DIAMOND: D6.3 – Open data management plan – Update 2

www.climate-diamond.eu 28/11/2025 D6.3 – Open data management plan – Update 2 WP6 – Open Page i D6.3 – Open Data Management Plan – Update 2 Disclaimer Funded by the European Union. Views and opinions expressed are however those of the author(s) only and do not necessarily reflect those of the European Union or the European Climate, Infrastructure and Environment Executive Agency (CINEA). Neither the European Union nor the granting authority can be held responsible for them. Copyright Message This report, if not confidential, is licensed under a Creative Commons Attribution 4.0 International License (CC BY 4.0); a copy is available here: https://creativecommons.org/licenses/by/4.0/. You are free to share (copy and redistribute the material in any medium or format) and adapt (remix, transform, and build upon the material for any purpose, even commercially) under the following terms: (i) attribution (you must give appropriate credit, provide a link to the license, and indicate if changes were made; you may do so in any reasonable manner, but not in any way that suggests the licensor endorses you or your use); (ii) no additional restrictions (you may not apply legal terms or technological measures that legally restrict others from doing anything the license permits). Grant Agreement Number 101081179 Acronym DIAMOND Full Title Delivering the next generation of open Integrated Assessment MΟdels for Netzero, sustainable Development Topic HORIZON-CL5-2022-D1-02 Funding scheme HORIZON EUROPE, RIA – Research and Innovation Action Start Date December 2022 Duration 48 Months Project URL http://www.climate-diamond.eu/ EU Project Advisor Silvia Vaghi Project Coordinator Institute of Communications and Computer Systems - ICCS Deliverable D6.3 – Open data management plan – Update 2 Work Package WP6 – Open Date of Delivery Contractual 30/11/2025 Actual 28/11/2025 Nature Report Dissemination Public Lead Beneficiary Institute of Communications and Computer Systems (ICCS) Responsible Author Stratis Alexandrou Email [email protected] ICCS Phone +30 210 772 3612 Contributors Georgios Xexakis (HOLISTIC) Reviewer(s) Konstantinos Koasidis, Alexandros Nikas (ICCS) Keywords data management; open source; open access; FAIR data Page ii D6.3 – Open Data Management Plan – Update 2 EC Summary Requirements 1. Changes with respect to the DoA The I2AM PARIS web platform has been revamped and rebranded, and is now referred to, as IAM PARIS web platform. No other changes with respect to the work described in the DoA. 2. Dissemination and uptake This report can be used internally by all consortium partners to guide all data management activities of the project. It can be also used by researchers outside the DIAMOND consortium as an example for informing their own data management strategy (especially for other Horizon Europe projects). 3. Short summary of results (<250 words) This report presents the third and final edition (Update 2) of the DIAMOND project’s open Data Management Plan (DMP). Like the previous DMP reports (D6.1 and D6.2), it describes the data that the project uses and generates, outlines the strategy for supporting Findable, Accessible, Interoperable, and Reusable (FAIR) data, and clarifies how resources and responsibilities are organised to ensure proper data management, data security, and attention to ethical requirements. Although this update does not markedly revise the practices established in earlier versions, it provides an updated overview of all datasets, journal articles, and project deliverables that have been released up to November 2025. These outputs are openly accessible through the DIAMOND website and its Zenodo community, while journal articles are also available through the respective publishers. In parallel with this report, the project continues to maintain a machine-actionable data management plan through OpenAIRE’s ARGOS service, which is kept up to date with metadata for all generated datasets. Although this document represents the last DMP report to be produced within DIAMOND, data management procedures will continue throughout the final project year. 4. Evidence of accomplishment This report and the DIAMOND maDMP hosted in ARGOS. Page iii D6.3 – Open Data Management Plan – Update 2 Preface DIAMOND will update, upgrade, and fully open six Integrated Assessment Models (IAMs) that are emblematic in scientific and policy processes, improving their sectoral and technological detail, spatiotemporal resolution, and geographic granularity. It will further enhance modelling capacity to assess the feasibility and desirability of Pariscompliant mitigation pathways, their interplay with adaptation, circular economy, and other SDGs, their distributional and equity effects, and their resilience to extremes, as well as robust risk management and investment strategies. This will be done via integration of tools and insights from psychology, finance research, behavioural and labour economics, operational research, and physical science. The project will develop a transdisciplinary scientific approach to legitimise the implementation process and co-create research questions that stretch the