Improving AI weather prediction models using global mass and energy conservation schemes

Yingkai Sha †, John S. Schreck †, William Chapman ‡, David John Gagne II † · 2025

Read Improving AI weather prediction models using global mass and energy conservation schemes by Yingkai Sha †, John S. Schreck †, William Chapman ‡, David John Gagne II † (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2501-05648.txt.

Description

Yingkai Sha, John Schreck, William Chapman and David Gagne wrote a 2025 paper on AI weather forecasts. AI models are strong at medium-range forecasts but can break basic physics rules. Their outputs may fail to conserve mass or energy. The team built add-on schemes to enforce those rules.

The schemes guard dry air mass, water balance and total energy for the whole globe. They are built as modules that fit many model types. Tests used one named AI model set on a coarse grid. Forecasts kept the physics rules better and scored higher, more so at longer lead times. Rain forecasts gained most, with less false drizzle.

Catalog record

CREATORYingkai Sha †, John S. Schreck †, William Chapman ‡, David John Gagne II †
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE88 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2501-05648

Browse the full library · About the Tower of Babel