Next Article in Journal
A Sustainability Assessment of the Greenseeker N Management Tool: A Lysimetric Experiment on Barley
Next Article in Special Issue
Environmental Impact Assessment of the Transportation Sector and Hybrid Vehicle Implications in Palestine
Previous Article in Journal
Analysis of Urban Morphological Effect on the Microclimate of the Urban Residential Area of Kampung Baru in Kuala Lumpur Using a Geospatial Approach
Previous Article in Special Issue
Measuring System-Level Impacts of Corporate Mobility as a Service (CMaaS) Based on Empirical Evidence
 
 
Article
Peer-Review Record

Method for a Multi-Vehicle, Simulation-Based Life Cycle Assessment and Application to Berlin’s Motorized Individual Transport

Sustainability 2020, 12(18), 7302; https://doi.org/10.3390/su12187302
by Anne Magdalene Syré *, Florian Heining and Dietmar Göhlich
Reviewer 1: Anonymous
Reviewer 2: Anonymous
Reviewer 3: Anonymous
Sustainability 2020, 12(18), 7302; https://doi.org/10.3390/su12187302
Submission received: 27 July 2020 / Revised: 3 September 2020 / Accepted: 3 September 2020 / Published: 6 September 2020

Round 1

Reviewer 1 Report

The Authors did a great research devoted to LCA of BEV and ICEV vehicles. A good review of the literature and comparison of the results of the presented method with the studies of other authors. However, there are some limitations and recommendations:

  1. In the introduction part, Figure 1 needs more explanation. Moreover, Moreover, there is no clear transition from the Figure to the main goal of the paper. The main goal and authors' main contributions should be in the end of the introduction. Please highlight them. 
  2. The figures are more readable and understandable when they are colored. Please use color Figure 2,3, 5-10, 12, 14, 16, 18.
  3. In 2.2. subsection there is a grammatical mistake in the title.
  4. There is a grammatical mistake in line 145.
  5.  Have you used specific equations in your environmental impact assessment? For example, the equation for reducing CO2 emissions? There are no formulas in your article. If yes, please highlight the references. 

 

Author Response

Dear Reviewer,

Thank you for your valuable feedback! We made the following changes according to your suggestions (All lines refer to the revised version in which the changes are not displayed (MS Word: “Track Changes” not enabled)):

  1. We reworked the introduction. We included a detailed description of the data presented in Figure 1 (lines 57 – 91). Furthermore, we emphasized the goal of our study (lines 111 - 137). The author's contributions are summarized at the end of the publication (before references).
  2. We colored all figures.
  3. We corrected the mistake
  4. We corrected the mistake
  5. We used the impact assessment methods provided by Ecoinvent (compare section 2.2.2.) and didn't use additional formulas.

Kind Regards,

The authors

 

Reviewer 2 Report

The methodology need to be more clearly and more sintetically presented, because the narative style is not appropriate.

Thus, the authors are advised to use a flowchart of all actions/steps in their methodology for the complex comparisons of different emissions in different conditions of operation for different type of passenger vehicles, and for the entire life cicle of them.

Moreover, for each step in this methodology all the used references need to be more emphasised as long as the results depend on specific previous research.

Author Response

Dear Reviewer,

Thank you for your valuable feedback! We made the following changes according to your suggestions (All lines refer to the revised version in which the changes are not displayed (MS Word: “Track Changes” not enabled)):

We reworked our introduction and introduced our methodology more clearly (line 111 – 143).

We added a flow-chart (line 144) and summarize the investigated cases before presenting the results (line 370 – 376).

Additionally, we made a couple of changes in section 2. Method to emphasize the used research.

Kind Regards,

The authors

Reviewer 3 Report

In the first part the article compares LCA results for Battery electric vehicles (BEV) and internal combustion engine vehicles (ICEV). In the second part the LCA results for different vehicle classes are used for calculating overall environmental impacts scaled up to the vehicle fleet (supposedly) for Berlin/Brandenburg in Germany. The first impression suggests that it is a very interesting topic but unfortunately the manuscript lacks clarity in describing the various assumptions for the investigated cases. In the present state the study is not very useful for the reader. Although it seems everything is explained it is very difficult to follow and in the end it remains for example unclear which vehicle distribution data is use for the study. The text (chapter 2.1.1) talks about vehicle distribution in Berlin/Brandenburg but then a diagram (fig.2) presents the German statistical data. Further the reference year used is unclear (is it 2017 or 2018 or extrapolated to 2050?) because it is not given for the later used scenarios.