frontiers of climate science, as well as establish vibrant communities of practice to transparently open model enhancements and to develop capacities, thereby lowering the entrance barriers to the established IAM community. ICCS INSTITUTE OF COMMUNICATIONS AND COMPUTER SYSTEMS EL BC3 ASOCIACION BC3 BASQUE CENTRE FOR CLIMATE CHANGE - KLIMA ALDAKETA IKERGAI ES CESAR KRATENA KURT AT CICERO CICERO SENTER FOR KLIMAFORSKNING NO CYI THE CYPRUS INSTITUTE CY E4SMA ENERGY ENGINEERING ECONOMIC ENVIRONMENT SYSTEMS MODELING AND ANALYSIS SRL IT HOLISTIC HOLISTIC IKE EL COMILLAS UNIVERSIDAD PONTIFICIA COMILLAS ES ISINNOVA ISTITUTO DI STUDI PER L'INTEGRAZIONE DEI SISTEMI (I.S.I.S) - SOCIETA'COOPERATIVA IT SEURECO SEURECO SOCIETE EUROPEENNE D'ECONOMIE SARL FR UM UNIVERSITEIT MAASTRICHT NL ESMIA ESMIA CONSULTANTS INC. CA USMF THE UNIVERSITY OF MARYLAND FOUNDATIION INC US UMD UNIVERSITY OF MARYLAND US EPFL ECOLE POLYTECHNIQUE FEDERALE DE LAUSANNE CH ETH EIDGENOESSISCHE TECHNISCHE HOCHSCHULE ZUERICH CH UNIBAS UNIVERSITAT BASEL CH Imperial IMPERIAL COLLEGE OF SCIENCE TECHNOLOGY AND MEDICINE UK Oxford THE CHANCELLOR, MASTERS AND SCHOLARS OF THE UNIVERSITY OF OXFORD UK UCL UNIVERSITY COLLEGE LONDON UK Page iv D6.3 – Open Data Management Plan – Update 2 Executive Summary This report is the final update to the initial version of the open Data Management Plan (DMP) of DIAMOND. In general, all versions of the DMP include the following information: 1. a description of the types and formats of data that are and will be generated and collected during the project, including the origin, size, and utility of the data; 2. a brief description of the project models that (will) process this data; 3. a strategy and an allocation of resources and responsibilities for achieving Findable, Accessible, Interoperable, and Reusable (FAIR) data; and 4. details on how we (will) ensure data security and correct handling of all ethical aspects. This update does not include any significant changes or additions to the data practices established in the initial DMP. Nevertheless, it provides an exhaustive documentation of all datasets, articles, and deliverables that have been published acknowledging DIAMOND until now. Overall, eleven datasets and twelve project deliverables have been released, as of November 2025, and are provided as open access on the DIAMOND project’s website and Zenodo community. In addition, thirty-eight journal articles acknowledging DIAMOND have been published (as fully open access). The articles are thus freely available on the websites of their publishers and are also referenced within the dedicated section of the DIAMOND website with the aim to increase their findability. In parallel with this report, a machine-actionable data management plan1 (maDMP) has been developed for DIAMOND using the ARGOS service of OpenAIRE. The maDMP has been and will be continuously updated with metadata for all datasets generated from project activities. Data management procedures will continue throughout the final project year, but this document represents the last DMP report to be produced within DIAMOND. 1 https://doi.org/10.5281/zenodo.7781424 Page v D6.3 – Open Data Management Plan – Update 2 Contents 1 Introduction .................................................................................................................................................... 1 2 Data Description ............................................................................................................................................. 3 2.1 Project objectives and implications for data collection and generation .................................................................. 3 2.2 Types and formats of data to be generated and collected ........................................................................................... 4 2.2.1 Data processing tools .......................................................................................................................................................... 4 2.2.2 Data inputs and outputs for models ............................................................................................................................. 6 2.2.3 Data used in scenario analysis ......................................................................................................................................... 8 2.3 Origin of the data and re-use of existing data ................................................................................................................... 9 2.4 Expected size of the data .......................................................................................................................................................... 11 2.5 Data utility ....................................................................................................................................................................................... 11 3 FAIR Data Guidelines.................................................................................................................................... 12 3.1 Making Data Findable ................................................................................................................................................................ 