Generating Realistic Geology Conditioned on Physical Measurements with Generative Adversarial Networks

TL;DR

This paper introduces a novel approach using Generative Adversarial Networks for realistic geological modeling conditioned on sparse physical measurements, capturing uncertainty and pattern diversity effectively.

Contribution

It demonstrates how semantic inpainting with GANs can generate diverse, realistic geological realizations that honor measurements and scale efficiently with data size.

Findings

Generates varied geological patterns matching physical data

Scales well with the number of data points

Produces state-of-the-art conditional samples

Abstract

An important problem in geostatistics is to build models of the subsurface of the Earth given physical measurements at sparse spatial locations. Typically, this is done using spatial interpolation methods or by reproducing patterns from a reference image. However, these algorithms fail to produce realistic patterns and do not exhibit the wide range of uncertainty inherent in the prediction of geology. In this paper, we show how semantic inpainting with Generative Adversarial Networks can be used to generate varied realizations of geology which honor physical measurements while matching the expected geological patterns. In contrast to other algorithms, our method scales well with the number of data points and mimics a distribution of patterns as opposed to a single pattern or image. The generated conditional samples are state of the art.