Finite-Element barotropic model for the Indian and Western Pacific Oceans: Tidal model-data comparisons and sensitivities. (September 2018)

• Record Type: Journal Article
• Authors: Pringle, William J.; Wirasaet, Damrongsak; Suhardjo, Andika; Meixner, Jessica; Westerink, Joannes J.; Kennedy, Andrew B.; Nong, Shangyao
• Is Part Of: Ocean modelling. Volume 129(2018)
• Page Start: 13
• Page End: 38
• Publication Date: 2018-09
• Subjects: Finite-element -- Unstructured grid -- Barotropic tides -- Bathymetry -- Internal tide energy conversion -- Bottom friction; Oceanography -- Periodicals; Océanographie -- Périodiques; Oceanography; Periodicals; 551.46
• DOI: 10.1016/j.ocemod.2018.07.003
• Languages: English
• ISSNs: 1463-5003

Abstract

Highlights: Mean RMS tidal elevation errors at coastal gauges are smaller than a data assimilated model. An absorption-generation sponge layer at lateral boundaries is necessary to help reduce resonant effects in the domain. Dissipative effects of two internal tide energy conversion parameterizations are compared. Bathymetry reduces mean continental shelf (26%) and coastal (30%) RMS tidal elevation errors. Spatially varying bottom friction coefficients reduce mean coastal RMS tidal elevation errors (9%). Abstract: In this study, a 9.6 million node large-scale unstructured grid finite-element forward barotropic model is developed and applied to understand the tidal dynamics and dissipation mechanisms of the Indian and western Pacific Oceans down to sub-kilometer scale at the coast. Tidal model-data comparisons are presented to assess the capabilities and limitations of our large-scale barotropic model. The average root-mean-square (RMS) discrepancies of tidal elevations at coastal tide gauges is 14 cm, which is ∼ 3 cm smaller than those of a state-of-the-art global data assimilated barotropic tidal model. Sensitivities to lateral boundary conditions, bathymetry, and dissipative processes are explored to guide future endeavors related to large-scale barotropic modeling in the region and other regions throughout the world. Lateral boundary conditions are found to induce adverse resonant effects on the lunar semi-diurnal modes when poorly placed elevation specified boundary conditions are used. This problem is largely resolved by using an absorption-generation layer at the boundary. Parameterization of internal tide energy conversion is identified as the most important aspect to control deep water solutions, and help reduce the RMS discrepancies of the entire system. Two forms of this parameterization are presented and their spatial distributions of dissipation are compared. Bathymetry has a negligible effect on the tidal solutions in deep water, but local high resolution bathymetry results in significant reductions to the average RMS discrepancies on the continental shelf (26%) and at the coast (30%). Implementing a spatially varying bottom friction coefficient based on sediment types decreases the average RMS discrepancy at the coast by 9% predominantly due to its positive effects in the Yellow Sea. The model is shown to capture a large amount of the tidal physics and has the potential for application to a range of barotropic problems such as wind-driven surge and tidal processes.