Exploration and Production Technologies

Mapping Alternations caused by Hydrocarbon Microseepages in Patrick Draw Area, Southwest Wyoming, Using Image Spectroscopy and Hyperspectral Remote Sensing


The project goal is to develop and implement a methodology for using remote sensing imagery to identify and map alteration zones caused by hydrocarbon microseepages in the Patrick Draw area of southwest Wyoming.

University of Houston, Houston, TX

Results from the Hyperion hyperspectral remote sensing work at the Patrick Draw oilfield in southwest Wyoming demonstrates a successful application of hyperspectral mapping techniques when applied to the identification of subtle geochemical alterations associated with hydrocarbon microseepage. These results are confirmed by geochemical work. The final report is available below under "Publications".

Using information gained from this project, a new methodology will be developed to remotely map alteration zones produced by hydrocarbon seepages. Calcite samples will be analyzed for carbon isotopes in order to confirm the hydrocarbon source for the alterations. This work can be used for mapping buried oil shales, tar sands, or other hydrocarbons. This will allow scientists to see the effects of hydrocarbons on natural vegetation. The techniques developed in this project could be applied for finding and mapping subsurface hydrocarbons in the United States and elsewhere.

The long-term seepage of hydrocarbons in the subsurface can cause a diverse array of chemical and mineralogical changes in the form of alterations in rocks and soils. The resulting alterations can include the formation of calcite, pyrite, uraninite, elemental sulfur, and certain magnetic iron oxides and iron sulfides; bleaching of red clay beds; clay mineral alteration; electrochemical changes; radiation anomalies; and geomorphic anomalies.

Alteration caused by hydrocarbon seepages has been reported from the Patrick Draw area in southwest Wyoming. Microseepages are vertical or near-vertical movements of hydrocarbons from the reservoir to the surface without clear visible evidence. The best available method for detection of hydrocarbon microseepage is the measurement of anomalous amounts of light hydrocarbons in soil gases and soils directly over petroleum deposits. Soil gas hydrocarbons have carbon isotope ratios very similar to the gases in the underlying deposits, while ratios for biogenic hydro-carbons differ from those values. This, along with the good compositional correlation between surface microseeping hydrocarbons and the type of underlying production, is convincing evidence of vertical hydrocarbon migration.

Detection of underlying hydrocarbon reservoir accumulations using remote sensing techniques had its inception with the identification of macroseeps. However, today we find ourselves relying on the detection of more subtle characteristics associated with petroleum reservoirs, such as microseeps. Microseepages are the result of vertical movement of light hydrocarbons from the reservoir to the surface through networks of fractures, faults, and bedding planes that provide permeable routes within the overlying rock. Microseepages express themselves at the surface in an array of alterations and anomalies, such as chemical or mineralogical changes in overlying soils and sediments. This project has developed remote sensing methodologies using NASA’s Hyperion hyperspectral imaging sensors and ground-truthing techniques in order to identify and map alterations caused by hydrocarbon microseepages whose presence has been confirmed through the use of soil gas surveys in previous studies. Also, using geochemical data, potential relationships between identified alterations and microseeping hydrocarbons was established in the Patrick Draw area so as to support connections drawn between microseepages and overlying alterations.

Training the classification of satellite imagery with spectral inputs of samples collected over previously defined areas of hydrocarbon microseepage resulted in the successful identification of an anomalous zone. Spectral Angle Mapper (SAM) as well as Mixture Tuned Matched Filter (MTMF) techniques were utilized for classification of images. SAM is a method for directly comparing image spectra to known spectral endmembers input by the user and/or defined from the image spectra itself. MTMF is a classification method that also provides a means of detecting specific materials based on matches to user endmember input or image-derived endmember spectra. However, unlike SAM, which picks the best match to a given spectrum in the image, MTMF first maximizes the response of the known endmembers and suppresses the response of the unknown background signal in the image prior to matching. SAM analysis proved to be more effective in the classification of anomalous areas, or areas that are distinct from their surroundings.

Geochemical characteristics of samples that defined this anomalous zone were then compared to the remaining non-anomalous samples using XRD, ICP, spectroscopy, and carbon isotope techniques. Spectroscopy results demonstrated higher proportions of clays within the anomalous samples compared with non-anomalous samples, along with greater amounts of quartz in non-anomalous samples relative to anomalous samples. XRD analyses demonstrated the increased presence of feldspars in non-anomalous samples compared with anomalous samples. Geochemical results provided evidence for the removal of soluble constituents from the weathering sites of anomalous samples versus non-anomalous samples. The d13C values range from -2.880/00 to as low as -45.320/00.

Although this project has laid the groundwork for the identification of hydrocarbon microseepages in the Patrick Draw area using remote sensing techniques, much room exists for the continued development and application of these methodologies. For example, due to the lack of outcrop in the study area, most analyses are based on soil samples. Because soils represent the weathered potion of the surrounding rock, their mineralogical signatures tend not to be as strong, both spectrally and geochemically. This situation is not always ideal for the Hyperion hyperspectral remote sensing system, which tends to be somewhat noisy and has a low spatial resolution of 30 by 30 meters. Thus, further analyses in the area might consider implementing a remote sensing system with a less noisy spectral resolution and higher spatial resolution so as to help characterize even more subtle alterations associated with hydrocarbon microseepages that can become lost in the already weathered soil conditions.

Current Status (July 2007) 
This project was completed on October 9, 2006.

This project was funded under a Historically Black Colleges and Universities grant.

Project Start: July 22, 2005 
Project End: October 9, 2006

Anticipated DOE Contribution: $19,995 
Performer Contribution: $0

Contact Information 
NETL – Daniel Ferguson (daniel.ferguson@netl.doe.gov or 918-699-2047) 
U. of Houston – Shuhab Khan (sdkhan@uh.edu or 713-743-3411)

Final Project Report [external site - OSTI)

Khan, S.D., Jacobson, S., “Hyperspectral Remote Sensing and Geochemistry Show Surface Alteration in SW Wyoming is Related to Hydrocarbon Microseepages,” Geological Society of America Bulletin.

Jacobson, S., “Identifying Surface Alterations Caused by Hydrocarbon Microseepages Using Hyperspectral Remote Sensing Techniques in the Patrick Draw Area of Southwest Wyoming,” MS thesis, University of Houston, 2006.

Jacobson, S., and Khan, S.D, “Mapping Alterations Caused by Hydrocarbon Microseepages in the Patrick Draw Area of Southwest Wyoming, Using Image Spectroscopy and Hyperspectral Remote Sensing,” American Association of Petroleum Geologists Annual Convention, 2006.

Jacobson, S., and Khan, S.D., “Mapping alterations caused by hydrocarbon microseepages in the Patrick Draw area of Southwest Wyoming, using image spectroscopy and hyperspectral remote sensing,” expanded abstract, Society of Exploration Geophysicists Annual Meeting, Vol. 24, No. NSE P1.3, Houston, TX, November 2005.

StayConnected Facebook Twitter LinkedIn RssFeed YouTube