both with a deadline for contributions by 11 February 2022.

The targeted survey was seeking in-depth views by organisations with an interest in, or working with EU rules on air quality. Therefore, the survey was specifically disseminated to targeted stakeholders, including competent authorities at different levels, private sector organisations, academics and civil society organisations. The targeted survey was distributed to a large network of relevant stakeholders to reach a great number of interested stakeholder in all EU Member States and all targeted stakeholder groups.

Part one of the targeted stakeholder survey received in total 139 replies representing 24 Member States. Part two of the survey received 93 replies representing 22 Member States. The number per stakeholder group and Member State for part one (i.e. policy area 1) and for part two (i.e. policy area 2 and 3) is listed in the boxes (A2.2 and A2.3) below.

On policy area 1, the majority of stakeholders saw difficulties to reach the recommended PM2.5 annual and peak concentrations by the WHO Air Quality Guidelines in the foreseeable future, while civil society & NGO representatives were the most optimistic and public authorities and research & academia the least optimistic. A majority was in favour of stricter PM2.5 annual limit values (60%, n=83) and short-term limit values (62%, n=86) that apply in all the territory and not at selected locations only (policy measure O1 and O2). For PM10 a relative majority (44%, n= 61 for annual and 48%, n= 66 for daily) of stakeholders saw the recommended PM10 levels by the WHO Air Quality Guidelines as feasible with significant additional effort, while civil society & NGOs were generally more optimistic. In addition, a relative majority of stakeholders (37%, n=51) were in favour of more stringent PM10 EU standards (especially on a long-term basis) and introducing an average exposure indicator target for short and long-term PM10 concentrations at a regional level (policy measure P1 to P3). 

For NO2, 38% (n=53) of the stakeholders saw the recommended annual levels by the WHO Air Quality Guidelines as feasible but only with significant effort and for the NO2 hourly concentrations which mirror the current EU standard a small relative majority of 31% (n=43) saw it feasible but only with some additional effort (policy measure Q2). A majority of 66% (n=93) of respondents were in favour of more stringent EU standards of the NO2 annual mean in a short-term timeframe (policy measure Q1). Also a relative majority indicated a full alignment with the WHO recommendation for a long-term standard (37%, n=51)). Again public authorities, civil society & NGOs and academia & research were the most ambitious stakeholders, whereas industry & business voted for less ambitious levels. As for PM2.5 and PM10, stakeholders are of the opinion that an average exposure indicator target at a more regional level would be appropriate. However, a low response rate for the question on NO2 average exposure indicates a low level of certainty across all stakeholder groups (policy measure Q3). On ozone (O3) a relative majority (38%, n=53) of stakeholders sees the 2021 WHO recommendations for annual ozone levels as feasible without additional effort. Public authorities and research & academia were the most optimistic stakeholder groups. A relative majority (26%, n=37) replied for being in favour of an ozone limit value, while 21% (n=28) replied for a target value. A relative majority of the stakeholders (31%, n=43) regarded the recommended ozone peak concentrations by the WHO as being feasible. However, for the short to medium term, 41% (n=57) of the respondents were in favour for the ozone short-term concentration that mirrors the current EU target value and only 43% (n=59) agreed to a full alignment with the WHO recommendation for long-term (policy measure R1 to R3).

On policy area 2, 83% (n=10) of the stakeholder group civil society & NGOs were strongly in favour of adjusting EU air quality standards upon publication of new scientific evidence while the other stakeholder groups and in particular industry & businesses but also public authorities were less supportive of this policy measure (policy measure A1). However, the policy measure to adjust EU air quality standards based on technical progress didn’t find full support across all stakeholder groups (policy measure A2). To establish short-term EU air quality standards for additional air pollutants found no majority as respondents of the different stakeholder groups answered with great variety (i.e. civil society & NGOs and research & academia largely in favour of this policy measure (B1) while industry & businesses and public authorities were less ambitious). With a relative majority (32%, n=30) of all replies, all stakeholders were largely in favour of the expansion of the exposure reduction target (B3), with public authorities the most supportive of this policy measure. The regular update of air quality plans was also supported by a relative majority of stakeholders (41%, n=38) while industry & business were the only stakeholder group being less supportive of this policy measure (C5). 

A policy measure that received minor support across most stakeholder groups (besides civil society & NGOs) was the further specification of the obligation to take measures to keep the exceedance period as short as possible with almost half of the respondents (45%, n=18) among public authorities that didn’t support this policy measure (C2). Public authorities with a majority (50%, n=21) of replies among public authorities respondents replied that they are not or largely not in favour of harmonising air quality plans (policy measure D2), while they were more supportive (38%, n=16) of establishing a requirement for Member States to involve specific actors in the air quality plan development (policy measure D1). This policy measure also found support across the other stakeholder groups. Regarding the policy measures about transboundary air pollution, a large relative majority (48%, n=45 for policy measure M1 and 36%, n=34 for policy measure M2) of all replies and across all stakeholder groups are in favour to use an agreed methodology when assessing transboundary air pollution and when it comes to cooperation and joint action on transboundary air pollution (noting that the response rate from industry & business was very low for those interventions).

Policy measures regarding additional enforcement tools (policy measures E1 to E4) in case of non-compliance had a very low response rate (27%, n= 44). The policy measure E2 on a specific provision that guarantees a right to compensation for damage and health found no support among public authorities while civil society & NGOs where largely supportive. Least support in this policy area was the introduction of an “access to justice” provision which was only largely supported by civil society & NGOs (policy measure E4). Policy measures regarding public air quality information was supported across all stakeholder categories with only minor reservations in regards to the regular up-to-date information from industry & business representatives (policy measure F1). A relative majority (40, n=38) of replies was in favour of requiring Member States to use harmonised air quality indices while comments from public authorities requested that this would be preferred as an additional index to the national index (policy measure F4). 

On policy area 3, the mandatory use of modelling as part of air quality assessment did not find large support, and comments made by stakeholders point to the current large uncertainty of harmonised model criteria (policy measure G2). To allow the use of indicative measurements to substitute fixed measurements as part of the air quality assessment was with a relative majority (43%, n=40) of all replies not supported among all stakeholder groups (policy measure G1), whereas the highest support (21%, n=12) was found in the stakeholder group of public authorities. More supported was the policy measure H2 that considers sampling points for PM2.5 and PM10 independently, especially among public authorities, research & academia and civil society & NGOs. The change of minimum number of sampling points was more favoured by civil society & NGOs (42%, n=5) and academia & research (41%, n=9) and by public authorities (14%, n=6) and found no support among industry & businesses (policy measure H1). To specify that sampling points with exceedances of limit values should be maintained (policy measure I1) found large support across the different stakeholder groups (noting that industry & businesses only had one reply). This result is similar to the policy measure that foresees the establishment of a protocol when a sampling point has to be relocated. Public authorities (43%, n=18) and academia & research (50%, n=11) were fully or largely agreeing to this policy measure (I3).

To include a requirement to monitor long-term trends if the fixed sampling point is discontinued with other techniques, such as indicative measurements, found less support, and stakeholders pointed to the uncertainty that occurs when different monitoring techniques are used for one measurement (policy measure I2). For the policy measure to further clarify macro-siting criteria for sampling points, a larger number of respondents (31%, n=29) was fully or more in favour, whereas a relative majority (44%, n=41) did not answer to this question. Also public authorities were more in favour (43%, n=18) of these policy measure, however a large number of respondents (31%, n=13) from public authorities did not reply (policy measure J1). For the policy measure to further clarify micro-siting criteria the same number of respondents from public authorities (n=13) either fully and to a large extend in favour of this policy measure or was to some extend or not at all supporting this policy measure as some indicated that current rules suffice, while NGOs & civil society were more supportive of this measure (policy measure J2). On the data quality requirements for sampling points to be further defined, a relative majority of replies to this policy measure (30%, n=27) were in favour, and academia & research and public authorities’ respondents were the largest supporters (policy measure K1). The mandatory up-to-date information on pollutant concentration was only to some extent or not at all supported across the stakeholder groups (29%, n=27). Various comments from public authorities’ stakeholders pointed out that up-to-date data would be important but it needs to be quality assured to present correct data information (policy measure K2). The introduction of standardized “modelling quality objectives” were supported largely (38%, n=26) with public authorities and academia & research being the greatest supporters of this policy measure (K3). To measure additional emerging air pollutants and to set a minimum number of sampling points for those was supported by a clear relative majority (43%, n=40 and 39%, n=37) from all stakeholders except from industry & business (policy measure L2). Similar support found the policy measure N1 on refining the minimum information to be included in an air quality plan (43%, n=40) across all stakeholders groups.

Interviews

Targeted interviews were conducted to complement the other consultation activities, in particular with representatives of regional and national public authorities, civil society & NGOs and academia & research. The interviews were conducted in April 2022 after the targeted stakeholder survey was closed and evaluated. A list of questions was sent to the stakeholders ahead of the interviews, which were then discussed during the meeting. The interviews focused on remaining gaps for policy area 2, notably on the feasibility, means of implementation and impacts of the various options considered. The main purpose of the interviews was to fill those information gaps identified from the evaluation of the targeted stakeholder survey. See Table A2.2.

2.3Stakeholder meetings 

First stakeholder meeting

The first stakeholder meeting took place on the 23 September 2021 and was attended by a total of 345 participants, either onsite or online according to COVID-19 restrictions at the time.

A total of 315 stakeholders from 27 Member States participated in the meeting, without considering the consultants contributing to the revision of the Ambient Air Quality Directives and EU officials involved. The aim of the first stakeholder meeting was to confirm the shortcomings identified and gather initial views on the ambition level from all stakeholder groups. 13 All relevant stakeholder groups as set out in the consultation strategy were represented during the stakeholder meeting. The stakeholder groups and the country is indicated below in box A2.4.

On policy area 1, the majority of civil society & NGOs argued strongly for full alignment of EU air quality standards with the WHO Air Quality Guidelines levels by 2030, whereas several public authorities commented on the need to consider the measurability and acceptability of future measures needed to attain a closer alignment. Industry & business stakeholders also cautioned that uncertainties related to technical feasibility, local issues, biogenic emissions and measurements remained. Stakeholders also stressed the need to look at additional pollutants, a more regular review of air quality standards, requested a location based limit value, argued for a regional exposure reduction target, and pointed to measurement uncertainties of air pollutants.

On policy area 2, the different stakeholder groups, supported in general the proposed elements to be tackled and possible policy measures presented. The raised topics and discussions focused on adding an explicit mechanism for adjusting air quality standards to technical and scientific progress, on expansion of actions required to address exceedance, specifying provisions to guide the development of air quality plans and on governance, and expanding the provision of information requirements. The importance of access to information was underlined by stakeholders, which is deemed crucial for the protection of public health and also directly connected to other provisions, for example, on access to justice. Stakeholders also agreed that the revision should also ensure that especially vulnerable groups have access to information.

