about forcing files

[mashinde]

Dear Frds,

I want to create forcing files for the roms. I read the 'readme' given in the forcing package on wiki roms page. But i do not understand how to make forcings using scripts/programs there.

Thanks


Mahesh

[kate]

I don't know of anyone using that old code anymore. I now use global forcings on their native grid and let ROMS do the interpolation (from CORE 2). We have to run a few NCO commands plus fix the times. What forcings are you hoping to use?

[mashinde]

Dear kate,

I am going to use ecmwf data for forcing....


regards
mahesh

[BenL]

Dear Kate,

I am running a the IRD version of ROMS_AGRIF. I realise this was a different iteration of the code, but I was hoping you might be able to help me with an attempt to hammer the CORE 2b dataset into a usable form for bulk forcing (via NCO or any other method).

Thanks

Ben

[kate]

I don't know that version of ROMS. Can it interpolate for you? For us, the trickiest thing was the clocks. The CORE runs start on a certain day and just count from there for time, throwing away all leap days. Does that work for you? We wanted a calendar that included leap days for tidal consistency so Enrique has a script to put them back in, setting the time origin to 1900.

The temperature was in Kelvins so there's a choice of running ncap2 or simply adding a temperature offset - and making sure that ROMS can parse the offset.

Here's what I get with ncdump -h on the Tair file:

Code: Select all

netcdf Tair.1948-2006 {
dimensions:
        lat = 94 ;
        lon = 192 ;
        tair_time = UNLIMITED ; // (86140 currently)
variables:
        double lat(lat) ;
                lat:long_name = "latitude" ;
                lat:units = "degrees_north" ;
                lat:point_spacing = "uneven" ;
                lat:axis = "Y" ;
        double lon(lon) ;
                lon:long_name = "longitude" ;
                lon:units = "degrees_east" ;
                lon:modulo = 360. ;
                lon:point_spacing = "even" ;
                lon:axis = "X" ;
        double tair_time(tair_time) ;
                tair_time:units = "days since 1900-01-01 00:00:00" ;
                tair_time:axis = "T" ;
                tair_time:bounds = "TIME_bnds" ;
                tair_time:time_origin = "1-JAN-1948" ;
                tair_time:calendar = "LEAP" ;
        float Tair(tair_time, lat, lon) ;
                Tair:missing_value = -1.e+34f ;
                Tair:_FillValue = -1.e+34f ;
                Tair:long_name = "IF Y[G=T_10] GT 60 THEN T_10_MOD1 + 0.49 * FACPAT ELSE T_10_MOD1surface air temperature" ;
                Tair:long_name_mod = "L=1:124" ;
                Tair:history = "From t_10.1948" ;
                Tair:units = "Celsius" ;
                Tair:coordinates = "lon lat" ;
                Tair:add_offset = -273.16f ;

// global attributes:
                :history = "Thu Aug 27 15:34:15 2009: ncatted -O -a add_offset,Tair,a,f,-273.16 Tair.1948-2006.nc\n",
                        "Tue Jul 14 11:03:38 2009: /usr/local/bin/em64t/ncks -p /hosts/pangea/data00/users/achaudhuri/PRS/Forcing/COREv2_48-06/ -v T_10_MOD -O t_10.1948-2006.15JUNE2009.nc /hosts/pangea/data00/users/achaudhuri/PRS/Forcing/ForcingFiles09/t_10.1948-2006.15JUNE2009_roms_Tair.nc\n",
                        "Tue May 12 16:02:38 2009: ncatted -O -a bounds,LAT,d,c,LAT_bnds t_10_mod.1948-2006.nomads.nc\n",
                        "Tue May 12 15:51:30 2009: ncks -x -v LAT_bnds t_10_mod.1948-2006.new.nc t_10_mod.1948-2006.nomads.nc\n",
                        "FERRET V6.11 (beta) 12-May-09" ;
                :Conventions = "CF-1.0" ;
                :nco_openmp_thread_number = 1 ;
                :type = "FORCING file" ;
                :orig_file = "/hosts/pangea/data00/users/achaudhuri/PRS/Forcing/COREv2_48-06/t_10.1948-2006.15JUNE2009.nc" ;

[BenL]

Hi Kate,

I have had some success so far - I just need to adapt the time stepping to accommodate the various temporal resolutions of the data sets, and address an upward long wave radiation issue, and I think I am set.

Thanks for your help,

Cheers

Ben