The Hubble Space Telescope recently observed an asteroid in the process of falling apart. Known as 6478 Gault, located in the main asteroid belt of our solar system between Mars and Jupiter, and roughly 2.5 miles wide, this asteroid is rotating so fast—about once every 2 hours—that it is literally flinging itself apart. The observations revealed two distinct tails of debris streaming from the asteroid, which consequently very much resembles a comet. As a close reading of the title of this post may indicate, our image processing of the data led to two representations of the image—hence a tale of two images of a two-tailed asteroid…see what I did there? (For more specifics on the nature of asteroid Gault, be sure to visit the Hubble press release.)
When HST observes a moving Solar System target, the observatory must track the target throughout the observation. This means that any background objects, like stars that otherwise appear stationary, will be smeared out during the course of an exposure—think of the motion blur you would see in a snapshot focused on a race car speeding down a track.
Additionally, the telescope is traveling through its orbit as exposures are taken, thus adding another element of motion to the data. My goal in processing this image was to obtain the cleanest possible view of the asteroid itself, taking advantage of the fact that there were several individual exposures taken. We can add data together to increase the quality of the desired signal and reduce the amount of background noise, including the star trails, in the final image. The end result of this process produces a pristine view of the asteroid (seen above).
Our colleagues at the European Space Agency took a different approach to this image, processing only a single exposure of the asteroid and keeping the star trails in the image, which presents us with an interesting learning opportunity.
This approach leads to a slightly noisier image, but also lends a sense of motion to the asteroid. Reproduced below are the data in a more raw state in black and white. In this image, you can clearly see the asteroid and star trails, but you can also see instrumental artifacts such as the gap between two detectors and a sort of “snow” that peppers the entire image. The snow is caused by cosmic rays (high energy particles, fragments of atoms racing through space in all directions) penetrating through the telescope and into the detectors as observations are made. They leave a tell-tale imprint in the data in the form of extremely bright pixels and streaks of pixels in all directions. One of the reasons that we take multiple exposures is to be able to more effectively remove this noise from the data. It can be removed from a single exposure, as was obviously the case with the ESA image, but some sacrifice is made in the image quality as a result.
As an interesting aside to this discussion, sometimes we can take advantage of the fact that we have multiple exposures to string the images together and create a movie of the observations. In the case of asteroid Gault, there was some speculation that we might be able to see changes in the debris tails over short time frames, so I prepared just such a movie. The short animation below combines five 6.5-minute exposures from Hubble taken on February 5, 2019 between 15:08:42 and 15:51:06 UT. Although the movie does not reveal any changes in the asteroid, it is nonetheless mesmerizing to watch. In just that short amount of time we see a great deal of motion. You can clearly see the streaks of background stars serving as a backdrop to the asteroid. We also see Hubble’s orbital movement reflected in the frames: Hubble moves at about 5 miles per second relative to the ground, and in the span of these observations moved about halfway around Earth!
The case of asteroid Gault highlights the fact that there are subjective, editorial decisions made in the process of preparing these images for press releases. The data are not manipulated in any way, only processed to calibrate the image and remove artifacts. Both representations ultimately show the data as they were collected from the telescope. Choices are made regarding how to approach the processing of the raw data to best convey the science story for each release. In this case, choices made early on in processing the data led to two very different and uniquely informative images of the same object.