DIAMOND: D6.2 – Open data management plan – Update 1

www.climate-diamond.eu 31/05/2024 D6.2 – Open data management plan – Update 1 Open data management plan – Update 1 1. ICCS PU - Public WP5 4. CICERO WP6 – Open Page i D6.2 – Open Data Management Plan 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 D6.2 – Open data management plan – Update 1 Work Package WP6 – Open Date of Delivery Contractual 31/05/2024 Actual 31/05/2024 Nature Report Dissemination Public Lead Beneficiary Institute of Communications and Computer Systems (ICCS) Responsible Author Konstantinos Koasidis Email [email protected] ICCS Phone +30 210 772 3612 Contributors Georgios Xexakis (HOLISTIC); Natasha Frilingou, Alexandros Nikas (ICCS) Reviewer(s) Clàudia Rodés-Bachs (BC3); Sonja Sechi (E4SMA) Keywords data management; open source; open access; FAIR data; Page ii D6.2 – Open Data Management Plan EC Summary Requirements 1. Changes with respect to the DoA No 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) The current report documents the second version of the open Data Management Plan (DMP) of DIAMOND. As in the initial DMP report (D6.1), this report includes a description of the data that will be used and generated during the project; a strategy and an allocation of resources and responsibilities for achieving Findable, Accessible, Interoperable, and Reusable (FAIR) data; and details on how we shall ensure data security and correct handling of all ethical aspects. While this updated report does not introduce any changes or additions to the data practices that have been established in the original DMP, it additionally includes a section documenting all project datasets, journal articles, and deliverables that have been published as of May 2024. All such publications and datasets are provided as open access on the DIAMOND project’s website and Zenodo community, and—in the case of journal articles—on the websites of their publishers. Again, in parallel with this report, we have developed and maintain a machine-actionable data management plan 1 (maDMP) using the ARGOS service of OpenAIRE. This maDMP is continuously updated with metadata for all datasets generated from project activities. The DMP report will be updated once more in Month 36 of the project (November 2025). 4. Evidence of accomplishment This report and the DIAMOND maDMP hosted in ARGOS. 1 https://doi.org/10.5281/zenodo.7781424 Page iii D6.2 – Open Data Management Plan 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 SYSTEM 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 D6.2 – Open Data Management Plan Executive Summary This report is an update to the initial version of the open Data Management Plan (DMP) of DIAMOND, which was published in March 2023 (Month 4). 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; 4. details on how we (will) ensure data security and correct handling of all ethical aspects. The current update does not include any 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 (and associated funding). Overall, five datasets and seven project deliverables have been released, as of May 2024, and are provided as open access on the DIAMOND project’s website and Zenodo community. In addition, eleven 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 plan 2 (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. The DMP report will be updated once more in Month 36 of the project (November 2025). 2 https://doi.org/10.5281/zenodo.7781424 Page v D6.2 – Open Data Management Plan 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 May 2024 .................................................................... 19 9 References ................................................................................................................................. 22 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 will be 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 Table 6. Open-access data from the base models that can be re-used by the new models .......................................... 10 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 Figures Figure 1. A screenshot of the Machine-Actionable Data Management Plan of DIAMOND in ARGOS ............................ 2 Page vi D6.2 – Open Data Management Plan List 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 D6.2 – Open Data Management Plan – Update 1 1 Introduction This report is the second 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 will cover 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 will also delineate DIAMOND’s Open Science strategy, i.e., how we will ensure that all data and project outputs will be open access and that all projectdeveloped applications and models will be open source, along with the required allocation of project resources to accomplish this strategy (Chapter 4). Apart from data accessibility, the report will discuss 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 OpenAIRE 3 (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 the dataset will be shared and whether it is linked to another project output, e.g., publications. The datasets themselves will be stored in Zenodo 4 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 will provide links to them. Through the ARGOS service, the maDMP can be read by both humans and machines and connect to OpenAIRE and European Open Science Cloud 5 , 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, without having to wait for the periodic updates of the DMP report (in this case, the final update in November 2025). Nevertheless, DMP reports will also provide summaries of all datasets that are currently present in the maDMP, as illustrated in Chapter 8 of the current report. Chapter 7 of this report provides a strategy on how new datasets will be added in the maDMP as well as on the roles of the different consortium partners in this process. Based on the experience of the partners with this strategy, adaptations will be made as needed. 