Locating microseismic sources using migration-based deconvolution

Abstract

We have developed a method for finding microseismic hypocenters from data recorded by arrays of triaxial motion sensors. The method reconstructs the elastic time-series signatures for possible microseismic sources at any point in 3D space, using full-waveform migration of the recorded vector wavefield. The imaging condition for the migration is based on a semblance-weighted deconvolution between two or more reconstructed source signatures, requiring similarity and simultaneity of the reconstructed signatures. This imaging condition eliminates the need for an absolute timing of the data, gives optimum resolution for the location of the microseismic sources—better than correlation-based approaches—and ensures numerical stability by adapting to the signal and noise conditions of the data. Because the method eliminates time-consuming phase picking and traditional event association, it should be well suited for fully or semiautomated data processing. The method was tested by an application to field data acquired with arrays of three-component receivers in two deep wells. Nevertheless, the formulation is equally applicable to data acquired by a distribution of single three-component receivers, or local arrays of these, deployed at the surface or in one or several shallow wells.

Passive seismic multicomponent imaging using geometrical optics theory

“…If the velocity and density are given, the passive seismic imaging/inversion problem is linear; thus, following the geometrical optics theory, forward modeling operator (Aki and Richards, 2009), and adjoint operator, often called beamforming, diffraction stacking, or Kirchhoff operator (Johnson and Dudgeon, 1993;Borcea et al, 2011), can be modified and, with a far-field displacement approximation, it is similar to Haldorsen et al (2013) formulation:…”

“…The imaging procedure presented here is a modification of a commonly used method in earthquake seismology for wave-field propagation analysis, where if a location/backazimuth of the seismic source is known, the 3-C data are rotated to so called radial, transverse and vertical components. The method is closest to Fuller et al (2007) and Haldorsen et al (2013) approaches. Comparing to the latter, the method uses the same concepts except the following:…”

Weighted-elastic-wave interferometric imaging of microseismic source location

“…Grigoli et al (2013Grigoli et al ( , 2014 used SSA to develop an imaging method by stacking STA/LTA-transformed waveforms based on traveltime. Haldorsen et al (2012Haldorsen et al ( , 2013 proposed a migration-based deconvolution method for microseismic event location and suggested that semblance-weighted (energy-weighted relevance) imaging is superior to the energy correlation imaging proposed by Artman et al (2010). Miao et al (2012aMiao et al ( , 2012b proposed a tomographic imaging method to locate microseismic events automatically and invert source parameters in mining industry.…”

Simultaneous microseismic event localization and source mechanism determination

“…Grigoli et al (2013) propose stacking of short time average to long time average ratio (STA/LTA) traces computed from the raw waveform data. Haldorsen et al (2013) use projected traces in the frequency domain for stacking. Rentsch et al (2007) apply the stacking procedure to the energy of 3C data weighted with a Gaussian-beam-type factor.…”