Hazards SEES Type 2: Hazard Prediction and Communication Dynamics in the Modern Information Environment

当当今世界面临危险威胁时，信息比比皆是。科学预测正在迅速改进，信息和通信技术极大地改变了人们获取、组合和共享信息的方式。然而，现有的备灾和预警范例为理解如何传达和使用预报和预警信息提供了有限的知识和工具。为了解决这一差距，该项目将在不断发展的气象预报和现代信息传递系统的背景下，调查不同来源和类型的信息之间的相互作用如何影响飓风逼近和登陆时的风险解释和行为反应。该项目汇集了计算机和信息科学、大气和相关科学以及社会和行为科学的概念、方法和专业知识，以解决三个研究问题：(1)随着飓风的逼近，有关飓风风险的不断演变的信息如何通过信息系统传播？(2)现代信息环境中的风险沟通如何缓解或加剧社会对飓风的脆弱性？尤其是对最脆弱的公众？(3)鉴于科学和技术的不断发展，如何改进飓风预报和预警信息及其沟通？这些问题将通过将对现实世界危险信息系统的研究与计算物理和社会建模相结合来进行研究。研究人员将从社交媒体流、访谈和焦点小组中收集和分析数据，其中包括对更脆弱人群的强调。这项研究将与高分辨率组合飓风和风暴潮模拟以及在计算实验室进行的实验相结合，该实验室将飓风和风暴潮模拟与追求、处理和传输此类信息的社会参与者的基于代理的建模相结合。这一努力的智力价值在于推进每个子学科的基本知识，并在各学科之间建立新的联系，以解决灾害研究中的综合问题。社交媒体、访谈和焦点小组研究将建立对数字和非数字领域中不同来源的危险信息如何被访问、组合、解释和采取行动的新理解。这项工作还将通过结合几个领域的理论来促进脆弱性科学，以检查随着灾害的逼近，风险传播、信息和通信技术以及社会网络如何影响脆弱性和适应能力。相关的天气和风暴潮模拟研究将建立对耦合的概率飓风和风暴潮预报以及风暴潮在整个提前期的可预测性的基本理解。从该模型的输出中得出的情景将被用于开发以新的方式结合地理空间、物理和社会信息的视觉风险沟通原型。基于飓风/智能体的耦合建模实验将使人们能够在受控环境下对信息系统动力学和紧急行为进行新的探索。通过跨研究组成部分的整合，该项目将建立对物理科学、社会和行为科学以及以人为中心的计算如何结合在一起以改变灾害风险沟通并帮助防止灾害成为灾难的新理解。相对于更广泛的影响，该项目的发现预计将对业务天气和风暴潮预报员、公职人员、紧急情况从业者、媒体人员和公众有用，以改进他们创建和传达有关飓风风险逼近的信息的方式。这些改进将推动采取行动，减少飓风的危害，增强能力，减少脆弱性，并随着时间的推移有助于提高可持续性。这些发现也将适用于其他有一定预测能力的危险。为了促进调查结果的应用，该项目包括与飓风预报、应急管理、媒体和数字志愿者社区的从业者合作伙伴的合作。该项目还将对高级和博士后研究人员和研究生进行跨学科途径和方法的交叉培训，帮助建立一支能够解决灾害复原力和可持续性等复杂问题的跨学科队伍。

When hazards threaten in today's world, information abounds. Scientific predictions are improving rapidly, and information and communication technology has dramatically changed how people access, combine, and share information. Yet existing paradigms for hazard preparedness and warning provide limited knowledge and tools for understanding how forecast and warning information is communicated and used. To address this gap, this project will investigate how interactions among different sources and types of information influence risk interpretations and behavioral responses as a hurricane approaches and comes onshore, in the context of evolving meteorological predictions and modern information-delivery systems. This project brings together concepts, methods, and expertise from computer and information science, atmospheric and related sciences, and social and behavioral sciences to address three research questions: (1) How does evolving information about hurricane risk propagate through the information system as a hurricane approaches? (2) How does communication of risk in the modern information environment alleviate or exacerbate societal vulnerability to hurricanes in general, and of the most vulnerable members of the public in particular?; and (3) Given ongoing developments in science and technology, how can hurricane forecast and warning information and its communication be improved? These questions will be investigated by integrating the study of the real-world hazard information system with computational physical and social modeling. Researchers will collect and analyze data from social media streams, interviews, and focus groups that include an emphasis on more vulnerable populations. This research will be combined with high-resolution ensemble hurricane and storm-surge modeling, as well as experiments in a computational laboratory that couples the hurricane and storm surge modeling with agent-based modeling of social actors who pursue, process, and transmit such information.The intellectual merit of this effort rests both in advancement of fundamental knowledge in each sub-discipline and in forging new connections among disciplines to address integrative questions in hazards research. The social media, interview, and focus group research will build new understanding of how hazard information from different sources, in the digital and non-digital spheres, is accessed, combined, interpreted, and acted upon. This work will also advance vulnerability science by combining theory from several fields to examine how risk communication, information and communication technology, and social networks influence vulnerability and adaptive capacities as a hazard approaches. Associated weather and storm surge modeling research will build fundamental understanding about coupled probabilistic hurricane and storm surge forecasting and the predictability of storm surge across lead times. Scenarios derived from the output of this modeling will be used to develop visual risk communication prototypes that combine geospatial physical and societal information in new ways. Coupled hurricane/agent-based modeling experiments will enable a novel exploration of information system dynamics and emergent behaviors in a controlled setting. Through integration across research components, this project will build new understanding of how physical sciences, social and behavioral sciences, and human-centered computing can join to transform hazard risk communication and help prevent hazards from becoming disasters.Relative to Broader Impacts, findings from this project are expected to be of utility to operational weather and storm surge forecasters, public officials, emergency practitioners, media personnel, and members of the public to improve how they create and communicate information about approaching hurricane risks. These improvements will motivate actions to reduce harm from hurricanes, enhance capacities, reduce vulnerabilities, and over time contribute to increased sustainability. These findings will also be applicable to other hazards for which there is some predictive skill. To promote application of findings, the project includes collaborations with practitioner partners in the hurricane forecasting, emergency management, media, and digital volunteer communities. The project will also cross-train senior and postdoctoral researchers and graduate students in interdisciplinary approaches and methods, helping to build an interdisciplinary workforce capable of solving the complex issues of hazard resilience and sustainability.