Link to USGS Home

graphic rule

San Antonio, Texas Audio-magnetotelluric Data Processing

graphic rule

The Audio-magnetotelluric (AMT) method uses natural electromagnetic fields to investigate the electrical conductivity structure of the earth. Natural sources of AMT fields above about one Hertz are are from thunderstorms worldwide. The electrical discharges from thunder storms radiates powerful EM fields that propagate great distances. It is assumed that the EM fields at the surface of the earth behave almost like plane waves, with most of their energy reflected but with a small amount propagating vertically downward into the earth. The amplitude, phase, and directional relationships between electric (E) and magnetic (H or B) fields on the surface depend on the distribution of electrical conductivity in the subsurface (Vozoff, 1991).

For this study we used the four channel Geometrics EH-4 Audio-magnetotelluric (AMT) system with a controlled source to fill in the middle frequencies (450-4,500 Hz) where the natural signal strength is often below the detection limits of the instrument. This instrument is designed to use lightning or atmospheric disturbances (some times called sferics) as an energy source. In some cases, atmospheric disturbances were close enough to forgo the use of the controlled source. The instrument measures the electrical field in the X direction (Ex) and the magnetic field in the Y direction (Hy). It also measures the orthognal field set (Ey and Hx). 

THEORY

The AMT impedance tensor (Z) contains four complex components which relate the measured electric (E) and magnetic (H) fields:

                                                       Equation 1. Contact author for explanation.                                                    (1)

The impedances are computed from spectra collected in the field using a local H field reference.

                                         Equation 2. Contact author for explanation.                                      (2)

                                         Equation 3. Contact author for explanation.                                      (3)

                                         Equation 4. Contact author for explanation.                                      (4)

                                         Equation 5. Contact author for explanation.                                      (5)

 

where R* is the local reference field.

            From the impedances computing the apparent resistivities and phases uses the four components of the impedance tensor (Zxx, Zxy, Zyx, Zyy). Apparent resistivities r(f) and corresponding phase f(f) are computed using:

                                                                   Equation 6. Contact author for explanation.                                                                 (6)

                                                              Equation 7. Contact author for explanation.                                                            (7)

Bostick depth transforms are calculated for each frequency using:

                                                            Equation 8. Contact author for explanation.                                                         (8)

where the Hilbert transform relationship between the apparent resistivity r and its phase f in degrees translated into the first quadrant and clipped to the range  is used.

                                                                  Equation 9. Contact author for explanation.                                                        (9)

and                                                         Equation 10. Contact author for explanation.                                                           (10)

            The cross-sections use the rotationally invariant impedances derived from the full tensor. The arithmetic average is computed using the following formula:

                                                    Equation 11. Contact author for explanation.                                                      (11)

   A more complete explanation of theory, instruments, processing and interpretation can be found in Vozoff (1987, 1991) and Spies and Frischknecht (1987).

The assembly of the 4 individual Audio-magnetotelluric survey areas around San Antonio, Texas was completed in several steps.

GENERAL DATA PROCESSING STEPS FOR ELECTRICAL SECTIONS:

  1. Raw field data are processed in the field into three file types. X-type files are the cross power files, Y-type files are the time series data, and Z-type files are the impedance files. All data were processed through a two stage coherency cutoff to insure data quality. In this case completely coherent signals are equal to one.  In stage one, data with coherencies less that 0.3 were rejected. In stage two, data with coherencies less than 0.5 were rejected.
  2. Data bases were constructed from the original Audio-magnetotelluric survey impedance data based upon the areal extent of the soundings.
  3. MT parameters are computed, then phase curves are edited to insure that spurious data points are within their respective first and third quadrants.
  4. A spline function is applied to the raw Rho (resistivity or r) and Phase curves to smooth them and frequencies with anomalous values (generated by equipment malfunction or noise) edited or inactivated as necessary.
  5. The curves are then interpolated.
  6. Bostick (1977) inversions are calculated for the sites and lines are then plotted as electrical sections.
  7. The soundings are grouped into site sets and plotted as electrical sections. Then, after inspection are replotted using a minimum curvature algorithm.
  8. The original station data are always available for inspection along with the interpolated data.
  9. The sections are shifted upward or downward to their elevations correct for topography.

|| San Antonio AMT || Index Map || Survey Information || Electrical Sections || USGS - Eastern Minerals Team || USGS - Geology || USGS ||


Contact: Herbert A. Pierce
[an error occurred while processing this directive]