[MITgcm-support] Exf package and interpolation in time

[Martin Losch]

Hi Georges,

maybe the names of the namelist variables are not very intuitive, but
for your example you used the wrong values. As far as I understand, you have
3-hourly wind, somehow interpolated onto the model grid. For a 3day run
you need 24/3 * 3(days) = 24 records, plus one or two records outside
your model run period (e.g. if your first forcing co-incides with model
time = 0, then your last one also co-insides with modeltime 24h*3 = 72*h
and you 25 records, correct?)
The exf-pkg needs to now
1. when is the first record (ustressstartdate1 and ustressstartdate2,
the first is the calendar day, e.g. Jan01,2010 would be 20100101, the
second is time, e.g., midnight would be 000000, 2.15pm would be 141500,
etc.)
2. what is the interval between two records (ustressperiod, in your case
it is 3h=10800sec, so ustressperiod=10800., and NOT 3days = 259200sec
3. you need the correct values in data.cal (the modelstartdate/time
cooincide with nIter0=0, or startTime=0. in data)

Martin

PS. with your parameters, you read the first and second forcing field at
the beginning, and then interpreted the second as being 3days away, so
that fac nearly 1 and 1-fac nearly zero for the beginning of your
integration (but not exactly)



Georges Djoumna wrote:
> Hi all,
> 
> I am still facing problems with time interpolation in EXF package.
> 
> I've done a test with 3-hourly observed data for the wind. I've  done a
> run for 3 days with   
> ustressperiod     = 259200.0,
> Deltat=150,
> ( I do interpolation in space)
> 
> In exf_set_gen.F  it is written 
> Interpolate linearly onto the  time.
>              genfld(i,j,bi,bj) = exf_inscal_gen * (
>                                fac * gen0(i,j,bi,bj) +
>                    (exf_one - fac) * gen1(i,j,bi,bj) )
> 
> so I print out gen0, gen1 and genfld. 
> gen0 is always the observed data at t=0, and gen1 is always zero which
> are not correct.
> 
> My question is how to fill in  gen1 with correct values? 
> 
> I've checked the binary file for uwindfile and it looks correct.
> 
> Suggestions are all welcome.
> 
> Thanks
> 
> Georges Djoumna 
> 
> 
> ------------------------------------------------------------------------
> 
> _______________________________________________
> MITgcm-support mailing list
> MITgcm-support at mitgcm.org
> http://mitgcm.org/mailman/listinfo/mitgcm-support

-- 
Martin Losch
Alfred Wegener Institute for Polar and Marine Research
Postfach 120161, 27515 Bremerhaven, Germany;
Tel./Fax: ++49(0471)4831-1872/1797