[MITgcm-support] Questions on configuring AIM surface boundary conditions

[Jean-Michel Campin]

Hi Fisher,

I will try to clarify few things:
> It seems that parameters like aim_useFMsurfBC and aim_useMMsurfFc control
> the use of boundary conditions, but I could not find detailed explanations 
> for these flags. Could you clarify their meanings? 

The description (in AIM_PARAMS.h + repeated in STDOUT) is rather short:
C     aim_useFMsurfBC :: select surface B.C. from Franco Molteni
C     aim_useMMsurfFc :: select Monthly Mean surface forcing (e.g., NCEP)
These 2 logical parameters allows to select between a simplified monthly
mean forcing, that repeats itself over a year ( aim_useMMsurfFc=T, aim_useFMsurfBC=F)
and the original Speedy code (from Franco Molteni) surface forcing (aim_useFMsurfBC=T, 
aim_useMMsurfFc=F, i.e., the default).
However, if you are using pkg/land (useLand=T, as in verification/aim.5l_cs/input/),
and assuming you don't turn off parts of the land code (keeping all 4 logical
parameters: land_calc_{grT,grW,snow,alb} set to True, i.e., the default), 
many surface fields will be replaced by the land prognostics variables so that the 
effective number of surface-BC fields to provide is reduced. 

And similarly, if you are using pkg/thsice (useThSIce=T, as in 
verification/aim.5l_cs/input.thSI/), few more surface fields will be replaced 
by pkg/thsice prognostics variables.

> Also, should the boundary-condition data be provided as monthly values, 
> or is another temporal format expected?
With aim_useMMsurfFc=T, monthly-mean values is the only option, with 1 file
per month (with 3 characters month suffix in the file name, see: 
verification/aim.5l_LatLon/input settings).

With aim_useFMsurfBC=T, all time records are in a single file. it still assumes
repeating forcing cycle, but since the repeating period is adjustable, if you provide
enough records to exceed the length of the simulation there will not be any repetition.
This is controlled by:
C     aim_surfForc_TimePeriod :: Length of forcing time period (e.g. 1 month)
C     aim_surfForc_NppCycle   :: Number of time period per Cycle (e.g. 12)
and the length of the repeating cycle is the product of the 2.

To summarize, with aim_useFMsurfBC=T, the surface BC files you may provide are:
 aim_LandFile : land fraction, loaded once (1 record) and kept fixed
 aim_albFile  : surface albedo over land, loaded once (1 record) and kept fixed
 aim_sstFile  : SST, multiple records

Not used if useThSIce=T:
 aim_oiceFile : seaice fraction, multiple records

Not used if useLand=T:
 aim_lstFile  : land surface Temp, multiple records
 aim_swcFile  : soil water content, multiple records
 aim_vegFile  : vegetation fraction, loaded once (1 record) and kept fixed
 aim_snowFile: snow depth over land, multiple records
Note: some might still be used as Land default initial condition (at startTime=0, 
     myIter=nIter0=0), something I have not checked.

Regarding this:
> If I want to prescribe only SST from data while letting the model internally 
> compute albedo, land-surface temperature, Sea Ice fraction, and other variables, 
> what is the correct way to set this up?
It seems to me that you will need both useLand=T and useThSIce=T (in data.pkg),
is this the case ?
And you still need to provide land-albedo (that will be modified where there
is snow).

Finally, regarding the attached "data.aimphys", what is the type of time-variation
you are targeting for the SST ?
If it's constant, I will just change aim_useFMsurfBC=F to True and set:
 aim_surfForc_NppCycle=1
(not the most efficient way but should do the job).

Cheers,
Jean-Michel

On Thu, Sep 18, 2025 at 01:14:17PM +0000, ht wrote:
> Dear Community,
> 
> I am currently trying to set up atmospheric and land-surface simulations using MITgcm, taking MITgcm/verification/aim.5l_cs as a reference but with my own topography and SST distribution.
> 
> I have several questions regarding the configuration files, especially data.aim:
> 
> -
> 
> It seems that parameters like aim_useFMsurfBC and aim_useMMsurfFc control the use of boundary conditions, but I could not find detailed explanations for these flags. Could you clarify their meanings? Also, should the boundary-condition data be provided as monthly values, or is another temporal format expected?
> 
> -
> 
> If I want to prescribe only SST from data while letting the model internally compute albedo, land-surface temperature, Sea Ice fraction, and other variables, what is the correct way to set this up?
> 
> -
> 
> I have attached my current data.aimphys file. Could you please check whether it is correctly configured for this purpose or if modifications are needed?
> 
> Thank you very much for your time and support.
> 
> Best regards,
> Fisher


> _______________________________________________
> MITgcm-support mailing list
> MITgcm-support at mitgcm.org
> http://mailman.mitgcm.org/mailman/listinfo/mitgcm-support