[RegCNET] Potential Bug in RegCM3 - ICBC.f

[bixq]

Thanks, Jonathan:

Yes, it's really a potential bug, and also FORTRAN compiler dependent.

I have updated the ICBC.f in both regcm.tar.gz and 
regcm_beta_20090916.tar.gz package.

Best regards,
Xunqiang Bi

On Thu, 21 Apr 2011, Jonathan Winter wrote:

> Hi Bi,
>
> I hope all is well in Trieste.  I understand that RegCM3 is no longer supported, but I recently downloaded a clean copy of the latest version of RegCM3 to compare to some of my preliminary RegCM4 simulations.  While looking over my ICBCs (NNRP2, OI_WK), I noticed that I had unrealistically high temperatures (ts and t) over land (350+ K).  This is what I believe is happening.  Line 6428 in subroutine cdc6hour references "ivar":
>
> status=nf_get_vara_int2(inet,ivar,start,count,work)
>
> In older versions of ICBC.f, "ivar" was not used:
>
> status=nf_get_vara_int2(inet,5,start,count,work)
>
> So the problem is that while "ivar" is defined the first time cdc5hour is called, in subsequent calls it is not defined and becomes 0, instead of what it should be (5).  I believe I was able to fix this problem by adding "ivar" to the common blocks.  So on line 6323:
>
> cjmw042111      COMMON /NNOPEN/ inet7(7),start(10),count(10)
>      COMMON /NNOPEN/ inet7(7),start(10),count(10), ivar
>
> Please let me know what you find and if this patch makes sense to you.  If you have any questions, let me know.
>
> Best,
>
> Jonathan
>

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
   Dr. Xunqiang Bi         email:bixq at ictp.it
   Earth System Physics Group
   The Abdus Salam ICTP
   Strada Costiera, 11
   P.O. BOX 586, 34100 Trieste, ITALY
   Tel: +39-040-2240302  Fax: +39-040-2240449
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~