[MITgcm-devel] r* changes 3 weeks ago?

[Alistair Adcroft]

It would seem the bug fix has changed the answers (based on ascii output 
but I've not yet made plots) for the internal wave example in the paper. 
W*=0 at top then and now. I suspect the solutions will be close enough 
given that W*=0.

A.

Thomas Haine wrote:
> Hi Alistair, J-M, 
> 
> I guess that the buggy z* code was less sensitive to bathymetry. With a
> new bathy and the bug-fix in place I have a stable code with the same
> timestep as before. I suspect the main bathy fix was to exclude all
> ocean cells shallower than 20m. I also removed all surface ocean cells
> with 3 sides of land.
> 
> Tom.
> 
> On Mon, 2005-08-08 at 19:58 -0400, Jean-Michel Campin wrote:
> 
>>Hi Alistair,
>>
>>I've just re-run the global_ocean.cs without realFreshWater,
>>and get almost zero (~truncation error) for W @ the surface
>>(in fact, it's w*), both with the bug in calc_r_star and without:
>>
>>with the bug:
>> k | --  Average --  | --  Std.Dev  -- | --     min  --  | --     max   -- |
>>  1 -1.5241674088E-24  1.2856393143E-21 -7.1272208023E-21  3.4752623419E-20
>>  2  2.5044944587E-10  3.0654052890E-06 -1.7805864466E-05  3.2450027450E-05
>>
>>and without:
>> k | --  Average --  | --  Std.Dev  -- | --     min  --  | --     max   -- |
>>  1 -1.8486143344E-23  1.0848723067E-21 -2.0861288253E-20  1.2560737602E-20
>>  2  2.6798489476E-10  3.0654806148E-06 -1.7807405140E-05  3.2452863292E-05
>>
>>My impression is that without OBCS, the z* seems to work.
>>
>>Jean-Michel
>>
>>On Thu, Jul 14, 2005 at 02:34:31PM -0400, Alistair Adcroft wrote:
>>
>>>JMC changed calc_r_star.F 3 weeks ago (June 19) with a "bug fix" and 
>>>reading the code it certainly looks like a treu bug fix.
>>>
>>>http://mitgcm.org/cgi-bin/viewcvs.cgi/MITgcm/model/src/calc_r_star.F.diff?r1=1.4&r2=1.5
>>>
>>>However, the Irminger Sea calculation (Tom Haine) is no longer stable 
>>>using z* where it was before the bug fix. What's up?
>>>
>>>How can we have produced the figures for the z* paper if this bug was 
>>>turning off z* over the slope region?
>>>
>>>Puzzled.
>>>
>>>A. + T.
>>>-- 
>>>Dr Alistair Adcroft            http://www.mit.edu/~adcroft
>>>MIT Climate Modeling Initiative        tel: (617) 253-5938
>>>EAPS 54-1624,  77 Massachusetts Ave,  Cambridge,  MA,  USA
>>>_______________________________________________
>>>MITgcm-devel mailing list
>>>MITgcm-devel at mitgcm.org
>>>http://mitgcm.org/mailman/listinfo/mitgcm-devel
>>
>>_______________________________________________
>>MITgcm-devel mailing list
>>MITgcm-devel at mitgcm.org
>>http://mitgcm.org/mailman/listinfo/mitgcm-devel


-- 
Dr Alistair Adcroft            http://www.mit.edu/~adcroft
MIT Climate Modeling Initiative        tel: (617) 253-5938
EAPS 54-1624,  77 Massachusetts Ave,  Cambridge,  MA,  USA