Sidebar
working_groups:computational_earth_sciences
Table of Contents
Computational Earth Sciences
Description of the group
The Computational Earth Sciences (CES) group is a multidisciplinary team within the Earth Sciences Department of the Barcelona Supercomputing Center (BSC) that develops, maintains, and optimises computational applications, tools and infrastructures supporting environmental and climate research.
CES brings together computer scientists and engineers who work closely with other scientific teams to ensure efficient, scalable, and sustainable use of Information Technology (IT) resources, with a special focus on High-Performance Computing (HPC). The group’s mission is to enable and accelerate scientific discovery in Earth system modelling, atmospheric processes, ocean information systems, and air quality studies through advanced computing, data management, and artificial intelligence (AI) techniques.
Main Objectives
The group’s activities focus on the following areas:
- Efficient use of HPC resources: Design, implement, and maintain computational methodologies and frameworks that enable Earth researchers to perform large-scale simulations and data analyses effectively. This includes ensuring adequate storage and computational resources.
- Software optimization and HPC adaptation: Research, develop and adapt scientific software and Earth system models to run efficiently on current and next-generation HPC architectures, moving towards exascale computing. This includes profiling, parallelization, GPU adaptation and co-design for novel technologies.
- Data management and dissemination: Build and maintain frameworks for storing, standardizing, and disseminating model outputs and datasets generated by the Earth Sciences Department, promoting open science and interoperability with external research and service communities.
- Computational workflows: Develop automated workflows and workflow management software for running complex Earth system model experiments across diverse HPC environments, ensuring portability, scalability, and efficiency through robust software engineering, collaboration, and continuous training.
- Scientific software development: Design, implement, and maintain efficient, portable, and well-documented tools for data processing, diagnostics, and visualization, ensuring high performance, scalability, and accessibility to support Earth system modelling, analysis, and collaborative research.
- IT infrastructure: Maintain an IT infrastructure that ensures an efficient working environment for the development of the BSC-ES research and services, including the provision of sufficient storage to reduce this specific vulnerability of the department.
- Good practices and collaborative workflows: Provide guidance on coding standards, version control, testing, and collaborative software development to improve productivity, ensure reproducibility, and facilitate onboarding of new team members.
- Training and technical support: Deliver continuous training and technical assistance to researchers and IT staff to ensure effective and sustainable use of computational resources and tools.
- Integration of Artificial Intelligence: Explore and apply AI and machine learning methods to solve Earth Sciences problems, such as speeding up numerical simulations, improving predictive skill, and creating human-machine interfaces. Additionally, support data-driven research in climate and atmospheric sciences.
- European integration and computational excellence: Actively engage in European and international research initiatives to foster collaboration, excellence, and impact in the field of computational Earth system sciences. Contribute to Horizon Europe projects, Centres of Excellence, and related communities to strengthen the group’s role within the European HPC and climate research ecosystem.
Organisation and Leadership
The CES group is led by Mario Acosta and Miguel Castrillo, and is organised into four specialised teams and eight units, each addressing complementary aspects of computational Earth sciences:
Teams
Units
Group Members
PhD Students
| Student | Advisor(s) | Project | Dates |
|---|---|---|---|
| Manuel G. Marciani | Gladys Utrera, Mario C. Acosta | Enhancing Workflow Efficiency in High-Performance Computing Environments for Earth Sciences | November 2023 - November 2026 |
| Sergi Palomas | Gladys Utrera, Mario C. Acosta | Optimizing Earth science simulations: enhancing performance metrics, benchmarking, and evaluation tools for high-performance architectures. | Dec 2023 - Dec 2026 |
| Okke van Eck | K. Anton Feenstra, Daan van den Berg, Mario C. Acosta | Investigating the Instance Hardness for HP-Model Protein Folding Problem Instances | June 2026 |
Research Outputs, Dissemination and Engagement
In this section, one can find some guidelines when starting working in CES
Software and Projects
| Software or Projects | Brief Description | Reference or Publications | Coordinator and Contact |
|---|---|---|---|
