Abstract: Since a quantitative study of the cloud radar, which belongs to State Key Laboratory of Severe Weather, Chinese Academy of Meteorological Sciences for measuring light precipitation has not been carried out so far, in this paper the radar observation data at Sanshui in Guangdong in 2013 are analyzed to determine the radar reliability and error. Examining the connection and difference between the cloud radar observation in a certain height and the calculation using disdrometer data on the ground, a methods is proposed to compare the velocity and spectral width calculated from raindrop size distribution with those observed by the cloud radar observation in a certain height for light precipitation. The source of the radar observation errors can be analyzed, and the quality of the data can be determined. The results showed that the radial velocity values which are affected by updraft have notable difference between those observed and calculated although both change roughly with a similar trend over time. The average error caused by updraft is 0.89 m·s-1. The spectral width from cloud radar and disdrometer data were corresponded well; its average error is 0.15 m·s-1.