Open Boundary Problem

Discussion of how to use ROMS on different regional and basin scale applications.

Moderators: arango, robertson

Post Reply
Message
Author
bibi951
Posts: 45
Joined: Tue Mar 17, 2009 4:06 pm
Location: cpeo,ocean university of china

Open Boundary Problem

#1 Post by bibi951 » Sun Jul 19, 2009 2:07 am

Hi,all
I do a simulation using Hycom's Global simulation product for my open boundary.
I have checked the Open Boundary section field after interpolating them to Roms's s- coordinate.It seems fine although not exactly the same.
But I have some doubt about the field consistent.I just use the same points for all the variables(ssh,t,s,u,v),and after interpolation to s-coordinate, I just integrate u,v to get ubar vbar.Is there something will cause trouble?
Besides, I didn't modify the field on land,so ssh may to very large.It doesn't matter,right?
I prepare the data in the order of ROMS index ,which means in horrizotal from 1 to xi(eta).and in vertical from 1 to L-layers.
I use only the open boundary forcing.
However,the result goes quite wrong.West Boundary Current is reversed.Here is the picture after 80 days or so.
http://photo.xiaonei.com/getphoto.do?id ... 1894331121
http://photo.xiaonei.com/getphoto.do?id ... 1894333244
I use OBC :fschapman m2flather m3radiation Tradiation.
Any idea?
Thanks in advance.

User avatar
kate
Posts: 3695
Joined: Wed Jul 02, 2003 5:29 pm
Location: IMS/UAF, USA

Re: Open Boundary Problem

#2 Post by kate » Mon Jul 20, 2009 5:02 pm

Hi Bibi,

I can't see your images, not having a login wherever that is.

Is your grid aligned with north-south-east-west? How about Hycom's grid? You may have to rotate the boundary u,v so that they are lined up with the ROMS grid.

bibi951
Posts: 45
Joined: Tue Mar 17, 2009 4:06 pm
Location: cpeo,ocean university of china

Re: Open Boundary Problem

#3 Post by bibi951 » Tue Jul 21, 2009 8:37 am

Thank you,Kate.
You are always so kind.
Sorry,I didn't realize the login problem.
Here is the new address
http://photo.blog.sina.com.cn/photo/46f ... 32f1f7e4be
http://photo.blog.sina.com.cn/photo/46f ... f81612#pic
and I will send you an email for pictures.
My grid is North-South-West-East(the angle is zero),hycom's product is along the longitutde and latitude.That should not be the problem.
Besides,I want to make sure this is right.Forgive me.The caculation is too expensive, so I don't want to do some tests blindly.
I prepare the data in the order of ROMS index ,which means in horrizotal from 1 to xi(eta).and in vertical from 1 to L-layers.
How about the velocity and ssh consistency? and the same line of data from hycom to provide Open Boundary data is OK?
Will the large value of ssh on land will cause trouble?
Thanks for any suggestion.

User avatar
kate
Posts: 3695
Joined: Wed Jul 02, 2003 5:29 pm
Location: IMS/UAF, USA

Re: Open Boundary Problem

#4 Post by kate » Tue Jul 21, 2009 4:19 pm

Your currents are more likely being generated by the initial conditions than the boundary conditions - you mean the ones up against the land, right? I don't think hycom is the wrong way to go for initial conditions, but it and roms need to agree on what density is at what depth. If you need to extrapolate into the land (where hycom has land), do that on level z surfaces before doing the vertical interpolations.

Large zeta in the land won't affect anything.

bibi951
Posts: 45
Joined: Tue Mar 17, 2009 4:06 pm
Location: cpeo,ocean university of china

Re: Open Boundary Problem

#5 Post by bibi951 » Wed Jul 22, 2009 2:41 am

Sorry, I didn't mention my initial condition is taken from SODA-1.4.2 and is climatology yearly mean.Both my soda and hycom data are on z-level surfaces.
Maybe the initial condition is the problem?
I will show pictures of the process from intial condition to the day 87;and the time varing East open boundary here
(title divided by 4 is the day number):Click the photo for the next one..
ssh:
http://photo.blog.sina.com.cn/photo/46f ... 49351ec970
velocity:
http://photo.blog.sina.com.cn/photo/46f ... 4931abbab5
boudary:
http://photo.blog.sina.com.cn/photo/46f ... eef918#pic
Here is the simulation gif pictures
http://rapidshare.com/files/258634548/zeta.gif
http://rapidshare.com/files/258630796/velocity.gif
I'm sorry it need 45 second waiting.Need your patience..
The vertical averege velocity(80days simulation) on the path of Kurishio is reversed.and I can't find the ubar pointing out the east boundary at any time.
I checked the Open boundary field after interpolation at some time,they seem OK comparing with the hycom's section(at least the feature is still there).