On policy area 3, the use of models to supplement assessment methods was welcomed, though it was noted this should not be at the expense of a reduced monitoring network. The importance of clear meta-data to describe a site to enable cross-city comparison was noted. Civil society & NGOs and public authorities advocated an increase in the number of PM2.5 monitoring stations, and more broadly set clearer requirements for the proportion between different types of monitoring stations, which would entail abandoning the PM2.5 /PM10 ratio. Concerns were expressed by several public authorities (both national and regional level) with regards to stability and sensitivity issues, hence participants noted that single sensors should not be used for compliance purposes and that uncertainties must be communicated transparently if such data is used.

Second stakeholder meeting

The second stakeholder meeting took place on the 4 April 2022 and was attended by a total of 285 participants, either onsite or online according to COVID-19 restrictions at the time.

A total of 257 stakeholders participated in the meeting, without considering the consultants contributing to the revision of the Ambient Air Quality Directives and EU officials involved. Stakeholders from 23 Member States were present at the meeting. The aim of the second stakeholder meeting was to collect feedback from stakeholders that would assist the Commission in its completion of the Impact Assessment. All relevant stakeholder groups as set out in the consultation strategy were represented during the stakeholder meeting. 14 The division of stakeholders per groups and per country is indicated below in Box A2.5.

On policy area 1, public authorities and civil society & NGO representatives expressed their preference for binding air pollutant standards. Several NGOs reiterated their preference for full alignment with the WHO Air Quality Guidelines levels by 2030 while one public authorities also expressed the same view. It was also raised by civil society & NGO representatives that they would be interested to see analysis of the percentiles for daily exceedances. Especially regarding NO2, one NGO stressed the future focus should be on daily limit values. Civil society & NGOs also expressed opinions on the relative effectiveness of target or limit values. It was pointed out that in certain Member States target values do not provide an effective incentive and as such the revised air quality rules should turn to limit values. Stakeholders from all stakeholder groups expressed their views regarding the definition of ‘short’ and ‘long’ term standards. Generally, civil society & NGO representatives were of the opinion that 2030 should already be considered as a ‘long term’ timeframe, while public authorities were more reserved and stated that 2040 might also be an appropriate long-term target. Industry & business pointed out that the transition in their sector would take time, and there is a need for the long-term targets to reflect that and align with sector plans and roadmaps. As such, industry & business expressed their preference for 2050 to be considered as the long-term target.

On policy area 2, representatives of public authorities stressed the importance of transboundary cooperation, which they proposed should be addressed by harmonisation of rules on air quality plans. This suggestion was also supported by certain civil society & NGO representatives. Furthermore on air quality plans it was suggested that their drafting starts with emission factors and should include absolute numbers. Civil society & NGO representatives also touched upon the topic of improvement of public information. They discussed that there is a need to harmonise the available information as well as to provide a link between the information provided and the health impacts of air pollution (e.g. by relying on colour coding of different health impacts). Linking health impacts with air quality will allow citizens to make more informed decisions. Making informed decisions can further be supported by providing citizens with real-life data, namely allowing citizens to limit their exposure levels. The list of pollutants on which real-life data are provided should be expanded and include pollen, for example. Furthermore, the topic of access to justice was also touched upon, namely that both the provisions on access to justice and on public information remain the largest gaps in the current Ambient Air Quality Directives. One attendee from research & academia stressed that penalties have to be more drastic in order to encourage Member States to take action. However, at the same it was highlighted that any legal action will be difficult if one relies on target values rather than limit values.

On policy area 3, public authorities’ representatives highlighted the importance with regards to sampling points of Annex III to Directive 2008/50/EC. One attendee from civil society & NGO pointed out that there can be issues with accuracy of modelling, due to the availability of data. As such, one has to approach modelling with caution. It is thus important to use data that is fit for purpose and regularly updated emission inventories. Civil society & NGOs also highlighted that clarification on the procedures for moving monitoring sites was crucial, as there are contentious cases where public authorities have shifted sites. One participant highlighted that there is a need for good real-time monitoring to be in place, ideally composed of several solutions (e.g. satellite and ground monitoring). Representatives of academia & research highlighted the need for monitoring requirements for ammonia, which are currently not present despite the potential severe impacts on biodiversity and ecosystems. This would also be beneficial for coherence with the NEC Directive. It was also pointed out that there is a need for monitoring both in urban and in rural areas, and for sampling points in residential areas, as wood burning still takes place and is subsidised in some Member States.

3Results of the stakeholder feedback 

3.1Feedback by stakeholder groups – Policy Area 1

Representatives from public authorities largely consider it is “not feasible, for the foreseeable future” that the recommended levels from the WHO Air Quality Guidelines for PM2.5 of 5 µg/m³ annually (incl. 36 out of 53 replies to the survey) or that the daily levels of 15 µg/m³ (incl. 25 out of 53 replies to the survey) are achievable. Further some representatives think that the PM2.5 annual level should be set on 10 µg/m³ (incl. 17 out of 53 replies to the survey) while more representatives favour an annual level of 15 µg/m³ for PM2.5 (incl. 19 out of 53 replies to the survey). 

For the average exposure indicator target, respondents from public authorities have no preferred approach but differ strongly. On PM10 levels, the representatives largely think the WHO recommended levels are not feasible in the foreseeable future (incl. 22 out of 53 replies to the survey) and there representatives are mostly favouring 20 µg/m³ or 30 µg/m³ for a PM10 annual level. The majority within this group is in favour of an average exposure indicator target of PM10 of an “ECO 15 at national level” for both short and long-term. For NO2 levels, public authorities think that 30 µg/m³ is the most feasible option and that the WHO recommended levels are only feasible with significant effort. The large majority is in favour that the NO2 levels and also PM2.5 and PM10 levels should apply in all territory, and is in favour that the type of standard should be a limit value. For ozone levels the representatives think that the most recent WHO recommendation is feasible, without additional effort and the majority is in favour for a level of 100 µg/m³ (incl. 13 out of 53 replies to the survey). For all heavy metals this stakeholder group thinks that the most recent WHO recommendations are feasible without additional effort. For benzo(a)pyrene, stakeholders don’t think that the WHO recommendations are feasible for the foreseeable future.

Representatives from civil society & NGOs largely think that the WHO recommended levels for PM2.5 are feasible with some additional effort (incl. 10 out of 12 replies to the survey). The representatives’ opinion on the levels are divided with around half favouring a PM2.5 level of 5 µg/m³ and the other half favouring a PM2.5 level of 10 µg/m³ (incl. 6 out of 12 replies for to the survey for each value). This stakeholder group thinks that PM2.5 levels should apply in all territory and a large majority is in favour to set a limit value as the type of standard. The representatives have the same opinion concerning the PM2.5 short-term standards; around half of the stakeholder group is in favour of a short-term PM2.5 level of 15 µg/m³ and the other half favours 25 µg/m³ (incl. 6 out of 12 replies for to the survey for each value). The respondents didn’t express many opinions about the favoured approach on the PM2.5 average exposure indicator target.

Concerning PM10 levels, respondents from civil society & NGOs expressed that the WHO recommendations are feasible, with some additional effort (incl. 11 out of 12 replies to the survey) and a majority is in favour of a PM10 annual level of 20 µg/m³. For the PM10 short-term concentrations, the majority is in favour for a level of 50 µg/m³ for short to medium term and for 45 µg/m³ in the long-term. Regarding NO2 annual levels around half of this stakeholder group is in favour of a NO2 level of 10 µg/m³ and the other half is favour of a NO2 level of 20 µg/m³. Similar to PM2.5 and PM10 the representatives think that the levels should apply in all territory and limit values is the preferred type of standard. For ozone levels the representatives believe that the WHO recommended levels are feasible, with some additional effort (incl. 11 out of 12 replies to the survey) and are equally in favour of an ozone level of 60 µg/m³ or 70 µg/m³ (incl. 6 out of 12 replies to the survey for each option). The ambition for heavy metals in regards of concentration levels are mostly quite low i.e. with the majority being in favour of a level for arsenic of 6 µg/m³.

A large majority of representatives from industry & business put forward that air quality standards for PM2.5 annual should be regulated by the EU. However, a majority also thinks that PM2.5 short-term concentrations should not be regulated by EU standards. This stakeholder group finds that the recommended PM2.5 annual levels by the WHO are feasible, but only with significant effort and a large majority (incl. 16 out of 26 replies to the survey) think that the PM2.5 level should be set at 25 µg/m³. Only one representative of this group expressed the opinion that PM2.5 level should be lower than 5 µg/m³ and another one expressed that the level should be set at 5 µg/m³. A majority of this stakeholder group is largely in favour of applying the PM2.5 annual levels at selected stations only (incl. 12 out of 26 replies to the survey), while only a minority is in favour of applying the levels in all territory. For the PM2.5 short-term levels the majority of this group thinks that the recommended levels by the WHO are not feasible, for the foreseeable future and are also largely in favour to not set a standard at all (incl. 14 out of 26 replies to the survey). However, asking the representatives what type of standard should apply only a minority repeated that no standard should be set while a majority expressed that they favour a limit value.

Most respondents from industry & business indicated that an “ECO at a more regional level” for both, short term and long term should be set (incl. 14 (short) and 16 (long) out of 26 replies to the survey). On PM10 levels, the majority in this group thinks that the WHO recommended levels are feasible, but only with significant effort and opinions were expressed that the PM10 annual level should be set at 30 µg/m³ and only applies at selected locations (incl. 12 out of 26 replies to the survey). For PM10 peak concentrations the majority of representatives is in favour of a level of 50 µg/m³, which corresponds to the current EU standard, but find that the standard should apply at selected locations only. Industry & business stakeholders are largely agreeing for PM10 on a “national emission ceiling at more regional level” for the average exposure indicator target for short and for long-term. For NO2 annual levels this stakeholder group largely expressed the view that the WHO recommended levels are not feasible, for the foreseeable future (incl. 17 out of 26 replies to the survey) and some expressed their opinion of being in favour of a NO2 annual level of 30 µg/m³. Some stakeholders expressed their opinion that the NO2 annual level should only apply at selected stations. Stakeholders also expressed that they are against an EU standard for NO2 short-term concentrations (incl. 13 out of 26 replies to the survey). Stakeholders think that a NO2 short-term level of 200 µg/m³ should apply i.e. which means no change to the current EU standard.

