Two finite-volume unstructured mesh models for large-scale ocean modeling

Non-staggered triangular grids have many advantages in performing river or ocean modeling with the finite-volume method. However, horizontal divergence errors may occur, especially in large-scale hydrostatic calculations with centrifugal acceleration. This paper proposes an unstructured finite-volume method with a filtered scheme to mitigate the divergence noise and avoid further influencing the velocities and water elevation. In hydrostatic pressure calculations, we apply the proposed method to three-dimensional curved channel flows. Approximations reduce the numerical errors after filtering the horizontal divergence operator, and the approximation is second-order accurate. Numerical results for the channel flow accurately calculate the velocity profile and surface elevation at different Froude numbers. Moreover, secondary flow features such as the vortex pattern and its movement along the channel sections are also well captured.

Download Full-text

Assessment of the Finite VolumE Sea Ice Ocean Model (FESOM2.0), Part I: Description of selected key model elements and comparison to its predecessor version

10.5194/gmd-2018-329 ◽

2019 ◽

Cited By ~ 1

Author(s):

Patrick Scholz ◽

Dmitry Sidorenko ◽

Ozgur Gurses ◽

Sergey Danilov ◽

Nikolay Koldunov ◽

...

Keyword(s):

Sea Ice ◽

Finite Volume ◽

Ocean Circulation ◽

Large Scale ◽

Unstructured Mesh ◽

Vertical Mixing ◽

Circulation Model ◽

Ocean Model ◽

Arbitrary Lagrangian Eulerian ◽

Nonlinear Free Surface

Abstract. The evaluation and model element description of the second version of the unstructured-mesh Finite-volumE Sea ice–Ocean circulation Model (FESOM2.0) is presented. The model sensitivity to arbitrary Lagrangian Eulerian (ALE) linear and nonlinear free surface formulation, Gent McWilliams eddy parameterisation, isoneutral Redi diffusion and different vertical mixing schemes is documented. The hydrographic biases, large scale circulation, numerical performance and scalability of FESOM2.0 are compared with its predecessor FESOM1.4. FESOM2.0 shows biases with a magnitude comparable to FESOM1.4 and it simulates a more realistic AMOC. Compared to its predecessor FESOM2.0 provides clearly defined fluxes and a three times higher throughput in terms of simulated years per day (SYPD). It is thus the first mature global unstructured-mesh ocean model with computational efficiency comparable to state-of-the-art structured-mesh ocean models. Other key elements of the model and new development will be described in following-up papers.

Download Full-text

Assessment of the Finite-volumE Sea ice-Ocean Model (FESOM2.0) – Part 1: Description of selected key model elements and comparison to its predecessor version

Geoscientific Model Development ◽

10.5194/gmd-12-4875-2019 ◽

2019 ◽

Vol 12 (11) ◽

pp. 4875-4899 ◽

Cited By ~ 2

Author(s):

Patrick Scholz ◽

Dmitry Sidorenko ◽

Ozgur Gurses ◽

Sergey Danilov ◽

Nikolay Koldunov ◽

...

Keyword(s):

Sea Ice ◽

Finite Volume ◽

Large Scale ◽

Unstructured Mesh ◽

Vertical Mixing ◽

Ocean Model ◽

Meridional Overturning Circulation ◽

Arbitrary Lagrangian Eulerian ◽

Element Approach ◽

Finite Volume Approach

Abstract. The evaluation and model element description of the second version of the unstructured-mesh Finite-volumE Sea ice-Ocean Model (FESOM2.0) are presented. The new version of the model takes advantage of the finite-volume approach, whereas its predecessor version, FESOM1.4 was based on the finite-element approach. The model sensitivity to arbitrary Lagrangian–Eulerian (ALE) linear and nonlinear free-surface formulation, Gent–McWilliams eddy parameterization, isoneutral Redi diffusion and different vertical mixing schemes is documented. The hydrographic biases, large-scale circulation, numerical performance and scalability of FESOM2.0 are compared with its predecessor, FESOM1.4. FESOM2.0 shows biases with a magnitude comparable to FESOM1.4 and simulates a more realistic Atlantic meridional overturning circulation (AMOC). Compared to its predecessor, FESOM2.0 provides clearly defined fluxes and a 3 times higher throughput in terms of simulated years per day (SYPD). It is thus the first mature global unstructured-mesh ocean model with computational efficiency comparable to state-of-the-art structured-mesh ocean models. Other key elements of the model and new development will be described in follow-up papers.

Download Full-text

Local quasigeoid modelling in Slovakia using the finite volume method on the discretized Earth's topography

Contributions to Geophysics and Geodesy ◽

10.31577/congeo.2020.50.3.1 ◽

2020 ◽

Vol 50 (3) ◽

pp. 287-302

Author(s):

Róbert ČUNDERLÍK ◽

Matej MEDĽA ◽

Karol MIKULA

Keyword(s):

Numerical Solution ◽

Finite Volume Method ◽

Finite Volume ◽

Large Scale ◽

Horizontal Resolution ◽

Geopotential Model ◽

Computational Domain ◽

Disturbing Potential ◽

Volume Method ◽

Upper Boundary

The paper presents local quasigeoid modelling in Slovakia using the finite volume method (FVM). FVM is used to solve numerically the fixed gravimetric boundary value problem (FGBVP) on a 3D unstructured mesh created above the real Earth's surface. Terrestrial gravimetric measurements as input data represent the oblique derivative boundary conditions on the Earth's topography. To handle such oblique derivative problem, its tangential components are considered as surface advection terms regularized by a surface diffusion. The FVM numerical solution is fixed to the GOCE-based satellite-only geopotential model on the upper boundary at the altitude of 230 km. The horizontal resolution of the 3D computational domain is 0.002 × 0.002 deg and its discretization in the radial direction is changing with altitude. The created unstructured 3D mesh of finite volumes consists of 454,577,577 unknowns. The FVM numerical solution of FGBVP on such a detailed mesh leads to large-scale parallel computations requiring 245 GB of internal memory. It results in the disturbing potential obtained in the whole 3D computational domain. Its values on the discretized Earth's surface are transformed into the local quasigeoid model that is tested at 404 GNSS/levelling benchmarks. The standard deviation of residuals is 2.8 cm and decreases to 2.6 cm after removing 9 identified outliers. It indicates high accuracy of the obtained FVM-based local quasigeoid model in Slovakia.

Download Full-text