Asymptotically well-balanced geostrophic reconstruction finite volumes numerical schemes for the 2D rotating NLSWE in spherical coordinates

A. González, M.J. Castro, J. Macías · 2025

Read Asymptotically well-balanced geostrophic reconstruction finite volumes numerical schemes for the 2D rotating NLSWE in spherical coordinates by A. González, M.J. Castro, J. Macías (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2510-16002.txt.

Introduction

Alejandro González del Pino, Manuel Jesús Castro Díaz, and Jorge Macías Sánchez build numerical schemes that respect geostrophic balance on a rotating sphere. Large-scale air and ocean flow is governed mainly by the balance between the Coriolis force and the pressure gradient. Schemes that break that balance corrupt forecasts, above all as the Rossby number shrinks toward zero.

Their finite-volume methods solve the two-dimensional rotating shallow-water equations in spherical coordinates at second and third order. By construction the schemes preserve geostrophic equilibrium in the small-Rossby limit, a property called asymptotic well-balancing. The end use is forecasting meteotsunamis, long ocean waves raised by air pressure disturbances that grow as they near the coast. Tests on analytic and real cases underscore how much long-run balance matters, the three authors report.

Catalog record

CREATORA. González, M.J. Castro, J. Macías
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE85 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2510-16002

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

Asymptotically well-balanced geostrophic reconstruction finite volumes numerical schemes for the 2D rotating NLSWE in spherical coordinates Thanks: Submitted to the editors DATE. A. González, M.J. Castro, J. Macías Abstract The dynamics of large-scale geophysical fluids is primarily governed by the balance between the Coriolis force and the pressure gradient. This phenomenon, known as geostrophic equilibrium, is the basis for the geostrophic model, which has proven to be extremely useful for understanding and forecasting large-scale atmospheric and oceanic dynamics. In the present work, we develop second- and third-order finite-volume numerical schemes applied to the 2D rotating shallow-water equations in spherical coordinates. These schemes are designed to preserve the geostrophic equilibrium in the limit as the Rossby number tends to zero. The final goal is to design reliable and efficient forecasting models for simulating meteotsunamis, long-wave events generated in the ocean by atmospheric pressure disturbances. These disturbances produce long waves of small amplitude that gradually amplify as they approach the coast. The numerical results for various analytical and real-world test cases underscore the importance of maintaining geostrophic equilibrium over time. keywords rotating shallow water equations, finite volumes, geostrophic equilibrium, GPU, meteotsunamis, asymptotic well-balanced Funding. This work was funded by Funded by ChEESE-2P (EU-EuroHPC JU-101093038) and DT-GEO project (HORIZON-INFRA-2021-TECH-01-01, number 101058129). This research has also been partially funded by MCIN/AEI/10.13039/501100011033 and by the “European Union NextGenerationEU/PRTR” through the Grant PDC2022‐133663‐C21 and by MCIN/AEI/10.13039/50110001103 and by “ERDF A way of making Europe,” by the European Union through the Grant PID2022‐137637NB‐C21. † † runningheads: AWB scheme for 2D RSWE in sphere / A. González, M.J. Castro and J. Macías AMS 86A15, 65L05, 65M99, 65M08, 65-04 1 Introduction Understanding atmospheric and oceanic dynamics has long captured the interest of scientists in many disciplines. All models aimed at addressing these problems use the Navier-Stokes equations as starting point, as they provide the complete set of equations describing fluid behavior. However, Navier-Stokes (NS) system of partial differential equations is too overly costly to directly solve them in big domains such as ocean basins or at global scale. To overcome this challenge, simplifying assumptions based on the specific problem are made, focusing on the dominant dynamics and neglecting less significant terms. This is where the Non-Linear Shallow Water Equations (NLSWE) come into play. The 2D NLSWE is a system of PDEs derived from the 3D Navier-Stokes system under the assumption that the horizontal scale far exceeds the vertical scale. As a result, horizontal dynamics dominate fluid movement, whereas vertical influences are comparatively negligible. In addition, the fluid pressure is assumed to be hydrostatic and the horizontal viscosity terms are neglected. To further simplify, the horizontal velocity field is depth-averaged, eliminating vertical dependence on the fluid velocities in the Navier-Stokes equations. The NLSWE are widely used and have demonstrated excellent