The representatives of academia & research largely believe that the recommended WHO levels for PM2.5 are not feasible, for the foreseeable future and are in favour of PM2.5 annual levels of 10 µg/m³ or 15 µg/m³. This stakeholder group is largely in favour of applying the levels in all territory and set a limit value as the type of standard for PM2.5. For the PM2.5 short-term concentrations the stakeholder group largely favours a PM2.5 short-term level of 25 µg/m³ for short-term and 15µg/m³ for long-term (incl. 16 out of 42 replies to the survey). On the average exposure indicator target, the opinions differ in this stakeholder group and also many didn’t express their opinion at all (incl. 24 out of 42 representatives didn’t reply to the survey on this matter). For PM10 annual values for short to medium term, a large share of representatives is in favour of 20 µg/m³, while many expressed of being in favour of 30 µg/m³ for PM10 levels. For long-term levels (with a view on year 2050), a large majority is in favour of 15 µg/m³. For long-term PM10 short-term concentrations this stakeholder group is the most ambitions with a large majority for a PM10 level of 45 µg/m³ and to a lesser extend for less than 45 µg/m³. The recommended NO2 levels recommended by the WHO are seen as feasible by the majority of this group (21 out of 42 replies to the survey), but only with significant effort. The majority thinks that a NO2 annual level of 30 µg/m³ for short-term and 10 µg/m³ for long-term should be put forward. Representatives of this stakeholder group think that NO2 levels should apply in all territory and the type of standard should be a limit value. For ozone the opinions of short term ozone levels differ strongly, while for the long-term levels a clear majority is in favour of ozone levels of 60 µg/m³. For heavy metals, this stakeholder group is in general more ambitious than other stakeholder groups in regards of air pollutant levels.

Representatives of EU citizens thought the most important option was to ensure achievement of the existing EU air quality standards. Regarding feasibility, EU citizens thought the most feasible option was to establish legally enforceable limit values for all air pollutants, while the least feasible option was to mandate that all air quality standards are met, either in general or everywhere. Representatives made some remarks, respectively on the need to act to protect human and environmental health, and the need to try to minimise economic impacts. Several EU citizens made comments linked to revising the Ambient Air Quality Directives, namely: the need to clearly assign responsibilities, setting targets for additional pollutants (indoor air quality, pollen), prioritising locations where people spend most of their time, strengthening monitoring in residential areas and strengthening enforcement to ensure compliance with standards. Finally, some stakeholders advocated for the ban of wood burning in residential areas.

3.2Feedback by stakeholder groups – Policy Area 2

Representatives of public authorities expressed largely positive feedback and a majority was in favour on policy measures regarding the periodically update of a list containing air pollutant of emerging concern, the establishment of additional short-term EU standards i.e. for PM2.5, the expansion of exposure reduction targets, an agreed methodology for transboundary air pollution and the obligation for Member States to provide specific health information to the public. Across all intervention areas 16 different levels of ambition and opinions were expressed by this stakeholder group, besides on topics that touch upon intervention area E, where only a minority from public authorities expressed their opinions (incl. 29 out of 43 no replies to the survey). In addition, the feedback that was received on those topics was mostly negative. Other policy measures that found only little consent among the public authorities were the introduction of a mechanism to adjust air quality standards based on technical progress, the obligation to introduce short-term action plans for each pollutant and the requirement for Member States to harmonise air quality plans.

Representatives from civil society & NGOs were more in favour of policy measures in area M and E and lesser to measures in area A and B. Representatives in particular expressed stronger opinions for topics related to intervention area A, B and F and to a lesser extent on topics that touch upon the other intervention areas. The large majority of this stakeholder group favoured the introduction of a mechanism to adjust EU air quality standards based on scientific advice. In comparison, no representative expressed to be largely in favour to adjust the EU air quality standards based on technical progress. Other policy measures that found high support among the stakeholder representatives touch on the following interventions: establishing short–term EU air quality for example for PM2.5, establish additional limit values for additional air pollutants, the introduction of an short-term action plan for each pollutant, the regular update of air quality plans, both policy measures regarding transboundary air pollution, the introduction of an explicit “access to justice” provision and the requirement to ensure more regular up-to-date data reporting. The policy measures that are included in intervention area A found in general the least support among the civil society & NGO representatives, i.e. the provision for Member States to adopt more stringent standards (2 out of 12 replies to the survey being in favour).

Representatives from industry & businesses favoured to a large extend the policy measure that introduces a mechanism to adjust EU air quality standards based on scientific advice (incl. 8 out of 14 replies to the survey). This stakeholder group expressed largely opinions in relation to topics that touch upon intervention area A and B, while other policy measures didn’t receive many (different) opinions. Only three policy measures (B1, B2 and D1) found some support expressed by the representatives from industry & businesses while other policy measures didn’t found strong support during the consultation period. This stakeholder group in particular not in favour of the establishment of short-term air quality standards such as for PM2.5. In addition, A1, A4 and B4 were the interventions where a large majority of representatives expressed not being in favour of those policy measures.

Representatives from academia & research expressed in general higher support for policy measures that touch upon intervention areas C and M and didn’t express strong support on policy measures that are included in intervention area E. Across all policy measures that were discussed, this stakeholder group expressed strong opinions for all policy measures except for those in area E (incl. 18 out of 22 no replies to the survey). A large majority of respondents are in favour to the periodically update of list for emerging air pollutants to ensure monitoring of those (13 out of 22 replies to the survey). Also policy measures regarding transboundary air pollution found among this stakeholder group large support (incl. 13 for M1 and 14 for M2 out of 22 replies to the survey). Defining alert thresholds and information thresholds for all air pollutants to alert the public was the policy measure that found least support in this stakeholder group.

The representatives of EU citizens thought all policy measures related to policy area 2 that were presented were almost equally important. They expressed a similar view regarding feasibility, citizens thought that all the measures presented were almost equally feasible (i.e. adjust EU air quality standards to the evolving technical and scientific progress, further define the different types of air quality standards and the actions their exceedances would trigger’ etc.). Respondents from EU citizens pointed out that better information (on air pollution in certain areas, on effects of air pollution and on what citizens can do to reduce pollution in their cities) is needed. Other measures EU citizens focused on: legally binding EU standards; extending the scope of air quality standards and monitoring to cover other pollutants harmful to health (e.g. mercury, black carbon, ultrafine particles and ammonia, and indoor air pollution) and restricting the right of corporations and individuals to make profits by conducting activities that curtail the right of current and future generations to a healthy, sustainable and naturally biodiverse environment. With regards to measures supporting implementation, this stakeholder group also supported improvements to the current provision on penalties.

3.3Feedback by stakeholder groups – Policy Area 3

Representatives of public authorities were largely in favour of the policy measure that foresees the introduction of standardised modelling quality objectives (incl. 19 out of 43 replies to the survey). Other policy measures that found strong support among the public authorities were the policy measure that requires monitoring long-term trend via indicative measurements or modelling, the measure on establishing a protocol if a sampling point needs to be relocated and the requirement to measure continuously certain emerging air pollutants. The policy measure that refines the minimum information for air quality plans has received as well largely positive feedback and was well discussed among this stakeholder group. The policy measure that found least support by this stakeholder group was, the simplification of the definitions of monitoring stations and/or sampling points with a majority not favouring this measure and a change of the minimum number of sampling points per air quality zone.

Representatives of civil society & NGOs were largely in favour of the policy measure that foresees the expansion of list of required and/or recommended volatile organic compounds (incl. 7 out of 12 replies to the survey) and the requirement of a regular review of the assessment regime following clear criteria (incl. 6 out of 12 replies to the survey). In general, this stakeholder group was more supportive across all policy measures and the majority of policy measures received positive feedback and that support. However, policy measures related to policy area 3 were the least discussed or raised by this stakeholder group. Similarly, policy measures in relation to intervention area K were hardly discussed by representatives. The policy measure that allows to use of indicative measurement to substitute fixed monitoring as part of the assessment was the measure that found the least support among the stakeholder group (incl. 7 out of 12 replies not favouring this measure in the survey).

Representatives from industry & businesses didn’t strongly discuss or support policy measures that are included in policy area 3. In the targeted survey, no policy measures was replied to with “fully” agree by this stakeholder group and in general this policy area had a low response rate in particular for intervention area G, H, I and J with up to 13 “no reply” or “no opinion” from out of 14 total replies. Representatives of this group did express being in favour to a large extent of the policy measure defining further data requirements for sampling points used for air quality data assessment. The least supported policy measure was a mandatory provision for up-to-date information on the pollutant concentration for certain air pollutants for a minimum number of sampling points per air quality zone (incl. 6 out of 14 replies to the survey).

Representatives from academia & research were largely in favour of the policy measure regarding the requirement of monitoring stations that measure continuously certain emerging air pollutants at “supersites” (incl. 16 out of 22 replies to the survey) followed by the introduction of standardised modelling quality objectives as a control mechanism (incl. 14 out of 22 replies to the survey). The policy measures that were least supported by representatives of this stakeholder group are: the policy measure which allows the use of indicative measurements to substitute fixed monitoring in some specified cases and the policy measure on simplifying the definitions type of monitoring stations and/or sampling point locations (incl. both 13 out of 22 replies to the survey). The stakeholder group was in general strongly engaging in this policy area and raised and discussed many policy measures.

Representatives of EU citizens think that the policy measures under this policy area are all almost equally important and regarding feasibility, EU citizens think that all measures are roughly equally feasible. A general support for improved monitoring and specifically in relation to ultrafine particles and hydrogen sulphide (in areas with industrial pollution) was expressed. Additionally, the harmonising of monitored data was also strongly supported. This stakeholder group also pointed out that for air quality plans a more stringent framework should be put forward setting out clear requirements and timelines in order to maximise their effectiveness.

4Other consultation activities 

4.1Ad-hoc contributions

In total 30 ad-hoc contributions (i.e. position papers, scientific studies and other documents) from 25 different stakeholders 17 were received throughout the duration of the revision period. Ad-hoc contributions were evaluated and analysed which policy area and policy option the ad-hoc contribution was targeting and took the information into account for the Impact Assessment and legislative proposal. The following table A2.3 lists the organisation names and further details from the received ad-hoc contributions.

4.2Third EU Clean Air Forum

The Third EU Clean Air Forum took place on 18 and 19 November 2022 in Madrid with the possibility to actively engage also via a smartphone application or watch the event online per web-stream. 18 Around 200 participants were present in the venue in Madrid and more than 500 participants attended the event online across the EU and other non-EU countries.

Stakeholder groups present at the event were mainly public authorities, environmental and non-governmental organisations, business associations and organisations, research and academia institutions and citizens. During the week of the event, the hashtag “#CleanAirEU” reached close to 27 million accounts on the social networking applications twitter.com and instagram.com globally.

High-level interventions and panel discussions with a wide range of stakeholders groups reflected on air quality issues and solutions, expressing further scope to improve the current legislation. The event focused in two sessions in particular on the “Revision of the Ambient Air Quality Directives” and “Access to justice and the right to clean air”.

4.3Inception Impact Assessment

The inception impact assessment was published on 17 December 2020 with a feedback period until 14 January 2021. Stakeholders were invited to provide feedback on the proposed inception impact assessment as outlined in the roadmap that was made public on the EU Have-Your-Say-Portal. 19 A total of 63 stakeholders from 12 Member States provided feedback on the inception impact assessment as indicated in box A2.6.

On policy area 1, the expressed ambition in the replies were predominantly for a high ambition level, calling for closer or full alignment of EU standards with the WHO recommendations.

