Green scenarios documentation

Transcription

1 ECMWF COPERNICUS REPORT Copernicus Atmosphere Monitoring Service Green scenarios documentation Configuration since 21/11/2016 Issued by: INERIS/ Laurence ROUÏL Date: 09/01/2017 Ref: CAMS71_2016SC1_D71.2.1b_201701_Green Scenarios Documentation_v3

2 This document has been produced in the context of the Copernicus Atmosphere Monitoring Service (CAMS). The activities leading to these results have been contracted by the European Centre for Medium-Range Weather Forecasts, operator of CAMS on behalf of the European Union (Delegation Agreement signed on 11/11/2014). All information in this document is provided "as is" and no guarantee or warranty is given that the information is fit for any particular purpose. The user thereof uses the information at its sole risk and liability. For the avoidance of all doubts, the European Commission and the European Centre for Medium-Range Weather Forecasts has no liability in respect of this document, which is merely representing the authors view.

3 Contributors INERIS Laurence ROUÏL Frédérik MELEUX TNO Hugo DENIER Van der GON Jeroen KUENEN CAMS71_2016SC1 Green Scenarios Documentations V3 Page 3 of 18

4 Table of Contents 1. Introduction 6 2. Reference Scenarios 9 3. Scenarios related to EU legislation Sector-oriented scenarios Traffic scenarios Industry scenario Agriculture scenario Residential heating Combined scenarios References 16 CAMS71_2016SC1 Green Scenarios Documentations V3 Page 4 of 18

5 Synthesis The green scenario toolbox set-up in September 2016 is based on a set of 10 new scenarios that are detailed in the present report. As a summary, the selected set of scenarios are the following: REF1: baseline scenario based on MACC/TNO emission inventory for 2011; used as the reference scenario to compute differences with the other ones; REF2: alternative reference scenario for the current situation based on new scientific knowledge. In particular, revised emission factors for road traffic and wood combustion are implemented in this scenario. Because of the potential sensitivity of the results obtained with this scenario (huge increase of PM concentrations in Europe compared to those obtained with the reference one) the forecasts are not published on the web pages; NEC2020: emission reduction scenario for 2020 as set in the revised NEC Directive 2016/2284/EU; NEC2030: emission reduction scenario for 2020 as set in the revised NEC Directive 2016/2284/EU; STRA30: road traffic emissions reduced by 30% compared to REF1; SIND30: Industry emissions reduced by 30% compared to REF1. SNAP sector 1,3, and 4 corresponding respectively to Combustion in energy and transformation industries, Combustion in manufacturing industries, and Production processes are targeted; SAGRI30: Agriculture emissions reduced by 30% compared to REF1; SRHBAT: Residential heating emissions reduced according to the performances of current best available technologies in this sector; STRAGR30: combined scenarios with road traffic and agriculture emissions reduced by 30% in each sector NEC_MTFR: Maximum Technical Feasible Technologies as defined in the negotiation process of the revision of the NEC Directive are applied to each activity sector for an maximum emission reduction. CAMS71_2016SC1 Green Scenarios Documentations V3 Page 5 of 18

