Abstract: On the basis of the 1:250000 DEM(digital elevation model) data and station information of Hefei, Wuhan, Yichang, Changde and Changsha weather radars and radar beam pattern (or power density function), the beam blockage coefficients is calculated. A beam bottom clearance of 150 m or more and an occultation of less than 60% are used to define a hybrid elevation angle which is used for calculating isobeam height based on radar altimetry. After iso-beam heights from individual radars are remapped onto the Cartesian grid, they are combined to produce a mosaic iso-beam height image. 7 radar pairs are composed, which have common covering area at 4　km height. The lowest scanning elevation angles of the 5 radars are analyzed using the radar raw data on 17-19，July 2004, the results show that the mean lowest elevation angles of Hefei and Wuhan radars are respectively 0.19°and 0.07° lower than 0.5°, the regulation of VCP21, and the others have no great difference. And then, after distance attenuation，and terrain obstruct and clutter are eliminated, the reflectivity difference of radar pair on the equidistant line are analyzed when they simultaneously observe. Because of different height above sea level of radar antenna, the objects' height at the same slant range with same elevation angle are dissimilar between radar pairs. In order that two contrast reflectivities come from same object more possible, the radar volume scan data are remapped onto 3D Cartesian grid, and for abating the attenuation influence, the reflectivities on equidistant line are selected, and for mitigating beam blockage and clutter, when the mean reflectivity difference between radar pair at same distance from two radars are calculated, only these points are considered whose height is higher than iso-beam height, viz. which are not blocked completely. Whereby, the following results can be approached: (1) Lowest scanning elevation angles can be used to examine the calibration of radar elevation angle. (2) Vertical cross sections of reflectivity on equidistant lines for radar pairs can be used to analyze the deviation of echo's position and intensity from radar-pair. Echo's vertical structures appear large differences on equidistant lines when Changde radar simultaneously observes with other adjacent 3 radars, its echo height is obviously lower. Nevertheless, the others' horizontal and vertical echoes are consistent well. (3) Reflectivity on equidistant line at same height from radar-pairs can be used to examine the calibration of radar azimuth. Because of undemanding space -time synchronization as well as the effects of atmospheric refraction and distance attenuation, the reflectivity variation trends of paired radars on the equidistant line at same height are incompletely consistent, and the mean and standard deviation of reflectivity differences between paired radars on equidistant line at same height change along with time, and so does correlation coefficient of reflectivities of radar pairs. (4) Wuhan radar's echo intensities are 2.4 dBz weaker than that of Hefei radar, 4.6　dBz lower than Yichang, 2.4 dBz lower than Changsha when it observes simultaneously with the others around. Yichang's is averagely 4.6 dBz stronger than Wuhan and 2.1 dBz stronger than Changsha. The variations are small between Wuhan and Hefei, and Wuhan and Changsha radars. Assumed the criterion based on Hefei and Changsha radars, Wuhan is about -2.4 dB system observation error, and Yichang about 2.4 dB.