Regional climate models (RCMs) can provide far more precise information than general circulation models (GCMs). However, RCMs depend on GCM results or re-analysis products providing boundary conditions, especially for future climate scenarios. Meanwhile, the capacity of RCMs to reproduce precipitation is strongly connected to its performance on circulation and moisture transport simulations in the low troposphere, which is the key problem with RCMs at present. In the Regional Climate Model Inter-comparison Project for East Asia (RMIP Ⅲ), the results of ECHAM5/MPI-OM (the European Centre-Hamburg model version 5/Max Planck Institute Ocean Model, simplified as E5OM here) are used to drive RCMs for the past (1978-2000) climate simulation and future (2038-70) climate scenarios. Therefore, it is necessary to test E5OM's ability to represent atmospheric circulation, which defines the large-scale circulation for RCMs. Here, comparisons between the E5OM results and NCEP/NCAR (simplified as NCEP) re-analysis data in the low troposphere for the years 1978 to 2000 are performed. The results show that E5OM results can generally reproduce atmospheric circulations in the low troposphere. However, differences can be detected in East Asian summer and winter monsoon simulations. For summer, there is an anti-cyclone circulation for the difference of wind vector at 850 hPa in Southeast China, the Indo-China Peninsula, the South China Sea, and the northwestern Pacific. For winter, due to the weaker northwesterly wind in Northeast Asia, the northeasterly wind from the Indo-China Peninsula to Taiwan in E5OM extends northward with greater intensity than that in NCEP. These differences will have a considerable influence on the low level atmospheric circulation and water vapor transport as well as the location and intensity of the precipitation. Therefore, when E5OM results are to be used as initial and boundary conditions to drive RCMs, the differences between NCEP and E5OM should be considered.