{"title":"机器学习在岩隙带水文学中的应用:综述","authors":"Xiang Li, John L. Nieber, Vipin Kumar","doi":"10.1002/vzj2.20361","DOIUrl":null,"url":null,"abstract":"Machine learning (ML) has been broadly applied for vadose zone applications in recent years. This article provides a comprehensive review of such developments. ML applications for variables corresponding to different complex vadose zone processes are summarized mostly in a prediction context. By analyzing and assessing these applications, we discovered extensive usages of classic ML models with relatively limited applications of deep learning (DL) approaches in general. We also recognized a lack of benchmark datasets for soil property research as well as limited integration of physics‐based vadose zone principles into the ML approaches. To facilitate this interdisciplinary research of ML in vadose zone characterization and processes, a paradigm of knowledge‐guided machine learning is suggested along with other data‐driven and ML model‐based research suggestions to advance future research.","PeriodicalId":23594,"journal":{"name":"Vadose Zone Journal","volume":null,"pages":null},"PeriodicalIF":2.5000,"publicationDate":"2024-06-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Machine learning applications in vadose zone hydrology: A review\",\"authors\":\"Xiang Li, John L. Nieber, Vipin Kumar\",\"doi\":\"10.1002/vzj2.20361\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"Machine learning (ML) has been broadly applied for vadose zone applications in recent years. This article provides a comprehensive review of such developments. ML applications for variables corresponding to different complex vadose zone processes are summarized mostly in a prediction context. By analyzing and assessing these applications, we discovered extensive usages of classic ML models with relatively limited applications of deep learning (DL) approaches in general. We also recognized a lack of benchmark datasets for soil property research as well as limited integration of physics‐based vadose zone principles into the ML approaches. To facilitate this interdisciplinary research of ML in vadose zone characterization and processes, a paradigm of knowledge‐guided machine learning is suggested along with other data‐driven and ML model‐based research suggestions to advance future research.\",\"PeriodicalId\":23594,\"journal\":{\"name\":\"Vadose Zone Journal\",\"volume\":null,\"pages\":null},\"PeriodicalIF\":2.5000,\"publicationDate\":\"2024-06-14\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Vadose Zone Journal\",\"FirstCategoryId\":\"89\",\"ListUrlMain\":\"https://doi.org/10.1002/vzj2.20361\",\"RegionNum\":3,\"RegionCategory\":\"地球科学\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q3\",\"JCRName\":\"ENVIRONMENTAL SCIENCES\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Vadose Zone Journal","FirstCategoryId":"89","ListUrlMain":"https://doi.org/10.1002/vzj2.20361","RegionNum":3,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q3","JCRName":"ENVIRONMENTAL SCIENCES","Score":null,"Total":0}
引用次数: 0
摘要
近年来,机器学习(ML)已广泛应用于地下蓄水层。本文全面回顾了这些发展。本文主要从预测的角度总结了 ML 在不同复杂渗流带过程变量中的应用。通过分析和评估这些应用,我们发现经典 ML 模型得到了广泛应用,而深度学习(DL)方法的应用则相对有限。我们还发现,土壤属性研究缺乏基准数据集,基于物理学原理的浸润带原理与 ML 方法的结合也很有限。为促进这一跨学科研究,我们提出了知识引导的机器学习范式,以及其他数据驱动和基于 ML 模型的研究建议,以推动未来的研究。
Machine learning applications in vadose zone hydrology: A review
Machine learning (ML) has been broadly applied for vadose zone applications in recent years. This article provides a comprehensive review of such developments. ML applications for variables corresponding to different complex vadose zone processes are summarized mostly in a prediction context. By analyzing and assessing these applications, we discovered extensive usages of classic ML models with relatively limited applications of deep learning (DL) approaches in general. We also recognized a lack of benchmark datasets for soil property research as well as limited integration of physics‐based vadose zone principles into the ML approaches. To facilitate this interdisciplinary research of ML in vadose zone characterization and processes, a paradigm of knowledge‐guided machine learning is suggested along with other data‐driven and ML model‐based research suggestions to advance future research.
期刊介绍:
Vadose Zone Journal is a unique publication outlet for interdisciplinary research and assessment of the vadose zone, the portion of the Critical Zone that comprises the Earth’s critical living surface down to groundwater. It is a peer-reviewed, international journal publishing reviews, original research, and special sections across a wide range of disciplines. Vadose Zone Journal reports fundamental and applied research from disciplinary and multidisciplinary investigations, including assessment and policy analyses, of the mostly unsaturated zone between the soil surface and the groundwater table. The goal is to disseminate information to facilitate science-based decision-making and sustainable management of the vadose zone. Examples of topic areas suitable for VZJ are variably saturated fluid flow, heat and solute transport in granular and fractured media, flow processes in the capillary fringe at or near the water table, water table management, regional and global climate change impacts on the vadose zone, carbon sequestration, design and performance of waste disposal facilities, long-term stewardship of contaminated sites in the vadose zone, biogeochemical transformation processes, microbial processes in shallow and deep formations, bioremediation, and the fate and transport of radionuclides, inorganic and organic chemicals, colloids, viruses, and microorganisms. Articles in VZJ also address yet-to-be-resolved issues, such as how to quantify heterogeneity of subsurface processes and properties, and how to couple physical, chemical, and biological processes across a range of spatial scales from the molecular to the global.