--> Integration of Geophysical, Geological, and Engineering Data to Continuously Adapt and Evolve Models for Optimization and Risk Analysis of an Ongoing Carbon Storage/eor Project at Farnsworth Texas, U.S.A.

2018 AAPG International Conference and Exhibition

Datapages, Inc.Print this page

Integration of Geophysical, Geological, and Engineering Data to Continuously Adapt and Evolve Models for Optimization and Risk Analysis of an Ongoing Carbon Storage/eor Project at Farnsworth Texas, U.S.A.

Abstract

The Southwest Partnership on Carbon Sequestration (SWP) is a CO2 carbon capture utilization and storage (CCUS) project sponsored by the U.S. Department of Energy, with a goal of permanently storing 1,000,000 tonnes of CO2. The SWP project is located in a mature waterflood undergoing conversion to enhanced oil recovery (EOR) at Farnsworth, Texas, USA. Utilized CO2 is anthropogenic, sourced from a fertilizer and an ethanol plant. The field has 15 CO2 injectors and stored more than 750,000 tonnes of CO2 between October 2013 and February 2018. Major project goals are optimizing the storage/production balance, ensuring storage permanence, and developing best practices for CCUS. The ongoing geotechnical data acquisition and analysis program at FWU continues to provide valuable contributions to our understanding of the subsurface characteristics controlling fluid flow and associated mechanical behavior. Recent achievements include: pre-stack depth migration of the surface 3D dataset for improved resolution of structural and stratigraphic interpretations and inversion products, acquisition of additional time-lapse VSP surveys for continuation of the plume tracking program, and deployment of a digital borehole seismic array for improved passive monitoring. Integrated modeling workflows have been developed for joint optimization of hydrocarbon production and CO2 sequestration, auditing/monitoring of CO2 storage, and mechanical risk assessment. Machine learning techniques are employed at various stages in these workflows from self-learning techniques for data driven lithological and mechanical facies identification with borehole and seismic data, to reduced order proxy modeling techniques to facilitate optimization workflows requiring numerous realizations of transient hydrodynamic and mechanical processes. Uncertainty characterization has been improved through application of Bayesian and stochastic techniques for hydraulic and geomechanical facies and property modeling. A high resolution coupled hydro-mechanical has been developed for containment and induced seismic risk assessment. This paper presents an overview of these recent data acquisition, analysis, and integration efforts. Funding for this project is provided by the U.S. Department of Energy under Award No. DE-FC26-05NT42591. Additional support has been provided by site operator Perdure Petroleum L.L.C.