Elastic full-waveform inversion is an ill-posed and highly non-linear data-fitting procedure that is sensitive to noise, inaccuracies of the starting model and the definition of multi-parameter classes. In this study, we investigate the performances of different minimisation functionals, such as the least-square norm (L2), the least-absolute-values norm (L1), and some combinations of both (the Huber and the so-called Hybrid criteria), with an application to a noisy offshore synthetic data set. The four functionals are implemented in a massively parallel, 2D elastic frequency-domain full-waveform inversion algorithm. Results show that, unlike the L2 norm, the L1 norm, the Huber and the Hybrid criteria allow for successful imaging of VP and VS models from noisy data in soft-seabed environment, where the P-to-S waves have a small footprint in the data. The Huber and the Hybrid criteria appear however to be sensitive to a threshold criterion, which requires tedious trial-and-error investigations for reliable estimation. The L1 norm provides a robust alternative to the L2 norm in the framework of efficient frequency-domain full-waveform inversion where a limited number of frequencies are involved in the inversion.


Article metrics loading...

Loading full text...

Full text loading...

This is a required field
Please enter a valid email address
Approval was a Success
Invalid data
An Error Occurred
Approval was partially successful, following selected items could not be processed due to error