problem of hc

Report or discuss software problems and other woes

Moderators: arango, robertson

Post Reply
Message
Author
limaolin
Posts: 45
Joined: Wed Nov 07, 2012 5:48 pm
Location: Dalian University

problem of hc

#1 Unread post by limaolin »

the problem is

NETCDF_CHECK_VAR - inconsistent value of variable: hc 5.000000E+00 2.496991E+01
in file: Projects_My/USEastCoast/coawst_ini.nc

I have set the TCLINE == 50.0d0, the minimum depth of grid is 5, the hc of ini.nc is 5 too. But where is 2.496991E+01 come from?

I have found the same problem in wiki, but I did not know how to solve the problem.

Thanks for reply.
Attachments
us_east_coast.in
(115.08 KiB) Downloaded 290 times
us_east_coast.h
(3.36 KiB) Downloaded 256 times

rsoutelino
Posts: 23
Joined: Tue Oct 07, 2008 11:27 am
Location: MetService - New Zealand
Contact:

Re: problem of hc

#2 Unread post by rsoutelino »

Depending on the combination of your vertical transformation and stretching functions, there restrictions for these parameters: https://www.myroms.org/wiki/index.php/V ... coordinate

limaolin
Posts: 45
Joined: Wed Nov 07, 2012 5:48 pm
Location: Dalian University

Re: problem of hc

#3 Unread post by limaolin »

Thanks for your reply.

I have set the Vtransform = 2 and Tcline = 5, the Tcline is the same as hmin. The error is corrected. Is it right?

But after that, there were other errors.

barack99
Posts: 17
Joined: Thu Aug 31, 2017 4:33 pm

Re: problem of hc

#4 Unread post by barack99 »

Hi,

I got a same error. Could someone help?

NETCDF_CHECK_VAR - inconsistent value of variable: hc

Actually, I select Vtransform = 2 and Vstretching = 4, so according to :

https://www.myroms.org/wiki/Vertical_S-coordinate --->> '' It works for both h c ≤ h m i n or h c > h m i n ''

In my domain, the grid was built based on: hmin =2m, and Tcl =16m and Vtransform = 2 and Vstretching = 4 as mentioned.

However, when I set Tcl =16m and Vtransform = 2 and Vstretching = 4 in ocean*.in, I got an error:

NETCDF_CHECK_VAR - inconsistent value of variable: hc 2.000000E+00 1.600000E+01
in file: initial_conditions.nc


then I tried to adjust Tcl = 2.0d0 in ocean*.in, another error message is:

NETCDF_CHECK_VAR - inconsistent value of variable: Tcline 1.600000E+01 2.000000E+00
in file: initial_conditions.nc
Barack

User avatar
kate
Posts: 4088
Joined: Wed Jul 02, 2003 5:29 pm
Location: CFOS/UAF, USA

Re: problem of hc

#5 Unread post by kate »

One value is coming from your ocean.in file, but the other value is coming from the initial_conditions.nc file. Where did that file come from? The software creating initial files has to know about the vertical stretching choices too, and set them consistently with how ROMS will be run.

barack99
Posts: 17
Joined: Thu Aug 31, 2017 4:33 pm

Re: problem of hc

#6 Unread post by barack99 »

Thanks Kate a lot!

However, when that is fixed, my model blow up after 6 minutes running, even if I reduced time step =5s...

Could you please help?

208 55866 00:17:20 7.582798E-03 2.280633E+02 2.280708E+02 9.305414E+10
(858,146,3) 1.712835E-02 2.473793E-01 2.289394E+00 5.344137E+00
209 55866 00:17:25 7.552532E-03 2.280621E+02 2.280696E+02 9.305238E+10
(858,146,3) 2.011910E-02 2.488418E-01 3.247929E+00 5.589116E+00
210 55866 00:17:30 7.521886E-03 2.280608E+02 2.280684E+02 9.305062E+10
(860,146,5) 1.186692E-01 2.654912E-01 2.150358E+01 5.828378E+00
211 55866 00:17:35 7.491077E-03 2.280596E+02 2.280671E+02 9.304886E+10
(858,146,4) 9.609532E-03 2.511305E-01 1.514073E+01 6.048339E+00
212 55866 00:17:40 7.460246E-03 2.280583E+02 2.280657E+02 9.304711E+10
(860,146,5) 1.903178E-01 2.576538E-01 2.121702E+00 1.738201E+01
213 55866 00:17:45 7.428715E-03 2.280569E+02 2.280643E+02 9.304536E+10
(859,147,5) 5.263803E-02 3.129963E-01 6.226965E+00 1.509292E+01
214 55866 00:17:50 7.399157E-03 2.280555E+02 2.280629E+02 9.304361E+10
(844,143,2) 1.884536E-01 1.234839E-01 2.146917E+01 2.895645E+01
DIAG speed trouble 28.956446301020534
DIAG speed ijk 39.951705537795448 859 147 5
1

Blowing-up: Saving latest model state into RESTART file

Node # 3 CPU: 403.595
Node # 1 CPU: 403.562
Node # 2 CPU: 403.549
WRT_RST - wrote re-start fields (Index=1,1) in record = 0000001

