Ground-Based Hyperspectral Remote Sensing and Terrestrial Laser Scanning of the Eagle Ford Formation

Abstract

This study uses ground-based hyperspectral remote sensing and terrestrial laser scanning data to map the Eagle Ford Formation in west Texas. The Eagle Ford Formation consists of alternating layers of limestones, marlstones and volcanic ashes with high total organic content deposited during the Cenomanian-Turonian oceanic anoxic event. Detailed remote sensing study of the outcrop can be utilized as an analog to the source rock and reservoir at the subsurface, and provide valuable geological information as well as foresight about hydrocarbon exploration. Hyperspectral remote sensing acquires electromagnetic radiation in numerous bands in a continuous spectrum and holds great power to resolve mineralogical compositions of scanned materials without physical damage. Ground-based hyperspectral imaging scans the geologic outcrops at close ranges with very fine spatial resolution (millimeters to centimeters). Pixel-based spectra matching of study material with reference standards are performed by spectral angle mapper algorithm, which revealed the variations of calcite and kaolinite concentrations among the alternating layers. Classifications allowed quantifying rhythmic layers of limestones, marlstones and ashes as well as other lithological variations in the Eagle Ford Formation. Laboratory spectroscopy is used to assist with mineral identification and image classification. Thin section petrography and X-ray diffraction data verified the classification results of hyperspectral remote sensing. Terrestrial laser scanning (TLS) is a novel LiDAR technique which provides fast and accurate 3D models and enables detailed stratigraphic and structural studies including bed thickness variations, lateral continuities, fracture density and orientations, etc. Stratigraphic bedding planes are automatically detected by machine learning techniques from the 3D geometry. And calibrated laser intensity data is utilized in lithology identification. Combining hyperspectral remote sensing and TLS data, this study creates 3D outcrop models with detailed mineralogical compositions, and provides geologic analogs to extract geo-mechanical characteristics. The utilization of these new techniques in geo-statistical analysis provides a workflow for employing remote sensing in resource exploration and exploitation.

AAPG Datapages/Search and Discovery Article #90259 ©2016 AAPG Annual Convention and Exhibition, Calgary, Alberta, Canada, June 19-22, 2016