CBET-EPSRC Efficient Surrogate Modeling for Sustainable Management of Complex Seawater Intrusion-Impacted Aquifers

人口稠密沿海地区的水管理是全球最紧迫的可持续性挑战之一。由于海水可能侵入地下水含水层，沿海地下水特别容易受到气候变化和海平面上升的影响。海水入侵减少了美国所有沿海地区的供水。这给社会带来了高昂的代价。受海水入侵影响的地下水需要昂贵的脱盐过程才能饮用，而灌溉用水可能无法使用，导致农田被废弃。未来的气候预测表明，海水入侵的问题将会恶化。然而，问题的规模尚不清楚，因此很难制定应对措施。虽然沿海地下水含水层的计算机模型可以用于预测海水入侵，但这些建模工作对即使是最快的计算机的能力也提出了挑战。我们建议通过开发比当前模型快几个数量级的模型来解决这一挑战。这将允许更广泛地考虑潜在的解决方案。这些建模方面的进展将与供水机构合作进行，目的是增加沿海社区地下水建模的效用。这些办法的成功发展和采用将有助于负责保护沿海含水层的机构制定可持续的管理战略，以保护稀少的水资源。海水入侵问题的解决方案包括更有效的抽水方案、减少需求和技术干预（如海水淡化）。然而，确定这些问题的最优解决方案会带来极大的计算需求。该项目将通过开发计算效率高、鲁棒性强、精度高的沿海地下水系统替代模型，极大地推动模拟优化（SO）的发展和应用。关于海水入侵问题的SO和代理模型的有限文献主要集中在简化的水文地质环境和管理策略的数学表示上。然而，现实的海水入侵问题涉及水文地质的复杂性，包括离散的岩性、断层和裂缝以及咸水-淡水混合带动力学。解决方案需要非线性目标函数和连续和离散决策变量，代表了广泛的工程部件。我们假设，这些水文地质和管理特征决定了准确和有效的替代模型的建立，而海水入侵问题的准确替代SO模型至少可以比全尺寸模型快一个数量级。计算工作量的减少将使我们能够研究更广泛的潜在海平面上升和气候变化影响，以及对这些影响的更广泛的潜在管理响应。为了实现这一目标，具体的项目目标是：i)开发SO测试问题，以提供对模型替代品的稳健评估；Ii)基于对测试用例含水层的管理，制定管理目标和约束，识别与测试用例相关的场景；iii)规划、培训和评估数据驱动的绩效。和模型驱动的?为测试案例含水层、一系列海平面速率、气候和地下水需求情景确定最佳管理方案的替代方案。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。

Water management in densely populated coastal regions is one of the most pressing sustainability challenges worldwide. Coastal groundwater is especially vulnerable to climate change and sea level rise due to the potential for seawater intrusion into groundwater aquifers. Seawater intrusion has reduced water supply in all coastal regions of the US. This has resulted in high costs to society. Groundwater affected by seawater intrusion requires expensive desalination processes to be made drinkable, while irrigation water could be rendered unusable leading to the abandonment of farmland. Future climate projections suggest the problem of seawater intrusion will worsen. However, the scale of the problem is unclear, making it difficult to devise responses. While computer models of coastal groundwater aquifers can be useful for predicting seawater intrusion, these modeling efforts challenge the capability of even the fastest computers. We propose to address this challenge by developing models that are orders of magnitude faster than current models. This will allow for a much broader consideration of potential solutions. These modeling advances will be made in collaboration with water supply agencies, with the goal of increasing the utility of groundwater modeling for coastal communities. Successful development and adoption of these approaches will help agencies tasked with the protection of coastal aquifers devise sustainable management strategies to protect scarce water resources.Solutions to seawater intrusion problems involve combinations of more efficient pumping schemes, demand reduction, and technological interventions such as desalination. However, determining optimal solutions for these problems poses extreme computational demands. This project will greatly advance the development and application of simulation-optimization (SO) by developing computationally efficient, robust, and accurate surrogate models for coastal groundwater systems. The limited literature on SO and surrogate modeling in seawater intrusion problems has focused on simplified hydrogeological settings and mathematical representations of management strategies. However, realistic seawater intrusion problems involve hydrogeological complexities, including discrete lithological facies, faults and fractures, and saltwater-freshwater mixing zone dynamics. Solutions necessitate nonlinear objective functions and continuous and discrete decision variables, representing a wide range of engineering components. We hypothesize that these hydrogeologic and management features determine the building of accurate and efficient surrogates, and accurate surrogate SO models for seawater intrusion problems can be at least an order of magnitude faster than full-scale models. The reduction in computational effort will allow us to investigate a broader range of potential sea level rise and climate change impacts and a wider range of potential management responses to these impacts. To achieve this goal, the specific project objectives are to: i) develop SO test problems to provide robust evaluation of model surrogates; ii) formulate management objectives and constraints based on management of the test case aquifers, and identify scenarios relevant to the test cases; and iii) program, train, and evaluate the performance of ?data-driven? and ?model-driven? surrogates to identify optimal management schemes for the test case aquifers, a range of sea level rates, climatology, and groundwater demand scenarios.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.