#include "geoslib_old_f.h"
#include "Calculators/CalcMigrate.hpp"
#include "Variogram/Vario.hpp"
#include "Basic/Utilities.hpp"
#include "Basic/OptDbg.hpp"
#include "Model/Model.hpp"
#include "Neigh/ANeigh.hpp"
#include "Db/Db.hpp"
#include "Db/DbGrid.hpp"
#include <math.h>

Classes
struct	LMlayers

Functions
static LMlayers *	lmlayers_free (LMlayers *lmlayers)

static int	st_get_number_drift (int irf_rank, int flag_ext)

static LMlayers *	lmlayers_alloc (int flag_same, int flag_vel, int flag_cumul, int flag_ext, int flag_z, int colrefd, int colreft, int colrefb, int irf_rank, int match_time, int nlayers)

static void	lmlayers_print (LMlayers *lmlayers)

static int	st_locate_sample_in_output (LMlayers lmlayers, Db dbin, DbGrid dbout, int iech, int igrid)

static void	st_check_layer (const char string, LMlayers lmlayers, int ilayer0)

static int	st_get_props_result (LMlayers lmlayers, Db dbout, int iech, int ilayer0, VectorDouble &props)

static int	st_get_props_data (LMlayers lmlayers, Db dbin, DbGrid *dbout, int iech, int ilayer0, VectorDouble &props)

static double	st_get_drift_result (LMlayers lmlayers, Db dbout, int iech, int ilayer0)

static double	st_get_drift_data (LMlayers lmlayers, Db dbin, DbGrid *dbout, int iech, int ilayer0)

static void	st_covariance_c00 (LMlayers lmlayers, Model model, const VectorDouble &prop1, MatrixSquareGeneral &covtab, double *c00)

static double	st_cij (LMlayers lmlayers, Model model, int ilayer, const VectorDouble &prop1, int jlayer, const VectorDouble &prop2, const double *dd, MatrixSquareGeneral &covtab)

static double	st_ci0 (LMlayers lmlayers, Model model, int ilayer, const VectorDouble &prop1, int jlayer, const double *dd, MatrixSquareGeneral &covtab)

static int	st_drift (LMlayers lmlayers, const double coor, double propval, double drext, int *ipos_loc, VectorDouble &b)

static int	st_lhs_one (LMlayers lmlayers, Db dbin, DbGrid dbout, Model model, VectorInt &seltab, int iech0, int ilayer0, double *coor, VectorDouble &prop0, VectorDouble &prop2, MatrixSquareGeneral &covtab, VectorDouble &b)

static int	st_rhs (LMlayers lmlayers, Db dbin, DbGrid dbout, Model model, double *coor, VectorInt &seltab, int iechout, int ilayer0, VectorDouble &prop0, VectorDouble &prop2, MatrixSquareGeneral &covtab, VectorDouble &b)

static int	st_lhs (LMlayers lmlayers, Db dbin, DbGrid dbout, Model model, VectorInt &seltab, VectorDouble &prop1, VectorDouble &prop2, MatrixSquareGeneral &covtab, double a, double acov)

static void	st_data_vector (LMlayers lmlayers, Db dbin, DbGrid *dbout, VectorInt &seltab, VectorDouble &zval)

static int	st_subtract_optimal_drift (LMlayers lmlayers, int verbose, Db dbin, DbGrid *dbout, VectorInt &seltab, VectorDouble &zval)

static int	st_get_close_sample (LMlayers lmlayers, Db dbin, int iech0, const double *coor)

static int	st_collocated_prepare (LMlayers lmlayers, int iechout, double coor, Db dbin, DbGrid dbout, Model model, VectorInt &seltab, double a, VectorDouble &zval, VectorDouble &prop1, VectorDouble &prop2, MatrixSquareGeneral &covtab, double b2, VectorDouble &baux, double ratio)

static void	st_estimate_regular (LMlayers lmlayers, int flag_std, double c00, double a, VectorDouble &b, double dual, double wgt, double estim, double stdev)

static void	st_estimate_bayes (LMlayers lmlayers, int flag_std, double c00, const double acov, VectorDouble &zval, VectorDouble &b, double wgt, double post_mean, double a0, double cc, double ss, const double gs, double estim, double stdev)

