DIAMOND: D3.2 – Model development strategy – Update 1 – Report

www.climate-diamond.eu 29/11/2024 D3.2 – MODEL DEVELOPMENT STRATEGY – UPDATE 1 WP3 – Update & Upgrade Page i D3.2 – Model development strategy – Update 1 Disclaimer Funded by the European Union. Views and opinions expressed are 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 D3.2 – Model development strategy – Update 1 Work Package WP3 – Update & Upgrade Date of Delivery Contractual 30/11/2024 Actual 29/11/2024 Nature Report Dissemination Public Lead Beneficiary Asociacion BC3 Basque Centre for Climate Change - Klima Aldaketa Ikergai (BC3) Responsible Author Dirk-Jan Van de Ven Email [email protected] BC3 Phone +34 94 401 46 90 (Ext) 136 Contributors Clàudia Rodés-Bachs, Russell Horowitz (BC3); Sonja Sechi, Maurizio Gargiulo (E4SMA); Constantinos Taliotis (CYI); Marc Vielle, Sigit Perdana (EPFL); Baptiste Boitier, Hayat Mekki (SEURECO); Anastasis Giannousakis, Giannis Tolios (ICCS) , Shivika Mittal (CICERO) Reviewer(s) Shivika Mittal (CICERO), Georgios Xexakis (HOLISTIC), Sigit Perdana (EPFL), Konstantinos Koasidis, Alexandros Nikas (ICCS) Keywords Integrated assessment; Model development; Climate policy; Sustainability Page ii D3.2 – Model development strategy – Update 1 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 The deliverable is intended for policymakers and the broader stakeholder community of DIAMOND and other IAM-related projects to showcase the capabilities and simple descriptions of the models in the armoury of the project. It is also intended for the scientific community of IAMs to present the added capabilities, expansions, and integrations that are performed in DIAMOND. 3. Short summary of results (<250 words) Integrated Assessment Models (IAMs) have historically been the dominant tools in mitigation policy assessment and the analysis of the implications of global and/or regional scenarios. In DIAMOND, a coordinated effort is undertaken to update, upgrade, expand, and integrate six different economy-wide IAMs. In a previous stage (D3.1), different modelling teams outlined initial development plans for their models. To enhance the broad application of the models, these plans must be updated to reflect stakeholder needs. This deliverable presents the updated model development strategies, incorporating progress and adjustments based on stakeholder feedback obtained in a dedicated workshop. Each IAM team completed an updated internal survey, showing progress and describing the integration of stakeholders’ insights in development strategies. Survey outputs show that all models plan to expand their geographical and sectoral/technological coverage. The models have updated their plans to improve the geographical detail in the EU and/or expand coverage in the rest of the world. In terms of technological coverage, the updated development strategies align with the higher technoeconomic detail demanded by stakeholders, particularly improving the representation of hydrogen and electricity storage technologies. In terms of policy representation, a large part of newly developed capacity will focus on trade policies, while all models will also be improved in terms of behavioural measures, such as transport modal shifts. Since the updated survey also highlighted data requirements for different input assumptions used by the models, this report provides a preliminary model harmonisation strategy that can be used in model development and future model intercomparison studies. 4. Evidence of accomplishment This report and the corresponding Zenodo dataset (https://zenodo.org/records/14216491) including the updated surveys for all models. Page iii D3.2 – Model development strategy – Update 1 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 D3.2 – Model development strategy – Update 1 Executive Summary Integrated Assessment Models (IAMs) have historically been the dominant tools in climate change mitigation policy assessment and the analysis of global and/or regional scenarios. While IAMs have largely contributed to global and regional policymaking and decarbonisation strategies, scientific literature has pointed out some weaknesses or areas of improvement that should be considered for the robustness of future IAM applications. First, IAMs do not represent socioeconomic and political barriers associated with the implementation of alternative policies. Second, there is misrepresentation of certain dynamics that can in turn lead to misinterpretation of derived outcomes, such as the common practice of using economy-wide carbon prices as an oversimplified proxy for global climate policies. One of the main aspects to be improved is achieving finer temporal and geographical scales to assess the distinct region-/country-specific solution space. Finally, another area of improvement for IAMs is to address the current lack of transparency and detailed documentation. IAMs have frequently been considered as “black boxes”, with opaque input assumptions that do not allow a full understanding of the uncertainties embedded in the produced outputs. In this context, DIAMOND aims to enhance six multi-sectoral IAMs by integrating stakeholder input and connecting them with sector-specific models, overcoming current limitations to ensure robust, reliable, and socially relevant outcomes. As outlined in the initial development plan (D3.1), the six IAMs within DIAMOND have heterogenous development plans, due to their systematically different modelling structures, including the geographical, temporal, and sectoral coverage, the representation of different variables and policy mechanisms, or input requirements. These model-specific development plans must be refined and updated to accurately reflect the needs and priorities of stakeholders, ensuring they remain practical, aligned with real-world objectives and problems, and impactful. The aim of this report is to update the initial model development plans, based on ongoing development efforts in the project so far, as well as drawing from suggestions for new capabilities and features provided by different stakeholders in the Requirements Workshop, hosted as part of the project’s transdisciplinary approach. Finally, this update also introduces a recommended data harmonisation protocol as a preparatory step towards model use within the project (WP5) as well as beyond. In order to gather the main information of the updated plans of the models, each IAM team has updated the internal surveys presented in D3.1, providing a comprehensive overview of their progress and detailing how they have incorporated stakeholders' insights into their development strategies. In terms of geographical coverage, the consulted stakeholders stressed the need for increasing the spatial granularity of models. As a response, different modelling teams have finetuned their plans to improve their representation of the European Union (EU) (e.g., by individual member states) and their coverage in the rest of the world. In terms of sectoral and technological coverage, different stakeholders demanded improving the level of technoeconomic detail, including aspects such as technological readiness or costs associated with storage technologies. Additionally, stakeholders highlighted the need for greater sectoral disaggregation, distinguishing between areas such as passenger versus freight transport or more detailed breakdowns of labour markets, particularly based on skill levels. Stakeholders also called for more accurate representation of trade policies and behavioural measures, such as transport modal shifts. The updated model development strategies align better with the stakeholders’ suggestions for increased technoeconomic and policy representation, with a focus on improving the representation of hydrogen and electricity storage systems and preparing models for scenarios focusing on behavioural change and trade policies. Finally, the updated internal survey includes information on the data sources for different input assumptions used by the different models. This deliverable includes early-stage insights for a model harmonisation strategy, which can assist modelling teams during the development stage, and which could also be applied in future project activities for participating or coordinating model inter-comparison exercises, a non-trivial process that is crucial for improving the consistency and robustness of scenario results. The steps outlined in this harmonisation strategy could also serve as a framework for future modelling exercises beyond the scope of this project. Page v D3.2 – Model development strategy – Update 1 Contents 1. INTRODUCTION............................................................................................................................... 1 2. UPDATING THE MODEL DEVELOPMENT STRATEGY ................................................................... 3 2.1 Initial development plan (D3.1) ............................................................................................................................................ 3 2.2 Incorporating stakeholder insights ..................................................................................................................................... 4 2.3 Representation of policies and measures ......................................................................................................................... 9 3. UPDATED MODEL DEVELOPMENT STRATEGIES ........................................................................ 12 3.1 From GCAM to GCAM-Europe ............................................................................................................................................ 12 3.1.1 Overview ...................................................................................................................................................................... 12 3.1.2 Development timeline ............................................................................................................................................ 16 3.1.3 Meetings (internal modelling team and external) ...................................................................................... 16 3.1.4 Completed development steps .......................................................................................................................... 18 3.1.5 Subsequent development steps......................................................................................................................... 22 3.1.6 GCAM-Europe references ..................................................................................................................................... 25 3.2 From TIMES-GEO to OMNIA .............................................................................................................................................. 26 3.2.1 Overview ...................................................................................................................................................................... 26 3.2.2 Development timeline ............................................................................................................................................ 30 3.2.3 Meetings (internal modelling team and external) ...................................................................................... 30 3.2.4 Completed development steps .......................................................................................................................... 32 3.3 From CLEWs to CLEWs-EU .................................................................................................................................................. 49 3.3.1 Overview ...................................................................................................................................................................... 49 3.3.2 Development timeline ............................................................................................................................................ 52 3.3.3 Meetings (internal modelling team and external) ...................................................................................... 52 3.3.4 Completed development steps .......................................................................................................................... 54 3.3.5 Planned development steps ................................................................................................................................ 63 3.3.6 References for CLEWs-EU ..................................................................................................................................... 65 3.4 From GEMINI-E3 to GEMINI-E3 EU ............................................................................................................................... 66 3.4.1 Overview ...................................................................................................................................................................... 66 3.4.2 Model development timeline .............................................................................................................................. 69 3.4.3 Meetings (internal modelling team and external) ...................................................................................... 69 3.4.4 Completed development steps .......................................................................................................................... 70 3.4.5 Planned development steps ................................................................................................................................ 73 3.4.6 GEMINI-E3 EU References .................................................................................................................................... 76 3.5 From NEMESIS to NEMESIS-World ................................................................................................................................ 77 3.5.1 Overview ...................................................................................................................................................................... 77 3.5.2 Model development timeline .............................................................................................................................. 80 3.5.3 Meetings (internal modelling team and external) ...................................................................................... 81 3.5.4 Completed model development steps ............................................................................................................ 82 3.5.5 Planned model development steps .................................................................................................................. 87 3.5.6 References for NEMESIS World .......................................................................................................................... 88 Page vi D3.2 – Model development strategy – Update 1 3.6 From PROMETHEUS to OPEN-PROM ............................................................................................................................. 90 3.6.1 Overview ...................................................................................................................................................................... 90 3.6.2 Model development timeline .............................................................................................................................. 93 3.6.3 Meetings (internal modelling team and external) ...................................................................................... 93 3.6.4 Completed model development steps ............................................................................................................ 93 3.6.5 Current State of OPEN-PROM ............................................................................................................................ 95 3.6.6 Planned model development steps .................................................................................................................. 95 4. MODEL HARMONISATION PROTOCOL ....................................................................................... 97 4.1 Context ....................................................................................................................................................................................... 97 4.2 Relevant harmonisation context in DIAMOND WP3 development exercises ................................................ 97 4.3 Recommended parameter input values ........................................................................................................................ 97 4.4 Soft recommendations by November 2024 for model harmonisation (anticipating future WP5 and beyond-project use of models) ................................................................................................................................................ 99 4.5 SSP-Ageing Report comparison .................................................................................................................................... 100 4.6 Data harmonisation protocol references: .................................................................................................................. 101 Bibliography .................................................................................................................................... 102 ANNEX .............................................................................................................................................. 106 Table of Figures Figure 1. Regional disaggregation in GCAM-Europe. ...................................................................................................................... 13 Figure 2. Summary of the resource, technology, and end-use demand representation in GCAM-Europe................ 15 Figure 3. Planned electricity grid regions GCAM-Europe ............................................................................................................... 20 Figure 4. Work scheme for openTEPES to provide input parameters consistent with GCAM-Europe structure. .... 21 Figure 5. Net Demand (subtracting intermittent electricity supply) cluster representation (left) and technology contribution by cluster (right) based on openTEPES outputs ....................................................................................................... 21 Figure 6. Updated trade structure in GCAM-Europe ........................................................................................................................ 22 Figure 7. Residential energy service output per period, region, service, and decile for the GCAM baseline scenario (EJ). (Sampedro et al, 2022). ....................................................................................................................................................................... 23 Figure 8. Regional disaggregation in OMNIA. .................................................................................................................................... 27 Figure 9. Summary of the resource, technology, and end-use demand representation in OMNIA. ............................. 29 Figure 10. OMNIA timeline: development phases and steps ........................................................................................................ 30 Figure 11. Disaggregation procedure of UNSD Energy Balance.................................................................................................. 33 Figure 12. Natural gas extraction chain in OMNIA ............................................................................................................................ 34 Figure 13. Liquefaction and regasification process for Natural gas in OMNIA ...................................................................... 35 Figure 14. OMNIA Model Residential Sector Disaggregation: A detailed framework capturing global energy demand by breaking down fuel types, end uses, heating and cooling profiles, and future technological needs across regions and intensity zones, enabling precise modelling of residential energy consumption patterns and efficiency opportunities. ............................................................................................................................................................................... 37 Figure 15. Examples of different decomissioning profiles between GEO-TIMES and OMNIA ........................................ 39 Figure 16. Evolution of the manufactured gas representation within the model ................................................................. 40 Figure 17. Example of availability factor subdivided by potential tier ....................................................................................... 40 Page vii D3.2 – Model development strategy – Update 1 Figure 18. Aluminium sector production pathways in OMNIA. R&S = Refining and smelting, BY = Base year, H2 = Hydrogen. .......................................................................................................................................................................................................... 