A Scalable Barotropic Mode Solver for the Parallel Ocean Program

Author(s):  
Yong Hu ◽  
Xiaomeng Huang ◽  
Xiaoge Wang ◽  
Haohuan Fu ◽  
Shizhen Xu ◽  
...  
2021 ◽  
Vol 13 (4) ◽  
Author(s):  
Hyun‐Gyu Kang ◽  
Katherine J. Evans ◽  
Mark R. Petersen ◽  
Philip W. Jones ◽  
Siddhartha Bishnu
Keyword(s):  

2016 ◽  
Vol 46 (6) ◽  
pp. 1963-1985 ◽  
Author(s):  
Lei Wang ◽  
Malte Jansen ◽  
Ryan Abernathey

AbstractThe phase speed spectrum of ocean mesoscale eddies is fundamental to understanding turbulent baroclinic flows. Since eddy phase propagation has been shown to modulate eddy fluxes, an understanding of eddy phase speeds is also of practical importance for the development of improved eddy parameterizations for coarse resolution ocean models. However, it is not totally clear whether and how linear Rossby wave theory can be used to explain the phase speed spectra in various weakly turbulent flow regimes. Using linear analysis, theoretical constraints are identified that control the eddy phase speed in a two-layer quasigeostrophic (QG) model. These constraints are then verified in a series of nonlinear two-layer QG simulations, spanning a range of parameters with potential relevance to the ocean. In the two-layer QG model, the strength of the inverse cascade exerts an important control on the eddy phase speed. If the inverse cascade is weak, the phase speed spectrum is reasonably well approximated by the phase speed of the linearly most unstable mode. A significant inverse cascade instead leads to barotropization, which in turn leads to mean phase speeds closer to those of barotropic-mode Rossby waves. The two-layer QG results are qualitatively consistent with the observed eddy phase speed spectra in the Antarctic Circumpolar Current and may also shed light on the interpretation of phase speed spectra observed in other regions.


2013 ◽  
Vol 43 (3) ◽  
pp. 548-562 ◽  
Author(s):  
K. Shafer Smith ◽  
Jacques Vanneste

Abstract Recent studies indicate that altimetric observations of the ocean’s mesoscale eddy field reflect the combined influence of surface buoyancy and interior potential vorticity anomalies. The former have a surface-trapped structure, while the latter are often well represented by the barotropic and first baroclinic modes. To assess the relative importance of each contribution to the signal, it is useful to project the observed field onto a set of modes that separates their influence in a natural way. However, the surface-trapped dynamics are not well represented by standard baroclinic modes; moreover, they are dependent on horizontal scale. Here the authors derive a modal decomposition that results from the simultaneous diagonalization of the energy and a generalization of potential enstrophy that includes contributions from the surface buoyancy fields. This approach yields a family of orthonormal bases that depend on two parameters; the standard baroclinic modes are recovered in a limiting case, while other choices provide modes that represent surface and interior dynamics in an efficient way. For constant stratification, these modes consist of symmetric and antisymmetric exponential modes that capture the surface dynamics and a series of oscillating modes that represent the interior dynamics. Motivated by the ocean, where shears are concentrated near the upper surface, the authors consider the special case of a quiescent lower surface. In this case, the interior modes are independent of wavenumber, and there is a single exponential surface mode that replaces the barotropic mode. The use and effectiveness of these modes is demonstrated by projecting the energy in a set of simulations of baroclinic turbulence.


2014 ◽  
Vol 7 (1) ◽  
pp. 267-281 ◽  
Author(s):  
B. van Werkhoven ◽  
J. Maassen ◽  
M. Kliphuis ◽  
H. A. Dijkstra ◽  
S. E. Brunnabend ◽  
...  

Abstract. The Parallel Ocean Program (POP) is used in many strongly eddying ocean circulation simulations. Ideally it would be desirable to be able to do thousand-year-long simulations, but the current performance of POP prohibits these types of simulations. In this work, using a new distributed computing approach, two methods to improve the performance of POP are presented. The first is a block-partitioning scheme for the optimization of the load balancing of POP such that it can be run efficiently in a multi-platform setting. The second is the implementation of part of the POP model code on graphics processing units (GPUs). We show that the combination of both innovations also leads to a substantial performance increase when running POP simultaneously over multiple computational platforms.


2021 ◽  
Author(s):  
Jin Hu ◽  
Emmanuel Garza ◽  
Carlos Perez-Arancibia ◽  
Constantine Sideris

2007 ◽  
Vol 39 (10-11) ◽  
pp. 849-864 ◽  
Author(s):  
O. V. Ivanova ◽  
M. Hammer ◽  
R. Stoffer ◽  
E. van Groesen
Keyword(s):  

Sign in / Sign up

Export Citation Format

Share Document