Interferometric Synthetic Aperture Radar (InSAR) coherence is important not only in determining measurement quality but also for extracting thematic information about objects on the ground in combination with backscattering coefficient. The decorrelation of repeat-pass InSAR caused by soil moisture change has received little attention in comparison with other sources of decorrelation. In this paper, we use ENVISAT ASAR data and laboratory experiments to analyze the repeat-pass InSAR coherence loss results due to soil moisture change. C-band ASAR data has high coherence over bare soil and grassland areas, which indicates that these two types of land cover are good choices for the analysis of InSAR coherence loss due to soil moisture change. In addition, spaceborne SAR with short revisit capability, has great potential for this specific application, even for agricultural fields. We conducted further analysis of the soil-sample laboratory data acquired in an anechoic chamber because of its controllable environment and the ability to exclude other sources of decorrelation. We found that the lower frequency range, 2-2.5 GHz, has the highest coherence and is the most insensitive to the initial soil moisture value. This indicates that the S band is more advantageous than the C band when using InSAR coherence to detect soil moisture change. This is true at least with respect to the S band's high coherence level and insensitivity to initial soil moisture values.