Comparing Statistical Rip Current Forecast Model Output with in Situ and Theoretical Rip Current Speeds

Rip currents are a major hazard for beachgoers, accounting for about 100 deaths per year and 80% of beach rescues in the United States. The National Weather Service is implementing a forecast system for hazardous rip currents based on a statistical model developed using lifeguard observations. Hindcasted likelihoods of hazardous rip currents compare well with lifeguard estimates of rip current intensity. Here, the modeled likelihood of rip currents is compared with in situ velocity observations collected at 5 to 15 locations in and near rip channels and on the neighboring shoals on an energetic sandy beach in Duck, NC. Over two months in 2012 with 0.3 to 1.5 m incident wave heights, five experiments were conducted in which channels were dredged across the surf zone to create bathymetry favorable to rip currents. Additional observations were collected for two months in 2013 with 0.3 to 3.5 m waves propagating over natural bathymetry that varied from alongshore uniform with a shore parallel sandbar to alongshore inhomogeneous with channeled crescentic sandbars. Observed hourly tidal elevation and significant wave height and direction were used to force the statistical model. Modeled likelihoods are consistent with observed rip current presence, and relationships between the modeled likelihood and observed maximum offshore-directed flow speed (across all sensors) are investigated for a wide range of conditions. In addition, the sensitivity of the statistical model to waves, tides, and bathymetry is compared with a theoretical parameterization of rip current speed based on the dynamics of rip currents near nonuniform bathymetry. The results will be used to improve and expand the hazardous rip current forecast system. Funded by NSF, ASD(R&E), and NOAA.