Inversion Strategy - Models and Methods of Magnetotellurics - page 495

and Main Mongolo-Okhotsky faults. These channels connect the sediments with the
conductive mantle. Although the starting resistivities of the crust and upper mantle
in this model remain unchanged, the experimental and model

⊥ -curves converge,
and misfits of their low-frequency branches do not exceed 20%. At the same time,
the -curves are weakly affected by the conductive faults, and the misfits of these

curves remain as small as during the TE inversion in the absence of the faults.
The inferred model, whose misfits do not exceed the assumed uncertainty of
the apparent-resistivity synthesis, may be considered a final result of the bimodal
interpretation of MT soundings in the class of mantle diapir models. In assessing the
adequacy of this result, two questions should be answered: (1) Is the asthenosphere
asymmetry reliably diagnosed? and (2) Is the anomalous mantle reliably outlined?
Analysis of the model shows that we can answer both questions in the positive.
By smoothing the asthenosphere asymmetry and excluding the anomalous mantle,
we conspicuously increase the model misfits. In the course of several model tests
we conclude that the resistivity of the anomalous mantle is about 50-100 Ohm m.

Assuming that the decrease of mantle velocity is caused by partial melting, such
resistivity values indicate that the concentration of the liquid phase does not exceed
a few percent. This estimate is consistent with seismic estimates by Krylov (1981).
What is the geological nature of the vertical conductive channels? Evidently, in
the upper and middle crust they may be interpreted as fluid-saturated fault zones.
But one might suppose that in the lower crust and upper mantle these channels are
associated with deep roots of the faults, characterized by vertical fracturing that
conveys mantle fluids.
Summing up, we can say that the mantle diapir hypothesis is fairly consistent
with the MT data.

·

12.6.6 Test of the Asthenosphere-Upwarp Hypothesis

Figure 12.37 shows starting models that correspond to the early and recent versions
of the asthenospheric-upwarp hypothesis (cf. Figure 12.30). The main elements of
these models are: (1) inhomogeneous sediments which are differentiated using a pri-
ori geoelectric data, (2) a homogeneous lithosphere with a resistivity of 10 4 Ohm · m,

(3) an asthenospheric upwarp with a resistivity 100 Ohm · m and with its roof at a

depth of 50 km.
Experimental longitudinal and transverse curves of

⊥ were interpreted
using the same algorithm as above. The latter includes: (1) correction of longitudinal
curves through their vertical translation, bringing their low-frequency branches
into coincidence with the model curves of

and

sm calculated from starting model, (2)

⊥ -curves.
The final results of TM inversion are presented in Fig. 12.38 (the early version of
the hypothesis) and Fig. 12.39 (the recent version of the hypothesis). Both hypothe-
ses give low-frequency TM inversion misfits that go far beyond a 20% confidence
boundary. Evidently, the models with asthenospheric upwarp that comes in contact
with the Moho are in conflict with the MT data.

-curves, and (3) inversion of the

inversion of corrected the