The temporal and spatial evolution characteristics of induced seismicity in the Changning shale gas field based on dense array

Abstract

Abstract In this paper, we performed microseismicity detection and location using the deep learning method and obtained a high-precision earthquake catalog in the Changning gas field, which is one of the largest shale gas production demonstration areas in China. We found that the spatial and temporal characteristics of seismicity in the region are indicative of its correlation with industrial operations. The distribution of earthquakes at depth reflected variations in reservoir depth and provided valuable constraints on it. The horizontal layered distribution of earthquakes at depth is due to the formation of fracture surfaces from interconnected fractures near the reservoir during hydraulic fracturing (HF) operations, which clearly demonstrates how HF operations impact seismicity. We suggested that the horizontally layered distribution was driven by two fundamental mechanisms: reactivation of pre-existing faults during HF and injection of high-pressure fluids into the reservoir, leading to fracture creation. Several M L ≥4 earthquakes, which did not occur on well-defined seismogenic faults, may have been triggered by pore elastic coupling resulting from regional stress accumulation and fluid injection. Significantly, the M L 4.9 seismic sequence occurring at the basement indicates that fracking has reactivated and promoted pre-existing faults, highlighting the need for further investigation into potential seismic hazards in the region.

Keywords

• The Changning shale gas field
• Induced seismicity
• Seismicity detection and location
• Deep learning
• The seismogenic mechanism