Pitfalls in crosshole seismic interpretation as a result of 3-D effects

Publisher Website
Google Scholar
Abstract
In surface seismics, much of the reflection data originate from more or less stratified structures with large lateral extensions. Consequently, the data can often be interpreted without introducing serious errors by treating the earth as a 2-D medium. In crosshole reservoir tomography, however, the target of interest is a finite object with geometric features comparable in size to the dominant wavelength of the source signal. Under these conditions, a significant amount of energy can bypass the object under investigation as a result of diffraction. When the velocity of the anomalous body is less than that of the surrounding medium, the bypassed energy can appear as the first arrival in the receiver borehole in accordance with Fermat’s principle. Crosshole seismic response of a number of simple 3-D reservoirs is investigated by finite‐difference models based on the acoustic wave equation. The nature of the signals computed along the receiver borehole confirms the presence of both the transmitted and the bypassed energy. Our observations indicate that unless the two surfaces at which the energy enters and leaves the reservoir are parallel, a 2-D inversion algorithm will lead to incorrect results. Moreover, a velocity anomaly as a result of a steam flood may have negligible effect on the first arrivals because most of the energy bypasses the anomaly in the third dimension. All these conditions point to potential pitfalls in standard 2-D tomographic inversion which considers the first arrivals only. More reliable results can be obtained by simultaneous recording of data along more than one borehole, identifying the bypassed signals, utilizing as many events as possible in the recorded data, and performing model‐driven interpretation.

This publication has 0 references indexed in Scilit: