the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 14 Dec 2023
Insights on the spatial distribution of global, national and sub-national GHG emissions in EDGARv8.0
Abstract. Knowing where emissions occur is essential for planning effective emission reduction measures and for atmospheric modelling. Emission inventories are typically compiled at national level and provide sector-specific emission estimates. Disaggregating national emissions on high-resolution grids requires spatial proxies that contain information on the location of different emission sources (e.g. point sources, linear and area sources). Knowing the correct allocation of emissions from point sources is essential to avoid the misallocating high emission levels. However, gathering information on point sources covering the entire globe and a wide temporal domain (1970 to present) is challenging due to limited data availability, accuracy of the reporting and completeness of data. The latest spatial proxies developed as part of the Emissions Database for Global Atmospheric Research (EDGARv8.0) provide the user with the possibility to work with different geographical details using a consistently developed GHG emissions database. A key novelty of EDGARv8.0 is the possibility to analyse sub-national GHG emissions over the European domain, but also over the US, China, India and main world countries. The relevance of using updated spatial information is assessed on the basis of regional case studies. The data can be accessed at https://doi.org/10.2905/b54d8149-2864-4fb9-96b9-5fd3a020c224 specific for EDGARv8.0 (Crippa, 2023a) and doi:10.2905/D67EEDA8-C03E-4421-95D0-0ADC460B9658 for the sub- national dataset (Crippa et al., 2023b).
- Preprint
(3495 KB) - Metadata XML
-
Supplement
(2368 KB) - BibTeX
- EndNote
Status: closed
-
RC1: 'Comment on essd-2023-514', Anonymous Referee #1, 28 Feb 2024
General comments
In this study the development of an updated GHG emissions database, EDGARv8.0, is outlined and supplemented with regional case studies. Updated emissions inventories, point source data, shipping emissions, and proxy methodologies are explored. This effort is aimed to improve the accuracy and consistency of the spatiotemporal distribution of emissions at national and subnational scales. The goal is to better inform climate mitigation and adaptation policy and assist climate modelers in understanding the impact of emissions on the earth and atmosphere.
This paper is rightfully within the scope of ESSD and should be published after minor revisions.
Specific comments
Line 69: It is not very clear what is meant by built-up surface information from GHSL. It would be helpful to explain this or refer to some source.
Line 141: How does EDGAR harmonize subnational and national data? Is there some scaling of the subnational data to match the national totals.
Line 148: What is meant by legal site?
Line 253: Can also explain in this section that venting is the release of flare gas (e.g., natural gas) without burning, which is distinct from flaring. Is venting included as an emission source in EDGAR (for CH4)?
Line 327: Is the impact of using these new gap-filling proxies implemented in other databases or validated through other studies?
Has consideration been given to incorporating emissions at height data as a potential feature? In our recent paper we find that SO2 injection height is a source of inter-model variability, so having a standardized set of data would be useful for climate models. https://acp.copernicus.org/articles/23/14779/2023/
The conclusion section can be improved with some more discussion on potential future works that address the limitations identified in the paper.
Technical corrections
Line 35: Perhaps say “Knowing where emissions are released…”
Line 113: Can remove “…also represented by…”
Line 123: Rephrase this to be more coherent, for example "…but also for other countries such as the United States, China, and India, by providing emissions at the state or province level."
Line 136: The word “cell” here is redundant.
Line 147: 1970-Present
Line 161: Can delete “...including the latest available information…”
Line 271: “2012 to 2022”
Figure 10: Is this showing CO2 equivalent?
Line 521: Can use the word species instead of substance.
Line 543: “…what is available…”
Citation: https://doi.org/10.5194/essd-2023-514-RC1 - AC2: 'Reply on RC1', Monica Crippa, 25 Mar 2024
-
RC2: 'Comment on essd-2023-514', Anonymous Referee #2, 08 Mar 2024
The paper presents a comprehensive overview of the spatial proxies used for EDGAR v8, specifically describing the updates that were made compared to earlier versions. Since EDGAR is a widely used and recognized source of emission information, this contribution is a welcome addition to the scientific community. I would recommend to publish this article but only after a number of comments have been addressed:
- The abstract contains a number of (nearly) copy & pasted sentences from the text, in particular from the introduction. For instance, the first 3-4 sentences of the abstract and the introduction are almost the same. It can probably be summarized in the abstract a bit more.
