In this paper, for selected station of 8 clusters in East Asia (Park, 2017) more (less) warming periods than the wintertime mean warming of intra-seasonal fluctuation curves were taken and their means were computed. Long term trends and synoptic features of the mean temperature changes were examined. In most clusters, around the third of January there were less warming periods (LWP) than the mean wintertime warming. On the contrary, in February and the first and second of January there were more warming periods (MWP) than the winter mean or LWPs having a warming trend with statistical signicance. Time series of the daily Siberian High indices showed they had been weakening in February and being stagnant around late January. In most stations, the mean temperatures of MWP or LWP had large negative correlation coecients with the Siberian high intensity. is result explains the occurrences of MWPs in most clusters in February and LWPs in late January. In cluster B there were LWPs in early February due to the influence of the Aleutian Low which were strengthening in that periods. Cluster E showed different features without LWPs in late January. The cluster is considered to be affected by its plateau environment of West Yúnnán and the Tibet Plateau which prevent cold air of the lower atmosphere in Northern Asia flowing southward, and by the regional atmospheric circulation of 500hPa surface centered in this region.

In this study, the intra-seasonal fluctuation (ISF) of wintertime temperature change in East Asia was classified by a cluster analysis of complete linkage. A ISF of temperature change was defined as a difference of synthesized harmonics (1 to 36 harmonic) of daily temperature averaged for 30 years (1951~1980, 1981~2010). Eight clusters were gained from the ISF curves of 96 stations in East Asia. Regions of the cluster C, G and A1 seem to be affected by the Siberian High (SH) center, whereas the cluster A1, A2, D, B and F by the SH main pathways. Regions of the cluster E are apart from the SH main pathways and appear to be in the area of influence of other factors. Wintertime temperatures in Northwest China (clusters C, G) and Northeast China (cluster A1) were increased very largely. In most clusters, around late January there were less warming periods than the winter mean of the mean ISF of the clusters, before and after this time there were more warming periods than the winter mean.