On policy area 2, addressing the enforcement and governance shortcomings, the ambition levels expressed in writing were outbalanced. Stakeholders had strong opinions of ambitions varying from the opinion that the Ambient Air Quality Directives do not need to be revised at all to the opinion that compensations for citizens who have to live surrounded by high air pollution needs to be granted.

On policy area 3, respondents addressed to a lesser extent issues in regards to this policy area. Topics that were raised by respondents with asking for high ambition was the need for more precise criteria for air quality monitoring and to consider a cooperation across different government levels when implementing air quality measures.

4.4Fit for Future Platform opinion on the ambient air quality legislation

The Fit for Future Platform is a high-level expert group that helps the European Commission in its efforts to simplify EU laws and to reduce related unnecessary costs, so as to deliver maximum benefits to citizens and businesses, in particular small and medium-sized enterprises. On 12 November 2021 the platform adopted its opinion to the “Ambient air quality legislation”  20 , which included the following suggestions (references in brackets refer to where these suggestions have been addressed in this impact assessment): 

·Review air quality standards to reflect latest scientific evidence and supplement limit values with regional exposure reduction targets (addressed in problem area I);

·Ensure coherence of action between different levels of governance to improvement the effectiveness of air quality measures and the implementation of the Ambient Air Quality Directives (addressed in problem area II);

·Improve monitoring networks to diminish discrepancies and enhance comparability across Member States; improve design of air quality plans and promote local/regional level action (addressed in problem area III);

·Monitoring of pollutants not currently covered by the Ambient Air Quality Directives such as Ultrafine Particles (PM0.1), black carbon and other components of PM, metals, and ammonia (addressed in problem area II);

·Simplify the legislative framework by bringing together directives 2008/50/EC and
2004/107/EC in a single directive (addressed as part of section 8.3 on administrative costs and REFIT);

·Ensure coherence with EU legislation, including urban and road transport, renewable energy and agricultural policies (addressed throughout this impact assessment, including modelling efforts and Annex 8);

·Address emission sources such as tyre and brake wear, non-exhaust traffic related particles, heavy goods vehicle refrigeration units, heating and power emissions, agriculture and wood burning (not directly addressed in this impact assessment, as they are covered under relevant EU legislation.) 21

5Use of stakeholder feedback

All of the stakeholder feedback as outlined under the sections here above was part of an extensive data collection process. The different consultation streams highlighted in this annex, as well as the modelling of scenarios for evidence gathering were combined in order to provide input for the impact assessment. The consultation activities aimed at informing the Ambient Air Quality Directives revision process, either by collecting evidence or by gathering the views of a broad array of stakeholders. The information gathered during the open public consultation (section 2.1) contributed to building the problem definition, and to designing potential (regulatory and non-regulatory) measures, including by seeking to understand the importance and feasibility of several potential measures according to different stakeholder groups. The targeted stakeholder survey (section 2.2) built on the results of the open public consultation and asked more specialised questions on the design, feasibility and potential impacts of different measures, which contributed to the assessment of these measures. The inputs gathered during the stakeholder meetings (section 2.3) also informed the revision process, by giving participants the opportunity to comment on the presentations given on the preliminary results of the project. Lastly the interviews were undertaken to fill in the knowledge gaps identified after the analysis of preceded consultation activities (section 2.2).

The data was examined to underpin the assessment of impacts of different policy options and the feasibility of their implementation. Data was analysed to identify contradictory or supportive statements and evidence to reach to conclusions for each of the stakeholder groups individually. In this context, all widely supported views are entirely considered in the final report, with less widely supported views identified as such.

Annex 3: Who is affected and how?

1.Practical implications of the initiative

This annex sets out the practical implications of the preferred policy package for the various types of stakeholders concerned. It describes the possible implications for public authorities or businesses of complying with the air quality standards and other measures set out in the revised legislation and indicates the likely costs to be incurred in meeting those, or, where quantitative information is not available, the nature and magnitude of such costs. It also presents the implications for the citizens.

Public authorities / Administrations

Increasing the stringency of standards can be expected to lead to an increase in the number of sites and zones in exceedance in the short term. As such, competent authorities will be required to develop and implement new or revise existing air quality plans in order to put in place a strategy to meet new standards. These plans will also require ongoing review and management. Hence, increasing standards will overall imply an increase in the competent authorities’ administrative burden. The degree to which this would affect each Member State would vary, provided that some would be closer to meeting new revised standards, while other would be further away from them. For those standards which could drive a large number of new exceedances with even a small change (e.g. PM2.5, PM10, NO2 and ozone), administrative costs are likely to be high. Where there is broad compliance with existing and proposed standards (e.g. SO2, CO, benzene, etc.), it could be assumed that administrative costs would at most be low.

Other sources of potentially high costs include the build-up of air quality modelling capacity where this is not developed yet as well as the installation of new monitoring stations, especially those for ensuring the additional monitoring of pollutants of emerging concern. The need to address poor air quality in hot spots requires action at local level in particular, some of which might be of non-technical nature and which would in any case differ considerably across municipalities and are therefore challenging to estimate.

Preferred policy options addressing governance and enforcement shortcomings will entail costs in relation to changing the way the Ambient Air Quality Directives are implemented (rather than them resulting from administrative burden of specific policy options suggested). Increasing the stringency of the existing policy framework will significantly increase the costs for those administrations currently in breach of the provisions of the Directives. Conversely, administrations currently compliant with the Ambient Air Quality Directives will have very limited additional costs other than those related to transition to the new regime.

Overall, total administrative costs are estimated to range from 75 to 106 million Euro per year in 2030, with costs in the preferred scenario estimated at 78 million Euro. These are costs that fall on public authorities. Some of the adjustment costs (see next sub-section) may fall on public authorities (such as through procuring materials and infrastructure, building ownership, changing vehicle fleets), but these have not been estimated separately here.

Businesses and the economy at large

Businesses and more generally employers will benefit from the reduction of negative health and (though less significant) non-health impacts associated with poor air quality. The improvement of air quality expected to follow from adopting the preferred policy package will have positive knock-on effects on the productivity of the EU workforce, both through reduced mortality and reduced morbidity (the latter causing absence through illness, including of dependent children, or lower productivity at work). The analysis has demonstrated that with a 10 μg/m3 headline limit value for PM2.5 as part of the preferred package, monetised benefits from reduced costs of health impact are estimated to be 40 or 119 billion EUR (2015 prices) in 2030, depending on the valuation approach chosen. 22 In either way, these represent a close to 30% decrease in costs compared to the baseline in 2030.

Material and ecosystem impacts are typically much smaller than health impacts. Benefits from reduced material damage are projected to amount to almost 200 million EUR in 2030 in the 10 μg/m3 scenario compared to the baseline; benefits from reduced crop damage to 254 million EUR, benefits from reduced forest damage to 287 million EUR, benefits from reduced ecosystem impacts between 706 (low estimate) and 2 117 (high estimate) million EUR (all 2015 prices).

At the same time, stricter air quality standards require investments such as installation of abatement measures that come at a cost. The costs increase with the stringency of the new standard. For the 10 µg/m3 standard, mitigation (or adjustment) costs beyond the baseline amount to around 5.6 billion EUR. Industry bears most of the costs, followed by agriculture. These two together bear above two thirds of the total costs. There are no direct administrative costs falling on businesses.

Taking these two sides of the equation together, the macroeconomic modelling undertaken shows that the market benefits of improved air quality outweigh the costs of abatement measures and other investments needed to meet stricter EU air quality standards. The key insight is that all scenarios, including the preferred one, improve aggregate economic outcomes in the EU compared to a situation of unchanged policy, when productivity gains of clean air are accounted for (the positive impact on the gross domestic product, GDP, and private consumption increases with the stringency of the scenario). With the exception of livestock-based agriculture, which sees a small percentage reduction, all sectors raise output compared to the baseline. Results further indicate enhanced competitiveness of the EU economy as indicated by an improved trade balanced and higher exports, again with productivity gains from clean air factored in.

Citizens and consumers

Citizens will enjoy health benefits from improved air quality. The benefits increase, as expected, with the ambition of the scenario. For PM2.5, premature death in the EU-27 caused by the exposure to air pollution at levels above the WHO guidelines reduces by around 50% under the 10 μg scenario compared to the baseline in 2030. In the same scenario, the additional reduction for NO2 compared to the baseline is 16%. Citizens will further enjoy benefits from reduced morbidity. A 10 μg headline limit value is projected to reduce the number of yearly cases of a range of health outcomes caused by the exposure to air pollution at levels above the WHO guidelines by around 50% in 2030 compared to the baseline. 23  

Citizens residing in hot spots areas are particularly vulnerable as a result of high exposure to air pollution and can thus be expected to benefit most from stricter air quality standards. Also citizens with existing medical conditions and citizens in sensitive groups may be at higher risk due to exposure and will therefore have more to gain on average from improved air quality. The analysis undertaken shows little distributional differences across scenarios. In other words, impacts on different age groups remain consistent across scenarios. Citizens vulnerable due to their lower socio-economic status (based on household income, unemployment rate and lack of higher education) have been shown to be disproportionately affected by poor air quality and will likewise benefit more on average from reduced air quality. 24 As with the analysis for sensitive groups, impacts on socio-economic groups remain consistent across scenarios, with the effects varying by pollutant and socio-economic group.

To meet the targets associated with the preferred package, some of the overall adjustment costs will be borne by households by switching to lower polluting devices such as for domestic heating. Some change in behaviour would likely be triggered by national or local strategies to abate pollutant emissions (such as a switch to cleaner modes of transport, including public transport). The extent of the costs borne by households for such measures will depend ultimately also on public policy choices made in Member States as regards financial and investment support mechanisms. There are no direct administrative costs falling on citizens.

The macroeconomic modelling undertaken shows that on an aggregate level, private consumption increases compared to the baseline in 2030 across scenarios, including the preferred one, when productivity gains from clean air are factored in. When taking into account further market and also non-market effects (avoided health care costs, years of life lost, loss of utility due to sicknesses etc.), which could not be taken into account in the macroeconomic modelling undertaken but is addressed separately, the overall benefit would become even larger.

An increased stringency of the legislative framework, as per the preferred policy options addressing governance and enforcement shortcomings will ensure higher compliance with the set objectives, and thus translate into environmental, social, and economic benefits for the wider public.

Other

The preferred policy package will improve air quality assessments done through monitoring and modelling, as well as data availability on air quality. This will be helpful for researchers that work on air quality monitoring and modelling, as well as for civil society organisations that work on improving air quality through awareness raising campaigns and other actions.

2.Summary of costs and benefits

II. Overview of administrative costs and the one-in-one-out scheme – Preferred option

The following provides an overview of the costs of the different policy options that form part of the preferred package. As presented in Section 6 of the main report, a 10 µg/m3 target for annual mean PM2.5 (policy-option I‑2) amounts to adjustment or mitigation costs of 5.6 billion Euro per year in 2030, with the most impacted sectors being industry, households and livestock.

