the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 03 Aug 2022
Twenty-five years of the IPCC Data Distribution Centre at the DKRZ and the Reference Data Archive for CMIP data
Michael Lautenschlager
Download
- Final revised paper (published on 03 Aug 2022)
- Preprint (discussion started on 01 Apr 2022)
Interactive discussion
Status: closed
-
RC1: 'Comment on gmd-2022-73', Karl E. Taylor, 12 May 2022
General comments:
The infrastructure that enables scientists and others to access, analyze, and cite climate model simulation output may be as important in advancing the science as the development of new models and the design of new diagnostic experiments. This manuscript addresses one aspect of the coordinated international effort to serve CMIP data, and in particular the use of that data in support of the IPCC assessment process. That the subject is perhaps somewhat at the periphery of GMD's focus does not detract from its importance in documenting the advances being made in sharing, curating, archiving, and enabling citation of climate data. It highlights the DDC's and the IPCC's increasing emphasis on FAIR principles (Findable, Accessible, Interoperable, Reusable), principles championed by CMIP long before the acronym appeared in the published literature. It provides an important review of DKRZ's ongoing leadership in establishing high standards for the the preservation and curation of climate data that will be of use to researchers for many years and will ensure availability of the data underpinning the IPCC assessments.
The paper overall should be of interest now and in the future as a record of the evolution in the DDC's hosting of climate model data serving the IPCC.
Specific comments:
- This article tells the history of the DDC from the perspective of one of its original partners, DKRZ. The DKRZ group is the one with primary responsibility for the DDC's collection of climate model data, which is produced primarily by the CMIP and CORDEX projects. Thus, the article's focus is on this. For readers unfamiliar with the DDC, however, there are some gaps in the summary that need filling.
If the focus is meant to be on the CMIP and CORDEX simulation data hosted by the DDC, then a much more complete discussion is needed describing how the DDC partners with the responsible projects (CMIP and CORDEX) to meet the special needs of the IPCC. Among the questions that should be addressed are:
- What distinguishes the DDC from the distributed CMIP and CORDEX data archives? Why is the DDC necessary, given that much of the model output would already be hosted and served by the projects generating the data? Does the DDC provide added-value products? (There is mention of the IPCC "atlas", but the description does not indicate what's contained in it. From the website it appears that except for a small amount of observed precipitation, temperature, and surface wind data, the data is otherwise based entirely on the CMIP and CORDEX data. It would be good to highlight how the atlas adds value to the original CMIP and CORDEX data.)
- Have there been special DDC requirements for metadata or data services that, were it not for the leadership provided by the DDC, might have been overlooked by CMIP and CORDEX? What data requirements and services were already part of CMIP and CORDEX planning, and what role did the DDC play in making sure any special needs of the DDC were met? Were there any DDC requirements that CMIP and CORDEX were unable to implement? Did DKRZ need to supplement the metadata or services already delivered by the CMIP/CORDEX projects before data could be served by the DDC?
If the intent of the article is to provide a balanced history of the DDC and how the role of DKRZ has evolved, then much more context is needed. The reader should not be left with questions like:
- What have the other DDC partners contributed? Do they hold replicas of the data hosted by DKRZ? Do they host data not available at DKRZ?
- If there is data hosted by the DDC other than CMIP/CORDEX data, is it structured in a common way and are uniform metadata requirements imposed? If so, does the DDC rewrite the data to meet the requirements or insist that the contributors meet the requirements? Who came up with any requirements for data other than the CMIP/CORDEX data?
- For in-text references like "https://ipcc-data.org/, last access: 10 March 2022", could this be made into a hidden link? If not, please delete the "last access: ...." part. The long reference to a date of marginal interest interrupts the sentence and makes it much more difficult to digest. If you need to indicate a date, do it only once and note that "this and all subsequent URL links were last accessed 10 March 2022".
- Similarly, in documenting meetings, the reading would be improved if the actual date were omitted (i.e., only include month and year). I don't think anyone cares what days the meetings were held.
