Cloud and Water Vapor Feedbacks to the El Nino Warming: Are They Still Biased in CMIP5 Models?

Previous evaluations for model simulations of the cloud and water vapor feedbacks in response to El Nino warming have singled out two common biases in the Phase 3 of Coupled Model Intercomparison Project (CMIP3) models: an underestimate of the negative feedback from the shortwave radiation forcing of clouds (SWCRF) and an overestimate of the positive feedback from the greenhouse effect of water vapor. Here we check whether these two biases are alleviated in the Phase 5 of Coupled Model Intercomparison Project (CMIP5) models. While improvements are found, particularly in the simulation of the negative feedback from SWCRF, the biases in the simulation of these two feedbacks remain prevalent and significant. Considering that the bias from SWCRF feedback is closely related to other biases from feedbacks such as precipitation, Walker circulation and longwave radiation forcing of clouds (LWCRF), the CMIP5 models are divided into two categories by their skills of simulating the SWCRF feedback. As the SWCRF feedback is generally simulated better in the models that have a higher skill in representing the zonal extent of the equatorial cold tongue, the excessive cold tongue is suggested as a key factor that is behind the persistence of the feedback biases. Owing to the air-sea interaction, a cold bias of SST may strongly affect the complicated feedbacks in climate models, which in turn causes a bias in the representation of the SST.