41 Figure 19. Improved representation of the steel sector in OMNIA ............................................................................................ 43 Figure 20. Visual representation of the approach to energy service demands projections. ............................................ 44 Figure 21. Soft-link OMNIA-ENGAGE ..................................................................................................................................................... 45 Figure 22. Modelling approach of OMNIA - MAgPIE soft-link ..................................................................................................... 47 Figure 23. Regional disaggregation in CLEWS-EU, disaggregating all 27 EU member states. ........................................ 50 Figure 24. Summary of the resource, technology, and end-use demand representation in CLEWS-EU. .................... 51 Figure 25. Simplified representation of CLEWs-EU energy module. .......................................................................................... 58 Figure 26. Primary level of land categories breakdown in the land module (CLEWs-EU). ................................................ 61 Figure 27. Simplified representation of CLEWs-EU land and water modules and illustration of key interactions with the energy module. ........................................................................................................................................................................................ 62 Figure 28. Regional disaggregation in GEMINI-E3 EU. .................................................................................................................... 67 Figure 29. Summary of the resource, technology, and end-use demand representation in GEMINI-E3-EU. ............ 68 Figure 30. New nested CES production structure – Industrial sectors (GEMINI-E3) ............................................................ 73 Figure 31. Regional disaggregation in NEMESIS-World ................................................................................................................. 78 Figure 32. Summary of the resource, technology, and end-use demand representation in NEMESIS-World. ......... 79 Figure 33. Gantt chart of the construction of the NEMESIS-World model .............................................................................. 81 Figure 34. General structure of MRIO databases ............................................................................................................................... 83 Figure 35. Regional disaggregation in OPEN-PROM. ...................................................................................................................... 91 Figure 36. Summary of the resource, technology, and end-use demand representation in OPEN-PROM. ............... 92 Figure 37. Timeline for the development of OPEN-PRΟΜ ............................................................................................................ 93 Figure 38. Comparison of preliminary OPEN-PROM pathways for CO2 emissions and final energy with different policy scenarios (upper graph) and with other IAMs for the 1.5C scenario (lower graph) ............................................... 96 Figure 39. Comparison of population and GDP projections for EU-27 between SSPs (v3.1) and EU Ageing Report 2024 ................................................................................................................................................................................................................... 100 Table of Tables Table 1. Overview of the contents of the initial survey...................................................................................................................... 4 Table 2. Summary of the stakeholder information processed by different modelling teams ........................................... 6 Table 3. Overview of the climate policies represented in each model. ..................................................................................... 10 Table 4. Overview of the energy, land and end-use demand measures represented in each model. .......................... 10 Table 5. Summary of model developments ......................................................................................................................................... 12 Table 6. List of model regions in GCAM-Europe ................................................................................................................................ 13 Table 7. GCAM-Europe kick-off meeting agenda .............................................................................................................................. 17 Table 8. GCAM-Europe meetings ............................................................................................................................................................. 17 Table 9. Country coverage of Eurostat data ......................................................................................................................................... 19 Table 10. List of model regions in OMNIA ........................................................................................................................................... 27 Table 11 OMNIA meetings key discussion points ............................................................................................................................. 31 Table 12. Fossil fuel extraction and processing data needs – For calibration years............................................................. 35 Table 13. Fossil fuel extraction and processing data needs – For future years ...................................................................... 36 Table 14. Sources used for the improvement of the power sector in Omnia ......................................................................... 38 Table 15. Data used in OMNIA aluminium sector ............................................................................................................................. 41 Page viii D3.2 – Model development strategy – Update 1 Table 16. Timeline for the finalisation of the disaggregated national CLEWs-EU model. ................................................. 52 Table 17. CLEWs-EU team meetings up to November 2024. ........................................................................................................ 53 Table 18. Naming convention sample for technologies and commodities in the energy module (CLEWs-EU)....... 55 Table 19. Naming convention sample for technologies in the land module (CLEWs-EU). ............................................... 56 Table 20. Naming convention sample for technologies in the water module (CLEWs-EU). ............................................. 56 Table 21. Key data requirements of the CLEWs-EU energy module. ......................................................................................... 59 Table 22. Key data requirements of the CLEWs-EU land and water modules. ....................................................................... 63 Table 23. GEMINI-E3 EU Regional description .................................................................................................................................... 67 Table 24. Timeline of the GEMINI-E3 EU progress and finalisation ............................................................................................ 69 Table 25. GEMINI-E3 EU Sectors description ....................................................................................................................................... 71 Table 26. Regions in Nemesis-World ...................................................................................................................................................... 78 Table 27. Summary of global MRIO databases. .................................................................................................................................. 84 Table 29. NEMESIS-World economic activities in each in region ................................................................................................ 86 Table 30. Relevant potential input data for WP3 model projections. All data is available through DIAMOND SharePoint ......................................................................................................................................................................................................... 98 Page 7 D3.2 – Model development strategy – Update 1 Topic GCAM-Europe OMNIA CLEWS-EU GEMINI-E3-EU NEMESIS-WORLD OPEN-PROM into highand low-intensity zones based on heating and cooling demands. Temporal demand profiles aligned with OMNIA's time slices capture hourly and seasonal variations, improving the model's accuracy. Power and Upstream - Explicitly model intra-EU trading of energy commodities and grid electricity, separated from extra-EU trade. - Include storage technologies in the electricity sector. - Implement different CCS retrofit technology options in the supply chain: refineries, fossil fuel power plants. Make sure all CCS options in the upstream sector are included. - Run sensitivities on hydrogen pipeline cost to study hydrogen transportation uncertainties. Add the option of existing natural gas pipeline retrofit into hydrogen pipelines. - Run sensitivities on extraEU trades to study a less dependent EU for energy supply. - Add new integrated storage technologies to power system modelling - Future technology options for electricity and heat generation included in the model will include fossiland biomass-fired combustion employing Carbon Capture and Storage (CCS) to assess the future importance of CCS (e.g., BECCS, gas with CCS). - Different storage options will be included in the model (e.g., batteries, pumpedhydro, green hydrogen production) to assess the importance and role of various means of storage in the energy transition. - The level of detail on crossborder electricity interconnections will be improved to allow projections on intra-EU electricity trade across scenarios. Potential implementation of CCS and BECCS plants in the model will be monitored to avoid overutilisation of these abatement technologies in simulations. - - Model CCS technologies such as BECCS and fossil fuel CCS plants - Model the need to have dispatchable energy sources available to support the system, such as GAS with CCS. - Model Storage in the power sector - Model which storage technologies are best suited for different countries/regions. In terms of cost and/or impacts on land use (e.g., for pumped hydro) Industry - Include DRI and H2 trade (at least intra-European) as an alternative option to industrial allocation. - Ensuring that the energy intensive sectors we are focusing on are represented in sufficient detail to provide insights on the role of CCS, material efficiency - Future technology options for industry included in the model will include fossiland biomass-fired combustion employing Carbon Capture and Storage (CCS). Sectoral reclassification: enable more reliable modelling of trade policies (taxes, duties, mandates and regulation) especially for hard to abate sectors. - In terms of capital restrictions, some policies acting on the cost of (imported) capital could be implemented in the NEMESIS model, but it - Model how circular economy policies impact decarbonisation and cost of mitigation in the cement sector - Model the Page 8 D3.2 – Model development strategy – Update 1 Topic GCAM-Europe OMNIA CLEWS-EU GEMINI-E3-EU NEMESIS-WORLD OPEN-PROM – Add material circularity options in steel, aluminum, cement. - Endogenously model efficiency costs in industry (potentially). and trade dynamics. We will also ensure that the OMNIA-ENGAGE model linkage is fit for purpose and able to address the important issues they raised. - The relevant technology options will be expanded to include biomass, solar thermal and green hydrogen to assess what the future role of renewable fuels will be in the industrial sector. means designing complete scenarios. - Disaggregate the results of employment by sector, educational level, types of jobs and gender, to model the potential winners and losers from energy transition. - To better represent financial aspects NEMESIS currently improved the modelling of capital in the EU version of the model by including several layers related to capital markets: central bank monetary rule, public. Nevertheless, so far it does not concern technologies per se. decarbonisation of Chemical and Petrochemical Sector (and reflect on the issue of data availability) - Model the role of renewable fuels in the industrial sector (i.e. Biofuels and Hydrogen). Water, land and other sectors - - - Detail in the model will be improved to capture climate change impact on crops and to assess impacts across the sectors on energy-water-land interlinkages. - In order to account for climate uncertainty, use of the FAO GAEZ tool will be adopted to derive different crop yield under different climate scenarios. - The level of detail in the land module will be improved to allow representation of energy content and CO2 emissions of different biomass options. - - - Modelling of Climate change impact on crops and to what extent the energy-water-land loop is completed in the models. - Representation of energy content and CO2 emissions of biomass. Page 9 D3.2 – Model development strategy – Update 1 Apart from the direct feedback to be use by the different teams to improve their models (WP3), stakeholders also raised some relevant feedback that could be further explored in WP4, which focuses on integrating IAMs with sectoral or bottom-up models to broaden the overall scope of analysis. In terms of high-level and cross sectoral research themes, several stakeholders outlined the importance of incorporating climate feedbacks into the models, particularly into the power sector, and into the projections of future crop yields and water availability. Integrating potential climate feedbacks is a recognised research priority for the integrated assessment community (Khan et al., 2021; Seneviratne et al., 2024), and is planned to be explored by the different modelling teams in WP4. The stakeholders also highlighted the need to consider the health impacts of air pollution, and the potential health co-benefits of decarbonisation. While it is not explicitly considered in this project, it is an active research line for some of the modelling teams (Sampedro et al., 2023) and could be further explored during the course of the project. For the residential and transport sectors, some modelling teams are planning to expand the model representation of different transportation modes to be able to capture the environmental implications of modal shifts (e.g., in WP5), which is an important option, considering the existing barriers for decarbonising the transport sector. Regarding the power sector, the consulted stakeholders showed their interest on improving temporal granularity (to better capture benefits of storage); modelling decentralised electricity systems, including energy communities; and representing intra-EU electricity trade and changes in the interconnector capacity. Stakeholders also provided feedback on the industry sector that could be explored by the modelling teams during the project, including the explicit consideration of carbon transportation costs or modelling additive manufacturing. Finally, on the land and water sectors, the consulted stakeholders showed interest on the potential integration of nutritional values of the different food sources into the models, as well as the representation of multifunctional land uses (e.g., agroforestry), wildfires, or biodiversity impacts. The Requirements Workshop was the key vessel for our stakeholder engagement approach to directly influence the model development strategy. However, as more engagement activities are planned for the next period, more insights may be useful for model development. The modelling teams will monitor these discussions and are prepared to further adapt this strategy based on potential feedback received. Any further such changes will be reported in the next update of the model development strategy. 2.3 Representation of policies and measures As part of the initial development plans, the different modelling teams also described the climate policies and broader energy, land and demand-side measures that could be represented with the models in the consortium. The implementation of different policies into the models is relevant for WP5, which focuses on exploration of alternative scenarios. Stakeholder input, as described in the previous section, is designed to steer the definition of policies in WP5, and therefore this information is crucial for the potential representation of policies and measures in the models in the context of WP3. Considering that IAMs are explicitly designed to analyse the effects and impacts associated with alternative climate policies, the original versions of the models within DIAMOND are already able to directly represent a suite of climate policies without any further modification, such as carbon prices, emission quotas, climate targets, energy tax/subsidies, or land protection policies. In addition, IAMs can also represent a wide range of economy-wide measures that are essential in a transition towards a carbon neutral economy. These measures include the use of cleaner or more efficient fuels or technologies in different end-use sectors (buildings, industry, transport, and agriculture), or additional mitigation options such as direct air capture, afforestation, or behavioural changes. The original models can also usually represent some of these options either endogenously or exogenously. The current development strategies included detailed plans for the improvement of the representation of alternative climate policies (Table 3) and additional energy-land-demand side measures (Table 4). Page 10 D3.2 – Model development strategy – Update 1 Table 3. Overview of the climate policies represented in each model. Category Policy measure GCAMEurope OMNIA CLEWSEU GEMINIE3-EU NEMESISWorld OPENPROM Emissions mitigation Tax Emissions target / quota Regulations (emissions standards, etc…) Global Temperature / Radiative Forcing target Financial supports (negative emissions, CDMs, Green Climate Fund) Energy Tax Subsidy Energy mix target Efficiency target Regulations (thermal regulation in buildings, banning of diesel cars in urban areas, etc…) Land Protected lands Production quotas Carbon sink pricing / Land use change emissions tax Afforestation targets Trade Carbon border tax on imports Carbon border supports on exports Regulations policies (certifications, Bestavailable technologies, standards, etc…) Policy not applicable in original or updated model Policies that could be directly implemented in the original model version Policies that could be implemented with the original version of the model, but with some modifications Policy is planned to be represented in the updated model version Option to implement the policy in the updated model version, but it still needs to be decided Table 4. Overview of the energy, land and end-use demand measures represented in each model. Category Policy measure GCAMEurope OMNIA CLEWSEU GEMINIE3-EU NEMESISWorld OPENPROM Buildings Fuel switch/energy mix Technology substitution Efficiency improvement Industry Fuel switch/energy mix Page 11 D3.2 – Model development strategy – Update 1 Category Policy measure GCAMEurope OMNIA CLEWSEU GEMINIE3-EU NEMESISWorld OPENPROM Technology substitution Efficiency improvement CCS Combined heat and power Fuel switch/energy mix Technology substitution Efficiency improvement CCS Transport Fuel switch/energy mix Technology substitution Efficiency improvement Modal shifts Agriculture Fuel switch/energy mix Technology substitution Efficiency improvement Land practices Agroforestry Animal husbandry practices Direct Air Capture Direct Air Capture Land Use, Land-Use Change and Forestry (LULUCF) Afforestation Behavioural Changes Behavioural Changes The measure is not represented in the original or updated model version The measure is represented endogenously in the original model version The measure is exogenous in the original version of the model The measure is planned to be incorporated or endogenised in the updated version of the model The representation or endogenisation of the measure in the original model version needs to be decided Page 12 D3.2 – Model development strategy – Update 1 3 UPDATED MODEL DEVELOPMENT STRATEGIES Table 5 summarises the planned model developments, and the following subsections detail the strategies of the six global models in the consortium. Table 5. Summary of model developments Original model New model Model type GCAM GCAM-Europe Partial equilibrium TIAM OMNIA Partial equilibrium Global CLEWS/GLUCOSE CLEWS-EU Linear optimisation GEMINI-E3 GEMINI-E3 EU Computable general equilibrium (CGE) NEMESIS NEMESIS-World Macroeconomic/Econometric PROMETHEUS OPEN-PROM Partial equilibrium Regarding the timeline, each modelling team has started the development of the model enhancements in Month 4 (M4), as indicated in D3.1. Likewise, the exploration of potential interconnection options with additional bottomup models (WP4) for some research areas (e.g., representation of labour/finance dynamics) has started at the very early stages of the project (M1). The following sections outline the updated model-specific development strategies per model (sections 3.1-3.6), which now incorporate stakeholder insights and suggestions from the dedicated “stakeholder requirements” workshop. This process ensures that the models will capture their needs and therefore will be relevant for policymaking. 