- Near line 55, why is there no mention of CMIP6? The section is labeled "The Reference Data Archive at the DDC at DKRZ". Isn't CMIP6 data included in the reference data archive? Also wasn't there CORDEX data archived too?
- Line 139: It wasn't clear what was meant by "bidirectional references between ...."
- Lines 150-155: The explanation here of what the procedure is. Please try to be a specific as possible.
- Line 156: Weren't ESGF, the WIP, and PCMDI essential in already having established metadata standards, catalogs, search engines, and more, along with long term preservation, which had gone a long way toward making CMIP data "FAIR"? I think that has to be recognized.
- Lines 180-193: There seems to be little point of including this list here. Add some content describing why the individual collaborations are needed, or eliminate the list (or move it to the "Acknowledgments").
- Line 216: It would be good to recognize the work over 30 years at PCMDI putting in place the "standardization of CMIP" data.
- Lines 245-255: The list of "gaps" needs to provide a bit information about what exactly needs improvement. It would be good to say for each item how improvement in the area will help users or data managers, or whomever.
Technical corrections:
I will separately send the authors a marked up version of their manuscript that includes editorial suggestions for improving clarity and readability, which they might consider.
Citation: https://doi.org/10.5194/gmd-2022-73-RC1 -
RC2: 'Comment on gmd-2022-73', Paul Durack, 02 Jun 2022
***Reviewer2: REF: gmd-2022-73 ("25 years IPCC Data Distribution Centre at DKRZ and the Reference Data Archive for CMIP data" by Dr Stockhause and Lautenschlager)***
General Comments
Many thanks for providing me with the opportunity to review the submission by Drs Stockhause and Lautenschlager. I must apologise for the tardiness in getting this completed.
I have been looking forward to reading this historical perspective, which lays out more than 25 years of coordination by the Data Distribution Centre (IPCC-DDC) in support of the UN Intergovernmental Panel on Climate Change (IPCC) Assessment Reports (ARs). These long-standing efforts are often assumed just to occur, however, without careful planning, organization and collaboration across the growing Coupled Model Intercomparison Project (CMIP, and the preceding Atmospheric MIP, AMIP) communities, such support and the resources required to deliver them would not be realized, and certainly not continue for the 25-year milestone documented in this new submission.
I found the submission to provide a comprehensive timeline of activities and participants, which while useful and informative, made the narrative a little harder to follow. A suggestion would be to collapse much of this into a timeline graphic, with key milestones identified through a temporal history, allowing the submission text to focus more on the milestone achievement, rather that the circumstance or the timings.
It would be useful to provide some overview as to what the role of the DDC is, how it functions, and its importance, the framing of the environment in which it operates is not clear, and the importance of this role needs to be highlighted. The placement of this within a very large (and growing) international community infrastructure would also be useful, for context.
There is a key point about the funding of these sustained (DDC, and its dependencies) activities, which is currently buried as a last paragraph. The long-term DDC funding strategy is an important point, maybe the most important point of this submission, and should be raised in the abstract. Clarification about the funding landscape with DDC’s funded as a national (operational) capability, which I do not believe is the case for DKRZ, and this could be a useful comment/idea to propose (alongside other partner organizations).
In addition, it would be useful to discuss how information/knowledge exchange can be facilitated from the ARx TSU offices. The AR6 WG1 TSU ramp down has already begun as a potential AR7 is discussed and leadership (and host organization etc) is identified. There is a requirement for much of the work, and resources to be preserved so that a likely AR7 WG1 TSU can pick up where the AR6 WG1 ended, building on, and not duplicating existing or previous efforts. This also depends upon CMIPx infrastructure providers, and this dependence, for e.g., CMIP5/6 ESGF federation needed to be operational for DDC to complete its archival role. Therefore, all infrastructure providers (ESGF nodes), modelling groups, etc need to be recognized as a key and dependent part of the effort, if there are problems (or failures), DDC can’t do its downstream job.
