Abstract: The coupled evolution of water and rock is one of the core problems in major water hazards prevention. In the whole-space geoelectric field detection, the natural electric field can be used to track water flow direction and its response is more advanced in time domain. As long as these advantages be combined to the spatial characterization capabilities of the primary electric field and the induced electric field for physical properties, the temporal and spatial accuracy of geoelectric field detection can be improved and some new breakthroughs in water-rock coupling evolution monitoring and emergency warning can be made. In the process of unsaturated seepage and its mutation, such as fractured rock mass seepage and broken rock mass seepage, the natural electric field is controlled by the streaming potential and is closely related to the evolution of water-rock coupling, including the water-passing area evolution. Therefore, the abnormal response of the natural electric field is defined as a near-source effect, induced by water approaching to the measuring point which appears in the whole-space monitoring. Its main features are as follows:(1) without water-passing event, the self-potential rises as the water approaches the measuring point, and decreases as the water moves away.(2) With water-passing event, the self-potential increases as water approaches the measuring point, falls as water covers the measuring point with the phase mutation(compared to other measuring points),and then resumes the upward trend, but finally decreases again as water moves away.(3) Due to the heterogeneity of fracture, only part of the measuring points meet water-passing events on the same survey line, which can be identified according to the self-potential falling characteristics and phase mutation, combining to the low resistance evolution characteristics in the apparent resistivity profile or resistivity profile. With the parallel detection technology of self-potential method and DC resistivity method, both the groundwater seepage state and the water-passing events can be effectively identified. Then the self-potential data and the DC resistivity data can be analyzed jointly. Thus, the water channels and water-rich areas evolution characteristics can be finely interpretated and the water-rock coupling evolution can be predicted. In this way, some new methods can be provided for major water hazards prevention and control in the area of mines, tunnels, subways and other underground projects.