skip to main content
OSTI.GOV title logo U.S. Department of Energy
Office of Scientific and Technical Information

Title: A Reduced Grid Method for a Parallel Global Ocean General Circulation Model

Abstract

A limitation of many explicit finite-difference global climate models is the timestep restriction caused by the decrease in cell size associated with the convergence of meridians near the poles. A computational grid in which the number of cells in the longitudinal direction is reduced toward high-latitudes, keeping the longitudinal width of the resulting cells as uniform as possible and increasing the allowable timestep, is applied to a three-dimensional primitive equation ocean-climate model. This ''reduced'' grid consists of subgrids which interact at interfaces along their northern and southern boundaries, where the resolution changes by a factor of three. Algorithms are developed to extend the finite difference techniques to this interface, focusing on the conservation required to perform long time integrations, while preserving the staggered spatial arrangement of variables and the numerics used on subgrids. The reduced grid eliminates the common alternative of filtering high-frequency modes from the solution at high-latitudes to allow a larger timestep and reduces execution time per model step by roughly 20 percent. The reduced grid model is implemented for parallel computer architectures with two-dimensional domain decomposition and message passing, with speedup results comparable to those of the original model. Both idealized and realistic model runs are presentedmore » to show the effect of the interface numerics on the model solution. First, a rectangular, mid-latitude, at-bottomed basin with vertical walls at the boundaries is driven only by surface wind stress to compare three resolutions of the standard grid to reduced grid cases which use various interface conditions. Next, a similar basin with wind stress, heat, and fresh water forcing is used to compare the results of a reduced grid with those of a standard grid result while exercising the full set of model equations. Finally, global model runs, with topography, forcing, and physical parameters similar to those used for ocean-climate studies, are advanced to a near equilibrium state for both the reduced grid and the standard grid. Differences between the two are presented for typical fields of interest, and very little degradation of the solution due to the reduced grid is observed.« less

Authors:
 [1]
+ Show Author Affiliations
  1. Univ. of California, Davis, CA (United States)
Publication Date:
Research Org.:
Lawrence Livermore National Lab. (LLNL), Livermore, CA (United States)
Sponsoring Org.:
USDOE Office of Science (SC)
OSTI Identifier:
791655
Report Number(s):
UCRL-LR-136903
TRN: US200302%%751
DOE Contract Number:
W-7405-Eng-48
Resource Type:
Thesis/Dissertation
Resource Relation:
Other Information: TH: Thesis (Ph.D.); Submitted to the Univ. of California, Davis, CA (US); PBD: 1 Dec 1999
Country of Publication:
United States
Language:
English
Subject:
16 TIDAL AND WAVE POWER; ALGORITHMS; CLIMATE MODELS; COMPUTER ARCHITECTURE; CONVERGENCE; FOCUSING; FRESH WATER; GENERAL CIRCULATION MODELS; RESOLUTION; TOPOGRAPHY

Citation Formats

Wickett, Michael Everett. A Reduced Grid Method for a Parallel Global Ocean General Circulation Model. United States: N. p., 1999. Web. doi:10.2172/791655.
Copy to clipboard
Wickett, Michael Everett. A Reduced Grid Method for a Parallel Global Ocean General Circulation Model. United States. doi:10.2172/791655.
Copy to clipboard
Wickett, Michael Everett. Wed . "A Reduced Grid Method for a Parallel Global Ocean General Circulation Model". United States. doi:10.2172/791655. https://www.osti.gov/servlets/purl/791655.
Copy to clipboard
@article{osti_791655,
title = {A Reduced Grid Method for a Parallel Global Ocean General Circulation Model},
author = {Wickett, Michael Everett},
abstractNote = {A limitation of many explicit finite-difference global climate models is the timestep restriction caused by the decrease in cell size associated with the convergence of meridians near the poles. A computational grid in which the number of cells in the longitudinal direction is reduced toward high-latitudes, keeping the longitudinal width of the resulting cells as uniform as possible and increasing the allowable timestep, is applied to a three-dimensional primitive equation ocean-climate model. This ''reduced'' grid consists of subgrids which interact at interfaces along their northern and southern boundaries, where the resolution changes by a factor of three. Algorithms are developed to extend the finite difference techniques to this interface, focusing on the conservation required to perform long time integrations, while preserving the staggered spatial arrangement of variables and the numerics used on subgrids. The reduced grid eliminates the common alternative of filtering high-frequency modes from the solution at high-latitudes to allow a larger timestep and reduces execution time per model step by roughly 20 percent. The reduced grid model is implemented for parallel computer architectures with two-dimensional domain decomposition and message passing, with speedup results comparable to those of the original model. Both idealized and realistic model runs are presented to show the effect of the interface numerics on the model solution. First, a rectangular, mid-latitude, at-bottomed basin with vertical walls at the boundaries is driven only by surface wind stress to compare three resolutions of the standard grid to reduced grid cases which use various interface conditions. Next, a similar basin with wind stress, heat, and fresh water forcing is used to compare the results of a reduced grid with those of a standard grid result while exercising the full set of model equations. Finally, global model runs, with topography, forcing, and physical parameters similar to those used for ocean-climate studies, are advanced to a near equilibrium state for both the reduced grid and the standard grid. Differences between the two are presented for typical fields of interest, and very little degradation of the solution due to the reduced grid is observed.},
doi = {10.2172/791655},
journal = {},
number = ,
volume = ,
place = {United States},
year = {Wed Dec 01 00:00:00 EST 1999},
month = {Wed Dec 01 00:00:00 EST 1999}
}
Copy to clipboard
Thesis/Dissertation:
View Thesis/Dissertation
DOI:  10.2172/791655
Other availability
Please see Document Availability for additional information on obtaining the full-text document. Library patrons may search WorldCat to identify libraries that hold this thesis or dissertation.
Save / Share:
Export Metadata
Save to My Library
You must Sign In or Create an Account in order to save documents to your library.