I have marked up some notes for consideration in the following pages, which I believe would need to be addressed before the contribution would be worthy of publication. The focus of the submission will be of interest to a wide readership and fits the remit of Geoscientific Model Development, however, recommend minor revisions are required to improve the present draft.
Specific Comments
Page 1, title: An alternative title could be “25-years of IPCC/CMIP data custodianship: the international Data Distribution Centre and the climate reference data archive”. While DKRZ has been a core player in this activity, including the institutional acronym in the title is not necessary in my view.
Page 1, lines 7-8: “..continuous developments in data standardization..” this achievement, amongst others has been dependent on a very active and engaged collaborative community, including contributions from US DOE’s PCMDI, the US NSF’s Unidata/UCAR, multiple EU IS-ENES phases, the ESGF federation partners and the 50+ modelling groups that are contributors to the CMIP6 and previous phases, amongst many, many others, etc. Without this coordinated community effort, the data archived by the DDC would not have been generated, and many of the latest phase developments (DataCite data DOIs) would not have been possible. It would be worth highlighting the role of this coordinated community at the outset.
It would be useful to document total hits, counts, or quantify the impact of the Data citation pages
Page 1, lines 9-11: “Examples of such milestones..” this sentence could be rewritten, with acronyms removed to ease of reading.
Page 1, line 14: “..DDC strategy, the DDC..” suggest a rewrite.
Page 1, line 14-15: “..positioned in the developing FAIR..” it was not obvious what this sentence was suggesting, a rewrite would be useful
Page 1, line 17: “..(DDC, https://??, last access..” suggest removing the last access and URL information to the reference list to ease reading
Page 1, lines 17-20: “..jointly managed by four partners.. ..DKRZ is the only remaining founding partner..” it wasn’t obvious if these four organizations currently comprise the DDC, or rather, are they the founding partners?
Page 2, line 31: “..selected as a shared DDC operation..” -> “..TO EXECUTE a shared DDC operation..”
Page 2, lines 34-35: “..Task Group on Climate Scenarios for Impact Assessment (TGCIA) expert group..” how are the scenarios, and their data managed (e.g. IS92, SRES, RCP and SSPs) – is this a core part of the DDC reference data?
Page 2, line 37: “..institutions to act as DDC and to provide..” -> “..institutions to act as THE DDC and to provide..”
Page 2, line 38: “This led to the DDC establishment..” redundant with first para sentence
Page 2, lines 39-46: this para could be replaced with a timeline graphic and rewritten to highlight salient points of what milestone(s) were achieved
Page 2, lines 49-50: The WCRP (expanded) CMIP is a better ordering (and less wordy) than CMIP of the WCRP.
Page 2, line 50: “..Second Assessment Report (SAR)” -> “..Second Assessment Report (SAR, 1995)”
Page 2, line 59: “As DDC Partner” -> “A DDC Partner”
Page 3, lines 69-71: It is not obvious why the sending of DVD and USB devices is of key importance. What are the regions, and how much was this service depended upon? Have these regions been better served by the more modern infrastructure provided by the latter CMIP5 and 6 phases? Is this service still required?
Page 3, line 71: “Data of the FAR were added to the DDC in 2008 in the original formats, in which they were rescued” it would be useful to this reader to know a little more about this activity – what were the original formats, was data rewritten, how was it rescued, from where/provenance, etc
Page 3, line 87-88: A more persistent ESGF reference is Williams et al., 2016; https://doi.org/10.1175/BAMS-D-15-00132.1
Page 4, Figure 2/3: what is the URL that these statistics relate to? Is the FAR and SR1.5 data numbers missing?
Page 4, line 96: a more persistent ES-DOC reference is Pascoe et al., 2020; https://doi.org/10.5194/gmd-13-2149-2020
Page 5, lines 106-107: “..long-term archived DDC AR5 data were made available through the ESGF” isn’t this around the wrong way? Wasn’t data published directly to the ESGF (numerous nodes) and once available, then pulled from the distributed ESGF federation and archived at the DKRZ DDC?