To assess the potential administrative burden placed on different actors, the EU’s Better Regulation Toolbox Standard Cost Model (SCM)  29 was used. The SCM uses information on: number of activities required, with the time required per activity and the cost per unit of time spent. The aim is to estimate additional costs (or cost reductions) of new policy options compared to the baseline scenario.

The following tables provides an overview of administrative costs related to monitoring, assessment, implementation and enforcement. The total administrative costs are estimated to range from 75 to 106 million Euro per year in 2030, increasing with the stringency of the scenario, with costs in the preferred scenario estimated at 78 million Euro per year.

Administrative costs estimates include:

·costs for all policy options and their individual measures included in the set of preferred policy options that are not linked to the level of ambition of revised EU air quality standards, which add up to about 75 million Euro per year – this includes approximately 4.8 million Euro per year related to better implementation and communication (see Table A3.1), and a further 70.3 million Euro per year related to improved monitoring and assessment (see Table A3.2);

·costs linked to the development of air quality plans, which depend on the number of exceedances above EU air quality standards to be expected in the target year 2030. This component hence depends on the level of ambition assumed via policy options I-1, I-2 or I-3, which adds up to between 1 and 31 million Euro per year. Table A3.3 provides an overview of costs related to exceedances per pollutant – based on assumption of residual exceedances based on the modelling that underpins this impact assessment.

Costs for consumers and businesses are represented jointly as there are no direct regulatory requirements for businesses stemming from the Ambient Air Quality Directives. This also means that there are no direct administrative costs to be borne by business or citizens. This means there is no need to look at potential off-setting measures as part of the Commission’s commitment to the ‘one-in-one-out’ scheme, and therefore the tables below do not contain the part of the template reserved for “costs related to the ‘one in, one out’ approach”.

Hence, most costs for consumer and businesses are indirect costs that cannot always be broken down into who will bear what share. The costs most clearly attributable are direct administrative and enforcement costs falling on the competent authorities in Member States.

3.Relevant sustainable development goals

Annex 4: Analytical methods

1.Overview

A range of interventions (or policy measures) have been defined to revise various elements of the Ambient Air Quality Directives. Each intervention would have a number of associated impacts, with the exact impacts, their size and significance depending on each individual intervention. Based on the Better Regulation Guidelines, 33 these interventions have been compared on the basis of how they address the objectives considering their effectiveness, efficiency and coherence. 

Twelve indicators (see Table A4.1) have been defined to capture and present the key economic, environmental, and social impacts associated with the interventions being considered. All interventions have been appraised against this set of indicators, to ensure consistency in the analysis and presentation of results.

Table A4.1 - Twelve indicators used as basis for in-depth assessment

Across each of these specific indicators, available evidence on the effectiveness, efficiency and coherence of the interventions has been collated, assessed and, where possible, quantified in comparison to the baseline. Where quantification was not possible, impacts were assessed in a qualitative way, clearly indicating the type of the most important impacts and their likely magnitude.

To support the assessment of impacts, three main sources of evidence were used: quantitative modelling, in particular focusing on the impacts of different air quality standards, detailed literature review and extensive stakeholder engagement. The remainder of this Annex focuses on presenting in further detail the approach taken to the quantitative modelling.

2.Quantitative modelling of air pollutant standards 

This section contains a general introduction to the modelling framework deployed in the support study for this impact assessment and a description of the most important elements of relevance for the assessment of policy options. More detailed descriptions including data sources for various underlying assumptions used in the modelling can be found in the annexes to the support study.

2.1Introduction of the modelling framework

Quantitative modelling has been conducted with a state-of-the-art modelling framework, including: the Greenhouse gas – Air pollution Interactions and Synergies (GAINS) model and MET Norway’s chemical transport model (EMEP CTM) with the uEMEP downscaling extension for fine resolution. This modelling assesses a number of effects, in particular: air pollutant emissions, concentrations, health and ecosystem impacts, feasibility to attain particular air quality targets as well as respective measures and their costs.

The GAINS integrated assessment model, developed at the International Institute for Applied Systems Analysis (IIASA), addresses air pollution impacts on human health from fine particulate matter (PM2.5) and ground level ozone (O3), vegetation damage caused by ground level ozone, the acidification of terrestrial and aquatic ecosystems and excess nitrogen deposition on soils. GAINS brings together data on economic development and structure, control potential and costs of emission sources, the formation and dispersion in the atmosphere of - as well as the inter-relations between - pollutants such as sulphur dioxide (SO2), nitrogen-oxides (NOx), particulate matter (PM), non-methane volatile organic compounds (NMVOCs) and ammonia (NH3). GAINS assesses more than 1 000 emission control measures for all EU Member States, computes the atmospheric dispersion of pollutants and analyses the costs and environmental impacts of pollution control strategies. In its optimisation mode, GAINS identifies the cost-effective emission control strategies that can be used to inform policy processes and international negotiations on mitigation of atmospheric air pollutants.

The EMEP CTM is a state of the art atmospheric chemistry transport model, and includes a recently developed novel, but well documented,  34 / 35 / 36 uEMEP downscaling module that allows the estimation of ambient air pollution concentrations down to a grid resolution of approximately 250x250 m2 for the whole of Europe. Downscaling is carried out where suitable high resolution emissions proxies are available. This includes the emissions from traffic, shipping, stationary combustion, off road combustion and aviation.

Annual mean concentrations are calculated with the EMEP model under different policy scenarios for the following pollutants and indicators: SO2, NO2 and NOX, PM2.5, PM10, NMVOC, O3, SOMO35, NH3, BaP, benzene and carbon monoxide (CO). Downscaling is applied to a selection of these pollutants (PM2.5, PM10, NO2, BaP, Benzene, CO and ozone) on annual mean concentrations. BaP is not normally explicitly modelled by the EMEP modelling suite. However, a BaP emissions inventory is available for present day emissions, though no scenario trends are available. By applying the same trends used for PM2.5 emissions to the BaP emissions, BaP can then be modelled explicitly by the EMEP modelling suite for all scenarios. Heavy metals, regulated under the Ambient Air Quality Directives cannot be quantitatively assessed with the EMEP CTM modelling suite. Therefore, these have been considered outside of the integrated modelling system through statistical analysis, by comparing different concentration thresholds to monitoring data for 2019. 37  

The integrated GAINS and EMEP models provide analysis of many of the impacts considered here. That said, some further calculations and post-processing was required to bring out further impacts associated with the interventions. This was the case for the assessment of health, social cost, and impacts on vulnerable groups. This also includes analysis by linking the GAINS model with the JRC-GEM-E3 model to explore macro-economic, GDP and employment effects. GEM-E3 is an applied general equilibrium model that covers the interactions between the economy, the energy system and the environment. It represents the whole economy and the interactions between key actors: firms, households and governments in the EU and in the rest of the world. Annex 5 (section 7 on macro-economic impacts) provides some further details on how GAINS results feed into GEM-E3.

All impacts are assessed compared to the baseline, in both a mid-term (2030) and long-term (2050) time horizon. The overall quantitative modelling flow is summarised in Figure A4.1.

Figure A4.1 – Modelling framework applied to assess the twelve indicators

2.2Application of GAINS

The policy options defined to address health and environmental outcome shortcomings (corresponding to policy area 1 in the support study) aim to attain closer alignment of air quality standards with the recently published WHO Air Quality Guidelines. The underlying analysis has assessed different policy scenarios (in line with the different policy options presented in this impact assessment) corresponding to different ambition levels. In addition, a Maximum (technically) Feasible Reductions (MFR, or MTFR) scenario was generated for both target years 2030 and 2050, which minimises emissions taking into account all available technologies irrespective of costs and thus represents the lower limit of emissions achievable with technical measures only.

The ‘headline indicator’ of the extent of the alignment with the revised WHO Air Quality Guidelines (and for expressing the level of ambition of different scenarios assessed) is the annual mean concentration of fine particulate matter (PM2.5), as this air pollutant at its current levels is associated with the most harmful effects on human health. The scenarios are defined based on assumptions of different PM2.5 levels as a headline indicator, but will also include assumptions for each pollutant covered by the current Ambient Air Quality Directives.

Results for the different policy scenarios have been assessed against a baseline, which includes existing and (in line with the Better Regulation guidelines) policies proposed. Annex 5 describes the baseline assumptions including the policies included.

The emission scenarios have been developed with the optimisation module of the GAINS model, which has been applied to identify cost-optimal strategies to achieve ambient PM2.5 concentrations in compliance with ambient air quality standards, where this is technically feasible through the model optimisation. The GAINS model includes a linearised approximation of the EMEP atmospheric model relating emissions of PM precursor pollutants to ambient concentrations on a (roughly) 7x7 km2 grid. GAINS can determine the cost optimal solution to achieve certain targets on ambient air quality. For the present analysis, the optimisation analysis is constrained to achieve PM2.5 compliance at this grid level. In case the model finds no feasible solution for achieving compliance in all grid cells, the constraints are relaxed to allow for a certain fraction of exceeding areas where additional local policy measures will be needed to achieve compliance. For such grid cells, the optimisation requires at least a 90% improvement of ambient PM2.5 concentrations towards the concentration levels attained in the maximum feasible reductions (MTFR) case. The cost optimisation is thus used to suggest the most cost effective national or EU wide emission control measures to bring ambient concentrations close to the ambient air quality limit values.

Given GAINS contains simplified atmospheric calculations based on a linear approximation of the EMEP CTM at 7 km resolution, it can only assess compliance at background level. It is important to note that compliance at hot spots, e.g., in busy street canyons, may require supplementary local measures (e.g., traffic restrictions, which cannot always be reflected in the GAINS model). To some degree, the question of compliance at hot spots may be addressed by adding a margin to the background PM2.5 concentration levels in the cost optimisation.

For any of the scenarios, if a feasible solution for attaining PM2.5 standards at background level is found, GAINS quantifies the related emissions of, at least, PM2.5, SO2, NOX, NH3, NMVOC, and CO in each Member State and economic sector. These are then fed into the EMEP CTM and uEMEP downscaling scheme to calculate ambient concentrations of air pollutants at fine resolution.

While the PM2.5 objectives are the driving indicator defining the different scenarios, different ambition levels for PM2.5 will have implications for the concentration levels of other air pollutants. Optimising for concentrations of multiple pollutants has not been considered feasible in the framework of this impact assessment. However, from the high-resolution calculations of the EMEP CTM, ambient concentrations of all pollutants covered in the model are estimated for the scenarios optimising around PM2.5. This allows quantification of the range of feasible concentration limits for other pollutants under each scenario.

2.3Concentration modelling methodology

Concentration modelling of the emission scenarios provided by GAINS is carried out using the EMEP CTM and uEMEP models. uEMEP calculates only annual mean concentrations.

