Round 1

Reviewer 1 Report

Line 44: Missing parentheses

line 48: please explain reasons why have not decreased as much as foreseen. I would suggest citing work like Anenberg that have analyzed the excess real-world emissions from ICE vehicles versus their certification emissions

Anenberg, S., Miller, J., Minjares, R. et al. Impacts and mitigation of excess diesel-related NO_x emissions in 11 major vehicle markets. Nature 545, 467–471 (2017). https://doi.org/10.1038/nature22086

line 51: “among others” I think you are saying among other reasons. It is unclear how much the emissions drop is from fleet turnover versus changes in activity, or a whole host of other reasons. It would be useful to look at this issue more broadly and find a reference of how these emissions have changed over the past decade or 2 and the impact of emission regulations on these changes (you do describe the Euro standards in the following paragraph)

line 91: should be section 4

section 2.1: it would be useful to provide a reference either external or in the supplement information with the table of the emission factors used in this study, differentiating euro standards, vehicle types, etc.

section 2.2: I’m unclear how you are using the emissions data from EDGAR. Above it says it’s for projections (line 97), but this section doesn’t explain that. Rather it says is using aerosol data from this inventory. It is unclear why the study didn’t collect this emissions data as part of the laboratory testing described in section 2.1. Later in the text it seems older vehicle emissions are using EDGAR data

Section 2.3: I would suggest removing the website URLs from the body of the text and just make sure that they are in the references. In addition it is difficult to understand the differences between the 3 scenarios as mentioned in my previous comment, a supplement table that describes the emission factors would be helpful as readers may not know the difference between Euro 6B, 6D-temp, and 6D vehicle emission factors

line 152: the acronym for non-methane VOCs is misspelled

figure 1: the color in figure 1G, does not fit the pattern used in the previous figures, in that 0 is yellow while in all the other graphs it is blue. I understand that this is the outlier in that the max is above 0 significantly, however it is something that might confuse readers.

Line 235: it seems a scenario that does include primary PM emissions would be beneficial. Otherwise, if you have justification to ignore it based on the difference beings small as compared to secondary formation, that should be stated.

Line 257: I’m not clear what the last sentence of this paragraph values mean compared to the values given in the 1^st sentence of the paragraph?

Figure 3: the 50 city chart is quite interesting and nicely presented, the exceedances charts are useful to to give an example of how the changes in vehicle emission standards will impact key polluted cities

line 393: I’m not sure what this sentence means that large reductions in emissions is not fully reflected in the reduction of concentrations. Please clarify

 

Author Response

Modelling the impact of the introduction of the EURO 6d-TEMP/6d regulation for light-duty vehicles on EU air quality.

 

We thank the Editor and reviewers for the constructive comments which have been very helpful in improving the manuscript. Please find below a point-to-point reply to the comments.

 

Editor

Please forgive my ignorance as a non-European editor, but it may be helpful to include a brief clarification as to why the modeling domain is EU27 + Norway, Switzerland, and the United Kingdom. Are there simply insufficient data for, e.g., Albania, Macedonia, and Serbia, or is there some other reason?

We consider EU27 + countries because these countries follow a common trend in the implementation of transport technologies. We added this to the text.

 

 

 

Reviewer 1

Line 44: Missing parentheses

Corrected, thank you.

 

 

line 48: please explain reasons why have not decreased as much as foreseen. I would suggest citing work like Anenberg that have analyzed the excess real-world emissions from ICE vehicles versus their certification emissions

Anenberg, S., Miller, J., Minjares, R. et al. Impacts and mitigation of excess diesel-related NOx emissions in 11 major vehicle markets. Nature 545, 467–471 (2017). https://doi.org/10.1038/nature22086

Thank you for the reference. We added the following to the text:

This could be explained by the fact that the reduction of NOx emissions by diesel cars has shown to be more challenging, due to the difference between real-world NOx emissions and certification limits under Euro 4 and 5 [8].

 

 

line 51: “among others” I think you are saying among other reasons. It is unclear how much the emissions drop is from fleet turnover versus changes in activity, or a whole host of other reasons. It would be useful to look at this issue more broadly and find a reference of how these emissions have changed over the past decade or 2 and the impact of emission regulations on these changes (you do describe the Euro standards in the following paragraph)

Thank you for this comment. We modified text with:

Since 1970, road transport activities increased in Europe by 1.7 times compared to nowadays (2018) and increased by more than 30% over the past 2 decades. Despite the increase of fuel combustion in this sector, most of air pollutant emissions have decreased over the past two decades (with the exception of NH3) thanks to the implementation of emission control technologies. Recent studies by [9, 10] found that in 2017 emissions from road transport were lower than in the previous year. In particular, emissions of nitrogen oxides decreased at a faster rate compared to the decreased activity [12] and specifically by 3%, and PM10 and PM2.5 by 1.4 % and 3.6 %, respectively [5] thanks, among others, to the implementation of the latest emission control technologies in new light-duty vehicles as they entered EU vehicle fleets.