Page 5, line 107: “..IPCC authors..” - > “..IPCC AR5 authors..”
Page 5, lines 108-109: “..data archive of ETH Zurich was added..” so is this a duplicate of the ESGF CMIP3/5 data holdings, which are identified as the “DDC AR5 Reference Data Archive”
Page 5, line 115: “Corona” -> COVID-19
Page 5, lines 115-117: “The lack of coordinating..” this sentence appears redundant, a rewrite of the para would be useful
Page 5, line 118: “The idea for the FAIR guidelines..” -> The implementation of the FAIR guidelines..
Page 5, line 122: “into the Sixth Assessment cycle.” -> “into the Sixth Assessment (AR6) cycle.”
Page 6, line 127: “The concept development of the FAIR Guidelines..” again, isn’t this IMPLEMENTATION of the guidelines as defined in Wilkinson et al. 2016?
Page 6, line 135: Again, isn’t this the IMPLEMENTATION of the FAIR guidelines?
Page 6, line 138: “..by the WGI TSU..” -> “..by the AR6 WGI TSU..”
Page 6, Figure 5: This graphic was difficult to see. Additionally, final versions of the pdfs, webpages etc are now available at https://www.ipcc.ch/report/ar6/wg1/.
Data and code access (WG1) has its own homepage too https://www.ipcc.ch/report/ar6/wg1/resources/data-access
Page 6, lines 144-150: Recent updates have occurred to the CMIP6 model data license information, see https://wcrp-cmip.github.io/CMIP6_CVs/docs/CMIP6_source_id_licenses.html
Page 7, line 161: “Sixths” -> Sixth Assessment (AR6) cycle
Page 7, line 169: “This active role of the DDC in AR6 increased the DDC’s visibility..”
Page 8, line 193: “..WIP..” -> “..the WCRP WGCM Infrastructure Panel (WIP)..”
Page 8, line 200: “..requirements of the IPCC in the WGCM..” -> “..requirements of the IPCC THROUGH the WGCM..”
Page 8, line 210: “..the WCRP project CMIP” -> “..the WCRP CMIP project..”
Page 8, lines 214-215: “..the WIP was setup..” -> ..the WIP was setup in 2014..”
Page 9, line 228: “coverages of ES-DOC..” while this statement may be correct, it comes across as a documented criticism. Rather the ES-DOC aspiration could be highlighted, with a recognition that for such a service to realize its potential, enhanced, and sustained funding is required, not only for the infrastructure to hold the information, but also on the modelling group side to ensure that this information is provided
Page 10, line 270: “..55th Session..” The details of all meetings and evolutions would most usefully be presented in a table or timeline, which can then allow the text to discuss the pertinent details, rather than documenting the dates, and times
Page 10, lines 271-274: The point about the long-term DDC funding strategy is an important point, maybe the most important point, and should be raised in the abstract
Page 10, line 289: “MPDI” -> MDPI
Page 11, lines 320-325: The final AR6 WG1 report is available, the SPM Masson-Delmotte et al., 2021 also has a DOI:
IPCC, 2021: Summary for Policymakers. In: Climate Change 2021: The Physical Science Basis. Contribution of Working Group I to the Sixth Assessment Report of the Intergovernmental Panel on Climate Change [Masson-Delmotte, V., P. Zhai, A. Pirani, S.L. Connors, C. Péan, S. Berger, N. Caud, Y. Chen, L. Goldfarb, M.I. Gomis, M. Huang, K. Leitzell, E. Lonnoy, J.B.R. Matthews, T.K. Maycock, T. Waterfield, O. Yelekçi, R. Yu, and B. Zhou (eds.)]. Cambridge University Press, Cambridge, United Kingdom and New York, NY, USA, pp. 3−32, doi:10.1017/9781009157896.001
Figures
Cleaner figures would improve the manuscript. In addition, a timeline graphic would be a useful (and engaging) way to present much of the timeline information.
