[MITgcm-support] (no subject)

[Martin Losch]

Kaveh, 

we cannot change the error message, as it is issued by the run-time system of the compiler. "Non-existing record number” usually means, that the model tries to read past the end of the file, i.e. that there are not enough records in your file. Expanding on Dimitris’s answer, your data.exf could look like this:
 swdown_nlon    = 2,
 swdown_nlat    = 2,
and then you need to provide the correct coordinate for
 swdown_lon0    = ,
 swdown_lon_inc = ,
 swdown_lat0    = .,
 swdown_lat_inc = 2*somevalue,
to let the model know about your forcing grid.

Martin

> On 18 Jun 2016, at 16:11, Dimitris Menemenlis <dmenemenlis at gmail.com> wrote:
> 
> you need to specify size and content of input files in data.exf
> e.g., MITgcm/verification/global_with_exf/input/data.exf
> 
>> On Jun 18, 2016, at 4:36 AM, kaveh Purkiani <kavehpurkiani at googlemail.com> wrote:
>> 
>> Dear MITgcm developer,
>> 
>> Hi,
>> 
>> I figured out that EXF_interpolation routine does not work when some of provided external input files are in different size (Nx, Ny).
>> 
>> I had aqh.bin, swdown.bin and lwdown.bin files in different size (2*2*time) as the rest of my external forcing files (30*30*time). I have had below massage error which I think also misleaded me. From the error massage I thought of  not providing the  proper start time with data.cal and startdate of external forcing, or not enough time record of external forcing fields, which obviously was not the case. 
>> 
>> Offline interpolating of aqh, swdown and lwdown to same size as the rest of external forcing helped to solve the problem.
>> 
>> Below you see the error massage.
>> 
>> At line 3612 of file exf_interp_read.f (unit = 9, file = './data-meteo/Swrad.bin')
>> Fortran runtime error: Non-existing record number
>> 
>> If this is correct, would you please add a new error massage to help users to find the error easier, or is it possible to fix the interpolation routine?
>> 
>> Thank you very much in advance.
>> 
>> Kaveh. 
> 
> 
> _______________________________________________
> MITgcm-support mailing list
> MITgcm-support at mitgcm.org
> http://mitgcm.org/mailman/listinfo/mitgcm-support