The maximum strain comparison for the first and second order tracking algorithm ε1 and ε2 (Equations 1.50–1.51) is presented in Figure 1.30 for FS = 50kHz and L0 = 10m. The second order algorithm can deliver measurements of fiber strain up to fiber breakage point (~10%) at frequencies of around 10 Hz.
Figure 1.29 Displacement noise comparison of DAS (with and without engineered fiber) with seismometer and geophone. 30 m DAS data are for synthetic gauge length.
Figure 1.30 Maximum strain comparison of first and second order algorithms for DAS.
We can now estimate the maximum DAS dynamic range D as:
(1.52)
Using the real noise level εmin = 0.03nanostrain from Miller et al. (2016), we can estimate D = 99dB for a maximum value ε1 = 2.9μstrain. This estimation gives the practical upper limit for seismic DAS at 100 Hz using Rayleigh scattering. Generally speaking, the second order tracking algorithm has limited applicability for a conventional DAS because flicker noise pulses can reach π and destroy measurements in accordance with Equation 1.49. Nevertheless, 120dB was achieved in Parker et al. (2014) when the fiber elongation zone was significantly smaller than the gauge length and pulsewidth, such that the flicker noise was suppressed. However, when a continuous seismic signal expands the reflectivity zone, then the reflection can disappear, and the signal has ambiguity. Fortunately, in engineered fibers, the scatter center zones are well defined, and so the reflectivity change is negligible. As a result, we can optimistically estimate a maximum D = 167dB for engineered fiber using εmin = 1picostrain and maximum ε2 = 220μstrain—see Figure 1.30.
The dynamic range of DAS with engineering fiber was tested during a dry alluvium geology series of chemical explosions, including 50,000 kg TNT‐equivalent at 300‐m depth‐of‐burial (Abbott et al., 2019). “Two orders of magnitude more data relative to traditional geophones/accelerometers” was successfully recorded.
Summarizing, we can conclude that theoretical estimations demonstrate that the performance of DAS with engineered fiber can potentially exceed that of conventional geophones and seismometers. In general, given that the overall sensitivity of a DAS system is a function of the coupling, cable, fiber, electronics, and digital signal processing, field data is most convincing, and, in the next section, we will discuss some examples of high definition seismic and microseismic data that demonstrate the benefits of the engineered fiber DAS solution as compared to conventional DAS and geophones.
1.3.3. Field Trial Results
A comparison of DAS with standard and engineered fiber for a seismic sweep signal is presented in Figure 1.31. This measurement was provided using two different fibers placed side by side in the same optical cable, so the elongation of both fibers was identical. The top graphs (a) and (c) demonstrate the difference between the time‐distance representation; the right panel (c), which represents engineered fiber, is visibly cleaner than the left panel (a). The detected seismic signal has the same shape (around 10 nm peak to peak for a channel 898) for engineered (d) and standard (b) fiber, except noise. Some change in amplitude (20%) can be explained by incomplete averaging of the DAS signal over distance, as is shown in Figure 1.7. There was less variation in the amplitude level for engineered fiber, and this stability can be important for 3D VSP, as was shown in Figure 1.21.
A comparison of DAS acoustic noise with standard and engineered fiber is presented in Figure 1.32. Noise spectral density versus distance is practically constant for engineered fiber (b) but varies significantly for standard fiber from channel to channel along distance (a). In other words, we can conclude that standard DAS noise depends on fiber randomness and can be far from the average value, but engineered fiber DAS noise is predictable. The SNR difference is emphasized by the signal Fourier transform in the bottom chart (c): the noise reduction for engineered fiber is nearly 20 dB as was expected from shot noise estimation (Equation 1.46).
Figure 1.31 Comparison of DAS with Rayleigh scattering [(a) and (b)] and engineered fiber [(c) and (d)] for a seismic sweep signal. Acoustic signals are measured in optical phase radians.
Figure 1.32 Comparison of DAS noise spectrums with Rayleigh scattering (a) and engineered fiber (b). Panel (c) represents acoustic noise spectrum density with respect to 1 rad/Hz0.5.
Source: Based on Richter et al. (2019).
Fine spatial resolution in combination with good sensitivity gives DAS a significant advantage for detection of microseismic events, particularly where a geophone chain cannot be readily positioned. Such measurements are used in fracking jobs, where a wireline fiber optic cable is pumped down into an already completed observation well (Richter et al., 2019). This gives the possibility to determine the frack height and well interference with unprecedented clarity.
A typical microseismic event is presented in Figure 1.33, where both S‐ and P‐waves are clearly visible, such that the distance from observation well to fracking event can be easily detected. Figure 1.34 shows how the same installation can be used to detect a “frac hit,” where a fracking zone and strain extends slowly from the well undergoing treatment to the observation well. This new data allows completion engineers to map the depth, azimuth, and speed of the fractures and feed that information back into the fracture models to validate and optimize the designs for the next operation.
Figure 1.33 Microseismic event in observation well detected by DAS with engineered fiber.
Figure 1.34 Example of low frequency (down to millihertz level) “slow strain” data,