New research reveals that the overall risk of a fatal car crash increases by about 34% during precipitation events in the contiguous United States. Compared to previous research, the use of a high-resolution radar precipitation reanalysis in this study, rather than data from in situ weather stations or police reports, results in much more precise information on the local conditions when a crash occurred.

New research published in the Bulletin of the American Meteorology Society reveals that precipitation events increase the likelihood of a fatal car crash in the continental United States by about 34%.

Lead author Scott Stevens of NC State’s North Carolina Institute for Climate Studies (NCICS), along with colleagues Carl Schreck and Ken Kunkel and co-authors from the Centers for Disease Control and Prevention and the University of Nebraska Medical Center, combined crash data for 2006–2011 from the National Highway Transportation Safety Administration with high-resolution radar data to identify crashes that took place during precipitation. Crashes where alcohol or drugs were noted in the police report were eliminated from the study to focus on precipitation as a causal factor.

The radar data comes from NOAA’s NEXRAD reanalysis, which was the result of collaboration between NCICS and two NOAA organizations: the National Centers for Environmental Information and the National Severe Storms Laboratory. It provides data at five-minute intervals for 0.01° x 0.01° latitude/longitude grid points.

The researchers evaluated the “relative risk” of precipitation events, calculated as 𝑝1/𝑝0, where 𝑝1 is the probability of a crash in the presence of precipitation, and 𝑝0 is the probability of a crash without precipitation. The article notes that for the 125,012 fatal crashes studied “NEXRAD indicates falling precipitation in 9,636 cases (7.7%), compared to a weighted climatology that indicates such conditions only 5.8% of the time. The overall relative risk of falling precipitation with regard to fatal crashes is 1.343, representing a 34% increase in the likelihood of a fatal crash during precipitation.”

These effects were significantly higher for heavier precipitation events, with a relative risk of 2.46 for heavy precipitation events compared to 1.27 for light precipitation. Although more crashes occur in the evening, the relative risk of precipitation peaks in morning hours.

The team also broke down the results across nine different regions. They found that the increase in risk was lowest in the Northeast (1.22) and highest in the Northern Rockies (1.74). Risks are also higher in the winter months, when precipitation is more likely to fall as snow or freezing rain.

Results from the 4th National Climate Assessment, including observed increases in annual average precipitation and in heavy precipitation events—changes that are expected to increase in the future with important regional and seasonal variations, suggest that these aspects of a changing climate could lead to an associated increase in fatal crashes.

The full paper is available online at:

The research was recently highlighted in an article in Claims Journal, the Associated Press, the Washington Post, and several local/regional news outlets. See In the Media for a full list of media coverage.

Stevens, S., C. Schreck, S. Saha, J. Bell, and K. Kunkel, 2019: Precipitation and fatal motor vehicle crashes: continental analysis with high-resolution radar data. Bulletin American Meteorological Society. In press.

(Story updated on April 29, 2019, to clarify affiliations and expand list of media coverage)