--> Abstract: Caprock Integrity Assessment: A Case Study of Natural Fracture Characterization in Mudstones of the Clearwater Formation, Alberta, Canada, by Queena Chou and Kevin Gillen; #90124 (2011)

Datapages, Inc.Print this page

AAPG ANNUAL CONFERENCE AND EXHIBITION
Making the Next Giant Leap in Geosciences
April 10-13, 2011, Houston, Texas, USA

Caprock Integrity Assessment: A Case Study of Natural Fracture Characterization in Mudstones of the Clearwater Formation, Alberta, Canada

Queena Chou1; Kevin Gillen2

(1) Geomechanics Services, Weatherford, Calgary, AB, Canada.

(2) Consultant, Burnaby, BC, Canada.

Caprock integrity plays a pivotal role in the maximum allowable operating pressures in steam-assisted recovery of heavy oil. Natural fracture networks represent potential pathways for breach of caprock by steam. This study details the methodology involved in the collection and processing of data and the assessment of the fracture architecture of a SAGD producing property in the Athabsaca area.

The caprock comprises both the Clearwater and Wabiskaw units that overlie the productive McMurray Formation oil sands. It consists of predominantly mudstones with interspersed silt and sand intervals, and averages a total package thickness of 50 meters. Seventeen boreholes penetrating this interval were available for study. Continuous coring into PVC sleeves was employed. In most wells, both resistivity-based and acoustic imaging logs were available over the same intervals as core. Procedures for data amalgamation were developed to ensure that redundant information from multiple sources within a single borehole were distilled down into a single set without duplication of the same fracture or fault. This combination yielded a more reliable assessment of fracture origin and character. For example, both acoustic logs and core confirm that the majority of fractures which appear resistive (closed) on the resistivity-based image logs are in fact NOT mineralized.

Contoured stereonets show that fracture orientations display a preferential orientation, and are somewhat coherent on a local scale. However, fracture populations can be quite variable between boreholes. The combination of core and image log data indicates that any one source is unlikely to yield a complete picture of fracture development in the vicinity of a borehole. In general, core and at least one image log are necessary.