3.1 From GCAM to GCAM-Europe 3.1.1 Overview GCAM (Global Change Analysis Model) is a dynamic-recursive model with technology-rich representations of the economy, energy sector, land use and water linked to a climate model that can be used to explore climate change mitigation policies including carbon taxes, carbon trading, regulations and accelerated deployment of energy technology. Regional population and labour productivity growth assumptions drive the energy, land-use and water systems employing numerous technology options to produce, transform, and provide energy services as well as to produce agricultural and forest products, and to determine land use and land cover. The model has been used to explore the potential role of emerging energy supply technologies and the impact on greenhouse gas emissions of specific policy measures or energy technology adoption including; carbon dioxide (CO2) capture and storage, bioenergy, hydrogen systems, nuclear energy, renewable energy technology, and energy demand in buildings, industry and the transportation sectors. Outputs of the model include projections of future energy supply and demand and the resulting greenhouse gas emissions, radiative forcing and climate effects of 16 greenhouse gases, aerosols and short-lived species, implications for future land cover and water use, contingent on assumptions about future population, economy, technology, and climate mitigation policy. Through separate modules, results in terms of emissions and water use can be used to estimate pre-mature mortality and water shortages by region. Page 13 D3.2 – Model development strategy – Update 1 For the development of GCAM-Europe, geographical disaggregation is increased for the European continent. GCAM by default divides the world in 32 regions, and the European continent is divided in five different regions: EU-12, EU-15, Europe Eastern, Europe-non-EU, and European Free Trade Association (EFTA). In GCAM-Europe, all European countries are disaggregated into individual model regions (see Figure 1 and Table 6). Having this level of detail will enable us to explore the country-level effects of European policy packages or transformational strategies, as well as the potential international effects (e.g., carbon leakage) for a representative set of nonEuropean regions (27) over the world. In terms of sectors, the model expands (for the newly created European countries, and there may be differences by country based on data availability) in various dimensions (Figure 2). The representation of electricity grids and regional trade, as well as demand segments, allows to have more sectoral detail in the power sector by reflecting energy storage. In final energy demand, predominantly in building energy demand, new demand categories are included as well as new technologies like heat pumps, driven by high data availability for European countries. Behind the overall sectoral representation, there is a further deep-dive in terms of consumer group representation, which is highlighted in section 3.1.5. Figure 1. Regional disaggregation in GCAM-Europe. Yellow-to-red colours represent the regional groups that are explicitly disaggregated for this model. White-to-blue regions represent model regional groups that were part of the original model. Table 6. List of model regions in GCAM-Europe Model region (default) Model region (new) Model region (new) Africa_Eastern Albania Lithuania Africa_Northern Belarus Luxembourg Africa_Southern Belgium Macedonia Africa_Western Bosnia and Herzegovina Malta Argentina Bulgaria Moldova Australia_NZ Croatia Netherlands Brazil Cyprus Norway Canada Czech Republic Poland Central America and Caribbean Denmark Portugal Central Asia Estonia Romania China Finland Serbia and Montenegro Colombia France Slovakia Page 14 D3.2 – Model development strategy – Update 1 India Germany Slovenia Indonesia Greece Spain Japan Hungary Sweden Mexico Iceland Switzerland Middle East Ireland Turkey Pakistan Italy United Kingdom Russia Latvia Ukraine South Africa South America_Northern South America_Southern South Asia South Korea Southeast Asia Taiwan D3.2 – Model development strategy – Update 1 Page 15 Figure 2. Summary of the resource, technology, and end-use demand representation in GCAM-Europe. White boxes indicate features already part of the original model. Grey boxes indicate features not part of the core nor the updated model. Dark yellow boxes indicate the features that will be explicitly developed for GCAM-Europe in the DIAMOND project, and which are not in the core version of the model D3.2 – Model development strategy – Update 1 Page 16 3.1.2 Development timeline • January 2024: Working “alpha” version with all countries as separate regions within global core structure • December 2024: Updated version (GCAM-Europe 7.1-Beta) o All data replaced by specific European data sources where available (Eurostat, ENTSO-E) o Grid regions for electricity trade between countries, based on openTEPES assumptions o Basic sectoral expansions based on data availability (e.g. building services and technologies, transport fuels) o Adapted commodity trade structure to represent EU trading block o Updated with GCAM core improvements (GCAM 7.1) o Open-source access through Github and technical description of model to be written and published in Q1 2025 in peer-reviewed journal and the I2AM PARIS Platform • June 2025: Updated version with initial expansions / WP4 links: o Household consumption based on actual HBS data for household groups (e.g. income levels) o Improvements in industry coverage and structure (ENGAGE) o Electricity module further updated emulating open-TEPES interactions o Other relatively straightforward potential improvements as listed in section 3.1.6.3 o Updated with GCAM core improvements (including potential base year update) • December 2025: Further updated version with deeper expansions / WP4 links: o Behavioural change aspects incorporated throughout model (UNIBAS) o Potential links with labour/financial modules (UM) o More challenging potential improvements as listed in section 3.1.6.3 o Updated with GCAM core improvements • June 2026: Final version (GCAM-Europe 7.X): o Potential bug / consistency fixes relative to previous version o Updated with GCAM core improvements o Full documentation (contents TBD) written down in D3.6 and academic publication o Open access branch ready through GitHub 3.1.3 Meetings (internal modelling team and external) The development of GCAM-Europe is a gradual process in which the European region is being disaggregated within the global version of the model through a step-by-step process. Largely, the structure of the model will follow that of the global version, with several additional deep-dives in sectors and sector-links based on data availability, similarly as in GCAM-USA (Binsted et al., 2021) and GCAM-China (Cui et al, 2021), among other regional GCAM versions. To coordinate efforts and gather input from developers at BC3, ICCS, and UMD, while considering stakeholder feedback (guided by ETH I the context of WP2 activities), a GCAM-Europe kick-off meeting was held D3.2 – Model development strategy – Update 1 Page 23 Figure 7. Residential energy service output per period, region, service, and decile for the GCAM baseline scenario (EJ). (Sampedro et al, 2022). Due to the lack of available harmonised data at a global level, in the core GCAM model the allocation of different residential energy services across subregional consumer groups is made using exogenous assumptions based on the form of the demand function within each of the sectors. In GCAM-Europe, we will improve this representation by using empirical data from the country-level Household Budget Surveys (HBS), which are available for all member states. In those European (non-EU) countries, where these data are not accessible, we will follow the process developed for core GCAM, based on the used functional forms. Beyond the residential sector, the implementation of consumer heterogeneity in further end-use sectors, namely food, transportation and municipal water, is expected to be integrated into the core in the near-term4. GCAMEurope will coordinate with the GCAM development team to implement multiple consumers in these end-use sectors following a similar structure that will be consistent will the core model. However, unlike the core model, GCAM-Europe will have a better representation of consumer heterogeneity by incorporating empirical country level household data (HBS). 3.1.5.2 Further WP4 links The links with additional sectoral and bottom-up models are in the preliminary phase, considering that the main focus has been the development of a beta version of GCAM-Europe. However, some initial conversations have been carried out with different modelling teams, in order to discuss potential integration activities: 4 Some initial implementation of consumer heterogeneity in multiple endues demand sectors has been developed during the GRAPHICS project: https://github.com/jonsampedro/gcam-core/tree/GCAM_GRAPHICS_v7 D3.2 – Model development strategy – Update 1 Page 24 • Link between GCAM and ENGAGE (UCL): The planned expansion of the industrial sector in GCAM-Europe will include inputs and insights of the ENGAGE module to develop additional capacities, including for example the endogenizing the availability and use of secondary metals relative to primary metals. • Adding household heterogeneity into GCAM (CESAR): The implementation of multiple consumers in GCAM-Europe, based on country-level HBS will be developed in coordination with the CESAR team. In addition, having the household structures for the different countries will allow to connect the model outcomes with “Medusa”5, an R package aimed at estimating the distributional impacts of alternative policies. While this package has originally been developed for Spain, the expansion to different EU member states and its connection with GCAM-Europe will be developed between BC3 and CESAR. • Labour and financial module (UM): GCAM will use inputs from UM to modify its default capital costs, (Weighted Average Cost of Capital, WACC), and other financial variables (e.g., discount rates), in order to have values which are differentiated by region and technology. Furthermore, the energy demand module developed at UM, focusing on differentiated household preferences with respect to large investments, is planned to be emulated in GCAM to reflect more realistic household investment behaviour. • Behavioural change (UNIBAS): Use of findings from studies on consumer behaviour to improve the parameters included in the different end-use demand functions, which will be particularly relevant when multiple consumers are implemented. For example, the adjustment of the income and price elasticities, or preference (“shareweight”) parameters for different countries and within-region consumer groups will be a direct and timely connection of the models. • openTEPES electricity model (IIT): After developing the initial GCAM-Europe version based on electricity grid regions as outlined above, the results of this version will be compared with comparable results from openTEPES to identify structural differences that may be better represented in openTEPES due to simplified assumptions in GCAM-Europe. Subsequently, the electricity module in GCAM-Europe may be adapted where necessary, with the aim to emulate openTEPES as detailed as possible. 3.1.5.3 Further sectoral coverage expansions to be explored Detailed, well-structured and updated data for European countries allows for potential sectoral expansions, of which the following ones have already been identified: • Energy supply sectors: • Improved district heat representation (crucial in Scandinavian and former Soviet-Union countries) • Hydrogen trade between European regions including e.g. blending in natural gas grids • Explicit representation of biogas potential (potentially linked to agricultural sector) • Industry energy demand: • More disaggregated industrial sectors to cover • Explicit representation of investments in efficiency improvements • Building energy demand: 5 https://github.com/bc3LC/medusa D3.2 – Model development strategy – Update 1 Page 25 • More disaggregated energy services in residential and commercial buildings, i.e. through detail in JRC-IDEES (Rozsai et al, 2024) • Representation of renewable heat pumps (aerothermal, geothermal, hydrothermal, solar thermal; link with cooling) • Endogenous representation of building efficiency improvements • Transport energy demand: • Separation of gasoline, diesel, LPG vehicles (improved representation of emission standards and air pollution) • Improved representation of transportation demand by country based on JRC-IDEES (Rozsai et al, 2024) • Separation of intra-EU and extra-EU international passenger transport Whether and when these expansions will be included depends on modelling feasibility and time/capacity constraints. 3.1.6 GCAM-Europe references - Armington, P.S., 1969. A Theory of Demand for Products Distinguished by Place of Production (Une théorie de la demande de produits différenciés d’après leur origine)(Una teoría de la demanda de productos distinguiéndolos según el lugar de producción). Staff Papers-International Monetary Fund 159–178. - Binsted, M., Iyer, G., Patel, P., Graham, N., Ou, Y., Khan, Z., Kholod, N., Narayan, K., Hejazi, M., Kim, S., 2021. GCAM-USA v5. 3_water_dispatch: Integrated modeling of subnational US energy, water, and land systems within a global framework. Geoscientific Model Development Discussions 1–39. - Cui, Ryna Yiyun, et al. "A plant-by-plant strategy for high-ambition coal power phaseout in China." Nature communications 12.1 (2021): 1468. - ENTSOE (2024), “TYNDP 2024: Draft Scenarios Report”. May 2024. https://2024.entsos-tyndpscenarios.eu/ - Rozsai, Mate; Jaxa-Rozen, Marc; Salvucci, Raffaele; Sikora, Przemyslaw; Tattini, Jacopo; Neuwahl, Frederik (2024): JRC-IDEES-2021. European Commission, Joint Research Centre (JRC) [Dataset] PID: http://data.europa.eu/89h/82322924-506a-4c9a-8532-2bdd30d69bf5 - Sampedro, J., Iyer, G., Msangi, S., Waldhoff, S., Hejazi, M., Edmonds, J.A., 2022. Implications of different income distributions for future residential energy demand in the US. Environmental Research Letters 17, 014031. D3.2 – Model development strategy – Update 1 Page 26 3.2 From TIMES-GEO to OMNIA 3.2.1 Overview TIMES is a modelling platform for local, national or multi-regional energy systems, which provides a technologyrich basis for estimating how energy system operations will evolve over a long-term, multiple-period time horizon (Loulou and Labriet, 2007). These energy system operations include the extraction of primary energy such as fossil fuels, the conversion of this primary energy into useful forms (such as electricity, hydrogen, solid heating fuels and liquid transport fuels), and the use of these fuels in a range of energy service applications (vehicular transport, building heating and cooling, and the powering of industrial manufacturing plants). In multi-region versions of the model, fuel trading between regions is also estimated. The TIMES framework is usually applied to the analysis of the entire energy sector but may also be applied to the detailed study of single sectors (e.g., the electricity and district heat sector). The framework can also be used to simulate the mitigation of non-CO2 greenhouse gases, including methane (CH4) and nitrous oxide (N2O). The TIMES-GEO Model Integrated Assessment Model, TIAM, is the multi-region, global version of TIMES, which combines an energy system representation of different regions with options to mitigate non-CO2 greenhouse gases as well as non-energy CO2 mitigation options. It can be used to explore a variety of questions on how to mitigate climate change through energy system transformations, as well as reductions in non-energy CO2 emissions and non-CO2 emissions. The Open-source MuNdus Integrated Assessment model (OMNIA) is a customised, open-access global IAM based on the TIMES framework. It provides a technology-rich basis for estimating how energy system investments and operations evolve over a long-term, multiple-period time horizon. It is a partial equilibrium, ‘perfect foresight’, welfare cost optimisation model. The OMNIA model integrates and expands the TIMES-GEO. Regional representation is configured on 29 regions (see Table 10 and Figure 8). Key technologies to decarbonise hard-toabate sectors are comprehensively represented and industry representation is enhanced based on ongoing research. The upstream sector, including fossil fuel extraction, is able to fully explore the implications of the transition. OMNIA also incorporates a spatially explicit representation of key energy carrier trades, including bioenergy, hydrogen, as well as commodities including captured CO2. Temporal representation of energy demand and supply is increased to properly model the increased penetration of weather-dependent renewables, storage, and new demand technologies, expanding the time slices to 20. OMNIA also integrates a climate module to estimate global average temperature changes of scenarios and includes abatement supply-cost curves for all major non-CO2 options, based on up-to-date assessments. In addition, with a newly calibrated and improved water module, the model is able to evaluate the interdependence between energy production and water consumption. D3.2 – Model development strategy – Update 1 Page 27 Figure 8. Regional disaggregation in OMNIA. Yellow-to-red colours represent the regional groups that are explicitly disaggregated for this model. White-to-blue regions represent model regional groups that were part of the original model. Table 10. List of model regions in OMNIA Model regions Description List of countries AFE Eastern Africa Ethiopia, Kenya, Sudan, Mauritius, Eritrea, South Sudan, Burundi, Comoros, Djibouti, Madagascar, Réunion, Rwanda, Somalia, Uganda AFN Northern Africa Egypt, Algeria, Morocco, Libya, Tunisia AFW Western Africa Democratic Republic of the Congo, Cote d Ivoire, Ghana, Cameroon, Gabon, Benin, Senegal, Togo, Niger, The Republic of Congo, Burkina Faso, Cape Verde, Central African Republic, Chad, Equatorial Guinea, Gambia, Guinea, Guinea-Bissau, Lesotho, Liberia, Malawi, Mali, Ma uritania, Sao Tome and Principe, Seychelles, Sierra Leone, Swaziland AFZ Southern Africa Tanzania, Angola, Mozambique, Zimbabwe, Zambia, Botswana, Namibia, South Africa NIG Nigeria Nigeria RUS Russian Federation Russian Federation ASC Central Asia Kazakhstan, Uzbekistan, Turkmenistan, Azerbaijan, Mongolia, Georgia, Kyrgyzstan, Armenia, Tajikistan, Afghanistan ASE Southeast Asia Thailand, Malaysia, Singapore, Myanmar, Cambodia, Brunei Darussalam, Lao People’s Democratic Republic, Democratic People's Republic of Korea, Cook Islands, East Timor, Fiji, French Polynesia, Kiribati, Macau, Maldives, New Caledonia, Palau, Papua New Guinea, Samoa, Solomon Islands, Tonga, Vanuatu CHN China Mainland People's Republic of China, Hong Kong IDN Indonesia, Philippines, Vietnam Indonesia, Philippines, Vietnam IND India India D3.2 – Model development strategy – Update 1 Page 28 ASO South Asia Bangladesh, Nepal, Sri Lanka, Pakistan, Bhutan JPN Japan Japan SKT South Korea, Taiwan Taiwan, South Korea ANZ Australia and New Zealand Australia, New Zealand USA United States United States CAN Canada Canada LAM Latin America Argentina, Venezuela, Colombia, Peru, Trinidad and Tobago, Ecuador, Guatemala, Cuba, Bolivia, Dominican Republic, Honduras, Paraguay, Uruguay, Costa Rica, El Salvador, Haiti, Panama, Nicaragua, Jamaica, Curacao, Suriname, Antigua and Barbuda, Aruba, Bahamas, Barbados, Belize, Bermuda, British Virgin Islands, Cayman Islands, Dominica, Falkland Islands (Malvinas), French Guiana, Grenada, Guadeloupe, Cooperative Republic of Guyana, Martinique, Montserrat, Puerto Rico, Saba, Saint Eustatius, Saint Kitts and Nevis, Saint Lucia, Saint Pierre and Miquelon, Saint Vincent and the Grenadines, Sint Maarten, the Turks and Caicos Islands BRA Brazil Brazil MEX Mexico Mexico CHL Chile Chile ENE Non-EU Eastern Europe Ukraine, Belarus, Serbia, Bosnia and Herzegovina, Moldova, Republic of North Macedonia, Kosovo, Albania, Montenegro ENW Non-EU Western Europe Norway, Switzerland, Iceland, United