Test bed for design of commercial monitoring approach - SECARB early test at Cranfield
Abstract
Monitoring a geologic sequestration site to demonstrate long term retention of injected CO2 is in some ways simple, requiring measurements confirming that no unexpected fluid migration has occurred. However, in the technical sphere, documenting a negative is in many ways difficult. Our research-oriented project deployed many more tools than will be recommended for commercial projects. The SECARB "early" monitoring program at Cranfield tested a suite of leakage detection and quantification methodologies over three years and during injection of more than three million tones of CO2. The research team focused on sensitivity and feasibility of four types of above-injection zone monitoring were studied: (1) pressure perturbation in selected above zone monitoring intervals (AZMI's), (2) time lapse geophysics, (3) groundwater geochemical surveillance, and (4) vadose zone gas surveillance. An AZMI's is a selected, laterally well connected, vertically confined permeable zone located in a position to intercept out-of zone fluid migration. Several types of pressure, temperature and geochemical sampling methods were used to test for allochthonous fluids and documented high sensitivity of the method, although need for improvement to minimize interference from the installation is needed. The sensitivity of well-bore, cross-well, and surface deployed time lapse geophysics was tested in an onshore environment with a complex fluid history. Significant variability of tool and inversion response to the changes in the reservoir zone provide a new metric informing the error bar that should be placed on such measurements. Deep-sourced methane interpreted on the basis of geologic and geochemical evidence to be a result of vertical migration over geologic time is found in the groundwater and vadose zone at this site. Near-surface methane biodegradation produces CO2 that can mimic a storage formation release. The distribution of gasses are complicated by near surface heterogeneity, both natural and engineered, and by historic oilfield practices. Identifying leakage of injected CO2 from depth is shown to be possible but to require isolation of signal from the complexities of the setting. Modeling the injection to test predictive capabilities and critical assessment of project results are underway to recommend best monitoring approaches to go forward to commercial applications. This research-oriented early project will lead to improved confidence in storage security in the following more commercially-oriented projects through knowledge sharing. An additional two years of collaborative data analysis and project integration are planned to analyze the large amounts of data collected.
- Publication:
-
AGU Fall Meeting Abstracts
- Pub Date:
- December 2011
- Bibcode:
- 2011AGUFM.H42C..02H
- Keywords:
-
- 1848 HYDROLOGY / Monitoring networks