Dissipation and Bathymetric Sensitivities in an Unstructured Mesh Global Tidal Model

Abstract

The primary dissipation mechanisms for global tides are boundary layer dissipation and internal tide dissipation from barotropic to baroclinic tidal conversion (Munk, 1997). From early estimates of total tidal dissipation characterized solely by boundary layer dissipation by Taylor andShaw (1920) andJeffreys (1921) to sophisticated estimates using altimeter data and assimilated tidal models performed by Egbert and Ray (2001) and Green and Nycander (2013), our understanding of where the astronomical energy imparted on the oceans is dissipated has grown tremendously. While there is some uncertainty in where tidal dissipation predominantly occurs in nature, it is relatively well accepted that tides dissipate approximately 3.5 TW of energy (Munk, 1997). What is more unclear is the distribution between internal tide dissipation and boundary layer dissipation. Munk (1997) estimates 2.6 TW of dissipation on the shelves and 0.9 TW dissipated through internal tides. Egbert and Ray (2001)

Global Barotropic Tide Modeling Using Inline Self‐Attraction and Loading in MPAS‐Ocean

“…This can be done by evaluating the integrals directly on the MPAS mesh so that the interpolation step is not needed, as pursued in a related study by our group. Comparison to tide gauge data shows that this first attempt at including tides results in slightly larger errors than tides in some other models (e.g., Blakely et al, 2022;Pringle et al, 2021;Schindelegger et al, 2018;Stammer et al, 2014). However, we expect that the RMS errors for the 𝐴𝐴 M2 tide could be improved with (a) variable resolution meshes with smaller minimum cell size, (b) optimization of parameters in the variable resolution mesh generation, (c) the addition of a spatially-dependent wave drag coefficient or a more sophisticated wave drag scheme (e.g., Green & Nycander, 2013), and (d) the inclusion of ice-shelf cavities.…”

“…We expect that once further improvements to the variable resolution mesh allow us to resolve the coastline in better detail, these errors might reduce even further. Additionally, allowing for different wave drag coefficients for different regions may help optimize drag specifically along shelves and coastlines (e.g., Blakely et al., 2022).…”

“…As such, we are planing a follow‐up study focused on the accuracy of tides under those circumstances. For now, the simulations in this paper do not include an explicit representation of ice shelf cavities, which have been shown to impact tidal accuracy, particularly in this region (Blakely et al., 2022; Stammer et al., 2014).…”

“…It was demonstrated by Arbic et al (2009) and Arbic and Garrett (2010) that regions of large resonant coastal tides, such as Hudson Bay, have a substantial "back effect" on the global ocean tidal system. Blakely et al (2022) showed that tidal errors can be improved by combining GEBCO bathymetry with various high-quality regional bathymetric data sets. Motivated by this, we include a comparison of two different global bathymetric data sets: GEBCO2021 (GEBCO Compilation Group, 2021) and SRTM15+ (Tozer et al, 2019), each with regional patching around Canada (Canadian Hydrographic Service, 2018), northern Australia (Beaman, 2016) and the great barrier reef (Beaman, 2020).…”

“…A parameterized topographic wave drag accounts for the energy dissipation that occurs when internal tides are generated from the tidal flow over rough topography in the presence of stratification. The importance of including this term in barotropic tidal simulations has been discussed in many papers (Arbic et al., 2004, 2008; Blakely et al., 2022; Egbert et al., 2004; Jayne & St. Laurent, 2001). Finally, we expect the inclusion of a full SAL calculation to increase the computational time of the simulations, as has been shown in previous implementations (Schindelegger et al., 2018; Shihora et al., 2022).…”