-
Notifications
You must be signed in to change notification settings - Fork 38
/
README.CICE
43 lines (34 loc) · 2.03 KB
/
README.CICE
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
The status of the CICE 5 coupling
I assume one would download CICE 5.1 and place it somewhere else. I have
a copy of clean_ice and comp_ice in the SeaIce directory. Both point to
the place you put your copy of CICE with the SRCDIR variable. They also
need to know where the compiled code goes in SYSTEM_USERDIR.
I have placed cice_cpl.diffs in the SeaIce directory, containing diffs
from the distribution CICE code (apply with "patch -p1 < cice_cpl.diffs"
from your SRCDIR). Also in that directory is a python script to create
a Netcdf grid file for CICE from the ROMS grid file. It is read with the
"roms_grid" and "nc" options for reading grids.
Since CICE only computes vertical fluxes where there's ice, I had the
scatter subroutine scatter by using aicea to partition between fluxes
from CICE and fluxes from the ROMS flux computations.
To build: execute SeaIce/comp_ice, then the ROMS build procedure. If you
have CICE_MODEL #defined, then it will want to link in libCICE.a. Make
sure to point the makefile to the location of this file and its .mod
friends with MY_CICE_DIR - this matches OBJDIR from comp_ice. (If you
use the build scripts, they need to set MY_CICE_DIR.)
CICE compiles in the tile size and number of ice categories, etc. If
these change, you need to run first clean_ice then comp_ice. The code
in SeaIce/Extra/ice_fakecpl.F uses modules from CICE and needs to be
recompiled after tile size changes too - I hand-coded the dependencies
in the makefile.
Status:
1. Running! It creates roughly the same amount of ice as Paul's ice
model. It is then less efficient at melting it - and Paul's isn't melting
enough to begin with. The ocean heat flux to the ice is computed in
ice_frazil.F, similar to how ice_mk.h does it for Paul's model.
cppdefs:
1. CICE_MODEL should be on, ICE_MODEL should be off.
2. CICE wants snowfall, so turn on SNOWFALL to read it from forcing
files. Otherwise hack in ANA_SNOW and all that.
3. If your rain field is total precip and you have a separate snow
field, subtract the snow from the rain with SNOW_FROM_RAIN.