Coupling Geostatistics to Detailed Reservoir Description Allows Better Visualization and More Accurate Characterization/Simulation of Turbidite Reservoirs: Elk Hills Oil Field, California

Malcolm E. Allan, Mark L. Wilson, and Jeff Wightman

The Elk Hills giant oilfleId, located in the southern San Joaquin Valley of California, has produced 1.1 billion barrels of oil from Miocene and shallow Pliocene reservoirs. 65% of the current 64,000 BOPD production is from the pressure-supported, deeper Miocene turbidite sands.

In the turbidite sands of the 315 structure, large porosity & permeability variations in the Main Body B and Western 315 sands cause problems with the efficiency of the waterflooding. These variations have now been quantifled and visualized using geostatistics. The end result is a more detailed reservoir characterization for simulation.

Traditional reservoir descriptions based on marker correlations, cross-sections and mapping do not provide enough detail to capture the short-scale stratigraphic heterogeneity needed for adequate reservoir simulation. These deterrninistic descriptions are inadequate to tie with production data as the thinly bedded sandIshale sequences blur into a falsely homogenous picture.

By studying the variability of the geologic & petrophysical data vertically within each wellbore and spatially from well to well, a geostatistical reservoir description has been developed. It captures the natural variability of the sands and shales that was lacking from earlier work. These geostatistical studies allow the geologic and petrophysical characteristics to be considered in a probabilistic model.

The end-product is a reservoir description that captures the variability of the reservoir sequences and can be used as a more realistic starting point for history matching and reservoir simulation.

AAPG Search and Discover Article #91019©1996 AAPG Convention and Exhibition 19-22 May 1996, San Diego, California