[MITgcm-support] Namelist Error with RBCS and multiple tracers

[Dustin Carroll]

Hi Hauke and Martin,

I think rbcsIniter is outdated and has been replaced with rbcsForcingOffset. Perhaps someone who has recently worked on the RBCS pkg can chime in.

Cheers,
Dustin


On Jun 16, 2014, at 1:13 AM, Martin Losch <Martin.Losch at awi.de> wrote:

> Hauke,
> 
> namelist errors are annoying to track down, because there are no good error diagnostics (some compilers are better than others, but MITgcm’s method of copying the files into a—modified—temporary file and then reading the temporary file, doesn’t make things easier to debug). Most often they are simple spelling errors (at least in my case).
> 
> I agree, It looks like the problem is in data.rbcs. From your file “12cores.err”, line
>> mitgcmuv           0000000000594D38  rbcs_readparms_          1907  rbcs_readparms.f
> tells you which namelist read in rbcs_readparms.f (with lower case “f”) is the problematic namelist, and the first line
>> forrtl: severe (19): invalid reference to variable in NAMELIST input, unit 11, file /sb/project/dfr-394-aa/hblank/MIT18/run11conttrac12/fortzM9hx2, line 12, position 17
> Gives you a clue where in this namlist you need to look. “/sb/project/dfr-394-aa/hblank/MIT18/run11conttrac12/fortzM9hx2” is the temporary file, I was talking about, and you don’t have access to that unfortunately. To get a similar file remove all lines starting with # in your data.rbcs 
> 
> If that does not work, try commeting out parts of the namelists and try again, until you have identified the line, where the problem is.
> (To be honest, I cannot see any problems in your data.rbcs, so narrow down the line where it happens would be very useful)
> 
> Martin
> 
> On Jun 16, 2014, at 3:08 AM, Hauke Blanken, Mr <hauke.blanken at mail.mcgill.ca> wrote:
> 
>> Hi Everyone,
>> 
>> I’m currently modelling a problem with 11 sources of passive tracer in my domain and am running into a namelist error (exit code 19), which I think is quite likely an easy fix, however I haven’t been able to solve it, being quite new to MITgcm. 
>> 
>> I’ve compiled the code for 11 tracers and 13 mask files and am using the RBCS package to restore the tracer field to it’s original value at each time step. I tested this setup using only one tracer and it worked fine. To add the remaining tracers I re-compiled, changing PTRACERS_num to 11 in PTRACERS_SIZE.h and maskLEN to 13 in RBCS.h, and modified the data.ptracers and data.rbcs files to add links to the appropriate tracer, mask and relaxation files. I’ve attached my data.ptracers, data.rbcs, error log and a sample STDOUT file (STDERR is empty). The error log and STDOUT seem to indicate that the error is in data.rbcs, though STDOUT terminates on data.rbcs on some time steps and data.ptracers on others. If anyone has insight on what might be happening here it would be hugely appreciated. And if there is other info that would be helpful in diagnosing the problem I’m more than happy to provide it.
>> 
>> Thanks in advance,
>> Hauke
>> 
>> Hauke Blanken
>> M.Eng. (Thesis) Candidate
>> Department of Civil Engineering and Applied Mechanics
>> McGill University
>> Phone: 778-871-3431
>> <12cores.err><data.ptracers><data.rbcs><STDOUT.0001>_______________________________________________
>> MITgcm-support mailing list
>> MITgcm-support at mitgcm.org
>> http://mitgcm.org/mailman/listinfo/mitgcm-support
> 
> 
> _______________________________________________
> MITgcm-support mailing list
> MITgcm-support at mitgcm.org
> http://mitgcm.org/mailman/listinfo/mitgcm-support
>