6 1. Introduction The CAMS green-scenario toolbox is a modelling tool dedicated to policy support for the management of air pollution episodes. It aims at forecasting, every day for the three next days, the impact of various air pollutants emission reduction scenarios on air pollutant (ozone, nitrogen dioxide PM10 and PM2.5) concentrations in Europe. Such forecasts allow the evaluation of the efficiency of the different control strategies tested and highlight the most sensitive emission sector(s) that drives targeted air pollution episode development. The first versions of this tool were developed during the MACC suites projects. The objective was to provide an information about the impact of various and theoretical sectoral emissions reduction strategies run every day in forecast mode. Initially this information was related to the evaluation of the efficiency of short term action plans on regulatory air pollutants and the reduction assumptions were supposed to reflect implementation of cuts in emissions that could applied during episode periods as emergency or short term control measures. According to the air quality Directive 2008/50/EC, such emergency plans are supposed to be implemented by the EU member States when air pollution episodes develop. Even if short term action plans are generally local plans (applied at the city level), the objective of the MACC toolbox was to provide decision makers with a quick assessment of the impact, throughout Europe, of sectoral control measures for air emissions according to where and when they are applied. Therefore, it was possible, for a given air pollution episode, to evaluate which activity sector and which geographical area were the most appropriate to apply emission reductions. Of course both sectors and location vary with the nature of the considered pollution episode, and the time of the year. Actually, implementation of short term control measures is not widely developed in Europe because: generally reducing emissions once the episode has developed is too late, some measures like traffic speed limits are not considered as very efficient, Their cost could be high (for instance when restricted road traffic areas are implemented, cost of public transport must be adapted). However, if these measures are decided sufficiently in advance (several days before the episode peak occurs) and implemented over relevant geographical areas, they could provide interesting results and avoid some situations when regulatory limit values are exceeded. In some countries (in France for instance) short term action plans are acknowledged in several legal texts as potential efficient means to reduce exposure to air pollution episodes. The new green scenario tool developed within the CAMS_71 service is, in its first stage, an evolution of the MACC tool. The basic modelling principles are the same, with some improvements in the version of the model used (CHIMERE developed by INERIS and CNRS) and its spatial resolution (0.25 instead of 0.5 ). CAMS71_2016SC1 Green Scenarios Documentations V3 Page 6 of 18

7 The main difference stands in the number of scenario simulated (10 instead of 6) and their nature. To complement theoretical emissions reductions formalized by a percentage of cut compared to a reference situation, some of the new scenarios are closer to emission reduction strategies that are likely to be implemented in the coming years to comply with various legislations. In particular, we propose to stick to the assumptions discussed within the recent negotiations ( ) related to the revision the National Emission Ceilings directive (NECD 2001/81/EC). This is the reason why it is more appropriate to define the product of the service as what if scenarios or green scenarios rather than short term or emergency scenarios. They reflect the impact of more voluntary and ambitious air pollution control policies that could apply over the European domain. However, we have kept the sectoral discrimination that is useful to help the decision maker in understanding the influence of the different anthropogenic sources on the air pollution levels in the European regions. Thus, this tool should support the policy maker not only in its decisions but also in the elaboration of its communication strategy towards the general public and the stakeholders. In this document we present and justify the choice of the ten scenarios that are implemented in the CAMS_71 green scenario toolbox (version 2016). They are likely to change in the next versions depending on the relevance of the results obtained that will be analyzed by the end of 2016, and the policy users feedback. A first set of 10 scenarios has been defined in summer 2016, and evaluated by TNO and INERIS regarding their feasibility, complementarity and relevance for the policy maker. Following this step review, a new set of scenarios was proposed and validated by ECMWF. Since September 2016, the green scenario toolbox is implemented with those 10 scenarios. They have been presented to policy users during a workshop we organized in the framework of CAMS_71 on the 28 th November 2016 in Brussels (hosted by the European Commission, DG ENV). We got no comment about the assumptions underlying the scenarios that were considered as relevant. For this reason, and to be able to elaborate an in-depth analysis of the model responses in various episode situations, we did not change the scenarios defined in September 2016 and suggest to run them at least until September Three kinds of scenarios are proposed: Reference scenarios that reflect the current situation. Actually two scenarios are proposed: one which is compliant with official reported emissions following regulatory frameworks set in Europe, and a second reference scenario which includes some modification of the first one to include latest scientific insights regarding some emission sectors; Official regulatory scenarios discussed in the negotiations of the NEC Directive for 2020 and 2030; Sectoral emission reduction scenarios, some of them being simple percentages of emission reductions compared to the reference situation and other reflecting more realistic assumptions, actual air pollution control measures, or higher levels of ambition. Those scenarios can be applied sector by sector to assess the weight of each of them. They also can result from a combination of targeted sectors to highlight the linkages that exist between them. These scenarios and justification of their choice are presented below. TNO and INERIS elaborated this analysis and translated assumptions bearing each scenario in emission inventories datasets ready to be digested by the CHIMERE model to run the forecasts. The impact in terms of air pollutant concentrations is presented on maps over the European domain. Maps of differences of CAMS71_2016SC1 Green Scenarios Documentations V3 Page 7 of 18

