Simulation of low radio frequency solar images using HART

The diagnostic potential of low radio frequency (< 300 MHz) solar observations has long been recognized. The radio waves are refracted by the smoothly and slowly varying large scale coronal structure and scattered by the small scale inhomogeneities. In addition, the presence of coronal magnetic fields make the coronal plasma dichroic in nature implying that even the unpolarized thermal radiation picks up some degree of polarization depending upon the details of the magnetic field geometry. The very same effects which impart the low radio frequencies its rich diagnostic power, also complicate the interpretation of these observations to extract coronal physics. A detailed analysis of coronal brightness temperature images necessarily requires a sophisticated understanding of coronal propagation and a robust and flexible numerical implementation to serve as a simulation tool. In anticipation of the solar images from the new generation of capable low radio frequency interferometers like the Murchison Widefield Array (MWA), we have been working on the design and development of a coronal propagation simulation tool. Christened Haystack and AOSS Ray Tracer (HART), this tool traces rays through a corona with specified electron density and temperature distributions. HART computes the appropriate radiative transfer to obtain the brightness temperature for each of the rays. This results in a simulated image corresponding to a specified observing frequency in each of the Stokes parameters. In view of the large number of pixels expected in the eventual images from the MWA and other instruments, and the large number of spectral slices for which these images would need to be simulated, considerable attention was paid to developing and implementing a robust and numerically efficient multi-threaded ray tracing algorithm. Here we describe the salient features of the flexible HART framework, presenting the current status of its implementation and the plans for near term development.