Abstract: The Sijiao village is the typical karst collapse area, in which karst collapse is uncertain, subtle and unexpected. Based on the characteristics of karst collapse, this paper selects the high density resistivity method, the active source surface wave method and the microtremor method to make a comprehensive study of this kind of karst collapse. The results indicate that the high density resistivity method can well divide the interface of rock and soil layer and the karst development zones, but the resolution is relatively low; the active source surface wave method can well divide the interface of rock and soil layer, and has a higher resolution, but the detection depth is relatively shallow; and the microtremor can accurately determine the karst collapse position but the anti-interference ability is relatively weak. The combination of the high density resistivity method, the active source surface wave method and the microtremor method can effectively identify the structure of the rock and soil layer in the collapse area, determine the location of the karst development zones, and predict the collapse risk areas, the collapse latent area and the relatively stable area. These work can determine the specific scope of the follow-up hazards treatment work and provides a scientific basis for the management of construction design. In addition, it is also found that the formation of collapse is related to the years of well pumping according to the abnormal characteristics of the three geophysical prospecting methods mentioned above.