[MITgcm-support] ABNORMAL END: S/R INI_THETA (when changing geometry)

[Baker, Jon]

Hello,


I am attempting to alter the geometry (number of grid points) of various tutorials and have encountered a new problem. After changing the number of grid points in the data file and SIZE.h file, the following error is shown when I run the global_oce_biogeochemistry and global_oce_latlon tutorials: ABNORMAL END: S/R INI_THETA .


The full error message is as follows:

 INI_THETA: found      5875 wet grid-pts with theta=0 identically.
 If this is intentional, you need to set checkIniTemp=.false. in "data", namelist PARM05
ABNORMAL END: S/R INI_THETA


I have looked up this error and found that it is because theta(i,j,k,bi,bj)=0. at at least one wet grid point (as described previously: http://mitgcm.org/pipermail/mitgcm-support/2008-June/005539.html).


However, I am having difficulty understanding why this is the case, since I believe tRef is set correctly (presumably this shouldn't change when the horizontal geometry is altered, as long as the vertical grid points remain unchanged). In the biogeochemistry tutorial which has 15 grid points in the vertical, the following setup is used in the data file:


&PARM01
 tRef=15*20.,
 sRef=15*35.,

Thus, the reference temperature is set to 20C at all 15 vertical grid points. This runs normally for the original geometrical setup (global) but has the above error when the geometry (number of horizontal grid points) is altered.

In the INI_THETA subroutine:

 theta<http://mitgcm.org/sealion/code_reference/vdb/names/4N1.htm>(I<http://mitgcm.org/sealion/code_reference/vdb/names/61.htm>,J<http://mitgcm.org/sealion/code_reference/vdb/names/8V.htm>,K<http://mitgcm.org/sealion/code_reference/vdb/names/9C.htm>,bi<http://mitgcm.org/sealion/code_reference/vdb/names/56.htm>,bj<http://mitgcm.org/sealion/code_reference/vdb/names/57.htm>) = tRef<http://mitgcm.org/sealion/code_reference/vdb/names/13O.htm>(K<http://mitgcm.org/sealion/code_reference/vdb/names/9C.htm>)

Therefore, the temperature should not be 0 at any of the wet grid points!

Interestingly, setting tRef=15*0. removes the error, replacing it with a different error:
'ABNORMAL END: S/R INI_SALT'

The natl_box tutorial does not have tRef or sRef defined in the data file and runs normally when I change the number of horizontal grid points.

The above link suggests that 'when you interpolated the temperature field, that you have the model read, you missed at least one point'. However, given the tRef that has been set (tRef=15*20.), interpolation should not be a problem, unless there is another file where the temperature is defined which I am unaware of.

Any help would be greatly appreciated!

Cheers,
Jon


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mitgcm.org/pipermail/mitgcm-support/attachments/20170127/6cfee0d9/attachment.htm>