User avatar
kate
Posts: 3695
Joined: Wed Jul 02, 2003 5:29 pm
Location: IMS/UAF, USA

Re: Open Boundary Problem

#6 Post by kate » Wed Jul 22, 2009 3:26 pm

So your initial currents look good, but then they go chaotic and then settle going the wrong way? Internal problems like that are still not due to boundary conditions, but something else. There's something up with your model physics. What terms in the equations do you have on? What is your equation of state?

bibi951
Posts: 45
Joined: Tue Mar 17, 2009 4:06 pm
Location: cpeo,ocean university of china

Re: Open Boundary Problem

#7 Post by bibi951 » Thu Jul 23, 2009 12:56 am

Thank you for teaching me how to diagnose where the problem may be,Kate.
I really appreciate it.
I should have shown the CPP flags before.
This is what I use
NANHAI NANHAI SHUZHISHIYAN
ANA_BTFLUX Analytical kinematic bottom temperature flux.
ANA_SMFLUX Analytical kinematic surface momentum flux.
ANA_STFLUX Analytical kinematic surface temperature flux.
ASSUMED_SHAPE Using assumed-shape arrays.
CURVGRID Orthogonal curvilinear grid.
DJ_GRADPS Parabolic Splines density Jacobian (Shchepetkin, 2002).
DOUBLE_PRECISION Double precision arithmetic.
EAST_FSCHAPMAN Eastern edge, free-surface, Chapman condition.
EAST_M2FLATHER Eastern edge, 2D momentum, Flather condition.
EAST_M3RADIATION Eastern edge, 3D momentum, radiation condition.
EAST_TRADIATION Eastern edge, tracers, radiation condition.
KANTHA_CLAYSON Kantha and Clayson stability function formulation.
MASKING Land/Sea masking.
MIX_ISO_TS Mixing of tracers along isopycnal surfaces.
MPI MPI distributed-memory configuration.
MY25_MIXING Mellor/Yamada Level-2.5 mixing closure.
NONLINEAR Nonlinear Model.
NONLIN_EOS Nonlinear Equation of State for seawater.
NORTH_FSCHAPMAN Northern edge, free-surface, Chapman condition.
NORTH_M2FLATHER Northern edge, 2D momentum, Flather condition.
NORTH_M3RADIATION Northern edge, 3D momentum, radiation condition.
NORTH_TRADIATION Northern edge, tracers, radiation condition.
N2S2_HORAVG Horizontal smoothing of buoyancy and shear.
POWER_LAW Power-law shape time-averaging barotropic filter.
PROFILE Time profiling activated .
K_GSCHEME Third-order upstream advection of TKE fields.
!RST_SINGLE Double precision fields in restart NetCDF file.
SALINITY Using salinity.
SOLVE3D Solving 3D Primitive Equations.
SOUTH_FSCHAPMAN Southern edge, free-surface, Chapman condition.
SOUTH_M2FLATHER Southern edge, 2D momentum, Flather condition.
SOUTH_M3RADIATION Southern edge, 3D momentum, radiation condition.
SOUTH_TRADIATION Southern edge, tracers, radiation condition.
SPLINES Conservative parabolic spline reconstruction.
TS_U3HADVECTION Third-order upstream horizontal advection of tracers.
TS_C4VADVECTION Fourth-order centered vertical advection of tracers.
TS_DIF2 Harmonic mixing of tracers.
UV_ADV Advection of momentum.
UV_COR Coriolis term.
UV_C2ADVECTION Second-order centered differences advection of momentum.
UV_QDRAG Quadratic bottom stress.
VAR_RHO_2D Variable density barotropic mode.
WESTERN_WALL Wall boundary at Western edge.
I set btflux,smflux,stflux to zero.
Anything wrong here?
Thanks

User avatar
kate
Posts: 3695
Joined: Wed Jul 02, 2003 5:29 pm
Location: IMS/UAF, USA

Re: Open Boundary Problem

#8 Post by kate » Thu Jul 23, 2009 1:30 pm

I don't see anything off hand. Have you looked at the coriolis parameter f in the grid file? Is it of the correct sign?

bibi951
Posts: 45
Joined: Tue Mar 17, 2009 4:06 pm
Location: cpeo,ocean university of china

Re: Open Boundary Problem

#9 Post by bibi951 » Fri Jul 24, 2009 12:54 am

Yes,parameter f is correct and it varies along latitude.
Maybe I should try another initial condition since it may conflict someway with the open boundary information,consindering they come from different model and differ in time.
I will report the result after this is done.

