CAREER: Predicting Transport, Mixing, and Reaction in Three-dimensional Heterogeneous Fractured Media Across Scales

全球约99%的未冻淡水储存在地下水系统中，地球上约75%的近地表地下水蓄水层已经破裂。最近的现场研究表明，地下水中存在新出现的污染物，如微塑料、病毒和有害细菌，而裂缝通常是污染物以异常快的速度长距离迁移的主要途径。然而，传统的地下水模型忽略了裂缝流动，因此往往无法预测污染物在地下的运移。为了确保水资源的可持续利用，迫切需要一种新的模型来预测裂缝介质中的流动和输运。该项目提出了一个综合的研究、教育和推广计划，将推进目前裂缝介质建模和裂缝岩石水文地质学教学的实践。该项目将首先结合视觉实验室实验和直接数值模拟来阐明在单个裂缝尺度下控制流动和输送的基本物理定律。确定的关键单裂缝规模过程将被纳入现场规模模型。最后，开发的模型将通过裂缝含水层现场实验进行验证。该项目有三个主要的外展和教学活动，与拟议的研究直接协同：(i)将成立一个专业的水文地质学家工作组，以弥合学术界、工业界和政府机构在裂隙岩石水文地质学领域的差距；（ii）将为K-12和大学水平的教育开发可视化地下水教学工具；最后（iii）将为新的城市实地课程开发基于受污染的裂隙含水层场地的可访问模块。传统的地下水模型通常将含水层视为二维连续多孔介质，从而忽略了控制裂缝含水层流动和输送的关键复杂性。最近的研究表明，复杂的三维（3D）流动可以改变整个混合和反应动力学，直至现场规模，但潜在的过程以及它们如何在裂缝网络中扩大，仍然没有得到很好的理解。该项目将建立对单裂缝尺度下三维流动对输运、混合和反应（TMR）影响的机理理解，并将这些知识转化为跨尺度的知识，以预测裂缝网络尺度上的过程。该项目首先结合了视觉实验室实验和直接数值模拟，以阐明裂缝非均质性（例如裂缝粗糙度和孔径可变性）和流动边界条件（雷诺数）之间的相互作用如何控制单裂缝尺度下的三维流动和TMR。然后将单裂缝尺度过程整合到裂缝网络尺度模型中，并开发出预测TMR的升级建模框架。将生成一个全面的数据集，并使用基于机器学习的方法进行分析，以生成一个强大的地图，将放大模型参数与关键介质特性和流动条件联系起来。开发的升级模型将通过裂缝含水层现场可控示踪试验进行验证。总的来说，该项目将改变我们对裂缝介质的理解和建模方式，并将为生物地球化学和流体流动耦合建模提供新的基础。该提案由地球科学部的水文科学和教育与人力资源项目共同资助。该奖项反映了美国国家科学基金会的法定使命，并通过使用基金会的知识价值和更广泛的影响审查标准进行评估，被认为值得支持。

About 99% of global unfrozen freshwater is stored in groundwater systems, and about 75% of near-surface groundwater aquifers on earth are fractured. Recent field studies have shown the presence of emerging contaminants such as microplastics, viruses, and harmful bacteria in groundwater, and fractures are often major pathways through which contaminants migrate long distances at anomalously fast rates. However, conventional groundwater models ignore fracture flow, and as a result, often fail to predict contaminant transport in the subsurface. To secure sustainable water resources, there is urgent need for a new model that predicts flow and transport in fractured media. This project presents an integrated research, education, and outreach plan that will advance the current practices of modeling fractured media and teaching fractured rock hydrogeology. This project will first combine visual laboratory experiments and direct numerical simulations to elucidate fundamental physical laws that govern flow and transport at single fracture scales. The identified key single-fracture scale processes will then be incorporated into field-scale models. Finally, the developed models will be validated through field experiments at a fractured aquifer site. The project has three major outreach and teaching activities that are directly synergistic with the proposed research: (i) a professional hydrogeologists’ working group will be formed to bridge the gap between academia, industry, and government agencies in the area of fractured rock hydrogeology, (ii) visual groundwater teaching tools will be developed for K-12 and college-level education, and lastly (iii) accessible modules based on a contaminated fractured aquifer site will be developed for a new urban field course.Conventional groundwater models often treat aquifers as two-dimensional continuous porous media, thereby missing critical complexities that govern flow and transport in fractured aquifers. Recent studies have shown that complex three-dimensional (3D) flows can alter the overall mixing and reaction dynamics up to the field scale, but the underlying processes and how they scale up in fracture networks are still not well understood. This project will establish a mechanistic understanding of 3D flow effects on Transport, Mixing, and Reaction (TMR) at single-fracture scales and translate that knowledge across scales to predict processes at fracture network scales. This project first combines visual laboratory experiments and direct numerical simulations to elucidate how the interplay between fracture heterogeneity (e.g., fracture roughness and aperture variability) and flow boundary conditions (Reynolds number) controls 3D flows and TMR at single fracture scales. The single-fracture scale processes will then be incorporated into fracture network-scale models, and an upscaled modeling framework that predicts TMR will be developed. A comprehensive data set will be generated and analyzed with machine-learning-based methods to yield a powerful map that links upscaled model parameters to key medium properties and flow conditions. The developed upscaled model will be validated through controlled field tracer experiments at a fractured aquifer site. Overall, this project will transform how we understand and model fractured media and will provide a new foundation for modeling coupled biogeochemistry and fluid flow.This proposal is co-funded by the Hydrologic Sciences and Education and Human Resources programs in the Division of Earth Sciences.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.