Official websites use .gov
A .gov website belongs to an official government organization in the United States.

Secure .gov websites use HTTPS
A lock ( ) or https:// means you’ve safely connected to the .gov website. Share sensitive information only on official, secure websites.

Ranking the Performance of Climate Models for Summertime Sea Ice Simulation

img-NOAA-arctic-ice

The Arctic has experienced substantial warming over the past 40 years with dramatic sea ice retreat and land ice melt during the summer. The sea ice melt is associated with a coupling, or linkage, between atmospheric warming and moistening in the summer and atmospheric circulation variability. Combined, these factors contribute to significant sea ice melt on a year-to-year as well as interdecadal basis. Researchers funded by both the CPO Climate Variability & Predictability (CVP) program and the Modeling, Analysis, Predictions and Projections (MAPP) program sought to answer the question: how well can climate models replicable the observed impacts of this summertime atmosphere-sea ice coupling on September sea ice? The researchers created a climate model metric focused on the statistical relationship between atmospheric temperature and humidity with sea ice in order to rank the performance of over 30 CMIP5 and CMIP6 climate models in reproducing the observed atmosphere-sea ice relationship. Their ranking and subsequent analysis, chronicled in the journal Climate Dynamics, reveals that even the highest ranked models seem to underestimate sea ice sensitivity to atmospheric forcing. In other words, the models have limitations in reproducing the full strength of the observed atmosphere-sea ice connection. The study authors offer several explanations of where exactly the limitations in the model may be occurring. For example, how the model simulates clouds in the Arctic. Overall, the models in both the highest and lowest ranking groups allow researchers to identify what works and what doesn’t work well in reproducing the observed atmosphere-sea ice connection. This study is also useful for improving seasonal prediction of Arctic sea ice, as study findings may also be applicable to the models used for prediction.

Read the study »

Image Credit: Collection of Dr. Pablo Clemente-Colon, Chief Scientist National Ice Center.

Scroll to Top