Elapsed CPU time (seconds):

Node # 0 CPU: 403.455
Total: 1614.161

Nonlinear ocean model elapsed time profile, Grid: 01

Allocation and array initialization .............. 0.920 ( 0.0570 %)
Ocean state initialization ....................... 0.122 ( 0.0075 %)
Reading of input data ............................ 3.259 ( 0.2019 %)
Processing of input data ......................... 1.291 ( 0.0800 %)
Computation of vertical boundary conditions ...... 0.910 ( 0.0564 %)
Computation of global information integrals ...... 10.429 ( 0.6461 %)
Writing of output data ........................... 10.656 ( 0.6602 %)
Model 2D kernel .................................. 1338.650 (82.9317 %)
Tidal forcing .................................... 1.310 ( 0.0811 %)
2D/3D coupling, vertical metrics ................. 11.855 ( 0.7344 %)
Omega vertical velocity .......................... 7.732 ( 0.4790 %)
Equation of state for seawater ................... 5.293 ( 0.3279 %)
3D equations right-side terms .................... 58.561 ( 3.6279 %)
3D equations predictor step ...................... 61.382 ( 3.8027 %)
Pressure gradient ................................ 7.213 ( 0.4468 %)
Corrector time-step for 3D momentum .............. 25.641 ( 1.5885 %)
Corrector time-step for tracers .................. 50.565 ( 3.1326 %)
Total: 1595.790 98.8619

Nonlinear model message Passage profile, Grid: 01

Message Passage: 2D halo exchanges ............... 13.627 ( 0.8442 %)
Message Passage: 3D halo exchanges ............... 1.787 ( 0.1107 %)
Message Passage: 4D halo exchanges ............... 1.282 ( 0.0794 %)
Message Passage: data broadcast .................. 9.435 ( 0.5845 %)
Message Passage: data reduction .................. 0.202 ( 0.0125 %)
Message Passage: data gathering .................. 2.390 ( 0.1480 %)
Message Passage: data scattering.................. 2.024 ( 0.1254 %)
Message Passage: boundary data gathering ......... 0.698 ( 0.0432 %)
Message Passage: point data gathering ............ 0.031 ( 0.0019 %)
Total: 31.476 1.9500

All percentages are with respect to total time = 1614.161


ROMS/TOMS - Output NetCDF summary for Grid 01:
number of time records written in HISTORY file = 00000004
number of time records written in RESTART file = 00000001
number of time records written in STATION file = 00000214

Analytical header files used:

/data/COAWST/Projects/DOMAIN/ana_btflux.h
/data/COAWST/Projects/DOMAIN/ana_smflux.h

ROMS/TOMS: DONE... Thursday - September 21, 2017 - 09:04:37 PM
Barack

User avatar
kate
Posts: 4088
Joined: Wed Jul 02, 2003 5:29 pm
Location: CFOS/UAF, USA

Re: problem of hc

#7 Unread post by kate »

It's time to look at the restart file it so nicely saved for you. The problem i,j,k is at (859 147 5) as reported. Where is that relative to the boundary conditions? Coastlines? Surface/bottom?

barack99
Posts: 17
Joined: Thu Aug 31, 2017 4:33 pm

Re: problem of hc

#8 Unread post by barack99 »

Thanks Kate!

I have looked at that point, but now I got another error message:

(089,052,5) 7.652039E-02 3.389343E-02 6.659950E-01 6.492329E+00
DIAG rho trouble 209.89901139715789 1
Blowing-up: Saving latest model state into RESTART file

What does 'DIAG rho trouble ' mean? and what could I do?

Regards
Barack
Barack

User avatar
kate
Posts: 4088
Joined: Wed Jul 02, 2003 5:29 pm
Location: CFOS/UAF, USA

Re: problem of hc

#9 Unread post by kate »

ROMS is checking for extreme velocities and extreme densities. You went from one to the other. The reporting doesn't always give the i,j,k values, not sure why. Still, look at the restart file for clues.

caozd999
Posts: 8
Joined: Tue Aug 19, 2014 4:35 pm
Location: UNIVERSITY OF FLORIDA

Re: problem of hc

#10 Unread post by caozd999 »

Hope it's not too late to reply.

I met with this problem today too. And I figured out how it happened to me.

1.I generated my initial file by the script roms_master_climatology_coawst_mw.m, which gives us 2 options on Vtransform:
1) Vtransform=1; and then hc = min(max(hmin,0),Tcline);
2) Vtransform=2; and then hc = Tcline;

2. I think some people still use Vtransform=1 option (although Vtransform=2 and Vstretching=4 are highly recommended), so they won't have any further hc issue.

3. If you set Vtransform=2, and you need pay attention to a function updatinit_coawst_mw called in the above script, in which there's no Vtransform option yet. And in this function, hc=min([hmin,Tcline]), which means no matter what value you choose for Vtransform, hc=min([hmin,Tcline]). This will give you the inconsistency issue if your hmin<Tcline.

I hope it helps. And please anyone is welcome to correct me if I'm wrong.
Thanks!
-zd

Post Reply