3 https://argos.openaire.eu/explore-plans/publicOverview/de2a89ba-9d7a-48ab-ab64-ecab2bd19674 4 https://zenodo.org/communities/diamond/ 5 https://open-science-cloud.ec.europa.eu Page 2 D6.2 – Open Data Management Plan – Update 1 Figure 1. A screenshot of the Machine-Actionable Data Management Plan of DIAMOND in ARGOS As most model development tasks of the project are still ongoing, while scenario modelling activities have not yet practically started, there has been no need to adapt the data management practices introduced in the first DMP. Nevertheless, an extra chapter has been added (Chapter 8), listing 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, 11 journal articles, five datasets, and seven deliverables have been published as of May 2024. The report DMP will be further updated in Month 36, documenting any potential changes to the project’s data practices as well as enriching the list of publications and datasets. Page 9 D6.2 – Open Data Management Plan – Update 1 Policy Type of measures Targets for afforestation and protected lands Targets Carbon pricing Economic Carbon sink pricing Economic Land use change emissions tax Economic Energy taxes Economic 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 will be based on the existing code and structure of their base models while their inputs will be based on existing data sources. These inputs will be enhanced significantly with the insights and datasets from the various disciplines participating in DIAMOND, including from psychology, finance, and industrial ecology, and will be 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 would be 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 will 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. Strategies to achieve fully open-access data inputs will be drawn out during the project in a case-by-case basis and will aim to substitute proprietary datasets with similar open-source data. In case this is not possible, another potential solution will be to convert input data in a prebuilt, binary format which can be only read by the models and can be potentially shared openly. For instance, the GCAM model uses such binary datasets for reading data from IEA balances. 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. Page 10 D6.2 – Open Data Management Plan – Update 1 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 energy efficiency NEMESIS 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. Page 11 D6.2 – Open Data Management Plan – Update 1 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 will guide both the new model development and scenario analysis of DIAMOND. During the project, stakeholders will be able to provide their perspectives on what is currently missing from the wide IAM ecosystem and what features are required for new models. Communities of practice will be 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 will be modelled during the project. These stakeholder perspectives will be recorded and stored 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 REINFORCE 24 , the total size of data that is expected to be collected, processed, and produced will be around 250GB. Most of this size is expected to result from the data inputs that will be 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 will be organised. On the other hand, the minutes of these meetings will be documented in a small number of reports which will not be 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. More accurate estimations will be available during the project and the next revision of the DMP. 2.5 Data utility The data outputs that will be produced by DIAMOND are expected to be useful to all expected audiences of the project. The new models will be directly useful to climate-economy modellers and other researchers while they will be also indirectly useful to policymakers, industry, and civil society representatives. The results of the scenario analysis will be 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). 24 https://paris-reinforce.eu/sites/default/files/2022-12/D8.8%20Data%20Management%20Plan-Update%202_v1.00_SUBMITTED.pdf Page 12 D6.2 – Open Data Management Plan – Update 1 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 will be uploaded in a Zenodo community that has been specifically created for the project 25 and where they will be automatically fitted with persistent digital object identifiers (DOIs). In the case of scientific publications, a DOI will be provided by the publishing journal, although we will still upload publications (or accepted manuscripts) in Zenodo and, potentially, also preprints. A versioning system will be 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 will 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 will add adequate keywords to facilitate document retrieval via search engines. We will also use 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 will also be archived in Zenodo and fitted with a DOI and adequate keywords. We will then link all datasets in Zenodo with the maDMP in ARGOS 26 and accompany them with rich metadata, using the Horizon Europe template provided in ARGOS (see Chapter 7 for more information). Similarly, all model code will be 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 I2AM 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 will be published in Zenodo under Creative Commons licenses. In most cases, we will publish project outputs using 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 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 25 https://zenodo.org/communities/diamond/ 26 https://doi.org/10.5281/zenodo.7781424 Page 13 D6.2 – Open Data Management Plan – Update 1 the list of journals that the organisations of project partners have a publishing agreement with. In that way, we will ensure that all project insights become immediately available and exploitable by project audiences. In case that a project deliverable is linked to a scientific publication, we will 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 will be published under open-source licenses. In some cases, such as GCAM-Europe, we will use the same license as the one for the base model. Table 7 shows the licenses that have been tentatively selected for the new models. 