Dynamical Imaging with Interferometry

3 Nov 2017  ·  Michael D. Johnson, Katherine L. Bouman, Lindy Blackburn, Andrew A. Chael, Julian Rosen, Hotaka Shiokawa, Freek Roelofs, Kazunori Akiyama, Vincent L. Fish, Sheperd S. Doeleman ·

By linking widely separated radio dishes, the technique of very long baseline interferometry (VLBI) can greatly enhance angular resolution in radio astronomy. However, at any given moment, a VLBI array only sparsely samples the information necessary to form an image. Conventional imaging techniques partially overcome this limitation by making the assumption that the observed cosmic source structure does not evolve over the duration of an observation, which enables VLBI networks to accumulate information as the Earth rotates and changes the projected array geometry. Although this assumption is appropriate for nearly all VLBI, it is almost certainly violated for submillimeter observations of the Galactic Center supermassive black hole, Sagittarius A* (Sgr A*), which has a gravitational timescale of only ~20 seconds and exhibits intra-hour variability. To address this challenge, we develop several techniques to reconstruct dynamical images ("movies") from interferometric data. Our techniques are applicable to both single-epoch and multi-epoch variability studies, and they are suitable for exploring many different physical processes including flaring regions, stable images with small time-dependent perturbations, steady accretion dynamics, or kinematics of relativistic jets. Moreover, dynamical imaging can be used to estimate time-averaged images from time-variable data, eliminating many spurious image artifacts that arise when using standard imaging methods. We demonstrate the effectiveness of our techniques using synthetic observations of simulated black hole systems and 7mm Very Long Baseline Array observations of M87, and we show that dynamical imaging is feasible for Event Horizon Telescope observations of Sgr A*.

PDF Abstract

Categories


Instrumentation and Methods for Astrophysics