- The introduction is not written very clearly, there seems to be a mix up of introductory sentences and sentences saying what this study is about throughout this section. I believe most necessary elements are there but the structure of this part is lacking, would recommend to rewrite this with a clear structure: introduce the topic and main issues there are, and then in the second part explain what this study is adding to that. See also some of the specific comments below.
- When the concept of emissions by NUTS region is introduced, there is a lot of attention given to the possible use of this inventory for regional policy analysis. However, most of the information was already available in the EDGAR inventory in its gridded form in previous versions. In fact, the addition of the NUTS regions to the emissions means these emissions by NUTS region can be quantified more precisely. In addition, the use of (often) global proxies makes that the NUTS2/3 region emissions are an approximation. For instance in the EU each country has its own (gridded) emission inventory which likely gives different NUTS2 level emissions from the EDGAR results presented here. The presented approach is however useful to provide a default if such national data is not available (which will be the case in many regions in the world). It would be good to add some reflections to the paper on the constraints of using subnational emission information, perhaps including some of the considerations mentioned here.
- The paper would benefit from a language review. This does not affect understanding of the article, but the use of proper English language should be ensured as it will make the paper better readable. Examples include “multi vessels information” (line 336), “accordingly with” (line 92), but also for instance the use of commas where not needed such as in line 132 (but also in other places in the paper).
More specific comments include the following:
- In the abstract some description of the results is missing. E.g. line 29-30 says “the relevance is assessed”. But what was the outcome? I think that should be summarized in the abstract.
- Abstract line 29: “main world countries”: rather use something like “other large countries”
- Line 41: please include here that EDGAR follows this approach of estimating country level emissions and then gridding them using spatial proxies. This may be trivial for some but not for everyone.
- Line 43-44 “although the resolution of underlying spatial information (…) may be higher”. I do not see the relevance of this addition at this point in the text?
- Line 46 “to weight national inventories” This is not clear, but I think what you mean is here to disaggregate national emissions to the grid? And “weight” should be “weigh” here.
- Line 48-49 “EU Member States” should rather be “Parties to the LRTAP Convention”, as CEIP is not related to EU but to the wider LRTAP Convention
- Lines 50-52 and 63-66 say the same thing twice. Suggest to integrate the part of lines 50-52 with the paragraph in lines 63-66.
- Line 54-56: again seems to be almost 1-to-1 copy of this sentence in the abstract
- Line 58: “power plant” should be “power plants” (plural)
- Line 59: “distributed” I suppose that here you mean “diffuse”?
- Line 89: “weight” should be “weigh” (same typo also in line 177)
- Line 91: Heaviside function: please explain or add reference
- Line 114: the correct sub-national information: what information? Please be specific.
- Line 118-119: NUTS3 is more detailed than NUTS2, and can easily be aggregated to NUTS2 in case needed. So how can the use of a more aggregated NUTS2 (compared to NUTS3) enhance the capability to represent sub-national emissions? Please explain or rephrase if something else is meant (probably the latter).
- Line 127-128: “more disaggregated statistics”. What is meant here? Please explain and/or give example.
- Line 162: add “spatially “ between “correctly” and “allocate”
- Line 163-170: It is known that the point source databases mentioned have some gaps and inconsistencies between each other. Please clarify if these databases have been used “as is” or if any corrections or gapfilling has taken place. It would also be good to add specifically what is the metric used for the proxy map: e.g. power plant capacity, reported emissions (CO2 or other?), or something else.
- Lines 175-178: if larger gaps exist, has there been any assessments on the consistency between energy consumption in power plants (from statistics) and the installed capacity as a means of cross-check?
- Line 198-199: the number of power plants has grown significantly in the database. Is this only a “real” growth in number of plants, or may it also be related to missing plants in earlier years?
- Line 203 “industrialised regions”: probably this refers to the regions like Europe and North America? But other regions are also partly industrialised. Please rephrase accordingly.
- Line 219: manufacturing of what? Probably manufacturing industry is meant, but that is an overarching term for the sectors mentioned thereafter. So if that is correct the word “manufacturing” is obsolete here.
