Optimized data masks for focussed solar tomography: background and artificial diagnostic experiments

Abstract

Context: The use of solar tomography for detecting subsurface features in the Sun is now well established. It customarily proceeds from an analysis of data on the solar surface given weightings in a predetermined geometric (e.g. centre-annulus) configuration.
Aims: We seek to improve these weightings by developing a scheme for choosing optimal combinations of data that maximise the contribution from signal at a desired horizontal and depth location.
Methods: We employ a subtractive optimally localized averaging (SOLA) scheme to pick weights for particular data points and analyse the quality of the results.
Results: We show in this work that particularly by using wave kernels instead of rays for modelling purposes, one can do well at localizing a measurement through our techniques while achieving desirable error-magnification properties.