static void	st_estimate (LMlayers lmlayers, Db dbin, DbGrid dbout, Model model, VectorInt &seltab, int flag_bayes, int flag_std, double a, VectorDouble &zval, double dual, VectorDouble &prop1, VectorDouble &prop2, MatrixSquareGeneral &covtab, VectorDouble &b, double b2, VectorDouble &baux, double wgt, double c00, VectorDouble &, double a0, double cc, double ss, double gs, double , double *post_mean)

static int	st_check_auxiliary_variables (LMlayers lmlayers, Db dbin, DbGrid *dbout, VectorInt &seltab)

static void	st_convert_results (LMlayers lmlayers, Db dbout, int flag_std)

static int	st_drift_data (LMlayers lmlayers, Db dbin, DbGrid *dbout, VectorInt &seltab, VectorDouble &prop1, VectorDouble &fftab)

static int	st_drift_bayes (LMlayers lmlayers, int verbose, double prior_mean, const double prior_vars, double acov, VectorDouble &zval, VectorDouble &fftab, double a0, double cc, double ss, double gs, double post_mean, double post_S)

int	multilayers_kriging (Db dbin, DbGrid dbout, Model model, ANeigh neigh, int flag_same, int flag_z, int flag_vel, int flag_cumul, int flag_ext, int flag_std, int flag_bayes, int irf_rank, int match_time, int dim_prior, double prior_mean, double prior_vars, int colrefd, int colreft, int colrefb, int verbose)

static int	st_evaluate_lag (LMlayers lmlayers, Db dbin, DbGrid dbout, Vario_Order vorder, int nlayers, int ifirst, int ilast, VectorDouble &zval, int nval, double distsum, int stat, VectorDouble &phia, VectorDouble &phib, double atab, double *btab)

static int	st_varioexp_chh (LMlayers lmlayers, int verbose, Db dbin, DbGrid dbout, Vario_Order vorder, VectorDouble &zval, int idir, Vario *vario)

int	multilayers_vario (Db dbin, DbGrid dbout, Vario *vario, int nlayers, int flag_vel, int flag_ext, int irf_rank, int match_time, int colrefd, int colreft, int verbose)

static int	st_get_prior (int nech, int npar, VectorDouble &zval, VectorDouble &fftab, double mean, double vars)

int	multilayers_get_prior (Db dbin, DbGrid dbout, Model model, int flag_same, int flag_vel, int flag_ext, int irf_rank, int match_time, int colrefd, int colreft, int colrefb, int verbose, int npar_arg, double mean, double vars)

Function Documentation

◆ lmlayers_alloc()

static LMlayers* lmlayers_alloc	(	int	flag_same,
		int	flag_vel,
		int	flag_cumul,
		int	flag_ext,
		int	flag_z,
		int	colrefd,
		int	colreft,
		int	colrefb,
		int	irf_rank,
		int	match_time,
		int	nlayers
	)

static

Allocate the Multi-Layers internal structure

Returns: Pointer to the allocated structure

Parameters

[in]	flag_same	1 if input and output files coincide
[in]	flag_vel	1 if work is performed in Velocity, 0 for Depth
[in]	flag_cumul	1 if work must be done on Depth; 0 on Thickness
[in]	flag_ext	1 if external drift must be used; 0 otherwise
[in]	flag_z	1 if the output must be provided in depth
[in]	colrefd	Rank of the reference depth variable
[in]	colreft	Rank of the reference Time variable in Dbout
[in]	colrefb	Rank of the reference Bottom Depth variable
[in]	irf_rank	Rank of the Intrinsic Random Function (0 or 1)
[in]	match_time	Pointer to the Time pointer (1 if defined as ELoc::F or 0 for ELoc::TIME)
[in]	nlayers	Number of layers

◆ lmlayers_free()

static LMlayers* lmlayers_free ( LMlayers * lmlayers )

static

Free the Multi-Layers internal structure

Returns: Pointer to the freed structure

Parameters

[in] lmlayers Pointer to the LMlayers structure to be freed

◆ lmlayers_print()

static void lmlayers_print ( LMlayers * lmlayers )

static

Print the Multi-Layers internal structure

Parameters

[in] lmlayers Pointer to the LMlayers structure

◆ multilayers_get_prior()

int multilayers_get_prior	(	Db *	dbin,
		DbGrid *	dbout,
		Model *	model,
		int	flag_same,
		int	flag_vel,
		int	flag_ext,
		int	irf_rank,
		int	match_time,
		int	colrefd,
		int	colreft,
		int	colrefb,
		int	verbose,
		int *	npar_arg,
		double **	mean,
		double **	vars
	)

