Binning Method for Mapping Irregularly Distributed Continuous Resistivity Profiling Data onto a Regular Grid for 3d Inversion
Tian Xu and John A. Dunbar
Baylor University, Waco, Texas
Continuous resistivity profiling (CRP) is an emerging direct current (DC) method that is increasingly used in shallow fresh water and marine environments to support hydrogeological and near surface geophysics studies. CRP is popular, because it allows efficient data acquisition. However, it is generally not possible to precisely control the spacing and origination of profiles during CRP data acquisition, resulting in irregularly distributed data. For 2D profiles, the irregular distribution is handled by extracting relatively straight segments for inversion and using either interpolation or irregular element widths in the inversion process to account for uneven data spacing along the profiles. 3D Inversion of CRP data sets is more challenging, because most commercial inversion codes are applicable only to parallel or sub-parallel survey lines, requiring data on a regular grid. In this paper, we develop a technique for binning irregularly distributed CRP data onto a regular grid, suitable for most 3D inversion codes. The method uses the GPS track line of the survey vessels as the assumed path of the electrode array to reconstruct the electrode locations through time during the survey. Individual apparent resistivity readings are assigned geographic locations at the center of each four-electrode array configuration. Apparent resistivity of like-array configurations within a spatial bin around a grid point is then interpolated to the center of the bin. This is repeated for each grid point and each array configuration. Finally, a set of binned readings is generated with each array configuration centered at the regular grid points and written to a standard input file type for a commercial 3D inversion code. We evaluate the method on synthetic data sets as well as field CRP data sets collected from two water reservoirs on Maui, Hawaii. Numerical results of the synthetic data inversion show that 20% and 10% model misfit between a “control group” and binned data can be achieved with data densities of 5 and 15 measurements in each bin, respectively. Also, the 3D inverted resistivity volumes of the binned field CRP data sets show good agreement with independently collected 2D profiles, showing the same subsurface features.
AAPG Search and Discovery Article #90182©2013 AAPG/SEG Student Expo, Houston, Texas, September 16-17, 2013