Satellite‐based measurements of surface deformation reveal fluid flow associated with the geological storage of carbon dioxide

Abstract

Interferometric Synthetic Aperture Radar (InSAR) data, gathered over the In Salah CO2 storage project in Algeria, provide an early indication that satellite‐based geodetic methods can be effective in monitoring the geological storage of carbon dioxide. An injected mass of 3 million tons of carbon dioxide from one of the first large‐scale carbon sequestration efforts, produces a measurable surface displacement of approximately 5 mm/year. Using geophysical inverse techniques, we are able to infer flow within the reservoir layer and within a seismically detected fracture/fault zone intersecting the reservoir. We find that, if we use the best available elastic Earth model, the fluid flow need only occur in the vicinity of the reservoir layer. However, flow associated with the injection of the carbon dioxide does appear to extend several kilometers laterally within the reservoir, following the fracture/fault zone.

Fluid injection into a confined porous layer

“…While the volumetric rates at which fluid enters each interstitial layer need not equal Q, since not all of the CO 2 necessarily leaks through to the upper horizons, they must nevertheless be comparable to or less than Q across the nine layers. At the In Salah test site, the density of carbon dioxide ρ ≈ 850 kg m −3 and its viscosity µ ≈ 8 × 10 −4 Pa s are both slightly larger than at Sleipner, owing to the higher pressures associated with its deeper injection (Vasco et al 2010). The aquifer has a thickness of 20 m, with no separation into interstitial layers, and having the significantly smaller porosity φ ≈ 0.15 and permeability k ≈ 10 −14 m 2 (Vasco et al 2010).…”

“…At the In Salah test site, the density of carbon dioxide ρ ≈ 850 kg m −3 and its viscosity µ ≈ 8 × 10 −4 Pa s are both slightly larger than at Sleipner, owing to the higher pressures associated with its deeper injection (Vasco et al 2010). The aquifer has a thickness of 20 m, with no separation into interstitial layers, and having the significantly smaller porosity φ ≈ 0.15 and permeability k ≈ 10 −14 m 2 (Vasco et al 2010). While the injection at Sleipner occurs at a single injection well, that at In Salah is achieved through three separate wells several kilometres apart.…”

“…Our interest stems mainly from the application to the emerging technology of carbon capture and storage (CCS), where carbon dioxide (CO 2 ) is injected at high pressures into confined aquifers deep underground (Orr 2009). At the In Salah test site in Algeria, CO 2 is injected roughly 2 km into a porous sandstone layer containing ambient salt water, forming a relatively shallow porous layer with a typical thickness of about 20 m through which the injected current of CO 2 flows (Vasco et al 2010).…”

Efficient global optimization of reservoir geomechanical parameters based on synthetic aperture radar-derived ground displacements

“…The consequent adjustment of the stress field and the volumetric deformation of the geologic formations can reactivate faults located in the surroundings of the reservoir (Trugman et al, 2014), induce seismic events (Lei et al, 2013), produce localized ground ruptures (Li et al, 2000), land subsidence (Fielding et al, 1998;Motagh et al, 2008;Chang et al, 2014), or uplift (Vasco et al, 2010;Teatini et al, 2011). Thus, reservoir management raises serious concerns in terms of human health, safety of structures and infrastructures, economic risks, and environmental and hydrologic impact (Morton et al, 2006).…”

“…Advanced 3D viscoelastoplastic models have been used to reproduce the geomechanical behavior of complex, multipool, faulted reservoirs worldwide (Ferronato et al, 2003;Vasco et al, 2010;Teatini et al, 2011;Castelletto et al, 2013;Rinaldi and Rutqvist, 2013). Despite significative developments, model reliability in predicting localized (e.g., along fault systems) and/or distributed (e.g., land subsidence) deformations of subsurface systems due to change in pore pressure is hampered by large uncertainties in model parameters.…”