CSLAM

Compatible grids

There are four grid resolutions in CAM that interface with CSLAM:

Resolution

Description

ne30pg3_ne30pg3_mg17

Approximately 1 degree CAM-SE-CSLAM

ne30pg2_ne30pg2_mg17

Approximately 1 degree CAM-SE-CSLAM with 1.5 degree physics grid

ne120pg3_ne120pg3_mt13

Approximately 1/4 degree CAM-SE-CSLAM

ne120pg2_ne120pg2_mt12

Approximately 1/4 degree CAM-SE-CSLAM with 3/8 degree physics grid

Location within CESM

cd $CESMROOT
grep -Rl CSLAM ./

  • ./components/cam/tools/topo_tool/cube_to_target/reconstruct.F90

    • Provides functions for performing conservative interpolation between cubed sphere and lat lon grids.

  • ./components/cam/src/dynamics/se/dyn_comp.F90

    • CAM interfaces to the SE Dynamical Core

  • ./components/cam/src/dynamics/se/interp_mod.F90

    • Module containing subroutines and functions for interpolation.

  • ./components/cam/src/dynamics/se/dycore/global_norms_mod.F90

    • Module for computing global integrals and CFL conditions

  • ./components/cam/src/dynamics/se/dycore/prim_advance_mod.F90

    • Contains logic to advance the model a single timestep

  • ./components/cam/src/dynamics/se/dycore/dimensions_mod.F90

    • This is a non-monophyletic module that contains node specification and other things such as, “scaling of viscosity in sponge layer.”

  • ./components/cam/src/dynamics/se/dycore/fvm_mod.F90

    • FVM_MOD File for the fvm project

  • ./components/cam/src/dynamics/se/dycore/prim_driver_mod.F90

    • Primary driver mod?

  • ./components/cam/src/dynamics/se/dycore/fvm_mapping.F90

    • Two things in this module:

      1. pg2->pg3 mapping as discussed in Herrington et al., 2019a 1 . The pg3 grid divides each GLL grid cell into 3x3 control volumes, while the pg2 grid divides each GLL grid cell into 2x2 control volumes. Herrington et al., 2019a claim that, “the effective resolution of the model is not degraded through the use of a coarser-resolution physics grid. Since the physics makes up about half the computational cost of the conventional CAM-SE-CSLAM configuration, the coarser physics grid may allow for significant cost savings with little to no downside.”

      2. pg3->GLL and GLL->pg3 mapping, (Herrington et al., 2019b 2 )

  • ./components/cam/src/dynamics/se/dycore/fvm_consistent_se_cslam.F90

  • ./components/cam/src/dynamics/se/dycore/prim_advection_mod.F90

  • ./components/cam/src/dynamics/se/dycore/prim_state_mod.F90

  • ./components/cam/src/dynamics/se/dycore/hybrid_mod.F90

  • ./components/cam/src/dynamics/se/dycore/fvm_analytic_mod.F90

  • ./components/cam/src/dynamics/se/dycore/fvm_control_volume_mod.F90

  • ./components/cam/src/dynamics/se/restart_dynamics.F90

  • ./components/cam/src/dynamics/se/dyn_grid.F90

  • ./components/cam/src/dynamics/se/dp_mapping.F90

  • ./ChangeLog

The meaning of the physics grid

When the documentation talks about the “physics” grid when CAM-SE is being used, it is referring to the finite volume method grid upon which CSLAM is being executed:

vim ./components/cam/doc/ChangeLog
[...]
components/cam/src/dynamics/se/dyn_comp.F90
[...]
the loop that sets analytic ICs for constituents directly on the
physics grid has been removed.  Instead all constituents are initially
set on the GLL grid, then mapped to the physics grid when CSLAM is used.
[...]

Running the physics grid at lower resolution

Herrington et al., 2019a 1 show that the physics grid can be run at lower resolution on the pg2 grid (which subdivides the cube sphere into 2x2 control volumes) with negligible ill-effects when compared to running at a higher resolution on the pg3 grid (which subdivides the cube sphere into 3x3 control volumes).

References

1(1,2)

Herrington, A. R., P. H. Lauritzen, K. A. Reed, S. Goldhaber, and B. E. Eaton, 2019a: Exploring a Lower-Resolution Physics Grid in CAM-SE-CSLAM. Journal of Advances in Modeling Earth Systems, 11, 1894–1916, https://doi.org/10.1029/2019MS001684.

2

Herrington, A. R., P. H. Lauritzen, M. A. Taylor, S. Goldhaber, B. E. Eaton, J. T. Bacmeister, K. A. Reed, and P. A. Ullrich, 2019b: Physics–Dynamics Coupling with Element-Based High-Order Galerkin Methods: Quasi-Equal-Area Physics Grid. Monthly Weather Review, 147, 69–84, https://doi.org/10.1175/MWR-D-18-0136.1.