Transition Function Parameter Matrix Type

A Transition Function Parameter Matrix controls the shape and size of patches generated for a given From/To transition, used by Patcher and Expander as their transitionParameters input. Each entry gives three parameters:

  • Mean Patch Size — the average size of a generated patch, in hectares.
  • Patch Size Variance — how much patch size varies around that mean, in the same units. A variance of zero produces uniformly-sized patches; a high variance relative to the mean produces widely varying patch sizes.
  • Patch Isometry — a dimensionless factor controlling how compact or elongated a patch's shape is. Low values (near 0) produce linear, stretched-out patches; higher values (1.5-2 in Dinamica's own published examples) produce compact, roundish patches that tend to nucleate around a point rather than stretch outward.

GUI Editor

Invalid Link
Graphical representation of the transition function parameter matrix editor

The editor presents a table with fixed From, To, Mean Patch Size, Patch Size Variance, and Patch Isometry columns.

EGO Script

A Transition Function Parameter Matrix constant has its own dedicated syntax in EGO Script. Each entry is written FROM→TO MEAN VARIANCE ISOMETRY, with FROM and TO as integers joined directly by with no space, followed by the three real-number parameters in order. As with Transition Matrix, a stationary entry (FROM equal to TO) cannot be written. Commas around each entry are optional, and none are needed between the three parameter values themselves:

[
  1->2 20.0 5.0 1.5,
  1->3 5.0 0.0 0.0,
  2->1 10.0 2.0 1.0
]

Automatic Conversions

  • Converted from: Table Type — the table must have the columns From* (Real), To* (Real), Mean_Patch_Size (Real), Patch_Size_Variance (Real), and Patch_Isometry (Real).
  • Converted to: Table Type, using that same column layout. Unlike Transition Matrix, Change Matrix, and Percent Matrix, this type does not convert to or from Lookup Table Type.

See Type System for the complete conversion reference across all types.