Section: ForestDrag

These parameters are active when ForestDrag is included in incflo.physics, and ForestForcing is listed as a source term in the ICNS (momentum) equation. Two forest representations are supported:

A legacy cylinder-based forest file, controlled by ForestDrag.forest_file.
A point-cloud forest description, controlled by ForestDrag.point_cloud_files.

The two approaches are mutually exclusive and cannot be used together in the same run.

ForestDrag.forest_file

type: String, optional, default = forest.amrwind

Input file for the legacy cylinder-based forest representation. If ForestDrag is enabled and ForestDrag.point_cloud_files is not provided, Kynema-SGF reads this file.

Each non-comment line of the file must contain eight values,

TreeType xc yc height diameter cd lai laimax

where TreeType selects the vertical LAD model, xc and yc are the forest-center coordinates, height and diameter define the cylindrical footprint, cd is the drag coefficient, lai is the leaf area index, and laimax gives the normalized height of the LAD maximum for the type-2 profile.

The supported legacy LAD models are the same as those described in the theory documentation: a uniform profile and an analytical vertically varying profile.

ForestDrag.point_cloud_files

type: List of strings, optional

List of point-cloud files, with one file per forest patch. When this parameter is present, the legacy cylinder-based input ForestDrag.forest_file must not be specified.

Each non-comment line of a point-cloud file must contain four values,

x y z lad

where x, y, and z are the point coordinates and lad is the leaf area density sampled at that point.

Blank lines are ignored. Text following # on a line is treated as a comment.

ForestDrag.coefficients_of_drag

type: List of reals, mandatory when ForestDrag.point_cloud_files is used

Drag coefficient for each point-cloud forest. The number of entries in this list must match the number of files listed in ForestDrag.point_cloud_files.

ForestDrag.point_neighbors

type: Integer, optional, default = 4

Number of nearest point-cloud samples used to reconstruct the local LAD at a cell center. Values are internally clamped to the range 1 through 8.

ForestDrag.point_interp_eps

type: Real, optional, default = 1.0e-12

Regularization parameter used in the inverse-distance weighting for the point-cloud forest model. This value avoids singular weights when a cell center is extremely close to a point-cloud sample.