Perturbed flow over and around an island can produce leeside vortices as well as a long wake region of reduced wind speed and altered thermodynamic structure. In this study we examine an island wake formed leeward of Kauai using the U.S. Navy’s Coupled Ocean/Atmosphere Mesoscale Prediction System (COAMPS) and the Wavewatch 3 (WW3) wave model. This investigation arose from radar performance tests conducted in December 1999 aboard the destroyer USS O’Kane in which Kauai’s wake noticeably impacted radar sea clutter returns. Both real data forecasts during the period of the radar tests and idealized sensitivity studies are conducted using COAMPS. To produce the real data forecasts, COAMPS is run triply nested with grid spacing of 27, 9, and 3 km; the wind field is used to drive WW3 on three meshes of the same spacing for an 84h period beginning 30 November 1999. To produce the idealized simulations, COAMPS is run with a single mesh of 3 km grid spacing and initialized with homogeneous conditions taken from an upwind sounding from the real data case. These conditions are fed in from the upwind boundary throughout the simulation; hence, in the idealized studies perturbations are due solely to the island, making wake features especially evident.

The real data forecast displays a ~200 km meandering wake trailing from Kauai, with strongly accelerated flow cornering the island flanks. The upwind flow is from the NNE throughout most of the 84 h period and a pronounced, persistent wind speed gradient exists across the wake boundaries throughout the period. A corresponding wake in sea state is clearly evidenced in the high-resolution WW3 results. The altered surface layer wind and thermodynamic fields within the wake produce an altered microwave refractivity profile as compared to that outside the wake. Insertion of the forecast COAMPS refractivity field and WW3 sea state into a radar propagation model has, in turn, permitted us to model the expected radar clutter returns. Results are found to agree well with the observed clutter distribution.

The idealized studies permit examination of wake sensitivity to mountain height and shape, as well as upwind conditions. Results vary from a no wake regime, to a long straight wake, to shedding Karman vortices as the island is raised from half to twice its true height. Results will be compared with previous studies and theoretical regime diagrams.