RAPID: Reconstruction of Hurricane Florence Flood Hydrographs (HF2Hs) for South Carolina's Critical Infrastructures Using Data Analytics Algorithms and In-situ Field Measurements

在飓风佛罗伦萨袭击南卡罗来纳州之后，本研究旨在收集洪水淹没/受损关键基础设施的高水位（HWMs）数据，以及来自交通摄像头和社交媒体的易损图像和视频片段。然后，调查人员将使用数据分析算法和HWMs数据重建佛罗伦萨飓风洪水曲线（HF2Hs），以估计洪水高度和覆盖道路和桥梁的淹没程度。以南卡罗来纳东部地区为例，该RAPID项目将解决以下问题：在关键基础设施上重建的洪水水文是否能对洪水阈值和频率提供有价值的见解？如果有，怎么做？为了解决这些问题，该团队由具有工程水文学、计算机科学和工程专业知识的成员组成，他们的定位是提供所需的易腐数据集的收集、检查和存档。收集易腐数据的方法结合了通过使用传统（卷尺、工程师规则等）和数据分析技术来加强易腐数据收集的更广泛目标，这两种技术都依赖于及时收集数据。重建的覆盖路线/道路和桥梁的洪水曲线将有助于了解关键基础设施如何应对飓风引发的洪水，这些洪水在全球许多地区持续存在广泛的挑战。收集到的数据将有助于开发新的洪水预测数值模型，这些模型将处理美国独特的需求和概念美国东南部集水区（浅含水层参数化）。数据分析算法的目标是灵活和可扩展，以收集和分析将通过开源公共存储库（例如GitHub）传播的大型数据集。数据的收集和整合旨在促进决策者和以技术为重点的机构之间的沟通/协作。该项目旨在对南卡罗来纳州产生直接影响，该州非常容易受到反复飓风事件的影响，并且面临日益增加的洪水威胁。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。

In the wake of Hurricane Florence in South Carolina, this research aims to collect high water marks (HWMs) data across flooded/damaged critical infrastructures, and perishable images and video footage from traffic cameras and social media outlets. The investigators will then reconstruct Hurricane Florence flood hydrographs (HF2Hs) using data analytics algorithms as well as HWMs data to estimate flood elevation and inundation extent over overtopped roads and bridges. Using the eastern portion of South Carolina (SC) as a case study, this RAPID project will address the following questions: Do reconstructed flood hydrographs over critical infrastructures provide valuable insight into flooding thresholds and frequencies? If so, how? To address these questions, the team consists of members with expertise in engineering hydrology and computer sciences and engineering who are positioned to deliver the needed collecting, examining, and archiving of perishable datasets. The methodology for collecting perishable data merges the broader objectives of enhancing perishable data collection through the use of traditional (tape measure, engineer's rule, etc.) and data analytics techniques, both of which depend on the timely collection of data. The reconstructed flood hydrographs for overtopped routes/roads and bridges will help understanding of how critical infrastructures respond to hurricane-induced flooding that presents persistent widespread challenges in many regions worldwide. The collected data will benefit the development of new numerical models for flood prediction that will deal with the unique needs and concepts of the U.S.'s southeast catchments (shallow aquifer parameterization). The data analytics algorithm is targeted be flexible and scalable to collect and analyze large sets of data which will be disseminated through open-source public repositories (e.g., GitHub). The collection and integration of data is targeted to facilitate communication/ collaboration between decision makers and technically-focused institutions. This project is intended to have an immediate impact on South Carolina, a state which is very vulnerable to repeated hurricane events and is under the threat of increasing floods.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.