The CO2CRC Otway International Test Centre in Victoria serves as a globally recognised research facility for advancing monitoring technologies that ensure safe and effective geological carbon storage (GCS). Stage 4 Otway Project is designed to test continuous and time-lapse seismic monitoring during controlled CO₂ injection into a heterogeneous saline aquifer at ~1.5 km depth. Between November 2024 and January 2025, 10 kt of CO₂-rich gas were injected. Monitoring utilised distributed acoustic sensing (DAS) across multiple wells, permanent surface orbital vibrators (SOVs), and time-lapse vertical seismic profiling (VSP) surveys to provide both active and passive seismic datasets.

For Stage 4, the final monitor dataset from the Stage 3 campaign in 2021 was adopted as the baseline. The first dedicated Stage 4 monitor (M9) survey was acquired in February 2025. A dedicated 4D processing workflow produced coherent and repeatable results across vintages, enabling robust time-lapse comparison. The 4D analysis successfully imaged the full 10 kt CO₂-rich gas plume, with anomalies clearly observed near the CRC-4 and CRC-5 wells, consistent with the injection interval and the known heterogeneity of the aquifer. These observations confirm the ability of DAS-enabled VSP to resolve plume migration and geometry at the reservoir scale.

A further 10 kt injection is scheduled for January–March 2026. Two additional monitor surveys are planned for December 2025 and March–April 2026, which will extend the dataset and further validate permanent, fibre-optic based seismic monitoring as a cost-effective and scalable solution for long-term CO₂ storage conformance and assurance.

Distributed acoustic sensing (DAS) is well-suited to detecting induced seismicity, but routine processing must cope with terabytes of data, heterogeneous backgrounds, and severe class imbalance. Classical STA/LTA—originally devised for single seismometers—requires adaptation to massive sensor arrays and modern acquisition geometries. We present a practical workflow that augments multi-channel coincidence STA/LTA with positive–unlabeled (PU) learning and distance-based ranking. From each trigger we compute compact temporal and spatial attributes tailored to DAS (e.g., temporal consistency/uniformity, spatial continuity, channel-gap statistics, coincidence efficiency), apply log transforms to mitigate heavy tails, and robustly scale with respect to the positive class. Without explicit negatives, we learn the geometry of the positive manifold from a small set of labeled microseismic examples and assign each unlabeled trigger an ensemble similarity score that combines distance to the positive centroid with k-nearest-neighbour proximity in standardized feature space. The system operates section-wise for long arrays with chunked I/O and caching, yielding auditable rankings that reduce tens of thousands of triggers to a few hundred candidates for expert review. We demonstrate the workflow on subsea DAS data and outline its extension to other acquisition geometries, including downhole and surface deployments. The approach is conservative, scalable, and interpretable: it prioritizes events similar to known positives while avoiding over-commitment to assumed negatives, and it is designed to accommodate future incorporation of geometry-specific priors as catalogues grow.

When CO2 is injected into deep saline aquifers, the resulting pressure build-up may cause micro-seismicity, fault reactivation, and induce damaging earthquakes. Continuous strain data are needed to measure vertical strain migration. Deploying a fiber-optic cable behind a well casing for subsurface geomechanical monitoring offers the opportunity to continuously track the deformation (strain) along the fiber-optic cable. 

 

We conducted water injection/production tests in both domestic and overseas sites and measured the strain profiles by using distributed fiber optic strain sensing (DFOSS) technique. Strain sensing based on Rayleigh scattering is focused on caprock deformation as well as well integrity monitoring due to the pressure build-up caused by CO2 injection, and the pressure communication between the injection/observation well and the ground water well when producing water for sampling following the regulation.

 Carbon capture and storage projects in deep reservoirs have
  significant monitoring obligations associated with regulatory needs  and social licence to operate. Downhole methods are becoming increasingly attractive because of their low surface footprint and ease of stakeholder engagement. When several wells are available, monitoring from VSP-based seismic methods is possible using downholeDAS fibre optics, and even lower footprint seismic methods are  possible using surface orbital vibrators combined with DAS.  Crosswell pressure tomography (PT) monitoring is also possible using test water injections in completions in the reservoir interval and downhole pressure gauges.  Monitoring confidence is enhanced when  all monitoring data can be explained with a single model or  range of models, which requires joint Bayesian inversion. However, simultaneous inversion of time-lapse seismic data and PT data is  generally a difficult problem involving coupled multi-physics  modelling. In the CCS context, highly buoyant and thin gas plumes make the problem susceptible to effective approximations that enable decoupling of the physics, and thereby a cascaded Bayesian inversion  for both seismic amplitudes and crosswell pressures. We show inversions for the Otway stage 3 CCS demonstration project using these ideas.