12 3.2 Making Data Openly Accessible ............................................................................................................................................. 12 3.3 Making Data Interoperable ...................................................................................................................................................... 14 3.4 Making Data Reusable ............................................................................................................................................................... 14 4 Allocation of Resources ............................................................................................................................... 15 5 Data Security................................................................................................................................................. 16 6 Ethical Aspects .............................................................................................................................................. 17 7 Machine actionable DMP in Argos ............................................................................................................. 18 8 Project publications and datasets as of November 2025 ......................................................................... 19 8.1 Publications .................................................................................................................................................................................... 19 8.2 Datasets & Repositories ............................................................................................................................................................ 22 8.3 Deliverables .................................................................................................................................................................................... 23 References ............................................................................................................................................................ 24 Table of Figures Figure 1. A screenshot of the Machine-Actionable Data Management Plan of DIAMOND in ARGOS .......................... 1 Table of Tables Table 1. Data types and formats per project activity .......................................................................................................................... 4 Table 2. Models developed by the DIAMOND project ...................................................................................................................... 5 Table 3. Auxiliary models that are linked to the new DIAMOND models .................................................................................. 6 Table 4. Indicative input and output data for project models ........................................................................................................ 6 Table 5. Indicative policies that can be explored in project scenarios ........................................................................................ 8 Page vi D6.3 – Open Data Management Plan – Update 2 Table 6. Open-access data from the base models that can be re-used by the new models ............................................. 9 Table 7. Selection of open-source licenses to use for new model code .................................................................................. 13 Table 8. Suggested options for open-access scientific publishing in DIAMOND ................................................................. 15 Table of Abbreviations DMP Data Management Plan maDMP machine-actionable Data Management Plan FAIR Findable, Accessible, Interoperable, and Reusable TRUST Transparency, Responsibility, User focus, Sustainability, and Technology CC Creative Commons GDPR General Data Protection Regulation Page 1 D6.3 – Open Data Management Plan – Update 2 1 Introduction This report is the third and final version of the open Data Management Plan (DMP) of the Horizon Europe DIAMOND project. The DMP outlines guidelines and procedures on the collection, generation, processing, and sharing of data throughout the duration of the project. The report mainly includes information on the expected data, including a description of model data inputs/outputs, data formats, data sources, as well as practices for data processing, curation, and preservation (Chapter 2). The report covers the whole lifecycle of project data, including data handling during and after the project, along with a detailed strategy to ensure that this data is Findable, Accessible, Interoperable, and Reusable (FAIR; see Chapter 3). It also delineates DIAMOND’s Open Science strategy, i.e., how we ensure that all data and project outputs are open access and that all projectdeveloped applications and models are open source, along with the required allocation of project resources to accomplish this strategy (Chapter 4). Apart from data accessibility, the report discusses security (Chapter 5) and ethical aspects (Chapter 6) related to data collection and use. In parallel with this document, a machine-actionable Data Management Plan (maDMP) was developed using the ARGOS