“3-D Close-the-Loop” Modeling of the Grosmont Reservoir

Abstract

The Grosmont platform in Alberta, Canada, holds one of the largest heavy oil accumulations in the world (400Bboe), of which 90Bboe are within Shell's current leasehold. Grosmont is a highly prolific, heterogeneous, fractured carbonate reservoir below an angular unconformity with well developed sinkholes on the top of the carbonate. Shell's primary target zone within the Grosmont platform is the Upper Ireton formation, a Late Devonian, Frasnian, intertidal platform dolomite. The reservoir depth of less than 400 meters provided the ability to acquire high resolution seismic. Using this seismic data in a “3D Close-the-Loop” study enabled us to check the validity of the existing reservoir model and to better understand spatial distribution of porosity in the Upper Ireton formation. This work integrated an existing static reservoir model from Petrel, with high resolution seismic data and well logs, calibrated with a carbonate rock physics model. The main purpose of this study was to use the best match between the synthetic seismic response from the existing static reservoir model and re-processed seismic data for future Petrel model updates. The work was executed in three steps: (1) 3D Check-the-Loop: comparison between the synthetic response of the original 3D static reservoir model from Petrel with the re-processed seismic data. (2) 3D Close-the-Loop: adjustment of various rock physics parameters using Shell proprietary inversion software in order to get the best match/fit between generated synthetic and re-processed seismic volumes. (3) QC and application of the results: update of porosity and layer thickness properties in the static reservoir model. After multiple iterations, we successfully executed a “3D Close-the-Loop” workflow using our Probabilistic Seismic Inversion (PSI) algorithm. The subsequent QC step showed a reasonable match between the porosities derived from XStream PSI inversion and the measured porosities at the available vertical and lateral blind wells within the seismic volume. High resolution, multiple trend porosity maps of each single zone were then used as new spatial porosity distributions in the static model instead of the old, single trend maps covering several zones.

AAPG Datapages/Search and Discovery Article #90216 ©2015 AAPG Annual Convention and Exhibition, Denver, CO., May 31 - June 3, 2015