Session: Government Agency Student Posters

Paper Number: 173779

The Impact of 2023 Lahaina Wildfire on Residents’ Future Expectations of Risk Communication and Evacuation Behavior 

Wildfire frequency and intensity are rising across the United States, yet recent events have revealed critical vulnerabilities in warning and communication systems. In August 2023, the Lahaina wildfire, fueled by Hurricane Dora, destroyed over 2,200 structures, claimed 101 lives, and inflicted $5.5 billion in damage. A cascading power‑line failure disabled all 21 local cell towers, leaving more than 14,466 subscribers without phone or internet access. The finding that 92% of residents did not receive official alerts as their first wildfire notification underscores systemic failures. The official evacuation orders have been proven by various studies to be the strongest predictor for evacuation. Delayed or absent alerts and evacuation orders can leave residents with insufficient time to react, substantially increasing their exposure to danger. Not only did risk communication fail, but the evacuation process also revealed critical weaknesses. The heavy traffic congestion, suboptimal route, and evacuation destination selection typified this event. Motivated by the cascading failures of warning and transportation infrastructures during the August 2023 Lahaina wildfire, this study investigates how such breakdowns undermine residents’ trust in official alerts, alter their expectations for warning receipt and accuracy, and shape their future protective‑action decisions.

Drawing on the Protective Action Decision Model (PADM), this study conducted a two‑wave household survey at the Lahaina Civic Center in November and December 2023. Amongst 660 valid responses, 317 (78 %) of the respondents reported their house lay within the burn zone. Survey items probed sources of information, warning receipt, threat appraisals, emotional reactions, past protective behaviors, and anticipated responses to future wildfires. This study applied descriptive statistics, correlation analyses, and multivariate regression techniques to examine these factors to residents’ expectations and intentions for future protective actions.

Preliminary findings reveal residents’ deep skepticism about the official warnings; 63% of respondents judged the likelihood of receiving an accurate future wildfire arrival‑time forecast as “not at all” to “moderately” likely, and 60% expressed similarly low confidence in receiving an evacuation notice. These two items were highly correlated  (r = .86, p < .01), indicating a general lack of faith in the official warnings and alerts. Those who missed warnings during the 2023 event reported significantly lower trust in future notifications, whereas attendance at preparedness meetings and prior evacuation advice from authorities were positively associated with greater confidence in warning delivery. Intentions for protective action displayed a complex pattern: over 60% reported they would evacuate before or immediately upon receiving an official warning, yet nearly half still expected to delay departure or remain to defend property despite the official evacuation orders. This mix of early‑evacuation planning and lingering defensiveness underscores the fragile nature of trust when infrastructure fails, particularly the communication channels that support infrastructure.

Evacuation preparation time expectations varied. Roughly 25% of residents believed they could complete preparations within 10 minutes, and 75% anticipated readiness within 40 minutes. These aligned with the 80% who departed within 30 minutes in the Lahaina wildfire 2023. Expectations for travel time were even more optimistic: 75% percent of residents believed they could reach safety within thirty‑five minutes of departure, and 96% percent expected to do so within one hour, despite the highway congestion observed during the actual event.

These findings demonstrate that infrastructure failures not only disrupt immediate disaster response but also significantly erode public trust, perpetuating a feedback loop that shapes residents' future protective behaviors. To rebuild residents' confidence, emergency managers should implement redundant, community-engaged communication channels, such as in-person preparedness workshops and localized siren networks. Evacuation strategies must address the coexistence of rapid-evacuation intent and defensive posture by integrating behavioral insights into system design. Ultimately, engineering robust, fail-safe warning infrastructures and engaging residents in preparedness activities will be essential for enhancing compliance and safety in future wildfires and other emergencies.

Presenting Author: Qianli Qiu School of Fire, Construction and Emergency Management, Oklahoma State University

Presenting Author Biography: Qianli Qiu is a Ph.D. student in Fire and Emergency Management at Oklahoma State University, mentored by Dr. Chen Chen. Their research focuses on mitigating disaster impacts on communities, with specific interests in risk communication strategies for hurricanes and wildfires and agent-based modeling for disaster scenarios. Qianli leverages micromobility systems to optimize post-disaster emergency supply delivery and risk communication, analyzing effects on evacuation behavior and community resilience. They presented an agent-based model and simulation findings at the 49th Natural Hazards Workshop. Qianli’s research aims to develop adaptive response strategies that address emergency logistics gaps and enhance community-level preparedness.

Authors:

Qianli Qiu School of Fire, Construction and Emergency Management, Oklahoma State University
Changjian Zhang Department of Civil and Environmental Engineering, University of Hawaii at Manoa
Michael K. Lindell Department of Urban Design and Planning, University of Washington, Seattle, WA
Yiwei Wang Department of Civil and Environmental Engineering, University of Hawaii at Manoa
Haizhong Wang School of Civil and Construction Engineering, Oregon State University
Guohui Zhang Department of Civil and Environmental Engineering, University of Hawaii at Manoa
Chen Chen School of Fire, Construction and Emergency Management, Oklahoma State University