A 600-year integration performed with the Bergen Climate Model and National Centers for Environmental Prediction/National Center for Atmospheric Research （NCEP/NCAR） reanalysis data were used to investigate the impact of strong tropical volcanic eruptions on the East Asian summer monsoon （EASM） and EASM rainfall.Both the simulation and NCEP/NCAR reanalysis data show a weakening of the EASM in strong eruption years.The model simulation suggests that North and South China experience droughts and the Yangtze-Huaihe River Valley experiences floods during eruption years.In response to strong tropical volcanic eruptions,the meridional air temperature gradient in the upper troposphere is enhanced,which leads to a southward shift and an increase of the East Asian subtropical westerly jet stream （EASWJ）.At the same time,the land-sea thermal contrast between the Asian land mass and Northwest Pacific Ocean is weakened.The southward shift and increase of the EASWJ and reduction of the land-sea thermal contrast all contribute to a weakening of the EASM and EASM rainfall anomaly.
The Twentieth Century Reanalysis （20thCR） dataset released in 2010 covers the period 1871-2010 and is one of the longest reanalysis datasets available worldwide. Using ERA-40, ERA-Interim and NCEP-NCAR reanalysis data, as well as HadSLP2 data and meteorological temperature records over eastern China, the performances of 20thCR in reproducing the spatial patterns and temporal variability of the East Asian winter monsoon （EAWM） are examined. Results indicate that 20thCR data： （1） can accurately reproduce the most typical configuration patterns of all sub-factors differences in the main circulation fields over East Asia involved in the EAWM system, albeit with some in comparison to ERA-40 reanalysis data; （2） is reliable and stable in describing the temporal variability of EAWM since the 1930s; and （3） can describe the high-frequency variability of EAWM better than the low-frequency fluctuations, especially in the early period. In conclusion, caution should be taken when using 20thCR data to study interdecadal variabilities or long-term trends of the EAWM, especially prior to the 1930s.