Multi-layers get the mean and prior matrices for Bayesian prior

Returns: Error return code

Parameters

[in]	dbin	Input Db structure
[in]	dbout	Output Db structure
[in]	model	Model structure
[in]	flag_same	1 if input and output files coincide
[in]	flag_vel	1 if work is performed in Velocity, 0 for Depth
[in]	flag_ext	1 if external drift must be used; 0 otherwise
[in]	irf_rank	Rank of the Intrinsic Random Function (0 or 1)
[in]	match_time	1 if external drift matches time; 0 otherwise
[in]	colrefd	Rank of the reference Depth variable in Dbout
[in]	colreft	Rank of the reference Time variable in Dbout
[in]	colrefb	Rank of the Bottom Depth variable in Dbout (or -1)
[in]	verbose	Verbose option
[out]	npar_arg	Number of drift terms
[out]	mean	Array of means
[out]	vars	Array of variances

◆ multilayers_kriging()

int multilayers_kriging	(	Db *	dbin,
		DbGrid *	dbout,
		Model *	model,
		ANeigh *	neigh,
		int	flag_same,
		int	flag_z,
		int	flag_vel,
		int	flag_cumul,
		int	flag_ext,
		int	flag_std,
		int	flag_bayes,
		int	irf_rank,
		int	match_time,
		int	dim_prior,
		double *	prior_mean,
		double *	prior_vars,
		int	colrefd,
		int	colreft,
		int	colrefb,
		int	verbose
	)

Multi-layers architecture estimation

Returns: Error return code

Parameters

[in]	dbin	Input Db structure
[in]	dbout	Output Db structure
[in]	model	Model structure
[in]	neigh	ANeigh structure
[in]	flag_same	1 if input and output files coincide
[in]	flag_z	1 if the output must be converted back into depth
[in]	flag_vel	1 if work is performed in Velocity, 0 for Depth
[in]	flag_cumul	1 if work is performed in Depth; 0 in Thickness
[in]	flag_ext	1 if external drift must be used; 0 otherwise
[in]	flag_std	1 if the estimation error must be calculated
[in]	flag_bayes	1 if the Bayesian hypothesis is used on drift coeffs
[in]	irf_rank	Rank of the Intrinsic Random Function (0 or 1)
[in]	match_time	1 if external drift matches time; 0 otherwise
[in]	dim_prior	Dimension of the prior information (for verification)
[in]	prior_mean	Vector of prior means for drift coefficients
[in]	prior_vars	Vector of prior variances for drift coefficients
[in]	colrefd	Rank of the reference Depth variable in Dbout
[in]	colreft	Rank of the reference Time variable in Dbout
[in]	colrefb	Rank of the Bottom Depth variable in Dbout (or -1)
[in]	verbose	Verbose option

◆ multilayers_vario()

int multilayers_vario	(	Db *	dbin,
		DbGrid *	dbout,
		Vario *	vario,
		int	nlayers,
		int	flag_vel,
		int	flag_ext,
		int	irf_rank,
		int	match_time,
		int	colrefd,
		int	colreft,
		int	verbose
	)

Multi-layers architecture experimental variogram

Returns: Error return code

Parameters

[in]	dbin	Input Db structure
[in]	dbout	Output Db structure
[in]	vario	Vario structure
[in]	nlayers	Number of layers
[in]	flag_vel	1 if work is performed in Velocity, 0 for Depth
[in]	flag_ext	1 if external drift must be used; 0 otherwise
[in]	irf_rank	Rank of the Intrinsic Random Function (0 or 1)
[in]	match_time	1 if external drift matches time; 0 otherwise
[in]	colrefd	Rank of the reference Depth variable in Dbout
[in]	colreft	Rank of the reference Time variable in Dbout
[in]	verbose	1 for a verbose option

◆ st_check_auxiliary_variables()

static int st_check_auxiliary_variables	(	LMlayers *	lmlayers,
		Db *	dbin,
		DbGrid *	dbout,
		VectorInt &	seltab
	)

static

Check all the auxiliary variables

Returns: The number of active samples

Parameters

[in]	lmlayers	LMlayers structure
[in]	dbin	Input Db structure
[in]	dbout	Output Db structure
[in,out]	seltab	Number of sample definition (0, 1 or 2)

◆ st_check_layer()

static void st_check_layer	(	const char *	string,
		LMlayers *	lmlayers,
		int	ilayer0
	)

static

Check if the target layer rank is consistent

Parameters

[in]	string	Name of the calling procedure
[in]	lmlayers	LMlayers structure
[in]	ilayer0	Rank of the target layer (starting from 1)

Remarks: If this target layer rank is not correct, mes_abort() is called; and the program is interrupted as this must never happen.