User avatar
kate
Posts: 3695
Joined: Wed Jul 02, 2003 5:29 pm
Location: IMS/UAF, USA

Re: Open Boundary Problem

#10 Post by kate » Fri Jul 24, 2009 4:35 pm

How about turning on DIAGNOSTICS? You could then see if there's something terribly wrong with coriolis balancing pressure gradients.

User avatar
susonic
Posts: 159
Joined: Tue Aug 21, 2007 5:44 pm
Location: Jeju National University
Contact:

Re: Open Boundary Problem

#11 Post by susonic » Sun Jul 26, 2009 2:35 am

How often is your BC value given in your domain?
Have you ever changed the BC code from RADIATION to CLAMPED for M3 and T ?
From my experience, CLAMPED BC code might be more promising.

-Peter

bibi951
Posts: 45
Joined: Tue Mar 17, 2009 4:06 pm
Location: cpeo,ocean university of china

Re: Open Boundary Problem

#12 Post by bibi951 » Sun Jul 26, 2009 8:33 am

Thanks for your suggustion.
My OB data is given daily to the interior domain.
I have not tried Clamped OBC yet,maybe that would help.
I am also thinking about nudging ,would 20 days of nudging time scale be OK for outflow , 0.5 day for inflow ?

bibi951
Posts: 45
Joined: Tue Mar 17, 2009 4:06 pm
Location: cpeo,ocean university of china

Re: Open Boundary Problem

#13 Post by bibi951 » Wed Jul 29, 2009 7:59 am

Hello,everyone.
I tried another initial condition derived from hycom, and used nudging mentioned above (20 days of nudging time scale for outflow , 0.5 day for inflow).It partly works(I don't know which contributes more to this, intialization or nudging),however not good enough.
ssh and barotropic velocity
http://photo.blog.sina.com.cn/u/1190472905
The title plus 90 is the day number of my simulation.(ie. 45+90=135 which means this is the 135 day's simulation.)Click the photo to see more clearly.
The kuroshio doesn't show up clearly, the surface elevation doesn't look good with too much strong holes(low ssh).So the velocity field is quite a mess.I also think Kuroshio water intrusion through luzon strait is too strong,and the Kuroshio(if is can be called Kuroshio) is too northward.
I wonder is there any model configuration I can do to change this simulation better?
Here is the standard output:

0.0000E+00 tnu2(01) Horizontal, harmonic mixing coefficient (m2/s)
for tracer 01: temp
0.0000E+00 tnu2(02) Horizontal, harmonic mixing coefficient (m2/s)
for tracer 02: salt
1.0000E-06 Akt_bak(01) Background vertical mixing coefficient (m2/s)
for tracer 01: temp
1.0000E-06 Akt_bak(02) Background vertical mixing coefficient (m2/s)
for tracer 02: salt
1.0000E-05 Akv_bak Background vertical mixing coefficient (m2/s)
for momentum.
5.0000E-06 Akk_bak Background vertical mixing coefficient (m2/s)
for turbulent energy.
5.0000E-06 Akp_bak Background vertical mixing coefficient (m2/s)
for turbulent generic statistical field.
3.0000E-04 rdrg Linear bottom drag coefficient (m/s).
2.5000E-03 rdrg2 Quadratic bottom drag coefficient.
2.0000E-02 Zob Bottom roughness (m).
1 Vtransform S-coordinate transformation equation.
1 Vstretching S-coordinate stretching function.
7.0000E+00 theta_s S-coordinate surface control parameter.
1.0000E-01 theta_b S-coordinate bottom control parameter.
50.000 Tcline S-coordinate surface/bottom layer width (m) used
in vertical coordinate stretching.
1025.000 rho0 Mean density (kg/m3) for Boussinesq approximation.
0.000 dstart Time-stamp assigned to model initialization (days).
20050101.00 time_ref Reference time for units attribute (yyyymmdd.dd)
1.0000E+01 Tnudg(01) Nudging/relaxation time scale (days)
for tracer 01: temp
1.0000E+01 Tnudg(02) Nudging/relaxation time scale (days)
for tracer 02: salt
0.0000E+00 Znudg Nudging/relaxation time scale (days)
for free-surface.
0.0000E+00 M2nudg Nudging/relaxation time scale (days)
for 2D momentum.
1.0000E+01 M3nudg Nudging/relaxation time scale (days)
for 3D momentum.
2.0000E+01 obcfac Factor between passive and active
open boundary conditions.
14.000 T0 Background potential temperature (C) constant.
35.000 S0 Background salinity (PSU) constant.
1.000 gamma2 Slipperiness variable: free-slip (1.0) or
no-slip (-1.0).
Thanks for any suggustion.