Kingdom, Gibraltar EUE Eastern Europe Union Poland, Czech Republic, Romania, Hungary, Bulgaria, Slovak Republic, Croatia, Lithuania, Slovenia, Estonia, Latvia EUW Western Europe Union Germany, Netherlands, Belgium, Sweden, Austria, Finland, Denmark, Ireland, Luxembourg EUM MediterraneanEurope Union France, Italy, Spain, Greece, Portugal, Cyprus, Malta MEA Middle East (Gulf States) Iran, Saudi Arabia, United Arab Emirates, Iraq, Qatar, Kuwait, Oman, Bahrain, Yemen MDA Mediterranean Asia Turkey, Israel, Syrian Arab Republic, Jordan, Lebanon D3.2 – Model development strategy – Update 1 Page 29 Figure 9. Summary of the resource, technology, and end-use demand representation in OMNIA. White boxes indicate features already part of the original model. Grey boxes indicate features not part of the core nor the updated model. Dark yellow boxes indicate the features that will be explicitly developed for OMNIA in the DIAMOND project, and which are not in the core version of the model. D3.2 – Model development strategy – Update 1 Page 30 3.2.2 Development timeline By the end of the year (December 2024), work on OMNIA focuses on delivering: • The final design for the demand projection module (working and usable version of the module, including reporting) • the new model for industrial sub-sectors • improved power and residential sectors • the calibrated transport sector for OMNIA regions based on 2019 UN energy balance and the preliminary report • the new updated upstream sector including the new supply chain for fossil fuels, and the new module for hydrogen supply chain • the key inputs file for transport and the bioenergy supply chain A first version of the new improved sectors is planned to be ready by December 2024. During the first quarter of 2025 the team will start to integrate and test all the developed main sectors (upstream, power, industry and residential, calibrated version of transport) and will complete the development of the remaining sectors (transport) and the additional modules (climate, water). Testing of the modules is scheduled to be continued until July 2025. The first official release of the model will be released as part of MS8) and feature all new and updated sectors and modules. In the second half of 2025, the team will carry out full model testing, including scenario analysis and will integrate all the necessary revisions and improvements to sectors and modules. The OMNIA timeline is presented in Figure 10. Figure 10. OMNIA timeline: development phases and steps 3.2.3 Meetings (internal modelling team and external) Several meetings were held in the two first years of the project for Omnia development including recurring meetings within the Omnia team, bilateral/multilateral meetings for the harmonisation process. D3.2 – Model development strategy – Update 1 Page 31 Table 11. OMNIA meetings key discussion points Type of meeting, description and dates Date Physical (Turin) Kick off Meeting First Design Meeting. Detailed planning of the improvements at sectoral level 3rd March 2023 Online Decision about priorities of model development and update of the GANTT 29th March 2023 Physical (Turin) NEMESIS-World, OMNIA Bilateral workshop in Torino: Discussions on how to share information between models and ensure possible soft linkage (E4SMA & SEURECO) 24th of August 2023 Hybrid (Maastricht) All Models Maastricht Workshop: How to improve finance, labour and distributional modelling 21st and 22nd of September 2023 Online Soft-Link Discussion about upstream design and ENGAGE softlink with industry sector 25th September 2023 Online OMNIA Discussion about the energy balance update, and soft links with openTEPES (COMILLAS) and behavioural change (UNIBAS) 23rd October 2023 Online OMNIA Discussion about the implementation of the new design of industry sector, proposal for the first release of OMNIA, planning of workshops and soft-linking activities 5th February 2024 Online Soft Link. Preliminary meeting for Climate and Water module integration (E4SMA and CICERO) 5th March 2024 Online Workshop on demand projections organised by UCL and with the participation of CICERO for WP5 integration 28th March 2024 Online Workshop Discussion on data source and preliminary design options for the OMNIA and ENGAGE soft link (OMNIA and UCL) 9th April 2024 Online Regional disaggregation proposals, Internal model sharing and workflow 23rd April 2024 Online Discussion on the possible links and data needed for the implementation of behavioural model (E4SMA, CICERO and UNIBAS) 23rd April 2024 Online Discussion on required disaggregation for the bioenergy sector in Omnia according to the MAgPIE granularity (E4SMA, UCL and CICERO) 14th May 2024 Online Update of the activities and implementation of improvements for the power sector, management of activities for the May GA in Lausanne 15th May 2024 Online Regional disaggregation finetuning, feedback on industry and upstream current model files to start the scoping phase 23rd May 2024 Physical (Lausanne) Discussion on the parameters exchanged between model and information needed for the surveys. 24th May 2024 Online Discussion about the energy balance for industry sector UCLE4SMA 8th July 2024 Online Discussion about the preliminary design of the residential and transport sector Imperial College-E4SMA 9th July 2024 Online Presentation of the advancements for the upstream, industry, residential and power sector. 12th September 2024 D3.2 – Model development strategy – Update 1 Page 32 Online Discussion about the definitive structure of the residential sector between Imperial College and E4SMA 19th September 2024 Online Bilateral Meeting Omnia-Nemesis World Discussion about data harmonisation between the two models 27th September 2024 3.2.4 Completed development steps The following sections cover key advancements and methodologies applied across various OMNIA model components, including updates to energy balances, sector-specific developments, and model linkages. Each subsection begins with the latest progress in updating energy data, restructuring supply chains, and advancing sectoral representations. This includes model enhancements for residential, power, industry, and water sectors, alongside detailed integration methods between OMNIA and other models. Data collection processes, methodology improvements, and specific structural adjustments are introduced for each module, providing a comprehensive overview of the ongoing work and the data sources underpinning these advancements. 3.2.4.1 Energy balance update Progress: The energy balance was successfully delivered at the beginning of the second year of the project and applied for the development of all sectors: The base year has been updated to 2019. The sources that have been used are: - UNSD Energy balances (https://data.un.org/SdmxBrowser/start) - UNSD Energy statistics (https://data.un.org/SdmxBrowser/start) The UNSD Energy statistics have been used to disaggregate the Energy Balances, which in the UNSD database are represented only with aggregated commodities: 1. Primary coal and peat 2. Coal and peat products 3. Primary Oil 4. Oil products 5. Biofuels and waste 6. Nuclear 7. Electricity 8. Heat 9. Renewables The procedure to disaggregate the energy balance is shown in the following Figure. D3.2 – Model development strategy – Update 1 Page 39 start year to the year when the plant was expected to be decommissioned. To improve this approach, a new set of decommissioning profiles was developed. These profiles now utilise a more detailed, data-driven method, incorporating information from the Global Energy Monitor (GEM) database. This GEM-based database covers more than 90% of global power plant capacity and includes detailed commissioning year data for each plant (source: Global Energy Monitor). By applying the technical life values from the previous model, we can now estimate the expected decommissioning year for each plant more accurately, resulting in a more realistic capacity retirement schedule. This improvement allows for the decommissioning of specific capacity percentages each year, based on actual plant lifespans rather than the generic average. Figure 15. Examples of different decomissioning profiles between GEO-TIMES and OMNIA Introduction of Manufactured Gas Commodity A new commodity, Manufactured Gas, has been introduced into the power sector model. This commodity aggregates three types of gases that were previously classified under Natural Gas: • Blast Furnace Gas • Coke Oven Gas • Gasworks Gas This disaggregation allows for a more accurate assessment of emissions, as the individual properties of these gases differ significantly from natural gas. By considering these gases separately, the model can evaluate plant emissions with higher precision. Additionally, a new set of technologies has been developed to handle the consumption of Manufactured Gas, adopting the same techno-economic characteristics as the natural gas technologies. D3.2 – Model development strategy – Update 1 Page 40 Figure 16. Evolution of the manufactured gas representation within the model New Storage Technology: Pumped Hydro Pumped Hydro has been added as a new storage technology in the model. This technology has been characterised using a detailed study by the National Renewable Energy Laboratory (NREL). Pumped Hydro offers significant potential for large-scale energy storage by using water to store energy in the form of potential energy, which can be converted back to electricity as needed. The techno-economic characterisation provided by NREL includes cost data, efficiency rates, and performance parameters, which have been integrated into the model. New Structure for Renewable Technologies: Solar and Wind Significant structural modifications have been made to the solar and wind technologies with respect to TIMESGEO model. In the original model, these technologies maintained the same availability factor regardless of the installed capacity. With these adjustments, however, the total potential of each technology has been divided into four tiers for wind technologies and three tiers for solar technologies, with each tier assigned a different availability factor. This allows the model to prioritize the installation of tiers with the highest availability factors first, moving to lower tiers only as capacity expands. Figure 17. Example of availability factor subdivided by potential tier 3.2.4.5 New modelling approach for Industry Recent progress: New improved more technology explicit representation for iron and steel, aluminium and cement design scoping completed. Data collection, energy balance disaggregation and first draft of design implementation D3.2 – Model development strategy – Update 1 Page 41 completed for aluminium in Oct 2024. For iron and steel and cement this is due to be completed in Dec 2024. As set out in MS5, our initial plans were to develop more technology rich representations of iron and steel, ammonia, aluminium and cement. Due to limitations with the UN energy balances that underpin the model we were unable to break out ammonia, i.e. because petrochemical feedstocks are not included in the balances. Nevertheless, this was not a sector that we intended to link with ENGAGE and so this does not impact our wider modelling plans. Therefore, within OMNIA the following sectors will be represented in a detailed, technology explicit manner (including primary and secondary production): • Iron and steel • Aluminium • Cement The structure of the aluminium sector which has been implemented in its current iteration is shown in Figure 18. The Base Year (BY) data (see Table 15) is used to model primary and secondary production in the base year. Process emissions from anode degradation are represented. For future years, new primary production options include a replacement fossil route using natural gas, an electricity only route and a route using hydrogen to provide the necessary thermal energy for refining and anode production. For the electricity and hydrogen routes, options are available to capture process emissions using CCS or use inert anodes that do not product process CO2. Future secondary production has a new fossil option and two low carbon routes, electricity only and hydrogen for thermal energy. Figure 18. Aluminium sector production pathways in OMNIA. R&S = Refining and smelting, BY = Base year, H2 = Hydrogen. The rest of the non-ferrous metals sector is then modelled using a generic energy service approach. The references for the data used in both aluminium and other non-ferrous metals will be made available in the model documentation. A sample of which are included in Table 15. Table 15. Data used in OMNIA aluminium sector Dataset Description Source D3.2 – Model development strategy – Update 1 Page 42 Energy balance Non-ferrous sector energy balances UN statistics Production statistics Primary aluminium production available from the US Geological Survey. Secondary aluminium production at regional level. Given the lower granularity of the secondary data, assumptions are made to disaggregate it. Primary: https://www.usgs.gov/centers/national-mineralsinformation-center/aluminum-statistics-andinformation Secondary: https://alucycle.international-aluminium.org/publicaccess/public-global-cycle/ https://www.statista.com/markets/407/topic/434/m etals/#overview Historic input energy mix Energy mix into primary and secondary aluminium production in base year. For secondary production, the required energy mix (electricity and thermal) is taken from the cited paper and assumptions made about what provides the thermal energy in each region in the base year. Primary production (2019): https://international-aluminium.org/resource/2019life-cycle-inventory-lci-data-and-environmentalmetrics/ Secondary production: https://www.mdpi.com/2673-4141/4/1/7 Technology parameterisation Data aggregated as necessary, i.e. we aggregate refining and smelting. https://missionpossiblepartnership.org/wpcontent/uploads/2023/03/MPP-AluminiumTechnical-Appendix.pdf https://european-aluminium.eu/wpcontent/uploads/2023/11/23-11-14-Net-Zero-by2050-Science-based-Decarbonisation-Pathways-forthe-European-Aluminium-Industry_FULL-REPORT.pdf For iron and steel, we are close to finalising the implementation of the new sector structure shown in Figure 19. The sector design is based on recently published work (see Pye et al, 2022) and models primary and secondary production, to enable modelling of the circular economy and the planned link with ENGAGE. D3.2 – Model development strategy – Update 1 Page 43 Figure 19. Improved representation of the steel sector in OMNIA The design of the sector was discussed with industry IPCC AR6 industry author Chris Bataille. Whilst we are not proposing to explicitly represent all material flows in OMNIA, the costs associated with raw iron ore inputs (e.g. through sinter or pellets) into steel production pathways will be included. 3.2.4.6 New approach for demand projections Recent progress: New general approach for energy services demand projections identified. Data collection and first draft implementation for residential and transport sectors to be completed by December 2024. The guiding principles in defining the new approach are, first, to introduce the dependency from multiple drivers, second, to have the possibility of modelling demand trajectories that are decoupled from historical trends (and hence from what can be obtained through regression analyses), and third, to keep the projections at a country level, provided that enough data is available. The expression for the annual variation – year t over year t-1 – for a generic energy service demand (ESD) can be expressed as follows: 𝐸𝐸𝐸𝐸𝐷𝐷𝑡𝑡 𝐸𝐸𝐸𝐸𝐷𝐷𝑡𝑡−1 = 𝑘𝑘𝑡𝑡× 𝑋𝑋1,𝑡𝑡 𝑋𝑋1,𝑡𝑡−1 × 𝑋𝑋2,𝑡𝑡 𝑋𝑋2,𝑡𝑡−1 × 𝑋𝑋3,𝑡𝑡 𝑋𝑋3,𝑡𝑡−1 × ⋯ × 𝐼𝐼𝑡𝑡 𝐼𝐼𝑡𝑡−1 Where kt is a decoupling factor to maintain the freedom to introduce trends not directly related to the drivers considered, X1, …, XN are the drivers of interest for the specific ESD and I is the intensity term, given by: 𝐼𝐼𝑡𝑡 = 𝐸𝐸𝐸𝐸𝐷𝐷𝑡𝑡 𝑋𝑋1,𝑡𝑡×𝑋𝑋2,𝑡𝑡×𝑋𝑋3,𝑡𝑡× ⋯ Drivers are grouped into primary drivers and intermediate levers. A description of the three elements composing the Equation is given in the following, while a visual representation summarising the approach is shown in Figure 20. - Primary drivers represent the impacts of socio-economic development on end-use demand and include population, GDP, GDP per capita, etc. Primary drivers are defined per country – or region – and represent the basis for building demand projections for different scenarios, that can be based on SSP framing as D3.2 – Model development strategy – Update 1 Page 44 well as other scenario frameworks. - Intermediate levers are used to provide additional flexibility and adjust the demand to reflect the impacts of sector-specific, service-oriented drivers (e.g., heating or cooling degree days, floor space per capita) while maintaining consistency with primary drivers. Future projections of intermediate levers vary depending on factors such as geographic location, economic development, population density. For example, air conditioner ownership has already reached saturation in some developed countries, while it is still increasing in most developing countries and will reach varying degrees of saturation at different rates in the future. Most of the intermediate levers are cross-cutting indicators which are influenced by factors such as income level, climate change, building conditions, transport mode. Therefore, data of intermediate levers can be estimated based on historical development trends, calculated using socioeconomic indicators with econometric approaches, or acquired from other sectoral-specific studies. - Intensity term denotes the final energy required to deliver a unit of service, and the definition depends on the drivers considered for each specific ESD. In general, the service-specific intensity term is assumed to remain constant. However, in practical applications, scenario modellers can adjust this parameter according to their scenario narratives, considering factors such as digitalisation and technological advancements using the decoupling factor kt. Figure 20. Visual representation of the approach to energy service demands projections. 3.2.4.7 Water module Progress: First design proposal of the water module shared by M6. Final design expected to be finalised by M26. Expected to be part of the first version of OMNIA (MS8) A first meeting among OMNIA developers on the possible design of the water module has been organised. A first design proposal sees the representation of the following water contributions: D3.2 – Model development strategy – Update 1 Page 45 - Water supply 1. Groundwater and surface water availability, through a link with scenarios from CICERO: The surface water availability in different basins is linked with different climate scenarios. 2. Cross water interactions among OMNIA regions - Water demand 1. Agricultural water demand and water demand for livestock will be taken from the Magpie model. 2. Water use in cooling of thermal power plants and resource extraction (endogenous representation): The water consumption for cooling in the power sector will be calculated using water use coefficients that vary across different technologies and regions. These coefficients are taken from Jin et al. (2019). The water consumption factors for fuel production are taken from Sprang et al (2014). 3. Desalination at regional level: Desalination technologies are incorporated in the model to convert seawater into fresh water for coastal areas. The mapping of coastal basins to regions is harmonised with the regional mapping of the GCAM model. 