Hi Jean Michel,
I looked (quickly) at the revision of seaice_check.F / init_fixed / readparms.
Several things I did not do quite right, or forgot, evidently. Thanks for putting in the time.
I looked (also quickly) through the lab_sea details below. I did not
catch any obvious problem in the params combinations. One thing from the
seaice-verif-notes.txt that I just recalled -- last week I offered to cook up a June
starting point, so that lab_sea finally tests melting conditions too. I have not done that yet...
As for the variable salinity with SItracers, I confirm your setup should be fine.
(To the extent that my code is all correct that is. Another thing I offered to do is to
double check conservation, coupling etc. for that part, and I have not done it yet...)
It may be useful that I elaborate on this setup :
x SItrFromOceanFrac(2) = 0.3, is indeed the counterpart to Dimitris' HSALT code (SEAICE_VARIABLE_SALINITY)
x SItrFromOcean0(2) = 4., instead would be a variable salinity but with constant salinity for new ice.
x if I recall correctly, Dimitris' HSALT code has flooding creating 0 salinity ice. What you did, Jean Michel, is
consistent with that approach. Indeed you left SItrFromFlood0(2) and SItrFromFloodFrac(2) at their 0 default.
You are also correct with regard to the pickup. Eventually I assume we would have a .NOT.pickupStrictlyMatch way of
using those old HSALT pickups. This requires mapping the extensive var. (HSALT) to the intensive var. (SItracer). Not done yet ...
Cheers,
Gael
On Mar 11, 2012, at 3:22 PM, Jean-Michel Campin wrote:
> Hi,
>
> Some details below of the new fwd lab_sea tests.
> If you have suggestions or if you see a bad combination of
> params, let me know.
> Also, I am not sure if this (lab_sea/input.salt_plume/data.seaice)
> is the right way of doing SEAICE_VARIABLE_SALINITY with SItracers ?
> Gael, can you confirm ?
> Right now, it seems that siHSALT from the pickup-file is not
> used to initialise the salinity SItracer.
>
> Cheers,
> Jean-Michel
>
> --------------------------------------------------------------------
> Details:
>
> lab_sea/code/SEAICE_OPTIONS.h
> I took the one that Martin (version.2) sent,
> - remove SEAICE_VARIABLE_FREEZING_POINT
> (has been removed completely from the seaice code)
> - define SEAICE_ALLOW_FREEDRIFT (so that it can be tested in input.fd)
> - update the description of SEAICE_MULTICATEGORY
> - some minor editing
>
> input/data.seaice:
> I took the one that Martin (version.2) sent, change the setting order
> (dynamics, seaice state & transport, thermodyn, IO params)
> set SEAICEheatConsFix=.TRUE., (as we discuss)
> and added
> SEAICE_multDim = 7, (but this is the current default)
> and also kept the 3rd namelist with the same 2 SItracers as before.
>
> input.fd/data.seaice: similar to input/data.seaice, except:
> - SEAICEuseFREEDRIFT = .TRUE.,
> SEAICEheatConsFix commented out (= default = F)
> - remove SItracers ;
> - SEAICEadvSchHeff = 33,
> # add small diffKh to test diffusion with multi-dim advect.
> SEAICEdiffKhHeff = 20.,
> - SEAICE_frazilFrac = 0.,
> so that we have 1 lab_sea test that does not freeze sea-water if below freezing.
>
> input.hb87/data.seaice: similar to input/data.seaice, except:
> - SEAICE_deltaTevp = 30.,
> - useHB87stressCoupling = .TRUE.,
> - no SItracers (as it used to be), no SEAICEheatConsFix
> - all 4 initial condition files (same as before)
> - use default Adv.Scheme with DIFF1 = 4.E-3,
> (so that it's tested in one of the lab-sea test)
> and also added:
> - SEAICE_multDim = 1, (will run single category)
> # strong ice-ocean turb-flux coeff:
> SEAICE_mcPheePiston= 0.001,
> SEAICE_frazilFrac = 0.5,
>
> input.salt_plume/data.seaice:
> has not been changed so much (kept old solve4temp)
> except: Adv-scheme=7 (since the ocean is also using 7),
> const. freezing point, + salinity as SItracer :
> SEAICE_salinityTracer = .TRUE.,
> # SEAICE_saltFrac = 0.3,
> #- constant seawater freezing point:
> SEAICE_tempFrz0 = -1.96,
> SEAICE_dTempFrz_dS = 0.,
> with 2 SItracers:
> SItrNumInUse = 2,
> SItrName(1) = 'ridge',
> SItrMate(1) = 'AREA',
> SItrName(2) = 'salinity',
> SItrFromOceanFrac(2) = 0.3,
>
> --------------------------------------------------------------------
> And here is the fwd-lab_sea relevant part of seaice-verif-notes.txt
> that Patrick sent:
>
> * lab_sea
> ---------
> - give up legacy
> - try different setups with different options
> forw1: input: exact baseline parameters + SEAICEheatConsFix=.TRUE. + LSR
> forw2: input.fd: SEAICEheatConsFix=.FALSE., SEAICEuseFREEDRIFT=.TRUE., --- start June 1st
> forw3: input.salt_plume: const. freezing point, old solve4temp + salinity as SItracer
> forw4: input.hb87: forw1 + EVP + useHB87stressCoupling
>
>
> _______________________________________________
> MITgcm-devel mailing list
> MITgcm-devel at mitgcm.org
> http://mitgcm.org/mailman/listinfo/mitgcm-devel