forked from NOAA-EMC/global-workflow
-
Notifications
You must be signed in to change notification settings - Fork 0
/
Release_Notes.gfs.v15.2.0.txt
261 lines (180 loc) · 10.6 KB
/
Release_Notes.gfs.v15.2.0.txt
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
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
GFS v15.2.0 RELEASE NOTES
PRELUDE (taken from GFS v15.2.0 SCN)
GFS version 15.1 was implemented into operation at the 12Z cycle on June 12, 2019. It was the first
GFS implementation with the finite volume cubed-sphere (FV3) dynamical core as the Weather Service’s
Next Generation Global Prediction System (NGGPS).
GFS version 15.2 is a minor upgrade. The major change to the system is to ingest new and replacement
satellite observations for data assimilation. It also contains a few other minor upgrades and bug fixes.
1) Assimilate new satellite observations
* GOES-17 AMVs
* GOES-17 has already replaced GOES-15 as the operational GOES-West satellite. Distribution of
GOES-15 products is expected to cease around December 2019 and active assimilation of GOES-17
AMVs is required to prevent a gap in data coverage.
* Metop-C AMSU and MHS
* Metop-C is now the primary satellite in the 9:30 AM orbit. Assimilation of these data provide
additional impact as well as adding robustness to the system.
* KOMPSAT-5 (GPS-RO)
* Provides additional robustness to the system.
* Addition changes are made to VIIRS AMV ingest code to allow continued use after an expected change
to the BUFR format.
2) Assimilate buoyb sea-surface temperature (SST) data
TAC2BUFR changes in the buoy network resulted in a reduction of available SST measurements from buoys
to 10% or less of expected levels. Obsproc and GSI changes were required to restore data counts to
previous levels.
3) New product: Graphical Turbulence Guidance (GTG)
Global Turbulence product generated using NCAR GTG algorithm will start being disseminated on NOMADS.
The product will be available 3 hourly from F06 to F36 and horizontal resolution will be a quarter
degree. gfs.t{CC}Z.gtg.0p25.grb2f{HHH}
4) Update the Unified Post Processor(UPP) to address a mask issue of several land surface fields over
water bodies in grib2 products.
This update will make GFS.v15.2 p-grid products to be more consistent with GFS.v14 products, and
remove spurious soil moisture along coastlines. These land surface fields include Soil Moisture,
Soil Temperature, Liquid Volumetric Soil Moisture, WEASD, Snow Depth, Water Runoff, GFLUX Ground
Heat Flux, WILT Wilting Point, and FLDCP Field Capacity.
Affected product files are:
gfs.t{CC}Z.pgrb2.0p25.F{HHH}
gfs.t{CC}Z.pgrb2b.0p25.g{HHH}
gfs.t{CC}Z.pgrb2.0p50.F{HHH}
gfs.t{CC}Z.pgrb2b.0p50.g{HHH}
gfs.t{CC}Z.pgrb2.1p00.F{HHH}
gfs.t{CC}Z.pgrb2b.1p00.g{HHH}
gfs.t{CC}Z.sfluxgrbf{HHH}.grib2
Where CC is cycle for 00, 06, 12, 18 UTC, and HHH is forecast hour.
5) The vessel icing program uses OISST as input. OISST will not be ported from WCOSS Phase 1 to Phase 3
after Phase 1 is decommissioned in 2020. A decision was made to move the vessel icing program
within the Unified Post Processor(UPP) and use GFS forecast skin temperature as input. Current vessel
icing product in operation (sice.tCCz.siceg) has a 1-deg resolution and is disseminated to the public at
ftp://ftp.ncep.noaa.gov/pub/data/nccf/com/omb/prod/sice.yyyymmdd/
New vessel icing product will be included as a variable (ICEG) in GFS p-grid products gfs.tCCz.pgrb2.xpxx.fxxx
and gfs.tCCz.pgrb2b.xpxx.fxxx at 0.25, 0.5, and 1.0-deg resolutions, and be disseminated to the public at
ftp://ftp.ncep.noaa.gov/pub/data/nccf/com/gfs/prod/gfs.yyyymmdd/
6) Added three stations to station time series bufr soundings:
006011 62.02N 6.76W TOR 00 Thorshvan, Denmark 54 Usr Rqst 4-19
999913 15.51S 128.15E WYN 00 Wyndham aerodrome Australia 4 Usr Rqst 1-19
999914 57.48N 7.36W EGPL 00 Benbecula, Scotland, UK 6 Usr Rqst 1-19
The affected output files are:
gfs_yyyymmddhh.sfc
gfs_yyyymmddhh.snd
gfs.tCCz.bufrsnd.tar.gz
Three additional files for the stations:
bufr.006011.yyyymmddhh
bufr.999913.yyyymmddhh
bufr.999914.yyyymmddhh
7) Reduction of water temperature biases in small lakes.
For small lakes adequate observations do not always exit to support the analysis of lake surface
temperature, often leading to significant departures from both the climatology and real-time observation.
Two changes were introduced to ensure that lake temperatures do not deviate from the climatology when
observations are not available. The first change is to replace a surface mask file at 0.5-degree
resolution with the one on the T1534 Gaussian grid (~13km) to prevent unrealistic SST climatology
from being used for updating the background of the near sea-surface temperature analysis over small
water bodies, such as those in the Great Salt Lake. The second change is to reduce the relaxation
time scale of the SST to climatology in GDAS forecast step from 90 days to 10 days.
8) Changes to NOAAPORT/SBN
Product Removals
* GADS FAX product which contains tropical surface analysis in TIF format with G4 compression.