3.2.4.8 Interlinkages with WP4 models The linkages between OMNIA and WP4 models have been discussed in more detail. The information required by OMNIA and the characteristics of each model have been considered to determine modifications and adjustments required to implement the linkages. Soft-link OMNIA-EnGAGE (T.4.1) The link between OMNIA and ENGAGE has been discussed in more detail and the outputs from each model that will be used as inputs in the other have been identified. The common assumptions that need to be set to ensure the coherency between both models have also been defined, while the particular scenarios and policies that will be tested with this integrated modelling framework are still under discussion. The next figure presents an example of scenarios and policies that can be considered using such an integrated framework. Figure 21. Soft-link OMNIA-ENGAGE D3.2 – Model development strategy – Update 1 Page 46 Key insights from ENGAGE will include changing demand for metals and construction materials, primarily cement, given how decarbonisation and circular economy policies and pathways impact demand. and how changes to steel, cement, aluminium stocks impact secondary production potential. Hence, the need for OMNIA to represent these sectors in detail. These links will allow to assess economic implications of different decarbonisation pathways/ net-zero targets in an integrated manner with circular economy policies, helping to highlight synergies, trade-offs, and rebound effects, as well as energy, emissions, and investment impacts of decarbonisation scenarios in the steel, aluminium and cement industries. This will lead to new IAM capabilities, as key economic activities related to circularity will be incorporated, eventually enabling to quantify how relevant interventions can contribute to energy, resource and emissions cuts, while increasing the process resolution for key circular economy sectors. The latest modifications in ENGAGE have focused on detailing the representation of the relevant sectors with the objective to allow for a full representation of circular economy measures and provide OMNIA with detailed information. For steel modelling, the three different pathways to produce steel and various intermediate inputs have been explicitly added. In addition, high and standard steel scrap qualities have been included and final steel products that are currently limited to the use of primary steel and high-quality scrap-based secondary production have been identified. This allows us a representation of the technical constrains to replacing primary steel by secondary steel from end-of-life scrap. For aluminium modelling, bauxite and aluminium production have been disaggregated into their distinct sub processes, intermediate and final products. Soft-link OMNIA – MAgPIE (T.4.2) The MAgPIE and OMNIA soft-link will cover different systems: - AFOLU emissions: MAgPIE can inherit GHG prices from OMNIA and generate corresponding land use and land use change GHG emissions, or provide a complementary analysis on land use emissions. CO2, CH4 and NO2 emissions from AFOLU sector for each scenario will be extracted from MAgpie model and in the OMINA - Bioenergy: Based on different socio-economic specifications, MAgPIE can generate biomass feedstock availability and related costs and GHG emissions. These are incorporated into OMNIA, which will estimate carbon prices and new demand levels for biomass feedstock. This is an iterative process which is run until there are minimal differences between MAgPIE and OMNIA. Note that this is only possible when OMNIA is complete, early 2025. Meanwhile, unilateral connection MAgPIE to OMNIA is being established, with MAgPIE providing cost supply curves for biomass feedstocks and AFOLU emissions. - Water: the two models can provide complementary analysis (water consumption from electricity and agriculture). As OMNIA will include the water sector assessing the water demand for energy production, MAgPIE will provide the water consumption for fertilizers and irrigation. The modelling approach for the OMNIA – MAgPIE soft-link is shown in Figure 22. D3.2 – Model development strategy – Update 1 Page 47 Figure 22. Modelling approach of OMNIA - MAgPIE soft-link Currently, MAgPIE has been reconfigured to run on OMNIA regions and on the same socio-economic settings, i.e. population and GDP trajectories. Also, the bioenergy sector design in OMNIA has been updated to match outputs from MAgPIE on five biomass feedstocks, i.e. forestall residues, agricultural residues, crops cultivated purposely for energy, first generation bioethanol and first-generation biodiesel. As of November 2024, first biomass feedstock cost-supply curves and related biomass and AFOLU GHG emissions are being prepared for OMNIA. Soft-link OMNIA – climate/physical impacts models (T.4.4) The soft-link between OMNIA and the climate physical impacts models of CICERO is going to be defined during the first semester of year 3. A first meeting was organised with CICERO team in November 2023 to start discussing the design of the link. A first proposal includes the possibility of OMNIA delivering emissions as output and CICERO models (openSCM) providing climate variables, e.g., global temperature. The subset of GHG emissions (CO2, CH4 and SO2, No2) from OMNIA model will be inputted to the infilling tool named Silicone to estimate the remaining gases. Once all the emissions trajectories for the given scenario are estimated, then these emissions will be inputted into a simple climate model to estimate the physical climate variables (like temperature and precipitation). These variables are downloaded using the pattern scaling model named METEOR model. In scenarios where we consider climate impacts, these regional climate variables will be used to assess changes in heating and cooling demand, crop yields, and other climate change related impacts. D3.2 – Model development strategy – Update 1 Page 48 Soft-link OMNIA – UNIBAS module on behavioural changes (T.4.6) After three meetings with the UNIBAS team, that is going to develop a specific module on behavioural changes to be linked to several IAMs, it was decided to primarily focus on 2 societal behaviours: - Adoption of insulation in households - EV adoption The UNIBAS module will provide as outputs the societal groups more prone to these two behaviours. These outputs will be converted into the share of population that is going to adopts house insulation and EVs and will be integrated into OMNIA demand drivers and constraints. UNIBAS team during the second year has developed an approach similar to be applied both OMNIA and CLEWs models. Soft-link OMNIA – openTEPES (T.4.7) The soft-link between OMNIA and openTEPES is going to be defined and developed during the first semester of year 3. Two meeting with the openTEPES team members has been already organised and the first proposal of soft-link sees openTEPES providing for the European regions: - Infrastructure costs per country - Energy corridors between OMNIA regions Concerning the approach to be applied to all the global regions, this will be defined during the first semester of year 3. D3.2 – Model development strategy – Update 1 Page 55 database, European Environment Agency (EEA) and relevant academic literature. • Calibration for land cover and crop production has been completed, • For the water system, all the sector specific withdrawal and consumption data is not reported by every European country. Therefore, assumptions are being made to fill in the gaps wherever needed. This activity might only be finalised when all the missing data at the national level is collected for the disaggregated model version. Time horizon and temporal resolution Even though the climate neutrality of the European Union is aimed to be achieved already by 2050, the modelling horizon of the project will be extended to 2070 with an annual resolution. The extension of the modelling horizon implies great uncertainty as to the technological advances and relevant cost assumptions, climatic conditions and energy demand projections to be expected so far into the future, but since impacts of a changing climate are projected to be more adverse in the latter half of the century, it is deemed crucial to be able to capture part of this timeframe. The base year will be 2018, but statistics up to 2021 will be accounted for and these years will be calibrated accordingly wherever possible. It was agreed during the project conception that the temporal resolution of the model would be enhanced from the 3 intra-annual parts adopted in the GLUCOSE (i.e., Global CLEWs) model. These will increase to 16 annual timesteps, which will consist of 4 seasons (i.e., winter, spring, summer, autumn) represented by a typical day; each consisting of 4 intra-day parts. Initially, the model will be developed with this temporal resolution, but in case the computational effort during the model optimisation permits further enhancement, this will be explored in future stages of the model development. Such an enhancement would strengthen the model’s capability to provide valuable insights regarding variable renewable energy integration. In this sense, the data collection process for time-variable statistics (e.g., electricity demand variability, renewable energy generation profiles, heating and cooling demand profiles etc.) will have to consider this plausibility. Naming Convention As the CLEWs-EU model is essentially being built from scratch, the way in which technologies and commodities would be represented within the model had to be agreed upon. Specifically, the short codes abbreviating these had to be formulated. This would facilitate the standardisation and automation of data entry and results extraction and visualisation. A 12-character naming convention was agreed for technologies, while a 6-character naming convention was adopted for commodities. Table 18 shows an extract of the naming convention adopted for the energy module, while Table 19 and Table 20 highlight the naming convention implemented for the land and water modules. The full explanation of the naming convention will be made available as part of the model documentation in the model’s GitHub repository, which is still under development. Table 18. Naming convention sample for technologies and commodities in the energy module (CLEWs-EU). Technology Country/Region (2 Characters) Module (1 Character) Sector (2 Characters) Fuel (3 Characters) Technology/ Destination (2 Characters) Type (2 Characters) EUEPTVPGSLPH Gasoline-fired plug-in hybrid passenger car in the EU EU (European Union) E (Energy) PT (Passenger Transport) GSL (Gasoline) VP (Passenger vehicle) PH (Plug-in hybrid) ATEGNELCDEIC AT (Austria) E (Energy) GN ELC (Electricity) DE (Germany) IC D3.2 – Model development strategy – Update 1 Page 56 Technology Country/Region (2 Characters) Module (1 Character) Sector (2 Characters) Fuel (3 Characters) Technology/ Destination (2 Characters) Type (2 Characters) Electricity interconnector between Austria and Germany (Grid Networks) (Interconnector) PLEEGCOAPPCS Coal power plant with CCS in Poland PL (Poland) E (Energy) EG (Electricity Generation) COA (Electricity) PP (Power Plant) CS (Carbon Capture and Storage) Commodity Country/Region (2 Characters) Module (1 Character) Fuel (3 Characters) ITEHY2 Hydrogen fuel in Italy IT Italy E (Energy) HY2 (Hydrogen) PTESEL Secondary electricity generated by power plants in Portugal PT Portugal E (Energy) SEL (Secondary electricity) Table 19. Naming convention sample for technologies in the land module (CLEWs-EU). Technology Country/ Region (2 Characters) Module (1 Character) Identifier 1 (3 Characters) Identifier 2 (3 Characters) Input Level (1 Character) Water supply (1 Character) Dummy (1 Character) EULIMPWHE000 Wheat imports in the EU EU (Europe) L (Land) IMP (Imports) WHE (Wheat) 0 (Dummy) 0 (Dummy) 0 (Dummy) EUL000WAT000 Land cover under water bodies EU (Europe) L (Land) 000 (Dummy) WAT (Water) 0 (Dummy) 0 (Dummy) 0 (Dummy) EUL000MAIHR0 Potential for rainfed maize cultivation under high input level EU (Europe) L (Land) 000 (Dummy) MAI (Maize) H (High Input Level) R (Rainfed type) 0 (Dummy) Table 20. Naming convention sample for technologies in the water module (CLEWs-EU). Technology Country/ Region (2 Characters) Module (1 Character) Identifier 1 (3 Characters) Identifier 2 (3 Characters) Identifier 3 (3 Characters) EUWDEMPUBGWT Ground water supply technology for public use EU (Europe) W (Water) DEM (Demand) PUB (Public supply) GWT (Ground water) EUWDEMAGRSUR Surface water supply technology for agricultural use EU (Europe) W (Water) DEM (Demand) AGR (Agricultural Use) SUR (Surface water) EUWMIN000PRC EU (Europe) W (Water) MIN 000 (Dummy) PRC (Precipitation) D3.2 – Model development strategy – Update 1 Page 57 Technology to produce water from Precipitation (Resource Technology) 3.3.4.2 Energy module The CLEWs-EU energy module is broken down into several subsectors. The sectoral and technological detail was discussed amongst the team and agreed upon during the kick-off and regular progress meetings. Enhancements in technological representation will be conducted to increase the number of options that can render net-zero emission pathways technically feasible. As an essential feature, hydrogen energy chains, currently missing from the Global CLEWs model, will be added to the energy sector module. More information on the module structure and the associated data requirements is provided in the next subsections. Module structure and technological representation Five overarching sectors are developed to represent the broader energy system. Each of these will include a large set of technologies to capture the current status of energy mix, as well as future technology options for decarbonisation. Specifically, the following sectors will be modelled: a) Primary energy supply – this relates to fossil fuel supply, nuclear fuel supply, renewable energy potential and hydrogen imports. Primary energy supply and transformation infrastructure (e.g., gas pipelines, LNG regasification terminals, oil refineries, etc.) will not be modelled explicitly, with the only exception of electricity interconnectors between countries in the disaggregated version of the CLEWs-EU model. Supply of biomass into the energy system will occur via an interlinkage with the land module of the model. b) Electricity and heat generation – fossil fuel, nuclear and renewable energy technologies will be represented, along with a simplified representation of grid networks to capture associated losses. Additional technology options to be represented include electricity storage, use of Carbon Capture and Storage (CCS) technologies, electrolysers for the production of green hydrogen; hydrogen production will also be possible through steam methane reforming (i.e., using natural gas as feedstock) with or without CCS. c) Transport – the transport sector will include road transport, which will be further broken down to passenger (passenger cars and buses) and freight (light commercial vans and heavy trucks), rail transport (passenger and freight), aviation and shipping. d) Buildings – this sector basically comprises of the households and service sectors. It will be broken down into end-use services (i.e., space heating, space cooling, cooking, sanitary hot water, lighting and appliances). e) Industry – the five most energy intensive industries – a considerable variation exists according to the latest energy balances (Eurostat, 2023) – will be represented separately for each country, while the rest of the industries will be lumped together in a sixth category7. The output of each of the industries will be represented in generic terms (i.e., PJ of energy services), similar to the approach previously employed by 7 The breakdown of the industrial sector will be based on the categories available in Eurostat’s energy balance: Iron & steel, Chemical & petrochemical, Non-ferrous metals, Non-metallic minerals, Transport equipment, Machinery, Mining & quarrying, Food, beverages & tobacco, Paper, pulp & printing, Wood & wood products, Construction, Textile & leather. D3.2 – Model development strategy – Update 1 Page 58 GCAM for some of the industries (JGCRI, n.d.). Figure 25. Simplified representation of CLEWs-EU energy module. It should be noted that the energy demands in all sectors are inputs to the model8. A simplified representation of the structure of the energy module is provided in Figure 25; further disaggregation of technologies will be available in the model than is depicted. For instance, the various vehicle technologies will be further broken down into 8 Estimation of energy demand projections need to account for official targets on efficiency and electrification; these need to be derived exogenously through the use of appropriate top-down and/or bottom-up models. D3.2 – Model development strategy – Update 1 Page 59 passenger cars, buses, light commercial vans and heavy trucks. Similarly, the structure of the industrial sector will be constructed for each of the five key industries (and other industries) to be represented in each country or the EU as a whole. Additionally, interlinkages with the land and water modules are presented in section 3.3.4.3. Data collection for energy module Since the CLEWs-EU model is developed from scratch, a large volume of data has to be collected and transformed for input. Table 21 provides a brief overview of some of the main data requirements relating to the energy module. Even though the aggregated regional EU version of CLEWs-EU will be developed first, information on all of the data items is collected at a national level to expedite the development of the disaggregated national version of the model. The only exception relates to technoeconomic assumptions (e.g., cost and efficiencies of technologies), for which generic data will be used. Table 21. Key data requirements of the CLEWs-EU energy module. Sector Data item Status Primary Source General National energy balances Complete (Eurostat, 2023) Technoeconomic assumptions across sectors Complete EU Reference Scenario 2020 (European Commission. Directorate General for Energy. et al., 2021) Electricity and heat Installed capacity by technology Complete Eurostat & ENTSO-E Electricity generation by technology Ongoing Eurostat, ENTSO-E Electricity demand load profile Complete ENTSO-E Renewable energy generation profile Complete ENTSO-E Transmission and distribution losses Complete Eurostat Grid interconnector capacity Complete ENTSO-E Transport Vehicle fleet statistics Complete Eurostat, JRC-IDEES 2021 Annual mileage per vehicle category Ongoing National statistics, JRC-IDEES 2021 Mobility demand projections Pending EU Reference Scenario 2020 (European Commission. Directorate General for Energy. et al., 2021), Soft-link with UCL demand projections Buildings Heating and Cooling Degree Days statistics Complete Eurostat Heating and Cooling Degree Days projections Pending (Climate-ADAPT, n.d., n.d.)– can also be linked with WP4 Statistics on demand by useful energy service in households and activity be each technology Complete Eurostat, JRC-IDEES 2021 Statistics on demand by useful energy service in commercial sector and activity be each technology Complete Eurostat, JRC-IDEES 2021 Industry Final energy demand by fuel by industry Ongoing Eurostat, JRC-IDEES 2021 Industrial activity projections Pending To be determined; Potential for own projections D3.2 – Model development strategy – Update 1 Page 60 3.3.4.3 Land and Water modules The water and land use modules are effectively interlinked in nature. Though we have different naming conventions and technologies to represent the two modules inside CLEWs-EU, their inputs and outputs are interdependent. The water module will represent the different primary sources, namely water from precipitation and snow melt, water from existing surface and ground water sources, water from neighbouring geographies by transboundary river exchanges, as well as secondary sources like water from treatment plants in some countries. In the land module, the total land available in the continent (in the engagement model) and in each country (in the disaggregated model) is divided into nine major land categories based on the European Space Agency’s land cover classifications. They are namely: grassland, cropland, snow cover, water bodies, shrubland, forest cover, built up land, barren land, other land (includes mangroves, lichens and wetlands). Representing land and water systems in a generic systems optimisation setup involves using relevant assumptions to simplify their characterisation and at the same time ensuring not a lot of the essential detail is lost. Effectively the land and water modules represent the balance of the two resources (i.e., land and water) in the respective systems. Despite accounting for the balance of these resources, some components like groundwater storage are not considered in these modules as it involves detailed sub surface flow modelling, which is outside the scope of the project. In the following section, a sample of the technology representation in the land and water modules is presented, along with some common data sources. Module structure and technological representation Figure 26 represents a diagrammatic illustration of the land component within a general CLEWs systems framework. The diagram breaks down "Total land" into various land uses and their subsequent products or services. Land Uses: This includes various classifications of land such as Cropland, Barren land, Forests, Pastures/Grasslands, Built-up land, and Water bodies. Each category of land use is dedicated to a specific type of production or ecosystem service. Production: From each type of land use, there is also a corresponding output: • Cropland is associated with the production of agricultural commodities like Maize, Rice, Coffee, Sugarcane, Bananas, and other crops. • Forests contribute to the production of wood and other forest products. • Pastures and Grasslands are tied to the production of meat and dairy, implying livestock farming. • Built-up land is related to residential, industrial, commercial, and transport services, indicating areas of human development and infrastructure. The arrows indicate the flow from land use to the type of production. The diagram encapsulates the interconnectedness of land use with various economic and environmental outputs, emphasising the integrated approach