Jan Vilhelm

Combining surface waves and common methods for shallow geophysical survey

Geophysical data were acquired during a survey of the Hluboka Fault in the Czech Republic, Central Europe. The recorded surface waves are studied in the frequency range 8–200 Hz. Phase velocity dispersion curves of Rayleigh and Love waves are determined from pairs of three-component seismograms with a 5 m receiver spacing by means of a frequency-time analysis along the profile. Rayleigh waves are analysed on the vertical (Z) and radial (R) components and Love waves on the transversal (T) component. Dispersion curves from the vertical component are then inverted to 1-D S-wave velocity models using the isometric method. A set of 1-D S-wave velocity models representing a pseudo 2-D S-wave velocity distribution along the profile is obtained. This velocity distribution is compared with the results of other geophysical methods and also with direct observation from a shallow paleoseismic trenching. A combination of the S-wave velocities obtained from the surface wave analysis and P-wave velocities from refraction tomography is used to estimate the Poisson ratio distribution. It is shown that the resolution capabilities of surface waves are comparable in this case with electric resistivity tomography in near surface medium and with P-wave tomography in the depths exceeding approx. 15 metres.

Assessment of Fracture Properties from P-Wave Velocity Distribution

Seismic measurements carried out on peridotite rock outcrop showed that seismic wave velocity depends on the direction of propagation. It was found out that the velocity is influenced on system of fractures. The fracture sets can cause velocity dispersion. The theory of displacement discontinuity was applied for an explanation of fractures influence on the seismic wave propagation. According to this theory, the fracture stiffness can be assessed on the basis of velocity directional dependence. The measurements made proved experimentally the validity of this approach, specifically for the frequency interval from 500 Hz to 700 kHz.

Back‐projection stacking of P‐ and S‐waves to determine location and focal mechanism of microseismic events recorded by a surface array

We present an automatic method of processing microseismic data acquired at the surface by a star-like array. The back-projection approach allows successive determination of the hypocenter position of each event and of its focal mechanisms. One-component vertical geophone groups and three-component accelerometers are employed to monitor both P- and S-waves. Hypocenter coordinates are determined in a grid by back-projection stacking of the short-time-average-to-long-time-average ratio of absolute amplitudes at vertical components and polarization norm derived from horizontal components of the P- and S-waves, respectively. To make the location process more efficient, calculation is started with a coarse grid and zoomed to the optimum hypocenter using an oct-tree algorithm. The focal mechanism is then determined by stacking the vertical component seismograms corrected for the theoretical P-wave polarity of the focal mechanism. The mechanism is resolved in the coordinate space of strike, dip, and rake angles. The method is tested on 34 selected events of a dataset of hydraulic fracture monitoring of a shale gas play in North America. It was found that, by including S-waves, the vertical accuracy of locations improved by a factor of two and is equal to approximately the horizontal location error. A twofold enhancement of horizontal location accuracy is achieved if a denser array of geophone groups is used instead of the sparse array of three-component seismometers. The determined focal mechanisms are similar to those obtained by other methods applied to the same dataset.

Automatic Processing of Microseismic Data: Determination of Hypocenter Position and Estimation of Focal Mechanism

We present an automatic method of processing microseismic monitoring data acquired at the surface with star-like array using the back projection approach that allows simultaneous determination of hypocenter position of events and an estimation of their focal mechanisms. Location coordinates are searched in a grid using two norms for seismogram stacking: STA/LTA for location and stack of real amplitudes corrected by the radiation pattern to estimate the focal mechanisms. From each point of the given grid we compute travel times to all geophones on the surface through the assumed velocity model. Afterwards we do migration of each trace using the computed travel time and stack the STA/LTA ratios of all traces. This process is repeated for each point of the grid. The most probable location corresponds to the maximum of the stack. In the second step a small and fine grid with center in the approximate location is used to refine the hypocenter position. In order to find source mechanisms we use additional grid search in source mechanism angles – strike and dip to find focal mechanism and correct polarity of seismograms according to computed theoretical polarities of the found focal mechanism. We test this method on real dataset of hydraulic fracturing monitoring of a shale gas play. We obtain similar hypocenter positions and focal mechanisms as provided by other methods applied to the same dataset.