◆ st_ci0()

static double st_ci0	(	LMlayers *	lmlayers,
		Model *	model,
		int	ilayer,
		const VectorDouble &	prop1,
		int	jlayer,
		const double *	dd,
		MatrixSquareGeneral &	covtab
	)

static

Calculate the covariance between data and target

Returns: The covariance terms or TEST

Parameters

[in]	lmlayers	LMlayers structure
[in]	model	Model
[in]	ilayer	Layer of interest (data point). Starting from 1
[in]	prop1	Working array at data point (Dimension: nlayers)
[in]	jlayer	Layer of interest (target point). Starting from 1
[in]	dd	Distance vector (or NULL for zero-distance)
[out]	covtab	Working array (Dimension = nlayers * nlayers)

Remarks: : As this function may return TEST, TEST value should be tested

◆ st_cij()

static double st_cij	(	LMlayers *	lmlayers,
		Model *	model,
		int	ilayer,
		const VectorDouble &	prop1,
		int	jlayer,
		const VectorDouble &	prop2,
		const double *	dd,
		MatrixSquareGeneral &	covtab
	)

static

Calculate the covariance between data and data

Returns: The covariance terms or TEST

Parameters

[in]	lmlayers	Pointer to the LMlayers structure to be freed
[in]	model	Model
[in]	ilayer	Layer of interest (first point). Starting from 1
[in]	prop1	Working array at first point (Dimension: nlayers)
[in]	jlayer	Layer of interest (second point). Starting from 1
[in]	prop2	Working array at second point (Dimension: nlayers)
[in]	dd	Distance vector (or NULL for zero-distance)
[out]	covtab	Working array (Dimension = nlayers * nlayers)

Remarks: : As this function may return TEST, TEST value should be tested

◆ st_collocated_prepare()

static int st_collocated_prepare	(	LMlayers *	lmlayers,
		int	iechout,
		double *	coor,
		Db *	dbin,
		DbGrid *	dbout,
		Model *	model,
		VectorInt &	seltab,
		double *	a,
		VectorDouble &	zval,
		VectorDouble &	prop1,
		VectorDouble &	prop2,
		MatrixSquareGeneral &	covtab,
		double *	b2,
		VectorDouble &	baux,
		double *	ratio
	)

static

Perform the per-calculation for estimation with collocated option

Parameters

[in]	lmlayers	LMlayers structure
[in]	iechout	Rank of the target
[in]	coor	Coordinates of the target
[in]	dbin	Input Db structure
[in]	dbout	Output Db structure
[in]	model	Model
[in]	seltab	Number of sample definition (0, 1 or 2)
[in]	a	L.H.S. (square) inverted matrix
[in]	zval	Data vector (extended)
[out]	prop1	Working array (Dimension: nlayers)
[out]	prop2	Working array (Dimension: nlayers)
[out]	covtab	Working array (Dimension = nlayers * nlayers)
[out]	b2	Working vector (Dimension = neq)
[out]	baux	Working vector (Dimension = neq)
[out]	ratio	Ratio value

◆ st_convert_results()

static void st_convert_results	(	LMlayers *	lmlayers,
		Db *	dbout,
		int	flag_std
	)

static

Convert the results in the Depth scale

Parameters

[in]	lmlayers	LMlayers structure
[in]	dbout	Output Db structure
[in]	flag_std	1 if the estimation error must be calculated

Remarks: The conversion is performed:; - if the calculations have been performed in Velocity or Thickness; - if the calculations have been performed in cumulative or not; The standard deviation is also transformed

◆ st_covariance_c00()

static void st_covariance_c00	(	LMlayers *	lmlayers,
		Model *	model,
		const VectorDouble &	prop1,
		MatrixSquareGeneral &	covtab,
		double *	c00
	)

static

Calculate the array of covariances for zero distance

Parameters

[in]	lmlayers	Pointer to the LMlayers structure to be freed
[in]	model	Model
[in]	prop1	Working array at first point (Dimension: nlayers)
[out]	covtab	Working array (Dimension = nlayers * nlayers)
[out]	c00	Returned array (Dimension = nlayers)

