Imaging subsurface solar flows by iterative helioseismic holography.
The main input data in helioseismology consists of extremely noisy, five-dimensional (2^2 spatial+1 temporal dimensions) cross-correlations of line-of-sight velocities at the solar surface. Due to the immense size of the input data, traditional approaches like time-distance helioseismology use only parts of the seismic information. Helioseismic holography, on the other hand, is a physically motivated averaging method consisting in backpropagating solar disturbances. This way helioseismic holography uses the whole seismic information to provide feature maps. Despite its great success in farside imaging, helioseismic holography is no quantitative regularization method at all. There are quantities like subsurface flows, which are in need of nonlinear inversions using the whole seismic information. This task can be tackled by iterative helioseismic holography, which combines converging iterative regularization methods with holographic backpropagation. We will present the theoretical framework of iterative helioseismic holography and show some preliminary results for non-linear inversions. In particular we are interested in the antisymmetric part of the solar differential rotation as traditional approaches like frequency splitting are not sensitive to this quantity. Afterwards we aim to step forwards to more complicated flows like convection and meridional flows.