8 concentrations between the scenarios and the reference situation are proposed as well. They are more illustrative of the efficiency of the tested control strategies and allow to limit the effect of the modelling uncertainties. Indeed, the difference between two modelling results obtained with the same model set-up is a more reliable variable than the absolute value of the concentrations and gives to the policy makers relevant information related to the influence of emission sources. Finally, it should be noted that the activity sectors will be described with the SNAP 1 nomenclature, compliant with the CHIMERE model s emission files. Three SNAP levels are used to describe in details emission activities. The first level is given below for information:. Table: list of the activity sectors according to the SNAP nomenclature SNAP number S1 S2 S3 S4 S5 Title Combustion in energy and transformation industries Non-industrial combustion plants Combustion in manufacturing industry Production processes Extraction and distribution of fossil fuels and geothermal energy S6 S7 S8 S9 S10 S11 Solvent use and other product use Road transport Other mobile sources and machinery Waste treatment and disposal Agriculture Other sources and sinks This report is an updated version of the report published in September Even if the choice of scenarios did not change, few more details are given in their description when needed. 1 Selected Nomenclature For Air Pollution CAMS71_2016SC1 Green Scenarios Documentations V3 Page 8 of 18

9 2. Reference Scenarios The reference scenario refers to the current situation in terms of emissions in Europe. It gives the reference or base case situation (in terms of air pollution patterns and levels for regulatory pollutants) which: Will be used for comparison against the emission reduction of green scenarios Will be used as the base case to which the emissions are applied. By definition, the reference scenario corresponds to the emission inventory used for the CAMS regional services developed in CAMS_50 that relate to European air quality forecasts, analyses and past re-analyses. Currently the MACC/TNO emission inventory developed for the year 2011 is used in those services and we propose to keep it as the reference scenario for the tool box until an updated CAMS inventory is made available. This emissions inventory has been developed by TNO and accounts for the emissions reported by the European countries in compliance with the Gothenburg protocol of the Convention on Long Range Transboundary Air Pollution (CLRTAP and for the countries belonging to the European Union, in compliance with the NEC Directive. Annual total emissions of sulfur dioxide, nitrogen oxides, Volatile Organic Compounds (VOCs), ammonia and PM are officially reported by sectors. Within former MACC suite projects, TNO worked on a high resolved description of the spatial and temporal distribution of these emissions. It resulted in a 7-8 km resolved inventory of high quality used for the MACC air quality services. The latest version relates to the year The MACC/TNO emission inventory should be updated to more recent years in the framework of CAMS in specific emission services implemented in Therefore, we propose to use as the reference case for CAMS_71 green scenarios tool in , the MACC/TNO emission inventory for the year Emission inventories result from the combination of emissions factors for a given pollutant and a unit of activity, and amount of this activity (for instance number of cars per kilometer, energy consumption...). The methodology and data used to elaborate emission inventories in the policy framework is described in guidebooks 2 edited by the European Environment Agency and the EMEP program of the CLRTAP. The 2011 MACC/TNO emission inventory is consistent with this methodology. However, this means it is built upon a number of assumptions regarding emissions factors, in particular, that could be more or less accurate and reliable. The recent dieselgate (2016) in the USA highlighted practices in the car industry that cause a substantial deviation between emissions under controlled test conditions and in the real world. A direct consequence can be an underestimation of the expected air pollutant emissions from road traffic. In a paper published by the ERMES group 3, the impact of the scandal itself on emission inventories was explained to be small, but this incident has shown that significant uncertainties may CAMS71_2016SC1 Green Scenarios Documentations V3 Page 9 of 18

