Full-sun synchronic EUV and coronal hole mapping using multi-instrument images: Data and software made available

Abstract:

A method for the automatic generation of EUV and coronal hole (CH) maps using simultaneous multi-instrument imaging data is described. Synchronized EUV images from STEREO/EUVI A&B 195Å and SDO/AIA 193Å undergo preprocessing steps that include PSF-deconvolution and the application of nonlinear data-derived intensity corrections that account for center-to-limb variations (limb-brightening) and inter-instrument intensity normalization. The latter two corrections are derived using a robust, systematic approach that takes advantage of unbiased long-term averages of data and serve to flatten the images by converting all pixel intensities to a unified disk center equivalent. While the number of applications are broad, we demonstrate how this technique is very useful for CH detection as it enables the use of a fast and simplified image segmentation algorithm to obtain consistent detection results. The multi-instrument nature of the technique also allows one to track evolving features consistently for longer periods than is possible with a single instrument, and preliminary results quantifying CH area and shape evolution are shown.

Most importantly, several data and software products are made available to the community for use. For the ~4 year period of 6/10/2010 to 8/18/2014, we provide synchronic EUV and coronal hole maps at 6-hour cadence as well as the data-derived limb brightening and inter-instrument correction factors that we applied. We also make available a ready-to-use MATLAB script EUV2CHM used to generate the maps, which loads EUV images, applies our preprocessing steps, and then uses our GPU-accelerated/CPU-multithreaded segmentation algorithm EZSEG to detect coronal holes.