[Mitgcm-support] Re: pkg/seaice and pkg/exf

[mitgcm-support at dev.mitgcm.org]

Hi Dimitris, 

Thanks for your help: the longwave forcing is fixed, but my problem
remains. I've got a few more questions:

(i) Which shortwave flux should I use if both ALLOW_KPP and ALLOW_ATM_TEMP
are defined?  It would help if these subtleties were documented somewhere.

(ii) I get a compile-time error in growth.F if ALLOW_RUNOFF is undef'ed,
not a runtime warning.

(iii) I'm using SEAICE_EXTERNAL_FORCING and EXF_NO_BULK_COMPUTATIONS so as
to use the pkg/seaice bulk formulae code. My evaporation field looks
wrong, however (it's O(10^-5) m/s and has both signs - I expect O(10^-8),
and only positive values). With EXF_READ_EVAP undef'ed and no evaporation
file defined I was expecting pkg/seaice to calculate the evap. In fact,
exf_getffields.F seems to be used to get it. I'm confused!

Tom.


On Sun, 5 Jan 2003, Dimitris Menemenlis wrote:

> 
> Tom, Happy New Year to you also and thank you for testing the sea-ice code.
> 
> >> I've been exploring your new seaice package (r1_p10) with the exf option
> >> selected (SEAICE_EXTERNAL_FORCING) in my 1deg N. Atlantic model. The code
> >> diverges after a few 100 steps.
> >> 
> >> A likely problem is the longwave radiation. In SEAICE_FFIELDS.h, lwflux is
> >> defined to be downward, hence we should expect negative values. But in
> >> seaice_readparms.F the longwave radiation has a minimum value of 60 Wm^-2.
> >> This seems contradictory.
> 
> For pkg/seaice bulk formulae you need to provide the "downward longwave
> radiation flux", that is, the part of longwave solar radiation that filters
> through the clouds, added to that absorbed and re-emitted downwards by the
> atmosphere.  That number is always positive, on average 300 W/m^2.
> 
> >> Also, your flo.labsea1979 data file has positive values in the range
> >> 100-300 Wm^-2. How did you get these numbers?  The Jan1979 NCEP net
> >> longwave radiation is in the range 20-90 Wm^-2 which seems more
> >> reasonable (attached).
> >> 
> >> I'm having other problems with the freshwater forcing, but they might be
> >> related to the radiative fluxes.
> 
> The flo.labsea1979 file contains a mean monthly climatology derived from the
> 1979-1999 NCEP "downward longwave radiation flux" fields, not from the "net
> longwave radiation" fields.  The later, as you point out, are usually negative
> because they include ocean-surface reflection as well as the black-body
> radiative contribution of SST and sea-ice.
> 
> >> Finally, the ALLOW_RUNOFF option doesn't work anymore when it's undef'ed.
> >> There's a trivial problem in growth.F
> 
> As configured, pkg/seaice requires that ALLOW_RUNOFF be defined.  Actually
> there is a check for this in pkg/seaice/seaice_check.F, which will abort the
> program if "ALLOW_RUNOFF" is undefined.  If runoff is unknown, both pkg/exf
> and pkg/seaice default to zero runoff, as is the case for all the
> verification/lab_sea test experiments.
> 
> Your questions do bring up the issue of sign convention and unit consistency
> between pkg/exf and pkg/seaice bulk formulae, which has been set aside, and
> which I need to sort out with Patrick and co.
> 
> Cheers, Dimitris
> 
> -- 
> Dimitris Menemenlis                      menemenlis at jpl.nasa.gov
> Jet Propulsion Lab, MS 300-323           tel: 818-354-1656
> 4800 Oak Grove Dr, Pasadena CA 91109     fax: 818-393-6720
>