RAPID: Local Residents' Responses to the 2018 Indonesia Earthquake and Tsunami

这个快速反应研究（RAPID）项目通过将社会科学预警研究、交通工程和伤亡建模联系起来，研究与地震和海啸灾害相关的疏散时间估计的基本问题。印度尼西亚帕卢和附近城市东加拉和马木居的这一毁灭性事件对美国研究人员具有重大科学意义，因为它与预测的卡斯卡迪亚俯冲带（CSZ）9.0级地震性质相似。帕卢事件对CSZ的沿海社区具有独特的重要影响。 首先，由7.5级走滑地震产生的20英尺高的海啸类似于太平洋西北（PNW）海岸产生海啸的类似地方来源。了解人们对帕卢地震和海啸的行为反应可以为改善PNW沿海社区和美国其他沿海地区的准备工作提供独特的见解。第二，帕卢事件代表了“最坏的情况”（地震、海啸、大量内陆泥浆流和液化，其中土壤破坏吞噬了1，400所房屋），在这种情况下，事件发生在大量人聚集在海滩参加节日的同时。很明显，由于海啸产生的大量内陆泥浆流，人口稠密的沿海地区遭受了严重破坏。前一次地震引起的强烈地面震动可能引发了土壤液化和侧向扩展。这些组合过程在溪流附近和填海土地上也很常见，这些过程可能会在预测的CSZ事件中在PNW海岸相对平坦的地区产生破坏性的泥石流。第三，帕卢独特的地理位置位于一个狭窄的10公里长的手指状海湾的尽头，这可能会放大相对较浅的海底快速移动的水和波浪能量，加剧海啸对受影响人民和社区的破坏力。在PNW海岸上有类似的社区位于这种类型的海湾。最后，许多太平洋西北沿岸社区也同样缺乏海啸防备。在CSZ，近场海啸的威胁直到最近才被认识到，因此许多社区的疏散计划都是基于远场海啸的情景，提供4-6小时的预警。以目前的地震和海啸预警技术很难预测这类事件，因此必须从帕卢事件中吸取教训。此外，对帕卢海啸疏散的研究将为其他类似类型的突发灾害（如山洪暴发）的疏散规划提供信息。 本项目以保护行动决策模型（PADM）为指导，收集人们对帕卢地震和海啸的行为反应的经验数据。问题包括：有多少人认为地震震动是海啸发生的环境线索？有多少人收到了来自官员、新闻媒体和同行的警告？人们如何应对威胁？ 本项目采用了以前在美属萨摩亚（2009年）、基督城（2011年）、东北（2011年）、库克海峡（2013年）和格拉斯米尔湖（2013年）使用的经过验证的调查程序。收集的数据集中在（1）官员决定发布海啸疏散警报所需的时间;（2）海啸警报的来源、渠道、信息和警报发布时间;（3）人们的疏散参与率、准备时间和出发时间;（4）海啸警报的发布时间。以及（4）该奖项反映了NSF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。

This Rapid Response Research (RAPID) project examines fundamental issues on the evacuation time estimates associated with earthquake and tsunami hazards by linking social science warning research, transportation engineering, and casualty modeling. This devastating event to Palu and the nearby cities of Donggala and Mamuju, Indonesia is of major scientific relevance to American researchers because it is similar in nature to the predicted M9.0 Cascadia Subduction Zone (CSZ) earthquake. The Palu event has uniquely important implications to the coastal communities in the CSZ. First, the 20-ft tsunami generated by a M7.5 strike-slip earthquake resembles a similar local source of tsunami generation on the Pacific Northwest (PNW) coast. Understanding people's behavioral responses to the Palu earthquake and tsunami can provide unique insights to improve the preparedness of PNW coastal communities and other coastal areas of United States. Second, the Palu event represents a "worst case scenario" (earthquake, tsunami, large inland flows of mud, and liquefaction, where soil failure swallowed 1,400 houses) in which the event happened at the same time a large number of people had congregated on a beach for a festival. It is clear that the densely populated coastal areas took heavy damage because of tsunami-generated large inland flows of mud. Intense ground shaking from the preceding earthquake may have triggered soil liquefaction and lateral spreading. These combined processes are also common near streams and on reclaimed land that can produce destructive mudslides in relatively flat areas on the PNW coast from the predicted CSZ event. Third, the unique geographic location of Palu sitting at the end of a narrow 10km-long finger-like bay likely amplified the fast-moving surge of water and wave energy on the relatively shallower seabed, exacerbating the damaging power of the tsunami on the affected people and communities. There are similar communities on the PNW coast located on this type of bay. Lastly, many PNW coastal communities suffer a similar lack of tsunami preparedness. In the CSZ, the threat of near-field tsunamis has been recognized only recently, so many communities have based their evacuation plans on far-field tsunami scenarios that provide 4-6 hours of forewarning. The difficulty of predicting such events with the current technology for earthquake and tsunami warning makes it essential to learn from the Palu event. In addition, a study of Palu tsunami evacuations will inform evacuation planning for other similar type rapid-onset disasters such as flash floods. This project uses the Protective Action Decision Model (PADM) as a guide to collecting empirical data on people's behavioral responses to the Palu earthquake and tsunami. Questions include: How many people recognized the earthquake shaking as an environmental cue for a tsunami onset? How many people received warnings from officials, news media, and peers? How did people respond to the threat? This project uses a validated survey procedure previously used in American Samoa (2009), Christchurch (2011), Tohoku (2011), Cook Strait (2013), and Lake Grassmere (2013). Collected data focus on (1) the amount of time it took officials to decide to issue tsunami evacuation warnings; (2) the tsunami warning sources, channels, messages and warning dissemination times; (3) people's evacuation participation rates, preparation times, and departure times; and (4) people's evacuation logistics.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.