A few sea ice related issues in the NSST analysis: Issue #18 #20

XuLi-NOAA · 2019-09-30T13:54:32Z

The UFS_UTILS Issue #18 is ready for review

Modify to resolve a few sea ice related issues in NSST analysis

GeorgeGayno-NOAA · 2019-09-30T15:08:24Z

sorc/global_cycle.fd/read_write_data.f90

+  call baclose(lu_sst,iret)
+  if (iret /= 0 ) then
+     write(6,*)'get_tf_clm_dim:  ***error*** close sst file'
+     stop


I would use 'mpi_abort' instead of 'stop'.

OK, I will use mpi_abort instead of stop

yangfanglin · 2019-09-30T15:34:26Z

ush/global_cycle.sh

@@ -234,6 +234,7 @@ HOMEgfs=${HOMEgfs:-$BASEDIR/gfs_ver.${gfs_ver}}
 EXECgfs=${EXECgfs:-$HOMEgfs/exec}
 FIXfv3=${FIXfv3:-$HOMEgfs/fix/fix_fv3_gmted2010/$CASE}
 FIXam=${FIXam:-$HOMEgfs/fix/fix_am}
+FIXsst=${FIXsst:-/scratch4/NCEPDEV/da/noscrub/Xu.Li/data/rtgsst/fix}


Default FIXsst needs to be defined as $Homegfs/fix/fix_sst or
add all FIXsst files to $HOMEgfs/fix/fix_am directory on all platforms,

FIXsst has been changed to be FIXam (are you looking at the newest version?)

yangfanglin

Xu,
Are this these changes going to be included in GFS.v16 ? If they are, have you run any experiments to show the impact?

XuLi-NOAA · 2019-09-30T16:08:40Z

Fanglin, Yes, and the tests look good. Xu

…

On Mon, Sep 30, 2019 at 11:43 AM Fanglin Yang ***@***.***> wrote: ***@***.**** commented on this pull request. Xu, Are this these changes going to be included in GFS.v16 ? If they are, have you run any experiments to show the impact? — You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub <#20>, or mute the thread <https://github.com/notifications/unsubscribe-auth/ANEMLJUPUKPVMGX7FO52BVDQMINB5ANCNFSM4I33ID6A> .

yangfanglin · 2019-09-30T17:10:06Z

Xu, would you like to present your tests at either the DA meeting or the GFS.v16 biweekly technical meeting ?

Other than that, the code looks fine to me.

GeorgeGayno-NOAA · 2019-09-30T17:21:32Z

I am not sure your branch is up-to-date with 'develop'. Can you do a merge to be certain?

XuLi-NOAA · 2019-09-30T17:33:16Z

Yes, I can do it, either one of the two meetings.

…

On Mon, Sep 30, 2019 at 1:10 PM Fanglin Yang ***@***.***> wrote: Xu, would you like to present your tests at either the DA meeting or the GFS.v16 biweekly technical meeting ? Other than that, the code looks fine to me. — You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub <#20>, or mute the thread <https://github.com/notifications/unsubscribe-auth/ANEMLJXDOWDPCTHEKBDZUDLQMIXG7ANCNFSM4I33ID6A> .

GeorgeGayno-NOAA · 2019-09-30T18:22:53Z

After your last update, cycle won't build:

read_write_data.f90(1576): error #6404: This name does not have a type, and must have an explicit type. [MPI_COMM_WORLD] CALL MPI_ABORT(MPI_COMM_WORLD, 111)

You need to add include "mpif.h"

…write_data.f90

XuLi-NOAA · 2019-09-30T18:33:09Z

Yes, I just fixed it and it builds now

…

On Mon, Sep 30, 2019 at 2:22 PM GeorgeGayno-NOAA ***@***.***> wrote: After your last update, cycle won't build: read_write_data.f90(1576): error #6404: This name does not have a type, and must have an explicit type. [MPI_COMM_WORLD] CALL MPI_ABORT(MPI_COMM_WORLD, 111) You need to add include "mpif.h" — You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub <#20>, or mute the thread <https://github.com/notifications/unsubscribe-auth/ANEMLJTPLI255TFHQOI56STQMI7X5ANCNFSM4I33ID6A> .