  • An ocean model with adaptive mesh refinement (AMR) capability is presented for simulating ocean circulation on decade time scales. The model closely resembles the LLNL ocean general circulation model with some components incorporated from other well known ocean models when appropriate. Spatial components are discretized using finite differences on a staggered grid where tracer and pressure variables are defined at cell centers and velocities at cell vertices (B-grid). Horizontal motion is modeled explicitly with leapfrog and Euler forward-backward time integration, and vertical motion is modeled semi-implicitly. New AMR strategies are presented for horizontal refinement on a B-grid, leapfrog time integration,more » and time integration of coupled systems with unequal time steps. These AMR capabilities are added to the LLNL software package SAMRAI (Structured Adaptive Mesh Refinement Application Infrastructure) and validated with standard benchmark tests. The ocean model is built on top of the amended SAMRAI library. The resulting model has the capability to dynamically increase resolution in localized areas of the domain. Limited basin tests are conducted using various refinement criteria and produce convergence trends in the model solution as refinement is increased. Carbon sequestration simulations are performed on decade time scales in domains the size of the North Atlantic and the global ocean. A suggestion is given for refinement criteria in such simulations. AMR predicts maximum pH changes and increases in CO 2 concentration near the injection sites that are virtually unattainable with a uniform high resolution due to extremely long run times. Fine scale details near the injection sites are achieved by AMR with shorter run times than the finest uniform resolution tested despite the need for enhanced parallel performance. The North Atlantic simulations show a reduction in passive tracer errors when AMR is applied instead of a uniform coarse resolution. No dramatic or persistent signs of error growth in the passive tracer outgassing or the ocean circulation are observed to result from AMR.« less

  • Two sets of idealized simulations of an atmospheric general circulation model are investigated in order to understand the sensitivities of climate to massive changes in atmospheric carbon dioxide concentration. C02 level is changed from one tenth to one hundred times the present-day amount. Cloud cover is predicted in the first set while it is prescribed in the other set. Comparing the two experimental sets clarifies the influence of the cloud feedback process on the climate sensitivities. The authors investigate changes in the globally and zonally averaged surface air temperature, precipitation and evaporation as the climate sensitivities in this study. Mainmore » features in the sensitivities are the following: (1) The cloud feedback process in this model tremendously enhances the global mean temperature sensitivity by interacting with the water vapor feedback process; (2) The change in meridional moist static energy transport tends to reduce the meridional temperature gradient at mid-latitude where the gradient is large; (3) The sensitivity of global mean precipitation (= evaporation) decreases at very high temperatures because the absorbed solar radiation at the ground is entirely used for evaporation; and (4) The latitudinal pattern of precipitation minus evaporation does not change very much as C02 level changes. In order to explain the above features, the authors present analyses on changes in the radiative fluxes at the top of the atmosphere, the surface energy balance, the internal climate variables and the meridional moist static energy transport. They also critically evaluate their results and discuss some paleoclimatic implications. A review on formulations of energy balance analyses is included as an appendix.« less

  • Water vapor not only participates in the radiative balance of the atmosphere and in cloud formation, it also participates in stratospheric chemistry and, due to the strong dependence of saturation on temperature, serves as a tracer for exposure of air to cold temperatures. The application of general circulation models (GCM`s) to stratospheric chemistry and transport both enables and requires a thorough investigation of stratospheric water vapor. The National Center for Atmospheric Research has redesigned its GCM, the CCM2, to enable studies of stratospheric chemistry and tracer transport, including that of water vapor. Simple methane chemistry provides an adequate representation ofmore » the upper stratospheric water vapor source in the CCM2. The CCM2`s water vapor distribution and seasonality compare favorably with observations in many respects, and the CCM2 fills gaps in the obsevations, yielding some new insights. For example, southern polar dehydration can affect midlatitude water mixing ratios by a few tenths of a ppmv. The annual cycle of water vapor in the tropical and subtropical lower stratosphere is dominated by drying at the tropical tropopause. Water vapor has a very long adjustment time, a factor 2-4 longer than for methane, a common long-lived tracer. In the lower stratosphere, however, two model deficiencies have a profound impact on simulated water vapor. The first is a cold temperature bias in the winter polar stratosphere, a deficiency common to GCM`s. The cold bias produces excessive dehydration in the southern hemisphere. This deficiency can be eliminated fairly simply by setting a minimum vapor pressure. The second deficiency, however, is not so easily remedied. Stratosphere-troposphere exchange in the tropics has a different character from the observed; for example, too little mass flux occurs under low mixing ratio conditions, so that the stratosphere is somewhat too moist.« less

  • The first phase of the proposed work is largely completed on schedule. Scientists at the San Diego Supercomputer Center (SDSC) succeeded in putting a version of the Hamburg isopycnal coordinate ocean model (OPYC) onto the INTEL parallel computer. Due to the slow run speeds of the OPYC on the parallel machine, another ocean is being model used during the first part of phase 2. The model chosen is the Large Scale Geostrophic (LSG) model form the Max Planck Institute.

  • Determine the structure of oceanic natural variability at time scales of decades to centuries; characterize the physical mechanisms responsible for the variability; determine the relative importance of heat, fresh water, and moment fluxes on the variability; determine the predictability of the variability on these times scales.