High-Resolution Mapping and Monitoring of Shallow Shear-Wave Velocity in Urban Pasadena with Distributed Acoustic Sensing

Date: 4/24/2019

Time: 06:00 PM

Room: Fifth Avenue

We excite resonance of Caltech's Millikan Library between 0.8 and 5 Hz using a specialized shaker installed on the roof of the nine-story building and record monochromatic Rayleigh waves generated by the building along the Pasadena Array. Operating almost continuously since May 2018, the Pasadena Array utilizes distributed acoustic sensing (DAS) technology to convert a >25-km optical fiber, originally installed by the city of Pasadena for telecommunications use, into a dense array of >5000 linear strainmeters. Rayleigh wave velocity at each frequency is computed from observed phase lag between any pair of stations, and then inverted to form a continuous shear-wave velocity model throughout the city with lateral resolution between 1 and 10 m. We benchmark our model with velocities calculated at coarser resolution from ambient noise correlation functions. Throughout Pasadena, Vs30 varies as much as 40 percent across spatial scales of less than 10 m, suggesting that current stochastic models of near-surface elastic heterogeneity may significantly underestimate the intensity of fractal randomness at large wavenumbers. Because measurement uncertainty decreases with shaking duration at each frequency and the Millikan Library source is fully repeatable, we are capable of high-precision time-lapse monitoring of shallow velocity structure. With new DAS arrays being installed throughout Southern California in 2019, the methods we demonstrate are designed to be generalizable, permitting robust, non-invasive, low-cost estimation of shear-wave velocity on a block-by-block scale in urban areas and hence a revolution of scale in the accuracy of ground motion predictions.

Presenting Author: Ethan F. Williams

Authors