- Line 220: “solvents” probably refers to the use of solvents, not the production of them. Please rephrase sentence accordingly.
- Line 238-243: What is used as a proxy in the end? Is that iron/steel plant capacity or another metric? In case capacity is used, please add some wording that this is an approximation as some of the installed capacity may actually not be in use.
- Line 253: The section is called “Venting and flaring” but in fact is only about flaring except for one mention of “venting” in line 267. Please clarify the differences and similarities between venting and flaring.
- Line 259-266: This part is a bit long to make a simple point that few countries make up a large part of the emissions. The point is already made in the first line (259-261) so perhaps the second sentence can be removed (no need to sum up countries beyond the top-4).
- Line 281: “fertilizer industries”: does this refer to fertilizer production or fertilizer use? Please clarify, also because the fertilizer production is already covered in the industrial section.
- Line 282-283: “satellite-derived NH3 point source database”: what do point sources refer to in this case? Probably not individual farms, since typically from satellites you get information at ~10km resolution at the very best.
- Line 287: “a fraction”: which fraction, how much to the so-called point source?
- Line 291: reference/short explanation of previous EDGAR proxy?
- Line 321: The ratio between which emissions? This is not clear.
- Line 337/Section 4: It is mentioned STEAM data were used from EDGAR v6 onwards. Does this imply this section describes the update for shipping in EDGAR v6 which is still the same in EDGAR v8? If yes please make that clear in the text, if not please add changes between different EDGAR versions.
- Related to the previous comment, which version of STEAM emissions is used? And is the same version of those STEAM emissions used in EDGAR v6, (v7), and v8?
- Line 341-344: The information on SECAs, NECAs, etc. and the impact on emissions is already included implicitly in the emissions calculated by STEAM. So why does it needs to be presented here and presented as a future update of EDGAR?
- Line 351: comparison to EDGAR v5 refers to the last version which was not yet using STEAM emissions, at least that is how it is understood. But what was the proxy used then? Please add that to inform the reader what the comparison shows.
- Line 358: remove “also”
- Line 442: Global administrative layer: reference?
- Line 526-527: Implementation may occur at subnational level, but also in many cases at national level.
- Line 534: “national” can be “(inter)national”
- In the acknowledgements, the authors seem to acknowledge mainly themselves and a range of the EDGAR databases which are basically the dataset corresponding to this paper. This reads very strange, and does not seem appropriate.
Citation: https://doi.org/10.5194/essd-2023-514-RC2 - AC1: 'Reply on RC2', Monica Crippa, 25 Mar 2024
Status: closed
-
RC1: 'Comment on essd-2023-514', Anonymous Referee #1, 28 Feb 2024
General comments
In this study the development of an updated GHG emissions database, EDGARv8.0, is outlined and supplemented with regional case studies. Updated emissions inventories, point source data, shipping emissions, and proxy methodologies are explored. This effort is aimed to improve the accuracy and consistency of the spatiotemporal distribution of emissions at national and subnational scales. The goal is to better inform climate mitigation and adaptation policy and assist climate modelers in understanding the impact of emissions on the earth and atmosphere.
This paper is rightfully within the scope of ESSD and should be published after minor revisions.
Specific comments
Line 69: It is not very clear what is meant by built-up surface information from GHSL. It would be helpful to explain this or refer to some source.
Line 141: How does EDGAR harmonize subnational and national data? Is there some scaling of the subnational data to match the national totals.
Line 148: What is meant by legal site?
Line 253: Can also explain in this section that venting is the release of flare gas (e.g., natural gas) without burning, which is distinct from flaring. Is venting included as an emission source in EDGAR (for CH4)?
Line 327: Is the impact of using these new gap-filling proxies implemented in other databases or validated through other studies?
Has consideration been given to incorporating emissions at height data as a potential feature? In our recent paper we find that SO2 injection height is a source of inter-model variability, so having a standardized set of data would be useful for climate models. https://acp.copernicus.org/articles/23/14779/2023/
The conclusion section can be improved with some more discussion on potential future works that address the limitations identified in the paper.
