[MITgcm-support] MITgcm-support Digest, Vol 194, Issue 30

[Martin Losch]

Hi David,

this is difficult to say, especially without the error message that you get at the end. It could be anything.

I have a few comments on your namelists: 
- in “data” you use 2nd order central difference for tracers without diffusion. I am surprised that this works for a longer integration. Probably because you set useCDscheme=.TRUE., which I would try to avoid (uncontrollable extra viscosity). I would try something stable, like 33 (which requires to useStaggerTimeStep=.TRUE.), or add diffusion. (I don’t see a gmredi parameter list)
- your free surface parameters appear inconsistent to me: you turn on the non-linear free surface, but turn of the “real” freshwater flux. Only with useRealFreshWaterFlux=.TRUE., you get the conservation properties of the non-linear free surface. I would also use r-star (select_rStar=2) to reduce the problem of surface cell running dry (probably your problem anyway).
- you don’t need to use globalFiles=.TRUE., unless it speeds up your output (usually doesn’t happen for small grids)
- you have free slip at the bottom, but no bottom drag

- in “data.seaice” you could try to set SEAICEpressReplFac = 0. to turn off replacement pressure. In long integrations, this commonly accepted regularization sometimes leads to massive seaice thickness increase in isolated grid points which will then make the ocean model blow up because the surface cell runs dry.
- in your configuration, I would not use EVP, but the default solver with SEAICEnonLinIterMax=10, and SEAICE_EPS = 1e-5, SEAICE_deltaMin = 2e-9,
- definitely turn off the diffusion of sea ice, that’s something really unphysical (I think). So “my” data.seaice for long intergrations (order 100years) would look like this (for recent code):
 &SEAICE_PARM01
 SEAICEnonLinIterMax=10,
 SEAICE_EPS = 1E-5, 
# SEAICEaEVPcoeff = 0.5,
# SEAICEnEVPstarSteps = 500,
# SEAICEuseEVPstar = .TRUE.,
# SEAICEuseEVPrev = .TRUE.,
# SEAICEaEVPalphaMin = 20
 SEAICEpressReplFac = 0.,
 SEAICE_deltaMin = 2e-9,
 SEAICE_salt0          = 4.,
 SEAICE_no_slip          = .TRUE.,
# default: SEAICEadvScheme       = 77,
# makes the ice thicker, if you need that
# SEAICE_multDim        =  7,
 &

 &SEAICE_PARM03
 &

Martin

> On 9. Sep 2019, at 17:59, David Vishny <davidvish at gmail.com> wrote:
> 
> Hi Martin,
> 
> After updating the model to the latest version, the sea ice dynamics do not cause the model to output NaNs right away, however ice thickness and velocity blow up within 100 years of model time, indicating there is some kind of numerical instability. If I run the model for 100 years without using sea ice dynamics, there is no issue. I’ve tried various things, including shortening the time steps in the “data” namelist file, playing with cg2d parameters, and changing the EVP scheme to aEVP, but I haven’t found a way to make the dynamics stable. Do you have any suggestions? I’ve attached some important files from my model setup.
> 
> 
> Thanks,
> 
> David Vishny
> 
> 
> <EXF_forcing_files><namelist><code>
>> Message: 1
>> Date: Fri, 30 Aug 2019 09:43:07 -0500
>> From: David Vishny <davidvish at gmail.com>
>> To: mitgcm-support at mitgcm.org
>> Subject: Re: [MITgcm-support] MITgcm-support Digest, Vol 194, Issue 28
>> Message-ID: <146B8FB9-E9CB-4531-8497-455772F5ADBC at gmail.com>
>> Content-Type: text/plain;	charset=utf-8
>> 
>> Hi Martin,
>> 
>> I?ve updated to the latest model version and everything seems to be working fine. Thank you for your help!
>> 
>> David Vishny