GeorgeGayno-NOAA · 2019-10-03T15:07:03Z

Xu, I ran stand-alone cycle using 20190730 data and compared your branch against 'develop'. Using a netcdf tool, I see differences in 'tsea', 'tisfc', 'tref', 'tfinc' and 'stc'. Is that expected? The 'stc' differences are very large. When viewed, the large differences are at sea ice. Here is a summary of tile 6 differences:

Variable Group  Count         Sum      AbsSum        Min          Max     Range       Mean    StdDev
tsea     /     100014    -2186.55     2186.76   -1.63343    0.0612983   1.69473 -0.0218624  0.030319
tisfc    /      15215    -1114.58     1114.58 -0.0849977 -1.54403e-05 0.0849823 -0.0732554 0.0224935
tref     /     111665    -9272.22     9272.22  -0.267241 -1.22118e-05  0.267229  -0.083036 0.0115932
tfinc    /      11564    -781.941     781.941 -0.0849977 -1.22118e-05 0.0849855 -0.0676186 0.0250677
stc      /     461264 5.19794e+06 5.24694e+06   -8.45253      38.8503   47.3028    11.2689   9.45556

Why are sea ice temperatures changing so much?

XuLi-NOAA · 2019-10-03T16:56:01Z

George, I have talked to Xingren and understand what happened exactly now. As we discussed, the line, soilt_tile(:,ij) = skint_tile(ij) , to assign stc needs to be removed. I have committed to my develop fork. Can you do the check with the statistics you did again? Thanks for finding out this problem. Xu

…

On Thu, Oct 3, 2019 at 11:07 AM GeorgeGayno-NOAA ***@***.***> wrote: Xu, I ran stand-alone cycle using 20190730 data and compared your branch against 'develop'. Using a netcdf tool, I see differences in 'tsea', 'tisfc', 'tref', 'tfinc' and 'stc'. Is that expected? The 'stc' differences are very large. When viewed, the large differences are at sea ice. Here is a summary of tile 6 differences: Variable Group Count Sum AbsSum Min Max Range Mean StdDev tsea / 100014 -2186.55 2186.76 -1.63343 0.0612983 1.69473 -0.0218624 0.030319 tisfc / 15215 -1114.58 1114.58 -0.0849977 -1.54403e-05 0.0849823 -0.0732554 0.0224935 tref / 111665 -9272.22 9272.22 -0.267241 -1.22118e-05 0.267229 -0.083036 0.0115932 tfinc / 11564 -781.941 781.941 -0.0849977 -1.22118e-05 0.0849855 -0.0676186 0.0250677 stc / 461264 5.19794e+06 5.24694e+06 -8.45253 38.8503 47.3028 11.2689 9.45556 Why are sea ice temperatures changing so much? — You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub <#20>, or mute the thread <https://github.com/notifications/unsubscribe-auth/ANEMLJSAAHMK6MJLKCIQ6NDQMYDBRANCNFSM4I33ID6A> .

SMoorthi-emc · 2019-10-03T19:14:10Z

I am ok with commit Moorthi

…

Sent from my iPhone

On Oct 3, 2019, at 12:56 PM, XuLi-NOAA ***@***.***> wrote: George, I have talked to Xingren and understand what happened exactly now. As we discussed, the line, soilt_tile(:,ij) = skint_tile(ij) , to assign stc needs to be removed. I have committed to my develop fork. Can you do the check with the statistics you did again? Thanks for finding out this problem. Xu On Thu, Oct 3, 2019 at 11:07 AM GeorgeGayno-NOAA ***@***.***> wrote: > Xu, I ran stand-alone cycle using 20190730 data and compared your branch > against 'develop'. Using a netcdf tool, I see differences in 'tsea', > 'tisfc', 'tref', 'tfinc' and 'stc'. Is that expected? The 'stc' differences > are very large. When viewed, the large differences are at sea ice. Here is > a summary of tile 6 differences: > > Variable Group Count Sum AbsSum Min Max Range Mean StdDev > tsea / 100014 -2186.55 2186.76 -1.63343 0.0612983 1.69473 -0.0218624 0.030319 > tisfc / 15215 -1114.58 1114.58 -0.0849977 -1.54403e-05 0.0849823 -0.0732554 0.0224935 > tref / 111665 -9272.22 9272.22 -0.267241 -1.22118e-05 0.267229 -0.083036 0.0115932 > tfinc / 11564 -781.941 781.941 -0.0849977 -1.22118e-05 0.0849855 -0.0676186 0.0250677 > stc / 461264 5.19794e+06 5.24694e+06 -8.45253 38.8503 47.3028 11.2689 9.45556 > > Why are sea ice temperatures changing so much? > > — > You are receiving this because you authored the thread. > Reply to this email directly, view it on GitHub > <#20>, > or mute the thread > <https://github.com/notifications/unsubscribe-auth/ANEMLJSAAHMK6MJLKCIQ6NDQMYDBRANCNFSM4I33ID6A> > . > — You are receiving this because your review was requested. Reply to this email directly, view it on GitHub, or mute the thread.

