Leveraging Crowdsourced Data to Assess Spatiotemporal Patterns of Resilience in Diverse Gulf Coast Communities Impacted by Natural Hazards

在美国，洪水事件的严重性和代价继续增加，往往对脆弱人群造成不成比例的影响，这些人对洪水的负面影响可能更敏感，适应能力更低。了解是什么使社区对洪灾变得脆弱或有弹性，对于制定能够减少未来灾害的负面影响的缓解行动至关重要。然而，现有的复原力评估框架往往不足，因为它们没有处理影响人们应对、适应和恢复自然灾害能力的重要的动态和高度局部化的因素。灾难复原力研究补助金(DRRG)项目将通过生成和利用高分辨率的众包数据集，并利用当地的知识和经验来研究影响复原力的因素(即暴露、敏感性和适应能力)如何随空间和时间变化，从而为地方一级的复原力评估开发一个自下而上的、社区驱动的框架。研究结果将对更有效的复原力建设产生影响。作为该项目的一部分，使用社区驱动的测绘平台Streetwyze的众包工作将提高人们对洪水及其在社区中的日常影响的认识，并将鼓励不同的声音参与数据收集，以支持当地的复原力规划工作。该项目将使用混合方法、社会技术的方法来研究如何利用众包数据集来(1)改善影响社区对洪水灾害的复原力的因素的时空特征，(2)制定新的衡量标准，以说明复原力的动态社会和物理性质，以及(3)鼓励更公平的能力建设，以减少未来洪水的影响，并增强不同人群的灾害复原力。该项目重点关注密西西比州沿海的洪水灾害，并与来自湾港和比洛克西市洪水易发地区的不同社区合作伙伴接触。两种新的众包技术，包括被动收集的流动性数据和主动生成的定性和图像数据，将用于表征暴露、敏感性和适应能力的精细空间和时间模式。调查将评估社区成员如何使用众包数据，并将评估人口和社会经济因素在影响参与众包努力方面的作用。将开发和应用地理空间、统计和机器学习模型，将众包数据集与传统传感器、卫星和调查数据相结合。模型输出将被用来开发新的测量方法，以改进对促进社区复原力的因素的评估。该奖项反映了NSF的法定使命，并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，被认为值得支持。

The severity and cost of flood events continue to increase in the US, often with disproportionate impacts on vulnerable populations who may have higher sensitivity to the negative effects of flooding and lower capacity to adapt. Understanding what makes communities vulnerable or resilient to flooding is critical to developing mitigation actions that can reduce the negative effects of future hazards. However, existing frameworks for assessing resilience often fall short, as they do not address important dynamic and highly localized factors that influence peoples’ ability to cope with, adapt to, and recover from natural hazards. This Disaster Resilience Research Grants (DRRG) project will develop a bottom-up, community-driven framework for local-level resilience assessment by generating and utilizing high-resolution crowdsourced datasets and leveraging local knowledge and experiences to examine how the factors contributing to resilience (i.e., exposure, sensitivity, and adaptive capacity) vary over space and time. Findings will have implications for more effective resilience building. As part of the project, crowdsourcing efforts using Streetwyze, a community-driven mapping platform, will increase the awareness of flooding and its daily impacts in communities and will encourage diverse voices to participate in the collection of data to support local resilience planning efforts.This project will use a mixed-methods, sociotechnical approach to examine how crowdsourced datasets can be leveraged to (1) improve the spatiotemporal characterization of factors that influence community resilience to flood disasters, (2) develop new metrics that account for the dynamic social and physical nature of resilience, and (3) encourage more equitable capacity-building to reduce the impacts of future floods and enhance disaster resilience across diverse populations. The project focuses on flood hazards in coastal Mississippi and engages with diverse community partners from flood-prone areas in the cities of Gulfport and Biloxi. Two novel crowdsourcing technologies that include passively collected mobility data and actively generated qualitative and imagery data will be used to characterize fine-scale spatial and temporal patterns of exposure, sensitivity, and adaptive capacity. Surveys will evaluate how community members use the crowdsourced data and will assess the role of demographic and socioeconomic factors in influencing participation in the crowdsourcing effort. Geospatial, statistical, and machine learning models will be developed and applied to integrate the crowdsourced datasets with conventional sensor, satellite, and survey data. Model outputs will be used to develop novel measurement approaches that improve assessments of the factors contributing to community resilience.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.