Climate changes around Japan associated with upper troposphere and stratosphere responses of global warming during December-January-February were investigated using the storyline approach and the Coupled Model Intercomparison Project 5 dataset. Climate change was calculated by subtracting the 1959-1990 mean in historical simulations from the 2068-2099 mean in the Representative Concentration Pathways 8.5. Four storylines in plausible future climates were discussed by considering two remote indices representing tropical amplification (tropical upper tropospheric temperature changes) and stratospheric vortex strength. Stratosphere-troposphere connections in terms of zonal wind responses are most pronounced in the storyline of high tropical amplification with strong stratospheric vortex in which the subtropical jet shifts substantially northward. The multimodel mean displays more (less) precipitation in the northern (southern) parts of Japan, while the storyline of high (low) tropical amplification with strong (weak) stratospheric vortex shows increasing (decreasing) precipitation in most parts of Japan.Projected precipitation changes around Japan depend heavily on the storyline adopted and the degree of global warming in these two storylines. Alternatively, precipitation changes depend mostly on the overall strength of global warming with minor influences from storylines in the case of low (high) tropical amplification with strong (weak) stratospheric vortex.

Large-scale trends related to the precipitation in central Japan were investigated in the Coupled Model Intercomparison Project 5 dataset. In the Representative Concentration Pathways 4.5 projections, surface temperature over central Japan increases by 1 to 4 K during the 21st century in most models. Focusing on the trend of surface-air temperature and precipitation in central Japan, these models were categorized into two groups: The precipitation significantly increases in 11 models (group A). On the other hand, the trends of precipitation are small in the remaining 20 models (group B).Analyses of high-frequency components based on daily data revealed that the difference in the precipitation prediction is associated with that in the storm activity around Japan.Relatively enhanced meridional surface-air temperature gradient in the subpolar region may contribute to reinforced storm activity. On the other hand, the zonal surface Sato et al., Mutimodel Climate Prediction over Central Japan in Winter 2 pressure gradient associated with the Siberian high and the Aleutian low is not strengthened in group A. The climatological wintertime monsoon does not appear to contribute to the difference of precipitation trend in central Japan between the two groups. Moreover, the influence of any other planetary-scale variations is not implied.

Many studies have suggested that mean precipitation associated with the East Asian Summer Monsoon (EASM) will be increased by the ongoing global warming, but its quantitative projection by climate models has large variability, with some models suggesting even decreases. We investigate the inter-model variability of projected centennial changes of the EASM separately for Baiu over Japan, Meiyu over eastern China, and Changma over Korea by using monthly-mean model outputs provided by the Coupled Model Intercomparison Project Phase 5 (CMIP5) project. Results with the Representative Concentration Pathway 4.5 and 8.5 scenarios (RCP4.5 and RCP8.5) are consolidated by normalizing with the global-mean near-surface air temperature changes. For all the three EASM land regions, inter-model differences in the mean precipitation changes are positively correlated with the southerly moisture flux changes to the south of the regions. The correlation is highest in June among the June-to-August months, whose reason may be because precipitation in early summer relies on large-scale southerly moisture transport. These changes are localized and nearly independent among the three regions where Baiu, Meiyu and Changma occur. The low-level southerly change to the south of Japan, which affects the Baiu precipitation change, is positively correlated with upper-tropospheric meridional wind to its north; it further exhibits a stationary Rossby-wave feature associated with the Silk-Road teleconnection. This study suggests that future changes in the EASM mean precipitation depend on circulation changes and more-or-less localized.