Algorithm for non-hydrostatic dynamics in the Regional Oceanic Modeling System

A nonhydrostatic version of FVCOM: 1. Validation experiments

“…Zijlema and Stelling [2005] attributed this to the first-order accuracy error in the pressure splitting inherent in this method. For this reason, recent nonhydrostatic model development efforts prefer to use the pressure correction method [Kanarska et al, 2007;Fringer et al, 2006]. Our analysis in the surface standing wave case found that the free surface damping and mismatch found by others are probably partly attributed to their specific treatment of the surface boundary condition for q.…”

“…No correction is made for the free surface elevation after q and the divergence-free velocity are determined at the internal mode step. This may be the reason why the computed amplitude and phase of the surface standing wave shown by Kanarska et al [2007] deviate from the analytical solution after 20 s (approximately four oscillation cycles), even though the grid resolution in their experiment is higher than that used in the FVCOM-NH simulation presented here. Excellent agreement between the FVCOM-NH-computed and the analytical free surface elevations in both surface standing and solitary waves in a flat-bottomed channel indicates that the variation of the nonhydrostatic pressure must be taken into account in calculating the free surface elevation in the external mode.…”

“…[49] A nonhydrostatic version of ROMS [Haidvogel et al, 2000] has been developed recently by Kanarska et al [2007]. Using the pressure correction method, the external mode in the nonhydrostatic version of ROMS is solved with the inclusion of nonhydrostatic pressure but with the assumption that q can be treated as a "slowly varying" force which does not vary over an internal time step.…”

“…This approach appears to maintain the accuracy of the predicted external velocities, which can be used to ensure the consistency between the barotropic and baroclinic fluxes. A small error, however, could still exist in the external velocities and subsequently in the free surface elevation since the depthintegrated nonhydrostatic pressure gradients must be "frozen" over the entire external substeps [Kanarska et al, 2007].…”

Development of a Novel Coupled Simulation Tool to Study Ocean-Estuary Exchange

“…To simulate the influence of coastal currents on exchange between San Francisco Bay and the coastal Pacific, the SUNTANS simulations will be nested within the ROMS model (Shchepetkin and McWilliams, 2005), which also has a nonhydrostatic module (Kanarska, et al, 2007). ROMS is discretized in horizontal curvilinear coordinates and has a generalized terrain-following vertical coordinate that is configured to enhance resolution near the sea surface, and the time-stepping algorithm is split-explicit.…”