Web-based spatiotemporal simulation modeling and visualization of tsunami inundation and potential human response

Abstract

Modeling spatiotemporal phenomena can provide insight into potential behavior of simulated objects during hypothetical events. Simulation frameworks can be a useful method of modeling these scenarios, and become more flexible when developed in a fashion that facilitates automated generation of output based on variable input parameters. By connecting a simulation framework to a web-based system, a user can assign input parameters of their choosing, run a simulation, and explore the output data in a dynamic, animated, map-based context using a standard web browser. The framework described here utilizes tsunami simulation data and user input to generate a combined web-based visualization and simulation model of human response to tsunami inundation. Input parameters pertaining to human population and community of interest are provided by the user and guide automated development of a simulation model scenario of spatiotemporal human response to a hypothetical tsunami inundation event. Simulated human movement is calculated at each time step using casualty model algorithms informed by behavioral research and variables such as water depth and road networks, while a mix of server-side and client-side code renders the mapping interface and supports user interaction within the web browser. Interactive controls included in the web-based simulation viewer allow the user to manipulate the map display and query underlying data either manually by time step or interactively while animation is underway. Although modeling of human movement has inherent limitations, integration of a formal casualty model with the automated simulation framework represents a unique quantitative approach for casualty determination and simulation modeling.

Keywords:

• simulation
• spatiotemporal data modeling
• geocomputation
• geovisualization
• web GIS

Acknowledgments

Toshitaka Katada provided helpful advice early in the study. Eric Geist provided high-resolution runup simulation data for the Seaside, Oregon area that were used in prototype development. Clatsop County personnel provided some of the GIS data used in this study. The authors would like to thank the anonymous reviewers for their valuable comments and suggestions.

Funding

Initial development of the simulation framework was supported by the National Science Foundation [grant number SES-0527520]. Some additional funding was provided by an NWACC 2008 Grant.