How can the human dimensions of disaster impacts be more accurately captured and represented in the analysis, modeling and simulation of disasters?
The human centered data on disaster impacts can be more accurately captured through the use of standardized measures of recovery in terms of housing; physical/mental stress; livelihood; individual social ties; collective community ties on individual; familial; community; and larger society system levels. These measures need to be implemented through long-term and longitudinal (if possible) methodologies. Disaster recovery is a long-term social process (Nigg; 1995; Wenger et al.; 1996); and is thus conceptualized as a function of time and other pre-existing conditions and post-event covariates. A set of cross-sectional data show static relationships among key variables; and a series of cross-sectional data is capable of illustrating aggregate changes over time. Longitudinal data can show more dynamic relationships on change; duration; time; lag; speed; and acceleration among exogenous; endogenous; and dependent variables over social change processes (Rose; 2000; Frees; 2004). Longitudinal research designs allow researchers to examine each individual recovery trajectory or a group of trajectories; so that they can explore; make inferences to; and eventually explain their relations to exogenous and endogenous explanatory variables. In other words; the advantages of longitudinal data go a long way in elucidating causality.
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?
Long-term longitudinal studies on resilience/recovery are still very new and rare in both Japan and the US. Each longitudinal study typically suffers from subject attritions (the more waves of the study; the fewer subjects with complete responses). One way to deal with the sample attritions is to append subjects from multiple sets of long term and longitudinal study data. Integrating existing Japan/US long-term/longitudinal study data also allows us to better explore the universal nature of resilience/recovery despite cultural/societal/linguistic/race/class/gender differences. It therefore makes sense to collaborate recovery data management efforts and to integrate/append existing Japan/US long-term/longitudinal study data bases by employing common languages of recovery/resilience constructs; by using identical measures of the same constructs and/or the development of the new technologies to create proxy variables that are comparable to the counterpart data sets.
In what ways can US-Japan collaborations advance these questions in new and important ways?
Shigeo Tatsuki and his colleagues began longitudinal life recovery surveys among the 1995 Kobe earthquake survivors in 6; 8 and 10 years after the incident. Out of the Kobe surveys the Seven Critical Element model (SCEM) of life recovery was developed. The validity and generalizability of the SCEM was then tested by the 2011 GEJE (Great East Japan Earthquake) longitudinal survey data which consisted of 2 to 5 and 10 years after the event. Tatsuki’s counterpart scientist; David Abramson has employed some of the major SCEM constructs/scales when he launched the longitudinal study of 2001 Hurricane Katrina survivors and has conducted the surveys to the same subjects in 1; 2; 3; 5 and 13 years after the Katrina disaster. Because these three (Kobe; GEJE and Katrina) longitudinal surveys shared the identical/similar constructs and the standardized measures from the long-term and longitudinal studies; it is possible to append/integrate cases from Kobe; Katrina and GEJE studies and to explore the key universal factors that determine the recovery/resilience of individuals/families despite cultural/societal/linguistic/race/class/gender differences.