Progress in Disaster Science最新文献

Wildfires can significantly impact regional and global climate, human health, and ecosystems, making it necessary to model their behavior and predict their outcomes. With increasing global temperatures and changing precipitation patterns due to climate change, the frequency and intensity of wildfires are expected to increase, heightening the requirement for accurate wildfire simulation models to support wildfire management and mitigation efforts. However, the interactions between the causative variables of wildfires and the wildfire bivariate characteristics have not been explored in wildfire modeling. Therefore, the copula function was applied to solve the complicated and nonlinear relationship of the dependence structure in wildfire statistics and the relationship between wildfire causative variables. Subsequently, we modeled wildfire characteristics globally using six machine learning models and compared the performances of the models. Specifically, the main conclusions were obtained as follows: (1) among the six machine learning models, long short-term memory had the best applicability in simulating wildfire characteristics; (2) when there were 4 predictors, the accuracy of wildfire characteristic simulation reached the average level; and (3) long short-term memory achieved excellent model performance within 56% of the global climate sub-regions. Overall, this analysis provides a reference to better understand wildfire and contributes to wildfire management.

Climate change-induced alterations in temperature and precipitation patterns pose a significant threat to flood risk worldwide. This research focuses on Java Island and presents a method to develop a flood hazard index map considering various future climate change scenarios. To simulate flood characteristics for various scenarios, an RRI (Rainfall-Runoff-Inundation) model was calibrated and validated using six different flood events in three basin areas. Furthermore, the baseline, historical 100-year return period, was compared against climate change scenarios representing near-future (2015–2060) and far-future (2061–2100) cycles, integrating six bias-corrected climate models under the SSP2.45 and SSP5.85 scenarios. This study reveals that the maximum predicted flood extents under both SSP2.45 and SSP5.85 scenarios surpass the baseline across all cycles, with propagation areas expanding with each scenario and cycle. Notably, the worst-case hazard score for all cycles and scenarios occurred in DKI Jakarta and Central Java due to high rainfall concentration and population density. In the case of DKI Jakarta, the hazard score was between 0.61 and 0.76, while, for Central Java, the hazard score was between 0.61 and 0.73. These findings are important for guiding non-governmental and governmental institutions in making robust flood mitigation policies to safeguard vulnerable communities in Java Island.

Academic interest in comprehending flood resilience concepts and their evaluation is growing with the increasing frequency and severity of disasters due to climate change (CC). This paper conducts a comprehensive and systematic literature review on urban flood resilience assessment (UFRA) to address the inadequacy of urban preparedness, response and lengthy recovery process. The review covers the UFRA process considering urban elements, methods, tools and techniques, and challenges in achieving resilience and provides guidance for constructing advance resilience assessment frameworks. Scopus and Web of Science databases were accessed by employing pertinent keywords. The review of UFRA literature reveals a prevalence of frameworks analysing isolated or clustered urban elements rather than a comprehensive framework for assessing flood resilience. Various methods, tools and techniques, such as multi-criteria decision-making methods (MCDM), hydrological modelling, household surveys, and geographic information system (GIS) techniques, are commonly employed in the flood resilience assessment of single or clustered elements. However, to achieve overall resilience in an urban area, there is a need for an integrated and comprehensive approach that accounts for the evolving risk landscape. Creating a comprehensive framework poses challenges due to the intricate nature of the urban system, which requires collaboration across multiple disciplines and integration of modern tools and techniques, including computational modelling such as agent-based modelling (ABM) and system dynamics approaches, artificial intelligence techniques, and cognitive fuzzy modelling. The outcome of this research serves as a valuable resource for researchers, academicians, and policymakers. It aids in understanding the urban flood resilience process, focusing on urban elements and their inter-relationships, assessment methods and approaches, dimensions and indicators necessary for achieving flood resilience scenarios comprehensively.

This paper presents a conceptual hybrid model that embodies a synergistic approach by integrating principles from operations research (OR) with steps from the design science research methodology (DSRM). The model is applied to develop an evacuation support system for a group of hospitals located in Western Sydney, Australia, exposed to the risk of floods induced by climate change. By combining these two approaches, the conceptual hybrid model effectively overcomes the limitations associated with traditional DSRM and provides a comprehensive research framework for addressing OR problems. The research findings hold significant implications for academia, professionals, and policy makers engaged in the field of disaster risk management. Moreover, this study offers valuable insights into enhancing the effectiveness and efficiency of OR-based solutions in practical scenarios, thus contributing to the advancement of knowledge in this area.

