forcing

[nageswararao]

Hi all,
I want to use BULK_FLUXES option in cppdefs.h and I had prepared monthly climatological forcing files of wind vectors,shortwave radiation flux,rainfall rate,air temperature,surface relative humidity,etc.... for indian ocean region. When I force the model with these fluxes I am getting the following errors. Please tell me what makes me to get this error. Can anyone guide me to rectify this error.
The error it showing is:

0: OPENCDF - cannot find attribute: type (global) in input NetCDF file: /home/revathy1/roms-2.2/tests1/IO/uwind.nc
0: GET_2DFLD - surface u-wind component, t = 15.0000
0: (File: uwind.nc, Rec=0001, Index=1)
0: (Min = -7.96056890E+00 Max = 6.09448481E+00)
0:
0: OPENCDF - cannot find attribute: type (global) in input NetCDF file: /home/revathy1/roms-2.2/tests1/IO/uwind.nc
0:
0: OPENCDF - cannot find attribute: type (global) in input NetCDF file: /home/revathy1/roms-2.2/tests1/IO/vwind.nc
0: GET_2DFLD - surface v-wind component, t = 15.0000
0: (File: vwind.nc, Rec=0001, Index=1)
0: (Min = -1.02274599E+01 Max = 9.29893780E+00)
0: GET_2DFLD - solar shortwave radiation flux, t = 15.0000
0: (File: roms_io_frc.nc, Rec=0001, Index=1)
0: (Min = 2.55495153E-05 Max = 8.10239792E-05)
0:
0: OPENCDF - cannot find attribute: type (global) in input NetCDF file: /home/revathy1/roms-2.2/tests1/IO/uwind.nc
0:
0: OPENCDF - cannot find attribute: type (global) in input NetCDF file: /home/revathy1/roms-2.2/tests1/IO/vwind.nc
0:
0: OPENCDF - cannot find attribute: type (global) in input NetCDF file: /home/revathy1/roms-2.2/tests1/IO/tair.nc
0: GET_2DFLD - surface air temperature, t = 15.0000
0: (File: tair.nc, Rec=0001, Index=1)
0: (Min = 1.15794601E+01 Max = 2.92110901E+01)
0:
0: OPENCDF - cannot find attribute: type (global) in input NetCDF file: /home/revathy1/roms-2.2/tests1/IO/uwind.nc
0:
0: OPENCDF - cannot find attribute: type (global) in input NetCDF file: /home/revathy1/roms-2.2/tests1/IO/vwind.nc
0:
0: OPENCDF - cannot find attribute: type (global) in input NetCDF file: /home/revathy1/roms-2.2/tests1/IO/tair.nc
0:
0: OPENCDF - cannot find attribute: type (global) in input NetCDF file: /home/revathy1/roms-2.2/tests1/IO/hair.nc
0: GET_2DFLD - surface air relative humidity, t = 15.0000
0: (File: hair.nc, Rec=0001, Index=1)
0: (Min = 6.16724610E-03 Max = 8.70529115E-03)
0:
0: OPENCDF - cannot find attribute: type (global) in input NetCDF file: /home/revathy1/roms-2.2/tests1/IO/uwind.nc
0:
0: OPENCDF - cannot find attribute: type (global) in input NetCDF file: /home/revathy1/roms-2.2/tests1/IO/vwind.nc
0:
0: OPENCDF - cannot find attribute: type (global) in input NetCDF file: /home/revathy1/roms-2.2/tests1/IO/tair.nc
0:
0: OPENCDF - cannot find attribute: type (global) in input NetCDF file: /home/revathy1/roms-2.2/tests1/IO/hair.nc
0:
0: OPENCDF - cannot find attribute: type (global) in input NetCDF file: /home/revathy1/roms-2.2/tests1/IO/srflx.nc
0:
0: OPENCDF - cannot find attribute: type (global) in input NetCDF file: /home/revathy1/roms-2.2/tests1/IO/rain.nc
0:
0: GET_2DFLD - unable to find requested variable: rain
0: in input NetCDF file: /home/revathy1/roms-2.2/tests1/IO/rain.nc
0:
0: ROMS/TOMS - Input error ............. exit_flag: 2
0:
0:
0: Elapsed CPU time (seconds):

Thanks in advance.

With regards,
G.NageswaraRao.

[kate]

Most of these are warnings you can fix by adding a global "type" attribute to your files. The type should be "FORCING", though it doesn't really matter. If you have the NCO package, you can do it with:

ncatted -a type,global,c,c,"FORCING" file.nc

Then you have the rain problem. Is there a rain variable in rain.nc?

[nageswararao]

Thanks Kate,
U are right. I think it stopped due to that attribute error but now I came to know that it is unable to find rain_time variable. Now I gave it and I had successfully run it.
But I had some doubt, where I can find surface air pressure, cloud factor, net longwave radiation COADS climatology data? Can anybody tell me, where I could found them and also the matlab file to interpolate them to the ROMS grid ?

Thanks in advance.

With regards,
G.NageswaraRao.

[wilkin]

There are many sources of surface meteorological data suitable for forcing ROMS. It all depends on your region and time scales of interest, and how you would like to specify the air-sea fluxes (whether as direct fluxes, or via the bulk formulae).

For example, the NOAA NOMADS opendap server has NCEP/DOE global montly average reanalysis fields
http://nomad3.ncep.noaa.gov/ncep_data/index.html
http://nomad3.ncep.noaa.gov:9090/dods/r ... b/pgb.info

Or there are many data sets available through the NVODS Live Access Server http://ferret.pmel.noaa.gov/NVODS/servl ... titem=3411

These are all gridded data sets and easily interpolated to your model in matlab. You don't need any special matlab code for that. I would suggest its generally better to do this data handling yourself so you can sure you are taking the data peparation steps you intend.

[nageswararao]

Sorry for the late reply.
Thank u Wilkin for ur suggestion and I had downloaded NCEP climatology fluxes and I could interpolate to ROMS grid also.

With regards,
G.NageswaraRao.