Local and partial evaluation of SH fields.

These do only require a call to shtns_create, but not to shtns_set_grid. More...

Functions
double	SH_to_point (shtns_cfg shtns, cplx *Qlm, double cost, double phi)
	Evaluate scalar SH representation Qlm at physical point defined by cost = cos(theta) and phi.
void	SH_to_grad_point (shtns_cfg shtns, cplx *DrSlm, cplx *Slm, double cost, double phi, double *gr, double *gt, double *gp)
void	SHqst_to_point (shtns_cfg shtns, cplx *Qlm, cplx *Slm, cplx *Tlm, double cost, double phi, double *vr, double *vt, double *vp)
	Evaluate vector SH representation Qlm at physical point defined by cost = cos(theta) and phi.
void	SHqst_to_lat (shtns_cfg shtns, cplx *Qlm, cplx *Slm, cplx *Tlm, double cost, double *vr, double *vt, double *vp, int nphi, int ltr, int mtr)
	synthesis at a given latitude, on nphi equispaced longitude points. More...
void	SH_to_lat (shtns_cfg shtns, cplx *Qlm, double cost, double *vr, int nphi, int ltr, int mtr)
	synthesis at a given latitude, on nphi equispaced longitude points. More...

Detailed Description

These do only require a call to shtns_create, but not to shtns_set_grid.

These functions are not optimized and can be relatively slow, but they provide good reference implemenation for the transforms.

Function Documentation

void SH_to_lat	(	shtns_cfg	shtns,
		cplx *	Qlm,
		double	cost,
		double *	vr,
		int	nphi,
		int	ltr,
		int	mtr
	)

synthesis at a given latitude, on nphi equispaced longitude points.

vr arrays must have nphi+2 doubles allocated (fftw requirement). It does not require a previous call to shtns_set_grid, but it is NOT thread-safe.

void SHqst_to_lat	(	shtns_cfg	shtns,
		cplx *	Qlm,
		cplx *	Slm,
		cplx *	Tlm,
		double	cost,
		double *	vr,
		double *	vt,
		double *	vp,
		int	nphi,
		int	ltr,
		int	mtr
	)

synthesis at a given latitude, on nphi equispaced longitude points.

vr, vt, and vp arrays must have nphi+2 doubles allocated (fftw requirement). It does not require a previous call to shtns_set_grid, but it is NOT thread-safe.