Inversion of acoustic waveforms for velocity and attenuation

In this dissertation, I present some new schemes and techniques for processing of acoustic waveform data. New objective functions are developed for waveform inversion. I use synthetic acoustic borehole waveform data to demonstrate their novel features. These schemes require neither prior deconvolution nor knowledge of the source-receiver wavelet. The most powerful of them is the fourwise processor, which is applicable to data sets from multiple shots and receivers even when each shot has a different unknown signature and each receiver has a different unknown impulse response. I apply this scheme to inversion for shear wave velocity from the ODP sonic data measured by an LSS tool, containing two uncalibrated transmitters and receivers. A 3-D search of waveform inversion gives stable results of formation S-wave velocity inverted from the monopole sonic data in a slow formation.; In addition, I develop new techniques to extract P- and S-wave attenuation from sonic logging data. The success of the techniques lies in the separation of attenuation-related wave attributes from other effects unrelated to attenuation. The wave attributes related to borehole diameter, formation density, and velocity changes are removed by use of synthetic sonograms. The novelty of the techniques is first to use a single receiver data set to extract a relative attenuation profile, then to correct it to absolute attenuation using multiple receiver data sets. The techniques are applicable to both monopole and dipole waveform log data for either P- or S-wave attenuation.