[12] International Energy Agency (IEA) (2019), Energy balance statistics for 1970-2017, http://www.iea.org/, Data release 2019, Available online:  https://www.iea.org/data-and-statistics/data-products (Accessed on 07 April 2022).

 

 

 

line 91: should be section 4

Corrected.

 

 

section 2.1: it would be useful to provide a reference either external or in the supplement information with the table of the emission factors used in this study, differentiating euro standards, vehicle types, etc.

We added to the Electronic Supplement (Table S1) the following information on the road traffic Emission Factors used in this study.

Sector/fuel/type	Substance	implied EF (kg/TJ)
TRO.ROA.DIE.LD+PC	NH3	0.550284988
TRO.ROA.DIE.LD+PC	NMVOC	10.58041207
TRO.ROA.DIE.LD+PC	NOx	256.1300959
TRO.ROA.MOG.LD+PC	NOx	67.45130685
TRO.ROA.MOG.LD+PC	NH3	22.49487648
TRO.ROA.MOG.LD+PC	NMVOC	65.60918433

LD= light duty, PC=passenger cars, DIE=diesel, MOG=motor gasoline

STU emission factors are provided in Table S2.

 

 

section 2.2: I’m unclear how you are using the emissions data from EDGAR. Above it says it’s for projections (line 97), but this section doesn’t explain that. Rather it says is using aerosol data from this inventory. It is unclear why the study didn’t collect this emissions data as part of the laboratory testing described in section 2.1. Later in the text it seems older vehicle emissions are using EDGAR data

 

We correct this and replaced the text with the following:

“The Emissions Database for Global Atmospheric Research (EDGAR) is a global inventory providing greenhouse gas and air pollutant emissions estimates for all countries over the time period 1970 till most recent years, covering all IPCC reporting categories, with the exception of Land Use,  Land Use Change and Forestry (LULUCF). Note that EDGAR follows the EEA Guidelines for reporting, with a highly detailed information/data on fleet composition. More detailed information about the emission inventory is given in [23] and references therein.

In this study, we used the emissions for aerosol and aerosol precursor gases from the EDGAR version 5.0 inventory (available at https://edgar.jrc.ec.europa.eu/dataset_ap50; [9, 20-22] as baseline scenario. Furthermore, we developed different emission scenarios coupling the activity data, technologies and abatement measures implemented in the EDGAR database with the emission factors retrieved from experiments performed at the VELA Laboratory.”

 

 

Section 2.3: I would suggest removing the website URLs from the body of the text and just make sure that they are in the references. In addition it is difficult to understand the differences between the 3 scenarios as mentioned in my previous comment, a supplement table that describes the emission factors would be helpful as readers may not know the difference between Euro 6B, 6D-temp, and 6D vehicle emission factors

As suggested by the Reviewer we removed the URLs and included them in the Reference list. Regarding the Emission Factors, we presented them in the Electronic Supplement.

 

 

line 152: the acronym for non-methane VOCs is misspelled

Corrected.

 

 

figure 1: the color in figure 1G, does not fit the pattern used in the previous figures, in that 0 is yellow while in all the other graphs it is blue. I understand that this is the outlier in that the max is above 0 significantly, however it is something that might confuse readers.

As suggested by the reviewer we changed to colour table for Figure 1g. It now reads as:

 

 

Line 235: it seems a scenario that does include primary PM emissions would be beneficial. Otherwise, if you have justification to ignore it based on the difference beings small as compared to secondary formation, that should be stated.

Primary PM from diesel Euro 6d-TEMP and 6d should be more or less the same as those in COPERT for Euro 5b, and Euro 6b since all diesels have particle filters since 5b. The reviewer is right that primary PM emissions by Euro 6d-TEMP and 6d gasoline are different. In this study, however, we investigated the impact of changing traffic emissions on Secondary Aerosol formation. Therefore, we added to the text that we changed the aerosol precursor emissions between SC1 and SC2, but we kept primary PM emissions the same.

 

 

Line 257: I’m not clear what the last sentence of this paragraph values mean compared to the values given in the 1st sentence of the paragraph?

We mean that the differences between the Reference Case and SC1 in seasonal averaged concentrations are similar to the differences in the yearly averaged NO2 concentrations. We made this clearer in the text.

