Preliminary TOUGH2 Model of King Island CO 2 Injection Christine Doughty Presented by Curt Oldenburg Earth Sciences Division Lawrence Berkeley National Laboratory October 25, 2011 Modeling Approach Characterization study no injection But injection can be simulated Properties obtained from nearby wells, core, well logging, and surface seismic can be used to define the model system Simulation of injection at Phase III scale or industrial scale can be carried out to understand local site and regional performance Two examples from other regions, one at each scale Preliminary King Island simulation results 2 Oldenburg p.1
West Coast Regional Carbon Sequestration Partnership GCS Pilot-Test Simulation (Chris Doughty (LBNL)) Denbury CO2-EOR at Cranfield, Mississippi Model Results SECARB Phase III CO2 Injection Injecting 200-500 T/day CO2 since December 2009 into water leg on flank of dome Initial Condition: brine saturated with CH4 Key Features in Breakthrough Curves Two monitoring wells with U-Tubes to collect samples, mass spec for gas composition Pulse of gas-phase CH4 forms at leading edge of plume as CO2 preferentially dissolves Multiple tracer pulses released Multiple CH4 peaks illustrate distinct flow paths above and below shale baffle Time-lapse well logging for saturation profiles Seismic and electrical resistivity imaging Low, wide peaks for later tracer releases evidence of self-sharpening plume front Model Development Injection-well log shows two high-permeability layers separated by a shale baffle Layered model, finer at wells CO2 injected over interval shown by black bar Simulations of a Full Deployment CO2 Storage Scenario in the Illinois Basin (Zhou and Birkholzer (LBNL)) W isconsin Iowa Ka nk ak ee -500 m Arc h Illinois Cincinnati Arch M is siss ippi Km 900 W isconsin Arch Groundwater Resource Region R iv er A rch 1000 Indiana ADM Site is no Illi 800 sin Ba 700 O 600 500 za rk D Core Injection Area om m e Kentucky -4300 m Missouri Pascola Arch 800 900 1000 1100 1200-300 -700-1100 -1500-1900 -2300-2700 -3100-3500 -3900-4300 1300 Km Elev. of the top of the Mt. Simon Fm. Concern has been raised that CO2 storage causes unacceptable pressure rise (e.g., Ehlig-Economides and Economides, J. Petrol. Tech., 2010). We have simulated the injection of 5 Gt CO2 over 50 years at 20 sites (5 Mt/yr at each site) in the Mt. Simon Formation. Simulated pressure rise is approx. 30 bars centered in the injection area. This pressure dissipates with time as brine flows into low-permeability cap rock over large areas. CO2 plumes from each of the 20 injection sites do not intermingle as 3D effects (vertical flows) allow efficient filling of pore space (Zhou and Birkholzer, GHGS&T, 2011). CO2 Saturation at 200 years Pressure Buildup (bar) at 50 years 0.6 0.5 0.4 0.3 0.2 0.1 200 years -1800-2000 m 30 25 20 15 10 5 1 Pressure rise (bar) at end of injection 140 km -2200 190 km CO2 saturation after 200 years Oldenburg p.2 0.6 0.5 0.4 0.3 0.2 0.1 0-2400 704 706 km 708 710 Vertical cross-section of CO2 saturation after 200 years for one site.
King Island Location Map 5 Preliminary Model Development Vertical Layering Structure Lateral Extent and Boundary Conditions Material Properties Initial Conditions Grid and Representation of well 6 Oldenburg p.3
Citizen Green Well Formation Tops and Core Intervals Core intervals Nortonville shale Domengine sand Capay shale King Island 1-28 well Citizen Green well Top of shale Top of sandstone Core interval Mokelumne River sand H&T shale Top Starkey sand shale & sand (Starkey) Peterson #1 sand shale (Starkey) Peterson #2 sand shale (Starkey) Four target sands: Domengine Mokelumne River Top Starkey Starkey/Peterson #1 7 Ignore gorges Approximate dip as uniform dip: 1.6 o up to ENE From Downey and Clinkenbeard, 2011 8 Oldenburg p.4
Axis aligned with dip direction West and south boundaries approximately aligned with faults East boundary open as Mokelumne may abut another permeable formation Closed Closed Constant Pressure Closed 9 Material Properties Rough estimates and literature values Sands Porosity 0.25-0.35 Permeability 50 500 md Anisotropic: k v /k h 0.01 to 0.1 Shales Porosity 0.10 Permeability 10 d Anisotropic: k v /k h 0.1 Properties are effective values that account for sub-grid-scale heterogeneity 10 Oldenburg p.5
Initial Conditions Single-phase liquid brine (salinity 0.05, or 50,000 ppm) Hydrostatic pressure profile Geothermal temperature gradient 11 Grid and Representation of Well Rectangular grid 24 layers 31 by 30 grid blocks per layer, 100 m resolution at well Diagonal well represented by stair-steps in rectangular grid Injection partitioned among grid block representing well according to permeabilitythickness product Does not account for different pressure gradients in well (CO 2 ) and formation (brine) Over-estimates injection at greater depth 12 Oldenburg p.6
Simulation of CO 2 Injection TOUGH2 numerical simulator Fully coupled multiphase fluid flow Isothermal simulations CO 2 exists as a supercritical phase and dissolves in brine Two cases 1 MT over 4 years (phase 3 size) 10 MT over 4 years (better problem for coarse grid) 13 CO 2 injection at 250,000 T/yr (t = 4 years) 14 Oldenburg p.7
CO 2 injection at 2,500,000 T/yr (t = 4 years) 15 Future work Incorporate actual geology Refine grid Simulate different injection depths Simulate post-injection period to study trapping Incorporate sub-grid-scale heterogeneity 16 Oldenburg p.8
CO 2 Inventory Evolution Doughty, C.A., Trans. Porous Med., 82, 49-76, 2010. 17 Acknowledgments This work was supported by the Assistant Secretary for Fossil Energy (DOE), Office of Coal and Power Systems, through the National Energy Technology Laboratory (NETL), U.S. Department of Energy, and by Lawrence Berkeley National Laboratory under Department of Energy Contract No. DE-AC02-05CH11231. 18 Oldenburg p.9