The accuracy of daily mean 2 meter air temperatures from five reanalyses are assessed against in-situ observations from Automatic Weather Stations in East Antarctica for 2005 to 2008. The five reanalyses all explain more than 70% of the average variance, and have annual root mean square errors （RMSE） between 3.4 and 6.9℃. The NOAA reanalyses, NCEP-1, NCEP-2 and 20CRv2, have cool biases of 2.5, 1.4 and 1.5℃, respectively. The ERA Interim and JCDAS reanalyses have warm biases of 1.7 and 2.0℃. All reanalyses generally perform better in the austral spring and worse in winter and autumn. They also show the best performance at an inland plateau site at 2800 m elevation, but are worst at Dome A, the summit of the East Antarctic ice sheet. In general, ERA Interim is superior to the other reanalyses, probably because of its 4D assimilation scheme. The three NOAA reanalyses perform worst; Their assimilation scheme is more constrained by limited observations and 20CRy2 has less input data, assimilating only surface pressure observations. Despite deficiencies and limitations, the reanalyses are still powerful tools for climate studies in the Antarctic region. However, more in-situ observations are required, especially from the vast interior of Antarctica.