GeorgeGayno-NOAA · 2019-10-03T20:34:07Z

After the bug fix for the sea ice column temperatures, the tile 6 difference (Xu's branch minus 'develop') is:

CHECK 20190730.000000.sfcanl_data.tile6.nc
Variable Group  Count      Sum  AbsSum        Min          Max     Range       Mean    StdDev
tsea     /     100014 -2186.55 2186.76   -1.63343    0.0612983   1.69473 -0.0218624  0.030319
tisfc    /      15215 -1114.58 1114.58 -0.0849977 -1.54403e-05 0.0849823 -0.0732554 0.0224935
tref     /     111665 -9272.22 9272.22  -0.267241 -1.22118e-05  0.267229  -0.083036 0.0115932
tfinc    /      11564 -781.941 781.941 -0.0849977 -1.22118e-05 0.0849855 -0.0676186 0.0250677
stc      /      60860 -4458.33 4458.33 -0.0849977 -1.54403e-05 0.0849823 -0.0732554  0.022493

Are these reasonable?

XuLi-NOAA · 2019-10-04T00:18:14Z

Yes, they are reasonable and as expected. Thanks!

…

On Thu, Oct 3, 2019 at 4:34 PM GeorgeGayno-NOAA ***@***.***> wrote: After the bug fix for the sea ice column temperatures, the tile 6 difference (Xu's branch minus 'develop') is: CHECK 20190730.000000.sfcanl_data.tile6.nc Variable Group Count Sum AbsSum Min Max Range Mean StdDev tsea / 100014 -2186.55 2186.76 -1.63343 0.0612983 1.69473 -0.0218624 0.030319 tisfc / 15215 -1114.58 1114.58 -0.0849977 -1.54403e-05 0.0849823 -0.0732554 0.0224935 tref / 111665 -9272.22 9272.22 -0.267241 -1.22118e-05 0.267229 -0.083036 0.0115932 tfinc / 11564 -781.941 781.941 -0.0849977 -1.22118e-05 0.0849855 -0.0676186 0.0250677 stc / 60860 -4458.33 4458.33 -0.0849977 -1.54403e-05 0.0849823 -0.0732554 0.022493 Are these reasonable? — You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub <#20>, or mute the thread <https://github.com/notifications/unsubscribe-auth/ANEMLJQAG5I4IYRY2CURVT3QMZJMBANCNFSM4I33ID6A> .

GeorgeGayno-NOAA · 2019-10-09T16:47:51Z

Your branch is not up-to-date with 'develop'.

GeorgeGayno-NOAA · 2019-10-09T16:57:12Z

sfcsub.F was updated for v15.2. The interpolation of substrate temperature was changed to not use a land mask. Those changes are not in Xu's branch.

GeorgeGayno-NOAA · 2019-10-09T17:06:54Z

sfcsub.F was recently updated in the forecast model:
see here

These updates should be incorporated into Xu's branch (in both cycle and chgres). sfcsub.F should be consistent between the forecast model and ufs_utils.

