Research Highlights
Electric Vehicle Charging Stations at Risk from Hazardous Events and Power Outages: Analytics and Resilience Implications
A geospatial analysis of natural hazard vulnerabilities and power outage risks to Level 2 and DC fast charging infrastructure in the United States
September 09, 2025


The upper panel presents the study methodology’s flowchart. The lower panel displays the spatial distribution of charging stations, power outage frequency, power outage duration, and FEMA’s Risk Index Score across Texas counties. The number of bars in each county represents the frequency of power outages that occurred in 2023. Bar length and color indicate power outage duration, while the map color scale represents the number of charging stations per county.
The Science Objective
The resilience of electric vehicle charging infrastructure presents a critical challenge as transportation electrification accelerates to mitigate climate change and reduce greenhouse gas emissions. While natural hazards and power outages threaten EV charging station operations, the quantitative relationships between hazard exposure and infrastructure placement patterns are not well understood. This study examines how natural hazard risks impact electric vehicle charging infrastructure across the United States through geospatial analysis of national risk index scores and charging station spatial distributions. We analyze correlations between hazard vulnerabilities and Level 2 and DC fast charging deployment patterns using 2023 electric disturbance event records. Our research identifies critical risk thresholds where power outage probabilities increase significantly, particularly in high-risk states like Texas. These findings provide evidence-based guidelines for prioritizing infrastructure fortification and resilience planning in hazard-prone regions.
Approach
- Conducted a comprehensive literature review to assess natural hazard impacts on EV charging infrastructure using the Dawson Framework methodology.
- Statistically associated hazard risk indices, U.S. national charger inventory, and power outage data.
- S. census-tract correlation / P-value matrix for Hazards Risk Score vs. Level-2 /DCFC ports counts.
- Feedback from data curves: population share vs. charger share by hazard class; population share vs. outage duration.
- Regression-discontinuity analysis shows that counties above a cutoff risk score value face higher power outage risk to their chargers.
Impact
- This research quantifies relationships between natural hazard risks and EV charging infrastructure distribution. The findings provide evidence-based foundations for risk-informed planning decisions.
- Level 2 chargers are clustered in high-risk, high-density census tracts while DCFC ports are more evenly distributed but still vulnerable to outages.
- Results show that demographic and economic factors drive infrastructure placement more than hazard risks. This informs policymakers about integrating resilience into balanced planning approaches.
- Weather-related events cause the majority of charging network disruptions. This highlights critical vulnerabilities requiring immediate attention from infrastructure planners.
- Recent extreme weather events exposed charging network vulnerabilities in major metropolitan areas. This provides concrete evidence for enhancing infrastructure resilience where EV adoption is rapid.
- The study identifies specific risk thresholds for power outage probability. This offers actionable tools for agencies to prioritize infrastructure investments in high-vulnerability counties.
Summary
This study uses a geospatial analysis to assess natural hazard vulnerabilities of electric vehicle charging infrastructure across the United States, revealing weak but statistically significant correlations between FEMA’s National Risk Index scores and Level 2/DC Fast Charging station distributions. The analysis identified severe weather and natural hazards as primary causes of power outages disrupting charging operations, with Texas serving as a critical case study where counties exceeding a risk index threshold of 97.18 face significantly increased power outage probabilities. Results demonstrate that high-risk census tracts disproportionately host Level 2 charging stations, while infrastructure disparities between urban and rural areas require differentiated resilience strategies.
Team Members

Wencheng Bao (Student)
UIUC

Eleftheria Kontou (Advisor)
UIUC
Publications
This paper has been accepted for publication in Renewable and Sustainable Energy Reviews (Article 116144)
Sign up for our mailing list
You’ll receive the latest updates on CLEETS research and programming.