Applying dynamically updated nearshore bathymetry estimates to operational nearshore wave modeling

Simulations of nearshore waves using the Steady-State Spectral Wave (STWAVE) Model with temporally-varying bathymetric boundary conditions were undertaken for a period of 11 months at the U.S. Army Corps of Engineers Field Research Facility in Duck, NC. Five sets of bathymetry were tested, two of which were derived from survey data (one evolving with each monthly bathymetric survey, one static) and three were derived from the depth inversion algorithm, cBathy, using Argus optical imagery data updated every half hour. The standard cBathy half-hourly Kalman filtered product was used along with two modified versions that filter depth estimates during large waves prior to being assimilated by the Kalman filter using an offshore wave height threshold or an optically derived wave breaking threshold to reduce bottom boundary condition errors. Bathymetry derived from the modified Kalman filter methods were first validated using continuous in-situ sonic altimeter data, and were found to improve RMSEs relative to the original cBathy bathymetry from an average of about 0.09 m and about 0.15 m offshore and onshore of the sandbar, respectively. Wave model results over the five bathymetric boundary conditions show that the thresholded cBathy bathymetry performs similarly (and slightly better in places) to simulations using an evolving bathymetry as new measurements are available. The wave height predictions using the static bathymetry were approximately equivalent in performance (if not, slightly better in places) to that of the original cBathy, while the evolving surveyed bathymetry had similar performance to the thresholded cBathy method.