Microseismic Event Relocation Using Arrival Times and Cross-correlation Coefficients – A Synthetic Study

Abstract

We present a feasibility study of seismic event location by jointly inverting P and S arrival times and values of maximum cross-correlation coefficients. This work is motivated by the well-known problem that the location by arrival times alone is occasionally only weakly constraint, and inaccurate measurements of arrival times may introduce significant errors in the event location. This in turn degrades, e.g., the mapping of fracture geometries in reservoirs by passive seismic monitoring. We introduce and test an algorithm, that inverts both measured arrival times and cross-correlation values for the hypocenter coordinates. For our test datasets, including realistic simulations of arrival time errors and uncertainties in the relation between waveform similarity and event separation, the source geometry is better recovered compared to the purely arrival time-based location. For high-quality manual picks, the accuracy of the relocated events is improved by more than 20%. In particular, outliers in the event locations can be effectively corrected.