This paper presents a comprehensive comparison of various machine learning models, namely U-Net, U-Net integrated with Vision Transformers (ViT), and Fourier Neural Operator (FNO), for time-dependent forward modelling in groundwater systems. Through testing on synthetic datasets, it is demonstrated that U-Net and U-Net + ViT models outperform FNO in accuracy and efficiency, especially in sparse data scenarios. These findings underscore the potential of U-Net-based models for groundwater modelling in real-world applications where data scarcity is prevalent.
@article{arxiv.2307.04010,
title = {Understanding the Efficacy of U-Net & Vision Transformer for Groundwater Numerical Modelling},
author = {Maria Luisa Taccari and Oded Ovadia and He Wang and Adar Kahana and Xiaohui Chen and Peter K. Jimack},
journal= {arXiv preprint arXiv:2307.04010},
year = {2023}
}