Winters occurred at 09/10 and 10/11 over Eurasia region have the different tropical conditions. However, Eurasia suffers the extreme cold winter with the strong Arctic warimg in both cases. So, comparing two winters can be ideal for understanding cold winter mechanism. It is known that large-scale teleconnection pattern influences cold surge characteristics in Eurasia and East Asian winter monsoon strength. We examined each factor's role contributing cold winters in Eurasia: one from Arctic (sea ice loss) and the other from tropics (El Nino and La Nina). To address the effect of Arctic sea ice cover and tropical SST forcing upon the recent 09/10 and 10/11 cold winter in Eurasia, we have conducted 5 experiments with the NCAR CAM3. The CAM3 control run (CTR) consists of a 100-year integration with a repeating seasonal cycle of climatological SST and sea ice concentration for the period 1979-2010 obtained from Met Office HadISST. The 09/10 (10/11) sea ice simulation, hereafter EXP1 (EXP3), consists of a 100-year integration with a repeating seasonal cycle of Arctic sea ice concentration for the period Apr2009-Mar2010 (Apr2010-Mar2011). For the EXP1 and EXP3 with prescribed sea ice concentration, SSTs are set to those in the control experiments so as to isolate the impact of the sea ice changes. In addition, the 09/10 (10/11) sea ice and SST simulation, hereafter EXP2 (EXP4), consists of the same experimental design as EXP1 (EXP3), but SSTs are set to real observations in each year. In cases of EXP2 and EXP4, we can investigate Arctic warming response to switch-on tropical SST perturbations during Apr2009-March2010 (El Nino-like) and Apr2010-Mar2011 (La Nina-like), respectively. Through our experimental design, we can obtain the modeled response to projected sea ice loss in Arctic (especially, Kara-Barents and Chukchi Seas). Removing sea ice cover allows for a more transfer of heat flux to the atmosphere, which shows that directly positive anomalies of overlying surface air and lower tropospheric temperatures are generated. Moreover, a signal of Arctic amplification in surface based warming generates a upper tropospheric teleconnection, which cause the development of the anomalous Siberian High and the cold surge in East Asia. The ENSO-related wave is in phase with the climatological stationary wave over the North Pacific, and on a planetary scale has been shown to amplify the zonal wavenumber 1. Therefore, the amplification of this wave tend to weaken the stratospheric polar vortex, which is highly correlated with climate change over Arctic and Eurasia regions. Finally, our main interest in this research acquire the contribution of an extreme Arctic sea ice loss to the extreme cold winters by upper tropospheric teleconnection, and furthermore we understand the relationship between Arctic warming and tropical SST forcing like El Nino or La Nina.