280 Raman‑based DTS for temperature, Brillouin‑based distributed strain/temperature (DSS/DTSS) for deformation, and Rayleigh‑based DAS for dynamic strain/vibration, Figure 3. This multi‑physics stack is widely reported in dams and TSFs and is increasingly deployed using the same fibre cables to minimise installation burden and maximise lifecycle value. [1][7][9] Figure 3: backscatter lights (right), travel through fiber optics and processed by interrogator (left) For TSFs, fibre layout is designed specifically to form a consequence‑based hazard model: (i) seepage‑sensitive zones (downstream toe, drains, abutments, underdrain manifolds), (ii) deformation‑sensitive zones (crest, raises, interfaces between lifts, abutments), and (iii) dynamic‑sensitive zones (areas affected by blasting, traffic, seismicity, and construction). Fibre installations can be modular and expand during raises by splicing additional cable segments and updating the light-pulse path and monitoring zone, an approach explicitly described for TSFs to maintain continuous coverage over time. [6][7]. 3.2 DTS for seepage and moisture pathways Distributed Temperature Sensing (DTS) enables seepage detection by continuously measuring temperature along fiber-optic cables embedded in the tailings or dam embankment. Natural seasonal temperature fluctuations in the reservoir act as a thermal forcing signal. Under normal (non-seeping) conditions, this heat propagates into the embankment by conduction. Seepage introduces advective heat transport, accelerating the thermal signal and creating detectable anomalies relative to the surrounding material. Passive DTS systems can resolve very small temperature changes and are therefore effective for early, spatially continuous identification of seepage pathways. This method is described in field monitoring literature and TSF‑specific implementations. [5][7][8]. Although not a TSF, a USACE embankment study provides a rigorous template for TSF practice: toe‑installed fibre, coordinate calibration, temperature anomaly observation during pumping, and coupled modelling to estimate expected responses under higher hydraulic loading, Figure 4. [4][5]
RkJQdWJsaXNoZXIy MTM0Mzk2