Remarks: : This array depends on the target location through proportions

◆ st_data_vector()

static void st_data_vector	(	LMlayers *	lmlayers,
		Db *	dbin,
		DbGrid *	dbout,
		VectorInt &	seltab,
		VectorDouble &	zval
	)

static

Establish the vector of data

Parameters

[in]	lmlayers	LMlayers structure
[in]	dbin	Input Db structure
[in]	dbout	Output Db structure
[in]	seltab	Number of sample definition (0, 1 or 2)
[out]	zval	The data vector (Dimension: neq)

◆ st_drift()

static int st_drift	(	LMlayers *	lmlayers,
		const double *	coor,
		double	propval,
		double	drext,
		int *	ipos_loc,
		VectorDouble &	b
	)

static

Calculate the drift terms

Returns: Error return code

Parameters

[in]	lmlayers	Pointer to the LMlayers structure to be freed
[in]	coor	Array of coordinates
[in]	propval	Value for the proportion (used if flag_cumul=TRUE)
[in]	drext	Value of the external drift
[in,out]	ipos_loc	Address for the first drift term. On output, address for the next term after the drift
[out]	b	Array for storing the drift

◆ st_drift_bayes()

static int st_drift_bayes	(	LMlayers *	lmlayers,
		int	verbose,
		double *	prior_mean,
		const double *	prior_vars,
		double *	acov,
		VectorDouble &	zval,
		VectorDouble &	fftab,
		double *	a0,
		double *	cc,
		double *	ss,
		double *	gs,
		double *	post_mean,
		double *	post_S
	)

static

Calculate the posterior mean and variances in Bayesian case

Returns: Error return code

Remarks: At the end of this function,; invS contains the inverse of prior_S; post_S contains the inverse of post_S

◆ st_drift_data()

static int st_drift_data	(	LMlayers *	lmlayers,
		Db *	dbin,
		DbGrid *	dbout,
		VectorInt &	seltab,
		VectorDouble &	prop1,
		VectorDouble &	fftab
	)

static

Fill the array of drift values at data points

Returns: 1 Error return code

Parameters

[in]	lmlayers	LMlayers structure
[in]	dbin	Input Db structure
[in]	dbout	Output Db structure
[in]	seltab	Number of sample definition (0, 1 or 2)
[in]	prop1	Working array (Dimension: nlayers)
[out]	fftab	Drift array (Dimension: npar[nrow] * nech[ncol])

◆ st_estimate()

static void st_estimate	(	LMlayers *	lmlayers,
		Db *	dbin,
		DbGrid *	dbout,
		Model *	model,
		VectorInt &	seltab,
		int	flag_bayes,
		int	flag_std,
		double *	a,
		VectorDouble &	zval,
		double *	dual,
		VectorDouble &	prop1,
		VectorDouble &	prop2,
		MatrixSquareGeneral &	covtab,
		VectorDouble &	b,
		double *	b2,
		VectorDouble &	baux,
		double *	wgt,
		double *	c00,
		VectorDouble &	,
		double *	a0,
		double *	cc,
		double *	ss,
		double *	gs,
		double *	,
		double *	post_mean
	)

static

Perform the estimation at the grid nodes

Parameters

[in]	lmlayers	LMlayers structure
[in]	dbin	Input Db structure
[in]	dbout	Output Db structure
[in]	model	Model
[in]	seltab	Number of sample definition (0, 1 or 2)
[in]	flag_bayes	1 if the Bayesian hypothesis is used on drift coeffs
[in]	flag_std	1 if the estimation error must be calculated
[in]	a	L.H.S. (square) inverted matrix
[in]	zval	Data vector (extended)
[in]	dual	Dual vector
[out]	prop1	Working array (Dimension: nlayers)
[out]	prop2	Working array (Dimension: nlayers)
[out]	covtab	Working array (Dimension = nlayers * nlayers)
[out]	b	Working vector (Dimension = neq)
[out]	b2	Working vector (Dimension = neq)
[out]	baux	Working vector (Dimension = neq)
[out]	wgt	Working array (Dimension = neq)
[out]	c00	Working array (Dimension = nlayers)
[out]	a0	Constant term
[out]	cc	Output value
[out]	ss	Output value
[out]	gs	Output value
[out]	post_mean	Array of posterior mean

