Standalone scripts¶

Although wrfxpy is meant to be an integrated system, some functionalities are exposed through separate scripts. These are detailed in this section.

Domain setup¶

The script domain_setup.sh accepts a domain configuration description and injects the domain configuration into a WPS nmelist file and into an input namelist file. Please refer to the domain configuration description in Forecasting.

Example:

./domain_setup.sh my_domains.json namelist.wps namelist.input

Assuming that my_domains.json contains the following:

{
  "1" : {
    "cell_size" : [1000, 1000],
    "domain_size" : [91, 91],
    "center_latlon" : [39.1, -105.9],
    "truelats" : [38.5, 39.6],
    "stand_lon" : -105.9,
    "time_step" : 5,
    "history_interval" : 15,
    "geog_res" : "0.3s",
    "subgrid_ratio" : [50, 50]
  }
}

Then both namelists will be setup for a single-domain configuration (1km grid cell size, 91 x 91 domain size, 20m fire grid).

Note

The namelist files are overwritten.

Grib retrieval and examination¶

The script grib_retr.sh accepts fourth arguments, the grib source identifier, the UTC start, the end time of a simulation in ESMF format and the ingest directory.

Example:

./grib_retr.sh HRRR 2016-03-26_14:00:00 2016-03-26_19:00:00 ingest

This will find out which GRIB2 files are required to perform this simulation and will download them into subdirectories of the ingest directory.

Tip

Using the wrfxpy ingest directory (or the same directory) consistently will make best use of the transparent local caching functionality. Any files that have already been downloaded are not re-downloaded.

The script grib_tool.sh allows the user to list the contents of a GRIB1/2 file and to convert it to a netCDF file.

Examples:

./grib_tool.sh list <grib-filename>

./grib_tool.sh to_netcdf <input-grib-filename> <message-to-convert> <output-netcdf-file>

Postprocessing¶

The script postprocess.sh accepts four arguments, the wrfout file to process, the variables to postprocess (or an instruction file, see below), the prefix on which to base the filenames and the skip (the script will process every skip-th frame). The script always generates PNG files and KMZ files for each variable and timestamp.

Example:

./postprocess.sh /path/to/wrfout T2,PSFC my_directory/file_prefix 1

Alternatively, instead of listing the variables, a more detailed configuration controlling the colormaps, ranges and other parameters can be specified:

./postprocess.sh /path/to/wrfout @var_instructions my_directory/file_prefix 1

Where the file var_instructions contains:

{
  "FGRNHFX" : {
      "name" : "Grnd Heat flux",
      "colorbar" : "W/m^2",
      "colormap" : "jet",
      "transparent_values" : [0, 1],
      "scale" : [0, 6]
  }
}

Will show the colorbar in W/m^2 units and change the displayed variable name to Grnd Heat flux, set the colormap to jet, ensure that values between 0 and 1 are not shown and fix the scale from 0 to 6.

Tip

For the default and more information on values that can be set, examine src/vis/var_wisdom.py.

Fuel moisture DA¶

The script apply_fmda.sh accepts a single wrfinput path argument and performs a data assimilation step using background covariance.

Example:

./apply_fmda.sh wrfinput_d01

The script will read in the timestamp from the wrfinput file, determine it’s physical extent (lat/lon) and download all observations of 10-hr fuel moisture valid at that time available in the region. Then the equilibrium fuel moisture content is computed and adjusted with respect to the observations using the background covariance. The updated values are written back into the fuel moisture file.

SSH Shuttle¶

The script ssh_shuttle.sh accepts a local directory a remote directory name and an identifier and uploads the entire local directory with simulation results to the remote host configured in conf.json and registers the simulation in the catalog.json file on the remote server.

Examples:

./ssh_shuttle.sh wksp/my-simulation/products test_fire_april test_fire_april

The script scans all the files in wksp/my-simulation/products and uses SFTP to put them onto the remote host. The remote directory must be either an absolute path or (recommended) should be relative to the remote host root setup in conf.json. The identifier will be used as the description and also as the key under which the simulation is stored in catalog.json on the remote host.

Data cleanup¶

The script cleanup.sh provides functionality to:

list all simulations that are available on a configured visualization server,
remove a selected simulation, freeing up diskspace.

Examples:

./cleanup.sh list

./cleanup.sh delete <simulation-id-from-list>