ON THE PHOTOMETRIC ACCURACY OF RHESSI IMAGING AND SPECTROSOCOPY

We compare the photometric accuracy of spectra and images in flares observed with the Ramaty High-Energy Solar Spectroscopic Imager (RHESSI) spacecraft. We test the accuracy of the photometry by comparing the photon fluxes obtained in different energy ranges from the spectral-fitting software SPEX with those fluxes contained in the images reconstructed with the Clean, MEM, MEM-Vis, Pixon, and Forward-fit algorithms. We quantify also the background fluxes, the fidelity of source geometries, and spatial spectra reconstructed with the five image reconstruction algorithms. We investigate the effects of grid selection, pixel size, field of view, and time intervals on the quality of image reconstruction. The detailed parameters and statistics are provided in an accompanying CD-ROM and web page. We find that Forward-fit, Pixon, and Clean have a robust convergence behavior and a photometric accuracy in the order of a few percent, while MEM does not converge optimally for large degrees of freedom (for large field of view and/or small pixel sizes), and MEM-Vis suffers in the case of time-variable sources. This comparative study documents the current status of the RHESSI spectral and imaging software, one year after launch.