Multi-layer, sharp-interface models of pore pressure buildup within the Illinois Basin due to basin-wide CO2 injection

We recently developed and applied a new parallel, multi-layer, finite-element model to the Illinois Basin in order to assess the spatial extent and magnitude of pore pressure increases resulting from the annual projected injection of 100 million metric tons of CO2. One focus of this work is to asses... Full description

Saved in:
Bibliographic Details
Online Access: Get full text
Authors:Zhang, Y.; Person, M.A.; Kerra, S.; Celia, M.A.; Nordbotten, J.M.; Bandilla, K.; Elliot, T.R.; Rupp, J.; Ellett, K.M.; Bowen, B.B.; Pickett, W.; Woolsey, E.E.
Volume Title:AGU 2011 fall meeting
Source:American Geophysical Union Fall Meeting, Vol.2011; American Geophysical Union 2011 fall meeting, San Francisco, CA, Dec. 5-9, 2011. Publisher: American Geophysical Union, Washington, DC, United States
Publication Date:2011
Note:In English
Subjects:Aquitards; Cambrian; Carbon dioxide; Carbon sequestration; Computer programs; Data processing; Fluid flow; Global change; Greenhouse gases; Hydrology; Injection; Models; Mount Simon Sandstone; Numerical models; Paleozoic; Permeability; Pollution; Reservoir rocks; Stress; Transport; Upper Cambrian; Waste management; Illinois Basin; United States
Record ID:2015030442
Copyright Information:GeoRef, Copyright 2020 American Geosciences Institute. Reference includes data supplied by, and/or abstract, Copyright, American Geophysical Union, Washington, DC, United States
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
We recently developed and applied a new parallel, multi-layer, finite-element model to the Illinois Basin in order to assess the spatial extent and magnitude of pore pressure increases resulting from the annual projected injection of 100 million metric tons of CO2. One focus of this work is to assess the potential for inducing a seismic event associated with low effective stress conditions around CO2 injection wells in the southern Illinois Basin where Mt Simon permeability is relatively low (< 50 mD). We used a sharp-interface formulation to represent a CO2, freshwater, and brine transport within each layer. A simple parallelization scheme was used in which fluid transport in each layer is solved on a separate processor. The layers are linked at the after each time step through vertical fluxes of fresh and saline water across their respective confining units. This model was validated, in part, by comparison to computed pore pressure distributions from a published 8-layer test case. Our Illinois Basin model represents spatial variations in porosity using a modified form of Athy's law. Permeability is logarithmically related to porosity so that permeability. Principal reservoirs represented in our model include the Mt. Simon Formation, the Knox Dolomite, Ordovician carbonates, Silurian-Devonian and Mississippian-Pennsylvanian sandstone/carbonates units. Key confining unit represented include the Eau Claire, Maquoketa, and New Albany Shales. A limited number of low-permeability faults were also included in the model. The permeability of fault elements were set to between 10-100 times lower than surrounding sediments. We calibrated our model using historical freshwater pumping data from the Chicago area (128 million gallons per day of H2O) as well as the salinity distribution across the Illinois Basin. We found that incorporating a stream network which included the Rock River near Chicago was important in reproducing pre-development head patterns in the Cambro-Ordovician aquifer system. This suggests that aquitards in this area are not perfectly confining. We were able to match the spatial extent (about 150 km) and maximum drawdown (270 m) around Chicago using our basin-scale model. We plan to simulate the injection of CO2 into the Mt Simon Formation in the central Illinois Basin and the Knox Dolomite in the southern Illinois Basin in Kentucky.