Saline aquifer CO2 storage : a demonstration project at the Sleipner Field : Work Area 5 (Geophysics) : gravity modelling of the CO2 bubble

A principal aim of the SACS project is to monitor the injected CO2 by geophysical methods and to develop a robust and repeatable monitoring and verification methodology for future CO2 sequestration operations. This report evaluates the applicability of microgravity surveys as a means of monitoring the future subsurface distribution and migration of the Sleipner CO2 bubble. Time-lapse seismic data acquired in 1999, after 2.3 MT of CO2 injection, show an exceptionally clear image of the CO2 bubble, characterised by very high reflection amplitudes. The outer envelope of the amplitude anomaly roughly defines an elliptical cylindrical ‘bubble envelope’, ~ 225 m high, with a major axis of ~ 1500 m oriented NNE and a minor axis of ~ 600 m. Gravity modelling was based on a number of scenarios. Two ‘in situ’ scenarios assume that the CO2 is entirely contained within the bubble envelope. The 1999 and 2001 in situ models assume respectively that 2.3 MT and 4MT of CO2 are contained within the envelope. Two migration scenarios are also modelled. The first assumes that 2.3MT of CO2 migrate vertically upwards into the overlying caprock succession to between depths of 375 and 600 m. The second migration model looks further ahead to the situation where 3 x 107 m3 (~ 10.5 – 21.0 MT depending on the density) of CO2 have been injected, and migrate laterally beneath the caprock at the top of the reservoir. Results depend strongly on the assumed density of the injected CO2 at reservoir conditions, which is subject to significant uncertainty. Only one, poorly-constrained, reservoir temperature measurement of 37 ° C is available. A density-depth profile based on this suggests that the density of CO2 in the reservoir is ~ 700 kgm-3. However the possibility of significantly lower densities cannot be discounted and modelling also includes a lower density case of 350 kgm-3. The 1999 and 2001 in situ cases produce anomalies which would be barely detectable if the higher density of CO2 is assumed. With the lower density however anomalies should be readily detectable with a modern seabed gravimeter. The vertical migration scenario indicates that large-scale vertical migration into the caprock, to depths where densities would be unequivocally lower, would be readily detected. The lateral migration scenario, whereby a single thin layer of CO2 migrates beneath the top reservoir seal, produces small anomalies which may be locally detectable but with insufficient resolution to enable effective migration mapping. However if lateral migration is via several layers, beneath intra-reservoir shales, then anomalies should be more usefully measurable. Obtaining time-lapse gravimeter readings directly above the bubble would appear to offer the best chance of obtaining useful information. Coupled with geometric information provided by the time-lapse seismic data, the gravity should be able to discriminate between the low and high CO2 density scenarios. This would provide important constraints on future reservoir modelling and also the volume estimates based on the seismic velocity pushdown effect. Related to this, gravity data would offer the potential to provide independent verification of the amount of CO2 sequestered. In addition gravimetric surveys above the bubble could provide an effective ‘early warning’ of major caprock breaching.