Rocky Mountain Section - 64th Annual Meeting (9–11 May 2012)

Paper No. 3
Presentation Time: 9:00 AM

SOFTWARE DEVELOPMENT FOR A 3D GRAVITY INVERSION AND APPLICATION TO STUDY OF THE BORDER RANGES FAULT SYSTEM, SOUTH-CENTRAL ALASKA


CARDENAS, Rolando, Program in Computational Science, University of Texas at El Paso, 500 W. University Avenue, El Paso, TX 79968, DOSER, Diane, Univ Texas - El Paso, El Paso, TX 79968 and BAKER, Mark, Geomedia Reasearch and development, 6040 Strahan, El Paso, TX 79932, roncardenas3@gmail.com

The Border Ranges Fault (BRFS) system bounds the Cook Inlet and Susitna Basins, an important petroleum province within south-central Alaska. An initial research goal is to test several plausible models of structure along the Border Ranges fault system by developing a novel, 3-D inversion software package. The inversion utilizes gravity data constrained with geophysical, borehole, and surface geological information. The novel inversion approach involves directly modeling known geology, initially free-air corrected data, and revising a priori uncertainties on the geologic model to allow comparisons to alternative interpretations.

Previous modeling of the BRFS using geophysical data has been limited due to the complexity of local geology and structure, both of shallow crustal features and the deeper subduction zone. Without a basic understanding of the geometry of the BRFS, its role in the formation and petroleum generation processes of the upper Cook Inlet and Susitna Basins is poorly understood. This technique to evaluate 3-D structure in regions of highly complex geology can be applied in other studies of energy resources.

The effectiveness of the software is in the user’s capability of including a priori constraints and gravity observation uncertainties to reduce the number of plausible numerical models. While similar inversion software packages utilize rectangular, hexahedronal, and triangular prisms, ours is unique in the application of semi-infinite vertical line elements to deal with the complex geometry of surface and sub-surface structures. In addition, the algorithm uses an initial distance test from a gravity station to the area of study to eliminate insignificant computations. Interval computations, Kalman filtering, and covariance between forward modeling, process noise, and digital elevation model uncertainties are considered in the model solution. On-going development will include a user interface with grid visualization capability and the means to observe a block of earth structure from different perspectives.

Handouts
  • GSA Meeting 2012 Oral Presentation (Albuquerque) - version 2.pptx (8.4 MB)