Description of software

Currently we have the ability to perform high resolution forecast runs of the UM in ensemble mode with MOPS observations either assimilated or not assimilated. The former runs generate innovation files that contain information about the MOPS precipitation observations and the corresponding precipitation forecast values at the MOPS observation locations.

When MOPS precipitation is not assimilated then no such data appear in the innovation files.

This program allows the user to

  1. Select two identical ensemble runs, one with MOPS observations being assimilated (the PPN runs) and one with no MOPS observations being assimilated. (the NOP runs).
  2. Specify the PPN and NOP innovation files.
  3. Specify the pp_001.MOPS fieldsfiles (in netCDF format) containing the precipitation foecasts for the NOP run.
  4. Extract the space and time coordinates of the MOPS observation from the PPN innovation files.
  5. Interpolate the fieldsfile precipitation forecasts to these MOPS observation locations/times.
  6. Write these interpolated forecasts and the MOPS precipitation observations to the NOP innovation files.

Thus the NOP files now contain precipitation information.

Location of source code

The IDL source code can be found at /home/h05/frbr/idl_source_code/

The code uses library routines written by Andy Heaps (NCAS, University of Reading) and these are accessed by adding these two lines into your .profile file:

export IDL_PATH=+/usr/local/itt
.  /data/nwp1/frbr/TIDLWorkspace/andy_lib/

Method of use - batch mode

  1. The user should make a copy of the IDL source code in his/her working directory.
  2. Next, create a text file called raininterp_controlfile in the working directory. This file should contain the following:
    107                                 ; start cycle (innovation files)
    107                                 ; end cycle (innovation files)
    /export/carrot/raid1/brugge/sm03/   ; no precip run innov
    /export/carrot/raid1/brugge/sm04/   ; precip run innov
    /export/carrot/raid1/brugge/sm03/   ; no precip run MOPS
    _qwq999.oper                        ; MOPS filename                    ; MOPS filename
    Innovations_qwq                     ; ppn innovation filename
    .txt                                ; ppn innovation filename
    Innovations_qwq                     ; nop innovation filename
    .txt                                ; nop innovation filename

    where the records should been changed to reflect the input data and output requirements of the user. The text to the right of the ';' symbols can be kept as a reminder of the meaning of the data records.

    Lines 3 to 5 are the full pathnames of the data directories. Lines 6 to 11 assume that the filenames are of the form:

    Note that the cycling is by innovation file cycle number; using innovation file cycle 'n' requires pp_001.MOPS file cycle 'n-1' because of the way the files are created.

  3. The idl code can now been run in interactively or in a batch job and will loop through the specified cycles.

The output innovation files will be added to the data directory of the NOP innovation files and be given names like ppn_added_Innovations_qwqCCC.txt.

Current code developments