Climate and Weather Modeling

The consequences of climate change are unpredictable, with extreme weather events affecting unexpected areas. Climate research relies on numerical simulations based on models developed over decades.

Advances in supercomputers now enable the simulation of phenomena like cloud formation or storms on a global scale. However, a key challenge is the gap between technological progress and model development cycles.
Adapting complex software, built over decades, to new hardware—quickly replaced by emerging architectures—is becoming increasingly difficult.

On Climate and Weather Modeling area, HANAMI will focus on two different projects:

1
Replicability and Skill of Earth System Models - Workflow and Interface
2
Optimising, Porting and Testing European and Japanese Weather and Climate Codes

Replicability and Skill of Earth System Models - Workflow and Interface

This project will develop a workflow for testing three aspects of Earth System Models (ESM):

1
Replicability
2
Projection Skill
3
Performance
Most model developers use their own tests to check performance, accuracy, and efficiency. However, the knowledge of how to get useful results is often kept within individual teams and rarely shared. Still, there are well-established methods accepted by the community. Researchers want to bring these methods together into a single workflow with a simple interface that works for any weather or climate model and can run on any cluster.

Optimising, Porting and Testing European and Japanese Weather and Climate Codes

The High-Performance Climate and Weather benchmark (HPCW) is a specialized benchmarking tool developed in the H2020 projects ESCAPE-2 and ESiWACE2 by ATOS and DKRZ, with support from ECMWF. It includes a set of tests that reflect the latest advances in high-performance weather and climate simulations.
HPCW offers a more practical way to evaluate computing systems than traditional benchmarks like High-Performance Linpack. Currently, it includes European climate and weather models, but this project aims to expand it to include Japanese models as well.

Meet the Team

WP4

Hanna Pawlowska

University of Warsaw (UW)

WP4

Joachim Biercamp

DKRZ

WP4

Kai Keller

Barcelona Supercomputing Centre (BSC)

WP4

Samuel Hatfield

ECMWF

WP4

Piotr Dziekan

University of Warsaw (UW)

WP4

Mario Acosta

Barcelona Supercomputing Centre (BSC)

WP4

Leo Arriola

WP4

Marta Alerany

WP4

Shin-ichiro Shima

WP4

Emmanuel Barillot

Keep exploring
our project!

Results

Learn more