| Autosubmit | Autosubmit is a flagship innovation open-source tool developed and led by CES that, since 2022, has consolidated its role as a production-grade workflow framework for large-scale Earth system experimentation. Crucially, Autosubmit has been adopted as the common workflow manager across models and institutions involved in the Climate DT of Destination Earth, enabling coordinated, reproducible, and fault-tolerant execution of complex simulation campaigns on EuroHPC JU systems. This adoption demonstrates CES’ leadership in delivering a shared digital backbone for a pan-European flagship initiative, transforming research software into a strategic infrastructure supporting Europe’s climate digital transformation. | D. Manubens-Gil, J. Vegas-Regidor, C. Prodhomme, O. Mula-Valls and F. J. Doblas-Reyes, “Seamless management of ensemble climate prediction experiments on HPC platforms,” 2016 International Conference on High Performance Computing & Simulation (HPCS), Innsbruck, Austria, 2016, pp. 895-900. https://doi.org/10.1109/HPCSim.2016.7568429 IEEE pdf | Coordinator: Miguel Castrillo miguel.castrillo@bsc.es Contact: Daniel Beltrán daniel.beltran@bsc.es, Miguel Castrillo miguel.castrillo@bsc.es |
| CALIOPE | CALIOPE is a long-standing innovation open-source application of CES, result of a collaboration with the Generalitat de Catalunya, that has been further consolidated since 2022. It is an operational, high-resolution air-quality modelling system providing daily forecast for next 48 hours of main air pollutants and supporting environmental assessment and public decision-making in three different spatial domains: Spain, Catalonia and Barcelona city, with the idea of extending in the near future the high-resolution urban model to other cities in Catalonia. By integrating advanced atmospheric models, data pipelines and diagnostics, CALIOPE translates state-of-the-art computational research into a policy-relevant service with direct societal impact in air quality and public health, empowering citizens, policymakers and technical users by monitoring and predicting the evolution of the main air pollutants affecting the quality of the air. This action exemplifies CES capacity to bridge scientific excellence and sustained operational deployment at territorial level. | Baldasano, J. M., Pay, M. T., Jorba, O., Gassó, S., & Jiménez-Guerrero, P. (2011). An annual assessment of air quality with the CALIOPE modeling system over Spain. Science of the Total Environment, 409(11), 2163-2178 https://doi.org/10.1016/j.scitotenv.2011.01.041 Benavides, J., Snyder, M., Guevara, M., Soret, A., Pérez García-Pando, C., Amato, F., Querol, X., & Jorba, O. (2019). CALIOPE-Urban v1. 0: coupling R-LINE with a mesoscale air quality modelling system for urban air quality forecasts over Barcelona city (Spain). Geoscientific Model Development, 12(7), 2811-2835 https://doi.org/10.5194/gmd-12-2811-2019 | Coordinator and Contact: Francesco Benincasa francesco.benincasa@bsc.es |
| ESMValTool | ESMValTool is a community diagnostics and performance metrics tool for the evaluation of Earth System Models (ESMs) that allows for routine comparison of models and observations. It includes a large collection of community recipes and observation data formatters to CMOR standards. ESMValCore is a software package which provides the core functionality for ESMValTool. It is a workflow to find CMIP data, and apply commonly used pre-processing functions. To get a first impression of what ESMValTool and ESMValCore can do for you, have a look at our blog posts Analysis-ready climate data with ESMValCore and ESMValTool: Recipes for solid climate science. | Righi, M., Andela, B., Eyring, V., Lauer, A., Predoi, V., Schlund, M., Vegas-Regidor, J., Bock, L., Brötz, B., de Mora, L., Diblen, F., Dreyer, L., Drost, N., Earnshaw, P., Hassler, B., Koldunov, N., Little, B., Loosveldt Tomas, S., and Zimmermann, K. (2020): Earth System Model Evaluation Tool (ESMValTool) v2.0 – technical overview, Geosci. Model Dev., 13, 1179–1199, https://doi.org/10.5194/gmd-13-1179-2020. | Coordinator (Project Co-PI): Birgit Hassler Contact: Pierre Antoine Bretonniere pierre-antoine.bretonniere@bsc.es |
| SUNSET | SUNSET is an R-based tool that provides climate services for sub-seasonal, seasonal and decadal climate forecast horizons. The tool post-processes climate forecast outputs by applying state-of-the-art methodologies to tailor climate products for each application and sector (e.g.: agriculture, energy, water management, or health). Its modular design allows the technicians and researchers to decide on the post-processing required steps, such as regridding, anomalies, downscaling, bias-adjustment methods, as well as the products definition by deciding on the forecast system and reference datasets, variables, and forecast horizon among others. The tool also allows the creation and visualization of climate forecast products, such as maps for the most likely terciles, and performs the verification of the products using user-defined metrics, which can be visualized on maps and scorecards. The integration of Autosubmit in the tool allows users to easily parallelize the computation in HPC machines. | N. Pérez-Zanón, V. Agudetse, Ll. Palma, A. Ho, C. Delgado-Torres, N. Milders, E. Duzenli, A. Llabrés-Brustenga, B. de Paula Kinoshita, P. Bretonnière, and A. Muñoz, “SUNSET: SUbseasoNal to decadal climate forecast post-processIng and asSEssmenT suite”, EMS Annual Meeting Abstracts, Vol. 21, EMS2024-361, 2024, updated on 11 Apr 2025 https://doi.org/10.5194/ems2024-361 | Coordinator: Núria Pérez-Zanón nuria.perez@bsc.es Contact: Victòria Agudetse victoria.agudetse@bsc.es |
Innovation Actions and Collaborations
| Type | Role of CES | Coordinator and Contact |
|---|---|---|
