- Climate models/tools are not specifically developed and/or tuned to provide deep insights into climate change impacts on Nordic environments, economies and societies as a whole
- Orchestration of compute and data resources across different sites is currently not possible while necessary for an efficient execution of ESM workflows (from the creation of an ESM experiment, post-processing and long-term archiving)
- Making the best use of EUROHPC requires fundamental upgrading of the entire workflows and strategies for tackling issues related to exascale computing (access to data, hardware failures, scalability, etc.)
- Insufficient e-infrastructure & services as well insufficient expertise to adapt and apply FAIR principles to Climate science
Towards Open Science, reproducibility of Earth System runs is seen as increasingly important along with producing accurate and FAIR climate data. However, improving the scientific accuracy of the Earth System Models requires huge computing and storage resources that are still provided locally e.g. at national level. This hinders bit-replicability and makes it difficult for researchers to compare their developments and confidently validate the scientific results. Earth System Modeling research is essential to assess the drastic impacts of climate change in the Nordics but cannot be expected to be carried out by other international teams who obviously focus on their own regional issues. The EuroHPC initiative is a unique opportunity for the Nordic countries towards a joint Climate Model Intercomparison Projects (CMIP) participation provided that all the necessary building blocks (containerization, workflow management, job orchestration, model diagnostics) for creating fully reproducible and efficient ESM workflows as well as for publishing FAIR climate data are in place.
NICEST-2 aims at
i) Improving the e-infrastructure to tackle these challenges in optimal conditions, with minimal waste of resources, and
ii) Setting up the building blocks for ESM workflows which are to be deployed as part of the Nordic European Open Science Cloud (EOSC), thereby realizing its full potential for the benefit of the Nordics.
- More ESM simulations done in less time through performance enhancements of the codes and optimization of the workflows for the entire community
- Homogeneous workflows for ESMs (NorESM, EC-Earth)
- Common diagnostic tools/workflows for highlighting climate impact in the Nordics
- Lower cost and better energy efficiency of the simulations
- Possibility to have a joint Nordic CMIP participation
- Benefit for all ESM researchers in the Nordic, not only those involved in CMIPs
- Climate data would be easier to access and exploit, even for non-specialists