Finding appropriate boundary conditions for high frequency forcing of Regional Simulations – California Current System as a case study.

Abstract

<p>Quite a handful of past studies have reported lack of energy near the tidal bands in high-resolution, regional model simulations&#8217; frequency-wave number spectra when compared to observations. A plausible reason for this discrepancy could be the lack of remotely generated internal tides in regional simulations. In this study, we consider the impact of remote internal tides on the energetics in regional simulations of the California Current System (CCS). The CCS is an eddy-rich mid-latitude region, where energetic NIWs and internal tidal waves coexist. We run high-resolution realistic regional simulations using the Regional Ocean Modelling System (ROMS). The ROMS simulations are boundary-forced with high-frequency offline data from the Hybrid Coordinate Ocean Model (HYCOM). We consider a year-long HYCOM expt_06.1 simulation with 8-km horizontal grid resolution and 41 depth layers. The HYCOM simulation is realistically forced with tides and atmospheric forcing.</p><p>Time-mean and depth-integrated internal tidal flux computed for the parent HYCOM domain shows radiation of remotely generated internal tide beams into the ROMS domain. These beams comprise mainly of modes 1 & 2. To ensure that we provide satisfactory open boundary conditions (OBCs) for our regional simulation, we conduct sensitivity runs using two main types of OBCs (clamped and adaptive OBCs). For the runs with clamped OBCs, we varied the sponge layer viscosities at the open boundaries from 100 to 800 m²/s. Both nudging parameters and sponge layer viscosities are varied for simulations with the adaptive OBCs.&#160; Although, we observe remotely generated internal tides in all our simulations, we find that the amount of internal tidal energy that is transmitted through the open boundaries and the internal tidal energetics in the interior of the domain depend on the nudging time scales, sponge layer width and/or viscosity values.</p><p>In the future, we plan to nest down to increasing high-resolution horizontal and vertical grids and perform simulations with different boundary forcings e.g. with total internal tides, stationary internal tides, and no internal tides. We will also force the ROMS model with unidirectional internal tides. We will evaluate the impacts of these scenarios on the internal tide energetics in the ROMS domain.</p>