modelTransform

Build a prefix-retained transform for model vertical grids.

Declaration 

 transform = modelTransform(self,options)

Parameters

  • self InternalModesBasis instance
  • options.component "F", "G", or "both"
  • options.nModes requested retained mode count
  • options.projectionTolerance leakage tolerance for prefix selection
  • options.maxConditionNumber condition-number limit for prefix selection
  • options.nTailCheck number of rejected tail modes checked for leakage
  • options.nonlinearAliasingPolicy "none" or "quadratic"
  • options.allowNoncanonical true to allow numerical wave-F projections

Returns

  • transform InternalModesTransform with prefix-retained modes

Discussion

Projection resolvability and nonlinear de-aliasing are treated as separate limits. If nonlinearAliasingPolicy is "quadratic", the transform automatically applies the two-thirds retained-mode cap and records that diagnostic separately from projection leakage.

\[n_{\mathrm{retained}} = \min(n_{\mathrm{requested}},n_{\mathrm{projection}}, n_{\mathrm{nonlinear}}).\]