In area A, more yellow patterns appear in the corrected image than that in the original one. Yonezawa and Takeuchi (2003) have shown that the maximum vertical displacement derived by InSAR and leveling measurement was about 3–4 cm in this area from the observation of 23 July 1995 to 15 March 1999. In our corrected image, it is about 1 cm from 8 February 1999 to 2 August 1999, which is consistent with Yonezawa and Takeuchi (2003). So the subsidence signal must have been partly counteracted by WSD. In area B,no subsidence has been measured; however, there is about one fringe in the interferogram. After wet delay correction, part of the fringe has been removed. The displacements in LOS at S0225, for instance, are 2.53 and 0.61 cm before and after correction, respectively. The GPS displacement at this station is 0.76 cm. So the accuracy has been increased about 1.8 cm at S0225. The coherence is less than 0.2 in area C, so we conclude that the red patterns are caused by the noise of decorrelation, instead of wet delay effect. Therefore, they cannot be removed by the correction of wet delay.