 

 

Figure 3: the 50 city chart is quite interesting and nicely presented, the exceedances charts are useful to to give an example of how the changes in vehicle emission standards will impact key polluted cities

Thank you.

 

 

line 393: I’m not sure what this sentence means that large reductions in emissions is not fully reflected in the reduction of concentrations. Please clarify

The Reviewer is right. A possible explanation for this is that the emissions of coal-fired power plants and other sources are weakening the effect of the filters in Diesel cars in these countries. We added an explanation to the text together with the reference to the NOx atlas [40]. That is,  “Degrauewe, B., Pisoni, E., Peduzzi, E., De Meij, A., Monforti-Ferrario, F., Bodis, K., Mascherpa, A., Astorga-Llorens, M., Thunis, P and Vignati, E., Urban NO2 Atlas, EUR 29943 EN, Publications Office of the European Union, Luxembourg, 2019, ISBN 978-92-76-10386-8, doi:10.2760/43523, JRC118193.”

 

Author Response File: Author Response.docx

Reviewer 2 Report

The manuscript investigated the impact of the introduction of the EURO 6d-TEMP/6d regulation on air quality. The results are comprehensive and the discussion is adequate. Only one question needs to be revised.

The model needs to be described in details and more parameters should be provided in the manuscript.

Author Response

Modelling the impact of the introduction of the EURO 6d-TEMP/6d regulation for light-duty vehicles on EU air quality.

 

We thank the Editor and reviewers for the constructive comments which have been very helpful in improving the manuscript. Please find below a point-to-point reply to the comments.

 

Editor

Please forgive my ignorance as a non-European editor, but it may be helpful to include a brief clarification as to why the modeling domain is EU27 + Norway, Switzerland, and the United Kingdom. Are there simply insufficient data for, e.g., Albania, Macedonia, and Serbia, or is there some other reason?

We consider EU27 + countries because these countries follow a common trend in the implementation of transport technologies. We added this to the text.

 

 

Reviewer 2

Comments and Suggestions for Authors

General remarks: The manuscripts deal with the environmental impacts of road transport and are in line with the journal aims & scope. The authors attempted to assess the impact of introducing the EURO 6d-TEMP/6d regulation for LDVs on air quality in the EU. In order to estimate the impact of introducing EURO standards, two hypothetical scenarios were simulated. The simulations were carried out using the EDGAR emission inventory database and the EMEP model. The results were compared with a baseline scenario that used 2015 EU vehicle fleet data.

The idea of the paper is quite clear, except for the use of different data sources of emission factors in the simulations. The COPERT emission factors (TIER 2) were used in the reference scenario and in the scenario 1. In scenario 2 emission factors derived from the STU measurements data were used. Why did the authors not use the COPERT emission factors for EURO 6 vehicles (after 2017) in the scenario 2?

The on-road emissions within the EDGAR emission model used in the exercise are based on the COPERT emissions model which matches the EMEP/EEA air pollutant emission inventory guidebook in its 2019 version. Exhaust emissions from light-duty vehicles follow the Tier 3 calculation approach. In the 2019 version, the available choice of emission factors covered up to Euro 6b emissions standard. The inventory guidebook and COPERT were updated in 2021 to cover as well the latest Euro emissions standard, namely Euro 6d-TEMP and Euro 6d. Scenario 2 aimed at studying the effect of the emissions control technologies advancement in vehicles type-approved according to the novel WLTP and RDE procedures as compared to previous Euro 6 vehicles type-approved according to the outdated laboratory NEDC procedure. Considering that the modelling exercise described in the manuscript was performed in 2021 prior to the last update of the inventory guidebook, the authors decided to use empirical emission factors for Euro 6d-TEMP light-duty vehicles as measured in-house with actual vehicles. The following sentence has been added to section 2.1 of the manuscript to make aware to the reader the reasoning behind the choice of emission factors in scenario 2: “Emission factors of Euro 6d-TEMP/6d light-duty vehicles have been derived ad-hoc for the study using data gathered by the STU as the 2019 version of the emission inventory guidebook used within EDGAR did not contain emission factors for such vehicles.”

 

 

The authors claim that the use of STU factors is a novel contribution. The authors should demonstrate that using the emission factors derived from the data collected by measuring emissions for only thirty vehicles is better than using the COPERT emission factors which correspond to the average emission value of a large number of cars. Moreover, the COPERT emission factors were estimated not only from the measurement data recorded during regulatory tests.

