Nonreflection seismic and inversion of surface and guided waves

D acoustic sensing (DAS) is a relatively recent development in the use of fiber-optic cable for measurement of ground motion. Discrete fiber-optic sensors, typically using a Bragg diffraction grating, have been in research and development and field testing for more than 15 years with geophysical applications at least 12 years old (Bostick, 2000, and summary in Keul et al., 2005). However, developments in recent years have sought to remove the need for point sensors by using the fiber cable itself as a sensor (Mestayer et al., 2011; Miller et al., 2012). Through Rayleigh scattering, light transmitted down the cable will continuously backscatter or “echo” light so that it can be sensed. Because light in an optical fiber travels at approximately 0.2 m/ns, a 10-ns pulse of light occupies about 2 m in the fiber as it propagates. The potential of DAS is that each 10 nanoseconds of time in the optical echo response can be associated with reflections coming from a 1-m portion of the fiber (two-way time of 10 ns). By generating a repeated pulse every 100 μs and continuously processing the returned optical signal, one can, in principle, interrogate each meter of up to 10 km of fiber at a 10-kHz sample rate. Local changes in the optical backscatter because of changes in the environment of the fiber can thus become the basis for using the fiber as a continuous array of sensors with nearly continuous sampling in both space and time. Because the technology for deploying fiber-optic cable in boreholes is well developed for thermal sensing (distributed temperature sensing, or DTS), a DAS system has the potential of having thousands of sensors permanently deployed in the subsurface, at relatively low cost. DAS systems currently use single-mode fiber, as opposed to the multimode fiber typically used for DTS, but the type of fiber does not affect deployment, and multiple fibers are easily deployed in a single capillary tube. Recent advances in optoelectronics and associated signal processing (Farhadiroushan et Thomas m. Daley, Barry m. FreiFelD, JonaThan aJo-Franklin, and shan Dou, Lawrence Berkeley National Laboratory roman Pevzner, Curtin University valeriya shulakova, CSIRO suDhenDu kashikar and Douglas e. miller, Silixa Julia goeTz, Jan henninges, and sTeFan lueTh, GFZs