of the CLEWs framework where changes or policies affecting land use have cascading effects on food, energy, water, and ecosystem services. D3.2 – Model development strategy – Update 1 Page 61 Figure 26. Primary level of land categories breakdown in the land module (CLEWs-EU). A more complex system diagram is illustrated in Figure 27 and portrays how the CLEWs-EU land and water modules will be represented, along with key interactions with the energy module. The diagram is organised into several columns, each representing different components of the CLEWs systems: • Land Use (left-most column): This column lists various types of land uses such as agricultural land (which will be divided into up to 10 different crop types), barren land, forests, built-up land, and water bodies. Each land type is linked to specific outputs, i.e., specific crops and the different aspects of the water system. • Precipitation and Sea Water (bottom rows and centre columns): This section details the sources (precipitation and sea water) and uses of water, distinguishing between surface water and groundwater, and their respective uses in agriculture, power generation (PWR), and other needs. • Energy Supply and Use (top right box): The diagram includes simplified forms of electricity generation, to show how these energy sources contribute to meeting the demands of different sectors, including transport, meat processing, and the transport and distribution (T&D) of water resources. • Interlinkages (arrows and lines): The arrows and lines represent the flow of resources and interconnections between the land, water, and energy components. For example, water from different sources is used for land irrigation, electricity generation requires water for cooling, and agriculture produces biofuels. D3.2 – Model development strategy – Update 1 Page 62 Figure 27. Simplified representation of CLEWs-EU land and water modules and illustration of key interactions with the energy module. Data collection for land and water modules The construction of the Land and Water modules within the CLEWs-EU framework necessitates an extensive acquisition and processing of data. Table 22 delineates a summary of the primary data requisites pertinent to the land and water components of the model. In alignment with the approach adopted for the energy module, we are amassing data at the individual country level, despite the initial development focus on a regional EU aggregate. This strategy is to facilitate a seamless transition to the more granular national models in subsequent phases. For land, this includes data on land use patterns, agricultural productivity, and potential changes due to climate or policy shifts. For water, we gather information on water availability, usage statistics across sectors, withdrawal limits, and quality standards. Exceptions are made for general hydrological parameters and land management D3.2 – Model development strategy – Update 1 Page 63 practices, where universally applicable figures are employed in the preliminary stages. Nevertheless, countryspecific calibration will be essential to enhance accuracy and relevance, particularly in reflecting the unique hydrological and agricultural conditions of each member state. Table 22. Key data requirements of the CLEWs-EU land and water modules. Sector Data item Status Primary Source Land Cover Area under different land cover classes Completed FAO, Eurostat, European Space Agency (WorldCover2021) Area under crop cultivation Completed Aquastat & Eurostat Agriculture Crop yields and production statistics Pending Global Agro-Ecological Ecological Zoning-GAEZ (FAO/IIASA) & Eurostat Crop yield variation under future climates Pending Global Agro-Ecological Ecological Zoning-GAEZ (FAO/IIASA) Fertilizer usage in Crop production Pending Eurostat Crop import statistics Partially completed Eurostat Costs involved in crop production Pending National Statistics, Eurostat & Derived entity in some cases Fuel usage in the agriculture and forestry sector Completed Energy Balances Water availability and Use Precipitation statistics Completed Eurostat and GAEZ Groundwater potential Pending National Statistics and Academic Journals Water use factors in different sectors Pending Eurostat Evapotranspiration under different land categories Pending GAEZ Water losses during transportation Pending Eurostat, National Statistics Demand projections for water in different sectors Pending Derived entity based on economic projections 3.3.5 Planned development steps 3.3.5.1 Module development efforts Current efforts of the CLEWs-EU team are focused on developing the disaggregated model version. The various sectors will be developed in parallel. Specifically, the following activities are planned: • Energy Module – Even though information for the aggregated version of the model was gathered from a variety of sources, the disaggregated version will utilise more detailed information. As such, the entire energy module will be updated with the most recent release of the JRC-IDEES database (Rózsai et al., 2024). o Electricity and heat generation – Relevant data is being gathered and transformed for input to OSeMOSYS, using complementary sources, such as Eurostat and ENTSO-E. An expanded list of technology options is being considered (e.g. to allow differentiation between dam and run-of-river hydropower), which can provide more detailed insights. The model shell will be developed before the end of the year, so that it can be directly populated with information. D3.2 – Model development strategy – Update 1 Page 64 o Buildings – Historical data has already been extracted from the JRC-IDEES 2021 database. Demand projections will be adopted once the relevant methodology being developed by UCL is finalised. Development of the model shell for each EU member state has started and should be finalised by the end of January 2025 (M26). o Transport – Relevant scripts are under development for the extraction of historical data at a national level from the JRC-IDEES 2021 database. The model shell for each EU member state is under development and should be populated with respective information by the end of January 2025 (M26). o Industry – Extraction of historical data and model development for this sector at a national level will commence by February 2025 (M27). • Land and Water Module o The representation of the desalination system is being finalised. This is applicable only to some European countries where a share of the water demand is met through desalination technologies. It should be completed by Dec 2024. o Sector specific water withdrawal and consumption numbers are being extracted from relevant literature (Fridman et al. 2024). The demands will be finalised by January 2025 (M26). o A simplified emissions accounting methodology is being implemented for the different Land Use categories. Will be tested by December 2024 and implemented across the EU27 countries in the model by February 2025 (M27). o A step-by-step method is being developed to apply the land and water module structure to all the EU27 countries in a streamlined manner. Structure will be developed by January 2025 (M26) and applied by March 2025 (M28). 3.3.5.2 Linkages with WP4 In an effort to go beyond the structural limitations of IAMs, the CLEWs-EU team is engaging with the modelling teams of WP4 – Integrate & Expand. Specifically, the envisioned integration will be conducted in collaboration with CICERO (Task 4.4), UM (Task 4.5) and UNIBAS (Task 4.6). Soft-links with the DIAMOND climate modelling and pattern scaling suite will be established within Task 4.4 to examine the effect of various climate scenarios on critical aspects of the food-energy-water nexus, such as water availability, land productivity and renewable energy output. During the physical meeting in Athens, it was agreed between the team members that a guiding document will be needed that identifies each parameter that needs to be adjusted across the various modules of the CLEWs-EU model, whenever a specific climate scenario will be represented9. This will facilitate the identification of soft-linking possibilities, while it will help avoid potential inconsistencies across the assumptions to be adopted for each scenario. The aforementioned document should be available when key sectors of the disaggregated version of the model will be ready (i.e., by M26 of the project). The CLEWs-EU model itself will be able to provide projections on greenhouse gas emissions with an annual resolution. 9 This includes but is not limited to parameters relevant with: annual precipitation, irrigation requirements, crop yield, water demand for domestic uses, hydropower output, thermal plants cooling requirements and efficiency, space heating and cooling demands in buildings. D3.2 – Model development strategy – Update 1 Page 71 we are also adding hydrogen as part of energy sectors represented in the model. Besides energy, sectors included in the ‘Manufacture’ group are also reclassified based on IPCC classification of manufacturing industries. The IPCC classification follows the International Standard industrial Classification of All Economic Activities (ISIC) Vol. 4. Two sectors under the ETS, the EII and petroleum sectors, are part of this manufacturing group. Non-energy-intensive industries are aggregated as Other Manufacturing Industries in this model. This energy-intensity-based division in manufacturing sectors follows the Industrial Classification from International Energy Outlook. Defining new regional and sectoral aggregation has been accomplished by March 2023. Table 25. GEMINI-E3 EU Sectors description NO Sectors ABB. GTAP Sectors S07 Agriculture AGRI pdr, wht, gro, v_f, osd, c_b, pfb, ocr, ctl, oap, rmk, wol, frs, fsh S01 Coal COAL coa S02 Oil OIL oil S03 Natural Gas NGAS gas, gdt S06 Hydrogen HYDR - S05 Electricity (Power Generation) ELY ely S04 Petroleum PETR p_c S08 Energy Intensive Industries EII oxt, ppp, chm, bph, rpp, nmm, i_s, nfm, fmp S09 Other Manufacturing Industries OMNI cmt, omt, vol, mil, pcr, sgr, ofd, b_t, tex, wap, lea, lum, fmp, ele, eeq, ome, mvh, otn, omf S12 Air Transport AIRT atp S10 Land Transport LANT otp S11 Water Transport WATT wtp S13 Services SERV wtr, cns, trd, afs, whs, cmn, ofi, ins, rsa, obs, ros, osg, edu, hht, dwe 3.4.4.2 Economic Database Update The core element of the Global Trade Analysis Project is the GTAP Data Base, a comprehensively documented global database that contains extensive data on bilateral trade, as well as information on transport and trade protection measures (such as taxes and subsidies). The GTAP Database serves as a representation of the global economy and is employed by a vast number of institutions worldwide as a crucial resource for conducting applied general equilibrium (AGE) analysis on matters related to the global economy. The GEMINI-E3 database is mainly based on this new GTAP database, including for GHG emissions. The new reference year of the model is now 2017. The model database has been updated by GTAP V.11A (Aguilar et al., 2022) in August 2023. 3.4.4.3 GEMINI-E3 coding and writing output interfaces The new databases as well as some other economic sources coming from OECD, IMF and IEA databases have been harmonised and processed by different GAMS programs and Excel sheets to create a harmonised database that can be used by GEMINI-E3. The Beta version of the model that is based on the specifications used into the PARISREINFORCE project (see the description of this version in I2AM PARIS platform10) has been created into the GAMS 10 https://www.i2am-paris.eu/detailed_model_doc/gemini_e3 D3.2 – Model development strategy – Update 1 Page 72 software environment and a reference scenario has been simulated. This step has several objectives: 1. To test the consistency of the database; 2. To check the solving property of the model; 3. To serve as a base for writing a new output interface that creates Excel templates; 4. To serve as a base for writing the new specifications of the model that are planned within the DIAMOND project. This step was also advertised through a video in a LinkedIn post. The GEMINI-E3 coding and writing output interfaces has been accomplished by end of October 2023. 3.4.4.4 Greenhouse Gases (GHG) The latest version of the GTAP database provides estimated data of CO2 by fuel and by user for each country/ region, for both combustion and non-energy combustion sources. The database also includes non-CO2 greenhouse gas emissions - CH4, N2O, and F-gases. Following no emissions update from CEDS and PRIMAP that previously used as database for GEMINI-E3 (2014 baseline year), the new GEMINI-E3 EU adopt emissions GTAP database of year 2017. To keep the consistency with previous classification of emission sources of CEDS (for CO2 and CH4), PRIMAP (for N2O) and EPA (for F gases) which were designed in line with the IPCC classification of AR6, the updated GHG emissions of GTAP are mapped to this classification. For this purpose, we refer to GTAP Memorandum No.32, Development of the Non-CO2 GHG Emissions Database for GTAP V.11A, Annex II: Definition, Units and Conventions of IPCC AR6 (IPCC, 2022). This mapping links emissions agents (GTAP sectors) and emission drivers (factors affecting the emissions level ex: output production, endowment used, or input used) to the previous related IPCC classification with previously estimated Marginal Abatement Cost. Defining the GHG definition and its implementation into GEMINI-E3 has been accomplished by end of October 2023. 3.4.4.5 Biofuel Following the database update in the new version of GEMINI-E3, biofuel used in electricity generation is updated to year 2017, from the latest version of GTAP Power 11. As in previous version, GTAP Power 11 biofuel is still aggregated in Other-Sector. We disaggregate biofuels with the same methodology as in the previous version of GEMINI-E3. First, we get the percentage of biomass power plant from total power generation for each country. This share is obtained from BP Statistical Review 2022, and then aggregate it into the 24 regions of GEMINI-E3. Several adjustments are made to ensure that the principle of market clearing is applied, i.e., demand of biofuel power plants equal to its supply. In addition to updating biomass, the new stage of model development is to consider the other uses of biomass/biofuel in the model. For this, we refer to IEA statistical data of 2017. The statistics represents 16 energy type for over 170 countries and regions and report aggregated number of biofuel and waste. To get the value biofuel energy balances, we deduct the waste components from the same source. The data is then aggregated. The aim is to obtain the data of total supply and total demand (intermediate goods and final consumption) of biofuel. Data collection on biofuel has been finalised in October 2023. 3.4.4.6 New CES Nesting Structure The introduction of new energy inputs (biofuel and hydrogen) as well as a better representation of energy D3.2 – Model development strategy – Update 1 Page 73 substitution within deep decarbonization scenarios require to reconsider the CES nesting structures for the production and consumption that we included in the former versions of GEMINI-E3. The design of this new nesting CES structure consider the work done within the DECARB project funded by the Swiss Federal Office of Energy where these two energy inputs are represented. This nesting CES structure details the energy that is used for transportation purpose, to the other uses (heat in industrial processes and space heating and cooling in buildings). Hydrogen can be used in industrial process and in transportation. Finally, the model assumes that biomass/biofuel can be used in industrial processes as well as in transport by blending biofuels with petroleum products. This new nesting structure is represented Error! Reference source not found.. Figure 30. New nested CES production structure – Industrial sectors (GEMINI-E3) This implementation of the new nesting CES structure has been finalised by December 2023. 3.4.5 Planned development steps 3.4.5.1 Air pollutants With such a high magnitude of air pollution-related externalities, implementing more stringent environmental policies (e.g., emission taxation or energy subsidies elimination) may result in significant co-benefits. While CO2 and non-CO2 GHG emissions are usually well represented in most global economic databases, air pollution flows are not included in many cases. The new version of GEMINI integrates the pollution database for the GTAP Data Base Version 11A. Emissions for nine substances are reported in the database: black carbon (BC), carbon monoxide (CO), ammonia (NH3), nonmethane volatile organic compounds (NMVOC), nitrogen oxides (NOx), organic carbon (OC), particulate matter 10 (PM10), particulate matter 2.5 (PM2.5) and sulphur dioxide (SO2). The dataset covers five reference years – 2004, 2007, 2011, 2014 and 2017. EDGAR Version 5.0 database is used as the main data source. To assist with emissions redistribution across consumption-based sources, the IIASA GAINS-based model and IPCC-derived emission factors are applied. Each emission flow is associated with one of the four sets of emission D3.2 – Model development strategy – Update 1 Page 74 drivers: output by industries, endowment by industries, input use by industries and household consumption. In addition, emissions from land use activities (biomass burning) are estimated by land cover types. These emissions are reported separately without association with emission drivers. Current Progress and Ongoing Task: By June 2024, air pollutant data had been aggregated in alignment with the model’s new structure, and a basic integration of air pollutants using a simplified modelling technique was successfully completed by August 2024. However, recognising that the drivers of air pollutants differ significantly across regions—due to factors such as industrialisation levels, economic activities, and policy frameworks—a comprehensive examination of emission control standards across various sectors and regions is essential to enhance the precision and reliability of the model's results. Consequently, a dedicated research effort is underway to analyse heterogeneity in emission control measures, considering the specific characteristics of different pollutants, industrial sectors, and geographic regions. Due to the wide scope of pollutant types and limited existing literature on modelling approaches, the completion of this task has been revised, with finalisation anticipated by Q1 of 2025. 3.4.5.2 Hydrogen Gronau et al. (2023) emphasise that few macroeconomic models incorporate hydrogen as an energy carrier to evaluate the macroeconomic impact of hydrogen-powered aviation, specifically through a Computable General Equilibrium (CGE) model. Similarly, Sylva et al. (2014) explore the potential of hydrogen-fueled vehicles by integrating a CGE model with an energy simulation model focused on road transportation. Gilmore et al. (2023) examine the energy mix and welfare implications of a deep decarbonization pathway that includes net-negative emission technologies. They employ Environment Canada's Multi-Sector, Multi-Regional CGE model in conjunction with the GCAM model, considering hydrogen technologies alongside negative emissions solutions such as Bioenergy with Carbon Capture and Storage (BECSS) and Direct Air Capture (DAC). In the GEMINI-E3 model, we introduce a new "hydrogen sector," which produces hydrogen through water electrolysis, with production costs informed by the International Energy Agency’s report, The Future of Hydrogen (IEA, 2019). These cost estimates are also utilised in both the GCAM and TIMES-based models (such as OMNIA). While we assume a uniform cost structure in USD across all regions, electricity prices are region-specific and are calculated within the model. Notably, electricity costs constitute the primary component of hydrogen production costs. Current Progress and Ongoing Task: 1. Data collection on hydrogen and design of the integration of this new sector into GEMINI-E3 has been accomplished in September 2023. 2. Following bilateral discussions with SEURECO, we recognise the need for collaborative efforts to determine the most effective approach for representing hydrogen within the GEMINI-E3-EU model and the NEMESIS model. This endeavour may involve collaboration with the OMNIA team to facilitate parameter sharing or establish linkages. 