Subsurface Property

Subsurface property characterization aims to accurately determine the physical properties of materials beneath the Earth's surface, crucial for applications like carbon capture and storage, resource exploration, and environmental monitoring. Current research heavily utilizes machine learning, particularly deep learning models like convolutional neural networks (CNNs), recurrent neural networks (RNNs), and generative adversarial networks (GANs), often coupled with Bayesian inversion techniques to improve accuracy and handle uncertainty in sparse datasets. These advancements enable more efficient and reliable estimations of properties such as porosity, permeability, and fluid content, improving decision-making in various fields and offering insights into complex subsurface processes. The integration of crowdsourced data and the development of comprehensive benchmark datasets are also contributing to the field's progress.