--> Circum-Arctic Petroleum Systems: Data Mining and Prediction of Physiochemical Properties Using Chemometrics and Paleoreconstruction, by Kenneth E. Peters, L. Scott Ramos, John E. Zumberge, Christopher R. Scotese, and Donald L. Gautier; #90052 (2006)

Datapages, Inc.Print this page

Circum-Arctic Petroleum Systems: Data Mining and Prediction of Physiochemical Properties Using Chemometrics and Paleoreconstruction

Kenneth E. Peters1, L. Scott Ramos2, John E. Zumberge3, Christopher R. Scotese4, and Donald L. Gautier5
1 U.S. Geological Survey, Menlo Park, CA
2 Infometrix, Inc, Bothell, WA
3 Geomark Research Inc, Houston, TX
4 University of Texas at Arlington, Arlington, TX
5 U. S. Geological Survey, Menlo Park, CA

The Circum-Arctic is one of the last major frontiers in petroleum exploration. As part of the World Energy Consortium organized by the U.S. Geological Survey, source- and age-related biomarker and isotopic data were measured for more than 1000 crude oil and seep samples collected above 55°N latitude. A unique, multi-tiered decision tree consisting of many chemometric (multivariate statistical) models created using Pirouette® allowed detailed classification of genetically related oil groups. The results show that our new chemometric approach is more versatile than conventional methods for oil-oil and oil-source rock correlation. Using 622 ‘training set' samples, an automated protocol was created using InStep™ to classify newly acquired samples of crude oils, seeps, and source-rock extracts, and assign corresponding confidence limits.

The geochemical data were also used to infer the age, lithology, organic matter input, and depositional environment of the source rock for each oil sample. Twenty-one oil groups were identified, mapped, and linked to their source rocks; examples include Upper Jurassic distal marine shale (West Siberia), Lower-Middle Jurassic paralic-deltaic marine shale (West Siberia), Triassic marine marl (North Slope), Devonian fluvial lacustrine shale (Scotland), Devonian marine carbonate (Western Canada Basin), and Precambrian marine marl (East Siberia). To better assess the original and present-day distributions of each petroleum system, paleo-latitudes and paleo-longitudes of the samples were reconstructed using PointTracker© and located on paleogeographic maps using Earth System History-GIS© (PALEOMAP Project, C. Scotese). These paleomaps can be used to predict the physiochemical properties of discoveries within the mapped areal extent of each petroleum system, including sulfur content, API gravity, and gas-to-oil ratio.