ARTS built-in documentation server

Workspace Method covmat1D


Create 1D covariance matrix.

Creates a 1D covariance matrix for two retrieval quantities on given 
 grids from a given functional form. Elements  of the covariance matrix
are computed as
 S_{i,j} = sigma_i * sigma_j * f(d_{i,j} / l_{i,j}) 
 where d_{i,j} is the distance between the two grid points and l_{i,j}
 the mean of the correlation lengths of the grid points.

 If a cutoff value co is given elements with absolute value less than this 
 are set to zero.

The following functional forms are available:
  "exp": f(x) = exp(-x) 
  "lin": f(x) = 1.0 - x, for x > 1.0, 0.0 otherwise 
  "gauss": f(x) = exp(-x^2) 

Authors: Simon Pfreundschuh


covmat1D( out, grid_1, grid_2, sigma_1, sigma_2, cls_1, cls_2, co, fname )


GOUTout(Matrix, Sparse)The matrix in which to store the covariance matrix.
GINgrid_1(Vector)The retrieval grid for the first retrieval quantity.
GINgrid_2(Vector, Default: [])The retrieval grid for the second retrieval quantity. (If empty taken as grid_1)
GINsigma_1(Vector)The variances of the first retrieval quantity.
GINsigma_2(Vector, Default: [])The variances of the second retrieval quantity.(If empty taken as sigma_1)
GINcls_1(Vector)The correlations lengths of the first retrieval quantity.
GINcls_2(Vector, Default: [])The correlations lengths of the second retrieval quantity.(If empty taken as cls_1)
GINco(Numeric, Default: 0.0)The cutoff value for covariance matrix elements.
GINfname(String)The name of the functional form to use.