accuracy in many applications [ 54 , 32 ] . However, when addressing large-scale geophysical flows, two critical factors must be considered: (1) the necessity of a spherical coordinate system, as the geometric terms associated to the Earth’s curvature cannot be neglected; and (2) a rotational framework, where the forces due to the Earth’s rotation may be retained as they play a significant role. When incorporating Earth’s rotational effects, a widely used model is the 2D rotating nonlinear shallow-water equations with variable bottom topography ( 1 ) (see [ 44 ] pp. 59-63 for a full derivation): \begin{cases}\partial_{t}h+\nabla\cdot(h\vec{u})=0\\ \partial_{t}\vec{u}+\vec{u}\cdot\nabla\vec{u}+\frac{1}{Fr^{2}}\nabla h+\frac{1}{Ro}2e_{z}\times\vec{u}=\frac{1}{Fr^{2}}\nabla H\end{cases} (1) where h is the water thickness; \vec{u} , the velocity vector; e_{z} , the normal unit vector in the vertical direction; H , the bottom topography measured from a fixed level of reference; Fr=\frac{U}{\sqrt{gB}} , the Froude number; Ro=\frac{U}{fL} , the Rossby number; g , the gravity acceleration and U , B , L are typical velocity, height, and length, respectively, and finally f is the Coriolis parameter. Note that the previous formulation is only valid for smooth solutions. If shocks appear in finite time, conserved variables should be used. In this work we will use the conservative formulation of system ( 1 ) The Rossby number, a dimensionless quantity defined as Ro=\frac{U}{fL} , is an important parameter in determining the regime of these fluids. When fluid simulations involve a time scale T=\frac{L}{U} that is significantly longer than the Earth’s rotation time scale (or equivalently, greater than the inverse of the Coriolis parameter, \frac{1}{f} ), the system operates in a low Rossby number regime, since the Rossby number can also be expressed as Ro=\frac{1}{fT}. Under these conditions, the Coriolis force dominates over the non-linear advective terms. As a result, the horizontal momentum equations of the fluid reach a balance between the Coriolis force and the pressure gradient [ 45 , 8 ] . This state, where the Coriolis force exactly balances the pressure gradient, is known as geostrophic equilibrium . The theory of geostrophic equilibrium is a powerful tool for understanding the dynamics of atmospheric and oceanic currents. For example, it explains why hurricanes or storms take on spiral shapes and accounts for the “Taylor Column” structure observed in rapidly rotating homogeneous incompressible fluids with flat bottom [ 37 , 48 ] . In Cartesian coordinates, geostrophic equilibrium can be written as: \begin{cases}$$g\,\partial_{x}(h-H)=fv,\\ g\,\partial_{y}(h-H)=-fu,$$\end{cases} (2) where (u,v) is the horizontal velocity field. Although this theory provides an excellent model for understanding how large-scale fluids flow from high-pressure to low-pressure regions -essentially driven by the geostrophic relation ( 2 ), which indicates that the movement is parallel to the isobars-, it has several important limitations. First, it describes a steady-state equilibrium, making it unsuitable for capturing time-evolving dynamics. Additionally, the assumption of nearly zero Rossby number may not hold throughout the system’s evolution, meaning that the effects of the non-linear advective terms in ( 1 ) should be considered during certain phases of the flow. To enhance the model, small perturbations around the geostrophic equilibrium are introduced, distinguishing between the velocities associated with the geostrophic and the non-geostrophic components, as their scales differ. This new consideration, along with the assumption of a low Froude and Rossby number of the same order of magnitude, leads to the quasi-geostrophic model. In this framework, the time dependence of the advective terms explicitly appears, and their influence in the fluid’s overall behavior is determined by the magnitude of the Rossby number, which is still assumed to be small Ro\ll 1 (for a full derivation of the quasi-geostrophic model, see Chapter 4 of [ 40 ] ). The 2D rotating NLSWE can be interpreted as an approximation of the quasi-geostrophic model in regimes with low Rossby and Froude numbers [ 40 ] . In these regimes, geostrophic balance plays a critical role. As a result, numerical schemes designed to solve these equations often focus on preserving an exact or highly accurate solution of the geostrophic equilibrium. In other words, these schemes are well-balanced for such steady states. Well-balanced schemes for a specific steady state -or a set of steady states- enable the accurate capture of small perturbations over the stationary state, even on coarse meshes, as the stationary state is resolved to machine precision (see [ 3 , 11 , 15 , 17 , 18 , 20 , 26 , 30 , 33 , 35 , 39 , 42 ] and references therein). It

Read the complete text