IMPLEMENTATION INSTRUCTIONS
* NOAA Vlab GIT is used to manage GFS.v15.2 code. The SPA(s) handling the GFS.v15.2 implementation need to have
permission to clone Vlab gerrit repositories. So far Wojciech Cencek has been given access to all GFS.v15.2
related git repositories. Please contact [email protected] or [email protected] if there is any VLAB
access issue and/or the individual code managers listed under item #6) below. Please follow the following
steps to install the package on WCOSS DELL
1) cd $NWROOTp3
2) mkdir gfs.v15.2.0
3) cd gfs.v15.2.0
4) git clone --recursive gerrit:global-workflow .
5) git checkout feature/dev-v15.2
6) cd sorc
7) ./checkout.sh
This script extracts the following GFS components from gerrit
MODEL -- tag nemsfv3_gfsv15.2.1 [email protected]
GSI -- tag fv3da_gfs.v15.2.0 [email protected]
UPP -- tag ncep_post_gtg.v1.1.4 [email protected]
WAFS -- tag gfs_wafs.v5.0.9 [email protected]
8) ./build_all.sh
*This script compiles all GFS components. Runtime output from the build for each package is written
to log files in directory logs. To build an individual program, for instance, gsi, use build_gsi.sh.
9) ./link_fv3gfs.sh nco dell
* Note: 1) ecflow suite definition and scripts are saved in gfs.v15.2.0/ecflow/ecf
2) ncep_post_gtg.v1.1.4 contains restricted GTG (Graphic Turbulence Guidance) code provided by
NCAR. Please do not post the GTG code in any public domain.
JOB CHANGES
* See docs/Release_Notes.gfs_downstream.v15.2.0.txt
SORC CHANGES
* sorc/
* checkout.sh - update to check out the following tags
* NEMSfv3gfs nemsfv3_gfsv15.2.1
* ProdGSI fv3da_gfsv15.2.0
* EMC_post_gtg ncep_post_gtg.v1.1.4
* EMC_gfs_wafs gfs_wafs.v5.0.9
* sorc/global_chgres.fd/
* sfcsub.F - bug fix. set logical variable (lmask) to be false for a land surface variable
* sorc/global_cycle.fd/
* sfcsub.F - bug fix. set logical variable (lmask) to be false for a land surface variable
SCRIPT CHANGES
* scripts/
* run_gfsmos_master.sh.cray - remove reference to COMROOTp1
* run_gfsmos_master.sh.dell - remove reference to COMROOTp1
* additional script changes documented in docs/Release_Notes.gfs_downstream.v15.2.0.txt
PARM/CONFIG CHANGES
* parm/
* Two files are modified to set a
* parm/config/config.base.emc.dyn - set 10 day relaxaion time scale to SST climatology in GDAS forecast
* parm/config/config.base.nco.static - set 10 day relaxaion time scale to SST climatology in GDAS forecast
* Two files were modified for adding three bufr stations:
* parm/product/bufr_stalist.meteo.gfs
* parm/product/bufr_stalist.meteo.gfs3
FIX CHANGES
* Files in fix/fix_gsi altered by GFS DA v15.2. See GFS DA v15.2 release notes
(sorc/gsi.fd/doc/Release_Notes.gfs_da.v15.2.0.txt) for details
PRODUCT CHANGES
* see SCN
RESOURCE INFORMATION
* Frequency of run
* No change from GFS v15.1
* Commonly used libraries, compiler, and modules are defined in gfs.v15.2.0/modulefiles. For nemsfv3gfs, gsi, upp, wafs
they maintain their own module files under gfs.v15.2.0/sorc/(fv3gfs gsi gfs_post global_wafs).fd/modulefiles
* GSI updated to use bufr/11.3.0
* Data retention time under $COMROOTp3 for GFS.v15.2 should be the same as GFS.v15.1.
* Disk space:
* About 4 Gb more per gdas cycle due to additional observation data in gdas and enkf diagnostic files
* Computational resources and run times:
* Due to processing additional observation data the runtime for the following jobs increases with
respect to GFS v15.1 as noted below
* gfs_analysis : about 30 seconds longer (27.4 minutes for GFS v15.1 -vs- 27.9 minutes for GFS v15.2)
* gdas_enkf_select_obs : about 1 minute longer (3.7 for GFS v15.1 -vs- 4.7 for GFS v15.2)
* gdas_enkf_innovate_obs_grp*: about 30 seconds longer (14.8 for GFS v15.1 -vs - 15.3 for GFS v15.2)
* gdas_enkf_update : about 20 seconds longer (6.4 for GFS v15.1 -vs- 6.7 for GFS v15.2)
PRE-IMPLEMENTATION TESTING REQUIREMENTS
* Which production jobs should be tested as part of this implementation?
* The entire GFS v15.2 package needs to be installed and tested. EMC can run the same date
and compare NCO and EMC output to confirm the EMC and NCO tests reproduce each other
* Does this change require a 30-day evaluation?
* No.
* Suggested evaluators
* Please contact the following EMC staff for the indicated components
[email protected] - MODEL
[email protected] - DA
[email protected] - UPP
[email protected] - WAFS
[email protected] - downstream products
DISSEMINATION INFORMATION
* Where should this output be sent?
* No change from GFS v15.1
* Who are the users?
* No change from GFS v15.1
* Which output files should be transferred from PROD WCOSS to DEV WCOSS?
* No change from GFS v15.1
* Directory changes
* No change from GFS v15.1
* File changes.
* See SCN
HPSS ARCHIVE
No change from GFS v15.1
JOB DEPENDENCIES & FLOW DIAGRAM
* No change from GFS v15.1
===========
Prepared by
Fanglin.Yang@noaa