service of the OpenAIRE2 (Figure 1). The maDMP is an online collection of rich metadata for all datasets that have been and will be developed during the project as well as specific guidelines for each dataset, for instance, on how a dataset is shared and whether it is linked to another project output, e.g., publications. The datasets themselves are stored in Zenodo3 which is a digital repository that follows the desired principles of Transparency, Responsibility, User focus, Sustainability and Technology (TRUST principles; see Lin et al., 2020), while the maDMP provides links to them. Through the ARGOS service, the maDMP can be read by both humans and machines and connects to OpenAIRE and European Open Science Cloud4, improving the reusability and findability potential of the data. The maDMP is a “living document”, to be updated immediately upon publication of a project dataset in Zenodo, even if there are no other periodic updates of the DMP report. Chapter 7 provides a strategy on how new datasets are added in the maDMP as well as on the roles of the different consortium partners in this process. Figure 1. A screenshot of the Machine-Actionable Data Management Plan of DIAMOND in ARGOS 2 https://argos.openaire.eu/explore-plans/publicOverview/de2a89ba-9d7a-48ab-ab64-ecab2bd19674 3 https://zenodo.org/communities/diamond/ 4 https://open-science-cloud.ec.europa.eu Page 2 D6.3 – Open Data Management Plan – Update 2 This final version of the DMP is mainly updated with summaries of all research products of the projects that have been released to date. Specifically, Chapter 8 lists all deliverables that have been produced in the context of the project as well as journal publications and accompanying datasets that have been acknowledging DIAMOND. Overall, thirty-seven journal articles, eleven datasets, and twelve deliverables have been published as of November 2025. Page 9 D6.3 – Open Data Management Plan – Update 2 Policy Type of measures Carbon border adjustments Economic COVID-19 recovery efforts Economic Regulations on emissions, energy, efficiency Regulations New standards on buildings, fuels, etc. Regulations Promotion of behavioural change Social International collaboration schemes (e.g., climate clubs) Foreign policy Changes in global supply chains (e.g., Europe’s current efforts to reduce its reliance on Russian natural gas) Foreign policy Notes: Some policies were adapted from the DMP of PARIS REINFORCE. 2.3 Origin of the data and re-use of existing data As discussed in the previous sections, the development of the new models are based on the existing code and structure of their base models while their inputs are based on existing data sources. These inputs are enhanced significantly with the insights and datasets from the various disciplines participating in DIAMOND, including from psychology, finance, and industrial ecology, and are based on an extensive review of the relevant literature (both peer-reviewed and grey) and available datasets. In addition, a significant goal of the project is to find ways to open-up both modelling code and input data. This indicates that all existing model code that cannot be opened needs to be adapted or circumvented to allow new models to be fully open source, while all data inputs need to be open access. The latter is one of the most challenging aspects of the Open Science strategy of the project, since many significant data sources for modelling are proprietary or not publicly available, such as the results from some of the analyses of IEA. An indicative list of data inputs used in the existing models that are publicly available and can be directly used by the new models is shown in Table 6. For more details, see Deliverable D6.7 “Open science protocols & diagnostics” on the project website30. Table 6. Open-access data from the base models that can be re-used by the new models Data sources Use Base models using the data EUROSTAT Energy balance sheets, energy prices, macroeconomic and sectoral activity data, population data and projection, physical activity data, CO2 emission factors and EU ETS registry for allocating emissions between ETS and non ETS NEMESIS EU reference scenarios Potential calibration of NEMESIS to nuclear and hydro share in electricity mix and exogeneous NEMESIS 30 https://climate-diamond.eu/publications/deliverables/ Page 10 D6.3 – Open Data Management Plan – Update 2 Data sources Use Base models using the data energy efficiency EPA abatement curves Calibration GCAM, PROMETHEUS IPCC Shared Socioeconomic Pathways Population qualification projections GCAM, PROMETHEUS, NEMESIS, TIMES-GEO FAOSTAT balances Food demand, agriculture GCAM UN World Population projections PROMETHEUS, GEMINI-E3 UNSD Energy balances TIMES-GEO ClimateWatchData GHG emissions PROMETHEUS OECD GEMINI-E3: data for calibration of indirect taxation and government expenditures; NEMESIS: data for calibration of long-term project of GDP in non-EU countries (complementary to Eurostat) NEMESIS, GEMINI-E3 IMF GEMINI-E3: data for calibration of indirect taxation and government expenditures; NEMESIS: data for short/medium-term GDP projections for non-EU countries NEMESIS, GEMINI-E3, CEPII alternative to OECD long-term