Of Atlantic Meridional Overturning Circulation in the CMIP6 Project

• Published in: Deep-sea research. Part II, Topical studies in oceanography Vol. 206; p. 105193
• Main Authors: Gong, Xun, Liu, Hailong, Wang, Fuchang, Heuzé, Céline
• Format: Journal Article
• Language: English
• Published: Elsevier Ltd 01.12.2022
• Subjects: AMOC; Climate Research; CMIP6; Fysisk geografi; Historical experiment; Klimatforskning; Oceanografi, hydrologi, vattenresurser; Oceanography, Hydrology, Water Resources; Physical Geography; Trend and variability; CMIP6; Historical experiment; Trend and variability; AMOC
• ISSN: 0967-0645; 1879-0100; 1879-0100
• DOI: 10.1016/j.dsr2.2022.105193

The Atlantic Meridional Overturning Circulation (AMOC) upper-cell circulation is widely linked to global oceans and climate. Here, we focus on a statistical overview about the modelled AMOCs on the basis of the historical simulations in the 5th and 6th phase of the Coupled Model Intercomparison Project (CMIP5 and CMIP6), including the modelled AMOC strength, cell structure, long-term trend and the variabilities on interannual, decadal and multi-decadal scales. Our results show that the multi-model averaged AMOC mean state of CMIP5 is insignificantly different from the CMIP6 results, meanwhile the corresponding multi-model averaged AMOC variability is reduced from CMIP5 to CMIP6 results. Moreover, the CMIP6 multi-model averaged AMOC becomes further distinct from the mean state of Rapid Climate Change (RAPID) observations. Overall, 7 out of the 18 CMIP6 models have suggested AMOC strengthening, meanwhile 6 models have indicated declining trends in the AMOC, with the rest 5 models in the variabilities with insignificant trends. Overall, the CMIP6 results have suggested pronounced modelling discrepancies in revealing AMOC trends, distinct from the more commonly weakening trend of the AMOCs in the CMIP5 simulations. Moreover, the multi-model averaged AMOC variabilities are comparable between CMIP5 and CMIP6 simulations, on inter-annual, decadal and multi-decadal time scales, with the discrepancies remaining among models.