Geophysics in Geothermal Exploration

104 Geophysics in Geothermal Exploration The processing classically applied to surface waves is based on spectral analysis and involves two steps. The first step is the construction of a dispersion curve (a plot of phase-velocity against frequency). The next step aims to obtain shear wave velocity (i.e. stiffness) profiles as a function of depth and horizontal position along the seismic survey line. The profiles are calculated using one of the two iterative purposes: matching the experimental dispersion curves to a theoretical curve derived from forward analytical modeling (usually 1D model) or an automatic least-squares approach. A geophysical survey was carried out in Yellowstone National Park (USA), in the Obsidian Pool Thermal Area. The goal of the seismic survey carried out at this site was to study shallow hydrothermal systems, characterize fluid pathways and improve understanding of the depths at which steam separates from liquid water. The area is characterized by extensive CO2 diffuse degassing and isolated thermal features with water temperatures between 21.9 and 84.0 °C. Seismic data were collected in July 2016 along a south-southwest−north-northeast transect, crossing a heat-flow anomaly between 50 and 120 m and a degassing feature between 86 and 96 m. The equipment and parameters used in the seismic survey were: • a 5.4 kg sledgehammer source swung onto a metal plate. The plate was hit five times at each position to increase the S/N, • 10 Geometrics Geode seismographs, with 24-channels in each one, • 4.5 Hz vertical component geophones spaced every 1 m, obtaining a 239 m long profile, • 25 shot gathers recorded every 10 m, • a sampling rate of 0.125 ms and a recording time of 0.75 s, to include the full surface wavefield. In addition, a GPS survey and airborne LiDAR data collection were carried out to extract the topography. The processing of the surface waves data was carried out using SWIP and readers can find supplementary information about this practical processing sequence in Pasquet and Bodet (2017). After field data windowing for validation of the 1D model hypothesis, the seismic record from its original time–distance domain was transformed into the frequency– phase-velocity domain. This step results in a set of frequency–phase-velocity pairs specifying dispersion curves. The experimental dispersion curves were identified in the f-k domain and the location of maxima energy were picked. The dispersion curve is a diagram of phase velocity versus frequency and Figure 2.39 (Top) shows examples of single dispersion curves from shots located at 0 m, 10 m, 50 m, and 60 m. Through the utilization of multi-shot acquisition setups, the inversion of the sets of dispersion curves leads to Pseudo-2D section of average S-wave velocity model (Figure 2.39, bottom) The S-wave velocity model is characterized by velocities ranging between 50 and 600 m/s, with higher shallow velocity below the heatflow anomaly observed between 50 and 120 m.

RkJQdWJsaXNoZXIy NjA3NzQ=