Re: Problem in converting CCSM hybrid data to pressure levels ...

Hi Ufuk,

I've used vinth2p_ecmwf to interpolate CCSM3 data (T, U,
V, Omega and Z) from hybrid to pressure levels (with
extrapolation=true) and have not any problem. The
interpolated values of U, V or Omega will be held constant
from the lowest model level at the points below ground
(e.g at the 850, 925 and 1000 hPa in parts of the
Antarctica region). The "goodness" of the comparison
between CCSM3 data with the reanalysis (NCEP or ERA40)
will depend on the CCSM3 performance. For example, I've
noted that the (sub-tropical) tropospheric and
stratospheric jets of CCSM3 (seen from the U field) are
more or less well positioned and separated from each other
(as in the reanalysis), but are stronger than those of the
reanalysis.

I've attached a NCL example script to interpolate the
CCSM3 T and U fields from hybrid to pressure levels
(10-1000 hPa), using vinth2p_ecmwf. Note that you will
need the additional fields of SURFACE GEOPOTENTIAL and
SURFACE PRESSURE.

Good Luck,

Carlos

Em Fri, 24 Oct 2008 15:58:49 +0200
  Ufuk Utku Turuncoglu <turuncu_at_be.itu.edu.tr> escreveu:
> Hi Dennis,
>
> The problem same for both case when i set the
>extrapolation option as True and False. I have problem in
>all interpolated vertical layer. In my case i will
>interpolate the hybrid to 18 pressure level. I check all
>of them but the results same.
>
> Thanks for your help,
>
> --ufuk
>
>
> Dennis Shea wrote:
>> Merhaba Ufuk
>>
>> The "vinth2p*" functions are directly out of
>> the old CCM Processor. No changes.
>>
>> In a separate email I will send you the code.
>>
>> It is not clear from the message below, but are you
>> talking about values that have been extrapolated?
>> [Your last argument is "True"]. I never use
>> extrapolation. For winds, the vinth2p* codes just
>> use the value at the lowest 'valid' level.
>>
>>
>>
>> D
>>
>>
>>
>> Ufuk Utku Turuncoglu wrote:
>>> Hi,
>>>
>>> I try to modify RegCM3 ICBC code to handle CCSM hybrid
>>>coordinate
>>> data. So, i create ic/bc files using modified code and
>>>run model
>>> using CCSM data (b30.030e.cam2) for a five year period.
>>>After
>>> investigation and comparing results, i realize that the
>>>precipitation
>>> pattern is not similar with observations.
>>>
>>> To check the results, i convert the raw CCSM data using
>>>both
>>> modified version of the RegCM3 ICBC code and NCL (using
>>>vinth2p_ecmwf
>>> and vinth2p). The results is not good in my case. When i
>>>compare the
>>> results with the NCEP/NCAR Reanalysis data (from CDC), u
>>>and v
>>> component of the wind is not similar with observations
>>>in NCL and
>>> modified ICBC converted case. I use the following
>>>commands to convert
>>> the hybrid levels,
>>>
>>> UonP =
>>>vinth2p(U(tt,:,:,:),hyam,hybm,pnew,PS(tt,:,:),1,P0mb,1,True)
>>> VonP =
>>>vinth2p(V(tt,:,:,:),hyam,hybm,pnew,PS(tt,:,:),3,P0mb,1,True)
>>>
>>> and also i try different interpolation types (lineer,
>>>log etc.) and
>>> vinth2p_ecmwf function but the results is similar.
>>>
>>> Is there anybody who faced this kind of problem? I also
>>>want to study
>>> the NCL vinth2p_ecmwf and vinth2p source codes. Where is
>>>the places
>>> of these functions in the source tar file?
>>>
>>> PS: If you want to see the NCL script, please let me
>>>know.
>>>
>>> Any help will be helpful for me.
>>> Best regards,
>>>
>>> Ufuk Turuncoglu
>>> ITU, Informatics Institute
>>>
>>>
>>
>
>
> --
> This message has been scanned for viruses and
> dangerous content by MailScanner, and is
> believed to be clean.
>
> _______________________________________________
> ncl-talk mailing list
> List instructions, subscriber options, unsubscribe:
> http://mailman.ucar.edu/mailman/listinfo/ncl-talk

_________________________________
Physics Department
University of Aveiro.
Campus Universitário de Santiago
3810-193 Aveiro.
Portugal
_________________________________

_______________________________________________
ncl-talk mailing list
List instructions, subscriber options, unsubscribe:
http://mailman.ucar.edu/mailman/listinfo/ncl-talk