GDP projections for non-EU countries NEMESIS World Input-Output Database To complete EUROSTAT on imports and exports NEMESIS OECD/ITF Transportation data TIMES-GEO Global Energy Assessment Energy end-use in transport and buildings TIMES-GEO Notes: The list is adapted from the DMP of PARIS REINFORCE. Apart from re-used data, a large volume of new data will be produced based on the co-creation activities of the project, as stakeholders’ feedback guides both the new model development and scenario analysis of DIAMOND. During the project, stakeholders can provide their perspectives on what is currently missing from the wide IAM ecosystem and what features are required for new models. Communities of practice are also invited to participate in and inform the development of the new models, including communities behind the creation and maintenance of the existing models such as the GCAM, IEA-ETSAP, and OSeMOSYS/CLEWs communities as well as wider modelling communities such as the Integrated Assessment Modelling Consortium (IAMC). These practitioners and researchers, along with policymakers and representatives from the industry and civil society, will also help define the scenarios that are modelled during the project. These stakeholder perspectives are recorded and stored Page 11 D6.3 – Open Data Management Plan – Update 2 appropriately, based on GDPR guidelines (see Chapter 5) and adequate ethical treatment (see Chapter 6). 2.4 Expected size of the data Based on previous modelling-based projects such as PARIS REINFORCE31, the total size of data that is expected to be collected, processed, and produced by the end of the project will be around 250GB. Most of this size is expected to result from the data inputs used in the new models, and, especially, from the sheer amounts of data outputs that will be produced. Other types of data that can be relatively heavy include the audio and video recordings of consortium and stakeholder meetings, which should be at a scale of hundreds of MB for each meeting, multiplied by a few dozens of meetings that are and will be organised. On the other hand, the minutes of these meetings are documented in a small number of reports which are not more than a few MB each. Model publications are also expected to not take too much space, considering that they will be around 60-80 pdf documents. 2.5 Data utility The data outputs that are and will be produced by DIAMOND are expected to be useful to all expected audiences of the project. The new models are directly useful to climate-economy modellers and other researchers while they are indirectly useful to policymakers, industry, and civil society representatives. The results of the scenario analysis are useful to the same target audiences and inform their policies, strategies, research, and other activities. We will especially focus our efforts to disseminate project data to the European Commission and EU agencies, national/local governments, businesses, energy-intensive industries, financial institutions, and researchers of the broader climate modelling landscape (mitigation, impacts, adaptation). 31 https://paris-reinforce.eu/sites/default/files/2022-12/D8.8%20Data%20Management%20Plan-Update%202_v1.00_SUBMITTED.pdf Page 12 D6.3 – Open Data Management Plan – Update 2 3 FAIR Data Guidelines 3.1 Making Data Findable We will ensure that all project outputs are findable by using adequate identifiers and metadata. All deliverables and policy briefs are uploaded in a Zenodo community that has been specifically created for the project32 and where they are automatically fitted with persistent digital object identifiers (DOIs). In the case of scientific publications, a DOI is provided by the publishing journal, although we still upload publications (or accepted manuscripts) in Zenodo and, potentially, also preprints. A versioning system is also employed to track the state of each publication and ensure that interested parties can find the version they are looking for. For this, we use Zenodo’s paradigm where each new version gets a separate DOI, but a top-level DOI is also available, resolving to the latest version available. When each document is uploaded in Zenodo or on a journal, we include adequate keywords to facilitate document retrieval via search engines. We are also using a consistent naming system for each type of publication: • Scientific publications: “author(s)_name(s)_year” (e.g., Smith_et_al_2023.pdf) • Policy briefs: “DIAMOND_Policy_Brief_Title (e.g., DIAMOND_Policy_Brief_IAMs_In_Policy.pdf) • Deliverable: “DIAMOND_DXX_Title” (e.g., DIAMOND_D6.2_Open_Data_Management_Plan_Update_1.pdf) Datasets of project outputs that underpin publications are also archived in Zenodo and fitted with a DOI and adequate keywords. We then link all datasets in Zenodo with the maDMP in ARGOS33 and accompany them with rich metadata, using the Horizon Europe template provided in ARGOS (see Chapter 7 for more information). Similarly, all model code are stored in public repositories in GitHub which will be then linked to Zenodo and the maDMP. All model code, documentation, inputs, and outputs will be also published in the IAM PARIS modelling platform which will be continuously promoted as