LEAPS-MPS: Computational Modeling to Characterize and Attribute Uncertainty in Future Coastal Risk

该项目旨在研究沿海地区如何减轻海平面变化和沿海洪水造成的损害。全球海平面上升和风暴加剧给沿海地区的人民和财产带来风险。管理这些风险的策略包括保护措施，如建造海堤，提升现有结构，以及远离海岸的搬迁。然而，不确定性是固有的地球物理过程，数学模型，以及用于校准这些模型的观测数据。 这些建模的不确定性导致不确定性的最佳策略，以抵御沿海灾害。这项研究将评估不同的地球物理和社会经济因素如何导致有效保护沿海地区的决策和成本的不确定性，以及潜在保护不足的沿海资产的估计损害的不确定性。该项目将为学生提供开发软件和进行计算机模型实验的培训机会。这些活动将通过参与项目增强学生对科学的认同感，支持代表性不足的少数群体学生的代表性和坚持性。此外，研究将在卡内基R2大学进行，通过该项目提供的资源将产生积极影响。该研究将通过利用沿海适应决策的数学结构和网格化的全球气候数据，调查海平面变化和沿海影响的模型。机器学习和统计工具将与现有的地球物理和社会经济模型相结合，将沿海适应决策与地球物理过程、气候和社会经济模型以及观测数据中的不确定性联系起来。这一耦合建模框架将作为一个实验室，以表征未来沿海适应成本和决策的不确定性。将使用监督机器学习和全球敏感性分析方法将不确定性分解并归因于风险和社会经济不确定性的地球物理驱动因素。这项研究将评估不同的不确定性在多大程度上影响沿海地区适应海平面变化所带来的风险的最佳战略。由于最优性通常是通过最小化预期损失来定义的，因此将在所采用的模型中实施额外的决策标准，从而能够探索风险规避和不完美信息的作用。一个标准的综合评估模型框架将被使用，这将有助于未来的努力，以扩大这项工作，以检查海平面灾害在气候风险综合评估中的更大作用。这个奖项反映了NSF的法定使命，并已被认为是值得通过使用基金会的智力价值和更广泛的影响审查标准进行评估的支持。

The project aims to examine how coastal areas can mitigate damages from sea-level change and coastal flooding. Rising global sea levels and intensifying storms cause risks for people and property in coastal areas. Strategies to manage these risks include protective measures like building seawalls, elevating existing structures, and relocation away from the coast. However, uncertainty is inherent in the geophysical processes, the mathematical models, and the observational data used to calibrate those models. These modeling uncertainties lead to uncertainty in the optimal strategy to defend against coastal hazards. This research will assess how different geophysical and socioeconomic factors lead to uncertainty in the decisions and costs to effectively protect coastal areas, and uncertainty in the estimated damages from potentially under-protecting coastal assets. The project will provide training opportunities for students to develop software and conduct computer model experiments. These activities will support the representation and persistence of students from underrepresented minority groups by enhancing students’ sense of science identity through engaging in projects. Further, the research will be conducted at a Carnegie R2 university, where the resources made available through this project will have positive impacts. The research will investigate models for sea-level change and coastal impacts by exploiting mathematical structures for coastal adaptation decision-making and gridded global climate data. Machine learning and statistical tools will be integrated with existing geophysical and socioeconomic models, bridging coastal adaptation decisions to uncertainties in geophysical processes, in climate and socioeconomic models, and in observational data. This coupled modeling framework will serve as a laboratory to characterize uncertainty in future coastal adaptation costs and decisions. The uncertainty will be decomposed and attributed to geophysical drivers of risk and socioeconomic uncertainties using supervised machine learning and global sensitivity analysis methods. The research will assess the extent to which different uncertainties affect the optimal strategies for adapting coastal areas to the risks posed by sea-level change. As optimality is typically defined by minimizing expected loss, additional decision-making criteria will be implemented in the models employed, thereby enabling an exploration of the role of risk aversion and imperfect information. A standard modeling framework for integrated assessment will be used, which will facilitate future efforts to expand this work to examine the larger role of sea-level hazards in integrated assessments of climate risks.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.