Technical corrections
Line 35: Perhaps say “Knowing where emissions are released…”
Line 113: Can remove “…also represented by…”
Line 123: Rephrase this to be more coherent, for example "…but also for other countries such as the United States, China, and India, by providing emissions at the state or province level."
Line 136: The word “cell” here is redundant.
Line 147: 1970-Present
Line 161: Can delete “...including the latest available information…”
Line 271: “2012 to 2022”
Figure 10: Is this showing CO2 equivalent?
Line 521: Can use the word species instead of substance.
Line 543: “…what is available…”
Citation: https://doi.org/10.5194/essd-2023-514-RC1 - AC2: 'Reply on RC1', Monica Crippa, 25 Mar 2024
-
RC2: 'Comment on essd-2023-514', Anonymous Referee #2, 08 Mar 2024
The paper presents a comprehensive overview of the spatial proxies used for EDGAR v8, specifically describing the updates that were made compared to earlier versions. Since EDGAR is a widely used and recognized source of emission information, this contribution is a welcome addition to the scientific community. I would recommend to publish this article but only after a number of comments have been addressed:
- The abstract contains a number of (nearly) copy & pasted sentences from the text, in particular from the introduction. For instance, the first 3-4 sentences of the abstract and the introduction are almost the same. It can probably be summarized in the abstract a bit more.
- The introduction is not written very clearly, there seems to be a mix up of introductory sentences and sentences saying what this study is about throughout this section. I believe most necessary elements are there but the structure of this part is lacking, would recommend to rewrite this with a clear structure: introduce the topic and main issues there are, and then in the second part explain what this study is adding to that. See also some of the specific comments below.
- When the concept of emissions by NUTS region is introduced, there is a lot of attention given to the possible use of this inventory for regional policy analysis. However, most of the information was already available in the EDGAR inventory in its gridded form in previous versions. In fact, the addition of the NUTS regions to the emissions means these emissions by NUTS region can be quantified more precisely. In addition, the use of (often) global proxies makes that the NUTS2/3 region emissions are an approximation. For instance in the EU each country has its own (gridded) emission inventory which likely gives different NUTS2 level emissions from the EDGAR results presented here. The presented approach is however useful to provide a default if such national data is not available (which will be the case in many regions in the world). It would be good to add some reflections to the paper on the constraints of using subnational emission information, perhaps including some of the considerations mentioned here.
- The paper would benefit from a language review. This does not affect understanding of the article, but the use of proper English language should be ensured as it will make the paper better readable. Examples include “multi vessels information” (line 336), “accordingly with” (line 92), but also for instance the use of commas where not needed such as in line 132 (but also in other places in the paper).
More specific comments include the following:
- In the abstract some description of the results is missing. E.g. line 29-30 says “the relevance is assessed”. But what was the outcome? I think that should be summarized in the abstract.
- Abstract line 29: “main world countries”: rather use something like “other large countries”
- Line 41: please include here that EDGAR follows this approach of estimating country level emissions and then gridding them using spatial proxies. This may be trivial for some but not for everyone.
- Line 43-44 “although the resolution of underlying spatial information (…) may be higher”. I do not see the relevance of this addition at this point in the text?
- Line 46 “to weight national inventories” This is not clear, but I think what you mean is here to disaggregate national emissions to the grid? And “weight” should be “weigh” here.
- Line 48-49 “EU Member States” should rather be “Parties to the LRTAP Convention”, as CEIP is not related to EU but to the wider LRTAP Convention
- Lines 50-52 and 63-66 say the same thing twice. Suggest to integrate the part of lines 50-52 with the paragraph in lines 63-66.
- Line 54-56: again seems to be almost 1-to-1 copy of this sentence in the abstract
- Line 58: “power plant” should be “power plants” (plural)
- Line 59: “distributed” I suppose that here you mean “diffuse”?
- Line 89: “weight” should be “weigh” (same typo also in line 177)
- Line 91: Heaviside function: please explain or add reference
- Line 114: the correct sub-national information: what information? Please be specific.
- Line 118-119: NUTS3 is more detailed than NUTS2, and can easily be aggregated to NUTS2 in case needed. So how can the use of a more aggregated NUTS2 (compared to NUTS3) enhance the capability to represent sub-national emissions? Please explain or rephrase if something else is meant (probably the latter).