Further it remains unclear according to which criterion the classification of vehicle classes have been done (tab.2). Because it seems strange that a “large capacity van” classified as a medium car and a “mid-range car” is classified as a large vehicle. Is it related to motor size or vehicle size? The criterion should be mentioned.

Viewing the results section it remains unclear why a rather complicated simulation model (MATsim) is used to model the daily mileage of the vehicle fleet. The results described very detailed according to the different LCA results but the differentiations related to impacts resulting from the composition of the vehicle fleet are very limited. It seems that it would have been also sufficient to take statistical data about vehicle life times and their mileage to receive the same depth of results related to the vehicle fleet.

The concept of the study is not clear. In the beginning it suggests that also other means of transport are considered (line 123 and 126) but it is not sufficiently defined what is understood by the term “transport system” in this study. The aim of the study suggests a new approach to calculate the fleet impacts, but a large part of the results section just presents the comparison of LCA results of ICEV and BEV. Since this results were just more or less reproduced from previous LCAs (mainly Ecoinvent data) nothing new is presented here. More effort could have been put in the fleet modelling and analysis but not so much for calculating the LCA. What is presented here, is just a repetition of already existing data, scaled up to a (not clear) fleet model of Berlin/Brandenburg. The main short coming is that it is difficult to understand, which assumptions have been made precisely for the different scenarios. The reader would not be able to reproduce the results. The text lacks at many points differentiated descriptions of the used input data. Also the discussion is sometimes difficult to follow because it is not precisely described which scenarios are compared with each other. The study should not be accepted without major changes. See detailed comments in the following for further illustration of the mentioned points (numbers refer to the respective lines):

58-61: How these examples are validating the statement in line 56? Are these examples giving a solution or an example for the stated problem? In the end also the study here is “just” relying on Ecoinvent data.

65-66: How this can be stated at this point? So far no results have been presented. It is not even clear at this point what the aim of the study is. Further, the Figure 1 is displaying a selection of results from various studies. How it was made sure that these studies are comparable e.g. in terms of vehicle mileage, system boundaries etc.?

74-79: The aims suggest that the modelling and presentation of LCA results in this study is related to the vehicle fleet. But in the results section large parts are covered by the comparison of LCA results for different vehicle types. Also the literature review in chapter 1 relates in large parts to study comparing BEV and ICEV. This doesn’t fit and confuses the reader.

Chapter 2: order of sub chapters is a little confusing. E.g. 2.1 would fit better after 2.2, because it should be described in the beginning what the subject of the investigation. Chapter 2.2 should be placed at the beginning of the chapter to get to know first the goal and scope of the study.

91: and figure 2: why is this graph displaying German statistics and not the one of Berlin/Brandenburg which seems to be the subject of the study?

95-96: Why is this fact highlighted? Is this data used for the case study (Berlin/Brandenburg) or why it is reported here?

Figure 2, labelling: it should be mentioned that the dark grey bars are showing the total vehicle stock and which types of vehicles (all or just ICEV?)

100: Reference for MATSim is missing

107: what are “10%” related to? not clear here

108: Is it just the road? Because above also public transport is mentioned (e.g. S-Bahn and U-Bahn)

109: it is not quite clear why map data is use additionally if the network covers entire road network of Berlin and Brandenburg.

111: Does the agents daily schedules also include the choice of transport options? Not clear here

119: impact categories covered by LCA or what is meant here?

120-121: what is meant with the text in brackets? ("compare" ?). not clear

122-123: which data is used to model the transport system of today and the future?

123: which vehicles are considered in the “transport system”? also public transport car sharing, car rent, commercial vehicles? Not clear

126: “transport system level” suggests, all means of transport are considered. It should be specified in the text, what is covered.

136-140: difficult to follow, because it seems this part is in a differnt order than it should be. 136 -140 seem to be more suitable in the beginning of the chapter, lines 133 to 136 could follow.

143-146: why then such a complicated model is used to model the daily mileage if just all days of the week are considered the same and the weekends are simply reduced by a factor. Then it would have been also sufficient to take statistical data about vehicle life times and mileage to get an information about the driven vehicle km. The concept of the study is not clear. In the beginning it suggests that also other means of transport are considered.

149: why 300%? Doesn't makes sense - just a few vehicles might be able to drice 619 000 km.

