After four days of hacking and collaborating together, March 28th, 2025, marked the final session of „The Final Countdown“ hackathon in Stockholm. Focused on the possible applications of Storm-Resolving Earth System Models (SR-ESM) in the renewable energy sector, this was also the last hackathon of the nextGEMS project. Some extraordinary accomplishments were made during the past 5 years since the start of this visionary project. The group was, for example, able to create 10, 5, and 2 kilometer-scale runs, and are close to release runs with a resolution of 1 kilometer. However, nothing stands out more than the solid and compromised community nextGEMS has built up through the years: talented and curious people working together to push forward high-resolution climate modeling and understanding the possibilities it comprises for a warming and changing planet.

The last hackathon day involved a small plot contest, won by scientist Matthias Aengenheyster from the European Center for Medium-Range Weather Forecasts (ECMWF). His visual shows wind gust speed and 2-minute averaged precipitation in an area around Japan for the coupled IFS-FESOM model simulation at 2.8 km resolution. Near the bottom, a tropical cyclone is approaching Japan, while another one near the top is transitioning from inter-tropical to an extra-tropical cyclone as it moves to cooler latitudes.

During the closing session, the five thematic groups were able to share their advancements with the audience. For instance, the Renewable Energy group talked about their efforts to monitor wind speed shifts in the central Sweden region with nextGEMS’ climate models, where some of the stakeholder companies had placed wind turbines. In parallel, the Storms & Radiation team shared their intentions to prepare 3 thematic research papers on diverse topics, such as climate sensitivity, feedback decomposition and tropical cloud organization—the last one with a special focus on deep convective clouds.

In an engaging presentation, some of the early-career scientists and first-time hackathon attendees who participated in the Storms & Land group, provided interesting insights of the snow coverage observations in the Iberian mountain range. Similarly, the Storms & Ocean team share their observations of mesoscale ocean circulation patterns (typically between 10 to 500 km in diameter) occurring at shallow depths, as well as their interest of observing the historical future of “El Niño” phenomenon. The Storms & Society thematic group conducted several interviews during the event and disseminated a final survey with the participants to finish their work on Climate Science storylines and the impact of hackathons in knowledge co-production. 

Finally, Climate Physics Director at the Max Planck Institute for Meteorology, Bjorn Stevens, closed the event remarking some of the useful applications nextGEMS’ models have enabled, such as testing hypotheses underpinning climate change, studying changes at the mesoscale or blocking statistics, and the representation of hydrological extremes worldwide. Furthermore, he mentioned some of the forthcoming activities for the community, spearheaded by nextGEMS, such as the upcoming Global Hackathon taking place in May, 2025.

The nextGEMS project has entered its final phase and will come to an end in August, 2025. But before going separate ways, our project  members and partners gather one last time for the sixth nextGEMS hackathon from March 24th to 28th. In the stylish surroundings of Stockholm city, the Swedish Museum of Natural History, the largest museum of the Nordic country, hosts “The Final Countdown”. This time, the participants´ challenge is centered around how the high-resolution capabilities enabled by nextGEMS simulations can support and enhance renewable energy applications in a changing climate.

The first day kicked off with the arrival of a diverse group of scientists, stakeholders, students, and climate enthusiasts that totaled 73 registered participants. Within the museum´s classic setting, the introductory session evolved into an active and engaging conversation. Representatives from the Max Planck Institute for Meteorology (MPI-M) and the European Center for Medium-Range Weather Forecasts (ECMWF), updated the audience on the progress being made with the simulations of the ICON and IFS-FESOM Earth System models. 

Tobias Becker, researcher from the ECMWF, presented insights on two simulations at 2.8 km resolution, produced with 14 months of new data using the IFS-FESOM model. These recent advancements add local granularity and allow to check if atmospheric phenomena previously analyzed at coarser or less detailed resolutions also show up at this higher resolution. Additionally, he reported on two 30-year simulations – historical and scenario-based – at 9 km resolution that should provide valuable information on how extreme events change in warming climate, such as tropical cyclones.

The different thematic groups—Storms & Land, Storms & Ocean, Storms & Radiation, and Storms & Society— discussed their newest achievements and upcoming challenges. Dragana Bojovic, from the Storms & Society group, for example, talked about the survey analysis from the past five hackathons, as well as of the work on renewable energy and fisheries storylines. This time, a new group joined the Stockholm hackathon: the renewable energy group. This group includes not only researchers, but also different industry stakeholders, such as people working at Vestas, Satkraft, Anemos, and local participants, addressing future energy scenarios for 2050.

To conclude the day, participants took part in an ice-breaker session, which included a micro-poster activity designed to enrich conversations and connections through the use of visualizations. Some of the first-time participants in the event, like Diego Garcia and Antonio Robles from Universidad Complutense de Madrid in Spain, shared posters illustrating their observations on historical data regarding snow coverage along the Spanish highlands and future changes in Tropical Basin interactions, created with the IFS-FESOM model.

Easy Gems is an online platform that consolidates information on high-resolution climate simulations produced by the nextGEMS project and other European initiatives, such as European Eddy-Rich ESMs (EERIE), WarmWorld, and the DYAMOND initiative. Developed by the German Climate Computing Center (DKRZ), this platform serves as a repository of best practices and a comprehensive guide to create climate models. According to DKRZ Senior Scientist Florian Ziemen, one of the key aspects behind the creation of Easy Gems was “the idea was to have one place where you can find everything you need when you want to analyze high-resolution simulations.”

