GSA 2020 Connects Online

Paper No. 193-6
Presentation Time: 10:55 AM

MULTIMINERAL PETROPHYSICAL ANALYSIS TO SOLVE FOR COMPLEX LITHOLOGIES OF UNCONVENTIONAL RESERVOIRS: A CASE STUDY FROM THE DUVERNAY SHALE IN ALBERTA, CANADA


BECERRA DELMORAL, Rafael and KONSTANTINOVSKAYA, Elena, Earth and Atmospheric Sciences, University of Alberta, 1-26 Earth Sciences Building, Edmonton, AB T6G 2E3, Canada

The Duvernay Shale is one of the most growing unconventional plays in North America. Located in western Canada, it has aroused a lot of interest among its producers regarding how to improve the completion of the horizontal wells and hydraulic fracture stimulation design through a deep understanding of the reservoir quality and variable geomechanics.

The complex lithology of this formation requires a complete set of geophysical logs, including advanced wireline logs to be able to solve for key mineral and fluid volumes through a conventional petrophysical analysis. However, advanced wireline datasets are not always available, especially in a number of wells that were logged when these technologies were not in the market, which forces the analyst to either discard these wells or to end up with the development of over-simplified models.

A case study is presented to show how by applying a multimineral petrophysical analysis (which determines the mineral composition of the formation by iterating the wireline tool responses with the geological constraints and the matrix parameters), and a basic quad-combo wireline dataset, one can overcome this challenge. The mineral model obtained as a result showed very good match with core measurements and advanced wireline logs in wells where these were available.

One critical step in the workflow consisted of selecting the proper matrix parameters so that the model reflected the varying formation characteristics, using as quality check the wells with core measurements and in a later step these parameters were used in the wells with no core measurements. Additionally, the construction of synthetic curves with the neural network technique was also included in the workflow, when some of the basic logs where not available.

The approach applied in this investigation allowed to model the complex lithology of the formation without the need of advanced wireline datasets and core measurements, and to determine how the mineral composition affects the geomechanical properties that are critical for selecting the landing zones of the horizontal wells and for hydraulic fracture stimulation design in ultra-low permeability formations like the Duvernay Shale.

This work was funded by NSERC through grants RGPIN 2019 04397 and DGECR 2019 00186.