Abstract: Based on the sea surface temperature (SST) data of the Hadley Center, the temperature data of 160 stations in China, and the reanalysis data of National Centers for Environmental Prediction/National Center for Atmospheric Research (NCEP/NCAR), the interannual variability of the North Atlantic SST anomaly (SSTA) in autumn and its association with early winter temperature in China are studied using a variety of statistical methods, such as empirical orthogonal function (EOF) and correlation analysis. The results show that the first mode of North Atlantic SSTA EOF in autumn is the North Atlantic horseshoe (NAH) pattern with negative (positive) SSTAs in southeastern Newfoundland and positive (negative) SSTAs in the subpolar, subtropical, and eastern North Atlantic, which revealed a variance of 20.5%. A significant positive correlation was observed between NAH SSTA in autumn and early winter in most parts of China. Further analysis shows that NAH SSTA in autumn can persist until early winter. When NAH SST is in a positive (negative) phase in autumn, the heating (cooling) of NAH SSTA over the subpolar and subtropical regions in early winter can cause divergence (convergence) in the upper troposphere, and two Rossby wave trains are stimulated in the north and south branches. The north branch of the wave train propagates northeastward from the North Atlantic subpolar to the vicinity of the Barents Sea and then propagates southeast along Siberia to China. The south branch of the wave train propagates eastward from the subtropical North Atlantic to China. Convergence (divergence) occurs in the upper troposphere over China under the influence of the south and north branches of wave trains. Through the sinking (rising) movement, the total cloud cover in the above area is reduced (increased), the shortwave radiation reaching the surface has increased (reduced), and the longwave radiation transmitted from the surface to the lower atmosphere has increased (reduced). Furthermore, the temperature in most areas of China is abnormally high (low) under diabatic heating. The effects of the NAH SSTA on atmospheric circulation, radiative forcing, and air temperature in early winter were simulated using NCAR Community Atmosphere Model version 5.3 (CAM5.3). The simulation results were consistent with the statistical analysis results of the observational data. This further confirms that the SST mode can affect the atmospheric circulation anomalies in East Asia by stimulating the teleconnected wave train, thereby affecting the interannual variation of air temperature anomalies in China through the effect of diabatic heating.