Summary
Microseismic imaging relies on accurate velocity models to produce reliable event properties. When the velocity model is incorrect event properties such as location, magnitude, and moment tensor may become un- dependable. Current methods to quality control (QC) or update the velocity parameters rely on arrivals with high signal-to-noise ratios that allow for picking or having known source locations such as perforation shots. Here we present a method to QC P- and S-wave velocity models without the need for picking or known source locations. This technique applies wave-equation migration and extended imaging conditions to passive seismic data. We first introduce the extended imaging conditions, then demonstrate the usefulness of the method through synthetic examples.
Introduction
Most methods to QC velocity models from microseismic events rely on perforation shots, or other sources with (assumed to be) known spatial and temporal origins. These procedures generally attempt to reduce the mismatch between the imaged or inverted location and the known location. Most borehole methods rely on picking arrivals (Prugger and Gendzwill, 1988) while surface methods, usually with low signal-to-noise data, use migration methods (Artman et al., 2010; Chambers et al., 2010). If focal locations do not match the known locations, it is common to update the velocity models by applying bulk shifts to velocity values or event locations, adjusting individual layers, or introducing anisotropic parameters (Maxwell, 2014). For high signal-to-noise data, travel time residuals from perforation shots can be inverted for velocity model updates (Warpinski et al., 2003). Using these procedures, velocity models are only calibrated along wave/ray paths from the injection well to the receiver array. Currently, the only process to calibrate away from the known shot locations is first-break travel-time tomography (Bardainne and Gaucher, 2010; Grechka and Yaskevich, 2013), which is not often viable for surface data. Artman et al. (2010) introduce multiple zero-lag passive imaging conditions (IC) and note that consistency of multiple images of the same event (i.e., separate images generated from the P- and S-wave data) indicate correctness of the velocity profiles.
