AAPG Bulletin, V. 82 (1998), No. 12 (December 1998), P. 2220-2245.

Heterogeneity Modeling and Geopseudo Upscaling Applied to Waterflood Performance Prediction of an Incised Valley Reservoir: Countess YY Pool, Southern Alberta, Canada¹

Madeleine Peijs-van Hilten,² Timothy R. Good,³ and Brian A. Zaitlin⁴

©Copyright 1998.  The American Association of Petroleum Geologists.  All Rights Reserved

¹Manuscript received April 2, 1997; revised manuscript received February 24, 1998; final acceptance March 21, 1998.
²Graduate of Heriot-Watt University. Current address: Halliburton Energy Services, Reservoir Description, 91-1SE-25J, 10200 Bellaire Boulevard, Houston, Texas 77072-5299; e-mail: madeleine.peijs@halliburton.com
³Mobil Technology Company, 13777 Midway Road, Dallas, Texas 75244-4390; e-mail: tim_good@email.mobil.com
⁴PanCanadian Petroleum Limited, 150 9th Avenue S.W., P.O. Box 2850, Calgary, Alberta, Canada T2P 2S5; e-mail: zaitlin@earthlink.net

This research project was done as the final part of a Master’s degree program in reservoir evaluation and management at the Petroleum Engineering Department of Heriot-Watt University in Edinburgh, Scotland. Many individuals provided assistance so that this study could be completed within the available period of three months. The Petroleum Engineering Department is acknowledged for providing computer facilities and probe-permeameter equipment. Philip Ringrose and Gillian Pickup are thanked for helpful information and discussions on geopseudo upscaling. We gratefully acknowledge PanCanadian Petroleum Limited, who operates the Countess YY Pool, for organizing and financing this project, for access to the data and the core facility, and for permission to publish this work. We especially thank Eugene Peters and Garth Syhlonyk for sharing their knowledge on the Countess YY pool and incised valley reservoirs in general. Reviews of the manuscript by Bob Dalrymple, Eugene Peters, Janok Bhattacharya, I. M. Yarus, and an anonymous reviewer are very much appreciated. 

Abstract

The objective of this study is to analyze the effects of different modeling approaches and various scales of geological heterogeneity on waterflood recovery and volumetrics of an incised valley reservoir. Seismic, well-log, and core data are integrated with an incised valley facies model to create cross sections used to perform two-phase 2-D (two-dimensional) fluid-flow simulations.

Core observations and probe-permeameter data are acquired to perform a geopseudo upscaling exercise, which simulates the effects of small-scale sedimentary structures on fluid flow. Applying this method and incorporating small-scale sedimentary structures in 2-D fluid flow simulations have proved to make a significant difference in individual-well oil recovery (up to 8%) depending on the facies types involved in a well’s drainage area. Incorporating variations in sand-body dimensions and connectivities has proved to have a major impact on field oil recovery (30% difference between extreme 2-D cases), whereas variations in incised valley size have the greatest impact on original oil-in-place values (21% between extreme 2-D cases).

A layercake model of an incised valley reservoir results in optimistic performance compared to a geopseudo upscaled model (11% higher oil recovery for a 2-D case). In a highly favorable scenario, an incised valley reservoir indeed may behave like a layercake, but it is more likely that it will not perform as well.

Not taking into account small-scale sedimentary structures, uncertainty in reservoir architecture, and incised valley size in reservoir simulation studies can introduce substantial errors in reserves estimation and production forecasting. Lessons learned from this 2-D study will be used in a future full-field three-dimensional waterflood simulation of the Countess YY pool.