Exposure calculations using the modelled concentrations are carried out at grid resolution of approximately 250x250 m2, matching the resolution of the available population density data, and these are used for the health impact assessment. Further to the exposure calculations, additional calculations at higher resolution, 50x50m2, are carried out to ascertain the impact of the emission scenarios at measurement station sites across Europe. Changes in concentrations at measurement sites are used to assess the possible level of attainment at these sites, in relation to the currently observed concentrations. All scenario simulations were using meteorological conditions for 2018. 38

Concentrations at individual stations cannot be expected to be perfectly matched with a Europe wide modelling approach. However robust statements about the likely distribution of concentration levels across stations can be made. Although no specific street canyon module was employed, experience has shown that concentrations generated with uEMEP for PM2.5, PM10, NOX and NO2 are comparable to measured roadside concentrations. For this impact assessment, the downscaling was extended to include O3, benzene and CO.

The analysis was limited to annual mean concentrations. For SO2 and the indicators that require temporal resolutions higher than annual mean, the EMEP model is used without downscaling. For the downscaled compounds of NO2, PM2.5 and PM10, statistical relationships based on observed concentrations are used to infer statements about likely compliance with short-term daily limit values, such as done in previous work for the Commission on the Thematic Strategy on Air Pollution 39 . No assessment of hourly indicators is carried out.

Benzo(a)pyrene (BaP) remains a problem in several countries in Europe and is mostly related to residential sector emissions (wood and coal combustion in stoves and small boilers). While BaP is not normally explicitly modelled by the tools used for this analysis, the analysis did include BaP using current day BaP emission inventories and linked them to PM2.5 emission scenarios to provide a quantitative assessment of BaP concentrations. This approach assumes that the ratio of BaP in PM2.5 will not change for any of the scenarios.

2.4Assessment of health impacts

The assessment includes premature mortality caused by long-term exposure to particulate matter (PM) and nitrogen dioxide (NO2), mortality caused by ozone (O3) peaks, and an estimate for the morbidity related to long-term and short-term exposure to particulate matter.

The WHO updated its Air Quality Guidelines in 2021. In the process of the update, also the systematic reviews on the current scientific knowledge concerning the mortality related to exposure to air pollution have been updated, leading in turn to updated concentration response functions. The latest WHO relationships are deployed in this analysis. The premature mortality are also estimated per age group (one-year interval). The analysis combines these estimates with the life expectancy (which can vary across Member States) from Eurostat, to assess the number of years of life lost (YLL). For future years, the evolution of the population is considered via the Eurostat projections.

To assess impacts on morbidity, the approach taken here is based on that taken in the second Clean Air Outlook, which in turn based its method on the health pathways and concentration response functions (CRF) recommended by the WHO in its Health risks of air pollution in Europe (HRAPIE) reports. 40 Acknowledging that there have been developments in the underlying evidence base since HRAPIE, but that the WHO has not undertaken a comprehensive recent review of morbidity pathways, a targeted review of literature was undertaken in preparation of the support study to explore whether there are other pathways for which evidence is stronger. The analysis only considers morbidity pathways associated with exposure to PM2.5.

In summary, this yields a three-tiered health impact assessment:

1.premature mortality caused by the long-term exposure to air pollution using the concentration response functions (CRF) recommended by the WHO;

2.morbidity caused by long-term and short-term exposure based on the HRAPIE recommendations from 2013 (chronic bronchitis in adults, bronchitis symptoms in children, cardiovascular hospital admissions, respiratory hospital admissions, infant mortality, restricted activity days and lost working-days);

3.morbidity effects beyond HRAPIE, to incorporate new insights that became apparent after the 2013 HRAPIE study, and to provide a more complete overview of the health impact due to air pollution. This covers three additional health outcomes in the primary analysis (asthma in children, lung cancer, stroke (CVA)), and three additional health effects in sensitivity analysis (COPD 41 , Diabetes Mellitus Type 2 and myocardial infarction).

For the main scenario analysis, quantification of health impacts for comparing the benefits of different policy options is limited to the impact of air pollution concentrations in excess of the revised WHO Air Quality Guidelines (from 2021). This approach has been adopted given that:

·The guideline exposure levels have been subject to extensive review work from WHO and represent an up-to-date overview of scientific knowledge on the subject, including on levels above which the health impacts are well documented;

·There is added uncertainty in the applicability of concentration response functions below the guideline exposure levels suggested by the WHO (also note that below these levels the contribution of natural sources of air pollution becomes more significant).

It is acknowledged, however, that this approach likely underestimates the total impact of air pollution on health (and thus also the likely benefits of action to improve air quality). For this reason, further quantification has been carried out to inform sensitivity runs (results of which are available in the support study).

This approach of assessing impacts above WHO Air Quality Guidelines levels is consistent with the approach the EEA adopts in its latest briefings on air quality in Europe 42 . It is also consistent with the approach followed in the preparation of the forthcoming Third Clean Air Outlook (publication foreseen for end 2022).

2.5Monetisation of health and non-health impacts

Calculating the costs to society of air pollution is a means of monetising the effects of air pollution, such that they can be more readily compared to the costs of mitigation action. To estimate the costs to society, the health impacts calculated in the previous step (such as number of deaths and of adverse health outcomes) were combined these with monetary impact values to capture the impact on: lost utility or welfare, lost labour (or productivity) and health care costs.

For human health impacts, the monetary values applied in the second Clean Air Outlook are used. The values are based on an extensive literature review of the latest approaches by organisations such as the OECD. The second Clean Air Outlook involved an extensive review of the literature available at the time, 43 and concluded in December 2020. Some of the health outcomes of the third tier go beyond what was covered in the second Clean Air Outlook. For these, a targeted literature review was undertaken to support the selection of appropriate monetary impact values for these pathways (available in the underlying support study). Also in line with the second Clean Air Outlook, results are presented for different approaches to monetising impacts: a VSL (value of statistical life) approach, which monetises the number of deaths, and a VOLY (value of statistical life year) approach, which instead monetises life years lost. For the aggregate assessment, the mortality effects associated with NO2 are excluded to avoid the risk of overlap with the mortality effects of PM2.5.

Materials damage has long been associated with emissions of SO2 and NOX. Damage values per unit emission for SO2 and NOx have been taken from the CASES study (CASES, 2008), and applied to the emissions changes observed in the integrated modelling.

Air pollution is also associated with a range of ecosystem impacts. Several of these impact pathways (but not all) have been monetised in the literature, most commonly: crop damage, forest damage and damage to ecosystems. Methods to monetise such effects stem from the ECLAIRE study 44 and are in line with what was done in the second Clean Air Outlook. For crops and forests, impacts from exposure to ozone on yields or productivity were taken into account. Forest damage reflects in addition reductions in carbon sequestration potential. For this, two different estimates of carbon mitigation costs were assumed, resulting in a low and high variant of forest damage (but only after 2030 do the assumptions for Low and High diverge). Impacts on ecosystems tend to be most significant out of the three. The analysis was limited to terrestrial ecosystems and the focus was on exceedance of the critical load for nitrogen in Natura 2000 sites, with valuation applied to the area subject to critical loads exceedance. No account was taken of exceedance of the critical load for acidification, because the area concerned is far less than that affected by eutrophication and there is potential for double counting if results for both effects are combined. A willingness to pay approach to valuation is adopted consistent with that used for other impacts assessed. A Low and High estimate is adopted to reflect uncertainty in the underlying valuation techniques.

3.Shortcomings of the modelling approach and sources of uncertainty

One of the major uncertainties in air quality modelling remains the correct representation of emissions, including how they are distributed in space. Modelling quality suffers where emission inventories (submitted by Member States) are deficient, e.g. because emissions are underestimated or unknown emission sources are not included. To some extent, these effects are reflected in the underlying modelling work when running sensitivity analyses with a bias adjustment. Modelling uncertainties in methodologies also lead to limitations. It is worth noting the EMEP and uEMEP models have been applied in countries where emissions are better known. Under these conditions the model performance is much improved.

During the course of the modelling some clear challenges in emissions have been found. These include:

·Separation and spatial distribution of national and international shipping emissions;

·Individual industries with large and uncertain emissions that can dominate the exposure in a whole city;

·Incorrect allocation of some residential heating emissions;

·Reported non-exhaust emissions that may not be adequately spatially distributed or quantified.

For the estimation of chronic mortality, the following limitations are observed:

·Only the mortality related to long-term exposure to PM, NO2 and O3 is considered. Other pollutants and mortality due to short-term exposure are not considered.

·Results for mortality are not corrected for overlaps between the different pollutants. As an indicative estimate for the order of magnitude of the overlap, HRAPIE suggests an overlap of 33%. 45 This number is, however, associated with a large uncertainty.

·Since the meteorological data is the same for each year under consideration, the impact of climate change is not considered (also holds for morbidity estimates).

·The uncertainty on the results is larger for the results reported per individual country, than for EU-27 total estimate (also holds for morbidity estimates).

For the estimation of morbidity, the following additional limitations are observed:

·Only the morbidity related to exposure to particulate matter is considered. Other pollutants are not considered.

·Future projections for the baseline incidence are unavailable for most health outcomes. The analysis therefore relies on the morbidity rates for the most recent year for the future baseline morbidity. Impacts due to improvements in health care, more / less healthy lifestyle etc. are hence not considered.

·In general, the uncertainty on the morbidity estimates is larger than the uncertainty on the mortality estimates, mostly due to more pronounced uncertainty in the input datasets (concentration response functions, baseline morbidity). When interpreting the results, the focus should therefore lie on relative differences between scenarios. 

Annex 5: Baseline, Maximum Technically Feasible Reduction and policy Scenarios – modelling results

This annex complements the description of baseline development in chapter 5 and of policy scenario developments in chapter 6 of the main report by providing further detailed results from the quantitative modelling. This means that modelled results that are already included in the main report are not repeated here. The underlying support study contains further, more disaggregated results, including tables with results per Member State.

1.Description of the baseline scenario

The starting point for the quantitative analysis is the baseline scenario, which provides a critical reference point against which to assess changes and impacts of the formulated policy options. It serves as the counterfactual for examining how the situation is expected to change in the case of no further changes to the Ambient Air Quality Directives. The baseline is defined by the current status of implementation of different obligations under the existing EU Directives relevant for air pollutant releases as well as national legislation, if stricter than the EU law. This defines the existing political and legal context at the EU and at the national level. The current status of implementation is well defined in several existing studies, not least the second Clean Air Outlook. This baseline builds on the backdrop of existing measures and policies already committed (including some which might require introduction of further measures in the near term).

In line with the Commission’s Better Regulation guidelines, policy proposals (even though still subject to modifications in the course of the policy making cycle) form part of the baseline assumptions. Policies and measures included in the baseline are considered to continue over the duration of the analysis period. Key elements of the baseline scenario that have been updated since the Second Clean Air Outlook include: 46  

·The broader EU policy environment and potential changes - including revised European Commission climate targets and related legislative proposals (Fit for 55) as well as of preliminary assumptions for the introduction of Euro 7;

·Confirmed changes at Member State level (i.e. adopted policies and measures as set out in National Air Pollution Control Programmes);

·Sulphur Emissions Control Area (SECA) in the Mediterranean Sea from 2025;

·Assumptions about the development in the non-EU countries, which are of relevance owing to the impact of transboundary pollution, in particular, new data and projections (energy and agriculture) for Western Balkan, Ukraine, Moldova, and Georgia from a recently completed EU funded project. 47  

2.Air pollutant emissions: baseline development and scope for MTFR reductions

Considering current economic and environmental policies included in the baseline for the EU-27 will result in continued decline in emissions of key air pollutants (Figure A5.1). Compared to 2015, emissions of PM2.5, NOx, and SO2 are estimated to drop by 50 to 70%, NMVOC by 25%, while for ammonia (NH3) only about 5% reduction is calculated by 2030. The trends are expected to continue towards 2050 but with much smaller further reductions.

