FINDING HIGH GROUND: SIMULATING AN EVACUATION UNDER THE THREAT FROM A LARGE LAHAR

In the vicinity of Mount Rainier, nearly 100,000 people live within communities that are vulnerable to large and destructive lahars. Because these large lahars are powerful enough to completely destroy all but the most robust man-made structures, evacuating to high ground, above the highest reaches of a lahar, is a critical lifesaving strategy.

Evacuations are complex social processes that depend on many factors including how many individuals are evacuating, how people perceive their own risk, how the population is distributed in space, the characteristics of the transportation network, and the choices individual evacuees make during the course of the evacuation. In many communities threatened by lahars from Mount Rainer, evacuees will have an important choice to make between whether to drive, or walk, to reach high ground.

To measure the impact of the choice between walking and driving on the time needed to evacuate the hazard zone, simulations of a hypothetical evacuation scenario were conducted. The scenario is set in and around the town of Orting, Washington, a suburban town with a population of >7500 people situated within approximately 32 km of Tacoma, Washington, an urban center with a population of >210,000. Orting is located at the confluence of two lahar prone rivers originating on Mount Rainier, approximately 50 km downstream from the volcano. The principal highway serving the community runs along the axis of the river valley offering limited means of vehicular escape. This talk will focus on explaining the results from the experiments in which an agent-based model employing a co-evolutionary learning algorithm was used to simulate a vehicle based evacuation. Results indicate that under a no-notice lahar scenario, evacuation success may be limited if those who can feasibly evacuate on foot, instead choose to drive.