Abstract: In a rapidly globalizing world, disasters and the way in which they are managed are changing. Social media, in conjunction with other online resources, now provide a wealth of information throughout the lifecycle of disasters and are relied upon by individuals and emergency responders alike. The study of such data as a lens for analysis has proved valuable in recent years, with many contributing to targeted emergency response protocols and improved methods for the management strategies of future crises. This study seeks to make a similar contribution by reporting on the use of such data for situational awareness during the case of hurricane Irma, which occurred between September and August 2017. Using a mixed methods approach the paper examines data from social media such as Twitter, as well as other online sources such as blogs and news media, to provide original insight into the disaster. A conceptual framework is then applied to determine the uses and users of social media, and to identify how these change throughout the course of the disaster, thus demonstrating situational awareness over time. The paper concludes with proposed improvements for disaster management and emergency response for future similar disasters, specifically in the hurricane season, in addition to more generalized hazards which are predicted to increase in their frequency and severity due to underlying issues such as climate change.

Abstract: Sentiment analysis has been widely researched in the domain of online review sites with the aim of generating summarized opinions of product users about different aspects of the products. However, there has been little work focusing on identifying the polarity of sentiments expressed by users during disaster events. Identifying sentiments expressed by users in an online social networking site can help understand the dynamics of the network, e.g., the main users' concerns, panics, and the emotional impacts of interactions among members. Data produced through social networking sites is seen as ubiquitous, rapid and accessible, and it is believed to empower average citizens to become more situationally aware during disasters and coordinate to help themselves. In this work, we perform sentiment classification of user posts in Twitter during the Hurricane Sandy and visualize these sentiments on a geographical map centered around the hurricane. We show how users' sentiments change according not only to users' locations, but also based on the distance from the disaster.

Abstract: Modern critical infrastructure (CI) systems are tightly coupled, resulting in unprecedented complexity and difficulty to predict, limit and control the consequences of disruptions caused by hazards. Therefore, a paradigm shift in disaster risk management is needed: instead of focusing on predicting events, resilience needs to be improved as a basis for adequate response to any event. This paper starts from a definition of CI resilience that provides a basis for quantitative and qualitative decision support. For the quantitative modelling approach, which aims at measuring the resilience of individual CIs, we focus on two CIs of fundamental importance for disaster response: transportation and power supply. The qualitative framework details relations between CIs. The results of this research are illustrated by a case study that analyses the impact of Hurricane Sandy. The findings highlight the need for a framework that combines qualitative and quantitative information from heterogeneous sources to improve disaster resilience.

Abstract: As the third hurricane the U.S. experienced in 2017, Hurricane María generated impacts that resulted in both short term and long term suffering in Puerto Rico. In this study, we aim to quantify the vulnerability of Puerto Ricans by taking region and society specific characteristics of the island into account. To do this, we follow Cutter et al.'s social vulnerability calculation, which is an inductive approach that aims to represent a society based on its characteristics. We adapted the Social Vulnerability Index (SoVI) for Puerto Rico by using data obtained from the U.S. Census Bureau. We analyzed the newly calculated SoVI for Puerto Rico and compared it with the existing deductive approach developed by the Center for Disease Control (CDC). Our findings show that the new index is able to capture some characteristics that the existing vulnerability index is unable to do.

Abstract: As shown by Hurricane Katrina, disposing of disaster-generated debris can be quite challenging. Extraordinary amounts of debris far exceeding typical annual amounts of solid waste are almost instantaneously deposited across a widespread area. Although the locations and amounts of debris can be easily summarized looking back after recovery activities have been completed, they are uncertain and difficult at best to estimate as debris operations begin to unfold. Further complicating matters is that the capacity of cleanup resources, which is dependent upon available equipment, labor, and subcontractors, can fluctuate during on-going cleanup operations. As a result, debris coordinators often modify initial resource assignments as more accurate debris estimates and more stable resource capacities become known. In this research, we develop a computer-based decision support system that incorporates a multi-stage programming model to assist decision makers with allocating debris cleanup resources immediately following a crisis event and during ongoing operations as debris volumes and resource capacities become known with increasing certainty.

