I’m having difficulty getting point sources to work for my model. I have followed the tutorial instructions to the letter, but I get the following runtime issue, whether using constant flow specified via croco.in, or time varying flow via the runoff netcdf file.
forrtl: severe (408): fort: (2): Subscript #1 of the array ISRC has value 1 which is greater than the upper bound of 0
I can’t be the first person to encounter such a problem, but I see no reference to the problem on the CROCO forum or in the mailing list communications.
I’ve also been unsuccessful trying to get PSOURCE/ANA_PSOURCE as well as PSOURCE/PSOURCE_NCFILE to work for the BENGUELA test case. The code runs well but I don’t get any river flow when I plot the results. But the RIVER test case works fine. I’ve checked to make sure the location of the point sources are on land points, and also used the make_runoff.m code to create a netcdf file. But to no avail.
Has anyone been successful with the PSOURCE setup for the BENGUELA test case? Please share your setup with us. Thanks.
Regarding the ISRC problem, my first guess would be to check the number of sources in your roms.in file and n_qbar variable in the runoff netcdf file.
Best,
Martin
Just to close this off, I found that this issue was linked to Msrc in param.h file which is referred to in the Ifremer specific section of the tutorials document for v1.2, whereas I was only looking at section 15 of the document and matlab scripts.
I sent out the same query and the corresponding solution via croco-users@inria.fr last week, apologies for the delay updating the forum.
The problem was due to a bug in the v1.2
It has been corrected into the v1.2.1 newly released.It is avalable on croco-ocean.org
Here the change log for the v1.2.1
ONLINE bulk: wind speed was not properly computed, with a consequent understimation of fluxes when using bulk ONLINE. This is now corrected
Runoff : Msrc parameter in param.h was wrongly set to 0, it is now corrected and set to 100
Plurimonth_scripts: DT was wrongly intialized with nesting. This is now corrected.
In 2D cases, the size of wrthis array was wrong. This is now corrected. A typo in def_diags has been corrected
CVTK: some continuous integration tests have been updated
The problem was due to a bug in the v1.2
It has been corrected into the v1.2.1 newly released.It is available on croco-ocean.org
Here the change log for the v1.2.1
ONLINE bulk: wind speed was not properly computed, with a consequent underestimation of fluxes when using bulk ONLINE. This is now corrected
Runoff : Msrc parameter in param.h was wrongly set to 0, it is now corrected and set to 100
Plurimonth_scripts: DT was wrongly intialized with nesting. This is now corrected.
In 2D cases, the size of wrthis array was wrong. This is now corrected. A typo in def_diags has been corrected
CVTK: some continuous integration tests have been updated
The problem was due to a bug in the v1.2
It has been corrected into the v1.2.1 newly released.It is available on croco-ocean.org
Here the change log for the v1.2.1
ONLINE bulk: wind speed was not properly computed, with a consequent underestimation of fluxes when using bulk ONLINE. This is now corrected
Runoff : Msrc parameter in param.h was wrongly set to 0, it is now corrected and set to 100
Plurimonth_scripts: DT was wrongly intialized with nesting. This is now corrected.
In 2D cases, the size of wrthis array was wrong. This is now corrected. A typo in def_diags has been corrected
CVTK: some continuous integration tests have been updated