Overall good progress is being made with the experience gained in both monitoring and modelling from demonstration and pilot projects. There is real progress in streamlining MMV at projects and reducing the costs of monitoring. Several sites have now demonstrated conformance of a modelling-monitoring loop, leading to an improvement in the understanding of this principle.
Key monitoring discussion points:
- The use of fibre-optic distributed acoustic sensors (DAS) at projects, including helical configured cables to overcome the limitations of directional signals.
- Reducing the level (and cost) of monitoring for commercial scale projects compared with the initial research-orientated projects. This is now happening at Shell’s QUEST project in Alberta where the research team are able to streamline their initial modelling, monitoring and verification (MMV) strategy without losing monitoring effectiveness, including the use of a new laser-based low-cost leakage detection technique over the well area.
- This reduced monitoring principle was also studied in the monitoring, reporting and verification (MRV) plan for Occidental’s CO2-EOR project.
- Leakage detection discussions were centred on the temporal and spatial complexity of near surface baselines and the implications for near-surface monitoring, its purpose, optimization and value to stakeholders, and hence the need for attribution methodologies to identify genuine leakage. An example from Japan on a new technique for doing this for offshore leakage was presented.
Key modelling discussion points:
- The complexities and challenges of upscaling from pore to core to reservoir, highlighting the importance of the influence of heterogeneity in the reservoir.
- Modelling flow in wellbores was discussed with several examples showing that it can require a different modelling approach.
- The US DOE’s (Department of Energy) NRAP programme has produced 10 ‘tools’ for reduced-order modelling of CO2 storage, these have been beta-tested and useful feedback was given in a dedicated session.