On the extraction of angle dependent wavelets from synthetic shear wave sonic logs

David Cho, Craig A. Coulombe, Gary F. Margrave

The extraction of angle dependent wavelets requires the use of a shear wave sonic log. However, shear wave measurements are often not acquired in a conventional logging suite and must be estimated to produce a synthetic result. The errors associated with the synthetic shear propagate through to the angle dependent reflectivity with a sine squared dependence of the incidence angle. Therefore, the reflectivity becomes unreliable at larger angles and a least squares extraction using the convolutional model could yield erroneous results. To reduce the errors associated with wavelet extraction at larger angles, a near angle wavelet was estimated with an acceptable amount of error using a least squares approach. Subsequently, an estimate of the angle dependent wavelet amplitude spectrum and a constant Q attenuation model was used to evolve the amplitude and phase respectively to estimate the wavelets at larger angles.