157-158: add “for LCA”. This phrase otherwise suggests, Ecoinvent is used to define vehicle classes.

table 2: not clear how (according to which criterion) these classes where defined - why is a large capacity van classified as a medium car and a mid-range car is a large vehicle. Is it related to motor size or vehicle size? The criterion should be mentioned.

163: what is the vehicle body and why the masses are differing so much? What is the remaining part? the engine, brakes, tires?  Also an electric vehicle has an engine and ICEVs have a small battery too. All this is not covered in the table 3 and in the text.

165: there is no drive train share displayed in the table 3

166-167: how many lightweight components in each case? this is not displayed anywhere- the reader will not look it up in reference 10

171-172: 100%? The total vehicle is produced with secondary materials? This cannot be true. There are a lot of plastic and rubber parts - might not be secondary materials.

172-173: what is the criterion for cut-off and what does it mean in tis case? Which parts are not considered?

179: how the values have been adapted? to which factor it was referred (motor size? vehicle size?)?

182: “from“ instead of „by“

192-195: how this was considered if it is stated at this point? If it doesn't play a role because the Ecoinvent dataset is used anyway, there is no need to refer to this study at this point. Would better fit to the discussion.

Table 6: why isn't the consumption for BEV and ICEV displayed in the same unit? It is not comparable for the reader without knowing the underlying assumed energy content of gasoline and diesel.

205: if the Case study is Berlin, why not the Berlin electricity mix is used?

216: above it was stated that the impacts of a future vehicle fleet (completely electrified) is also modeled. Which is the reference year and which electricity mix is assumed for it?

223: is 2050 the reference year? or is it just used to have a scenario for 100% renewable power mix? Because if 2050 will be the reference year also for a full electric vehicle fleet, efficiency gains and technology improvements have to be considered in modelling and assessing BEV manufacturing and use within the LCA. But it seems here the present (or more or less already outdated) ecoinvent data was used for modelling the present state and the future

235: what do the reference in brackets mean? very confusing. please make an entry in the list of references if it is another study or refer to a chapter number, if this is a reference to another chapter

239: why vehicle segments and size categories then have been defined before?

240-245: it is not clear what have been done here. It should be sufficiently explained which data is modified and how.

252-254: which emission factors have been used? could be displayed in a table. It is not clear what have been done exactly.

286-287: it is still not clear why the MATsim Scenario have been used at all, if it is not providing a complete picture of the Berlin traffic.

305-306: which grid mix was considered for the results in fig 3?

365-366: not realy comparable - figures in fig 10 are 1000 times higher (since given in kg)

366-367: it should be added in this phrase that is is compared with the ICEV base case.

368-369: this doesn't seem to be logic. Because if vehicles don't run, the proportion of fixed costs or emissions from production etc. increases, because they are allocated to a lower mileage.

371: what is the distribution case for ICEV and BEV?? 100% ICEV or BEV respectively? not clearly described before

371: projected to which year? just 2017? or in the future?

372-373: emissions of ICEV and BEV cases are compared to what?

374: please insert "chapter" or chapter number or at least quotation marks in this remarks. It is very confusing if there is just written "compare".

377-378: the breakeven between which options? please describe precisely

378-380: this is everything which is found in comparing transport system GWP? It is not sufficiently described which cases are advantageous and which would e.g. be the optimal share of BEV and when this would be reached etc.

figure 10 (and following figures): what is meant by the category “transport” in the case of ICEV? does it refer to battery logistics or logistics for other components? This is not clear.

441: what is the BEV base case? it should be made clearer before. if it is in table 9 the column "Distribution BEV 2017" this should be named as "base case" otherwise it is too confusing for the reader

444: smaller values in which impact categories?

453-454: the case has only 27% small vehicles and much higher shares of medium and small vehicles - and other shares haven't been investigated - how can this statement be made? it is not backed up by any results from this study

499-501: wasn't that clear before? This study doesn't seem to add many new results. Just two options have been compared: fleet without BEV and fleet with BEV and additionally BEV fleet fuelled with renewable electricity. For sure these effects can be analysed, but it is not necessary to invent a new method for it. Sort of that investigations have been done before.

503-505: What an insight! Then the current policies cannot be that wrong. Thanks for proofing this. Sorry, but this is not a very new finding at all. If different aspects are meant it should be formulated accordingly

508-510: doesn't seem to be a logic conclusion because if more users share a reduced number of cars it would be the opposite case of the standstill case. From an LCA point of view the use of vehicles increases and the share of production impacts are allocated to a higher mileage, which decreases the impacts per km. The standstill case is representing under-utilized cars. They have been produced but are not used. Mainly for BEV results would be worse per km than for frequently used cars.

