Information Technology Reference
In-Depth Information
On this way we can localize a three-dimensional structure and roughly estimate its
depth and strike.
2. Let the quasi-homogeneous geoelectric medium be covered by the areal mag-
netotelluric survey. In that case the three-dimensional interpretation can be reduced
to the quasi-one-dimensional inversion of apparent resistivities
eff
,
brd
or
rms
computed from one of the scalar invariants
Z
eff
,
Z
brd
or
Z
rms
of the impedance ten-
sor. Such an approach is referred to as
scalarization
of the magnetelluric inversion.
3. In the case of the locally inhomogeneous geoelectric medium the three-
dimensional interpretation can be accomplished by successive partial inversions
informatively connected with each other through starting models and stabilizing
functionals. Proceeding from the experience of two-dimensional interpretation, we
assume that the most efficient is the following succession of inversions: (1) inversion
of the magnetovariational response functions (tipper, horizontal magnetic tensor) -
being free of near-surface static distortions, this inversion yields sufficiently reliable
image of the geoelectric medium (particularly of its conductive zones), (2) inversion
of the magnetotelluric phase response functions (phases of impedance tensor, phase
tensor) - this inversion is also free of near-surface static distortions, it controls and
edits the results of the magnevariational inversion, and (3) inversion of the mag-
netotelluric amplitude response functions (apparent resistivities) - this inversion is
subjected to strong near-surface static distortions, but it can give additional infor-
mation on upper layers of the sedimentary cover and galvanic connections that exist
in the sedimentary cover and consolidated Earth's crust. Such a three-stage interpre-
tation can be significantly simplified by scalarization of the partial inversion (scalar
invariants of tipper, horizontal magnetic tensor, impedance tensor, phase tensor are
in use).
4. Method of the successive partial inversion provides necessary stability of the
three-dimensional interpretation under condition of sufficiently simple structure of
geoelectric medium (single simple-shaped conductive inclusions, absence of strong
horizontal and vertical ruptures and dislocations, etc). Studying complex-structured
media, we should divide the three-dimensional interpretation into two stages. At the
first stage the smoothing (or quasi-one dimensional) inversion is conducted and a
simplified schematic model of the investigated medium is constructed. At the sec-
ond stage this model is included into the blocky starting model as the fixed first
approximation. The obtained basis is correlated with the a priori information and
with the results of the qualitative analysis of MV-MT data. On this way it is supple-
mented with a set of local hypothetical inhomogeneities, whose free parameters are
determined by the method of successive partial inversions. Hypotheses, which yield
the minimal model misfit, are accepted as the most plausible. Thus, the smoothing
(or quasi-one-dimensional) inversion is combined with the hypotheses tests. This
complex provides stable sufficiently detailed three-dimensional interpretation per-
formed in the class of models with a small number of free parameters.