DUNCAN, W.S. , Gulf Coast Consulting Group, Brookeland, TX

As we look forward to the future of the petroleum industry it is beneficial to re-evaluate concepts and methodologies. Identifying seeps as an indicator of hydrocarbon reserves is such a method. This old idea exploited with the versatile new technology of hyperspectral remote sensing is yielding impressive results.

The initial proof-of-concept study successfully located existing wells by pinpointing their locations from seal leaks using hyperspectral data. The seal leak results were used in this study to classify larger areas and identify natural seeps. To assess the accuracy, ground truthing verified classification of seeps in areas of known production and in environments such as abandoned stream channels. Alteration minerals' resulting from seep activity were also used for assessment. Several sites where alteration minerals were identified correlated with areas of seep detection.

Hyperspectral data were used to classify seeps in a variety of environments ranging from open water, timber (including pine, hardwood, and mixed timber) and pastureland. Hydrocarbon seep classifications were ground truthed and found to be in existing oil fields, areas exhibiting geologic features such as abandoned stream channels and crevasse splays, and sites associated with mineral alteration. These findings demonstrate the potential hyperspectral remote sensing holds in petroleum exploration. In addition to generating prospect leads, when integrated with other geophysical techniques hyperspectral remote sensing will enhance petroleum I exploration success rates.

AAPG Search and Discovery Article #90908©2000 GCAGS, Houston, Texas