Monday, December 30, 2013

Was Comet ISON Seen in AIA 4500?

A recent blog post at The Astronomy Stackexchange asked whether an AIA image taken during the perihelion passage of Comet ISON showed the comet. It does not show the comet. The portion of the image where the bright spot is visible has a complicated pattern that is only partially removed during image processing. I will only discuss the image, not the other points made in the blog post.

The AIA 4500 image described in the blog is shown on the left. An arrow points to a bright area that is claimed to be fragments of Comet ISON breaking up near perihelion. A large format version is available from our jpeg archive on the SDO website.

This bright area is an artifact resulting from stray light entering the telescope through a pinhole leak in one of the front filters. Our science data files are available from the SDO JSOC as fits (Flexible Image Transport System) files, the astronomy way to store image data. Each CCD image moves through a series of image processing to subtract the bias field, correct for dark current, subtract the flat field, remove proton hits, and other known effects. The final science data is the processed image. You need special software to work with fits files, so we convert the images to jpeg files to make them easier for a larger audience to use.

I downloaded the fits file that leads to the AIA 4500 image at 1800 UTC, opened it in fv (a free fits viewer), and made a screenshot. Here is that original image, with an inverted color table to enhance the pattern to the left and below the image of the Sun in the upper right. This pattern is a result of a pinhole light leak in a front filter and scattering inside the telescope. While the spacecraft is moving to the offpoint, the brightness of the pattern changes, and its structure is a little smeared as well. With the solar disk out of the center of the field of view, the pattern stands out more clearly than usual; however, a similar stray light pattern is visible in all AIA 4500 images. We have decided to stop serving the AIA 4500 Å images because of this pattern (see the blog post on December 17, 2013.)

AIA 4500 images were taken once an hour during the Comet ISON perihelion passage. It was only chance that this one was taken while we were moving. The AIA 4500 image an hour later, which is much closer to perihelion and a much better exposure, is shown in the second screenshot. The pattern is now much closer to the usual pattern and the processed image at the SDO website shows no sign of the comet, but there is a little brightness in the same area as the earlier image (1289 pixels over and 1379 pixels down). This shows the difficulty in removing the pattern, especially in regions with no direct solar input.

SDO Science Team members were very disappointed to not see Comet ISON as it flew by the Sun. We are convinced that SDO data does not reveal the presence of the cometary detritus.