Author Response

Dear Reviewer,

Thank you for your valuable feedback! According to your suggestions, we made the following changes (All lines refer to the revised version in which the changes are not displayed (MS Word: “Track Changes” not enabled); Your suggestions are in italic):

In the first part the article compares LCA results for Battery electric vehicles (BEV) and internal combustion engine vehicles (ICEV). In the second part the LCA results for different vehicle classes are used for calculating overall environmental impacts scaled up to the vehicle fleet (supposedly) for Berlin/Brandenburg in Germany. The first impression suggests that it is a very interesting topic but unfortunately the manuscript lacks clarity in describing the various assumptions for the investigated cases. In the present state the study is not very useful for the reader. Although it seems everything is explained it is very difficult to follow and in the end it remains for example unclear which vehicle distribution data is use for the study. The text (chapter 2.1.1) talks about vehicle distribution in Berlin/Brandenburg but then a diagram (fig.2) presents the German statistical data. Further the reference year used is unclear (is it 2017 or 2018 or extrapolated to 2050?) because it is not given for the later used scenarios.

We summarized the investigated cases (line 370 – 377) and reworked most parts according to your suggestions (detailed description is below) to clarify our assumptions. Partly, we had to use specific data for Germany, as Berlin-specific data wasn’t available. We included the reference year for the investigation cases.

Further it remains unclear according to which criterion the classification of vehicle classes have been done (tab.2). Because it seems strange that a “large capacity van” classified as a medium car and a “mid-range car” is classified as a large vehicle. Is it related to motor size or vehicle size? The criterion should be mentioned.

We adapted the vehicle classes according to [43]. They include motor size, vehicle weight, type of usage.

Viewing the results section it remains unclear why a rather complicated simulation model (MATsim) is used to model the daily mileage of the vehicle fleet. The results described very detailed according to the different LCA results but the differentiations related to impacts resulting from the composition of the vehicle fleet are very limited. It seems that it would have been also sufficient to take statistical data about vehicle life times and their mileage to receive the same depth of results related to the vehicle fleet.

The goal of our study was not only to present the results of our case study but also to introduce a new methodology. We emphasized this in our introduction (lines 111 – 137). We integrated simulation results so that new results can easily be analyzed (lines 565 – 569). The use case serves as a verification for the method. Nonetheless, we displayed several cases in which we investigated a different vehicle distribution, the influence of standstill, and a renewable electricity mix.

The concept of the study is not clear. In the beginning it suggests that also other means of transport are considered (line 123 and 126) but it is not sufficiently defined what is understood by the term “transport system” in this study. The aim of the study suggests a new approach to calculate the fleet impacts, but a large part of the results section just presents the comparison of LCA results of ICEV and BEV.

There are two levels: the method describes how a transport system represented by an agent-based transport simulation can be environmentally analyzed (Other MATSim scenarios can be integrated and allow the analysis of different means of transport); The case study investigates the motorized individual transport, represented by the MATSim Open Berlin Scenario. We changed the term transport system to motorized individual transport in the use case parts. Additionally, we added a Flow Chart (Figure 2.).

Since this results were just more or less reproduced from previous LCAs (mainly Ecoinvent data) nothing new is presented here. More effort could have been put in the fleet modelling and analysis but not so much for calculating the LCA. What is presented here, is just a repetition of already existing data, scaled up to a (not clear) fleet model of Berlin/Brandenburg. The main short coming is that it is difficult to understand, which assumptions have been made precisely for the different scenarios. The reader would not be able to reproduce the results. The text lacks at many points differentiated descriptions of the used input data. Also the discussion is sometimes difficult to follow because it is not precisely described which scenarios are compared with each other. The study should not be accepted without major changes. See detailed comments in the following for further illustration of the mentioned points (numbers refer to the respective lines):

58-61: How these examples are validating the statement in line 56? Are these examples giving a solution or an example for the stated problem? In the end also the study here is “just” relying on Ecoinvent data.

These are examples of solutions, we added an explanation (line 47).

65-66: How this can be stated at this point? So far no results have been presented. It is not even clear at this point what the aim of the study is. Further, the Figure 1 is displaying a selection of results from various studies. How it was made sure that these studies are comparable e.g. in terms of vehicle mileage, system boundaries etc.?

