MeteoSwiss and SDSC join forces to refine weather forecasts using AI

MeteoSwiss and SDSC join forces to refine weather forecasts using AI
MeteoSwiss and SDSC join forces to refine weather forecasts using AI

MeteoSwiss and the Swiss Data Science Center have signed a contract aimed at exploiting the full potential of AI in weather and climate forecasting. The collaboration will be able to rely on the CSCS Alps supercomputer.

The Federal Office of Meteorology and Climatology (MétéoSuisse) has announced the signing of a framework contract with the Swiss Data Science Center (SDSC). This partnership aims to develop advanced methods leveraging artificial intelligence (AI) to improve weather forecasting and climate analysis. This project will make it possible to exploit 160 years of Swiss meteorological data, paving the way for new climate analyzes and more reliable forecasts.

A key goal of this partnership is to improve the accuracy and speed of forecasts, both for immediate weather events such as storms and for long-term climate change. MeteoSwiss and the SDSC will work in close collaboration with the Swiss Center for Scientific Computing (CSCS), using the Alps supercomputer.

Over the next four years, the partners will focus their efforts on projects that will gradually integrate these methods into existing MeteoSwiss systems. The expected improvements concern short and medium term forecasts, with increased precision for the next ten days. The technical developments envisaged also aim to increase the automation of the processing of new data and the control of the quality of measurements. The data collected and the new methods developed will be made available to the scientific community.

As a reminder, last May, MétéoSuisse implemented ICON, a new model taking advantage of the Alps supercomputer. As part of a recent report entirely devoted to the contribution of AI and data to weather forecasts, the editorial staff spoke with Lionel Moret, head of a team dedicated to machine learning at MétéoSuisse. He mentions, among other things, a project to create an emulator capable of reproducing the operation of ICON while mobilizing much fewer resources.

Business

-

-

PREV Nuclear energy companies weighed down on Wall Street after the rejection of an Amazon project
NEXT Port of Ouchy in Lausanne: start of the public inquiry