Smoothing of an All-sky Survey Map with a Fisher-von Mises Function

A convolution of the all-sky survey data with a smoothing function is crucial in calculating the smooth surface brightness of the sky survey data. The convolution is usually performed using a spherical version of the convolution theorem. However, a Gaussian function, applicable only in the flat-sky approximation, has been usually adopted as a smoothing kernel. In this paper, we present an exact analytic solution of the spherical-harmonics transformation of a Fisher-von Mises function, the mathematical version of a Gaussian function in spherical space. We also obtain the approximate solutions exp[-l(l + 1)/2k]. The exact and the approximate solutions may be useful when an astrophysical survey map is convolved with a smoothing function of k>1.