We aim to address the need for fundamentally new strategies in analyzing the big-data output from next-generation climate models.The new German-community climate model ICON is a next-generation model (Zängl et al.,2015). Thanks to its triangular grid, ICON runs effectively on tens of thousands of CPUs and harvests advances in supercomputing.
In contrast to previous climate models this new model is based on a triangular grid. To provide fast computing algorithms one can thus no longer work with a cube-grid structure.
The main idea of this project is to use a technique from geometric group theory to translate the triangle structure into a parallel grid and back and thus to provide a methods to integrate existing fast algorithms into the new model.
This project won the "Best grant proposal award 2018" by the YIN@KIT