References
Masson-Delmotte et al., 2021 https://doi.org/10.1017/9781009157896.001 or https://www.ipcc.ch/report/ar6/wg1/ as a placeholder
Pascoe et al., 2020; https://doi.org/10.5194/gmd-13-2149-2020
Williams et al., 2016; https://doi.org/10.1175/BAMS-D-15-00132.1
Citation: https://doi.org/10.5194/gmd-2022-73-RC2 -
RC3: 'Comment on gmd-2022-73', David Huard, 27 Jun 2022
General comments
The paper is a useful contribution as it discusses the important role played by data curators in the climate science ecosystem. I would like to see the DDC history expanded a little, and more context given for some of the discussion points of the paper.
Specific comments
6 Review the abstract, I don't think it quite captures the essence of the paper, and the syntax is awkward at times.
13 Not clear how visibility and exposure pose "a challenge and opportunity to operationalize ...".
14 There is not much talk in the paper about this long term DDC strategy
15 It's not clear which part of the paper discusses the political and societal realm of climate change.
16 The history section starts with the current composition. Wouldn't it be clearer to describe the DDC history in chronological order?
20 Shouldn't CEDA be part of that list, as they're still actively engaged in the final data archival and curate final datasets from AR6.
34 Would be interesting to know what situation spurred the creation of the DDC.
42 Here and elsewhere, there is a lot left unsaid on the reasons for decisions taken. I understand getting into the historical details is not the objective, but it leaves the reader without a storyline.
52 Explain why long-term archives are useful. What actual problems do they solve? Is it to provide original model data to future researchers? If so, can you cite papers using those data?
58 Do these downloads eventually translate into publications? Is the DDC cited often as a data provider for CMIP data from previous cycles?
102 No "NO" branch on "Data QC L3 passed"? "Replicated" bubble should be tied with arrow to diagram, no? Why is data replication a necessary condition for DDC archival?
123 The IPCC is not a research activity, but an assessment of the science. So I'm not sure the sentence "The FAIR data principles describe requirements for datasets to become an integral part of the research environment." really captures the IPCC motivation for this. Also, in my mind the IPCC FAIR Guidelines is a document mostly meant for IPCC authors, so it's not clear what the TRUST principles have to do with this, as it pertains to the archive curator.
144 An important outcome of the FAIR guidelines is also to give due credit to IPCC "chapter scientists", who are often heavily involved in IPCC graphics preparation.
173 How many authors ended up making active use of these virtual workspaces?
198 I think its worth adding a sentence describing the actual CMIP6 licensing issue.
267 "The importance of data has reached the policymakers." The example here seems a bit anecdotal? Can we really say this then follow-up with the fact that DDCs face funding uncertainties?
271 I think it would be useful to mention how DDC activities were funded up to now, what challenges does DDC face now, and discuss the rationale for changing the funding model.Technical corrections
10 IPCC Sixth Assessment Report
11 Suggest to refer to the "IPCC FAIR Guidelines", not just FAIR guidelines, to avoid confusion.
13 has become more exposed -> has gained exposure and visibility
20 "DKRZ is the only remaining founding partner, who has been operating the DDC for 25 years." I would keep that comment for later, after discussing the formation of the DDC. Maybe rephrase since there were only two founding members if I understood correctly.
21 Sentence on MoU is unconnected to rest of text
27 Review sentence "The core role..." Not clear what "support in data" means.
35 The events are not in chronological order. It makes the presentation confusing.
41 syntax (again).
54 "downstream users without specific knowledge of climate model applications." What kind of users would fit this description?
55 via ca ?
