ORIGIN AND GENERATION TIMING OF COOK INLET PETROLEUM: CIRCUM-ARCTIC CHEMOMETRICS AND 1-D GEOHISTORY MODELING

PETERS, Kenneth E.¹, MAGOON, Leslie B.¹, and HUDSON, Travis², (1) U.S. Geological Survey, 345 Middlefield Road, MS 969, Menlo Park, CA 94025, [email protected], (2) Applied Geology, Inc, 902 Vista del Mar, Sequim, WA 98382

Source- and age-related biomarker and isotopic data were used to infer the age, lithology, organic matter input, depositional environment, and identity of the source rock for >1000 Circum-Arctic crude oil and seep samples collected above ~60°N latitude. A unique, multi-tiered decision tree consisting of many chemometric (multivariate statistical) models allowed classification of 22 genetically distinct Circum-Arctic oil groups. An automated protocol based on the decision tree was created to classify and assign confidence limits for newly acquired samples of crude oils, seeps, and source-rock extracts. The decision tree classified Cook Inlet oil samples from Trading Bay, McArthur River, Middle Ground, Beaver Creek, and Swanson River fields as originating from Jurassic distal marine shale similar to that found in the North Slope and Sverdrup Basin. Age-related biomarker ratios, such as C₂₆ tricyclic terpane/Ts, exclude the possibility of Triassic or older source rock. The results of this new approach support previous studies suggesting a Middle Jurassic Tuxedni Group source rock in the Cook Inlet.

A calibrated one-dimensional (1-D) petroleum systems model located in the depocenter between the Middle Ground Shoal and Swanson River fields in the Upper Cook Inlet indicates that petroleum generation from the Tuxedni Group source rock began as early as 22 Ma and achieved the gas generation zone by 9 Ma (Early and Late Miocene time, respectively). Older models required that petroleum generated in Paleocene time migrate updip to the basin flanks where it was trapped stratigraphically until structural traps formed during Miocene deformation. In these older models, petroleum generated prior to deposition of the Hemlock Conglomerate (30 Ma) could be lost due to erosion, but this is not a constraint in the new model. Risk analysis of the 1-D model indicates that heat flow is the key parameter affecting calculated vitrinite reflectance (R_o). R_o was calculated using 50 risk runs that sampled a Gaussian distribution of heat flow with 5% standard deviation that was centered on that used in the 1-D model (60 mW/m²). The mean value for present-day R_o of the Tuxedni Group source rock at 27,100 ft. depth in the 1-D location is 2.9 ±0.6%.