10 still exist in the estimation of road transport exhaust emissions. The consequences for environmental studies are severe, even we do not have yet enough elements to appreciate them. Road traffic is not the only one sector for which high uncertainties in emission estimations can lead to an underestimation of their impact on air quality. The question of emissions from residential heating sector is particularly sensitive in Europe, and especially the evaluation of the contribution of biomass burning and wood combustion. This contribution is likely to impact the occurrence of particulate matter (PM) episodes in winter in some European regions especially when meteorological conditions are so stable that air pollutants accumulate in the lowest layers of the atmosphere. Recently, (Denier van der Gon et al, 2015) demonstrated that current PM emission inventories in Europe do not account correctly for wood combustion emissions. Both emission factors and activity data have been revised with a significant impact on PM emission patterns in Europe, especially in some areas where PM10 concentrations are exceeded in wintertime. Considering those two examples we propose to elaborate an alternative reference scenario (and emission inventory) corresponding to the current situation, but with revised emissions factors and activity data for the road traffic sector (NOx) and residential heating (PM10), supposed to reflect new scientific understanding of emissions in those sectors and possibly closer to real world emissions. The difference between concentrations forecasted with the official TNO-MACC emission inventory (2011) (REF1) and those forecasted with this alternative, modified inventory (REF2) will give an idea of the impact of the uncertainty associated with official emissions data reported through legal processes. Moreover, the science-based reference is important for the comparison to the sectorspecific green scenarios discussed later; if the science based REF2 estimates a sector to be a higher emitter, the impact of the green scenario addressing that sector may be larger than derived from comparison with the REF1. For this first 2016 version of the green scenarios toolbox, the REF2 scenario proposes scientific adjustments for traffic and wood combustion sectors, but it will be possible to adapt this scenario as a service evolution, according to new knowledge for those or other sectors. This scenario is qualified as science-based scenario to reflect the fact that it reports the very last scientific knowledge on emission factors and activity data. So the TNO-MACC emission inventory (REF1) will be kept as the reference scenario to assess the impact of emission reduction strategies. The science-based scenario REF2 should be considered as a complementary information that could help, in some cases to understand discrepancies between modelled forecasts and observations. In the first months of the service implementation, we recommend to not present the REF2 scenario forecast and related maps on the CAMS_71 web site but allow for internal evaluation of the product. The results may be confusing for users not fully aware of the assumptions behind this scenario, and at this stage, it is not possible to anticipate the results we will obtain with CHIMERE CTM. This decision has been welcome and confirmed by the policy users who attended the 1 st CAMS policy user workshop on the 28 th November The few results presented for this scenario raised their interest but bring many questions about the current emission data reported officially by the countries. A testing period of three months was proposed initially; we recommend to extend it until June 2017 (to be able to analyze the model response to the REF2 CAMS71_2016SC1 Green Scenarios Documentations V3 Page 10 of 18