At the same time, the goal was to build something that was not tied strictly to a single project, thereby ensuring the sustainability of the platform and its data even after individual projects conclude. This approach prevents the platform from ending up in the „website dumpster,“ Ziemen explained. As a result, Easy Gems is designed to extend beyond individual projects by offering access to all simulation data—or „simulation gems“—hosted at DKRZ, while also functioning as a how-to resource or e-book.

Since nextGEMS drives the development of two European storm-resolving Earth-system models — the ECMWF Integrated Forecasting System (IFS) and the Icosahedral Nonhydrostatic Weather and Climate Model (ICON) — Easy Gems includes details on three development cycles of these Earth System Models, as well as some pre-final simulations. This encompasses simulations at various horizontal resolutions and evolving model configurations.

Additionally, the platform offers guidance on logging data, plotting data, and applying a variety of best practices in data processing. It is entirely user-driven, meaning that the Easy Gems community actively contributes and keeps the content up-to-date. Beyond serving as a guide, the platform acts as a comprehensive documentation tool for the project, providing access to project outputs that are further illustrated and explained.

The platform is organized into three main sections: Simulations, Processing, and Contribute. The Simulations section provides detailed information on the simulations currently available, while the Processing section offers tips and example scripts for data handling. Finally, the Contribute section explains how users can collaborate and become part of this community-driven effort.

Easy Gems encourages contributions from anyone, regardless of their background. Contributions can range from reporting errors and requesting additional articles to suggesting clearer descriptions or improved illustrations. To contribute and interact with Easy Gems, users need a DKRZ account. Additionally, additions require approval from other members of the community to ensure that the input is correct and avoid redundancies. Currently, there are 3,260 registered accounts, with approximately 490 users granted access to all the data hosted on the platform.

If you’re interested, feel free to check out Easy Gems here!

Nowadays, Machine Learning (ML) can assist in identifying climate models based on daily output. How? Firstly, it is important to understand ML as the process of developing algorithms that enable computers to learn and make predictions or decisions based on data, without being explicitly programmed for each scenario. In that context, ML techniques, such as Convolutional Neural Networks (or CNNs), are increasingly utilized in Climate Science to evaluate climate models; identify model characteristics; and assess model performance in comparison to observational data.

The study “Identifying climate models based on their daily output using machine learning”, by researchers Lukas Brunner and Sebastian Sippel, shed light on the use of ML classifiers – such as the CNNs mentioned before. Specifically, on how they can be trained to robustly identify climate models using daily temperature output. 

By analyzing individual daily temperature maps, ML methods can separate models from observations and from each other, even in the presence of considerable noise from internal variability on specific weather timescales (Brunner & Sippel, 2023). Internal variability refers to the fluctuations in the climate system that arise from various processes within the Earth’s atmosphere, oceans, and land surfaces that we might refer to as weather. Hence, the ML approach allows for the identification of models and observations based on short timescales, providing new ways to evaluate and interpret model differences.

Separating models from observations, and from each other

The study used daily temperature maps from 43 Coupled Model Intercomparison Projects (CMIP6) models and four different observational datasets. Additionally, ICON-Sapphire, one of the Earth system models developed by nextGEMS, was utilized as an experimental km-scale model. With that basis, two different statistical and ML methods were used to separate models from observations, and from each other.

Firstly, through logistic regression, the researchers were able to distinguish between models and observations because it allows the appreciation of the learned coefficients (Brunner & Sippel, 2023). The coefficients learned by the logistic regression classifier reveal that many well-known climatological model biases are already emerging as important for identifying daily maps. Nonetheless, other regions like the Arctic are not relevant for daily classification at all. 

It is important to mention that logistic regression is a linear method and, after bias correction with the mean seasonal cycle, it is no longer skilful. To complement logistic regression, a second methodnamely CNN, specially due to the possibility of obtaining more trainable parameters that can also lean more complex, non-linear relationships within the data. (Brunner & Sippel, 2023)

Main findings and future directions

Some of the main results of this research work are related to the high accuracy achieved by CNN classifiers in identifying models and observational datasets, even when faced with complex classification tasks. Overall accuracy of 83% was achieved in identifying 43 models and four observational datasets (Brunner & Sippel, 2023). Moreover, CNNs could pick up unique patterns specific to each dataset, enabling successful separation from other datasets. Generally, it is important to take away that dependencies between models – and observations – emerge even on daily time scales.

On another note, Brunner and Sippel (2023) clarified that misclassifications often occurred within model families or were related to common “ancestors”, indicating shared features among related datasets. However, the study revealed the ability of the CNN to correctly identify a significant portion of test samples, even those from distant time periods and under different climate scenarios.

The authors anticipate a planned follow-up study that aims to analyze the origin of classification skill in more detail, using explainable ML techniques and domain-specific approaches from Climate Sciences. In other words, the follow-up study will investigate the coupling of atmosphere and ocean; surface energy balance in models; and targeted masking of regions to understand model performance dependencies. 

If you are interested in working with this method, feel free to do so! The researchers have made the code used in the paper freely available on Github.

References:

Brunner, L., & Sippel, S. (2023). Identifying climate models based on their daily output using machine learning. Environmental Data Science, 2. https://doi.org/10.1017/eds.2023.23