3. Additionally, several areas require further examination, including methodologies for integrating hydrogen data into regional statistical frameworks, defining mechanisms for representing international hydrogen trade dynamics, and accurately modelling the hydrogen production supply chain. Given the relatively nascent state of hydrogen technology and the unique challenges associated with this topic, this project remains ongoing, with completion anticipated by Q2 2025. D3.2 – Model development strategy – Update 1 Page 75 3.4.5.3 Interlinkages within WP4 Following the objectives of WP4 to expand IAM representation of real-key world process our team has participated in a two-day workshop held on 2122 September 2023 in Maastricht, Netherlands, with the primary objective of exploring the feasibility of establishing a soft linkage of GEMINI-E3 with the modules developed by the participating WP4 team. This initiative aims to integrate key components of our economic landscape, including the financial sector, labour market, and heterogeneity in final demand. The workshop provided a valuable platform for collaborative discussions and knowledge exchange, allowing us to engage with experts in the field and gain insights into potential synergies and challenges. Through these deliberations, we have taken significant strides toward enhancing the integration of GEMINI across diverse economic domains, paving the way for more robust and comprehensive analyses in the future. Our focus was on seamlessly incorporating heterogeneity in consumer behaviour and representing diverse skill sets within the labour force through this integration. Households Heterogeneity The deliverable emphasises the critical role of incorporating household heterogeneity and behavioural factors in climate-related models to enhance the assessment of policy and scenario impacts. It underscores that household consumption expenditure is essential to achieving a climate-neutral economy, generally accounting for approximately 60% of GDP. While Integrated Assessment Models (IAMs) have historically lacked household heterogeneity, recent models such as EUROGREEN, E3ME, PET, IMAGE-TIMER, and MESSAGE-ACCESS have been developed to address this limitation. The new version of GEMINI model will introduce heterogeneity of Households, assisted by CESAR (Task 4.3) providing microdata of European Union Statistics on Income and Living Conditions (EU-SILC) and Households Budget Survey (HBS). Heterogeneity captured in GEMINI-EU will be based on quintile income level and potential different structures of households. Current Progress and Ongoing Task: Data exchange has been maintained between CESAR and EPFL; the completion of the data update, including the full dataset, will rely on the ongoing database management efforts for GCAM, anticipated to conclude by April to May 2025. Consequently, the finalisation of this task is projected for Q2 2025. Modelling household heterogeneity in CGE models faces challenges, particularly with regard to significant negative savings observed in some countries, such as Greece and Poland. Addressing this issue requires careful calibration and consideration of household behaviours that impact savings rates. In response, the EPFL team has initiated discussions and brainstorming sessions with other modelling teams beyond DIAMOND, to conduct an extensive literature review, to refine approaches and enhance model accuracy in representing diverse household financial behaviours. Labour Dynamic The discussions and collaborative efforts during the workshop have resulted in a clear pathway for GEMINI to effectively capture the nuances of consumer diversity and skill differentiation within the labour market. UM and EPFL decided that work would begin on an empirically validated model that aids in quantifying labour transition costs within the scope of GEMINI-EU’s labour taxonomy. The model will focus on skill levels and occupational types (distinguished at a minimum by high/low skilled, green/brown and white/blue collar designations) Current Progress and Ongoing Task: Regular correspondence and online meetings are conducted to share updates on the management of the new labour dataset. Full integration of labour heterogeneity is anticipated to be finalised by Q2 2025. D3.2 – Model development strategy – Update 1 Page 76 Linking IAMS to climate/ physical impact models The previous version of the GEMINI-E3 model has demonstrated its capacity to simulate climate impacts (Gonseth & Vielle, 2019; Vöhringer et al., 2019, Joshi et al., 2016, Labriet et al., 2015), though its focus was primarily on specific regions with deterministic projections of emissions outcomes. The updated model seeks to address greater complexity by incorporating relative confidence levels in climate outcomes, achieved through potential linkages to physically manifested climate change models. This interlinkage aims to strengthen the model’s ability to represent the dynamic interconnections between emissions and climate impacts more robustly. Current Progress and Ongoing Task: Discussions regarding these enhancements are still in the early stages between EPFL and CICERO. EPFL is currently awaiting input from CICERO regarding the potential climate data and information that can be integrated into the GEMINI-EU framework. 3.4.6 GEMINI-E3 EU References Aguiar, A., Chepeliev, M., Corong, E., & van der Mensbrugghe, D. (2022). The global trade analysis project (GTAP) data base: Version 11. Journal of Global Economic Analysis, 7(2). Gilmore E., Ghosh, M., Johnston, P., Shahid Siddiqui M., Macaluso, N. (2023). Modeling the energy mix and economic costs of deep decarbonization scenarios in a CGE framework, Energy and Climate Change, Vol 4, December Gonseth C., Vielle M. (2019). A General Equilibrium Assessment of Climate Change Impacts on Swiss Winter Tourism with Adaptation, Environmental Modeling and Assessment, 24(3). Gronau S., Hoelzen J., Mueller T., Hanke-Rauschenbach R. (2023) Hydrogen-powered aviation in Germany: A macroeconomic perspective and methodological approach of fuel supply chain integration into an economy-wide dataset, International Journal of Hydrogen Energy, Vol 48., Issue 15, February IEA. (2019), The Future of Hydrogen, IEA, Paris https://www.iea.org/reports/the-future-of-hydrogen, License: CC BY 4.0 IPCC. (2022). Climate Change 2022: Impacts, Adaptation and Vulnerability. Available at: https://www.ipcc.ch/report/ar6/wg2/downloads/report/IPCC_AR6_WGII_FullReport.pdf Joshi, S.R., Vielle, M., Babonneau, F., Edwards, N.R., Holden, P.B. (2016). Physical and Economic Consequences of Sea-Level Rise: A Coupled GIS and CGE Analysis Under Uncertainties, Environmental and Resource Economics, 2016, 65(4), pp. 813–839Labriet, M., Joshi, S.R., Vielle, M., ... Loulou, R., Babonneau, F. (2015)., Worldwide impacts of climate change on energy for heating and cooling, Mitigation and Adaptation Strategies for Global Change, 2015, 20(7), pp. 1111–1136 Sylva, C. M., Ferreira, A. F., Bento J.P. (2014) Impact of hydrogen in the road transport sector for Portugal 20102050, Energy Procedia, 57, 207-214. Vöhringer, F, Vielle, M., Thalmann, P., … Stocker, D., Thurm, B. (2019). Cost and Benefits of Climate Changes in Switzerland, Climate Change Economics, 10(2). 19500052 D3.2 – Model development strategy – Update 1 Page 77 3.5 From NEMESIS to NEMESIS-World 3.5.1 Overview The NEMESIS model (New Econometric Model of Evaluation by Sectoral Interdependency and Supply) is a sectoral detailed macroeconomic model for the European Union (Boitier et al. 2018). It is a system of economic models for every European country (including the United Kingdom), devoted to study issues that link economic development, competitiveness, employment and public accounts to economic policies, and notably all structural policies involving long term effects. The essential purpose of the model is to provide a consolidated framework to realise “Business as Usual" (BAU) scenarios (or other alternative scenarios), up to 30 to 40 years, and to assess the socioeconomic impact of the implementation of all additional policies not already implemented in the BAU. The main mechanisms of the model are based on the behaviour of representative agents: firms, households, government and rest of the world. From the supply side, the model distinguishes 30 different economic activities that produce goods and services through production functions and, to do so, uses production factors: capital, energy, low and high-qualified labour and other intermediate consumption. All these economic activities are interrelated by inter-sectoral exchanges (conversion matrices) and external trades with other EU countries and the rest of the world. NEMESIS includes a detailed energy-environment module that allows the model to deal with climate mitigation policies, at an EU and Member State level. This module enhances the representation of the energy system of each Member State by detailing the energy used (ten different products) by each economic activity but also by households. It also details the power generation sector by allocating electricity production between different technologies through diffusion curves. CO2 emissions from fossil fuel combustion (FFC) are then calculated from fossil fuel consumption whereas other GHGs (CO2 from other sources, CH4, N2O, HFCs, PFCs and SF6) are either calculated by other modelled sectors or calibrated on external studies. Furthermore, the model can be relatively easily linked to other tools. In case of a linkage with more detailed energy-system models, NEMESIS is then used to assess the socioeconomic impacts at EU and Member State level, with the energy system modelling being delegated to the detailed energy models. The NEMESIS - World model is under development, and it will be a new multi-sector multi-region macroeconomic model, inspired by the NEMESIS model (EU version). It will be a system of economic models for 22 regions in the World (see Figure 31 and Table 26), devoted to study issues that link economic development, competitiveness, employment and notably structural policies involving long term effects, such as climate policy. The essential purpose of the model will be to assess the socioeconomic impact of some policies in a scenario comparative analysis framework. The main mechanisms of the model will be based on the behaviour of representative agents: firms, households and governments. From the supply side, the model distinguishes 58 different economic activities that produce goods and services through production functions and, to do so, use production factors: capital, energy, low and high-qualified labour and other intermediate consumption. All these economic activities are interrelated by inter-sectoral exchanges (conversion matrices) and external trades. NEMESIS-World will include a detailed energy-environment module that allows the model to deal with climate mitigation policies. This module enhances the representation of the energy-system of each region by detailing the energy used by each economic activity but also by households. It will also detail the power generation sector by allocating electricity production between different technologies through diffusion curves (see Figure 32). D3.2 – Model development strategy – Update 1 Page 78 Figure 31. Regional disaggregation in NEMESIS-World Yellow-to-red colours represent the regional groups that are explicitly disaggregated for this model. White-to-blue regions represent model regional groups that were part of the original model. Table 26. Regions in Nemesis-World European Union countries/ regions Non-EU countries/ regions AU Australia BR Brazil CA Canada CN China DE Germany ES Spain FR France ID Indonesia IN India IT Italia JP Japan KR South Korea MX Mexico OE Other EU RU Russia UK United-Kingdom US United States of America WA Rest of World - Asia & Pacific WE Rest of World - Europe WF Rest of World - Africa WL Rest of World - America WM Rest of World - Middle East D3.2 – Model development strategy – Update 1 Page 79 Figure 32. Summary of the resource, technology, and end-use demand representation in NEMESIS-World. White boxes indicate features already part of the original model. Grey boxes indicate features not part of the core nor the updated model. Dark yellow boxes indicate the features that will be, and light-yellow boxes features that might be (still to be decided), explicitly developed for NEMESIS-World in the DIAMOND project, and which are not in the core version of the model. Page 80 D3.2 – Model development strategy – Update 1 3.5.2 Model development timeline SEURECO has established a general working plan up to November 2025 (M36), with detailed process being presented in Figure 33 organised in four main activities: 1. The model construction per se with the following steps (from M2 to M36): a. the general modelling approach, started in M1 and finalised in M10 b. the identification and selection of the software(s) for input data processing, model coding and solving and outputs data processing (M3 to M12). c. the general characterisation of the model coverage (geographical, sector, energy product, etc.) (M9-M17) d. the coding of the economic model (M16-M25) e. The coding of the energy/climate model (M24-M34) f. The testing and validation of the model (M24-M36) g. The realisation of a reference scenario (M32-36) 2. The data collection, analysis and processing with (M6 – M29): a. Taking stock of existing datasets (M6 – M13) b. A benchmarking of key databases (M6 – M15) c. The processing of the selected databases (M12 – M30) 3. The model’s documentation (M13-M36) as well as the strategy to open the model (licensing, files organisation, coding documentation, code repository, etc. – M25-M36) and its implementation (M32M36). 4. the link with other activities in the DIAMOND project to take advantage of potential synergies in the development and upgrade of other models (WP3) but also on the general modelling improvements (WP4). Page 87 D3.2 – Model development strategy – Update 1 3.5.5 Planned model development steps 3.5.5.1 Interlinkages within WP4 Decile of households’ income by income source and by consumption purposes SEURECO has started a data collection process to enhance the representation of household heterogeneity. Eurostat databases, called “Income, consumption and wealth” (Eurostat, 2023a), “Structure of consumption expenditure by income quintile and COICOP consumption purpose” (Eurostat, 2023b) and “Reconciling social surveys with household accounts” Eurostat (2023c), have been combined to calculate households’ consumption expenditures by consumption purposes, households’ income decile and the source of income for each households’ income decile. SEURECO also developed a module to parse the data into the NEMESIS model (EU version, so far). For now, the module splits the households by income decile with differentiated share of income by source and with aggregated income sources coming from the agent accounts of the model. Furthermore, the module splits the total as well as each individual expenditures by consumption purposes by households’ income decile. The module is still in development, and further testing must be done on: (i) the impacts of different modelling options (e.g., use the total aggregated consumption function as a driver and use shares for total consumption expenditures for each household decile or directly use households’ decile total consumption functions, the total being the sum); (ii) the link of the consumption purpose by households’ income decile with the energy module; (iii) some analytical tests to assess the impacts of some shocks and how these influence the model results; (iv) the disaggregation of behavioural equations (i.e., parameters) by households’ decile Ultimately, the module will allow the distributional impact assessment of policies impacting income sources and/or relative cost of the different consumption purposes on households. Economic climate damages by type of impacts We completed a literature review on the expected economic impacts of 8 different climate impacts (sea level rise, river flooding, drought as well as induced economic impacts on labour productivity, agriculture yields and energy demand and supply, forestry and fishery economic activities) in EU and we estimate climate damage curves for all EU economies for 5 of these damages (SLR, river flooding, labour productivity, drought and, agriculture production). This work has been presented at the ECEMP 2024 Conference. Labour market In November 2024, we collected data for EU countries on the composition of national employment by sector, occupation and gender based on Labour Force Survey (Eurostat, 2024b). Due to confidentiality and statistical representativeness thresholds, data availability was weak for low-size countries and/or sectors. We then applied a methodology to fill in the dataset using available information or applying “closest” principle when no information was available. We will then use this information for an end-of-pipe split of the model’s employment results by economic activity, occupation and gender. Thereafter, we will engage a collaborative work with UM to compare our labour demand projections with their labour supply projections. Finance and capital markets We have improved the modelling of interest rates. We have considered that the interest rates paid by end users are the result of various successive stages. Firstly, Central Banks define short term rate for refining according to Page 88 D3.2 – Model development strategy – Update 1 their monetary policy that we assumed can be modelled using a Taylor rule. The long-term risk-free rate is then determined and on the top it, a national spread is applied depending on national key macro-financial indicators (government current balance, national public debt, external trade balance, etc.), defining as such the long-term national public bond rate (10 years). Thereafter, this rate is used to define the national interest rate applied to the private sectors that are also differentiated according to sectoral indicators on profitability and financial leverage. 3.5.5.2 Concrete next steps We identify three-time horizons for the finalisation of the model, out of linkage with WP4. Next steps – Winter 2024-2025 Up to the end of the 2024 and beginning of 2025, we will: • Finalise the processing of economic data (December 2024 / January 2025) • Complete the coding of the economic model (February 2025) • Define the parameters of the economic model: estimates, calibrations, literature review, etc. (March 2025 • Start collecting and processing energy and GHG emissions data (December 2024 to March 2025) • Adapting the coding of the economic model for the implementation of the energy module (March 2025) • Document the model, from data collection to coding (from December 2024 to March 2025) Following steps – Spring to Summer 2025 Thereafter, during spring and summer 2025, we plan to: • Finalise the data collection and processing for the energy/climate module (April 2025) • Code and calibrate the energy/climate module (June 2025) • Collect and process input data for the model (May 2025) • Testing and validation of the model (May-June 2025) • Document the model, from data collection to coding (Milestone 8, May 2025) Final steps – Autumn 2025 Finally in Autumn 2025, we will finalise the model publication completing outstanding work on previous steps, continue testing the model, run a reference scenario to be used in DIAMOND (with common set of assumptions), finalise the documentation of the model and ensure the implementation of the open-sourcing strategy for the model. 3.5.6 References for NEMESIS World • Bjelle, E. L., Többen, J., Stadler, K., Kastner, T., Theurl, M. C., Erb, K.-H., Olsen, K.-S., Wiebe, K. S. and R. Wood, 2020, Adding country resolution to EXIOBASE: impacts on land use embodied in trade, Journal of Economic Structures, vol. 9(14). Doi: 10.1186/s40008-020-0182-y. • Eurostat, 2022, FIGARO Methodology, link: https://ec.europa.eu/eurostat/documents/51957/12767369/Figaro-methodology.pdf/487255c9-903a0cb7-f5e2-73f37e35f196?t=1620750704022. Page 89 D3.2 – Model development strategy – Update 1 • Eurostat, 2023a, Share of households and economic resources by income, consumption and wealth quantiles - experimental statistics. https://ec.europa.eu/eurostat/web/products-datasets/-/icw_res_01. • Eurostat, 2023b, Structure of consumption expenditure by income quintile and COICOP consumption purpose. https://ec.europa.eu/eurostat/web/products-datasets/-/hbs_str_t223 • Eurostat, 2023c, Distribution of income and consumption for the household sector, Eurostat centralised exercise – methodological note. https://ec.europa.eu/eurostat/documents/7894008/17065819/Methodological-note.pdf/58f5ad43-ca2d9ca6-6b5c-72f553097be2?t=1689779418892 • Eurostat, 2024a, Eurostat, 2024x, Cross-classification of gross fixed capital formation by industry and by asset (flows) - nama_10_nfa_fl. Link: https://ec.europa.eu/eurostat/databrowser/view/nama_10_nfa_fl/default/table?lang=en&category=na1 0.nama10.nama_10_nfa. Accessed in September 2024 • Eurostat, 2024b, LFS series - detailed annual survey results (lfsa), Link: https://ec.europa.eu/eurostat/cache/metadata/en/lfsa_esms.htm. Accessed in October 2024. • Huo, J., Che, P., Hubacek, K., Zeng, H. and D., Guan, 2022, Full-scale, near real-time multi-regional input– output table for the global emerging economies (EMERGING), Journal of Industrial Ecology, vol. 26(4). Doi: 10.1111/jiec.13264. • Lenzen, M., Moran, D., Kanemoto, K. and A., Geschke, 2013, Building EORA: A global multi-region input– output database at high country and sector resolution, Economic Systems Research, vol. 25(1), pp. 20-49. Doi: 10.1080/09535314.2013.769938. • Lenzen, M, Geschke, A., West, J., Fry, J., Malik, A., Giljum, S., Milà i Canals, L., Piñero, P., Lutter, S., Wiedmann, T., Li, M., Sevenster, M., Potočnik, J., Teixeira, I., Van Voore, M., Nansai, K. and H., Schandl, 2022, Implementing the material footprint to measure progress towards Sustainable Development Goals 8 and 12, Nature Sustainability, vol. 5(157-166). Doi: 10.1038/s41893-021-00811-6. • OECD, 2021, Inter-Country Input-Output tables, link: http://oe.cd/icio. • OECD, 2024a, Annual capital formation by economic activity. Link: https://data-explorer.oecd.org. Accessed in September 2024. • OECD, 2024b, Annual fixed assets by economic activity and by asset. Link: https://dataexplorer.oecd.org. Accessed in September 2024 • OECD, 2024c, Annual non-financial accounts by institutional sector (Expenditure). Link: https://dataexplorer.oecd.org. Accessed in October 2024. • OECD, 2024d, Annual non-financial accounts by institutional sector (Revenue). Link: https://dataexplorer.oecd.org. Accessed in October 2024. • Stadler, K., Wood, R., Bulavskaya, T., Södersten, C.