11 scenario during the whole winter period). We will also use evaluation capacities of the CAMS_50 forecasting and analysis system, to assess objectively if this alternative reference scenario can give better results for forecasting PM concentrations than the official one. A synthesis of the results will be proposed by the mid Scenarios related to EU legislation It is important in the perspective of policy decision support, to be able to provide an objective information on the impact of the future mandatory emission control strategies that will apply in Europe thanks to legislation. In the European Union, the National Emission Ceilings Directive (NECD 81/2001/EC) sets ceilings in total emissions of four pollutants: sulfur dioxide (SO2), nitrogen oxides (NOx), Volatile Organic Compounds (VOC) and ammonia. In 2013 the process of revision of the NEC Directive started. It followed the achievement of another important legislative text: the amended Gothenburg Protocol of the Convention on Long Range Transboundary Air Pollution revised in December The Protocol is likely to be ratified by the 51 Parties to the Convention when the NEC Directive concerns the Member States of the European Union. It is important to note that a new ceiling is set in the Gothenburg protocol for an additional pollutant, the fine particulate matter (PM2.5). Since 2013, discussions and negotiations about the level of ambition of the revised NEC Directive developed and an agreement between the European Commission, the European Parliament and the European Council, has just been found the 30 th June This agreement states that: From 2020 to 2029, the national emission ceilings agreed in the Gothenburg Protocol should apply in all Member States In 2030, a new and more ambitious stage should be reached to ensure that mortality of European Citizens due to air pollution should be reduced by 50% compared the situation in to The new NEC Directive referred 2016/2284/EU which abrogates the 2001 Directive, has been finally published on the 14 th December We propose to simulate and forecast every day what would be the levels of ambient ozone, nitrogen dioxide and PM concentrations in 2020 and 2030 with the emission control regulation that should apply with the new NEC Directive. We will build up two scenarios based on the assumptions discussed in the negotiations for the Directive. In September 2016 the assumptions that bear the 2030 objectives were not public. So we used the emission reductions for 2030 as proposed by the European Commission in their Air quality policy package (EC, 2013). These reduction percentages are in turn based on calculations performed by the International Institute on Applied Systems Analysis (IIASA) in support of the implementation of the Thematic Strategy on Air Pollution (TSAP) (EC 2013). Detailed descriptions of the assumptions that bear negotiated control strategies have been elaborated IIASA in numerous reports they published for the European Commission ( (Amann et al. 2013). We have chosen to use the reductions proposed in the 2013 Commission proposal. We proposed to update this scenario with the official agreed reduction percentages once CAMS71_2016SC1 Green Scenarios Documentations V3 Page 11 of 18

12 the negotiated emission reduction commitments are formally released. This will be done by the beginning of 2017 and the scenario will be updated before March The proposed emission reductions are presented as percentages of emission reductions for the five pollutants country by country, compared to the emissions in We will use them as support to distribute the emission reductions sector by sector. The same geographical distribution patterns as for the reference scenarios will be applied. The reduction percentages will be applied to the TNO_MACC-III emission database for the year 2005 which has been developed in the MACC-III project. The two regulatory scenarios will be named NEC2020 and NEC2030 respectively for each targeted year. 4. Sector-oriented scenarios 4.1 Traffic scenarios Traffic scenario aims at evaluating the impact of road transport on air quality. In the former MACC toolbox the traffic scenario was a reduction of 30% of all pollutants emitted by the traffic sector (except resuspension). This is not, of course, realistic and certainly too ambitious compared to what can be done, but: The objective was to illustrate the impact and contribution of road traffic on the air pollutant levels we observe and forecast. A large amount of reduction is more illustrative in that perspective; This percentage of reduction can roughly be considered as representative of the impact of short term actions like alternate traffic schemes; Finally, it will give to the policy maker a kind of upper bound of what can be achieved reducing road traffic emissions. Sometimes the impact will be small compared to the effort requested to reduce the emissions to that level. And sometimes it will be very efficient. This is important to know for the policy maker to take appropriate decisions and manage his communication. For all these reasons we suggest to keep this theoretical scenario (-30% in road traffic emissions) in the list of the CAMS_71 scenarios. This reduction will be applied to the whole European domain to assess where the strategy could be the most efficient depending on the nature of the episode. This scenario will be named STRA30. It may be interesting to compare also with a more realistic scenario that corresponds to actual control strategies of traffic emissions. Generally, such strategies apply to the local (city) scale. They can refer for example to speed limits, or restricted access of heavy vehicles to the city centers, or the famous low emissions zones, protected areas in the city where no vehicles or limited number of vehicles can circulate. TNO worked on such scenarios in the INTERREG project JOAQUIN and this experience allows to define a relevant traffic scenario applied in the main city areas of the domain. However, based on the JOAQUIN experience we do not encourage selecting this scenario. The impact was highly local with very limited impact (< 5%) on the regional scale. Quantification of the impact of low emission CAMS71_2016SC1 Green Scenarios Documentations V3 Page 12 of 18

