Waveform-based source-parameter estimation for elastic full waveform inversion

Estimating source parameters (e.g.,wavelet and source mechanisms) is an important aspect of both active- and passive-source elastic full waveform inversion. Waveform-based source-parameter estimation potentially leads deeper insights into the nature of earthquakes and active sources. Accuracy of source parameters affects estimation of elastic parameter distributions, and vice versa as source-path effects (i.e., a trade-off between velocity and source parameters). We use synthetic and field active-source cross-well datasets, and demonstrate clearly that the source-parameter estimation depends on the quality of the velocity models, geometry of sources and receivers, and available data components. We further demonstrate that incorrect source parameters degrade the quality of elastic models obtained by full waveform inversion.