XuLi-NOAA · 2019-10-09T17:24:29Z

Hi, I thought it is already included in my fork develop. Thanks for the remind. It is in now. XU

…

On Wed, Oct 9, 2019 at 1:06 PM GeorgeGayno-NOAA ***@***.***> wrote: sfcsub.F was recently updated in the forecast model: see here <https://vlab.ncep.noaa.gov/redmine/projects/comfv3/repository/diff/gfsphysics/physics/sfcsub.F?utf8=%E2%9C%93&rev=01e034fe0b44bbcbafda51f1624e2bded6c1faef&rev_to=3dc7e9b1fc341bae78c4aac37582e4046a330c2d> These updates should be incorporated into Xu's branch (in both cycle and chgres). sfcsub.F should be consistent between the forecast model and ufs_utils. — You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub <#20>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ANEMLJV6XAEVO233ZTKU6LTQNYFS7ANCNFSM4I33ID6A> .

…d the definition of lbms (logical => logical*1)

GeorgeGayno-NOAA · 2019-10-10T19:58:34Z

The sfcsub.F module contains a variable - 'tgice' - that is 271.2. That variable is used in several computations. Also, variable 'tsfomn' is 271.2 and is used to range check the lower bound of open water. Do these need to be changed for salinity?

XuLi-NOAA · 2019-10-10T20:41:34Z

Hi, George, It is a good question. As Moothi told me, the atmospheric model is still not ready yet to use the calculated salinity dependent water temperature for the water/ice mixed grids, therefore, the constant value of 271.2 for tgice and tsfomn will be used still. The salinity dependent water temperature affects the NSST tref, and therefore tsea (or tmpsfc), in the NSST adjust part, which is ont in sfcsub.F. Therefore, this time, we don't need to change tgice and tsfomn in sfcsub.F. Or we may need think about more to see if the salinity dependent water temperature can be applied to more areas, like the water/ice mixed grids (currently, the water portion is assigned as 271.2, which is too low by more than 1 degree for lakes with lower salinity). XU

…

On Thu, Oct 10, 2019 at 3:58 PM GeorgeGayno-NOAA ***@***.***> wrote: The sfcsub.F module contains a variable - 'tgice' - that is 271.2. That variable is used in several computations. Also, variable 'tsfomn' is 271.2 and is used to range check the lower bound of open water. Do these need to be changed for salinity? — You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub <#20>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ANEMLJVCJ76CKPVFANVRVBLQN6COXANCNFSM4I33ID6A> .

GeorgeGayno-NOAA · 2019-10-11T14:57:05Z

ush/global_cycle.sh

@@ -58,6 +58,8 @@
 #     FNMXIC        Input maximum sea ice climatology GRIB file.
 #                   Defaults to ${FIXam}/global_maxice.2x2.grb
 #     FNTSFC        Input SST climatology GRIB file.
+#                   Defaults to ${FIXam}/global_salclm.t1534.3072.1536.nc
+#     FNSALC        Input Salinity climatology netcdf file.


There is a typo here and in the next line.

Fixed and pushed.
Thanks!

GeorgeGayno-NOAA

Ready to merge.

XuLi-NOAA and others added 3 commits September 26, 2019 15:00

UFS_UTILS Issue #18:

406d9dc

Modify to resolve a few sea ice related issues in NSST analysis

Issue #18, fix a few problems in cycle.f90

f5fe742

Issue #18, commit the fix file, FNSALC, definition in global_cycle.sh

e5b4858

GeorgeGayno-NOAA requested review from SMoorthi-emc and yangfanglin September 30, 2019 14:25

GeorgeGayno-NOAA reviewed Sep 30, 2019

View reviewed changes

yangfanglin reviewed Sep 30, 2019

View reviewed changes

XuLi-NOAA added 2 commits September 30, 2019 13:54

UFS_UTILS Issue #18: use mpi_abort to replace stop

5431926

Merge remote-tracking branch 'upstream/develop' into develop

5453e1e

add a comment of FNSLAC in global_cycle.sh and include mpi.h in read_…

c9b0ad6

…write_data.f90

Fix a problem found by George: stc becomes too cold due to the change

b85999e

GeorgeGayno-NOAA removed the request for review from SMoorthi-emc October 3, 2019 17:00

XuLi-NOAA added 2 commits October 9, 2019 17:16

Commit change in sfcsub.F

e1bb0bf

Commit the change in sfcsub.F for global_chgres.fd

9bf6d8e

XuLi-NOAA added 2 commits October 10, 2019 15:23

Commit the changes in sfcsub.F, including some fortran code format am…

f84965f

…d the definition of lbms (logical => logical*1)

