-
Notifications
You must be signed in to change notification settings - Fork 1
/
Copy pathCHANGES
120 lines (90 loc) · 5.03 KB
/
CHANGES
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
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
CM1 Numerical Model, release 21.0 (cm1r21.0)
20 April 2022
Summary of changes.
-------------------------------------------------------------
ccccccccccccccccccccccccccccccccccccccccccccccccccccccccccccc
-------------------------------------------------------------
1. New options/features.
-------------------
cm1r21.0:
- Officially added the "LES with Mesoscale Model" configuration, in which
an inner part of the domain (ie, an "inner fine mesh") uses large-eddy
simulation (LES) and all other parts of the domain use a mesoscale
model configuration in which a planetary boundary layer (PBL)
parameterization is used. User must set cm1setup=4, and should turn
on eddy recycling. See param17 and param18 sections of namelist.input
(and README.namelist) for more information.
- On a restart, users can now interpolate to a different model grid, which
can have higher (or lower) resolution. No flags/options need to be set
by users; rather, if the code detects on a restart that the model grid
is different than it was in the previous simulation (which wrote the
restart file) then all data are interpolated onto the new grid
automatically. See the "Changing resolution on a restart" section of
README.restart for more information.
- Added the option to use the "two-part" subgrid model with time-averages
(rather than only with spatial averages). Useful for hurricane LES, among
other applications with horizontal heterogeneity. See t2p_avg option in
namelist.input.
- Added new pre-configured cases in the "config_files" subroutine:
- Hurricane simulation using the "LES within mesoscale model" setup
(hurricane_les_within_mm)
- High-resolution, small-domain idealized LES with hurricane winds
at a coast (les_HurrCoast)
- Wind tunnel case with an immersed cube
(les_ib_windtunnel)
- Added many more diagnostics into the azimuthal-averaging code, including:
max/min windspeeds; total upward/downward mass flux; fraction of grid
points that exceed certain thresholds (eg, 95% relative humidity, 40 dBZ
reflectivity); hydrometeor fall velocities; PBL tendencies for radial and
tangential velocity components; turbulent fluctuations and variances
(e.g., <w'v'>, <v'v'>); and more.
(azimavg.F)
- Added "domain location" as an output variable for simulations that use
the "adaptive moving domain" option. (So, users easily see how far a
hurricane moved during a simulation, and plot its track, for example.)
- Added a namelist option ("outunits") so user can specify whether they
want the spatial variables in output files (eg, netcdf, grads) to be in
km or meters.
- Moved options for the large-scale (domain-average) nudging code to
namelist.input; see the param19 section.
- Moved options for immersed boundaries to namelist.input; see the param20
section.
- Moved options for the simple hurricane boundary layer setup to
namelist.input; see the param21 section.
- Added namelist options to use a gradual increase (i.e., "ramp up") of
the subgrid turbulence model at a beginning of LES runs when turbulence
has not yet developed. See "ramp_sgs" and "ramp_time" in README.namelist.
- Added some text to the standard output that notifies users when large
arrays are being allocated. Hopefully, this will make it clearer when
CM1 crashes because memory limits have been exceeded.
- Added the ability to write out fall velocities of hydrometeors from the
Thompson microphysics scheme.
-------------------------------------------------------------
ccccccccccccccccccccccccccccccccccccccccccccccccccccccccccccc
-------------------------------------------------------------
2. Changes, fixes, modifications, etc.
-------------------
cm1r21.0:
- Updated the eddy recycling code, based on experiences using it for several
applications. Among other changes, the eddy recycler now captures/injects
perturbations from a time-average, rather than the full fields. Also,
variables related to the eddy recycling code are now included in
namelist.input, for convenience (see param18 section of namelist).
- Added some diagnostics variables to output files called recy_cap and
recy_inj so users can see exactly where the "capture" and "injection"
regions are located when using the eddy recycling code.
- Fixed a problem with the code hanging when there aren't enough grid points
to do domain decomposition.
(param.F)
- Fixed values of wspa and wspan in cm1out_diag files.
(domaindiag.F)
- Fixed a divide-by-zero problem when lsnudge_ramp_time is set to zero.
(solve1.F)
- Tried to fix a sneaky problem where the adaptive time stepping algorithm
will gradually ramp-down the time step to a very, very small value.
Hopefully, this bug is now fixed.
(misclibs.F)
- Fixed a problem that would sometimes cause the model to crash when using
the isnd=17 option (ie, when ignoring the wind profile in input_sounding
files).
(base.F)