- Line 127-128: “more disaggregated statistics”. What is meant here? Please explain and/or give example.
- Line 162: add “spatially “ between “correctly” and “allocate”
- Line 163-170: It is known that the point source databases mentioned have some gaps and inconsistencies between each other. Please clarify if these databases have been used “as is” or if any corrections or gapfilling has taken place. It would also be good to add specifically what is the metric used for the proxy map: e.g. power plant capacity, reported emissions (CO2 or other?), or something else.
- Lines 175-178: if larger gaps exist, has there been any assessments on the consistency between energy consumption in power plants (from statistics) and the installed capacity as a means of cross-check?
- Line 198-199: the number of power plants has grown significantly in the database. Is this only a “real” growth in number of plants, or may it also be related to missing plants in earlier years?
- Line 203 “industrialised regions”: probably this refers to the regions like Europe and North America? But other regions are also partly industrialised. Please rephrase accordingly.
- Line 219: manufacturing of what? Probably manufacturing industry is meant, but that is an overarching term for the sectors mentioned thereafter. So if that is correct the word “manufacturing” is obsolete here.
- Line 220: “solvents” probably refers to the use of solvents, not the production of them. Please rephrase sentence accordingly.
- Line 238-243: What is used as a proxy in the end? Is that iron/steel plant capacity or another metric? In case capacity is used, please add some wording that this is an approximation as some of the installed capacity may actually not be in use.
- Line 253: The section is called “Venting and flaring” but in fact is only about flaring except for one mention of “venting” in line 267. Please clarify the differences and similarities between venting and flaring.
- Line 259-266: This part is a bit long to make a simple point that few countries make up a large part of the emissions. The point is already made in the first line (259-261) so perhaps the second sentence can be removed (no need to sum up countries beyond the top-4).
- Line 281: “fertilizer industries”: does this refer to fertilizer production or fertilizer use? Please clarify, also because the fertilizer production is already covered in the industrial section.
- Line 282-283: “satellite-derived NH3 point source database”: what do point sources refer to in this case? Probably not individual farms, since typically from satellites you get information at ~10km resolution at the very best.
- Line 287: “a fraction”: which fraction, how much to the so-called point source?
- Line 291: reference/short explanation of previous EDGAR proxy?
- Line 321: The ratio between which emissions? This is not clear.
- Line 337/Section 4: It is mentioned STEAM data were used from EDGAR v6 onwards. Does this imply this section describes the update for shipping in EDGAR v6 which is still the same in EDGAR v8? If yes please make that clear in the text, if not please add changes between different EDGAR versions.
- Related to the previous comment, which version of STEAM emissions is used? And is the same version of those STEAM emissions used in EDGAR v6, (v7), and v8?
- Line 341-344: The information on SECAs, NECAs, etc. and the impact on emissions is already included implicitly in the emissions calculated by STEAM. So why does it needs to be presented here and presented as a future update of EDGAR?
- Line 351: comparison to EDGAR v5 refers to the last version which was not yet using STEAM emissions, at least that is how it is understood. But what was the proxy used then? Please add that to inform the reader what the comparison shows.
- Line 358: remove “also”
- Line 442: Global administrative layer: reference?
- Line 526-527: Implementation may occur at subnational level, but also in many cases at national level.
- Line 534: “national” can be “(inter)national”
- In the acknowledgements, the authors seem to acknowledge mainly themselves and a range of the EDGAR databases which are basically the dataset corresponding to this paper. This reads very strange, and does not seem appropriate.
Citation: https://doi.org/10.5194/essd-2023-514-RC2 - AC1: 'Reply on RC2', Monica Crippa, 25 Mar 2024
Data sets
EDGARv8.0_GHG Monica Crippa et al. https://doi.org/10.2905/B54d8149-2864-4FB9-96B9-5FD3A020C224
EDGARv8.0 Greenhouse Gas Emissions at sub-national level Monica Crippa et al. https://doi.org/10.2905/D67EEDA8-C03E-4421-95D0-0ADC460B9658
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 484 | 95 | 32 | 611 | 35 | 27 | 23 |
- HTML: 484
- PDF: 95
- XML: 32
- Total: 611
- Supplement: 35
- BibTeX: 27
- EndNote: 23
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1