Blow up for Upwelling case

[goutam]

Hi,
I am new to ROMS. I am just trying to run the UPWELLING test case in the Peru region( -85E to -75E longitude and -15N to -7N latitude). When I define ANA_INITIAL the model runs fine. But when I undefine ANA_INITIAL, and use my own input initial file. The model blows up.
The initial conditions used are u,v,ubar,vbar=0
Can someone tell me what's the problem.

Thanks,
Goutam

[kate]

Are you providing the other fields it needs in the initial file? T,S, zeta? The model should tell you if they are missing. Is it telling you why it's blowing up?

[goutam]

Hi Kate,
Thank you for the response. I will just tell everything in detail. I use the Roms Tools package by Pierrick Penven. I used that and generated the following:-
roms_grd.nc
roms_ini.nc('u,v,ubar,vbar,zeta,temp,salt' values included)
roms_clm.nc
roms_frc.nc

So when I wanted to run the model for the first time, this is exactly what error it gave:
GET_STATE - unable to find model variable: ^H^E^@^@<0a>^]^H0^@^@^@ in input NetCDF file: roms_ini.nc

So when I looked into the roms_ini.nc file, I saw the variable 'scrum_time', which I didn't find in the 'varinfo.dat' file.

So I changed the 'ocean_time' variable in the 'varinfo.dat' file to 'scrum_time'. The model started running. But after running for a few time steps it blows up.
This is what it shows after running for some 48 time steps:
ROMS/TOMS - Blows up ................ exit_flag: 1

Also one more thing, I have figured that the number of time steps after which the model blows up varies upon the TCLINE value which I give.

Can you just tell me how to stop the blow up and also correctly estimate the TCLINE and DT(time step) value.

Thanks
Goutam

[kate]

It sounds like the tools need an update to change them over to ocean_time.

Is the number of steps it runs dependent on dt? That's one way to know that the timestep is the problem. As for tcline, we usually let it default to the minimum depth. How are you setting the other stretching parameters?

[goutam]

The stretching parameters I used are:-
theta_s=7.0
theta_b=0.0

Goutam

[goutam]

I also wanted to know how to set the HC parameter? Are the HC, TCLINE and HMIN parameters related?

Goutam

[kate]

From set_scoord.F:

hc(ng)=MIN(hmin(ng),Tcline(ng))

Thus, setting Tcline to 200 when hmin is 50 means that Tcline isn't doing anything at all. That's the case I was referring to.

[goutam]

Hi,
I am actually trying to model the Gulf of Kutch, India , where water depth is much than 40 metres. So I gave
hmin=5m
tcline=5m
hc=5m
It wasn't working.

But when I give
hmin=5m
tcline=3m
hc=3m
It is working.

Also I gave
theta_s=5.0
theta_b=0.9

Thanks.
Goutam

[bettykjohn]

Dear Gautham

I want to know whether your problem solved or not
because i got the same error while executing (ocean_time error)
so if your case is executing well then it can apply to my case

Thank you
Betty

[pramaneeraja]

I get the following error when I try and compile the upwelling test case.
analytical.f90(319): error #5082: Syntax error, found ':' when expecting one of: % . = =>
ana_grid.h: no values provided for Xsize, Esize, depth, f0, beta.
----------------^
analytical.f90(1053): error #5082: Syntax error, found ':' when expecting one of: % . = =>
ana_vmix.h: no values provided for Akv.
----------------^
analytical.f90(1062): error #5082: Syntax error, found ':' when expecting one of: % . = =>
ana_vmix.h: no values provided for Akt.
----------------^
compilation aborted for analytical.f90 (code 1)
make: *** [Build/analytical.o] Error 1
root@neeraja-Compaq-Presario-CQ40-Notebook-PC:/home/neeraja/Desktop/ROMS/roms-3.1# ana_grid.h: no values provided for Xsize, Esize, depth, f0, beta.
ana_grid.h:: command not found
root@neeraja-Compaq-Presario-CQ40-Notebook-PC:/home/neeraja/Desktop/ROMS/roms-3.1# ----------------^
----------------^: command not found

Any idea why such a problem persists..

[kate]

What exactly are you doing? Build script or Makefile? Can you look at your analytical.f90 to see what's there? I suspect it somehow doesn't know you want the UPWELLING case, so make sure the UPWELLING code gets into analytical.f90. Have you discovered the ROMS wiki? https://www.myroms.org/wiki/index.php/D ... ion_Portal

[pramaneeraja]

Hello,

Thanks for your help.Upwelling test case is working just fine now. I am using make file. But when I try other test cases , I get error in compilation itself.Apart from changes in make file for ROMS_APPLICATION, do i need to make any other changes.Yes, I have explored the wikiROMS but except for upwelling case it gives no details for any other test case.

Any help regarding the test cases especially pertaining to river plume would really be great.While compilation I get following error.
def_avg.f90(1244): error #6606: The block construct names must match, and they do not. [QUERY]
END IF QUERY
-------------^
def_avg.f90(838): error #6321: An unterminated block exists.
QUERY : IF (.not.ldef) THEN
^
compilation aborted for def_avg.f90 (code 1)
make: *** [Build/def_avg.o] Error 1

In the make file I have directed the ROMS_APPLICATION to riverplume1.I have also made changes in Compiler directory in order to use ifort.

[kate]

You need to look at def_avg.f90 to see what it's complaining about. Do you agree that there's a syntax error in the file? Where did it come from in the source file? In this case, the source file is def_avg.F. Can you see how it should be changed there? If there's a ROMS bug, report it via trac.