Understanding Intermodel Diversity of CMIP5 Climate Models in Simulating East Asian Marginal Sea Surface Temperature in the Near Future (2020-2049)

Using 14 Coupled Model Intercomparison Projections Phase 5 (CMIP5) climate models, we examined the intermodel diversity when simulating East Asian Marginal Seas (EAMSs) sea surface temperature (SST) in the near future period (2020-2049) under four different Representative Concentration Pathway runs. We classified two groups for the CMIP5 climate models: for models that simulate SSTs in the EAMS that are higher (H_EAMS) and lower (L_EAMS) than the ensemble mean, respectively. Results show that compared to L_EAMS, H_EAMS tends to simulate weaker westerlies in the western-to-central North Pacific, together with a weaker Aleutian Low intensity, which causes higher EAMS SSTs through a reduction in latent heat flux. Furthermore, H_EAMS is characterized by cooler SST, less precipitation, and stronger trade winds in the central-to-eastern tropical Pacific than in L_EAMS. We argued that the intermodel diversity of simulated tropical Pacific SST is associated with the diversity of EAMS SST, which is related to atmospheric teleconnections from the tropics to the western-to-central North Pacific. It is also found that the bifurcation latitude of the North Equatorial Current is lower in H_EAMS than in L_EAMS, which is associated with the difference of tropical Pacific mean state between H_EAMS and L_EAMS. A lower bifurcation latitude transports more warm water into the EAMS, resulting in warmer SSTs in the H_EAMS than in the L_EAMS. These results show the importance of correctly simulating the tropical Pacific mean state to reduce the uncertainty in EAMS SST during the near-future period.We acknowledge the World Climate Research Programme's Working Group on Coupled Modelling and climate modeling groups for providing the model output. For CMIP, the U.S. Department of Energy's Program for Climate Model Diagnosis and Intercomparison provides coordinating support and leads the development of software infrastructure in partnership with the Global Organization for Earth System Science Portals (https://cmip.llnl.gov/cmip5/data_portal.html).SST data ERSST V5 were taken from https://www.ncdc.noaa.gov/dataaccess/marineocean-data/extended-reconstructed-sea-surface-temperature-ersst-v5.This work was supported by National Research Foundation Grant NRF-2018R1A5A1024958