How can the human dimensions of disaster impacts be more accurately captured and represented in the analysis, modeling and simulation of disasters?
The social characteristics of communities - including social capital (the ties that bind us horizontally to residents and vertically to decision makers) and social infrastructure (the spaces and places that maintain and build social capital)- matter tremendously during shocks and disasters; as a growing body of response has demonstrated. But consistent datasets from the neighborhood to the national level on this topic are challenging to find in the US and Japan. Engineering; evacuation; and even extreme weather response models that operate without these parameters cannot capture the full complexity of human behaviors. For example; to understand the ways that disaster communication impacts human behaviors; we cannot rely solely on estimates of social ties generated by generic census questions or out of date estimates of voter turnout. Better and more accurate models of evacuation behavior have instead drawn on real time data from social media platforms such as Facebook; mobility data from cell phones; and on the ground measurements of trust; civic engagement; and community norms. Additional data on the quality and quantity of community centers; libraries; and faith based organizations in communities add additional nuance to such models.
What type of data and supporting research infrastructure would be necessary to enable novel, transdisciplinary approaches to answering these and other human-centered disaster questions?
To accurately capture the social characteristics of communities - such as their reservoirs of social capital and their social infrastructure facilities - we need to think broadly about the kinds of data available to us; ranging from individual level mobility data to social media and geocoded business platform data to interviews; surveys; and fieldwork. First; we need more researchers to be able to access data like Facebook's Data for Good (DFG) project; and we need more platforms (Nextdoor; Twitter; Instagram; etc.) to make this data available in more locations. Next; we need more scholars and institutions in Japan and the United States to post data on publicly available sites such as Harvard's DataVerse Network (DVN) and ICPSR. I have made a number of attempts to ask scholars to make their data publicly available with little success (despite claims from journals about the need to post relevant datasets). Third; we need more high quality neighborhood information on Japan and the United States; available beyond the subscription services of marketing firms like ESRI. My team and I have spent a great deal of time having to scrape the shi-cho-son (city /town/ village) level websites to gather relevant information
In what ways can US-Japan collaborations advance these questions in new and important ways?
There are several ways that US-Japan collaborations can advance these questions; especially as new reports have underscored that Japan is falling behind in terms of its scientific advancement. (1) We need more regular cross national collaboration between institutions; with more scholarly and undergraduate and graduate student exchanges and conferences explicitly focused on building up Japan-US teams (especially for early career researchers). (2) We need more regular engagement between engineers; hard scientists; and social scientists (to avoid moments such as that during COVID19 when medical personnel expressed surprise at the lack of uptake in vaccination and wished that they had thought about the problem from the perspective of social scientists. (3) We also need to speak with policy makers and decision makers directly. (4) We need more fellowships and funds structured to push forward this kind of policy-applied; disaster-focused questions.