Velocity continuation with Fourier neural operators for accelerated uncertainty quantification

TL;DR

This paper introduces a Fourier neural operator surrogate for velocity continuation in seismic imaging, significantly reducing computational costs in uncertainty quantification by accurately mapping seismic images across different background models.

Contribution

The work presents a novel Fourier neural operator surrogate trained on limited data to efficiently predict seismic images for new background models, accelerating uncertainty quantification.

Findings

Surrogate accurately predicts seismic images for new models.

Method reduces computational costs in uncertainty analysis.

Demonstrated effectiveness on realistic seismic data.

Abstract

Seismic imaging is an ill-posed inverse problem that is challenged by noisy data and modeling inaccuracies -- due to errors in the background squared-slowness model. Uncertainty quantification is essential for determining how variability in the background models affects seismic imaging. Due to the costs associated with the forward Born modeling operator as well as the high dimensionality of seismic images, quantification of uncertainty is computationally expensive. As such, the main contribution of this work is a survey-specific Fourier neural operator surrogate to velocity continuation that maps seismic images associated with one background model to another virtually for free. While being trained with only 200 background and seismic image pairs, this surrogate is able to accurately predict seismic images associated with new background models, thus accelerating seismic imaging uncertainty quantification. We support our method with a realistic data example in which we quantify seismic imaging uncertainties using a Fourier neural operator surrogate, illustrating how variations in background models affect the position of reflectors in a seismic image.