-J, Simas, M., Schmidt, S., Usubiaga, A., Acosta-Fernández, J., Kuenen, J., Bruckner, M., Giljum, S., Lutter, S., Merciai, S., Schmidt, J. H., Theurl, M.C., Plutzar, C., Kastner, T., Eisenmenger, N., Erb, K.-H., de Koning A. and A., Tukker, 2018, EXIOBASE 3: Developing a Time Series of Detailed Environmentally Extended Multi-Regional Input-Output Tables, Journal of Industrial Ecology, vol. 22(3). Doi: 10.1111/jiec.12715. Page 90 D3.2 – Model development strategy – Update 1 3.6 From PROMETHEUS to OPEN-PROM 3.6.1 Overview OPEN-PROM (“open PROMETHEUS”) is the advanced and open-source version of the PROMETHEUS model. The two models are being developed by ICCS-E3M in Athens, Greece. PROMETHEUS is a global energy system model covering in detail the complex interactions between energy demand, supply, and energy prices at the regional and global level. Its main objectives are: 1) Assess climate change mitigation pathways and low-emission development strategies for the medium and long-term 2) Analyse the energy system, economic and emission implications of a wide spectrum of energy and climate policy measures, differentiated by region and sector) 3) Explore the economics of fossil fuel production and quantify the impacts of climate policies on the evolution of global energy prices. The PROMETHEUS model provides detailed projections of energy demand, supply, power generation mix, energy-related carbon emissions, energy prices and investment to the future covering the global energy system. PROMETHEUS is a fully fledged energy demand and supply simulation model aiming at addressing energy system analysis, energy price projections, power generation planning and climate change mitigation policies. PROMETHEUS contains relations and/or exogenous variables for all the main quantities, which are of interest in the context of general energy systems analysis. These include demographic and economic activity indicators, primary and final energy consumption by main fuel, fuel resources and prices, CO2 emissions, greenhouse gases concentrations and technology dynamics (for power generation, road transport, hydrogen production and industrial and residential end-use technologies). PROMETHEUS quantifies CO2 emissions and incorporates environmentally oriented emission abatement technologies (like RES, electric vehicles, CCS, energy efficiency) and policy instruments. The latter includes both market-based instruments such as cap and trade systems with differential application per region and sector specific policies and measures focusing on specific carbon emitting activities. Key characteristics of the model, that are particularly pertinent for performing the analysis of the implications of alternative climate abatement scenarios, include world supply/demand resolution for determining the prices of internationally traded fuels and technology dynamics mechanisms for simulating spill-over effects for technological improvements (increased uptake of a new technology in one part of the world leads to improvements through learning by experience which eventually benefits the energy systems in other parts of the World). PROMETHEUS is designed to provide mediumand long-term energy system projections and system restructuring up to 2050, both in the demand and the supply sides. The model produces analytical quantitative results in the form of detailed energy balances in the period 2015 to 2050 annually. The model can support impact assessment of specific energy and environment policies and measures, applied at regional and global level, including price signals, such as taxation, subsidies, technology, and energy efficiency promoting policies, RES supporting policies, environmental policies, and technology standards. OPEN-PROM is in essence PROMETHEUS, but enhanced by expanding its regional resolution (from 10 to approximately 40 regions/countries, focusing on major emitters and EU countries, see Figure 35), time horizon (from 2050 to 2100 with annual time steps), sectoral coverage (various industrial subsectors, building types, transport modes including international shipping and aviation), and technology representation: it will include new mitigation options required for the transition to net zero, like hydrogen, synthetic e-fuels, bioenergy with carbon capture and storage (BECCS), direct air capture (DAC), advanced biofuels, carbon capture, utilization and sequestration (CCUS), etc (see Figure 36). These enhancements will be underpinned by the development of a comprehensive and open database, input data framework, ensuring consistency between various datasets related to energy system, emissions, technology data, and policies. Current representation of fossil fuel production and trade will be expanded to simulate trade in new clean energy forms (e.g., hydrogen, synthetic fuels) to allow the analysis of geopolitical implications of the transition to net zero. The model will also be expanded to incorporate Page 91 D3.2 – Model development strategy – Update 1 abatement cost-supply curves for all major non-CO2 gases by region. Several model features will be redesigned to allow soft linking with the models and methods developed in WP4 and in particular with MAGPIE (on land use), ENGAGE (for circular economy), finance and labour dynamics household heterogeneity (FIDELIO), and electricity (OpenTEPES). All model enhancements will be validated in a series of modelling benchmark scenarios, such as exogenous shocks (Harmsen et al 2021) and standard mitigation scenarios toward net-zero. Figure 35. Regional disaggregation in OPEN-PROM. Yellow-to-red colours represent the regional groups that are explicitly disaggregated for this model. White-to-blue regions represent model regional groups that were part of the original model or slightly updated multi-country aggregations. D3.2 – Model development strategy – Update 1 Page 92 Figure 36. Summary of the resource, technology, and end-use demand representation in OPEN-PROM. White boxes indicate features already part of the original model. Grey boxes indicate features not part of the core nor the updated model. Dark yellow boxes indicate the features that will be, and light-yellow boxes features that might be (still to be decided), explicitly developed for OPEN-PROM in the DIAMOND project, and which are not in the core version of the model. Page 93 D3.2 – Model development strategy – Update 1 3.6.2 Model development timeline 1. Development of a transparent, well documented, and reproducible model (ongoing) M4-M43 2. Transparent, well documented, flexible, and reproducible input data generation (ongoing) M5-M43 3. Extend time horizon to 2100 M6-M14 4. Calibration M13-M25 5. Hindcasting M35-36 6. Validate reference scenario (current policies) M33-34 7. Policy scenarios M18-M23 8. Validation against AR6, etc. M35-36 9. Output routines and dashboards M15-M20 10. Soft-coupling with other models (WP4) M10-M43 11. Scenario/sensitivity/uncertainty analyses M35-M43 12. Modularisation M35-M43 Figure 37. Timeline for the development of OPEN-PRΟΜ 3.6.3 Meetings (internal modelling team and external) The core modelling team of OPEN-PROM holds meetings daily, with the purpose of discussing model development progress, coordinating and assigning tasks to the suitable team members, and allocating resources in an optimal manner. Apart from that, regular meetings with other ICCS/E3M teams take place in an ad-hoc manner to discuss data and modelling issues. 3.6.4 Completed model development steps • July 2023: Start migrating everything to GitHub to make a full open model version; Automate data handling and management; Defined new regions and sectors; Development of new naming conventions; Identify list of technologies to be integrated; Identify the use of data sources; • August 2023: Create model wiki: https://github.com/e3modelling/OPEN-PROM/tree/main/Tutorials Open Github repo: https://github.com/e3modelling/OPEN-PROM Developoment of R package “mrprom”, for use with MADRat (open input data framework): Page 94 D3.2 – Model development strategy – Update 1 https://github.com/e3modelling/mrprom • September 2023: Load data for: Activity in sectors (GEM-E3), Fuel consumption & price (per country, sector, fuel) (ENERDATA); • October 2023: Incorporate more data on technology costs (EU Reference Scenario 2020 of the European Commission); Develop model results validation tools against historical and reference data • November 2023: initial version of transport module with 6 transport modes and various consumer groups implemented, • December 2023: Started validating the model results. • January 2024: Completed the initial preparation and processing of input data. Completed the power generation module. • February 2024: Added current carbon prices for the default scenario. • March 2024: Testing the validation tool of I2AM PARIS. • April 2024: Discussion on how to approach trade modelling. Started working on outprom: a suite of tools containing the OPEN-PROM output routines. • May 2024: Derivation of targets of calibration. Testing and developing different calibration approaches. • June 2024: Switched to the updated Shared Socioeconomic Pathways (SSP) scenarios. • July 2024: Constantly improving and updating the model input data. Tested the Initial approach for scaling. • August 2024: Tried a second approach for scaling. Started working on the power module documentation. • September 2024: Conducted preliminary model runs for the 1.5C scenario. • October 2024: Added a selection mechanism for different scenarios, i.e. NPi default, 1.5C and 2C. Added carbon prices for the 1.5C and 2C scenarios. Page 95 D3.2 – Model development strategy – Update 1 3.6.5 Current State of OPEN-PROM At this point, OPEN-PROM consists of the electricity generation, transportation, and industry modules, based on the PROMETHEUS model architecture. The OPEN-PROM model is currently functional, and generates preliminary results for all regions, until the year 2100. In addition, we have recently implemented a mechanism to run the main model scenarios, i.e. NPi (“National Policies implemented”), and deeper mitigation scenarios, consistent with the 1.5C and 2C temperature targets. Each scenario loads different exogenous carbon prices and consequently outputs the respective pathways for the global energy system. Our focus right now is on model calibration, with the purpose of aligning output with specific climate mitigation targets. Furthermore, we are also validating model results, and comparing them to other leading IAMs, such as REMIND and WITCH. As seen in Figure 40, the model can generate final energy projections for all the main scenarios. The team also included the same projections from other external IAMs, including REMIND-MAgPIE and IMACLIM, for demonstration and comparison purposes. Furthermore, CO2 emissions projections generated by OPEN-PROM and other IAMs for comparison are also included in Figure 38. It should be noted that OPEN-PROM remains under heavy development, hence the output is only shared to demonstrate its current progress and state of functionality. 3.6.6 Planned model development steps The calibrated model at the state described in the previous paragraph is expected to be the content of the initial model version to be included in MS8 (July 2025 – M32). Next steps and aspirations regarding the development of OPEN-PROM include: • Expansion of the fossil fuel production representation by use of fossil fuel extraction curves. M33-M36 • Improvement and expansion of trade representation to simulate trade of new clean energy forms (e.g., hydrogen, synthetic fuels) and to allow analysis of geopolitical implications of the transition to net zero. M33-M38 • Introduction of new mitigation options for the transition to net zero, like hydrogen, synthetic e-fuels, direct air capture (DAC), advanced biofuels, and carbon capture, utilisation and sequestration (CCUS). M33-M43 • Integration of abatement cost-supply curves for all major non-CO2 gases by region. M33-M7 • Improved sectoral coverage (various industrial subsectors, building types, transport modes including international shipping and aviation). M33-M43 Concerning soft-links, discussions are already underway for linking OPEN-PROM with MAgPIE, a leading land use and agricultural model, the MAGICC modelling suite, and the OpenTEPES electricity model, all of which are currently in the planning phase—see also detailed progress of WP4. Similar discussions are currently formulated regarding the potential links between OPEN-PROM and the circularity (i.e., ENGAGE), household heterogeneity, and the financial dimensions. Page 96 D3.2 – Model development strategy – Update 1 Figure 38. Comparison of preliminary OPEN-PROM pathways for CO2 emissions and final energy with different policy scenarios (upper graph) and with other IAMs for the 1.5C scenario (lower graph) Page 103 D3.2 – Model development strategy – Update 1 Khan, Z., Iyer, G., Patel, P., Kim, S., Hejazi, M., Burleyson, C., Wise, M., 2021. Impacts of long-term temperature change and variability on electricity investments. Nature communications 12, 1–12. Krey, V., Guo, F., Kolp, P., Zhou, W., Schaeffer, R., Awasthy, A., Bertram, C., de Boer, H.-S., Fragkos, P., Fujimori, S., He, C., Iyer, G., Keramidas, K., Köberle, A.C., Oshiro, K., Reis, L.A., Shoai-Tehrani, B., Vishwanathan, S., Capros, P., Drouet, L., Edmonds, J.E., Garg, A., Gernaat, D.E.H.J., Jiang, K., Kannavou, M., Kitous, A., Kriegler, E., Luderer, G., Mathur, R., Muratori, M., Sano, F., van Vuuren, D.P., 2019. Looking under the hood: A comparison of technoeconomic assumptions across national and global integrated assessment models. Energy 172, 1254–1267. https://doi.org/10.1016/j.energy.2018.12.131 Loulou, R., Labriet, M., 2007. ETSAP-TIAM: The TIMES integrated assessment model Part I: Model structure. Comput. Manag. Sci 5, 7–40. Miller, C.A., Wyborn, C., 2020. Co-production in global sustainability: histories and theories. Environmental Science & Policy 113, 88–95. Mulugetta, Y., Sokona, Y., Trotter, P.A., Fankhauser, S., Omukuti, J., Somavilla Croxatto, L., Steffen, B., Tesfamichael, M., Abraham, E., Adam, J.-P., 2022. Africa needs context-relevant evidence to shape its clean energy future. Nature Energy 1–8. Nikas, A., Gambhir, A., Boitier, B., 2023. Promoting sustainable transitions across the globe. Special Issue in Renewable and Sustainable Energy Transition. Nikas, A., Skalidakis, S., Sorman, A.H., Galende-Sanchez, E., Koasidis, K., Serepas, F., Van de Ven, D.-J., Moreno, J., Karamaneas, A., Koutsellis, T., 2021. Integrating integrated assessment modelling in support of the Paris Agreement: The I2AM PARIS platform. Presented at the 2021 12th International Conference on Information, Intelligence, Systems & Applications (IISA), IEEE, pp. 1–8. O’Neill, B.C., Kriegler, E., Riahi, K., Ebi, K.L., Hallegatte, S., Carter, T.R., Mathur, R., van Vuuren, D.P., 2014. A new scenario framework for climate change research: the concept of shared socioeconomic pathways. Climatic Change 122, 387–400. https://doi.org/10.1007/s10584-013-0905-2 Peng, W., Iyer, G., Bosetti, V., Chaturvedi, V., Edmonds, J., Fawcett, A.A., Hallegatte, S., Victor, D.G., van Vuuren, D., Weyant, J., 2021. Climate policy models need to get real about people—here’s how. Pindyck, R.S., 2017. The use and misuse of models for climate policy. Review of Environmental Economics and Policy. Pindyck, R.S., 2013. Climate change policy: what do the models tell us? Journal of Economic Literature 51, 860– 872. Pirani, A., Fuglestvedt, J.S., Byers, E., O’Neill, B., Riahi, K., Lee, J.-Y., Marotzke, J., Rose, S.K., Schaeffer, R., Tebaldi, C., 2024. Scenarios in IPCC assessments: lessons from AR6 and opportunities for AR7. npj Climate Action 3, 1. Poblete-Cazenave, M., Pachauri, S., Byers, E., Mastrucci, A., Van Ruijven, B., 2021. Global scenarios of household access to modern energy services. Nature Energy. Pye, S., Welsby, D., McDowall, W., Reinauer, T., Dessens, O., Winning, M., Calzadilla, A., & Bataille, C. (2022). Regional Page 104 D3.2 – Model development strategy – Update 1 uptake of direct reduction iron production using hydrogen under climate policy. Energy and Climate Change, 3, 100087. https://doi.org/https://doi.org/10.1016/j.egycc.2022.100087 Rao, N.D., van Ruijven, B.J., Riahi, K., Bosetti, V., 2017. Improving poverty and inequality modelling in climate research. Nature Climate Change 7, 857–862. Robertson, S., 2021. Transparency, trust, and integrated assessment models: An ethical consideration for the Intergovernmental Panel on Climate Change. Wiley Interdisciplinary Reviews: Climate Change 12, e679. Rosen, R.A., 2015. IAMs and peer review. Nature Climate Change 5, 390–390. Sampedro, J., 2021. Accounting for access and inequality. Nature Energy 6, 773–774. https://doi.org/10.1038/s41560-021-00872-z Sampedro, J., Markandya, A., Rodés-Bachs, C., Van de Ven, D.-J., 2023. Short-term health co-benefits of existing climate policies: the need for more ambitious and integrated policy action. The Lancet Planetary Health 7, e540– e541. Secretariat, U., 2016. Just transition of the workforce, and the creation of decent work and quality jobs. FCCC/TP/2016/7. UNFCCC. Seneviratne, S., Quilcaille, Y., Windisch, M., Gudmundsson, L., Biess, B., Jaeger, F., Hauser, M., Hirschi, M., 2024. Using regional ESM emulators to assess climate feedbacks to IAMs: The" FASTMIP" experimental protocol. Copernicus Meetings. Skea, J., Shukla, P., Al Khourdajie, A., McCollum, D., 2021. Intergovernmental Panel on Climate Change: Transparency and integrated assessment modeling. Wiley Interdisciplinary Reviews: Climate Change 12, e727. Sognnaes, I., Gambhir, A., van de Ven, D.-J., Nikas, A., Anger-Kraavi, A., Bui, H., Campagnolo, L., Delpiazzo, E., Doukas, H., Giarola, S., 2021. A multi-model analysis of long-term emissions and warming implications of current mitigation efforts. Nature Climate Change 11, 1055–1062. United Nations, 2008. International Standard Industrial Classification of All Economic Activities - Statistical Papers (No. Sales No. E.03.XVII.4, No.4, Rev.31). United Nations publication. van Beek, L., Hajer, M., Pelzer, P., van Vuuren, D., Cassen, C., 2020. Anticipating futures through models: the rise of Integrated Assessment Modelling in the climate science-policy interface since 1970. Global Environmental Change 65, 102191. Van der Sluijs, J.P., 2002. A way out of the credibility crisis of models used in integrated environmental assessment. Futures 34, 133–146. van der Wijst, K.-I., Bosello, F., Dasgupta, S., Drouet, L., Emmerling, J., Hof, A., Leimbach, M., Parrado, R., Piontek, F., Standardi, G., van Vuuren, D., 2023. New damage curves and multimodel analysis suggest lower optimal temperature. Nature Climate Change. https://doi.org/10.1038/s41558-023-01636-1 Van Ruijven, B.J., O’Neill, B.C., Chateau, J., 2015. Methods for including income distribution in global CGE models for long-term climate change research. Energy Economics 51, 530–543. Page 105 D3.2 – Model development strategy – Update 1 Weyant, J., 2017. Some contributions of integrated assessment models of global climate change. Review of Environmental Economics and Policy. Wilson, C., Guivarch, C., Kriegler, E., Van Ruijven, B., Van Vuuren, D.P., Krey, V., Schwanitz, V.J., Thompson, E.L., 2021. Evaluating process-based integrated assessment models of climate change mitigation. Climatic Change 166, 1– 22. Wise, M., Patel, P., Khan, Z., Kim, S.H., Hejazi, M., Iyer, G., 2019. Representing power sector detail and flexibility in a multi-sector model. Energy Strategy Reviews 26, 100411. Page 106 D3.2 – Model development strategy – Update 1 ANNEX Model documentation survey The internal surveys with the updated information on the IAMs, including their regional, sectoral, policy coverage, as well as the calibration data and input assumptions, and the identified stakeholder inputs are publicly available on Zenodo here: https://zenodo.org/records/14216491