65 proprietary
72 syntax
76 Not sure what the value of this table is.
87 distributed
90 Use white background
115 Use expression "COVID-19 pandemic"
116 lack -> delayed start
127 I think it would be clearer if the history of the IPCC FAIR Guidelines was concentrated in one paragraph.
133 Confusing that FAIR Guidelines were approved on the first TG-Data meeting, yet the official version was only adopted in 2022. I think TG-Data approved an early draft version in 2019, giving it the mandate to discuss it subsequently with IPCC authors at the various Lead Author Meetings. Version 1.0 got approved at the last TG-Data teleconference in 2022.
137 the analysis script applied to produce the figures -> the analysis scripts generating the figures
137 I feel the word "collected" undervalues the considerable work of IPCC authors to prepare clean, documented versions of figure generation scripts.
145 I don't think DFG is a familiar acronym outside of Germany. Clarify reference.
153 What is "use metadata"?
156 Orphan sentence.
157 "Use metadata"? Should there be a step about the creation of entries in the DDC catalog?
161 Sixths ... -> AR6
178 Not convinced the list of partners is really relevant here, it is not accompanied by text describing the actual collaborations.
236 ... evolving requirements ...
236 ... have adapted to developments ...
237 AR6 stands for ... review syntax
240 On the one hand ... on the other hand is used to express contrasting points of view. This is not the case here.
242 und -> and
262 Link with "Good Scientific Practice" unclear.Citation: https://doi.org/10.5194/gmd-2022-73-RC3 -
AC1: 'Comment on gmd-2022-73 (answering RC1-RC3)', Martina Stockhause, 06 Jul 2022
General comments:
Thanks to the reviewers for their detailed reviews with rich comments and valuable suggestions for improvements that add depth to the paper. The draft focused on the DDC and overemphasized technical aspects. The work of other groups providing standards and infrastructure were mentioned within the CMIP6 section but not for previous years. We added DDC services’ dependencies on the efforts of these groups in the pre-CMIP6 sections to acknowledge their work. In this context, the position of the DDC within the complex project and infrastructure landscape was sharpened to answer the question of its relation to the CMIP data archive and the ESGF infrastructure. More details are given on the archival process and the long-term preservation and curation measures for the DDC’s Reference Data Archive.
We took up the suggestion to discuss funding options and the long-term DDC strategy in the conclusion section, thereby shifting the focus away from technical towards institutional aspects. These changes are reflected in the abstract. The suggested figure on the DDC history has been introduced, enabling us to concentrate on the achievements/milestones in the text. The comment to expand the DDC history to provide reasons behind decisions was taken up exemplary for the EM on the future of TGICA. However, most of the IPCC-induced changes related to the Task Group rather than the DDC.
We have not addressed some comments that were outside the scope of this article, which is the balanced history of the DDC and the evolution of the role of the DDC Partner DKRZ. We do not discuss the contributions of our DDC Partners (e.g. the WGI Atlas) or further important IPCC-related input datasets outside of DKRZ’s responsibility. We do not discuss CMIP and the CMIP infrastructure, but only their roles and importance for the DDC.
Specific comments:
- FAIR and IPCC FAIR Guidelines: The FAIR principles formulated by Wilkinson et al. (2016) were interpreted in multiple ways. For example, the GO-FAIR initiative focusses on the technical aspects and the role of PIDs. The IPCC’s interpretation takes the essence of the FAIR principles and formulates a workflow approach to target them. IPCC also included the long-term aspect, which is not part of the FAIR principles but formulated in the TRUST principles for research repositories by Lin et al. (2020) and implemented in the CoreTrusSeal criteria, included in the WDS Regular Membership application process. In this regard, the IPCC FAIR Guidelines are more than an implementation of the FAIR principles but the specific data management concept of IPCC TG-Data aiming at enhanced transparency of the IPCC outputs and in principal allows the reproducibility of figures within the report.