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 both model code and input data will be finalised later in the project. Table 7. Selection of open-source licenses to use for new model code New models Open-source licence GCAM-Europe Educational Community License v2.027 (same as base model, GCAM) OMNIA MIT28 CLEWs-EU Apache License 2.029 (same as base model, OSeMOSYS) GEMINI-E3 EU Undecided yet NEMESIS-World MIT OPEN-PROM GNU AGPLv330, adding the “Commons Clause” to restrict commercial use31 In terms of file formats, we will strive to use well-known formats that can be opened by freeware software such as pdf, csv, txt, mp3, and mp4 files. We will strive 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 will also ensure that all open project outputs will remain available for as long as possible. All project publications and datasets will be 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 I2AM 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 27 https://choosealicense.com/licenses/ecl-2.0/ 28 https://choosealicense.com/licenses/mit/ 29 https://choosealicense.com/licenses/apache-2.0/ 30 https://choosealicense.com/licenses/agpl-3.0/ 31 https://commonsclause.com/ Page 14 D6.2 – Open Data Management Plan – Update 1 service of the EU will be also consulted to find these mechanisms. 3.3 Making Data Interoperable We will 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 will be 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 will be formatted based on the IPCC AR6 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 will be added in DIAMOND’s maDMP in ARGOS, using the format template of Horizon Europe. As the maDMP is machine actionable, it will be fairly 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 will also support their uptake by interested parties. Similarly, all scientific papers will be published under open-access licenses and will be 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 will also ensure the reusability of datasets by using the Horizon Europe metadata scheme in the maDMP of the project. For each dataset, the scheme will provide 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 will be formatted using the reporting templates of IPCC AR6 (IAMC format 32 ) to ensure reusability by the wider modelling community of integrated assessment and climate change mitigation. 32 https://docs.ece.iiasa.ac.at/iamc.html Page 15 D6.2 – Open Data Management Plan – Update 1 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 will be uploaded in Zenodo which is free to use, and we will 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 I2AM 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 will have the overview of all data management in the project. The same partners will ensure that all project data and publications are uploaded in Zenodo and linked to the project website, the maDMP, and the I2AM PARIS platform. The platform will be further developed by HOLISTIC and ICCS, including creating new visualisations and applications based on the project’s data. All project partners will be 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 will be 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 D6.2 – Open Data Management Plan – Update 1 5 Data Security We will 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 will 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 will be important as SharePoint will be also used to store contact details of project stakeholders that will need to be fully secure. Similarly, the personal data of newsletter subscribers will be 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 I2AM 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 will be 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 D6.2 – Open Data Management Plan – Update 1 6 Ethical Aspects All data collection and management activities of the project will be 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 not ethical or legal aspects apart from respecting the licenses of databases that will be used as sources of modelling inputs. In contrast, ethical aspects become especially important in all the cocreation activities of the project (WP2), where we will collect the perspectives of different project stakeholders through workshops, interviews, and surveys. For the workshops we will use the Chatham House Rule ("when a meeting, or part thereof, is held under the Chatham House Rule 33 , 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 will heed this rule and will avoid linking individuals to specific statements. The same process will be used for documenting interviews, while surveys will 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 will ask for an explicit and clear informed consent from the participants. The form used to get this informed consent and further details on ethical aspects of the stakeholder engagement activities of the project will be included within the deliverables of WP2. As mentioned in Chapter 5, all contact details and feedback of project stakeholders will be stored in the project’s secure SharePoint instance while the contact details of the newsletter subscribers will be stored in the MailerLite account of HOLISTIC. Both the SharePoint and the MailerLite instances are hosted within the EU and are GDPR compliant. We will avoid exchanging contact details through unsecure channels such as emails. No personal data transfer will take place from EU to non-EU countries and vice versa. The only information to be collected will be lists of research questions to be extracted by project stakeholders, without capturing who expressed which opinion (anonymisation). 33 https://www.chathamhouse.org/about-us/chatham-house-rule Page 18 D6.2 – Open Data Management Plan – Update 1 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 project 34 . The maDMP will be updated whenever a project dataset is created or modified during the project and a summary of its contents will also be provided in the final update of this report (D6.3). Personnel from ICCS and HOLISTIC will be responsible for creating entries for new datasets in the maDMP while all project partners will be 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 (and its final update, D6.3) 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 will be used for new model code (expected to be released in the upcoming months of the project lifetime), where links 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 will be 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 will be also uploaded in Zenodo. All these processes will be evaluated based on the experience of data managers and creators and may be adapted and optimised further during the project. 34 https://doi.org/10.5281/zenodo.7781424