windkit.create_wasp_site_effects#

windkit.create_wasp_site_effects(output_locs, n_sectors=12, not_empty=True, site_effects=None, seed=9876538, **kwargs)[source]#

Create empty site-factors dataset.

Parameters:

  • output_locs (xarray.Dataset) – Output geospatial information

  • n_sectors (int) – Number of sectors. Defaults to 12.

  • not_empty (bool) – If true, the empty dataset is filled with random meaningful data.

  • site_effects (list of strings, string, or None) – Variables to include in the output. "all" includes every available variable, including flow_sep_height. A single variable name is also accepted. If omitted, the legacy default variable list is preserved.

  • seed (int) – Seed for the random data, defaults to 9876538.

  • kwargs (dict) – Additional arguments.

Returns:

ds – Empty site factors dataset.

Return type:

xarray.Dataset

Notes

Height-aware variables (displ and the eight speedup/turning variables) are stored float32, C-contiguous, with sector trailing the full spatial batch. Height-independent variables (site_elev, rix, z0meso, slfmeso, flow_sep_height, dirrix) use the horizontal batch instead — never a separate height axis — and are stored the same way. Both are the reverse of the compiled core’s declared argument order, so PyWAsP’s compiled boundary can take a Fortran-ordered view instead of converting a copy (windkit#807). In a point structure, points that share a (west_east, south_north) location get the height-independent values of the first point at that location. Neither the dtype nor the dimension order is a supported API contract: create_wasp_site_effects is a test-data utility.