The key drivers of emissions change towards 2030 are different for each pollutant:

·for PM2.5 most of the reduction is due to reduced use of coal and biomass in the residential sector and transition to cleaner technologies;

·for NOx recent legislation and fuel trends (less diesel and increase of hybrid and full electric vehicles) are the key drivers;  

·for SO2, first strong reduction in coal use in power plants and then residential coal use decline are among major factors;

·For NMVOC, reduction in residential heating sector (see PM2.5) and transport (see NOx) are key contributors;

·For NH3, the (limited) decline is mostly driven by structural changes (livestock numbers), including reduction of mineral nitrogen fertilizer application.

Figure A5.1 – Trends of air pollutant emissions in the EU-27; baseline scenario (GAINS)

For NH3, further emission reductions are expected from recently proposed revision of the IED, including cattle and reducing the farm size threshold for pigs and poultry. 48 While the modelling baseline used here    does not include this proposal (implementation in the GAINS model is under way within the study supporting the third Clean Air Outlook), the impact assessment study of the revised IED estimated the potential ammonia reductions at about 155 kt per year, about 4.4% of total EU ammonia emissions.

To define the scope of the maximum mitigation potential based on available technologies, a maximum technically feasible reduction (MTFR) scenario for 2030 and 2050 was modelled in GAINS (Figure A5.2). Key elements to note:

·Lifetime of installed capacity is respected, i.e. no premature scrapping of existing equipment is considered;

·No further structural (e.g. fuel switch) or behaviour-driven (e.g. lifestyle choices of reducing meat/diary intake) measures are considered beyond what is included in the baseline, neither at the local nor regional level;

·Potential local and technological constraints are taken into account to the extent that they are reflected in the model drawing on previous Member State consultations and technology information; 

·Any potential financial constraints are ignored (in other words do not hinder the take-up of measures).

Figure A5.2 – Scope for further emission mitigation of air pollutant emissions in 2030 and 2050 in the EU-27. Changes shown relative to 2015 (GAINS model)

Since current legislation is expected to reduce emissions of SO2 and NOx, further potential is rather limited and even declines in the long-term owing to the reduced used of fossil fuels already in the baseline. Remaining limited potential has been identified for industrial process. For PM2.5, key further mitigation can be achieved in the residential sector and also by enforcing bans on open burning of various agricultural residues that in spite of existing legislation are still burned, while there would remain very limited, if any, potential to further reduce emissions from power or industrial sectors. For NMVOC, apart from some potential in residential sector and agricultural burning, further reductions in solvent use applications were estimated. For ammonia, mitigation of emissions from mineral nitrogen fertilizer application and livestock offer significant reduction potential assuming that measures addressing housing, storage, and application of manures on land would be introduced in an integrated manner (as proposed in the revised IED), but for a much larger number of farms than is currently the case as per baseline assumptions, especially for cattle.

The mitigation potential shown above (as well as the results that include sectoral breakdown shown in chapter 6 of the main SWD) vary strongly between Member States depending on structure of emission sources and local constraints. 49 The following figures (A5.3 to A5.7) reflect this variation. They present the reduction of emissions of PM2.5 and its precursors compared to the baseline as calculated in the GAINS model for policy and MTFR scenarios, showing a disaggregated sector/measure resolution and with results per Member State. 

Figure A5.3 – Reduction of PM2.5 emissions, split by Member State (2030)

Figure A5.4 – Reduction of SO2 emissions, split by Member State (2030)

Figure A5.5 – Reduction of NOX emissions, split by Member State (2030)

Figure A5.6 – Reduction of NH3 emissions, split by Member State (2030)

Figure A5.7 – Reduction of VOC emissions, split by Member State (2030)

3.Air pollutant concentration projections

The main report includes detailed results for the baseline and policy scenarios showing the number of stations with remaining exceedances above different limit values. To complement these results, this section includes a number of maps that show the geographical distribution of air pollutant concentrations in the baseline, various optimisation and MTFR scenarios.

The below maps (Figures A5.8) show concentrations of fine particulate matter (PM2.5) in 2030 indicate that already under baseline assumptions substantial improvements in air quality over the coming decades can be expected, with most of Europe reaching concentration levels below 10 μg/m3. Areas including parts of Central Europe, the Baltics, Italy and Northern parts of France move even to concentration levels of below 5 μg/m3.

The optimisation scenarios bring further improvements. Significant parts of Europe remain above 5 μg/m3 in the OPT-5 scenario (which corresponds to policy option I-1), including Northern Italy (Figure A5.10), the border region of Czechia, Poland and Slovakia (see also Figure A5.11), as well as southern regions along the Mediterranean coast of the EU. This remains the case in the MTFR scenario, which would bring little additional improvements.

Going to 2050 (Figure A5.9), additional areas reach concentration levels below 5 μg/m3, notably most remaining parts of Poland and Hungary, as well as of Belgium and the Netherlands. Elevated levels of fine particulate matter (PM2.5) in Southern Europe are rather persistent, which is explained by the fact that much of the pollution is due to natural sources, and Sahara dust and sea spray in particular (which the current Ambient Air Quality Directives allow to be deducted from air pollution levels reported).

Figure A5.10 highlights the specific dynamics in the Po Valley region in Northern Italy, where specific meteorological and orographic circumstances lead to reduced dispersion, and elevated emission levels from residential heating (including biomass burning) as well as agricultural emissions represent particular challenges. While under the preferred policy option the area exposed to PM2.5 concentration levels above 10 µg/m3 reduces significantly by 2030, some hotspots would be expected to remain across optimisation scenarios (and significant parts remain are just below 10 μg/m3).

Similarly, for much of Eastern Europe (see for example Figure A5.11), residential heating (often reliant on fossil fuel combustion) and industry production facilities today lead to elevated PM2.5 concentration levels. Under the preferred policy option and based on the measures taken to address these emissions, the area exposed to PM2.5 concentration levels above 10 µg/m3 reduces almost to zero by 2030.

As regards the concentration levels of particulate matter (PM10) for the baseline and MTFR, most areas in the EU reach concentration levels of below 15 µg/m3 in 2030 already in the baseline when looking at background concentration levels (Figure A5.12). The remaining areas above 15 µg/m3 are along the Mediterranean cost and as such can be explained by pollution from natural sources.

Zooming in to a finer resolution, however, indicates that local peaks of concentration levels of particulate matter (PM10) can be expected to remain under all scenarios analysed. These can be linked, for example to constant levels of non-exhaust emissions from transport: Figure A5.13 illustrate the particular case of Stockholm, where pollution peaks can be seen along the main road traffic axes.

The baseline assumptions related to improved vehicle emissions standards and increased electrification of road transport lead to reductions of nitrogen dioxide (NO2) concentration levels in urban centres across the EU from 2020 to 2030, with further reductions in the MTFR and towards 2050 (Figure A5.14). A large area with noticeable improvements in NO2 concentration levels is the area spanning parts of Belgium and the Netherlands as well as the Ruhr area in Germany. Zooming into this region (Figure A5.15) shows that areas with an annual mean concentration of over 20 µg/m3 are markedly reduced in 2030, already in the baseline scenario, with remaining areas above 20 µg/m3 being situated mainly around ports.

Moving from 2020 to 2030 in the baseline already brings marked improvements to ozone (O3) (26th highest maximum 8-hour daily running mean) concentrations levels with large areas where concentrations levels are reduced to below 100 µg/m3, and few remaining areas with levels above 120 µg/m3 (in Northern Italy). Further reductions are observed in the MTFR in 2030 and even more pronounced in baseline and MTFR in 2050, when most parts of the EU have levels of 80 to 100 µg/m3, with levels above 100 µg/m3 remaining primarily in Romania, Northern Italy, in some parts around the Mediterranean cost as well as parts of North-Western Europe (Figure A5.16).

Figure A5.17 shows low levels of sulphur dioxide (SO2) concentrations levels of below 40 µg/m3 prevailing throughout the EU. This is explained by the strong reduction in coal use in power plants as well as in residential coal use in line with EU energy and climate policy that form part of the baseline assumptions.

Carbon monoxide (CO) concentration levels are below 1 mg/m3 in most of the EU, with little changes between baseline and MTFR. Smaller patches are between 1 and 2 mg/m3.

Most of the EU has concentration levels for benzene (C6H6) below 0.8 mg/m3 already in the baseline in 2020. Parts of Northern Italy show more elevated levels between 0.8 and 3.4 mg/m3. This area is reduced significantly already in the baseline in 2030.

Concentrations levels for benzo(a)pyrene (BaP) are above 1 ng/m3 in significant parts of the EU in the baseline in 2020, notably in Poland, in Northern Italy and in more localised places in Southern, Central and Eastern Europe. Already the baseline assumptions reduce these areas in 2030 to a good extent, most notably in Poland. In the MTFR in 2030 and going towards 2050, there are very limited areas left with concentration levels exceeding (in Poland, Northern Italy and Greece) 1 ng/m3.

Maps for fine particulate matter - PM2.5

Figure A5.8 - PM2.5 concentrations for baseline 2020 and a range of optimised (OPT) scenarios, including MTFR for 2030. Calculations are made on the EMEP 0.1o grid For details (including on bias correction), please see the underpinning support study on the revision of the Ambient Air Quality Directives.

Base 2020                    Base 2030

OPT-15 2030 (Policy Option I-3)            OPT-10 2030 (Policy Option I-2)    

OPT-05 2030 (Policy Option I-1)            MTFR 2030

Figure A5.9 - PM2.5 concentrations for baseline 2020 and a range of optimised (OPT) scenarios, including MTFR for 2050. Calculations are made on the uEMEP 250 m grid. For details (including on bias correction), please see the underpinning support study on the revision of the Ambient Air Quality Directives.

Base 2020                    Base 2050

OPT-15 2050 (Policy Option I-3a)            OPT-10 2050 (Policy Option I-2a)

OPT-05 2050 (Policy Option I-1a)            MTFR 2050

Figure A5.10 - Focus: Region in Northern Italy. PM2.5 annual mean concentrations for baseline 2020 and a range of optimised (OPT) scenarios, including MTFR for 2030. Calculations are made on the uEMEP 250 m grid. Note the change in colour scale to emphasize concentrations between 5 and 12 µg/m3. For details (including on bias correction), please see the underpinning support study on the revision of the Ambient Air Quality Directives.

