HRRC study finding uncertainty’s effects on public hazard decisions

In the first large-scale study of its kind, Texas A&M researchers are investigating how storm forecast graphics, such as those depicting potential storm paths with cone-shaped regions of uncertainty, influence public decision making during hazard events.

Findings from the study led by Michael Lindell and Carla Prater, faculty fellows at the Hazard Reduction and Recovery, will help develop improved methods and tools for accurately presenting predictive simulations of hurricanes, wildfires and other hazards resulting in better informed individual and public policy decisions.

Researchers Michael Lindell, professor of urban planning, and Carla Prater, senior lecturer of urban planning and the HRRC’s associate director, plan to use the study’s findings to develop a set of improved methods and tools to accurately present predictive simulations of hurricanes, wildfires and other hazards resulting in better informed individual and public policy decisions.

Anyone with an Internet connection will soon be able to participate in the four-year, $319,125 National Science Foundation-funded [project] (http://www.sci.utah.edu/vis-uncertainty) .

Participants will record the manner in which they view hazard forecasts, including how they search for hazard information in maps and tables, the order in which they access individual elements and the amount of time they spend viewing it, allowing researchers to make recommendations on the most effective ways to present hazard visualization information.

“This will also allow researchers to overcome a limitation of laboratory experiments, the difficulty of including participants other than students and local residents who can come to the computer laboratory,” said Walter Gillis Peacock, director of the HRRC.

The project will test new ways of displaying uncertainties about hurricanes’ tracks and other characteristics such as their intensity, size, and forward movement speed, all relevant to evacuation decisions.

The project will also test new ways of integrating information about hazard and evacuation uncertainties, including as the length of time needed to clear the risk area, the cost to local government and residents, and the potential loss of life.

Researchers have completed one experiment in the study, finding that peoples’ conclusions about a hurricane strike probability were unchanged whether they viewed a forecast track only, an uncertainty cone only, or a forecast track with an uncertainty cone.

In another finding, strike probabilities were judged to be higher for category 4 hurricanes than for category 1 hurricanes having the same tracks; in other words, people’s judgments of an event’s likelihood are affected by the severity of its consequences, a finding consistent with other risk perception studies.

Lindell and Prater are conducting the research with collaborators Mary Hegarty, professor of psychology at the University of California, Santa Barbara, Donald House, chairman of the division of visual computing at Clemson University and Ross Whitaker, professor of computing at the University of Utah.

The project will build on an existing grant in which Lindell, Prater and House are studying how people process information about approaching hurricanes and how it influences their evacuation plans.