/[MITgcm]/MITgcm/verification/lab_sea/README

Diff of /MITgcm/verification/lab_sea/README

Parent Directory | Revision Log | View Revision Graph Revision Graph | View Patch Patch

-revision 1.1.2.9 by dimitri,
Wed Apr 30 04:55:13 2003 UTC
+revision 1.1.2.10 by dimitri,
Thu May 22 20:49:14 2003 UTC
 Line 142 
 of open-water bulk fluxes are carried ou
  To run Experiment 1:
    cd ../../../verification/lab_sea/input
    ln -sf DATA.10hours data
-   ln -sf DATA.SEAICE.adi data.seaice
+   ln -sf DATA.SEAICE.lsr data.seaice
    mitgcmuv_1 >&! output.txt
    mkdir exp1
    mv *tave.0000000010.data exp1
  Use matlab script lookat_exp1.m to compare the output
- of exp1 with that from release1_p10 sea-ice code:
+ of exp1 with that from release1_p16 sea-ice code:
    cd ../../../verification/lab_sea/matlab
    matlab
    lookat_exp1
 Line 157 
 of exp1 with that from release1_p10 sea-
  Instructions for running Experiment 2
  =====================================
- This is a 1-cpu test of the LSR solver.  The solution is
+ This is a 1-cpu test of the ADI solver.  The solution is
- compared to that of experiment 1, which used the ADI solver.
+ compared to that of experiment 1, which used the LSR solver.
  To run Experiment 2:
    cd ../../../verification/lab_sea/input
    ln -sf DATA.10hours data
-   ln -sf DATA.SEAICE.lsr data.seaice
+   ln -sf DATA.SEAICE.adi data.seaice
    mitgcmuv_1 >&! output.txt
    mkdir exp2
    mv *tave.0000000010.data exp2
 Line 201 
 results from exp3a and exp3b:
    matlab
    lookat_exp3
- If the dynamic ice solvers are correct, they should converge to a
+ Both solvers should and do converge to solutions
- solution that is constant at each latitude.
+ that are constant at each latitude.  But note that
+ the two solutions are not exactly the same because
+ the two solvers represent slightly different
+ approximations of the viscous/plastic equations.
  Instructions for running Experiment 4
-Line 214 
 tile edges for sea-ice dynamic solvers.
+Line 217 
 tile edges for sea-ice dynamic solvers.
  To run Experiment 4:
    cd ../../../verification/lab_sea/input
    ln -sf DATA.10hours data
-   ln -sf DATA.SEAICE.adi data.seaice
+   ln -sf DATA.SEAICE.lsr data.seaice
    mpirun -np 2 mitgcmuv_2x1
    mkdir exp4
    mv *tave.0000000010.data exp4
-Line 225 
 the 2-cpu output to that of exp1:
+Line 228 
 the 2-cpu output to that of exp1:
    matlab
    lookat_exp4
- The inaccuracy at the tile boundary can be decreased by increasing
+ The accuracy of the dynamic solvers is increased by
- NPSEUDO in data.seaice at the expense of computation time.
+ reducing parameter LSR_ERROR for the LSR solver and
- Also as sea-ice ages (thickens) the difference at the tiles becomes
+ by increasing parameter NPSEUDO for the ADI solver.
- increasingly small.  For solver pkg/seaice/adi.F a value of NPSEUDO=10
+ The computational cost/accuracy trade-off is quite
- appears adequate for forward integrations but cannot be used for sea-ice
+ high for the ADI solver but relatively small for
- adjoint model.  With NPSEUDO=300 or greater, the tile disappears, but the
+ the LSR solver, which is the default solver.
- computational cost is very large (11 times that of teh forward model).
  Instructions for running Experiment 5
-Line 243 
 convention used by seaice_get_forcing.F
+Line 245 
 convention used by seaice_get_forcing.F
  To run Experiment 5:
    cd ../../../verification/lab_sea/input
    ln -sf      DATA.2years             data
-   ln -sf  DATA.SEAICE.adi      data.seaice
+   ln -sf  DATA.SEAICE.lsr      data.seaice
    ln -sf  evap.labsea1979  evap.labsea1980
    ln -sf   flo.labsea1979   flo.labsea1980
    ln -sf   fsh.labsea1979   fsh.labsea1980
-Line 260 
 the 2-cpu output to SMMR-SSM/I data:
+Line 262 
 the 2-cpu output to SMMR-SSM/I data:
    matlab
    lookat_exp5
- Disclaimer:
+ The comparison here is just for fun.
- The comparison here is just for fun, not really supposed
+ The configuration is too coarse and artificial
- to look anything like the data.
+ to reproduce realistic sea-ice conditions.
- Otherwise it would put a lot of people out of business :-)
  Instructions for running Experiment 6
  =====================================
-Line 285 
 the output of exp6 to that of exp1:
+Line 287 
 the output of exp6 to that of exp1:
    matlab
    lookat_exp6
  Instructions for running Experiment 7
  =====================================
-Line 319 
 This experiment uses pkg/gmredi, pkg/kpp
+Line 322 
 This experiment uses pkg/gmredi, pkg/kpp
  To run Experiment 8:
    cd ../../../verification/lab_sea/input
    ln -sf DATA.10hours_exf data
-   ln -sf DATA.SEAICE.adi_exf data.seaice
+   ln -sf DATA.SEAICE.exf data.seaice
    mitgcmuv_exf_forcing >&! output.txt
    mkdir exp8
    mv *tave.0000000010.data exp8
-Line 341 
 open-water surface fluxes are provided b
+Line 344 
 open-water surface fluxes are provided b
  To run Experiment 9:
    cd ../../../verification/lab_sea/input
    ln -sf DATA.10hours_exf data
-   ln -sf DATA.SEAICE.adi_exf data.seaice
+   ln -sf DATA.SEAICE.exf data.seaice
    mitgcmuv_exf_fluxes >&! output.txt
    mkdir exp9
    mv *tave.0000000010.data exp9
-Line 362 
 with bulk formula computation of evapora
+Line 365 
 with bulk formula computation of evapora
  To run Experiment 10:
    cd ../../../verification/lab_sea/input
    ln -sf DATA.10hours_exf data
-   ln -sf DATA.SEAICE.adi_exf data.seaice
+   ln -sf DATA.SEAICE.exf data.seaice
    mitgcmuv_compute_evap >&! output.txt
    mkdir exp10
    mv *tave.0000000010.data exp10

 Legend:



Removed from v.1.1.2.9
 


changed lines


 
Added in v.1.1.2.10
 Legend:



Removed from v.1.1.2.9
 


changed lines


 
Added in v.1.1.2.10
-Removed from v.1.1.2.9
+Added in v.1.1.2.10

	ViewVC Help
Powered by ViewVC 1.1.22