Abstract: Based on the 25 models in the Phase 5 of Coupled Model Intercomparison Project (CMIP5) piControl simulation, the present study assessed the spatial distribution of the sea surface salinity (SSS) for the Eastern-Pacific (EP) and the Central Pacific (CP) El Niño in the tropical Pacific Ocean and explored the relationships among SSS, the sea surface temperature (SST) and precipitation. The results illustrate that: (1) Most of CMIP5 models can realistically reproduce the features of the two types of El Niño. The spatial skill scores of SST simulation show the best performance, followed by those of precipitation and SSS. The simulated horizontal distributions of SST and precipitation anomalies for the EP El Niño were better than those for the CP El Niño, but the results were opposite for SSS simulations. During the CP El Niño period, the positions of maximal SST, precipitation and SSS anomalies were clearly shifted to the west and slightly weaker compared with that during the EP El Niño. (2) The correlations among SST, precipitation and SSS in the CP El Niño were higher than that in the EP El Niño, which showed that SST directly affected precipitation, which then subsequently affected SSS significantly. In addition, SSS had an obvious feedback on SST. Compared with that in EP El Niño, the interaction between SST and SSS might be weaker because of the horizontal advection, the nonlocal effects and other related ocean physics. (3) Based on each SSS skill score simulated, CMIP5 models were divided into two groups. It was found that the maximal position of the variability with the low (high) scores models were located westward (eastward) in the equatorial Pacific, probably due to the position of the equatorial Pacific cold tongue. When the Pacific cold tongue extended westward, the warm SST anomalies moved westward remarkably during the El Niño events. This resulted in westward shift of precipitation and SSS anomalies at the same time. Meanwhile, the negative SSS anomalies extended southeastward in the low skill score models, which were probably attributed to the effects of the southeastern branch of the double ITCZ that caused more precipitation and freshen SSS. The SSS variability showed a close relationship with SST associated with the evolution of ENSO. Furthermore, the present study in the variation of simulated SSS spatial distribution and related physical fields can provide some information for improving climate prediction in the future.