an integrated hub of information for the climate-economy and energy modelling community. For each dataset, we will include links to all related project publications in Zenodo and scientific journals to further increase findability. Finally, both the platform and the project website will include enough relevant content and an optimised sitemap structure to ensure findability in search engines like Google and Bing. 3.2 Making Data Openly Accessible All project deliverables, policy briefs, and datasets are published in Zenodo under Creative Commons licenses. In most cases, we have published project outputs with the highly permissive CC BY license (version 4.0). Exceptionally, we may also consider less permissive licenses such as CC BY-NC to restrict commercial uses of project outcome, but only when this is necessary for specific and clear reasons by the author(s) of the publication or the dataset. Similarly, scientific publications of the project have been and will be published in journals offering open-access options in compliance with the Horizon Europe rules. When possible, we will publish in fully open-access journals, also considering the Open Research Europe publishing platform. In case that the available fully open-access options do not align with the scope of a publication, we will select a journal that offers a gold open-access option from the list of journals that the organisations of project partners have a publishing agreement with. In that way, we ensure that all project insights become immediately available and exploitable by project audiences. In case 32 https://zenodo.org/communities/diamond/ 33 https://doi.org/10.5281/zenodo.7781424 Page 13 D6.3 – Open Data Management Plan – Update 2 that a project deliverable is linked to a scientific publication, we still add the deliverable in Zenodo, but we may add an appropriate embargo period so that the deliverable becomes publicly available only after the paper is published. All new models developed by the project have been published under open licenses. Table 7 shows the licenses that have been so far selected for the new models (for more details see D6.7). It is noted that these licenses refer to the model code and not the input data. In the case of input data, there are still some pending issues with the use of publicly restricted datasets in some of the base models and for which we are exploring open-access alternatives. The selection of licences for input data will be finalised towards the end of the project with the delivery of the final version of each model. Table 7. Selection of open-source licenses to use for new model code New models Open-source licence GCAM-Europe Educational Community License v2.034 (same as base model, GCAM) OMNIA Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International Public License35 CLEWs-EU MIT License 2.036 (same as base model) GEMINI-E3 EU GNU LESSER GENERAL PUBLIC LICENSE Version 2.137 NEMESIS-World Apache License 2.038 adding the “Commons Clause” 39 OPEN-PROM GNU AGPLv340, adding the “Commons Clause”37 In terms of file formats, we strive to use well-known formats that can be opened by freeware software such as pdf, csv, txt, mp3, and mp4 files. We aim to avoid sharing content through proprietary file formats such as docx and xlsx. When this is not feasible (for instance, when we need to publish spreadsheet files with multiple tabs), we will accompany files in proprietary formats with links to compatible freeware software such as the Open Office suite. We also ensure that all open project outputs remain available for as long as possible. All project publications and datasets are published in established online repositories such as Zenodo and GitHub where high availability is expected for many years to come. This is especially the case for Zenodo where its operation is supported by the EU. The project website will be also kept online for at least three years after the project’s end to support the dissemination and findability of project outcomes. Project coordinator ICCS and project partner HOLISTIC will also ensure the longevity of the IAM PARIS platform for at least three years after DIAMOND’s end (till around 2030). The partners will also undertake efforts to ensure the platform’s sustainability, either by future projects that will take over its operation or through appropriate funding mechanisms. The Horizon Results Booster service of the EU will be also consulted to find these mechanisms. 34 https://choosealicense.com/licenses/ecl-2.0/ 35 https://creativecommons.org/licenses/by-nc-sa/4.0/deed.en 36 https://choosealicense.com/licenses/apache-2.0/ 37 https://www.gnu.org/licenses/old-licenses/lgpl-2.1.en.html 38 https://www.apache.org/licenses/LICENSE-2.0 39 https://commonsclause.com/ 40 https://choosealicense.com/licenses/agpl-3.0/ Page 14 D6.3 – Open Data Management Plan – Update 2 3.3 Making Data Interoperable We achieve high interoperability of project data by using adequate data formats and by providing informative metadata. As suggested in Section 3.2, all project datasets and reports of the project are shared through widespread formats such as pdf, csv, and txt, avoiding proprietary formats when possible. The documentation of all new models along with the results of the scenario analysis are formatted based on the IAMC reporting templates, ensuring that the wider climate-economy modelling community can use them, while also achieving interoperability with other relevant platforms of the community such as the Scenario Explorers by IIASA. Metadata for all project datasets are added in DIAMOND’s maDMP in ARGOS, using the format template of Horizon Europe. As the maDMP is machine actionable, it will be easy to read the internal representation of metadata and potentially convert it to another format, further ensuring the interoperability of project datasets. 