◆ st_estimate_bayes()

static void st_estimate_bayes	(	LMlayers *	lmlayers,
		int	flag_std,
		double	c00,
		const double *	acov,
		VectorDouble &	zval,
		VectorDouble &	b,
		double *	wgt,
		double *	post_mean,
		double *	a0,
		double *	cc,
		double *	ss,
		const double *	gs,
		double *	estim,
		double *	stdev
	)

static

Perform the estimation at the grid nodes in the bayesian case

Parameters

[in]	lmlayers	LMlayers structure
[in]	flag_std	1 if the estimation error must be calculated
[in]	c00	Variance at target
[in]	acov	L.H.S. (square) inverted matrix
[in]	zval	Data vector
[in]	b	Working vector (Dimension = neq)
[in]	wgt	Working array (Dimension = neq)
[out]	post_mean	Array of posterior mean
[out]	a0	Constant term
[out]	cc	Output value
[out]	ss	Output value
[out]	gs	Output value
[out]	estim	Estimated value
[out]	stdev	Standard deviation of estimation error

◆ st_estimate_regular()

static void st_estimate_regular	(	LMlayers *	lmlayers,
		int	flag_std,
		double	c00,
		double *	a,
		VectorDouble &	b,
		double *	dual,
		double *	wgt,
		double *	estim,
		double *	stdev
	)

static

Perform the estimation at the grid nodes in regular case

Parameters

[in]	lmlayers	LMlayers structure
[in]	flag_std	1 if the estimation error must be calculated
[in]	c00	Variance for target
[in]	a	L.H.S. (square) inverted matrix
[in]	b	Working vector (Dimension = neq)
[in]	dual	Dual vector
[in]	wgt	Working array (Dimension = neq)
[out]	estim	Estimated value
[out]	stdev	Standard deviation of estimation error

◆ st_evaluate_lag()

static int st_evaluate_lag	(	LMlayers *	lmlayers,
		Db *	dbin,
		DbGrid *	dbout,
		Vario_Order *	vorder,
		int	nlayers,
		int	ifirst,
		int	ilast,
		VectorDouble &	zval,
		int *	nval,
		double *	distsum,
		int *	stat,
		VectorDouble &	phia,
		VectorDouble &	phib,
		double *	atab,
		double *	btab
	)

static

Evaluate the sill matrix for a given lag

Returns: Error return code (proportions not calculatable)

Parameters

[in]	lmlayers	LMlayers structure
[in]	dbin	Input Db structure
[in]	dbout	Output Db structure
[in]	vorder	Vario_Order structure
[in]	nlayers	Number of layers
[in]	ifirst	Index of the first pair (included)
[in]	ilast	Index of the last pair (excluded)
[in]	zval	Array containing the sample values
[out]	nval	Number of relevant pairs
[out]	distsum	Average distance
[out]	stat	Working array (Dimension: nlayers * nlayers)
[out]	phia	Working array for proportions (Dimension: nlayers)
[out]	phib	Working array for proportions (Dimension: nlayers)
[out]	atab	Working array (Dimension: nhalf * nhalf)
[out]	btab	Working array (Dimension: nhalf)

◆ st_get_close_sample()

static int st_get_close_sample	(	LMlayers *	lmlayers,
		Db *	dbin,
		int	iech0,
		const double *	coor
	)

static

Check if an intercept with the bottom layer is located close enough to the current sample

Returns: 1 if a duplicate must be generated; 0 otherwise

Parameters

[in]	lmlayers	LMlayers structure
[in]	dbin	Input Db structure
[in]	iech0	Rank of sample to be discarded (or -1)
[in]	coor	Coordinates of the target

◆ st_get_drift_data()

static double st_get_drift_data	(	LMlayers *	lmlayers,
		Db *	dbin,
		DbGrid *	dbout,
		int	iech,
		int	ilayer0
	)

static

Return the external drift value at an input location for a target layer

Returns: The external drift value of TEST

Parameters

[in]	lmlayers	LMlayers structure
[in]	dbin	Input Db structure
[in]	dbout	Output Db structure
[in]	iech	Rank of the target sample (in output Db)
[in]	ilayer0	Rank of the target layer (Starting from 1)

◆ st_get_drift_result()

static double st_get_drift_result	(	LMlayers *	lmlayers,
		Db *	dbout,
		int	iech,
		int	ilayer0
	)

static

Return the external drift value at an output location for a target layer

Returns: The external drift value of TEST

Parameters

[in]	lmlayers	LMlayers structure
[in]	dbout	Output Db structure
[in]	iech	Rank of the target sample (in output Db)
[in]	ilayer0	Rank of the target layer (Starting from 1)

