Back to search

CLIMIT-Forskning, utvikling og demo av CO2-håndtering

Bayesian monitoring design.

Alternative title: Bayesisk monitorerings design.

Awarded: NOK 6.6 mill.

Carbon Capture and Storage (CCS) is the concept of capturing CO2 from large point sources and storing t in geological formations. The purpose is to reduce the greenhouse gas forcing and ocean acidification caused by the post-industrial increased anthropogenic emissions of CO2. To be an effective technology, the volume of CO2 to be stored will have to be substantial. Mappings have shown large capacity and plans are being developed to store large amounts of CO2 underneath the North Sea, and the Norwegian government is aiming for implementing at least one full scale project within 2022. Storage projects will be designed to keep the stored CO2 within the intended formations, and the formations and injection wells will be monitored by standard technologies to assure detection of unexpected events. However, due to the amount needed to be stored, we might have to use less than perfect formations. Hence , there will always be a small possibility that CO2 may migrate toward the seafloor undetected. As a precaution, the marine environment will have to be monitored for indications of a leak, which is also required by international regulations such as the London and the OSPAR conventions. Added complexity from present marine operations are the area to be monitored. The CO2 might reach the seafloor far away from the injection site, and the signal of a leak may be camouflaged within the natural variability in CO2 concentration. Marine waters are also notoriously hostile to instruments and marine operations are costly. False alarms, especially, will mobilize additional resources to confirm a leak may become costly and should be avoided. Due to this, it is important that the monitoring program is lean but adequate to assure detection of signals from a leak. BayMoDe addressed design of monitoring programs to detect anomalies in the marine environment as indications of a leak. Methodology from Bayesian statistics offers quantification of uncertainties, minimized during design, and assurance that alarms being real indications of a leak. The project has partners from UK (Plymouth Marine Laboratory and Heriot-Watt University) and Norway (NORCE and University of Bergen. In addition, the project worked closely with the HORIZON2020 funded project STEMM-CCS (, which among others built an environmental baseline for Goldeneye in UK sector in the North Sea. The large-scale release experiment performed in 2019 gave valuable insight into the footprint of a leak of CO2 through sediments. Results from the project have been presented at IEAGHG network meetings and at GHGT13 in Lausanne in November 2016, with subsequent proceedings publication. The project took part in four presentations or posters at GHGT14, with subsequent proceedings . Nine peer review articles have been published and several others are on their way. The project has been active in the CO2 Storage Forum meetings arranged by the Norwegian Petroleum Directorate and in various IEAGHG network meetings. A trans-disciplinary meeting on marine management was arranged in Bergen in May 2017, bringing together different disciplines within ocean sciences, marine legislation and industry. The meeting had international participants from Japan and the UK. As a follow up, the project took part in arranging a workshop on "Linking Local Responsible Research and Innovation for Global Sustainable Marine and Maritime research" as part of the Sustainable Development Goals conference at UiB in 2018. The Baymode project has been an important project, allowing us to keep a relative high visibility on our activities related to environmental monitoring as part of the assurance monitoring of offshore storage projects. This visibility was part of the reasons that the «4th International Workshop on Offshore Geologic CO2 Storage», combined with the «STEMM-CCS Open Science Meeting», was arranged in Bergen in February 2020. The project was also instrumental in establishing the consortium behind the ACTOM project, funded by the ACT Era-Net consortium. We have also secured funding from the Academy agreement between Equinor and the University of Bergen to continue the to analysis data from the STEMM-CCS experiment. Hence, the activities will continue after the completion of the Baymode project.

The BayMoDe project has contributed to: - A PhD degree in applied mathematics, using machine learning for environmental monitoring of offshore storage sites. - More than 10 peer reviewed manuscripts published or under review on marine monitoring of storage sites. - Establishment of the ACTOM project, that will operationalize BayMoDe research. - Assured increased Norwegian activities related to marine environmental monitoring, and assurance monitoring in general, of storage sites.

With the current plans to utilize the large theoretical storage capacity in the North Sea there is a need for economical monitoring technologies that will either assure storage integrity or detect adverse effects or unwanted release events over large areas. This is also required by international regulations and agreements. BayMoDe addresses the design and operation of monitoring programs aiming to detecting anomalies in the marine environment. The suggested probabilistic approach offers quantification of uncertainties in the program, and this uncertainty is minimised during design. This further reduces the chance of false alarms that will accelerate the cost significantly. This approach will automatically filter out any outliers in a time series; a single leak indication will not automatically sound the alarm but rather increase our awareness by increasing our belief that a leak is on going. Subsequent measurements might reduce or increase our awareness, only when the number of indications reaches a threshold will the extra resources be mobilized. To test the ability and usefulness of Bayes theorem in the context of environmental monitoring we aim to design a data analysis framework, including monitoring design capabilities, in which the Bayesian approach is the core data treatment. The three main building blocks in the framework will be a probabilistic map of potential leak locations, environmental baseline statistics, and predictions of leak footprint characteristics. The former two will be part of a site characterization, while the latter will in addition depend on characteristics of seeps. Even though the focus here is on seafloor monitoring, the approach has the potential to simplify documentation of uncertainty in all monitoring methods. As such the method might accelerate implementation of large-scale storage projects through better procedures for designing and maintaining monitoring programs.

Publications from Cristin

Funding scheme:

CLIMIT-Forskning, utvikling og demo av CO2-håndtering