Due to the enormous advances in computational power, machine-learning algorithms have recently experienced significant breakthroughs in handling and processing complex and big data. Water resources problems have always been among the issues that researchers have attempted to leverage such advances for various purposes, such as hydrological modeling, fluid flow and transport in porous media, and characterization of geo-systems. As such, there is an overwhelming interest in developing machine-learning techniques for taking advantage of the big data that are becoming increasingly available from various sources, such as satellite images, sensors, drones, geophysical data, pore-scale imaging, etc., and discovering the relationships between the important variables. In this special issue, researchers that are active in this emerging area present their work on the use of machine-learning algorithms, together with big data for addressing critical issues in water resources and similar (sub)surface problems. In particular, the emphasis is on works that are related to novel applications of machine learning for solving the problems that either have never been explored in water resources, or those that require considerable computational power using the traditional methods. In addition, approaches that are based on the combination of physics-based modeling of geo-systems and machine learning are particularly of interest. The results reported by the papers published in this special issue have enriched our understating of important phenomena related to various issues regarding water in geo-systems, and have presented new insights into the applications of machine learning and data-driven techniques for the aforementioned problems.

由于计算能力的巨大进步，机器学习算法最近在处理和处理复杂的大数据方面经历了重大突破。水资源问题一直是研究人员试图将这些进展用于各种目的的问题之一，例如水文建模，流体在多孔介质中的流动和运输以及地质系统的表征。因此，人们对开发利用机器学习技术以利用大数据的兴趣非常浓厚，这些技术已从各种来源（例如卫星图像，传感器，无人机，地球物理数据，孔隙尺度成像等）越来越多地获得，并发现重要变量之间的关系。在本期特刊中，活跃在这个新兴领域的研究人员展示了他们在使用机器学习算法以及大数据方面的工作，这些数据用于解决水资源中的关键问题和类似的（地下）地表问题。特别地，重点在于与机器学习的新颖应用相关的作品，以解决水资源中从未探索过的问题，或者使用传统方法需要大量计算能力的问题。此外，基于物理的地理系统建模与机器学习相结合的方法特别受关注。本期特刊发表的论文所报告的结果丰富了我们对与地质系统中与水有关的各种问题相关的重要现象的低估，