2nd Meeting of the Risk Assessment Network

Publication Overview

The meeting continued the progress made at launch network meeting in developing our understanding of the status of risk assessment in its application to CCS and developing the role that risk assessment can play.

Publication Summary

The main conclusions from the meeting can be summarized as follows:

Site characterization is a key component in ensuring that the storage sites selectedcan effectively contain CO2 for 1000’s of years. Risk Assessment (RA)is onetool that can be used in the early screening of storage sites. RA and sitecharacterization work in an iterative manner, over different project stages frompreliminary screening to permitting to implementation.2. Risk assessment studies can provide guidance on likely seepage rates fromstorage sites but they cannot define the impacts of leakage. EnvironmentalImpact Assessments (EIA) can provide the framework for assessing the long termimpacts of leakage. However, there is little research work underway currentlythat is addressing specifically the effects of CO2 leaks and their potential impactsthat could allow an EIA to be compiled. This is a major research gap.3. A communication exercise with regulators has been undertaken to gauge theirexpectations for risk assessment and to make them aware of the current status ofRA. As a result of this process regulators are better informed on both the rolethat RA can play and its current technical status.4. However, RA is only part of the message that needs to be given to regulators;remediation is another important issue as well. Also, we need to get the messageover that we are not promoting innovatory technology to avoid over regulation.5. RA can also be considered as part of a Risk Management framework. RA is themeans of identifying, estimating or calculating and evaluating potential risks.Risk management on the other hand deals with assessing, monitoring &remediating risks to conform to risk acceptance levels.6. Natural analogues could be used to build confidence in CCS. By building up adatabase of events from natural and industrial analogues comparable to those thatcould occur from a CO2 storage reservoir you can build a risk matrix that allowsyou to compare and communicate the risks of CCS in a way that is readilyunderstandable to those outside the CCS community.

Four recently reported RA cases studies on potential geological storage sites werereviewed; three were based on aquifers and one on an oil field operation. Itshould be emphasized that several of these cases were not full blown riskassessments but were really scoping studies. The results of such studies shouldtherefore be treated with some care when communicated outside of the technicalcommunity. The aquifer based assessments generally suffered from a lack of data,which is not unsurprising, which required a lot of assumptions to be made. Theoil field case was much better characterized which allowed a more detailed riskassessment process to be undertaken. The oil field study gives us someconfidence that CO2 can be retained in that formation for 1000’s years the samedegree of confidence cannot be drawn from the aquifer studies.8. The RA case studies completed to date have contributed significantly to thelearning process for undertaking such studies which will be of benefit in thefuture and help to allow us to better define the data requirements needed tocomplete a good robust risk assessment.9. More RA studies are needed to help develop confidence in the techniques andmodels used as well in the results they generate

The meeting has raised a number of issues that warrant further consideration at future network meetings. These include:

• On the issue of site selection we need to define how much characterization isneeded to allow a formal risk assessment to be completed
• We need to ask the question whether a full blown quantitative risk assessment isrequired to give regulators confidence that a storage site is secure? Or could asimpler screening assessment be sufficient to generate confidence in CO2 storage?
• Also now that we have some experience of using FEP databases for riskscreening and scenario development can we design a screening process involvinga simpler FEP database?
• There is a desire by regulators and project implementers alike to see thedevelopment of a RA standard or protocol. We need to decide how best toproceed to develop such a protocol or standard
• Similarly we need to process to peer review models and benchmark RA tools andapproaches. We also need to decide how best to proceed to develop abenchmarking process.

In addition, the meeting has identified that within the RA community there is a need to try and harmonize the terminology used to allow the community to effectively communicate amongst itself let alone to outside bodies. In summary, it was clear that we have gained a lot of knowledge from the RA cases completed to date, but that learning is far from complete and we need more case studies to be undertaken to build our confidence in the tools, models and approaches used. Also the RA studies to date have only been undertaken on storage reservoirs, we also need studies on the full CCS chain to help convince the public that the whole system is safe not just the storage reservoir.