[MITgcm-support] External forcing for deltaT < 1 sec

[Dustin Carroll]

Hi Jody,

I meant that it wasn’t incrementing sequentially through my forcing files — but it was when deltaT > 1. I’ll recheck my settings and look at get_periodic_interval.F.
That’s really good to know about the interpolation between the last and first record…especially at such small time steps where things can blow up quickly. I’m sure other folks will find that useful as well.

Best,
Dustin

On Apr 11, 2014, at 8:24 PM, Jody Klymak <jklymak at uvic.ca> wrote:

> 
> On Apr 11, 2014, at  13:09 PM, Dustin Carroll <dcarroll at uoregon.edu> wrote:
> 
>> I tried using the basic periodicExternalForcing but that didn’t properly loop through my forcing files properly when deltaT < 1. Can anyone confirm that periodicExternalForcing works with deltaT < 1 or recommend a method for time dependent bc's with these small time steps?
> 
> Hi Dustin,
> 
> Not sure if I understand "didn't properly loop".  If deltaT>1. does it "properly loop"?  The source code you want to look at is get_periodic_interval.F, maybe that has a clue.
> 
> One thing to be careful about, which has got me in the past, is that your first record in your OBET_hr.bin etc is actually centered at 0.5*externForcingPeriod (=1800s), and that at t=0, the boundary condition is actually a linear interpolation between the records at -1800 s and at +1800 s i.e. between the *last* record in your file and the first.  If you are doing something with a linear initial ramp, and the last value you have in your hourly forcing is large, your t=0 startup will be a lot bumpier than you were hoping.
> 
> The justification for that convention is that a lot of folks force with monthly climatologies etc, which are most appropriately centered on the period.  
> 
> Cheers,  Jody
> 
> 
> 
> 
> _______________________________________________
> MITgcm-support mailing list
> MITgcm-support at mitgcm.org
> http://mitgcm.org/mailman/listinfo/mitgcm-support
>