We thank the reviewer for his comment. As indicated in the previous comment, the emission factors from STU used in Scenario 2 were considered by the authors the best available source of data in 2021 at the time when the modelling exercise was conducted. It is true that STU emission factors are based on a limited number of vehicles, but at the time of the modelling exercise such a dataset constituted a unique source of liable emission factors. The emission factors were derived for different powertrains using vehicles from different segments, manufacturers, and after-treatment systems. STU emissions were measured with laboratory grade instrumentation on chassis dynamometers that simulate the real road loads of the vehicles according to the update dyno settings in the WLTP. It is therefore assumed that the exhaust emissions are an accurate representation of the emissions in real life operation. On the other hand, it is unclear how the Euro 6d-TEMP and Euro 6d emission factors in COPERT have been produced as there is a lack of publicly available documentation of the project (how many vehicles were tested? Which characteristics had the vehicles? Which driving conditions were covered? Which is the uncertainty of the measurements in case PEMS were used? It is still unclear whether the emission factors in the guidebook, revision of 2021, are mature enough as for example NOx from Euro 6d-TEMP LCV are 3-4 times higher as the ones reported in the available literature (see Figure below). At the same time, we are not claiming that one dataset is better than the other, but simply using both to provide a more complete picture of the issue. Title x-axis is velocity (km/hr).

 

 

In my opinion, the emission databases used in simulations are not comparable in terms of credibility. Therefore the scope of the presented differences in the total annual emissions may raise doubts. The analysis of the results obtained for the EU vehicle fleet in 2015 should be supported by fleet composition statistics for the same year instead of 2017 data (lines 190-193).

We know that COPERT is the de-facto standard on emission factors, and this is why we propose to use it in this study. Our idea was to complement COPERT results with updated emission factors not yet available within COPERT (based on STU emission factors). As 6dtemp was not available in COPERT (was up to 6a and 6b available), when the study was done, we used the emission factors for 6dtemp by STU.

For what regards credibility, it is worth mentioning that the STU chairs the ERMES group (European Research for Mobile Emission Sources) and provides emission factors to it in a regular basis based on its experimental activity in the Vehicle Emissions Laboratories. The STU has supported technically the development and implementation of a number of emissions regulations at EU and United Nations level and STU personnel chairs a number of technical groups on vehicle emissions. In addition, the STU has as well a number of relevant scientific contributions that demonstrate its technical capacity to generate emission factors:

• https://doi.org/10.1016/j.atmosenv.2013.01.006
• https://doi.org/10.1016/j.atmosenv.2012.09.062
• https://www.sciencedirect.com/science/article/pii/S0048969713010814
• https://doi.org/10.1021/es2008424
• https://doi.org/10.3389/fenvs.2015.00082
• https://doi.org/10.3390/atmos10050243
• https://doi.org/10.1186/s12302-020-00407-5
• https://doi.org/10.1016/j.envres.2019.108572
• https://doi.org/10.3390/ijerph15020304

 

 

In the part concerning the results for the pollutant concentration (3.2) three main periods are considered. Unfortunately, there is no explanation as to how the authors take into account different traffic pattern (e.g. given total trip length, number of trips) of LDVs during these periods and changes in emission factors depending on the different share of the cold/cool emission. The latter is especially important when discussing air pollution levels in cities.

In both scenarios, only the emission factors are changed. The activity data is kept constant. Changes in emission rates are relative to the change in the emission factors only. The text in the manuscript has been updated to explicitly indicate the former.

Emissions are estimated in EDGAR using a bottom up methodology, following the methodology described in the IPCC (Intergovernmental Panel for Climate Change) and EEA (European Environmental Agency) reporting guidelines*. Fuel consumption statistics are based on the figures from IEA (International Energy Agency)’s energy balances**, which are disaggregated according to vehicle category and standards, using statistics from vehicle population and yearly mileage provided by EMISIA*** and Tier 3 fuel economy indicators. The latter were quantified employing the correlations, based on COPERT model, included as supplementary information in the original 2019 release of the EEA Guidelines. These indicators account for fractions of mileage driven in different traffic settings, average speeds, the influence of some  flue gas treatments, and assumptions on ‘cold start ‘and ‘warm engines’. Tier 2 exhaust emission factors in the guidelines, expressed in g/km, were converted in g/TJ using the afore mentioned Tier 3 fuel economy indicators.

The three scenarios, discussed in this paper, were developed by considering the fuel consumption for different vehicle categories modelled for 2015 and assumptions on the widespread of emission standards among these categories. In the case of the Reference Scenario (REF), emissions are estimated for the actual 2015 fleet. Scenario 1 (SC1) presents emissions in a hypothetical world in which all the fleet were made up by units which followed the EURO 6 standards existing in 2015 (EURO 6 b). In the original 2019 guidelines, emission factors for 6b were equal to 6d, and did not account for on road measurements. Scenario 2 (SC2) builds on SC1 but using emission factors, which account for both cycle test and real driving conditions.