prakash22
Posts: 24
Joined: Thu Jan 08, 2015 11:16 pm
Location: iit delhi

Re: Open Boundary Problem

#14 Post by prakash22 » Sun Jan 03, 2016 10:10 am

hi bibi

Could you tell me which tool are you using to create boundary and initial file using hycom global data. if it is possible then could you help me.

That will very helpful.


thanks in Advance
-Prakash

User avatar
kate
Posts: 3695
Joined: Wed Jul 02, 2003 5:29 pm
Location: IMS/UAF, USA

Re: Open Boundary Problem

#15 Post by kate » Sun Jan 03, 2016 5:50 pm

There are Python tools to do this in pyroms. I'm sure others have Matlab tools. Did you have a preference?

prakash22
Posts: 24
Joined: Thu Jan 08, 2015 11:16 pm
Location: iit delhi

Re: Open Boundary Problem

#16 Post by prakash22 » Mon Jan 04, 2016 4:46 am

No, I don't have any preference. So Could you tell me about Matlab tool or can you send me the link of tool.

User avatar
kate
Posts: 3695
Joined: Wed Jul 02, 2003 5:29 pm
Location: IMS/UAF, USA

Re: Open Boundary Problem

#17 Post by kate » Mon Jan 04, 2016 6:05 pm

For the matlab, start here.

prakash22
Posts: 24
Joined: Thu Jan 08, 2015 11:16 pm
Location: iit delhi

Re: Open Boundary Problem

#18 Post by prakash22 » Tue Jan 05, 2016 11:31 am

Hi kate

I install pyroms,pyroms_tool successfully but I am not able to create boundary file.

I have few doubt like
when I ran make_bdry_file.py then I am getting error like _iso , _average file is not found. I copied these file to current directory now I am getting error like


Traceback (most recent call last):
File "make_bdry_file.py", line 41, in <module>
src_grd = pyroms_toolbox.Grid_HYCOM.get_nc_Grid_HYCOM(src_grd_file)
File "/usr/local/python_tool//lib/python2.7/site-packages/pyroms_toolbox/Grid_HYCOM/get_nc_Grid_HYCOM.py", line 16, in get_nc_Grid_HYCOM
File "netCDF4/_netCDF4.pyx", line 1795, in netCDF4._netCDF4.Dataset.__init__ (netCDF4/_netCDF4.c:11938)
RuntimeError: No such file or directory



another doubt is that, Is these tools need Hycom global grid and ROMS grid information

if is it so then from where I can get hycom global grid data ( if it is possible send me the link of these data)

please help me

thanks in Advance
-ravi

User avatar
kate
Posts: 3695
Joined: Wed Jul 02, 2003 5:29 pm
Location: IMS/UAF, USA

Re: Open Boundary Problem

#19 Post by kate » Tue Jan 05, 2016 6:24 pm

Oops, I added the python script for downloading the HYCOM grid to the Arctic_HYCOM example. You need to have both grids, you need to have the PYROMS_GRIDID_FILE environment variable set to point to the gridid.txt file and you need to edit it for your grid. Then you should be able to run make_remap_weights_file if scrip.so was installed properly. Once that is done, you can move on to making the boundary files.

prakash22
Posts: 24
Joined: Thu Jan 08, 2015 11:16 pm
Location: iit delhi

Re: Open Boundary Problem

#20 Post by prakash22 » Wed Jan 06, 2016 4:43 am

But in place of hycom global grid which data I will provide

User avatar
kate
Posts: 3695
Joined: Wed Jul 02, 2003 5:29 pm
Location: IMS/UAF, USA

Re: Open Boundary Problem

#21 Post by kate » Wed Jan 06, 2016 5:55 am

Ah, are you downloading a subset of the HYCOM global grid? For the Arctic example I downloaded a subset covering the north pole. I also have an example covering a non-polar subset which I can add as another example. OK, done.

Post Reply