Introduction
A monitoring network composed of a vertical VSP array and a subsurface array can reliably locate microseismic events by using a traditional seismological location method. This distributed array method, which depends only on the first arrival times of P and/or S phases, avoids using timeconsuming, inconvenient and less-reliable measurements of signal azimuth and inclination angle to locate event hypocenters. Numerical analyses of hypocenter uncertainties demonstrate that events located by time-terms alone can achieve better location reliability than hypocenters located by range & bearing methods common to single VSP receiver configurations. The distributed array method is also applicable to those events identified with only S-phase arrivals.
A single, borehole vertical array of 3-component geophones is commonly used to record P- and S-wave signals which yield information of azimuth, angle of inclination and P, S-phase first arrival times of identifiable events (Oye and Roth, 2003). This information is used to estimate hypocenter locations of passive microseisms in oil reservoirs. However, the uncertainty of the hypocenter location using this kind of single VSP array is generally on the order of tens of meters and often requires several additional conditions to be met (Fabriol, 2001; Phillips et al., 1998; Rutledge et al, 1998). Using this method, the measurement of azimuth and angle of inclination of an event from hodogram is often severely affected by the quality of P-phase arrivals such as S/N ratio, impulsivity, and the assumed velocity model (Vadale, 1986). Another weakness of the range & bearing method is that when the S/N ratios of P-phase first arrivals are low and the hodograms are ambiguous (a quite common occurrence in microseismic monitoring), location reliabilities are problematic. Under these conditions, the measurement and evaluation of azimuth and angle of inclination are timeconsuming and variable and are dependent on the time window for measurement, hence highly dependent on the experience and skill-level of the analyst. For many typical microseismic surveys, short-comings of the range & bearing method affect the efficiency and reliability of the data processing and hypocenter location results. In traditional earthquake seismology, hypocenter location methods using P- and/or S-phase first arrivals have long been used and proven to be more accurate and reliable than methods using a combination of azimuth, angle of inclination, and time difference between P- and S-phases. Obviously, the straight line geometry of a single VSP vertical array is poorly suited for the application of the arrival time method. Chen (2007) proposed a similar monitoring method that used multiple VSP arrays. In this presentation, we propose to use subsurface geophone arrays with improved S/N ratio in combination with a VSP vertical array to form a distributed network with geometry suitable for the distributed array hypocenter location method. To examine the reliability of hypocenter event locations recorded by such a network, we have calculated the uncertainties of hypocenter location using several possible array geometries.
To determine the uncertainty of hypocenter location, an error ellipsoid is generally calculated to depict the error distribution.