Merge remote-tracking branch 'upstream/develop' into develop

206ca7a

GeorgeGayno-NOAA reviewed Oct 11, 2019

View reviewed changes

fix a typo comment line in global_cycle.sh

d4aa42e

GeorgeGayno-NOAA self-requested a review October 11, 2019 16:30

GeorgeGayno-NOAA approved these changes Oct 11, 2019

View reviewed changes

GeorgeGayno-NOAA merged commit 1b76994 into ufs-community:develop Oct 11, 2019

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

A few sea ice related issues in the NSST analysis: Issue #18 #20

A few sea ice related issues in the NSST analysis: Issue #18 #20

XuLi-NOAA commented Sep 30, 2019

GeorgeGayno-NOAA Sep 30, 2019

XuLi-NOAA Sep 30, 2019

yangfanglin Sep 30, 2019 •

edited

Loading

XuLi-NOAA Sep 30, 2019

yangfanglin left a comment

XuLi-NOAA commented Sep 30, 2019 via email

yangfanglin commented Sep 30, 2019

GeorgeGayno-NOAA commented Sep 30, 2019

XuLi-NOAA commented Sep 30, 2019 via email

GeorgeGayno-NOAA commented Sep 30, 2019

XuLi-NOAA commented Sep 30, 2019 via email

GeorgeGayno-NOAA commented Oct 3, 2019

XuLi-NOAA commented Oct 3, 2019 via email

SMoorthi-emc commented Oct 3, 2019 via email

GeorgeGayno-NOAA commented Oct 3, 2019

XuLi-NOAA commented Oct 4, 2019 via email

GeorgeGayno-NOAA commented Oct 9, 2019

GeorgeGayno-NOAA commented Oct 9, 2019

GeorgeGayno-NOAA commented Oct 9, 2019

XuLi-NOAA commented Oct 9, 2019 via email

GeorgeGayno-NOAA commented Oct 10, 2019

XuLi-NOAA commented Oct 10, 2019 via email

GeorgeGayno-NOAA Oct 11, 2019

XuLi-NOAA Oct 11, 2019

GeorgeGayno-NOAA left a comment

A few sea ice related issues in the NSST analysis: Issue #18 #20

A few sea ice related issues in the NSST analysis: Issue #18 #20

Conversation

XuLi-NOAA commented Sep 30, 2019

GeorgeGayno-NOAA Sep 30, 2019

Choose a reason for hiding this comment

XuLi-NOAA Sep 30, 2019

Choose a reason for hiding this comment

yangfanglin Sep 30, 2019 • edited Loading

Choose a reason for hiding this comment

XuLi-NOAA Sep 30, 2019

Choose a reason for hiding this comment

yangfanglin left a comment

Choose a reason for hiding this comment

XuLi-NOAA commented Sep 30, 2019 via email

yangfanglin commented Sep 30, 2019

GeorgeGayno-NOAA commented Sep 30, 2019

XuLi-NOAA commented Sep 30, 2019 via email

GeorgeGayno-NOAA commented Sep 30, 2019

XuLi-NOAA commented Sep 30, 2019 via email

GeorgeGayno-NOAA commented Oct 3, 2019

XuLi-NOAA commented Oct 3, 2019 via email

SMoorthi-emc commented Oct 3, 2019 via email

GeorgeGayno-NOAA commented Oct 3, 2019

XuLi-NOAA commented Oct 4, 2019 via email

GeorgeGayno-NOAA commented Oct 9, 2019

GeorgeGayno-NOAA commented Oct 9, 2019

GeorgeGayno-NOAA commented Oct 9, 2019

XuLi-NOAA commented Oct 9, 2019 via email

GeorgeGayno-NOAA commented Oct 10, 2019

XuLi-NOAA commented Oct 10, 2019 via email

GeorgeGayno-NOAA Oct 11, 2019

Choose a reason for hiding this comment

XuLi-NOAA Oct 11, 2019

Choose a reason for hiding this comment

GeorgeGayno-NOAA left a comment

Choose a reason for hiding this comment

yangfanglin Sep 30, 2019 •

edited

Loading