Text in the manuscript has been updated to better explain the philosophy behind the scenario formulation.

 

* IPCC, 1997. Houghton J T, Meira Filho L G, Lim B, Treanton,K, Mamaty I, Bonduki Y, Griggs, D J, Callander, BA (Eds.), Revised 1996 IPCC Guidelines for National Greenhouse Gas Inventories, Vols. 1-3.

EEA (European Environmental Agency). EMEP/EEA air pollutant emission inventory guidebook 2019. Available on line at: https://www.eea.europa.eu/publications/emep-eea-guidebook-2019.

** IEA (International Energy Agency), 2017. Energy balance statistics for 1970-2015, http://www.iea.org/

*** EMISIA. COPERT data: Ready to go vehicle fleet , activity, emissions and energy consumption data for the EU28 countries (plus 5 additional countries).

 

 

Considering the above, I would suggest to select one emission database for research and perform calculations again. If the authors were to use the STU emission factors for all scenarios, the necessary explanation should be provided in the manuscript and not in the Electronic Supplement (by the way, the supplement has not been made available).

We respectfully disagree with this suggestion. We believe that the three scenarios have features that make them unique and relevant for readers. Particularly, the scenarios allow to appreciate the effects of key sources of uncertainty when modelling emissions from road transport. The differences between REF and SC1 are consequence of different assumptions for fleet whilst SC1 and SC2 highlight the impact on emissions and concentration of using emission factors for vehicles complying with the latest emission standards which account for real driving conditions. We think that the insights of the comparison exercise can be of significance for the research community as a way to weigh the effect of uncertainties in different inputs for the emission modelling.

 

 

Detailed remarks:

double or multiple references are not necessary in many places of the manuscript (e.g. in lines 32, 36,  65, 67, 115, 116, 433); abbreviations should be explained when first used (see lines 110, 118)

We believe that’s justified to have in some places more than one reference to underline our findings in this manuscript. Recent articles in Applied Science use 4 or more references in a sentence.

Regarding the abbreviations we corrected this as suggested. Thank you.

 

 

more explanation should be given in the methodology section, e.g. about the emission factors (Tier method)

This is described in the Electronic supplement, which unfortunately was not made available during the first review process.

 

 

reference should be numbered and numbers should be placed in square brackets (see line 135)

Done, corrected.

 

 

Website should be presented as single reference  (see lines 56, 126, 138, 147, 193 etc.). one abbreviation NMVOC or VOC should be used

We went carefully through the text and made the necessary changes and put the URLs in the Reference list. Reviewer 3 had a similar comment.

 

 

in line 183 the authors say that in general NH3 emission is higher in SC1 than in SC2, but in line 204 it is stated that in general NH3 emission in SC1 is lower than in SC2

NH3 emissions by SC1 are lower than by the Reference Case as mentioned in line 183, but as the reviewer correctly remarks, NH3 emissions by SC1 are lower than SC2. This can be explained by the implementation of the after treatment system in diesel vehicles in SC2, namely the selective catalytic reduction (SCR) catalysts in which NH3 can be produced as a by-product. As mentioned in the text, vehicles in SC2 need to comply with stringent NOx limits on the road, therefore vehicles in SC2 make use of the SCR catalysts. A consequence of the use of the SCR catalyst is that more NH3 is emitted by SC2 than by the Euro 6 pre-RDE vehicles (as in SC1).

 

 

lines 197-198: “the mass emitted in Scenario 1 is lower than in the Reference case SC1” ?

Corrected.

 

 

line 451 : “which the WHO As there” ?

Corrected, thank you. The sentence now reads as: “In order to understand how often people are exposed to high PM levels, we look at the number of days in the year for which the WHO PM2.5 daily limit value are exceeded.”

 

Author Response File: Author Response.docx

Reviewer 3 Report

General remarks:

The manuscripts deal with the environmental impacts of road transport and are in line with the journal aims & scope.

The authors attempted to assess the impact of introducing the EURO 6d-TEMP/6d regulation for LDVs on air quality in the EU. In order to estimate the impact of introducing EURO standards, two hypothetical scenarios were simulated. The simulations were carried out using the EDGAR emission inventory database and the EMEP model. The results were compared with a baseline scenario that used 2015 EU vehicle fleet data.

The idea of the paper is quite clear, except for the use of different data sources of emission factors in the simulations. The COPERT emission factors (TIER 2) were used in the reference scenario and in the scenario 1. In scenario 2 emission factors derived from the STU measurements data were used. Why did the authors not use the COPERT emission factors for EURO 6 vehicles (after 2017) in the scenario 2?   

