Modeling and Epistemic Uncertainty Analysis of Faults as Conduit-Barriers to Fluid Flow and Salinization in Siliciclastic Aquifer Systems

断层在沉积盆地中作为流体流动的管道和障碍的作用引起了广泛的兴趣。对断层在碳氢化合物的迁移和圈闭、二氧化碳在封存过程中的泄漏以及在浅层含水层系统和深部盆地环境中作为流动的管道和障碍的作用的努力引起了人们的极大兴趣。该项目将在严格分析认识不确定性的背景下，研究断层作为硅质碎屑含水层系统中流体流动和地下水盐化的管道屏障。路易斯安那州巴吞鲁日断裂拥有大量的历史和野外资料，为研究和检验各向异性断层渗透率的概念提供了理想的野外区域。此外，断层在断层以北的一个主要市政和工业淡水来源的盐碱化过程中扮演着一个尚未完全了解的角色。该项目将检验四个假设：(1)Bense和Person(2006)断层渗透率方法可以用来表征断层的水力性质；(2)有可能确定深水含水层沿断层面的垂直泄漏或跨断层的横向泄漏是盐碱化的主要原因；(3)贝叶斯模型平均(BMA)是在模型参数和模型结构中各种认知不确定性来源下量化盐度预测不确定性的有效方法；以及(4)拟议的实验设计是减少盐化预测中认识性不确定性的有效方法。项目成果将大大提高对基于BMA的实验设计在减少盐碱化模型中认知不确定性方面的有效性的理解。该项目还将提高对巴吞鲁日断层在硅质碎屑含水层系统中地下水盐化中的了解。该项目提供了一种手段，用于表征硅质碎屑岩序列中断层的渗透性，并开发技术，以量化和减少流体流动和污染物跨断层和沿断层传输的建模中的不确定性。BMA分析将推动对此类研究中不确定性来源的研究。特别是，该项目将大大提高南山含水层系统流体流动和盐碱化的数值模拟和不确定性分析技术，并将提供该地区地下地质、水文地质和水文地球化学的详细知识。南山含水层系统中的咸水入侵目前是路易斯安那州政府、当地自来水公司和行业面临的一个紧迫问题。含水层系统目前提供高质量的地下水，供四个教区的居民使用，但由于引入咸水，地下水有受到损害的危险。该项目的成果将有助于指导巴吞鲁日地区以及其他类似断层和松散的硅质碎屑含水层系统今后的盐碱化建模和补救工作。拟议的研究对整个社会的一个非常实际的好处将是更好地了解盐碱化的过程和速度，并制定一种补救大巴吞鲁日地区南山含水层系统的方法。

There has been extensive interest in the role of faults as conduits and barriers to fluid flow in sedimentary basins. Much of the interest has been generated by the attempt to understand the role of faults in the migration and entrapment of hydrocarbons, the leakage of carbon dioxide during sequestration, and as conduits and barriers to flow in both shallow aquifer systems and in deep basinal settings. The project will study a fault as a conduit-barrier to fluid flow and groundwater salinization in a siliciclastic aquifer system under the context of a rigorous analysis of epistemic uncertainty. The Baton Rouge fault in Louisiana provides an ideal field area to investigate and test the concept of anisotropic fault permeability because of the large amount of historical and field data which exists for the area. In addition, the fault is playing an incompletely understood role in the salinization of a major municipal and industrial source of fresh water north of the fault. The project will test four hypotheses: (1) that the Bense and Person (2006) fault permeability method can be used to characterize the hydraulic properties of the fault, (2) that it is possible to determine whether vertical leakage up the fault plane from the deep saltwater aquifers or lateral leakage across the fault is the dominant cause of salinization, (3) that the Bayesian model averaging (BMA) is a valid method for quantifying salinity prediction uncertainties under a variety of sources of epistemic uncertainty in model parameters and model structure, and (4) that the proposed experimental designs are a valid approach in reducing epistemic uncertainties in salinization predictions. The project outcomes will significantly increase the understanding of the effectiveness of BMA-based experimental designs for reducing epistemic uncertainties in salinization models. The project will also improve the understanding of the Baton Rouge fault in groundwater salinization in a siliciclastic aquifer system. The project provides a means for characterizing the permeability of faults in siliciclastic sequences and for developing techniques for quantifying and reducing uncertainties in the modeling of fluid flow and transport of contaminants across and along faults. The BMA analysis will advance the study of sources of uncertainty in such studies. In particular, the project will significantly advance techniques in the numerical modeling and uncertainty analysis of fluid flow and salinization of the Southern Hills aquifer system and will provide detailed knowledge of the subsurface geology, hydrogeology, and hydrogeochemistry of the region. Saltwater intrusion in the Southern Hills aquifer system is currently an urgent issue for the Louisiana state government, local water utility companies, and industry. The aquifer system currently provides high-quality groundwater that is utilized by the citizens of a four-parish area, but is in danger of being compromised by the introduction of saline waters. The results of project will help guide future salinization modeling and remediation efforts both in the Baton Rouge area and in other similar faulted and unconsolidated siliciclastic aquifer systems. One very practical benefit of the proposed research to society as a whole will be a better understanding of the processes and rates of salinization and the development of a means of remediation of the Southern Hills aquifer system in the greater Baton Rouge area.