Microseismic Source Inversion in Anisotropic Media
Microseismic source mechanisms are of interest in hydraulic stimulation
because of what they may reveal about the induced or enhanced fracture network.
Since rocks are in general anisotropic, particularly shales, it is of interest
to consider what impact anisotropy may have on microseismic sources and their
inversion. The strongest and most prevalent type of anisotropy in sedimentary
rocks is transverse isotropy with a symmetry axis normal to bedding or vertical
(so-called VTI). While the theory used is valid for general anisotropy we
consider presently only layered VTI models.
An earthquake may be represented as a displacement discontinuity across
a plane, composed of a double couple (DC) or shear slip component and an
opening (or closing) component. While this type of source description seems
physically intuitive for microseismic sources generated by hydraulic
stimulation, it is an incomplete description of a moment tensor; the addition
of a pure explosion (or implosion) component is necessary. We are calling such
a three part Explosion + Opening + Slip source an “EOS” source,
from which the moment tensor can be constructed. A theory has been developed to
decompose any moment tensor M into an EOS source, given the anisotropic medium
local to the source (Chapman and Leaney, 2011).
Previously we looked at the influence of anisotropy at the event
location on the source including radiation patterns (Leaney and Chapman, 2010)
for different EOS sources and found that if displacement was not in the
symmetry axis of the medium, then the presence of anisotropy would produce
false non-DC components and the focal plane solutions would be significantly
distorted from their true orientations. In the present work we recover such an
EOS source using three steps: 1. a ray-based, frequency domain, linear
inversion for the moment tensor and source function; 2. a nonlinear inversion
to obtain the scalar moment from the estimated source function; 3. a new
decomposition of the moment tensor given the anisotropic medium local to the
source (Chapman and Leaney, 2011). We illustrate this new anisotropic moment
tensor inversion and decomposition on synthetic data. Interpretation strategies
that these new analysis tools make possible are being explored.