The authors claim that the use of STU factors is a novel contribution. The authors should demonstrate that using the emission factors derived from the data collected by measuring emissions for only thirty vehicles is better than using the COPERT emission factors which correspond to the average emission value of a large number of cars. Moreover, the COPERT emission factors were estimated not only from the measurement data recorded during regulatory tests.

In my opinion, the emission databases used in simulations are not comparable in terms of credibility. Therefore the scope of the presented differences in the total annual emissions may raise doubts. The analysis of the results obtained for the EU vehicle fleet in 2015 should be supported by fleet composition statistics for the same year instead of 2017 data (lines 190-193).

In the part concerning the results for the pollutant concentration (3.2) three main periods are considered. Unfortunately, there is no  explanation as to how the authors take into account different traffic pattern (e.g. given total trip length, number of trips) of LDVs during these periods and changes in emission factors depending on the different share of the cold/cool emission. The later is especially important when discussing air pollution levels in cities.  

Considering the above, I would suggest to select one emission database for research and perform calculations again. If the authors were to use the STU emission factors for all scenarios, the necessary explanation should be provided in the manuscript and not in the Electronic Supplement (by the way, the supplement has not been made available).

 

Detailed remarks:

• double or multiple references are not necessary in many places of the manuscript (e.g. in lines 32, 36,  65, 67, 115, 116, 433)
• abbreviations should be explained when first used (see lines 110, 118)
• more explanation should be given in the methodology section, e.g. about the emission factors (Tier method)
• reference should be numbered and numbers should be placed in square brackets (see line 135)
• Website should be presented as single reference  (see lines 56, 126, 138, 147, 193 etc.)
• one abbreviation NMVOC or VOC should be used
• in line 183 the authors say that in general NH3 emission is higher in SC1 than in SC2, but in line 204 it is stated that in general NH3 emission in SC1 is lower than in SC2 
• lines 197-198: “the mass emitted in Scenario 1 is lower than in the Reference case SC1” ?
• line 451 : “which the WHO As there” ?

 

Author Response

Modelling the impact of the introduction of the EURO 6d-TEMP/6d regulation for light-duty vehicles on EU air quality.

 

We thank the Editor and reviewers for the constructive comments which have been very helpful in improving the manuscript. Please find below a point-to-point reply to the comments.

 

Editor

Please forgive my ignorance as a non-European editor, but it may be helpful to include a brief clarification as to why the modeling domain is EU27 + Norway, Switzerland, and the United Kingdom. Are there simply insufficient data for, e.g., Albania, Macedonia, and Serbia, or is there some other reason?

We consider EU27 + countries because these countries follow a common trend in the implementation of transport technologies. We added this to the text.

 

 

Reviewer 3

General remarks: The manuscripts deal with the environmental impacts of road transport and are in line with the journal aims & scope. The authors attempted to assess the impact of introducing the EURO 6d-TEMP/6d regulation for LDVs on air quality in the EU. In order to estimate the impact of introducing EURO standards, two hypothetical scenarios were simulated. The simulations were carried out using the EDGAR emission inventory database and the EMEP model. The results were compared with a baseline scenario that used 2015 EU vehicle fleet data.

The idea of the paper is quite clear, except for the use of different data sources of emission factors in the simulations. The COPERT emission factors (TIER 2) were used in the reference scenario and in the scenario 1. In scenario 2 emission factors derived from the STU measurements data were used. Why did the authors not use the COPERT emission factors for EURO 6 vehicles (after 2017) in the scenario 2?

The on-road emissions within the EDGAR emission model used in the exercise are based on the COPERT emissions model which matches the EMEP/EEA air pollutant emission inventory guidebook in its 2019 version. Exhaust emissions from light-duty vehicles follow the Tier 3 calculation approach. In the 2019 version, the available choice of emission factors covered up to Euro 6b emissions standard. The inventory guidebook and COPERT were updated in 2021 to cover as well the latest Euro emissions standard, namely Euro 6d-TEMP and Euro 6d. Scenario 2 aimed at studying the effect of the emissions control technologies advancement in vehicles type-approved according to the novel WLTP and RDE procedures as compared to previous Euro 6 vehicles type-approved according to the outdated laboratory NEDC procedure. Considering that the modelling exercise described in the manuscript was performed in 2021 prior to the last update of the inventory guidebook, the authors decided to use empirical emission factors for Euro 6d-TEMP light-duty vehicles as measured in-house with actual vehicles. The following sentence has been added to section 2.1 of the manuscript to make aware to the reader the reasoning behind the choice of emission factors in scenario 2: “Emission factors of Euro 6d-TEMP/6d light-duty vehicles have been derived ad-hoc for the study using data gathered by the STU as the 2019 version of the emission inventory guidebook used within EDGAR did not contain emission factors for such vehicles.”