| Contract with public administration AEMET-SDS-WAS Sand and Dust Storms Warning Advisory and Assessment System operational services | CES is in charge of leading the developments and operational services of the Barcelona Dust Regional Center, which includes the daily run of the in-house numerical model MONARCH and the collection of 14 more models’ outputs from external partners. This is done through official contracts in collaboration with AEMET. It’s the first dust center officially recognized by the WMO, delivering daily operational dust forecasts, multi-model products and evaluation plots and statistics against Earth Observations. All mentioned products are generated in a High Availability configuration infrastructure, consisting in the same setup replicated in AEMET and BSC facilities: a Virtual Machine exposing a web portal, an interactive dashboard displaying maps and plots, and a data server providing binary data, everything connected to an HPC cluster. The current contract started in 2022 (€1.175.000 funded) but this is a collaboration of more than one decade between CES, AEMET and the WMO, becoming a reference for academy, research institutes, national meteorological agencies and policy makers. | TBC |
| BSC AI Factory | CES has a critical role in the BSC AI Factory , in which it contributes software, data, and AI expert services, while leading the Software Services work package. The BSC AI Factory (€42,000,000 funded, 2025–2027) is a strategic pillar of Europe’s emerging AI innovation ecosystem. The initiative delivers a comprehensive portfolio of AI-oriented services, advanced training, and networking infrastructures tailored to accelerate AI adoption across industry, particularly among SMEs and start-ups. This role positions CES at the forefront of integrating artificial intelligence with scientific computing infrastructures, strengthening its capacity to translate AI innovation into Earth system applications and reinforcing the strategic alignment between digital transformation, environmental sciences, and sustainable supercomputing. | TBC |
| Industrial collaboration with NVIDIA and HUAWEI on GPU porting and optimisation of the NEMO ocean model | In recent years, CES has led a strategic industrial collaboration through two contracts with NVIDIA (PI: Francisco Doblas-Reyes, one year, €350,000 funded, 2025) and HUAWEI (PI: Mario Acosta, two years, €400,000 funded, 2022 and 2023) focused on the GPU porting and optimisation of the NEMO ocean model, a cornerstone of European ocean and Earth system modelling. NEMO is used by major European institutions and infrastructures and underpins key initiatives such as Destination Earth and EuroHPC JU-funded climate projects. This collaboration positions CES as a trusted partner for industrial co-design, strengthening European readiness for heterogeneous supercomputing systems and ensuring the long-term sustainability and performance of a critical scientific asset. | TBC |
Publications
See peer publications here
Other Publications
See other publications here
Posters
See posters here
Invited Talks
See information about the invited talks here
Technical Memoranda
See the technical memoranda here
Events and Visits
See events and visits (co-)organised by CES here
Templates and Resources
Presentations
Oboarding Guide
PDPs
Consent forms
Standardized informed consent templates have been developed to be used across all projects and activities. These templates have been designed to take into account both current uses of personal data and potential future needs. Please read the guidance document explaining how to use these templates and addressing some frequently asked questions: how_to_use_the_es_consent_forms.pdf
| Template | Description | Download |
|---|---|---|
| Informed Consent Form – Earth Sciences | Main informed consent document | es_consent_form.docx |
| Informed Consent Form – Online Version | Version to be used when consent is collected through an online platform | online_consent_form.docx |
| Informed Consent – Full Information Notice | Complete information notice to be linked from the online consent form | full_information_notice.docx |
GDPR compliance check list
The Legal, DPO, and COMSEG teams have prepared a form that must be completed by the Principal Investigator (PI) for any new project or proposal involving personal or sensitive data. This includes anonymized, pseudonymized, or raw data that requires access or processing by BSC-CNS staff or systems. Based on the type of data your project will handle, the form will guide you on whom to contact and outline the necessary approvals to ensure compliance with data protection regulations and the required technical security measures.
The completed form must be submitted to COMSEG at comseg@bsc.es. Approval from both COMSEG and the DPO is mandatory before any project can begin. No project implementation can start without their express approval. Once the form has been approved by COMSEG and the DPO, the PI must send it to the Legal team, along with a request for the review of the corresponding legal documentation.
| Template | Description | Download |
|---|---|---|
| GDPR Form | Form that must be completed by the Principal Investigator (PI) for any new project or proposal involving personal or sensitive data. | GDPR form |
CES Meetings
working_groups/computational_earth_sciences.txt · Last modified: 2026/03/06 10:05 by ysamper