- Target group of the IPCC FAIR Guidelines and TRUST principles/long-term data preservation: The IPCC FAIR Guidelines cover several data management aspects. We see the guidelines as collection of individual guidelines on specific aspects. The long-term data preservation aspect had been one part from the beginning (see Stockhause et al., 2019). It ensures reusability of the data on the long-term. It is a prerequisite for making data citable. Without data curation and keeping the data accessible, a data reference will not direct a reader to the data on the long-term. We do not think that the IPCC FAIR Guidelines formulated by Pirani et al. (2022) are specific enough to serve directly as IPCC authors but explain our approach towards FAIR data within the IPCC. IPCC authors need more specific and practical information on what to provide and in what form.
- Credit for “IPCC chapter scientists” for final data: Users citing the data give credit. A precondition is the long-term preservation and DOI assignment to the data. Thus, we added this credit aspect to the long-term preservation bullet point.
- CEDA’s role: CEDA is no longer a formal DDC Partner, UK support has transitioned from CEDA to MetadataWorks, but CEDA is currently contributing to the AR6 final data archival.
- FAIR Guidelines’ approval by TG-Data: We agree that the formulation was confusing and revised that part as suggested.
- Relation between downloads and citations: That is an important point but difficult to estimate. The data citation process is still broken in several aspects: 1. The cultural change to cite data in addition to papers is not completed. Many datasets are still not cited. 2. If the data is cited, many publishers do not expose data citations in their crossref metadata esp. for old papers. Thus as repository, we only know about a subset of cited datasets. In summary, the accessible data citations (using Scholix) are still too incomplete to give a reliable picture of the data citations.
- ES-DOC coverage: We rephrased the sentence to be neutral, but do not discuss the reasons for its poor coverage because of the paper’s scope.
- Usage of Virtual Workspaces by IPCC authors: DKRZ did not separate IS-ENES or DKRZ users from IPCC authors. Some users might have multiple roles. We cannot give reliable numbers on IPCC authors using the Virtual Workspaces.
- Replica at other DDC Partners: The DDC Partners are responsible to ensure the long-term availability of their data shares, according to their answers to this CoreTrustSeal criterion. The DDC Partner DKRZ stores an off-cite copy of the DDC data at the Max Planck Computing and Data Facility (MPCDF) in Garching, Germany. We have included that in the text.
- Data requirements other than CMIP and CORDEX for the DDC: The role of the DDC is the long-term preservation of data underpinning the IPCC ARs. IPCC authors select the data and the WG TSUs gather this information on data usage.
- Downstream user requirements: Using model output data requires some knowledge on spatial-temporal resolution related to model grids and knowledge about experiment design and used standards. Outside of the modeling community, this knowledge is only partially present. Examples for downstream users are climate impact researchers or experts from insurance companies.
- Recognition of PCMDI’s effort: We added that to the historical sections. This dependency on PCMDI’s work for the pre-CMIP6 assessment cycles was important for the DDC and still is important as it lay the foundation for the current CMIP data standard.
- DDC gap analysis: These gaps were identified but not further analyzed due to higher priorities for the DDC tasks related to implementation of the IPCC FAIR Guidelines into the AR6 and the lack of staff to target these gaps.
- Chronological order in the history section: We want to introduce the DDC in the first part before we describe its history.
- Progress made in the IPCC AR6 WGI report and CMIP6 licenses was recognized.
- Suggested title “25-years of IPCC/CMIP data custodianship: the international Data Distribution Centre and the climate reference data archive”: We decided not to change the title to keep the focus on the DDC Partner DKRZ and not move it towards CMIP.
- Figure 4 on the CMIP5 quality procedure: We have deleted it, as it was not clear and required more explanations than we have intended to provide in our review paper.
- Delete table 1: We moved the table to the Appendix, as it is not essential for the paper but provides further in-depths information on DDC’s core variables.
- In-text URL references: To provide access dates for each URL is a journal’s requirement. We discussed the reviewer’s suggestions with Copernicus but Copernicus recommended to leave it as it is.
- Thanks for the additionally provided marked up word document.
Citation: https://doi.org/10.5194/gmd-2022-73-AC1
