From Bureau of Economic Geology, The University of Texas at Austin (www.beg.utexas.edu).
For more information, please contact the author.

AAPG Annual Convention, Calgary, Alberta, Canada, June 19–22, 2005

Ordovician Karst Mapping with Hyperspectral and LIDAR Data

Jerome A. Bellian, Charles Kerans, Richard A. Beck, Jerry Lucia, and Robert G. Loucks

Abstract:

The Franklin Mountains in El Paso, Texas, expose a world-class Ordovician paleocave system. Recent mapping of this system using tripod-mounted lidar, GPS, DEM data, and hyperspectral satellite imagery has revealed mappable spectral patterns that allow us to better understand the paleocave development, collapse, and dolomitization in three dimensions.
The area is characterized by excellent midlatitude illumination and excellent outcrop exposure in steep and varied topography. Preliminary studies using a full-range spectroradiometer (400–2,500 nanometers) show that a variety of calcite- and dolomite-rich facies can be differentiated using hyperspectral reflectance spectra. Atmospherically corrected hyperspectral imagery from the EO-1 spacecraft's Hyperion instrument enables us to differentiate facies at 30 meter spectral resolution and resolve these carbonate facies although additional tasks over areas of interest are required.

Recent work in the Mesozoic of Utah suggests that off-nadir summer acquisitions will be useful in this and other areas of steep topography.

Ground-truth spectral acquisition during satellite overpasses also improves overall classification accuracy. Extensive averaging of training spectra maximizes the signal-to-noise ratio of the training spectra. These practices maximize the effectiveness of full-range hyperspectral facies mapping for hydrocarbon-exploration-model development. The result is a new paradigm for three-dimensional, quantitative, hybrid hyperspectral/topographic facies analysis that leverages rock-reflectance spectra invisible to the human eye.