The article examines how Super Typhoon Muifa in September 2022 highlighted the limitations of conventional landslide monitoring. It showcases distributed acoustic sensing (DAS) with buried optical fibers as a continuous, high-resolution solution for tracking slope stability during extreme weather.
A Chinese DAS network captured micro-deformation and seismo-acoustic signals over a 50-km corridor. This offered new insights into when and where landslides are likely to occur, even under cloud cover and heavy rainfall.
Muifa’s impact and the limits of traditional monitoring
During Muifa, rapid, widespread slope failures occurred under intense rainfall. Cloud cover limited optical or satellite observations and restricted the effectiveness of many conventional sensors.
This event underscored the need for real-time, densely sampled monitoring that can operate in challenging conditions and across mixed urban–mountainous terrains. The study shows that relying solely on standard instruments can miss progressive precursors and fail to provide timely warnings in the most at-risk zones.
Researchers explored an approach that leverages existing fiber networks to infer ground motion. DAS, paired with phase-sensitive optical time-domain reflectometry, converts optical phase changes along buried fibers into continuous measurements of soil strain rate.
This method opens the possibility of near-real-time slope monitoring over long transects with high spatial density. It works even where traditional stations are sparse or infeasible.
DAS technology and its application to slope monitoring
Distributing sensing along a 50-km segment in Zhejiang Province, the team converted raw phase data to strain-rate signals with a target spatial resolution of about 50 m and an analysis cadence of 1 Hz. The buried cables, installed at roughly 1 m depth, provided strong soil coupling, enabling detection of sub-millimeter deformations and progressive precursory signals that are often missed by conventional sensors.
DAS offers long-range coverage at a lower per-kilometer cost than dense networks of geophones or conventional instruments. This makes it a compelling option for real-time landslide early warning during extreme weather events.
The study’s signal-processing approach combined Fourier and S-transform analyses to separate micro-deformation noise from other environmental vibrations. It also aligned DAS-derived signals with meteorological data, including wind, humidity, and rainfall.
By focusing on low-frequency energy (0–0.5 Hz), the researchers could identify energy signatures associated with slope instability, even under heavy atmospheric loading. This integration of geophysical signals with environmental data is a key step toward actionable hazard assessment in complex terrain.
What the Zhejiang deployment accomplished
Analysis of the two confirmed landslides during Muifa revealed pronounced energy surges in the DAS spectrograms, with magnitudes ranging from 20 to 80 dB. These surges concentrated at elevations of roughly 200–300 m and in zones with high topographic variability, highlighting how topography modulates landslide susceptibility.
High-resolution time–frequency analysis distinguished two signal patterns: a persistent, lower-energy micro-deformation consistent with retrogressive scarp formation, and abrupt, high-energy bursts signaling rapid triggering zones. This dual pattern provides a more nuanced view of slope behavior than would be possible with traditional sensors alone.
The study demonstrates that DAS can differentiate stable deformation from precursory activity. The ability to track micro-motions and sudden bursts over a wide area enables operators to prioritize response in sectors most likely to slide.
This improves both forecasting and emergency management during storms that produce intense rainfall and strong winds.
Three DAS-derived metrics guiding forecasting and response
The authors propose an indicator-based framework built on three DAS-derived metrics to link environmental disturbance with landslide occurrence and to enhance forecasting and rapid response. These metrics include:
- Low-frequency energy content in the 0–0.5 Hz band, associated with slow deformation precursors.
- Energy surges and bursts in DAS spectrograms, indicative of rapid triggering zones during intense rainfall or wind events.
- Spatiotemporal patterns of energy distribution, identifying elevations and topographic settings with elevated instability risk (e.g., 200–300 m and high variability zones).
These indicators, when correlated with meteorological conditions, can improve situational awareness and shorten the window between detection and action. The approach also underscores DAS’s potential to augment, rather than replace, existing hazard-monitoring frameworks.
Looking ahead: benefits, gaps, and integration
While promising, the study notes that fully leveraging DAS under extreme conditions requires further work.
Challenges include system optimization, robust data interpretation in the presence of environmental noise, and seamless integration with hazard-management workflows.
Here is the source article for this story: Monitoring landslide disturbances using distributed acoustic sensing under extreme weather conditions