We added a detailed description of Figure 1. and the assumptions of the respective authors (line 57 – 91).

74-79: The aims suggest that the modelling and presentation of LCA results in this study is related to the vehicle fleet. But in the results section large parts are covered by the comparison of LCA results for different vehicle types. Also the literature review in chapter 1 relates in large parts to study comparing BEV and ICEV. This doesn’t fit and confuses the reader.

We reworked the last part of our introduction to emphasize the goal of our study and integrated more literature (line 111 - 137).

Chapter 2: order of sub chapters is a little confusing. E.g. 2.1 would fit better after 2.2, because it should be described in the beginning what the subject of the investigation. Chapter 2.2 should be placed at the beginning of the chapter to get to know first the goal and scope of the study.

We changed it accordingly.

91: and figure 2: why is this graph displaying German statistics and not the one of Berlin/Brandenburg which seems to be the subject of the study?

Sadly, there is no data available for Berlin/Brandenburg. Therefore, we took average German data. This is now explained in the text (line 158).

95-96: Why is this fact highlighted? Is this data used for the case study (Berlin/Brandenburg) or why it is reported here?

Yes, both vehicle distributions will be used (compare lines 316 – 324 and lines 370 – 377).

Figure 2, labelling: it should be mentioned that the dark grey bars are showing the total vehicle stock and which types of vehicles (all or just ICEV?)

We added this in the figure (now Figure 3.).

100: Reference for MATSim is missing

We added the reference (now line 167 – 168).

107: what are “10%” related to? not clear here

We explained it now (line 175 – 177)

108: Is it just the road? Because above also public transport is mentioned (e.g. S-Bahn and U-Bahn)

We explained the public transport in MATSim (line 178 – 180)

109: it is not quite clear why map data is use additionally if the network covers entire road network of Berlin and Brandenburg.

The map data in MATSim is from OpenStreetMap and wasn’t changed by us. It is just an explanation for the reader, we emphasized this (line 181).

111: Does the agents daily schedules also include the choice of transport options? Not clear here

Yes, we added this (line 183)

119: impact categories covered by LCA or what is meant here?

Yes, we added this (line 140)

120-121: what is meant with the text in brackets? ("compare" ?). not clear

We meant section 2.1.2., we corrected this (line 210).

122-123: which data is used to model the transport system of today and the future?

We do not provide a future scenario. We investigate the impacts of today’s ICEV transport and complete electrified transport (with today's technology). We emphasized this (line 190 – 193)

123: which vehicles are considered in the “transport system”? also public transport car sharing, car rent, commercial vehicles? Not clear

We investigated the whole motorized individual transport. We added this (line 139 – 143 and line 192 - 193). No changes in operation compared to the MIT today are considered to allow a comparison with existing studies. Applying new modes of transportation such as vehicle or ride-sharing services are studies we will use the verified LCA-Method for in future research.

126: “transport system level” suggests, all means of transport are considered. It should be specified in the text, what is covered.

We changed this (line 195).

136-140: difficult to follow, because it seems this part is in a differnt order than it should be. 136 -140 seem to be more suitable in the beginning of the chapter, lines 133 to 136 could follow.

We changed this accordingly (line 202 – 209).

143-146: why then such a complicated model is used to model the daily mileage if just all days of the week are considered the same and the weekends are simply reduced by a factor. Then it would have been also sufficient to take statistical data about vehicle life times and mileage to get an information about the driven vehicle km. The concept of the study is not clear. In the beginning it suggests that also other means of transport are considered.

We took statistical data (12.6 years) and scaled the simulation. This way, every vehicle has different lifetime mileages depending on their agents’ daily schedule. The average lifetime mileage of a vehicle then calculates to 206,396 km. If we would take statistical data, we had to assume an average lifetime mileage for each vehicle, neglecting that vehicles have different lifetime mileages depending on their owner’s behaviors.

149: why 300%? Doesn't makes sense - just a few vehicles might be able to drice 619 000 km.

We didn’t want to claim that this is possible. The assumption enables the investigation of the influence of high lifetime mileages. We added this in lines 219 – 221.

157-158: add “for LCA”. This phrase otherwise suggests, Ecoinvent is used to define vehicle classes.

We added this (line 230).

table 2: not clear how (according to which criterion) these classes where defined - why is a large capacity van classified as a medium car and a mid-range car is a large vehicle. Is it related to motor size or vehicle size? The criterion should be mentioned.