◆ st_get_number_drift()

static int st_get_number_drift	(	int	irf_rank,
		int	flag_ext
	)

static

Returns the number of drift functions

Returns: Number of drift conditions

Parameters

[in]	irf_rank	Rank of the Intrinsic Random Function (0 or 1)
[in]	flag_ext	1 if external drift must be used; 0 otherwise

◆ st_get_prior()

static int st_get_prior	(	int	nech,
		int	npar,
		VectorDouble &	zval,
		VectorDouble &	fftab,
		double *	mean,
		double *	vars
	)

static

Determine the mean and variance of drift coefficients

Returns: Error return code

Parameters

[in]	nech	Number of samples
[in]	npar	Number of drift coefficients
[in]	zval	The data vector (Dimension: neq)
[in]	fftab	Drift array (Dimension: npar[nrow] * nech[ncol])
[out]	mean	Array of means
[out]	vars	Array of variances

◆ st_get_props_data()

static int st_get_props_data	(	LMlayers *	lmlayers,
		Db *	dbin,
		DbGrid *	dbout,
		int	iech,
		int	ilayer0,
		VectorDouble &	props
	)

static

Fill the proportion vector at a data location, up to the target layer

Returns: 1 if the proportion vector cannot be defined; 0 otherwise

Parameters

[in]	lmlayers	LMlayers structure
[in]	dbin	Input Db structure
[in]	dbout	Output Db structure
[in]	iech	Rank of the target sample
[in]	ilayer0	Rank of the target layer (starting from 1)
[out]	props	Working array (Dimension: nlayers)

◆ st_get_props_result()

static int st_get_props_result	(	LMlayers *	lmlayers,
		Db *	dbout,
		int	iech,
		int	ilayer0,
		VectorDouble &	props
	)

static

Fill the proportion vector at a output location, up to the target layer

Returns: 1 if the proportion vector cannot be defined; 0 otherwise

Parameters

[in]	lmlayers	LMlayers structure
[in]	dbout	Output Db structure
[in]	iech	Rank of the target sample (in output Db)
[in]	ilayer0	Rank of the target layer (starting from 1)
[out]	props	Working array (Dimension: nlayers)

◆ st_lhs()

static int st_lhs	(	LMlayers *	lmlayers,
		Db *	dbin,
		DbGrid *	dbout,
		Model *	model,
		VectorInt &	seltab,
		VectorDouble &	prop1,
		VectorDouble &	prop2,
		MatrixSquareGeneral &	covtab,
		double *	a,
		double *	acov
	)

static

Calculates the Kriging L.H.S.

Returns: : 1 if the L.H.S. vector cannot be calculated; 0 otherwise

Parameters

[in]	lmlayers	LMlayers structure
[in]	dbin	Input Db structure
[in]	dbout	Output Db structure
[in]	model	Model
[in]	seltab	Number of sample definition (0, 1 or 2)
[out]	prop1	Working array (Dimension: nlayers)
[out]	prop2	Working array (Dimension: nlayers)
[out]	covtab	Working array (Dimension = nlayers * nlayers)
[out]	a	L.H.S. (square) matrix
[out]	acov	L.H.S. (square) covariance matrix

◆ st_lhs_one()

static int st_lhs_one	(	LMlayers *	lmlayers,
		Db *	dbin,
		DbGrid *	dbout,
		Model *	model,
		VectorInt &	seltab,
		int	iech0,
		int	ilayer0,
		double *	coor,
		VectorDouble &	prop0,
		VectorDouble &	prop2,
		MatrixSquareGeneral &	covtab,
		VectorDouble &	b
	)

static

Calculates the L.H.S. for one data

Returns: : 1 if the L.H.S. vector cannot be calculated; 0 otherwise

Parameters

[in]	lmlayers	LMlayers structure
[in]	dbin	Input Db structure
[in]	dbout	Output Db structure
[in]	model	Model
[in]	seltab	Number of sample definition (0, 1 or 2)
[in]	iech0	Rank of the target sample (used for ext. drift)
[in]	ilayer0	Rank of the layer of interest (Starting from 1)
[in]	coor	Coordinates of the data
[in]	prop0	Working array at first point (Dimension: nlayers)
[out]	prop2	Working array (Dimension: nlayers)
[out]	covtab	Working array (Dimension = nlayers * nlayers)
[out]	b	R.H.S. vector (Dimension = neq)

