Southern Ocean climate models show inaccuracies in cloud and heat data

A study published in Geophysical Research Letters has identified discrepancies in climate models regarding the Southern Ocean. Data collected by the Japanese research icebreaker R/V Shirase during the 64th Antarctic research expedition (JARE64) between December 2022 and March 2023 revealed that major atmospheric reanalysis datasets and climate models may be inaccurate.

Researchers found that the ERA5 and MERRA-2 datasets overstate the frequency of low clouds compared to actual shipboard measurements. Additionally, the ERA5, MERRA-2, and CAM-ATRAS models all underestimated downward longwave radiation, which is the heat emitted by clouds and the atmosphere back toward the surface.

The study, led by Kazutoshi Sato, Jun Inoue, and Hitoshi Matsui, suggests these errors stem from simulated clouds containing excessive ice, which reduces heat emission, and model temperatures that are colder than observed levels. Because the Southern Ocean significantly influences global climate, these inaccuracies can impact projections for sea ice changes, regional weather, and global warming.