Computing Location Uncertainty for Imaged Sources
Event name: Fifth EAGE Passive Seismic Workshop
Session: Integrated Interpretation
Publication date: 28 September 2014
Info: Extended abstract, PDF ( 1.34Mb )
Price: € 20
A method is presented for establishing the precision of microseismic location estimates derived from an imaging -based location algorithm. The technique is based upon calculating the probability density function (PDF) for the stack amplitude derived at the computed point source position. This allows construction of an uncertainty volume where contours of the values enclose regions that are X% likely to contain the source, where X is the chosen certainty level. The procedure provides a data derived estimate of location precision rather than assuming some precision a priori. Furthermore it does not require a visible signal in the data, which makes it particularly suited to imaging based location algorithms applied to surface microseismic data. This advantage is extended by the applicability of the method in cases where non-linear stacking operators (such as semblance) are being used. The method does not make any assumptions regarding the shape of the error surface, thus event uncertainty can be computed even in cases where the error surface is discontinuous. In this paper the method is applied to a synthetic surface dataset designed to mimic the results of a hydraulic fracturing.