Abstract: Massive crisis open data is not fully utilized to identify humanitarian needs because most of it is not in a structured format, thus hindering machines to interpret it automatically and process it in a short time into useful information for decision makers. To address these problems, the paper presents a method which merges ontologies and logic rules to represent the humanitarian needs and recommend appropriate humanitarian responses. The main advantage of the method is to identify humanitarian needs and to prioritize humanitarian responses automatically so that the decision makers are not overwhelmed with massive and unrelated information and can focus more on implementing the solutions. The method is implemented on real data from the Hurricane Wilma crisis. The use of the method in the hurricane Wilma crisis shows the potential abilities to identify the humanitarian needs in specific places and to prioritize humanitarian responses in real time.

Abstract: Survey researchers need to, Find out what assumptions evacuation modelers are making and collect empirical data to replace incorrect assumptions;, Obtain data on the costs of evacuation to households, businesses, and local government; and, Extend their analyses to address the logistics of evacuation and the process of re-entry. Evacuation modelers need to, Incorporate available empirical data on household evacuation behavior, and, Generate estimates of the uncertainties in their analyses. Cognitive scientists need to, Conduct experiments on hurricane tracking and evacuation decision making to better understand these processes, and, Develop training programs, information displays, and performance aids to assist local officials who have little or no previous experience in hurricane evacuation decision making.

Abstract: Several nations have signaled their intent to phase out petroleum-based engines for passenger vehicles and promote a transition to battery electric vehicles (BEVs). While researchers have established the long-term environmental benefits of BEVs, there are critical considerations for policymakers in areas prone to natural disasters. This research intends to develop a geospatial-based model to explore and simulate the evacuation of BEVs during a disaster. This work-in-progress (WiPe) paper examines the variables essential to creating an effective hurricane simulation. The final simulation model is intended to allow for the evaluation of BEV policy options under a variety of scenarios. We describe the considerations made during the development of this geospatial agent-based simulation under various hurricane parameters. Finally, we mention the expected benefits of our work and hint at possible policy directions.

Abstract: Trends over the past decades when coupled with recent disaster events call into serious question whether our typical reactions to natural disasters will be sufficient for what we can expect in the future. This paper summarizes current events and scientific understanding of our planet to provide insights of the authors into what should be the basis for future policies and plans.

Abstract: While Hurricane Irma struck the Cuban Southern coasts, thousands of tourists were evacuated from this area and relocated in the Varadero peninsula. In a couple of days, all means for families and friends to connect with the tourists were down, leaving them without any information about the on-site situation. This paper focuses on the volunteer citizens' initiative to take advantage of social media, to enhance their own situational awareness in Varadero area, supporting fellow citizens to identify and localize their relatives. In particular, two Facebook groups that were created at the time are analyzed and their messages' content and objectives categorized. We will show that once more, social media has constituted opportunities for citizens to engage a specific response to the crisis, but at the same time has raised specific ethical and social issues.

Abstract: To this date, research on crisis informatics has focused on the detection of trust in Twitter data through the use of message structure, sentiment, propagation and author. Little research has examined the effects of perceived emotion of these messages in the crisis response domain. Toward detecting useful messages in case of crisis, we examine perceived emotions of these messages and how the different emotions affect the perceived usefulness and trustworthiness. Our analysis is carried out on two datasets gathered from Twitter concerning hurricane Sandy in 2012 and the Boston Bombing 2013. The results indicate that there is a significant difference in the perceived emotions that contribute towards the perceived trustworthiness and usefulness. This could have impacts on how messages from social media data are analyzed for use in crisis response.

Abstract: We utilize the 2012 Hurricane Sandy dataset to investigate methods to measure network stability during a crisis. While previous research on information distribution has focused on individuals that are most connected, or most willing to share information, we examined this dataset for indicators of network stability. The value of this measure is to identify the points of failure within the network. The findings in this paper provide support for the use of social network analysis within the realm of crisis response to identify the points of failure within the network.