[MITgcm-support] paleo runs and model drift

[Martin Losch]

Hi Andrew,
there's an (undocumented) flag ALLOW_BALANCE_FLUXES. If that defined in CPP_OPTIONS.h you can use balanceEmPmR = .true., in data (PARM01) to balance the fresh water flux at each time step (to give you zero salinity and SSH drift), and balanceQnet = .true., to balance the heat flux. Neither of these flags are very physical, but they do the job. I introduced them out of despair, but I found out later that especially for fresh water, this technique is not too uncommon with OGCMs on long time scales (see e.g. Griffies et al, 2009, Ocean Modelling, Appendix B3).

Martin

On Apr 4, 2010, at 4:36 PM, Holly Dail wrote:

> Andrew -
> 
> The following is an example of work related to the paleo problem:
> Wunsch and P. Heimbach, How long to oceanic tracer and proxy equilibrium?, Quat. Sci. Rev. 27 (2008), pp. 637–651
> 
> I'm working on the LGM using the MIT GCM for my thesis, but haven't yet written anything up.
> 
> As for drift in long model runs, I've tried several approaches to balancing heat and FW inputs to reduce long term drift in mean T/S.  I'm happy to discuss those with you.  Which specific drift problem are you concerned with?
> 
> Holly
> 
> 
> On Apr 4, 2010, at Apr 4 , 8:43 AM, Andrew Keats wrote:
> 
>> Hi,
>> 
>> Can anyone point me to journal articles or conference papers where MITgcm has been used for paleoclimate/paleooceanographic applications?  I am having a hard time finding anything using Google Scholar and other search engines.
>> 
>> Also, has anyone done a quantitative assessment of the model drift?  For multiple-thousand year runs, can drift be compensated for without using relaxation?
>> 
>> Thanks,
>> 
>> Andrew
>> 
>> 
>> --
>> Andrew Keats
>> NSERC Postdoctoral Fellow
>> Department of Physics and Physical Oceanography
>> Memorial University of Newfoundland
>> 
>> 
>> 
>> 
>> 
>> 
>> _______________________________________________
>> 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