Measuring pressure performance of a large saline aquifer during industrial-scale CO2 injection: The Utsira Sand, Norwegian North Sea

Abstract

The Sleipner injection project has stored around 14Mt of CO2 in the Utsira Sand and provides a unique opportunity to monitor the pressure response of a large saline aquifer to industrial-scale CO2 injection. There is no downhole pressure monitoring at Sleipner, but the 4D seismic programme provides an opportunity to test whether reliable indications of pressure change can be obtained from time-lapse seismic. Velocity–stress relationships for sandstones, calibrated against measured data from the Utsira Sand, indicate that pore pressure increases of <1MPa should produce measurable travel-time increases through the reservoir. Time-lapse datasets were used to assess travel-time changes by accurately mapping the top and base of the reservoir on successive repeat surveys outside of the plume saturation footprint. Measured time-shifts are of the order of a very few milliseconds, with significant scatter about a mean value due to travel-time ‘jitter’. The ‘jitter’ is due to non-perfect repeatability of the time-lapse surveys and shows a Gaussian distribution providing a useful statistical tool for determining the mean. Observed mean time-shifts through the Utsira Sand were 0.097ms in 2001 and 0.175ms in 2006. These correspond to mean pressure increases of less than 0.1MPa. An idealised noise-free reservoir ‘impulse response’ was computed, taking into account lateral reservoir thickness variation. Convolving this with the repeatability noise distribution gives a realistic predicted reservoir response. Comparing this with the observed time-shifts again indicates a pressure increase less than 0.1MPa. Flow simulations indicate that pressure increases should range from <0.1MPa for an uncompartmentalised reservoir to >1MPa if strong flow barriers are present, so the results are consistent with the Utsira reservoir having wide lateral hydraulic connectivity.