 

 

The authors claim that the use of STU factors is a novel contribution. The authors should demonstrate that using the emission factors derived from the data collected by measuring emissions for only thirty vehicles is better than using the COPERT emission factors which correspond to the average emission value of a large number of cars. Moreover, the COPERT emission factors were estimated not only from the measurement data recorded during regulatory tests.

We thank the reviewer for his comment. As indicated in the previous comment, the emission factors from STU used in Scenario 2 were considered by the authors the best available source of data in 2021 at the time when the modelling exercise was conducted. It is true that STU emission factors are based on a limited number of vehicles, but at the time of the modelling exercise such a dataset constituted a unique source of liable emission factors. The emission factors were derived for different powertrains using vehicles from different segments, manufacturers, and after-treatment systems. STU emissions were measured with laboratory grade instrumentation on chassis dynamometers that simulate the real road loads of the vehicles according to the update dyno settings in the WLTP. It is therefore assumed that the exhaust emissions are an accurate representation of the emissions in real life operation. On the other hand, it is unclear how the Euro 6d-TEMP and Euro 6d emission factors in COPERT have been produced as there is a lack of publicly available documentation of the project (how many vehicles were tested? Which characteristics had the vehicles? Which driving conditions were covered? Which is the uncertainty of the measurements in case PEMS were used? It is still unclear whether the emission factors in the guidebook, revision of 2021, are mature enough as for example NOx from Euro 6d-TEMP LCV are 3-4 times higher as the ones reported in the available literature (see Figure below). At the same time, we are not claiming that one dataset is better than the other, but simply using both to provide a more complete picture of the issue. Title x-axis is velocity (km/hr).

 

 

In my opinion, the emission databases used in simulations are not comparable in terms of credibility. Therefore the scope of the presented differences in the total annual emissions may raise doubts. The analysis of the results obtained for the EU vehicle fleet in 2015 should be supported by fleet composition statistics for the same year instead of 2017 data (lines 190-193).

We know that COPERT is the de-facto standard on emission factors, and this is why we propose to use it in this study. Our idea was to complement COPERT results with updated emission factors not yet available within COPERT (based on STU emission factors). As 6dtemp was not available in COPERT (was up to 6a and 6b available), when the study was done, we used the emission factors for 6dtemp by STU.

For what regards credibility, it is worth mentioning that the STU chairs the ERMES group (European Research for Mobile Emission Sources) and provides emission factors to it in a regular basis based on its experimental activity in the Vehicle Emissions Laboratories. The STU has supported technically the development and implementation of a number of emissions regulations at EU and United Nations level and STU personnel chairs a number of technical groups on vehicle emissions. In addition, the STU has as well a number of relevant scientific contributions that demonstrate its technical capacity to generate emission factors:

• https://doi.org/10.1016/j.atmosenv.2013.01.006
• https://doi.org/10.1016/j.atmosenv.2012.09.062
• https://www.sciencedirect.com/science/article/pii/S0048969713010814
• https://doi.org/10.1021/es2008424
• https://doi.org/10.3389/fenvs.2015.00082
• https://doi.org/10.3390/atmos10050243
• https://doi.org/10.1186/s12302-020-00407-5
• https://doi.org/10.1016/j.envres.2019.108572
• https://doi.org/10.3390/ijerph15020304

 

 

In the part concerning the results for the pollutant concentration (3.2) three main periods are considered. Unfortunately, there is no explanation as to how the authors take into account different traffic pattern (e.g. given total trip length, number of trips) of LDVs during these periods and changes in emission factors depending on the different share of the cold/cool emission. The latter is especially important when discussing air pollution levels in cities.

In both scenarios, only the emission factors are changed. The activity data is kept constant. Changes in emission rates are relative to the change in the emission factors only. The text in the manuscript has been updated to explicitly indicate the former.

Emissions are estimated in EDGAR using a bottom up methodology, following the methodology described in the IPCC (Intergovernmental Panel for Climate Change) and EEA (European Environmental Agency) reporting guidelines*. Fuel consumption statistics are based on the figures from IEA (International Energy Agency)’s energy balances**, which are disaggregated according to vehicle category and standards, using statistics from vehicle population and yearly mileage provided by EMISIA*** and Tier 3 fuel economy indicators. The latter were quantified employing the correlations, based on COPERT model, included as supplementary information in the original 2019 release of the EEA Guidelines. These indicators account for fractions of mileage driven in different traffic settings, average speeds, the influence of some  flue gas treatments, and assumptions on ‘cold start ‘and ‘warm engines’. Tier 2 exhaust emission factors in the guidelines, expressed in g/km, were converted in g/TJ using the afore mentioned Tier 3 fuel economy indicators.