 

 

 

 

 

 

We explained it now (line 240 – 243).

171-172: 100%? The total vehicle is produced with secondary materials? This cannot be true. There are a lot of plastic and rubber parts - might not be secondary materials.

No, not 100 %. We changed the sentence to avoid this misunderstanding (line 246).

172-173: what is the criterion for cut-off and what does it mean in tis case? Which parts are not considered?

The cut-off system model is the allocation model from Ecoinvent. “Cut Off “ means in short, that recyclable materials are burden-free for recycling processes and that the recycling process is allocated to the recycled materials. We added this in lines 249 – 251.

179: how the values have been adapted? to which factor it was referred (motor size? vehicle size?)?

The values are adapted according to the vehicle weight. We added this in lines 257 – 258.

182: “from“ instead of „by“

Yes, we changed this (line 261).

192-195: how this was considered if it is stated at this point? If it doesn't play a role because the Ecoinvent dataset is used anyway, there is no need to refer to this study at this point. Would better fit to the discussion.

Yes, that’s right. We moved this part to the discussion (line 558 – 562)

Table 6: why isn't the consumption for BEV and ICEV displayed in the same unit? It is not comparable for the reader without knowing the underlying assumed energy content of gasoline and diesel.

We displayed the common units for fuel/electricity consumption. It is possible to calculate the respective energy content of the fuel. However, we don’t aim to compare energy consumption and prefer the common units for both so that the reader can follow the values. We didn’t change this.

205: if the Case study is Berlin, why not the Berlin electricity mix is used?

There aren’t detailed current data for the electricity mix of Berlin and Brandenburg.  

216: above it was stated that the impacts of a future vehicle fleet (completely electrified) is also modeled. Which is the reference year and which electricity mix is assumed for it?

We model a theoretical completely electrified vehicle fleet with today’s technology. We use the grid mix from 2018 for all cases but the renewable energy case. For this case, we assume a complete renewable grid mix and we calculate this grid mix according to a theoretical grid mix for the year 2050 like it is presented in [51]. However, we don’t change any other assumptions. We added an explanation in lines 300 – 303.

223: is 2050 the reference year? or is it just used to have a scenario for 100% renewable power mix? Because if 2050 will be the reference year also for a full electric vehicle fleet, efficiency gains and technology improvements have to be considered in modelling and assessing BEV manufacturing and use within the LCA. But it seems here the present (or more or less already outdated) ecoinvent data was used for modelling the present state and the future

We did not investigate a future scenario assuming new technologies. We assumed a completely electrified vehicle fleet with today’s technologies. As one reference case, we investigate the influence of a completely renewable grid mix. (line 300 – 303)

235: what do the reference in brackets mean? very confusing. please make an entry in the list of references if it is another study or refer to a chapter number, if this is a reference to another chapter

We replaced it with the section number (line 314).

239: why vehicle segments and size categories then have been defined before?

We defined the vehicle classes to demonstrate the influence of a different vehicle distribution. We made this more comprehensible in lines 317 – 324.

240-245: it is not clear what have been done here. It should be sufficiently explained which data is modified and how.

We calculate the distribution for the vehicle classes small, medium, and large according to the vehicle distribution in Germany presented in Figure 3. We use the vehicle segments and their classification (small, medium, and large presented in Table 2). For clarification, we adjusted line 317 – 324 and Table 9.

252-254: which emission factors have been used? could be displayed in a table. It is not clear what have been done exactly.

We added the respective emission factors in lines 333 – 334.

286-287: it is still not clear why the MATsim Scenario have been used at all, if it is not providing a complete picture of the Berlin traffic.

The MATSim provides a complete picture of Berlin’s traffic. In the 10 % scenario agents and road capacities are scaled-down due to computing times, we added the explanation in lines 174 – 177.

305-306: which grid mix was considered for the results in fig 3?

The grid mix from 2018 was considered. We added this in lines 371 – 378.

365-366: not realy comparable - figures in fig 10 are 1000 times higher (since given in kg)

There was a typo in the figure. We fixed it (now Figure 11.).

366-367: it should be added in this phrase that is is compared with the ICEV base case.

Yes, we added this (line 452).

368-369: this doesn't seem to be logic. Because if vehicles don't run, the proportion of fixed costs or emissions from production etc. increases, because they are allocated to a lower mileage.

