[Wrf-users] wps intermediate file using ncl with NASA MERRA

[Dominikus Heinzeller]

Hi Nikhil,

Forcing WRF with model level data is possible and often done - for example for the ECMWF operational forecasts with 138 hybrid levels nowadays. For ECMWF data, a utility program called calc_ecmwf_p.exe comes with WPS, which takes a text file with coefficients a and b to generate additional intermediate files (prefix PRES). These contain pressure, geopotential height (and relative humidity) for each model level, grid  point and time step. The ECMWF model levels are defined as p(k) = a(k) + b(k)*psfc, see the following links:

http://rda.ucar.edu/datasets/ds627.0/docs/Eta_coordinate
http://www.ecmwf.int/en/forecasts/documentation-and-support/137-model-levels
https://www.met.reading.ac.uk/wrf/?page_id=58

For MERRA, the sigma coordinate is very similar, except that it corresponds to the pressure at the bottom edge of the grid box, whereas you need the pressure at the center (http://wiki.seas.harvard.edu/geos-chem/index.php/Notes_regarding_GEOS_vertical_grids, note the typo in the definition of pcenter). You could either re-calculate the indices a and b in order to use calc_ecmwf_p.exe or you could write your own little tool that calculates the missing pressure data and writes it into intermediate format. Template code in Fortran can be found here:

Writing Meteorological Data to the Intermediate Format <http://www2.mmm.ucar.edu/wrf/users/docs/user_guide_V3/users_guide_chap3.htm#_Writing_Meteorological_Data>

Cheers

Dom/

> On 1/02/2016, at 1:17 PM, Nikhil <nikhilgarg.gju at gmail.com> wrote:
> 
> 
>> In my little experience, I have found that usually data used in WRF from sources for example CFSR, NCEP FNL or ECMWF ERA Interim are interpolated to pressure levels and then ingested in WRF model.
>> I have recently tried to write the WRF input files using the MERRA data, but I have kept the data on the model levels which in case of MERRA are GEOS vertical levels.
>> I was wondering whether real.exe can figure out the vertical distribution
>> from the GEOS level to the eta level or do I need to provide some
>> information (example vertical levels) in namelist files for real or metgrid. 
>> I have gone through the WRF manual and I could not find any specific
>> information on the data structure needed by WRF model, apart from a
>> specific format for intermediate file. For my model, I have used GLDAS
>> data for soil moisture etc. I have seen some cases on ncl (ncarg) website where they show how the intermediate file are written and I see that usually data in those example scripts is given on pressure levels.
>> I appreciate any help in this matter.
> _______________________________________________
> Wrf-users mailing list
> Wrf-users at ucar.edu
> http://mailman.ucar.edu/mailman/listinfo/wrf-users

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.ucar.edu/pipermail/wrf-users/attachments/20160202/29ba1bc0/attachment.html