Point Mode

Point mode is a special mode in CHM that evaluates the set of modules only at a specific set of points, i.e., 1D mode. Specifically, only the triangles associated with the output points are run. Four criteria must be met to take advantage of this mode

1. point_mode is defined in the option configuration section

2. The module point_mode is used

3. Only parallel::data modules are used

4. Have only timeseries outputs at named points

The last constraint limits the module selection to only those modules that operate in a column mode and do not have any requirement on the surrounding mesh elements. This prohibits the use of modules such as blowing snow and shadowing.

Step 1. Set options flag

In the option section of the configuration, set point_mode

option:
{
       "point_mode": {
            //empty
         },
}

Point mode works with either netcdf or ascii file input forcing. If used with ASCII forcing data, a station name may be optionally given to force that specific station to be used. If this is done then only 1 output point / triangle is possible to ensure a 1:1 mapping of the two.

option:
{
       "point_mode": {
            "forcing": "my_station"
         },
}

Please see option for more details.

Step 2. Add point_mode to modules

The point_mode module acts as a shim and allows for passing met from the forcing file through to the other modules un touched. This is useful if you have an insitu met station collection data that you wish to use directly in, e.g., the snow module. Please see the point_mode module for more information.

Step 3. Only use data parallel

Only column mode (1D) models are available in point_mode. Currently the only way to confirm this is by checking the source code of a module, although CHM will inform you at runtime if a parallel::domain module is chosend

Step 4. Timeseries output

Only timeseries output at named points can be used. E.g.,

"output":
{
   "UpperClearing":
   {
      "longitude": -115.175362,
      "latitude": 50.956547,
      "file": "uc.txt",
      "type": "timeseries"
   },
   "VistaView":
   {
      "longitude": -115.172169,
      "latitude": 50.970944,
      "file": "vv.txt",
      "type": "timeseries"
   }
},