Calculate the eigenfunctions of the spherical-cap concentration problem for a single angular order.


call SHReturnTapersM (theta0, lmax, m, tapers, eigenvalues, shannon, degrees, ntapers, exitstatus)


theta0 : input, real(dp)
The angular radius of the spherical cap in radians.
lmax : input, integer
The spherical harmonic bandwidth of the localization windows.
m : input, integer
The angular order of the localization windows.
tapers : output, real(dp), dimension (lmax+1, lmax+1)
The spherical harmonic coefficients of the lmax+1 localization windows, arranged in columns. The first and last rows of each column correspond to spherical harmonic degrees 0 and lmax, respectively, and the columns are arranged from best to worst concentrated. Only the first ntapers columns are non-zero.
eigenvalues : output, real(dp), dimension (lmax+1)
The concentration factors of the localization windows.
shannon : output, optional, real(dp)
The Shannon number, which is the trace of the concentration kernel.
degrees : input, integer, optional, dimension (lmax+1)
List of degrees to use when computing the eigenfunctions. Only those degrees where degrees(l+1) is non-zero will be employed.
ntapers : output, integer, optional
The number of non-zero tapers.
exitstatus : output, optional, integer
If present, instead of executing a STOP when an error is encountered, the variable exitstatus will be returned describing the error. 0 = No errors; 1 = Improper dimensions of input array; 2 = Improper bounds for input variable; 3 = Error allocating memory; 4 = File IO error.


SHReturnTapersM will calculate the eigenfunctions of the spherical-cap concentration problem for a singule angular order. The 4pi normalized spherical harmonic coefficients of each window are given in the columns of tapers, and the corresponding concentration factors are given in eigenvaules. The columns of tapers are ordered from best to worst concentrated, and the first and last rows of each column correspond to spherical harmonic degrees 0 and lmax, respectively. The localization windows are normalized such that they have unit power. If the optional vector degrees is specified, then the eigenfunctions will be computed using only those degrees l where degrees(l+1) is not zero.

When possible, the eigenfunctions are calculated using the kernel of Grunbaum et al. 1982 and the eigenvalues are then calculated by integration using the definition of the space-concentration problem. Use of the Grunbaum et al. kernel is prefered over the space-concentration kernel as the eigenfunctions of the later are unreliable when there are several eigenvalues identical (within machine precision) to either 1 or zero. If, the optional parameter degrees is specified, and at least one element is zero for degrees greater or equal to abs(m), then the eigenfunctions and eigenvalues will instead be computed directly using the space-concentration kernel.


Grunbaum, F. A., L. Longhi, and M. Perlstadt, Differential operators commuting with finite convolution integral operators: Some non-abelian examples, SIAM, J. Appl. Math. 42, 941–955, 1982.

Simons, F. J., F. A. Dahlen, and M. A. Wieczorek, Spatiospectral concentration on a sphere, SIAM Review, 48, 504-536, 2006.

Wieczorek, M. A. and F. J. Simons, Localized spectral analysis on the sphere, Geophys. J. Int., 162, 655-675, 2005.

See also

shreturntapers, computedg82, computedm

Tags: fortran
Edit me