3 |
Notes on tags used in MITgcmUV |
Notes on tags used in MITgcmUV |
4 |
============================== |
============================== |
5 |
|
|
6 |
|
checkpoint53b_pre |
7 |
|
|
8 |
|
o created pkg/mom_common in preparation for variable viscosity code |
9 |
|
o renamed and split ini_mnc_io |
10 |
|
o pkg/land: only liquid water can diffuse or run-off |
11 |
|
o AIM using land: soil wetness availability account for snow |
12 |
|
o added printMapIncludesZeros to EEPARMS namelist |
13 |
|
|
14 |
|
checkpoint53a_post |
15 |
|
o restore TAMC compatiility for exf_constants.h |
16 |
|
(no usage of _d in .h) |
17 |
|
o Additional storing in KPP in an attempt to fix |
18 |
|
TAMC problem with adjoint of KPP code. |
19 |
|
(no success...) |
20 |
|
Identical code works fine with TAF. |
21 |
|
o Fixed verification/natl_box_adjoint experiments. |
22 |
|
Initial conditions for Experiments 4 and 5 are now |
23 |
|
set to "SEAICE_initialHEFF=0", which is more reasonable, |
24 |
|
and hence gives results closer to that of Experiment 2. |
25 |
|
Experiment 6 requires "SEAICE_initialHEFF=1" and is working once again. |
26 |
|
|
27 |
|
checkpoint53 |
28 |
|
o stable checkpoint with up to date adjoint |
29 |
|
TLM tests only run for 1st grdchk at the moment. Its unclear if |
30 |
|
this is a code problem or a problem with resetting code for g_ |
31 |
|
variables that is being autogenerated wrong. |
32 |
|
Otherwise: |
33 |
|
o code is testing well on Linux (P4 and AMD), SGI Alitix and Irix. |
34 |
|
o IBM and HPQ tests could not be run because NCAR (IBM) and NCCS (HPQ) |
35 |
|
are both still in chaos following their security messes. |
36 |
|
|
37 |
checkpoint52n_post |
checkpoint52n_post |
38 |
o various build (mostly genmake2) changes including: |
o various build (mostly genmake2) changes including: |
39 |
- ability to build/run more easily within Cygwin and MacOS X |
- ability to build/run more easily within Cygwin and MacOS X |