IMAGING INDUCED AND NATURAL FRACTURES USING FULL WAVEFORM MICROSEISMIC DATA

The development of unconventional reservoirs relies on hydraulic fracturing, a technique that stimulates fracture network and enhances permeability by injecting high-pressure fluids into production formations. Microseismic monitoring is currently the key technique that can provide real-time information about the geometry of stimulated fracture network.

The important task in microseismic monitoring is locating induced fractures (microseismic events). Most of the standard automated location routines are currently based on the P- and S-wave phases information using the Geiger algorithm adopted from earthquake seismology. The location is performed by minimizing the traveltime residuals between calculated and observed (picked) arrival times of body waves (mainly the first P and S onsets), using iterative inversion algorithms. The increasing interest on microseismic monitoring operations, particularly for oil and gas reservoirs applications, pushed the development of new location methods based on migration techniques used in reflection seismology. Migration-based location methods do not require phase picking nor phase identification and directly exploit the waveform information contained in full waveform seismograms.

In this talk, I will show the development of imaging induced and natural fractures using full waveform information. In the first part, I introduce a distributed microseismic fracture imaging algorithm. The key innovation is that microseismic imaging cross correlate each receiver wavefield in both space and time from virtual sources. The method is capable of producing high-resolution images of multiple source locations, even when the signal-to-noise ratio (SNR) is low. In the second part, I present a workflow of using microseismic data for imaging natural fractures. Unlike conventional fracture imaging that often adopts reflection-type seismic imaging with known source, my approach relies on the data without the need for passive source information. I use two types of information from passive data: transmitted waves and scattering (coda) waves. Numerical examples include surface and borehole acquisition scenarios. Finally, I will discuss the prospective research in full waveform inversion of event location and velocity model.