Southern Ocean Solar Reflection Biases in CMIP6 Models Linked to Cloud Phase and Vertical Structure Representations
International audience Over the Southern Ocean (SO, 40°S–70°S), climate models have consistently underestimated solar reflection. Here we evaluate the relationship between cloud profiles, cloud phase and radiation over the SO in Coupled Model Intercomparison Project Phase 6 (CMIP6) models against Cl...
Published in: | Geophysical Research Letters |
---|---|
Main Authors: | , , , |
Other Authors: | , , , , , , , , , , , , , , |
Format: | Article in Journal/Newspaper |
Language: | English |
Published: |
HAL CCSD
2022
|
Subjects: | |
Online Access: | https://hal-insu.archives-ouvertes.fr/insu-03872791 https://hal-insu.archives-ouvertes.fr/insu-03872791/document https://hal-insu.archives-ouvertes.fr/insu-03872791/file/Geophysical%20Research%20Letters%20-%202022%20-%20Cesana%20-%20Southern%20Ocean%20Solar%20Reflection%20Biases%20in%20CMIP6%20Models%20Linked%20to%20Cloud%20Phase.pdf https://doi.org/10.1029/2022GL099777 |
Summary: | International audience Over the Southern Ocean (SO, 40°S–70°S), climate models have consistently underestimated solar reflection. Here we evaluate the relationship between cloud profiles, cloud phase and radiation over the SO in Coupled Model Intercomparison Project Phase 6 (CMIP6) models against Clouds and the Earth's Radiant Energy System and Cloud-Aerosol Lidar and Infrared Pathfinder Satellite Observations. We find that the lack of solar reflection is slightly improved in CMIP6 models compared to CMIP5's, attributable to a better representation of cloud fraction and phase. We show that clouds have a different vertical structure and radiative effect south and north of where the 0°C isotherm meets the surface (∼55°S). Although the models capture the greater vertical extent of clouds south of 55°S, they fail to reproduce the observed increase in solar reflection, which we pinpoint to cloud phase biases. Increasing CMIP6 supercooled liquid cloud opacity should help reduce their persistent shortwave biases. |
---|