3.4 Making Data Reusable As suggested in the previous section on open access, by releasing all deliverables and datasets through CC BY license we also support their uptake by interested parties. Similarly, all scientific papers are published under openaccess licenses and are made available directly after acceptance by the journals. We will use more restrictive licenses, e.g., including clauses to limit commercial use, only in exceptional cases and when it is fully necessary. We also ensure the reusability of datasets by using the Horizon Europe metadata scheme in the maDMP of the project. For each dataset, the scheme provides a short description of the data, links with publications and other datasets, and guidelines for the specific dataset related to FAIR practices, allocation of resources, and security and ethical aspects. Lastly, all modelling documentation and results are formatted using the reporting templates of IPCC AR6 (IAMC format41) to ensure reusability by the wider modelling community of integrated assessment and climate change mitigation. 41 https://docs.ece.iiasa.ac.at/iamc.html Page 15 D6.3 – Open Data Management Plan – Update 2 4 Allocation of Resources Most of the practices described in Chapter 3 for ensuring FAIR data in the project are not requiring any costs from the project. All deliverables and datasets are uploaded in Zenodo which is free to use, and we also use the free version of GitHub to store project code. Similarly, the documentation of datasets in the maDMP in ARGOS is also free of charge. Other activities for ensuring FAIR data (see Chapter 3) require resources which have been considered in the project’s Grant Agreement. The extension and hosting of IAM PARIS requires funds that have been budgeted under WP6, while funds on the development and maintenance of the project website have been considered in the budget of WP7 and the defined purchase costs for ICCS and HOLISTIC. We have also earmarked a part of the budget for publishing in open access journals. Most of this budget is managed by the project coordinator ICCS while all partners have some funds available for individual open access publications related to their work in the project. Suggested options for publishing in open access journals are shown in Table 8. Table 8. Suggested options for open-access scientific publishing in DIAMOND Access type Funder Fees License Publish in a fully Open Access journal Paid through DIAMOND’s Grant Article Processing Charges indicatively ranging between 130€ and 9,500€ CC BY 4.0 CC BY-NC-ND 4.0 Publish in a journal that has the option of Gold Open Access Paid through publishing agreements between the organisations of project partners and the publisher In terms of responsibilities of project partners, ICCS and HOLISTIC have the overview of all data management in the project. The same partners ensure that all project data and publications are uploaded in Zenodo and linked to the project website, the maDMP, and the IAM PARIS platform. The platform is further developed by HOLISTIC and ICCS, including creating new visualisations and applications based on the project’s data. All project partners are responsible for correct data handling and curation based on the guidelines of the DMP, including that model code is frequently uploaded in the DIAMOND community in GitHub. As also mentioned above, HOLISTIC is responsible for keeping the website and the platform online and all related project data available for at least three years after the end of the project. For the platform, both ICCS and HOLISTIC are exploring ways to further extend its lifetime. Page 16 D6.3 – Open Data Management Plan – Update 2 5 Data Security We ensure the security of all project data through robust data storage and secure platforms for communication and data exchange. On the latter, ICCS has created a dedicated workspace in its enterprise version of Microsoft Teams as an internal communication system for video-calls and chatting among project partners (see also Milestone 2). This system is also connected to a secure instance of Microsoft SharePoint which serve as the exclusive data exchanging system for the project. The management and security of these systems is supported by the admins and the Data Protection Officer (DPO) of ICCS, while their servers are within the EU (Greece), ensuring compliance with GDPR and relevant EU legislation. This is important as SharePoint is also used to store contact details of project stakeholders that need to be fully secure. Similarly, the personal data of newsletter subscribers are