Baseline monitoring for time-lapse pressure tomography: Initial results from Otway Stage 3

Abstract

As part of the CO2CRC Otway stage 3 project, time-lapse cross-well pressure tomography methods are being developed to demonstrate unobtrusive, continuous and low-cost monitoring of an injected CO2 plume. Active pressure tomography relies on a robust baseline set of water injections and environmental monitoring to characterise the aquifer in its base state, before CO2 has been injected. Here, we report the initial results from such baseline testing.

In late 2019, 4 new monitoring wells were drilled (CRC-4 to 7), in addition to the existing CRC-2 well (where CO2 injection occurred during the stage 2C project) and the CRC-3 well which is the primary CO2 injection well for stage 3. These wells were completed in the Paaratte aquifer formation at depth ≈1500m TVD. Two sets of baseline test sequences were performed, with water injection occurring sequentially at CRC-4, 5, 6 and 7, with all wells (including CRC-2 and 3) continuously monitoring the subsurface pressure. High-quality downhole, surface and weather data were acquired for each sequence, with monitored downhole pressures generally repeatable to within the base environmental noise level.

An adjoint Bayesian inversion method was developed to invert for the depth-averaged aquifer petrophysical properties, based on the monitored pressure signals. The Bayesian framework utilised prior aquifer information - interpreted faults and an independently developed geological static model - to inform and constrain the inversion. The resulting baseline model is able to capture the complex time-series pressure propagation across the aquifer subject to the various water injections. Heterogeneous regions of permeability, porosity and aquifer thickness account for the subtle variations in pressure peak arrival time and magnitude observed at the various wells, with spatial variations at a resolution that is a small fraction of the well spacing. The baseline injections and aquifer model provide a robust base for repeat water injections once CO2 has been injected in the next stages of the project.

Keywords: CCS monitoring, pressure tomography, Bayesian inversion, environmental response