Future exemplary education should foster inclusive and respectful learning environments to meet new challenges like digital inequality and power concentration. In the new normal of education due to COVID-19, inclusive online disaster risk reduction (DRR) education is essential. Therefore, this article aims to investigate the current status quo of inclusive online and distance DRR education and its benefits. Expert interviews were conducted with 40 educators with experience in 13 countries. The experts were selected based on their experience in designing and/or delivering courses/modules related to disaster management and/or disaster risk reduction at the tertiary level. The interview questions covered 3 key aspects i.e., 1. the effectiveness of online delivery methods, 2. the status quo of online and distant DRR education and 3. the unique benefits of online education for DRR. The key findings suggest that an online setting works best when it is scientifically designed for the right audience, the right subject area, and the right mix. In creating inclusivity in DRR education, the digital divide needs to be acknowledged and interactive learning should be diversely designed. This study identifies gaps in digital disaster education, urging policy and practice changes to support diverse DRR communities beyond education providers and recipients.

As minority groups, foreign nationals are considered vulnerable because of cultural, linguistic, and religious differences; however, some may work as critical stakeholders in disaster risk reduction (DRR). In Japan, some mosque communities (mainly composed of foreign nationals) were engaged in relief activities immediately after disasters; however, these reports were scattered and did not clarify long-term activities. Therefore, we aimed to summarize the activities and roles of mosques located in areas affected by recent natural hazard-related disasters in the short (that is, response) and long term (that is, recovery and mitigation). We targeted three mosques located in areas affected by the 2011 Great East Japan Earthquake, and the 2016 Kumamoto Earthquake in Japan. We conducted a literature review and interviewed mosque administrators. The results showed that all mosques worked as distribution centers in the short term, receiving relief supplies from Muslims all over Japan and distributing them to affected people. The relief goods included halal foods that fit foreign minorities, but these goods were distributed not only to them but also to other people regardless of religion and nationality. Some mosques also functioned as evacuation shelters, soup kitchens, and accommodations for volunteers. Support for affected people was further facilitated by intermediaries (such as private organizations and key local individuals) between the mosques and local communities or governments. In contrast to the short term, we did not find prominent recovery and mitigation activities by the mosques. These comprehensive findings contribute to the promotion of inclusive and community-based DRR activities.

The Sendai Framework for Disaster Risk Reduction 2015–2030 emphasizes the importance of learning from experiences to strengthen preparedness and “Build Back Better” through risk reduction in the phases of recovery, rehabilitation, and reconstruction. On December 26, 2004, Aceh Province in Indonesia was affected by an earthquake and tsunami that caused infrastructure damage and resulted in the death of 160,000–200,000 people. Mc. Caughey simulation (2017) showed that 50% of respondents in Banda Aceh, preferred mosques as tsunami evacuation sites in the future. Therefore, this study aimed to examine the function of mosques as tsunami evacuation buildings through a community-based approach. The method involved participatory observation and in-depth interviews, guided by the principles of Sense of Place Attachment and Behavior Theory. The results showed that the Baitul Musyahadah Mosque satisfied the criteria for serving as a tsunami evacuation facility. A significant proportion of the community (38%) selected mosques as their preferred evacuation sites, considering factors such as ease of horizontal access (22%), ease of vertical access (22%), the call to prayer (12%), and the available space (4%). The analysis of sociodemographic factors indicated that individuals aged 20–30 expressed attachment towards the physical aesthetics of the mosques, while those aged 31–50 exhibited attachment towards the symbolic significance. This study recommended integrating the design concept of mosques to accommodate worship activities and emergency evacuation, thereby contributing to national and international endeavors aimed at mitigating tsunami risks.

This study used the functional resonance analysis method (FRAM) to analyze a real-world example of an emergency-response process, one that occurred in Koriyama city, Japan, during Typhoon Hagibis (known in Japan as Typhoon No. 19, or Reiwa 1 East Japan Typhoon). The analysis revealed nine functions that constitute the municipality's emergency-response process (including Public Communication, Provision/Management of Evacuation Shelters, Evacuation, and Rescue Operations) as well as the interactive relationships between these functions. It also revealed how variabilities in the timeliness and accuracy of the functions resonated with (reinforced) each other, culminating in an incident involving a backlog of emergency calls. From these results, two recommendations were derived: reviewing the public communication system and building a framework for minimizing variability in evacuation behavior. And also, based on the FRAM analysis we could contextualize the interview data and the individual phenomena described fragmentally in the retrospective report (used for evaluating the disaster response), thereby gaining an understanding of the incident generation mechanism. In conclusion, the study demonstrated FRAM can be an effective tool for retrospective evaluation of disaster response or revising a disaster plan.

Locating, routing and deploying essential goods and equipment are proactive disaster management strategies which received attention during recent decades. Many artificial intelligence (AI) based methods have been applied to respond to disasters in the past decade. However, there lacks a systematic review on these approaches. This paper reviews such papers published over the period of 2012–2022. These publications were examined according to their goal of using AI-based methods (e.g., for disaster management or for essential goods and equipment locating and deployment). We examined the approaches adopted and their specific application areas within the broad spectrum of disaster management. Based on our review, we recommend a few areas which could benefit from AI-based methods, especially for the less explored area of locating and routing problem during disasters. This research would be helpful for academics and practitioners alike in effectively adopting AI methods to improve the resilience and response in disastrous events.