13 zones demands city-scale modelling and regional scale modelling as done in CAMS_71 is not the right tool for this policy question. 4.2 Industry scenario Although industry sector already reduced its emissions substantially over time to comply with emissions limit values (ELV) set in the legislation, its impact on air quality is still significant and in case of episodes, a supplementary effort could be requested. This is the reason why in the CAMS_71 proposal in response to the ITT, we proposed a scenario reducing by 10% industrial emissions (including the energy combustion and transformation processes). The MACC experience, however, showed that the impact of this reduction is relatively low, and from the discussions that were held in the NEC directive negotiations it was shown that more ambitious targets could be set. So finally we recommend to define an industry scenario abating emissions by 30%. The question of the sectors to which applying this reduction is important. 5 SNAP sectors are potentially concerned: SNAP number S1 S3 S4 S5 S6 Title Combustion in energy and transformation industries Combustion in manufacturing industry Production processes Extraction and distribution of fossil fuels and geothermal energy Solvent use and other product use Complementary analysis by the CAMS_71 team shows that SNAP 5 and 6 do hardly contribute to the selected air pollutants, especially for SO2, NOX and PM (less than 2% of the total emissions). For VOC emissions SNAP 5 and 6 are more important but including too many SNAP sectors makes it less transparent what measure and sector combination is effective This could complicate the interpretation of sectoral importance for air quality. So, we recommend to focus the industry scenario on sectors SNAP 1, 3 and 4 which actually include industry and power generation. The scenario will be named SIND Agriculture scenario Agriculture was a very sensitive stake during the negotiations of the revised NEC directive. The PM episodes that impacted a large part of Europe in spring of the 10 past years highlighted the important role played by ammonia emissions in the chemical formation of secondary inorganic particles. Ammonia in the atmosphere is oxidized and combines with nitrogen oxides to create ammonium nitrate in particular. Ammonium nitrate is the dominant species in PM during spring episodes that develop during the fertilizing period in agricultural countries. Under appropriate meteorological conditions, some exceedances of the daily PM10 limit value can even occur. Over the 20 past years, CAMS71_2016SC1 Green Scenarios Documentations V3 Page 13 of 18

14 although it was in the list of pollutants in the Gothenburg Protocol and the NEC Directive, ammonia has never been drastically reduced in Europe (less than 10% in general). With the increasing concern of policy makers regarding exposure to fine particles, things changed recently. Several studies (Bessagnet et al, 2014) demonstrated the importance of this chemical compound and the nonlinearity of the response, in terms of PM ambient concentrations to ammonia emission reductions. In particular, it is shown that to be effective in Europe, it is necessary to significantly reduce ammonia emissions, by more than 20% compared to the current situation. This is the reason we had in the former MACC toolbox a scenario abating ammonia emissions by 30%, and we propose to keep it for the new tool. It will be referred as the SAGR30 scenario. It should also be noted that an advantage of the -30% for several sectors is that communication of the scenario results is easier, even though in absolute terms the emissions reductions of -30% on e.g. traffic or agriculture are quite different. 4.4 Residential heating The impact of residential heating emissions on PM concentrations in winter time is high. First because meteorological conditions (cold weather, clear sky and thermal inversions) can create situations where the PM emitted cannot disperse and stay stuck close to the ground especially in the cities. And second because when the weather is very cold, use of residential heating and wood burning increases significantly. We have mentioned in the introduction the questions of the big uncertainties that can impact the estimations of emissions from this sector. On the other side, it is expected that development of new efficient stoves (best available technologies) in European countries could help in limiting emissions and the impact of this sector on air quality. In the former MACC toolbox, a theoretical scenario with a 30% emission reduction from residential heating was proposed. For the present we propose to focus on a best available technologies implementation scenario that will simulate use of efficient stoves in the European cities. An indication of the impact of such a scenario can be found in Fountoukis et al. (2014) which showed substantial reductions for several PM components in winter time. The simulations of Fountoukis et al. show a large decrease of fine organic aerosol (more than 60%) in urban and suburban areas during winter and decreases of 30 50% in elemental carbon levels in large parts of Europe. It will be named SRHBAT scenario. 4.5 Combined scenarios To complete the sectoral analysis, we propose to explore two more combined scenarios. Over the last years, development of spring PM episodes in Western Europe with PM10 concentrations values exceeding the limit value of 50 µg/m 3 (daily average) in large parts of countries became a real CAMS71_2016SC1 Green Scenarios Documentations V3 Page 14 of 18

