Circum-Arctic Petroleum Systems Defined Using Biomarkers, Isotopes, and Chemometrics

Kenneth E. Peters¹, L. Scott Ramos², and John E. Zumberge^3
1Schlumberger, Mill Valley, CA.
²Infometrix, Inc., Bothell, WA.
³GeoMark Research, Ltd., Houston, TX.

Because petroleum systems involve multiple elements and processes, traditional exploration for subsurface traps is prone to error. Costly exploration failures, e.g., the Mukluk well in Alaska, show that traps may lack oil and gas due to various factors, such inadequate migration pathways from the source rock. Understanding the link between petroleum and the source rock reduces exploration risk, e.g., identification of the source rock for accumulations provides information on migration pathways and whether nearby structures might have filled. Unfortunately, many oil-to-source rock correlations and studies of mixed oils are unreliable because they are based on only a few parameters and seldom include statistical evaluation of uncertainty.

This paper describes new chemometric methods that use geochemical data to define petroleum systems and de-convolute oil mixtures in the circum-Arctic. Source- and age-related biomarker and isotope data were measured for ~1000 oil samples collected above ~55 degrees N latitude. A multi-tiered chemometric decision tree identified 31 oil families based on a training set of 622 samples. ‘Decision-tree chemometrics’ uses principal component and other statistical analyses to classify and assign confidence limits for oil and source-rock extract samples. For example, the method identifies seven oil families in West Siberia, four in East Siberia, and two in the Volga-Ural Basin and the corresponding source rocks.

Seventy-four of the above oil samples from the Barrow arch in Alaska were studied to assess relative contributions from different source rocks to the giant Prudhoe Bay Field. Alternating least squares of concentration data (ALS-C) for 46 biomarkers was used to de-convolute the mixtures. ALS-C results for 23 oil samples from the prolific Ivishak Formation reservoir in the field indicate similar contributions from Triassic Shublik Formation and Cretaceous Hue-GRZ source rocks (37% each), and less from the Jurassic Kingak Shale (26%). These results differ from published interpretations that most oil in the field originated from the Shublik Formation. Unlike conventional methods to assess mixtures, ALS-C does not require that pure end member oils be identified prior to analysis or that laboratory mixtures of these oils be prepared to evaluate mixing. Further application of these methods could significantly improve understanding of the origins of crude oil in other areas of the circum-Arctic, thus reducing exploration risk.

AAPG Search and Discover Article #90096©2009 AAPG 3-P Arctic Conference and Exhibition, Moscow, Russia