The naming of our cases was confusing: The (previously called standstill case) is now called zero standstill case. The base case included standstill vehicles, the (now) zero standstill case doesn’t. We clarified this for all figures and the results, discussion, and conclusion section.

371: what is the distribution case for ICEV and BEV?? 100% ICEV or BEV respectively? not clearly described before

We included a summary of our cases to clarify this (line 371 – 377).

371: projected to which year? just 2017? or in the future?

We used the new BEV registrations from the year 2017 for the vehicle distribution case for ICEVs and BEVs. For ICEVs the respective share of gasoline and diesel-fueled ICEVs remains the same as in the base case. We clarified this in lines 371 – 377 and 457 – 458. The reference year for the other assumptions (e.g. grid mix) remains the same.

372-373: emissions of ICEV and BEV cases are compared to what?

Compared to the respective base cases for both. We added this (lines 459 – 460).

374: please insert "chapter" or chapter number or at least quotation marks in this remarks. It is very confusing if there is just written "compare".

We added the section number (line 461).

377-378: the breakeven between which options? please describe precisely

The breakeven between the ICEV and BEV base case. We described it more precisely in lines 464 – 467.

378-380: this is everything which is found in comparing transport system GWP? It is not sufficiently described which cases are advantageous and which would e.g. be the optimal share of BEV and when this would be reached etc.

We didn’t calculate the optimal share of BEVs. We investigated the cases described in lines 370 – 377. We found that a vehicle distribution consisting of small vehicles, fewer standstill vehicles, and a renewable grid mix is advantageous in comparison to the base cases. Additionally, we found that a complete electrified MIT emits less CO2-Eq. per km than today’s MIT operating with ICEVs. Both are described in lines 449 – 468 and discussed in lines 517 – 537.

figure 10 (and following figures): what is meant by the category “transport” in the case of ICEV? does it refer to battery logistics or logistics for other components? This is not clear.

We meant all emissions which occurred during the use phase and are not related to the fuel or electricity driving consumption of the vehicles. We changed the term to “operation” to clarify this (Figures 4; 6 - 11; 13; 15; 17; 19; 21).

441: what is the BEV base case? it should be made clearer before. if it is in table 9 the column "Distribution BEV 2017" this should be named as "base case" otherwise it is too confusing for the reader

The BEV base case uses the same distribution as the ICEV base case. We changed Table 9 and summarized the cases in lines 370 - 377 to make this more comprehensive.

444: smaller values in which impact categories?

In all impact categories, we added this in line 520.

453-454: the case has only 27% small vehicles and much higher shares of medium and small vehicles - and other shares haven't been investigated - how can this statement be made? it is not backed up by any results from this study

The distribution case investigates a theoretical vehicle distribution with 50.9 % small vehicles (compare Table 9) for both, ICEVs and BEVs. The distribution case shows advantages for all impact categories in comparison to the respective base case (compare results). Therefore, the statement is valid.

499-501: wasn't that clear before? This study doesn't seem to add many new results. Just two options have been compared: fleet without BEV and fleet with BEV and additionally BEV fleet fuelled with renewable electricity. For sure these effects can be analysed, but it is not necessary to invent a new method for it. Sort of that investigations have been done before.

The new method enables the analysis of agent-based transport simulation results. We emphasize this in our introduction (line 111 – 137), in our discussion (line 566 – 570), and changed this sentence accordingly (line 580 – 581).

503-505: What an insight! Then the current policies cannot be that wrong. Thanks for proofing this. Sorry, but this is not a very new finding at all. If different aspects are meant it should be formulated accordingly

We didn’t say that this is a new finding, we just summarized our results.

508-510: doesn't seem to be a logic conclusion because if more users share a reduced number of cars it would be the opposite case of the standstill case. From an LCA point of view the use of vehicles increases and the share of production impacts are allocated to a higher mileage, which decreases the impacts per km. The standstill case is representing under-utilized cars. They have been produced but are not used. Mainly for BEV results would be worse per km than for frequently used cars.

The naming of our cases was confusing: The (previously called standstill case) is now called zero standstill case. The base case included standstill vehicles, the (now) zero standstill case doesn’t. We clarified this for all figures and the results, discussion, and conclusion section.

Thank you again for your detailed review! Your comments and suggestions helped us to clarify our goals, cases, and assumptions.

Kind Regards,

The authors

Round 2

Reviewer 3 Report

can be acepted with the latest changes

Author Response

Thank you again!

Back to TopTop