Skip to main content
U.S. flag

An official website of the United States government

Predicting surf zone injuries along the Delaware coast using a Bayesian network

August 14, 2019

Personnel at Beebe Healthcare in Lewes, Delaware, collected surf zone injury (SZI) data for eight summer seasons from 2010 through 2017. Data include, but are not limited to, time of injury, gender, age, and activity. More than 2000 SZI events, including 196 spinal injuries and 6 fatalities, occurred at the five most populated beaches along the 25 miles of Atlantic-fronting coast. SZI are predominantly wave related incidents associated with wading (50.1%), body surfing (18.4%), and body boarding (13.3%). The episodic nature of SZI indicate the importance of linking the environmental conditions and human behavior in the surf zone to predict days with high injury rates. Higher order statistics are necessary to effectively consider all associated factors related to SZI. Two Bayesian networks (BN) were constructed to model SZI and predict changes in injury rate (proportion of injuries to bathers) and injury likelihood (probability of at least one injury occurrence) on an hourly basis. The models incorporate environmental data collected by weather stations, wave gauges, and researcher personnel on the beach. The models include prior (e.g., historic) information to infer relationships between provided parameters. Sensitivity analysis determined the most influential parameters related to injury rates were significant wave height, foreshore slope, and water temperature. Exposure parameters (e.g., air temperature) influenced the number of people in the water, resulting in strong correlation between injury likelihood and the related meteorological conditions (variance reduction > 0.4%). Log likelihood ratio (LLR) scores indicate the network predicts SZI likelihood during any specified hour with more skill than prior predictions with the best performing model improving prediction 69.1% of the time (LLR = 69.1%). An alternative BN predicting injury rate performed worse with the prior probability model out predicting the injury rate network (positive LLR = 36.7%). Issues persist with predicting SZI that have an LLR ≪ -1 (< 5% of 2017 injuries) and occur in conditions different than when most other SZI occur. Better understanding of SZI will improve awareness techniques to both educate beachgoers and assist beach patrol decision making during high risk conditions.

Publication Year 2019
Title Predicting surf zone injuries along the Delaware coast using a Bayesian network
DOI 10.1007/s11069-019-03697-y
Authors Matthew Doelp, Jack A. Puleo, Nathaniel G. Plant
Publication Type Article
Publication Subtype Journal Article
Series Title Natural Hazards
Index ID 70207595
Record Source USGS Publications Warehouse
USGS Organization St. Petersburg Coastal and Marine Science Center