Abstract:The Automated Surface Observing System is currently replacing conventi onal observations in China. There are differences between the old and new measur ing systems. From a climatological viewpoint，it probably causes the errors when developing the homogeneous datasets and investigating climatic change，and anal yzing the extreme weather events. So it is necessary to compare the data from th e two systems. By means of computing the annual average differences of 21 weathe r stations in the Central China in 2005, and sorting analysis hourly data of 6 national climate stations according to cloud amount，wind speed, day and night, the differences in sitecharacteristics, solar radiation（IR cooling in night）, instrument system and observing practice are considered the main factors. It is showed that the annual average difference of temperature is almost within ±0.2 ℃. Among the above mentioned four factors, the instrument bias plays a part ro le，about 0.1℃ at most stations，the bias about local effect and observing prac tice are very little，but the bias of radiation is the largest.