A strategy for automated analysis of passive microseismic data to image seismic anisotropy and fracture characteristics

Abstract

Monitoring of induced seismicity is gaining importance in a broad range of industrial operations from hydrocarbon reservoirs to mining to geothermal fields. Such passive seismic monitoring mainly aims at identifying fractures, which is of special interest for safety and productivity reasons. By analysing shear‐wave splitting it is possible to determine the anisotropy of the rock, which may be caused by sedimentary layering and/or aligned fractures, which in turn offers insight into the state of stress in the reservoir. We present a workflow strategy for automatic and effective processing of passive microseismic data sets, which are ever increasing in size. The automation provides an objective quality control of the shear‐wave splitting measurements and is based on characteristic differences between the two independent eigenvalue and cross‐correlation splitting techniques. These differences are summarized in a quality index for each measurement, allowing identification of an appropriate quality threshold. Measurements above this threshold are considered to be of good quality and are used in further interpretation. We suggest an automated inversion scheme using rock physics theory to test for best correlation of the data with various combinations of fracture density, its strike and the background anisotropy. This fully automatic workflow is then tested on a synthetic and a real microseismic data set.

The limits of ray theory when measuring shear wave splitting in the lowermost mantle withScSwaves

“…Synthetic waveforms were measured for shear wave splitting using SHEBA (Wuestefeld et al 2010), which implements the minimumeigenvalue method (Silver & Chan 1991) to retrieve a set of splitting parameters which best linearise an arrival's particle motion. The program has been modified to take account of recent work highlighting the under-reporting of formal uncertainties in the measurement in the original formulation (Walsh et al 2013).…”

When do fractured media become seismically anisotropic? Some implications on quantifying fracture properties

“…The rotated waveforms are correlated using the same windows, where the time lag associated with the largest correlation peak gives a measure the SWS time difference. By comparing the similarity in the calculated time differences from the two methods a Q S WS value is defined, where values close to one represent good splitting and values close to negative one are good nulls (i.e., no SWS) (Wüstefeld et al, 2010). When the value of Q S WS is close to zero, the data quality of the splitting is poor or inconclusive.…”

“…Typically one window size is chosen but a range of window start and end times are evaluated to cover the maximum possible time delays that could be expected. A grid search of analysis windows over these intervals allows for a much faster calculation of SWS parameters than manual picking and provides a measure of the overall SWS quality Q (Wüstefeld et al, 2010).…”

“…We use the algorithm SHEBA (e.g., Teanby et al, 2004b;Wüstefeld et al, 2010) to compute the SWS parameters. The analysis of SWS requires first rotating the three component waveforms into a local ray coordinate frame, where the P-wave energy will be constrained to the ray direction (P) and the shear-wave energy on the two remaining components (S V and S H ).…”

Development of Software for Processing Microseismic Data with Application to Observation of Shear-wave Splitting

“…S-wave splitting was observed in microseimic data (Wuestefeld, et al 2010), and anisotropic velocity models were build and used for microsesmic event location (e.g. Grechka and Yaskevich 2014).…”