[MITgcm-support] Format of initialization data and relaxation data

Martin Losch Martin.Losch at awi.de
Wed Feb 1 06:26:50 EST 2012


Hi Liu Jinliang,

see answers below. (And I really recommend reading the documenation.)

Martin
On Feb 1, 2012, at 8:47 AM, liujinliang.627 wrote:

>  Hello,
> 
>     I'm trying to have a 10 years run for climatology with the example "\verification\tutorial_global_oce_latlon".
> 
> 1.  But I don't know how to set initial tracer data "hydrogThetaFile" and "hydrogSaltFile". 
>    
>      If I want to run for 10 years(Global Ocean Simulation, 1 degree * 1 degree, Nx=360, Ny=160,Nr=15 ), which format should I use for "lev_t.bin" and "lev_s.bin"?  Can it be in an (x,y) or (x,y,r) fashion or  others ?  which software should be use for input, matlab or fortran? 
hydrogThetaFile/hydrogSaltFile require 3D fields (x,y,r). 
The format the all input files (including hydrogThetaFile/hydrogSaltFile) is ieee-be. You can use any tool to generate such files. In matlab it is very simple:
t = zeros(nx,ny,nr); % here you need to use realistic values of course
fid = fopen('theta.init','w','b');
fwrite(fid,t,'real*4'); % readBinaryPrec=32, for readBinaryPrec=64 use 'real*8'
fclose(fid).
In Fortran you have to open your file with ACCESS='DIRECT' (and make sure that you write the correct precision and endian type)
> 
> 2.  Can the relaxation term be turned off ?  If could, which part should I change?
yes. In your example like this (no relaxation for either temperature nor salinity):
 tauThetaClimRelax=  0.,
 tauSaltClimRelax = 0.,
> 
> 3.  If I want to using wind force for 10 years continuously (externForcingPeriod = 1month, externForcingCycle = 10years ), what format should "zonalWindFile" and  "meridWindFile" be ?  Should it be in  (Nx,Ny) from first month to the end( i.e. (Nx*12*10, Ny) ), or in (Nx, Ny, t) (t is the number of months) fashion?
If you want to precribe monthly data for 10 years (and all years are different), then your data needs to have  the format (Nx,Ny,10*12) (and ieee-be). But setting these flags requires that all forcing data is in this format, i.e., also for these
 thetaClimFile=  'lev_sst.bin',
 saltClimFile=   'lev_sss.bin',
 surfQFile=      'ncep_qnet.bin',
 EmPmRFile=      'ncep_emp.bin',

To have more freedom (and more problems) use the exf-package. 
> 
> Thank you !
> 
> Liu Jinliang
>  
> 
> 
> _______________________________________________
> MITgcm-support mailing list
> MITgcm-support at mitgcm.org
> http://mitgcm.org/mailman/listinfo/mitgcm-support




More information about the MITgcm-support mailing list