Biases and improvements of the boreal winter–spring equatorial undercurrent in the Indian Ocean in the CMIP5 and CMIP6 models

We assessed the performance of state-of-the-art coupled models in reproducing equatorial undercurrent (EUC) Indian Ocean based on outputs Coupled Model Intercomparison Project Phase 6 (CMIP6) and compared with 5 (CMIP5) models. Our results showed that CMIP6 reproduced boreal winter–spring EUC more realistically than CMIP5 models, although both generations underestimated strength observations. This underestimation can be attributed to excessively strong westward-extended cold tongue Pacific. In CMIP a stronger winter-mean favors zonal sea surface temperature gradient, which forces easterly wind bias over western This, turn, contributes an acceleration Walker circulation. enhanced circulation Indo-Pacific directly causes lower level westerly drives shallow west–deep east thermocline tilt bias, ultimately leading weak via wind-induced processes. Compared overall improvement traced back degree bias. suggest efforts should made reduce mean-state Pacific further improve simulation projection atmospheric oceanic circulations Ocean.