[MITgcm-support] Bug in rbcs_fields_load.F

[Christopher L. Wolfe]

Hi modelers,

There is a bug in rbcs_fields_load.F that causes parts of the ptracer  
relaxation fields to be loaded into the tile overlap region instead  
of the tile interior. This can cause obscure errors in the relaxation  
pattern.

Here is a diff of the incorrect, old file (<) compared to the new,  
corrected file (>):

162c162
<      &       'RS', Nr, rbcptracer0(1,1,1,1,1,iTracer), intime0,  
myThid )
---
 >      &       'RS', Nr, rbcptracer0(1-Olx,1-Oly,1,1,1,iTracer),  
intime0, myThid )
164c164
<      &       'RS', Nr, rbcptracer1(1,1,1,1,1,iTracer), intime1,  
myThid )
---
 >      &       'RS', Nr, rbcptracer1(1-Olx,1-Oly,1,1,1,iTracer),  
intime1, myThid )
182,183c182,183
<            _EXCH_XYZ_R4(rbcptracer0(1,1,1,1,1,iTracer),myThid)
<            _EXCH_XYZ_R4(rbcptracer1(1,1,1,1,1,iTracer),myThid)
---
 >            _EXCH_XYZ_R4(rbcptracer0(1-Olx,1-Oly, 
1,1,1,iTracer),myThid)
 >            _EXCH_XYZ_R4(rbcptracer1(1-Olx,1-Oly, 
1,1,1,iTracer),myThid)

This diff is against checkpoint58q_post, but I don't believe that  
this file has changed in the recent updates.

If anyone knows who maintains the RBCS package, please give them a  
heads up for me.

Christopher


-----------------------------------------------------------
Dr. Christopher L. Wolfe                   858-534-4560
Physical Oceanography Research Division    OAR 357
Scripps Institution of Oceanography, UCSD  clwolfe at ucsd.edu
-----------------------------------------------------------