High-Resolution Nested Circulation Modeling in Near-Shore Regions Using Delft3D

Abstract:

Ocean circulation modeling in near-shore regions can be accomplished by nesting a high-resolution model in fields obtained from a global or basin-scale operational forecast model; however, the accuracy obtained using this strategy can be limited by coarse spatial and temporal resolution of the operational forecast models. The objective of this study was to investigate nesting of a sub-mesoscale model for near-shore circulation in output fields from an operational forecast model in order to determine both the appropriate boundary condition type, as well as the spatial and temporal sampling requirements, for boundary- and initial-condition specification. A high resolution (~50 m) Delft3D model for the San Diego Bay region for a four-day period in June 2014 is the focus of this study. Boundary and initial condition data is obtained via nesting in the output from the Southern California (SOCAL) regional Navy Coastal Ocean Model (NCOM), with ~3.5 km spatial resolution and three-hour temporal sampling, produced daily by the Naval Oceanographic Office. To assess the impact of different nesting approaches, the Delft3D model results are compared to contemporaneous data for velocity, sea surface heights, temperature, and salinity in the region and also to the low-resolution forecast fields. Results show that the fidelity and consistency of the nested model depend on boundary-condition type, as well as details of the boundary-condition specification, and on the turbulence closure implementation. Results also exhibit sensitivity to bathymetry differences between the low- and high-resolution models; offshore, bathymetry such as ETOPO2 (with 2-minute spatial resolution) is sufficient, but near the shore a higher resolution bathymetry, such as the 1-arc-second Southern California Coastal Relief Model v.2, is required. The results demonstrate the ability to nest high-resolution models within operational forecasts to yield consistent circulation predictions for near-shore regions at higher spatial and temporal resolution than currently available from regional or basin-scale operational models.