Most of the light from the Sun is visible wavelengths. By most I mean sunlight at visible wavelengths is 1 million times brighter than the ultraviolet and extreme ultraviolet wavelengths measured by AIA and EVE. Those instruments use filters on the front of their telescopes to keep the visible light photons from reaching the CCDs.
The filters on the AIA telescopes are thin metal foils that are supported by meshes. The rows and columns of the meshes are rotated about 45° from the top of the Sun and are separated by a rotation of about 10°. Unfortunately, the mesh grid diffracts photons and a point of light is seen as a pattern like the image on the left. This image represents the 171 Å “point spread function,” or PSF. For the 171 Å images, almost 2/3 of photons end up in the “correct” location, at the center of the crosshairs. But 1/3 of the photons are diffracted to other places in the image. The amount of diffraction is wavelength-dependent, so each image has its own corresponding PSF. However, the pattern in each wavelength is similar to the one on the left. Planet transits can be used to calibrate that software by providing a disk that should always be completely black but sometimes isn't.
You might remember seeing the PSF pattern as the eight-spoked pattern can appear over a flaring location (like the AIA 131 Å image from August 2014 on the left). The important thing to remember is that this pattern is not happening on the Sun - it is caused by the telescope. Even better, we can use software to “put” the light back into the right place.
Mercury, which will be roughly 36 million miles from the Sun at the time of the transit, should appear completely black against the solar disk. However, photons diffracted from the filter mesh may make Mercury appear temporarily brighter as it passes through a diffraction pattern from a particularly bright region. The AIA team used observations from the Transit of Venus in 2012 to refine their understanding of the PSF and improve our ability to apply the correction to the science data. It is hoped that we can once again take advantage of a should-be-dark planet to further calibrate our images.
Not to worry, we will get another chance to verify the calibration during the next Mercury transit on November 11, 2019!
You can safely watch the transit at http://mercurytransit.gsfc.nasa.gov. My thanks to the SDO scientists, engineers, and web programmers that make this SDO Data Event possible.
Always use sun-safe optics to look at the Sun.
Enjoy!