Fibre optic sensing of fracture processes: from small-scale laboratory experiments to multi-scale applications

Distributed fibre optic sensing (DFOS) is increasingly used for subsurface monitoring due to its ability to provide dense spatial coverage and a broad range of strain measurements, from slow quasi-static deformation to rapid dynamic signals. Distributed strain sensing (DSS) provides detailed measurements of strain evolution along the fibre, enabling identification of strain localisation. However, linking these measurements to underlying fracturing and failure processes remains challenging, as it is unclear how strain localisation, acoustic emission (AE) activity, and observed fracture development relate to the same underlying fracture process. These signals are often analysed separately, limiting the ability to consistently relate DFOS observations to fracture processes and to understand how these relationships evolve across spatial scales.

In this study we present results from the first stage of a multi-scale experimental campaign designed to investigate how fracture processes are expressed in DSS, AE, and high-speed imaging within a single controlled experiment. The focus is on controlled laboratory experiments on cylindrical sandstone samples of approximately 6 cm in diameter and 12 cm in height. Cyclic loading is used as a controlled probe of damage evolution, allowing progressive activation and reactivation of deformation and fracturing processes over repeated loading cycles. The loading protocol is based on monotonic failure tests and consists of stepwise increases in displacement amplitude with repeated loading–unloading cycles at each level.

High-resolution DSS measurements are conducted on the sample surface to capture the development of strain localisation. In addition, experiments include configurations with fibres deployed both on the sample surface and within a borehole drilled through the sample, enabling direct comparison between externally observed deformation and internal strain response. These measurements are complemented by AE monitoring using sensors located on the sample surface to track microcracking activity and by high-speed camera imaging to observe fracture initiation and propagation. The cyclic protocol enables identification of the onset and evolution of localised deformation, as well as changes in signal response between successive loading cycles. The experiments focus on how the different measurement techniques respond to the same evolving damage state and how signals recorded at the surface relate to those observed within the sample.

Preliminary results show how DSS and AE signals evolve in time and space during fracture nucleation and propagation, and how these relate to directly observed fracture development. The comparison between surface and borehole measurements provides insight into how internal deformation processes are expressed in fibre optic signals, with implications for interpreting borehole-based monitoring data. These observations provide a basis for identifying robust indicators of fracture evolution and assessing their sensitivity to loading history.

This centimetre-scale study forms the foundation for subsequent experiments at larger scales, where similar protocols will be applied to investigate the consistency of observed relationships under more complex conditions. With our multi-scale approach, we aim to improve the interpretation of distributed fibre optic measurements and support the development of more reliable, physics-based monitoring strategies for subsurface systems.