◆ st_locate_sample_in_output()

static int st_locate_sample_in_output	(	LMlayers *	lmlayers,
		Db *	dbin,
		DbGrid *	dbout,
		int	iech,
		int *	igrid
	)

static

Returns the absolute grid node absolute index which is the closest to a given sample of a Db In the case of same input and output file, simply return 'iech'

Returns: 1 if the sample does not belong to the grid; 0 otherwise

Parameters

[in]	lmlayers	LMlayers structure
[in]	dbin	Input Db structure
[in]	dbout	Output Db structure
[in]	iech	Rank in the input Db
[out]	igrid	Rank of the node in the output Db

◆ st_rhs()

static int st_rhs	(	LMlayers *	lmlayers,
		Db *	dbin,
		DbGrid *	dbout,
		Model *	model,
		double *	coor,
		VectorInt &	seltab,
		int	iechout,
		int	ilayer0,
		VectorDouble &	prop0,
		VectorDouble &	prop2,
		MatrixSquareGeneral &	covtab,
		VectorDouble &	b
	)

static

Calculates the Kriging R.H.S.

Returns: : 1 if the R.H.S. has not been calculated; 0 otherwise

Parameters

[in]	lmlayers	LMlayers structure
[in]	dbin	Input Db structure
[in]	dbout	Output Db structure
[in]	model	Model
[in]	coor	Coordinates of the target sample
[in]	seltab	Number of sample definition (0, 1 or 2)
[in]	iechout	Rank of the target sample
[in]	ilayer0	Rank of the layer of interest (Starting from 1)
[out]	prop0	Working array (Dimension: nlayers)
[out]	prop2	Working array (Dimension: nlayers)
[out]	covtab	Working array (Dimension = nlayers * nlayers)
[out]	b	R.H.S. vector (Dimension = neq)

◆ st_subtract_optimal_drift()

static int st_subtract_optimal_drift	(	LMlayers *	lmlayers,
		int	verbose,
		Db *	dbin,
		DbGrid *	dbout,
		VectorInt &	seltab,
		VectorDouble &	zval
	)

static

Calculate the Drift and subtract it from the Data

Parameters

[in]	lmlayers	LMlayers structure
[in]	verbose	1 for a verbose option
[in]	dbin	Input Db structure
[in]	dbout	Output Db structure
[in]	seltab	Number of sample definition (0, 1 or 2)
[out]	zval	The data vector (Dimension: neq)

Remarks: In the current version, the optimal coefficients of the Drift; are output using the set_keypair mechanism using the keyword:; "Optim_Drift_Coeffs" which returns 'ipos' values

◆ st_varioexp_chh()

static int st_varioexp_chh	(	LMlayers *	lmlayers,
		int	verbose,
		Db *	dbin,
		DbGrid *	dbout,
		Vario_Order *	vorder,
		VectorDouble &	zval,
		int	idir,
		Vario *	vario
	)

static

Calculate the variogram for each lag per direction

Returns: Error return code

Parameters

[in]	lmlayers	LMlayers structure
[in]	verbose	1 for a verbose option
[in]	dbin	Input Db structure
[in]	dbout	Output Db structure
[in]	vorder	Vario_Order structure
[in]	zval	Data vector
[in]	idir	Rank of the Direction
[out]	vario	Vario structure

Classes

Functions

Function Documentation

◆ lmlayers_alloc()

◆ lmlayers_free()

◆ lmlayers_print()

◆ multilayers_get_prior()

◆ multilayers_kriging()

◆ multilayers_vario()

◆ st_check_auxiliary_variables()

◆ st_check_layer()

◆ st_ci0()

◆ st_cij()

◆ st_collocated_prepare()

◆ st_convert_results()

◆ st_covariance_c00()

◆ st_data_vector()

◆ st_drift()

◆ st_drift_bayes()

◆ st_drift_data()

◆ st_estimate()

◆ st_estimate_bayes()

◆ st_estimate_regular()

◆ st_evaluate_lag()

◆ st_get_close_sample()

◆ st_get_drift_data()

◆ st_get_drift_result()

◆ st_get_number_drift()

◆ st_get_prior()

◆ st_get_props_data()

◆ st_get_props_result()

◆ st_lhs()

◆ st_lhs_one()

◆ st_locate_sample_in_output()

◆ st_rhs()

◆ st_subtract_optimal_drift()

◆ st_varioexp_chh()