Visualization Far-Side Magnetic Images from Helioseismic Holography

Active regions on the Sun are the sources of energetic phenomena (e.g., solar flares and coronal mass ejections); which present a direct threat to telecommunications and power transmission on the Earth and pose significant hazards to astronauts and spacecraft. Seismic imaging of far-side of the Sun allows anticipation of the appearance of large active regions ahead of their arrival on the front-side. This improves space weather forecasting. Direct imaging of the far-side is currently not available and we rely on seismic holography to infer the far-side magnetic field. However, mapping between holograms and magnetic field images is non-trivial. In this work, We use supervised Convolution Neural Networks (CNNs) to map far-side magnetic images from helioseismic holography images.