stored in the MailerLite account of HOLISTIC which is also GDPR-compliant. Apart from the SharePoint, other data storage systems used in the project include the databases of the project website and the IAM PARIS platform. For both databases, HOLISTIC and ICCS have implemented disaster recovery and backup policies to ensure that the data is safe from loss caused by a disaster such as a critical systems failure, fire, theft, or natural disaster. A similar process is followed by ICCS for the SharePoint system while there is also a versioning system in place that protects the users from accidentally deleting or modifying data. The project’s communities in Zenodo and GitHub are also used to store data during the process, and, most importantly, to preserve all created datasets and publications after the end of the project. The possibility of data loss in these repositories is very low, as all files and documents are stored in multiple online servers that ensure redundancy. Additionally, it is highly unlikely that these repositories will close operations and, even in that case, they will migrate all content to suitable archives such as the servers of the Software Heritage Foundation and Internet Archive. Page 17 D6.3 – Open Data Management Plan – Update 2 6 Ethical Aspects All data collection and management activities of the project are compliant with the EU GDPR regulation and with the national privacy and data protection laws of the countries of each partner. For most activities related to model development, there are no ethical or legal aspects apart from respecting the licenses of databases that are used as sources of modelling inputs. In contrast, ethical aspects are especially important in all the co-creation activities of the project (WP2), where we collect the perspectives of different project stakeholders through workshops, interviews, and surveys. For the workshops we use the Chatham House Rule ("when a meeting, or part thereof, is held under the Chatham House Rule42, participants are free to use the information received, but neither the identity nor the affiliation of the speaker(s), nor that of any other participant, may be revealed."). Minutes of the workshops follow this rule and avoid linking individuals to specific statements. The same process is used for documenting interviews, while surveys also avoid questions on personal details other than those that are needed for research reasons (e.g., whether a respondent is from academia or policymaking). In all engagement activities, we ask for an explicit and clear informed consent from the participants. As mentioned in Chapter 5, all contact details and feedback of project stakeholders are stored in the project’s secure SharePoint instance while the contact details of the newsletter subscribers are stored in the MailerLite account of HOLISTIC. Both the SharePoint and the MailerLite instances are hosted within the EU and are GDPR compliant. We avoid exchanging contact details through unsecure channels such as emails. No personal data transfer take place from EU to non-EU countries and vice versa. The only information to be collected include lists of research questions to be extracted by project stakeholders, without capturing who expressed which opinion (anonymisation). 42 https://www.chathamhouse.org/about-us/chatham-house-rule Page 18 D6.3 – Open Data Management Plan – Update 2 7 Machine actionable DMP in Argos In parallel with this report, the DIAMOND maDMP in ARGOS provides a detailed overview of all datasets generated, curated, or managed during the project43. The maDMP is updated whenever a project dataset is created or modified during the project. Personnel from ICCS and HOLISTIC are responsible for creating entries for new datasets in the maDMP while all project partners are responsible for checking that the metadata provided for their datasets in the maDMP is correct. The following process is being used for creating a new dataset in maDMP: 1. The project partner(s) that created the dataset (henceforth called “data creators”) share it with ICCS and HOLISTIC (“data managers”). 2. The data managers then upload the dataset on Zenodo. 3. The data managers also create a new dataset in the maDMP and prefill it by searching the name of the dataset through the search engine of ARGOS. 4. The data managers complete the metadata for the dataset based on guidance from this report as well as link the dataset to relevant deliverables or scientific publications. 5. The data creators are invited to check these metadata and ensure their accuracy. 6. The data managers update the maDMP with the new dataset. A similar process is used for new model code, where links are and will be created between the model repositories in GitHub and Zenodo to ensure that the code base of the models receives a unique DOI. This Zenodo entry is then linked to the maDMP which will be documented with adequate metadata as above. In some cases, such as the OMNIA model, the entire raw input database is also uploaded in Zenodo. All these processes are evaluated based on the experience of data managers and creators and may be adapted and optimised further during the project. 43 https://doi.org/10.5281/zenodo.7781424