Base 2020                    Base 2030

OPT-15 2030 (Policy Option I-3)            OPT-10 2030 (Policy Option I-2)    

OPT-05 2030 (Policy Option I-1)            MTFR 2030

Figure A5.11 – Focus: Region in Central Europe. PM2.5 annual mean concentrations for baseline 2020 and a range of optimised (OPT) scenarios, including MTFR for 2030. Calculations are made on the uEMEP 250 m grid. Note the change in colour scale to emphasize concentrations between 5 and 12 µg/m3. For details (including on bias correction), please see the underpinning support study on the revision of the Ambient Air Quality Directives.

Base 2020                    Base 2030

OPT-15 2030 (Policy Option I-3)            OPT-10 2030 (Policy Option I-2)    

OPT-05 2030 (Policy Option I-1)            MTFR 2030

Maps for particulate matter - PM10

Figure A5.12 - PM10 annual mean concentrations for baseline (Base) and MTFR for 2020, 2030 and 2050. Calculations are made on the uEMEP 250 m grid. For details (including on bias correction), please see the underpinning support study on the revision of the Ambient Air Quality Directives.

Base 2020

Base 2030                    MTFR 2030

Base 2050                    MTFR 2050

Figure A5.13 – Focus: Region in Scandinavia. PM10 annual mean concentrations for baseline 2020 and a range of optimised (OPT) scenarios, including MTFR for 2030. Region shown is the city of Stockholm in Sweden. Calculations are made on the uEMEP 250 m grid. Note the change in colour scale to emphasize concentrations between 7.5 and 20 µg/m3. For details (including on bias correction), please see the underpinning support study on the revision of the Ambient Air Quality Directives.

Base 2020                    Base 2030

OPT-15 2030 (Policy Option I-3)            OPT-10 2030 (Policy Option I-2)    

OPT-05 2030 (Policy Option I-1)            MTFR 2030

Maps for nitrogen dioxide - NO2

Figure A5.14 - NO2 concentrations for baseline (Base) and MTFR for 2020, 2030 and 2050. Calculations are made on the uEMEP 250 m grid. For details (including on bias correction), please see the underpinning support study on the revision of the Ambient Air Quality Directives.

Base 2020    

Base 2030                    MTFR 2030

Base 2050                    MTFR 2050

Figure A5.15 – Focus: Region in North-Western Europe. NO2 annual mean concentrations for baseline 2020 and a range of optimised (OPT) scenarios, including MTFR for 2030. Region in North-Western Europe including Belgium, Germany and The Netherlands. Calculations are made on the uEMEP 250 m grid. Note the change in colour scale to emphasize concentrations between 10 and 25 µg/m3. For details (including on bias correction), please see the underpinning support study on the revision of the Ambient Air Quality Directives.

Base 2020                    Base 2030

OPT-15 2030 (Policy Option I-3)            OPT-10 2030 (Policy Option I-2)    

OPT-05 2030 (Policy Option I-1)            MTFR 2030

Maps for ozone – O3

Figure A5.16 - O3 (26th highest maximum 8 hour daily running mean) concentrations for baseline (Base) and MTFR for 2020, 2030 and 2050. Calculations are made on the EMEP 0.1o grid. For details (including on bias correction), please see the underpinning support study on the revision of the Ambient Air Quality Directives.

Base 2020    

Base 2030                    MTFR 2030

Base 2050                    MTFR 2050

Maps for sulphur dioxide - SO2

Figure A5.17 - SO2 (99th percentile daily mean) concentrations for baseline (Base) and MTFR for 2020, 2030 and 2050. Annual means are calculated and converted to 99th percentiles. Calculations are made on the EMEP 0.1o grid. For details (including on bias correction), please see the underpinning support study on the revision of the Ambient Air Quality Directives.

Base 2020    

Base 2030                    MTFR 2030

Base 2050                    MTFR 2050

Maps for carbon monoxide - CO

Figure A5.18 - CO (highest maximum 8 hour daily running mean) concentrations for baseline (Base) and MTFR for 2020, 2030 and 2050. Calculations are made on the EMEP 0.1o grid. For details (including on bias correction), please see the underpinning support study on the revision of the Ambient Air Quality Directives.

Base 2020

Base 2030                    MTFR 2030

Base 2050                    MTFR 2050

(1)

   Council (2020), Council conclusions 6650/20 (accessed: 10.06.2022)

(2)

   COM (2022), Have your say - Air quality - revision of EU rules  (accessed: 10.06.2022)

(3)

     Contract no. 09029901/2021/848269/SFRA/ENV.C.3, implementing Framework Contract no. ENV.F.l/FRA/2019/0001

(4)

     European Parliament (2021), resolution of 25 March 2021 on the implementation of the Ambient Air Quality Directives (accessed: 10.06.2022)

(5)

     COM (2021), Upgrading the ambition of EU Air Quality legislation  (accessed: 10.06.2022)

(6)

   COM (2021), public consultation on Air quality- revision of EU rules (accessed: 10.06.2022) 

(7)

     WHO (2021), WHO global air quality guidelines  (accessed: 10.06.2022)

(8)

     COM (2021), EU third Clean Air Forum (accessed: 10.06.2022)  

(9)

 COM (2021) AAQDs revision - consultation strategy - final (accessed: 04.08.2022)

(10)

     COM (2021) AAQDs revision - consultation strategy - final (accessed: 04.08.2022)

(11)

  COM (2021), OPC- Factual summary report - final (accessed: 04.08.2022) 

(12)

     COM (2021), Have your say portal  (accessed: 04.08.2022)

(13)

     COM (2021), First stakeholder meeting summary report- final (accessed: 04.08.2022)

(14)

 COM (2022), Second stakeholder meeting briefing paper (accessed: 04.08.2022)

(15)

     ECO=Exposure concentration obligation

(16)

     For a complete overview of all intervention areas, see Annex 6 to the SWD.

(17)

     Two each from: Federal Environment Agency (UBA Germany), Ministry of Transport, Baden-Wuerttemberg (Germany), Po valley regions (Italy) and three from Client Earth.

(18)

     COM (2021), Third Clean Air Forum Events page (accessed: 04.08.2022)

(19)

     COM (2021), Have your say portal (accessed: 04.08.2022)

(20)

     COM (2022), Fit for Future Platform Opinion reference: 2021/SBGR1/04 (accessed: 04.08.2022)

(21)

     Including Directives 2010/75/EU (on industrial emissions), 2009/125/EC (on eco-design), as well as EC Regulations  443/2009 and 510/2011 (on emission standards for vehicles), Regulations (EU) 2016/427 , (EU) 2016/646 , and (EU) 2017/1154 (on real driving emissions), Directive (EU)2016/2284  on the reduction of national emissions of certain atmospheric pollutants, as well as relevant published or upcoming proposals, such as on the revision of the Industrial Emissions Directive, and on Euro 7 standards for road vehicles.

(22)

     In line with the second Clean Air Outlook, results are presented for different approaches to monetising impacts: a ‘VSL’ or value of statistical life approach, which monetises the number of deaths (yielding the 119 €bn), and a VOLY or value of statistical life year approach (40 €bn), which instead monetises life years lost.

(23)

     These include infant mortality, (chronic) bronchitis in children (in adults), cardiovascular as well as respiratory hospital admissions, restricted activity days, lost working days, stroke, lung cancer and asthma in children.

(24)

     EEA (2019), EEA Report No 22/2018 (accessed: 10.06.2022)

(25)

     See previous section.

(26)

     EEA (2019), EEA Report No 22/2018 (accessed: 10.06.2022)

(27)

      Special Eurobarometer 497  (accessed: 10.06.2022)

(28)

COM (2021), Open Public Consultation on “Air quality – revision of EU rules”  (accessed: 10.06.2022)

(29)

     COM (2022), Tool #60 - the standard cost model for estimating administrative costs  (accessed: 10.06.2022)

(30)

     This and following tables categorise costs as follows: ‘low’ means costs of <100k, ‘medium’ 100k to 1 million, high >1 million EUR.

(31)

     This and following tables categorise costs as follows: ‘low’ means costs of <100k, ‘medium’ 100k to 1 million, high >1 million EUR.

(32)

     This and following tables categorise costs as follows: ‘low’ means costs of <100k, ‘medium’ 100k to 1 million, high >1 million EUR.

(33)

SWD(2021) 305 final

(34)

     Denby, B. R., Gauss, M., Wind, P., Mu, Q., Grøtting Wærsted, E., Fagerli, H., Valdebenito, A., and Klein, H. (2020): Description of the uEMEP_v5 downscaling approach for the EMEP MSC-W chemistry transport model, Geosci. Model Dev., 13, 6303–6323, Description of the uEMEP_v5 downscaling approach for the EMEP MSC-W chemistry transport model, Geosci. Model Dev., 13, 6303–6323 (accessed: 10.06.2022)

(35)

     Mu, Q., Denby, B. R., Wærsted, E. G., and Fagerli, H. (2022): Downscaling of air pollutants in Europe using uEMEP_v6, Geosci. Model Dev., 15, 449–465, Downscaling of air pollutants in Europe using uEMEP_v6, Geosci. Model Dev., 15, 449–465 (accessed: 10.06.2022)

(36)

     Transboundary particulate matter, photo-oxidants, acidifying and eutrophying components. EMEP Status Report 2020  (accessed: 10.06.2022)

(37)

     See annex 4 of the underlying support study for more detail.

(38)

     In addition, control runs for the 2015 baseline scenario were performed using meteorological data for 2015.  

(39)

     See IIASA, Thematic Strategy on Air Pollution (TSAP) Reports (accessed: 15.06.2022)

(40)

     See WHO (2013), Health risks of air pollution in Europe – HRAPIE project Recommendations for concentration–response functions for cost–benefit analysis of particulate matter, ozone and nitrogen dioxide  (accessed: 15.06.2022)

(41)

     Chronic obstructive pulmonary disease; COPD impacts are not included in the aggregated results for the valuation of health impacts (as per Annex 5.5) given concerns about the overlap with chronic bronchitis.

(42)

     EEA (2022), Europe’s air quality status 2022  (accessed: 10.06.2022)

(43)

     Available in the annex to the support study for the Second Clean Air Outlook report, as well as in the annex of the support study underlying this impact assessment.

(44)

     Europa.eu (2015), Effects of Climate Change on Air Pollution Impacts- final. -nr. 282910  (accessed: 10.06.2022)

(45)

     See WHO (2013), Health risks of air pollution in Europe – HRAPIE project recommendations for concentration–response functions for cost–benefit analysis of particulate matter, ozone and nitrogen dioxide (accessed: 16.06.2022)

(46)

     See Appendix 3 of the support study for a full list of policies included in the baseline.

(47)

     Extension of the EU Energy and Climate Modelling Capacity to include the Energy Community and its Nine Contracting Parties (ENER/2020/OP/0005)

(48)

COM(2022) 156 final/2 (accessed: 04.08.2022)

(49)

See main report of the underlying support study for estimates per Member State.