At its core, when searching for the best fit model to predict seismic measurements, XWITM swaps a conventional point-to-point misfit measure to one based on an adaptive cost function (Warner and Guasch, 2016). This gives a superior quality gradient or update direction in terms of pointing towards the global minimum misfit minimiser. Further it gives longer length scales deeper down, in particular length scales which setting derivative constraints can latch onto for flooding large scale structures like salt. The results below show the combination of an adaptive cost function (AWITM) with spatial derivative constraints (CWI).
XWITM with directional derivative constraints
Automated salt flooding leading to global minimum convergence
XWITM without directional derivative constraints
Unable to recover salt body leading to local minimum convergence
Run Full Waveform Inversion on the Cloud
Use XWITM on AWS to discover a step change in accuracy and resolution of your velocity model.