An Examination of the Impact of Probabilistic vs. Deterministic Information on the Decision-Making Practices of Emergency Managers: 2016 HWT Case Study

Hazardous weather events can result in billions of dollars in infrastructure damage and loss of life.  Currently, a substantial body of research is focused on investigating ways to reduce the loss of life and property during such events.  Of particular importance is the development of tools that will positively impact decision-making during hazardous weather.  The 2016 Hazardous Weather Testbed (HWT) using the Probabilistic Hazard Information (PHI) tool conducted at the National Weather Center offered a unique opportunity to examine the role of probabilistic information in emergency managers’ (EMs) interactions with weather forecasters and their decisions during severe weather events.  Through the use of surveys, focus groups, and severe weather simulation sessions (both canned cases and real-time severe weather events) conducted during the HWT, this study examines: (a) the extent to which EMs preferred working with numbers (i.e., 80%) over words (i.e., very likely); (b) the influence of the type of event EMs were responding to (i.e., tornadoes, lightning, severe thunderstorms) on their preferences for numerical or textual information; and (c) the EMs’ understanding of probabilistic information.  Previous research has examined the implications of presenting probabilistic information in different ways, but with probability of precipitation.  To our knowledge, there is scant research that examines the role of probabilistic information in EMs’ decision making during convective/severe weather events, which is a unique contribution of the present research.  Implications of these findings will be addressed in the proposed talk.