15 issue (see discussion on agriculture scenario). Reducing ammonia emissions is a relevant an important objective, but the physico-chemical processes that explain PM formation from ammonia emissions are also driven by NOx emissions. Those are mainly due to road traffic. This is the reason why it makes sense to evaluate the impact of a combined scenario STRA30 + SAGR30, and to compare it to results related to STRA30 and SAGR30 taken individually. It will demonstrate the complexity of the responses of air pollutant concentration to sectoral emission control policies. This scenario will be named STRAGR30. For the same reasons, it will be instructive to build up a combined scenario with reductions applied in the four activity sectors. Here we have two options: either applying in each sector the theoretical emissions reduction previously defined (-30% emissions for road traffic, industry, agriculture and highest possible emission reduction for residential heating), or evaluating even more ambitious assumptions based on the implementation of best available technologies (BAT) in the various sectors. IIASA elaborated such a scenario, called MTFR (Maximum Technical Feasible Reduction Scenario) which assumes implementation of ambitious end-of-pipe measures to abate emissions. Even if they are not considered yet as appropriate options for operational implementation (mainly because of costs aspects) in the perspective of a what if analysis for policy makers it is interesting to investigate them. This is our recommendation for the second combined scenario that we will call NEC_MTRF. Indeed, the quantitative evaluation of the potential emission reductions by sector and by country with the implementation of BATs is described in the IIASA s TSAP reports considered in by policy makers for the NECD negotiation. CAMS71_2016SC1 Green Scenarios Documentations V3 Page 15 of 18

16 5. References Bessagnet B., Beauchamp, M., Guerreiro, C., de Leeuw, F., Tsyro, S., Colette, A., Meleux, F., Rouïl, L., Ruyssenaars, P., Sauter, F., Velders, G. J. M., Foltescue, V. L., van Aardennee, J. (2014) Can further mitigation of ammonia emissions reduce exceedances of particulate matter air quality standards?, Environmental Science & Policy, Volume 44, Pages Denier Van der Gon H.A.C, R. Bergström, C. Fountoukis, C. Johansson, S.N Pandis, D. simpson, and A. J.H Vissechedijk, 2015, Partculate matter emissions from wood combustion in Europe: revised estimates and an evaluation, Atmos. Chem. Phys., 15, Fountoukis, C., T. Butler, M. G. Lawrence, H.A.C. Denier van der Gon, A. J. H. Visschedijk, P. Charalampidis, C. Pilinis, and S. N. Pandis, Impacts of controlling biomass burning emissions on wintertime carbonaceous aerosol in Europe, Atmospheric Environment, 87, p DOI: Amann M, Borken Kleefeld J, Cofala J, et al. (2014) The Final Policy Scenarios of the EU Clean Air Policy Package. TSAP Report #11. International Institute for Applied Systems Analysis, Laxenburg, Austria. EC (2013) Proposal for a Directive of the European Parliament and of the Council on the reduction of national emissions of certain atmospheric pollutants and amending Directive 2003/35/EC, European Commission (EC), Brussels, Belgium CAMS71_2016SC1 Green Scenarios Documentations V3 Page 16 of 18

17 CAMS71_2016SC1 Green Scenarios Documentations V3 Page 17 of 18

18 ECMWF - Shinfield Park, Reading RG2 9AX, UK Contact: info@copernicus-atmosphere.eu atmosphere.copernicus.eu copernicus.eu ecmwf.int