The three scenarios, discussed in this paper, were developed by considering the fuel consumption for different vehicle categories modelled for 2015 and assumptions on the widespread of emission standards among these categories. In the case of the Reference Scenario (REF), emissions are estimated for the actual 2015 fleet. Scenario 1 (SC1) presents emissions in a hypothetical world in which all the fleet were made up by units which followed the EURO 6 standards existing in 2015 (EURO 6 b). In the original 2019 guidelines, emission factors for 6b were equal to 6d, and did not account for on road measurements. Scenario 2 (SC2) builds on SC1 but using emission factors, which account for both cycle test and real driving conditions.

Text in the manuscript has been updated to better explain the philosophy behind the scenario formulation.

 

* IPCC, 1997. Houghton J T, Meira Filho L G, Lim B, Treanton,K, Mamaty I, Bonduki Y, Griggs, D J, Callander, BA (Eds.), Revised 1996 IPCC Guidelines for National Greenhouse Gas Inventories, Vols. 1-3.

EEA (European Environmental Agency). EMEP/EEA air pollutant emission inventory guidebook 2019. Available on line at: https://www.eea.europa.eu/publications/emep-eea-guidebook-2019.

** IEA (International Energy Agency), 2017. Energy balance statistics for 1970-2015, http://www.iea.org/

*** EMISIA. COPERT data: Ready to go vehicle fleet , activity, emissions and energy consumption data for the EU28 countries (plus 5 additional countries).

 

 

Considering the above, I would suggest to select one emission database for research and perform calculations again. If the authors were to use the STU emission factors for all scenarios, the necessary explanation should be provided in the manuscript and not in the Electronic Supplement (by the way, the supplement has not been made available).

We respectfully disagree with this suggestion. We believe that the three scenarios have features that make them unique and relevant for readers. Particularly, the scenarios allow to appreciate the effects of key sources of uncertainty when modelling emissions from road transport. The differences between REF and SC1 are consequence of different assumptions for fleet whilst SC1 and SC2 highlight the impact on emissions and concentration of using emission factors for vehicles complying with the latest emission standards which account for real driving conditions. We think that the insights of the comparison exercise can be of significance for the research community as a way to weigh the effect of uncertainties in different inputs for the emission modelling.

 

 

Detailed remarks:

double or multiple references are not necessary in many places of the manuscript (e.g. in lines 32, 36,  65, 67, 115, 116, 433); abbreviations should be explained when first used (see lines 110, 118)

We believe that’s justified to have in some places more than one reference to underline our findings in this manuscript. Recent articles in Applied Science use 4 or more references in a sentence.

Regarding the abbreviations we corrected this as suggested. Thank you.

 

 

more explanation should be given in the methodology section, e.g. about the emission factors (Tier method)

This is described in the Electronic supplement, which unfortunately was not made available during the first review process.

 

 

reference should be numbered and numbers should be placed in square brackets (see line 135)

Done, corrected.

 

 

Website should be presented as single reference  (see lines 56, 126, 138, 147, 193 etc.). one abbreviation NMVOC or VOC should be used

We went carefully through the text and made the necessary changes and put the URLs in the Reference list. Reviewer 3 had a similar comment.

 

 

in line 183 the authors say that in general NH3 emission is higher in SC1 than in SC2, but in line 204 it is stated that in general NH3 emission in SC1 is lower than in SC2

NH3 emissions by SC1 are lower than by the Reference Case as mentioned in line 183, but as the reviewer correctly remarks, NH3 emissions by SC1 are lower than SC2. This can be explained by the implementation of the after treatment system in diesel vehicles in SC2, namely the selective catalytic reduction (SCR) catalysts in which NH3 can be produced as a by-product. As mentioned in the text, vehicles in SC2 need to comply with stringent NOx limits on the road, therefore vehicles in SC2 make use of the SCR catalysts. A consequence of the use of the SCR catalyst is that more NH3 is emitted by SC2 than by the Euro 6 pre-RDE vehicles (as in SC1).

 

 

lines 197-198: “the mass emitted in Scenario 1 is lower than in the Reference case SC1” ?

Corrected.

 

 

line 451 : “which the WHO As there” ?

Corrected, thank you. The sentence now reads as: “In order to understand how often people are exposed to high PM levels, we look at the number of days in the year for which the WHO PM2.5 daily limit value are exceeded.”

Author Response File: Author Response.docx

Round 2

Reviewer 3 Report

Thank you for the detailed explanations.