Cost function for fitting Fluspect leaf parameters to measured reflectance/transmittance

get.COST_4Fluspect.for.SCOPE builds the residual vector between simulated and measured leaf reflectance/transmittance for a given parameter vector, intended for use inside a non-linear least-squares optimizer when calibrating leaf biochemistry/structure parameters against measured spectra. Note: this function calls an unqualified fluspect_B_CX_PSI_PSII_combined(), which is neither defined in this package nor exported by ToolsRTM - it will fail with "could not find function" until that dependency is resolved. Not exported; needs review before use.

Usage

get.COST_4Fluspect.for.SCOPE(params, measurement, input)

Arguments

params: numeric vector of length 8, in order: Cab, Cdm, Cw, Cs, Cca, Cant, Cx, N - candidate leaf biochemistry/structure parameters.
measurement: list. Measured spectra, with elements refl, tran (measured reflectance/transmittance) and std (measurement uncertainty), all as a function of wavelength.
input: list of length 6: leafbio (baseline leaf parameters), optipar (Fluspect optical parameters), spectral (spectral configuration, including wlP model wavelengths and wlM measurement wavelengths), include (list of 0/1 flags selecting which parameters in params are actually varied), target (character, "0" for combined reflectance+transmittance residuals, "1" for reflectance only, otherwise transmittance only), and range (list with wlmin/wlmax defining the wavelength range used to compute residuals).

Value

A list with: er (residual vector, or two-column matrix of reflectance/transmittance residuals when target == "0"), refl/tran (simulated reflectance/transmittance interpolated to spectral$wlM), and leafopt (full Fluspect output).

Author

Carlos Camino

Examples

if (FALSE) { # \dontrun{
cost <- get.COST_4Fluspect.for.SCOPE(params, measurement, input)
} # }