There has been mounting evidence of a ninth planet far out in the solar system, weighing around 5 to 20 Earth masses (around the size of Neptune). This theoretical planet is estimated to be 200-600 times farther from the Sun than the earth (even beyond Neptune or Pluto). Astronomers classify objects like this as Extreme Trans-Neptunian Objects (ETNOs).
The evidence for its existence comes from the strange clustering of other (much smaller) discovered ETNOs, which is highly unlikely to occur naturally. In fact, it is estimated that there is only a 0.007% probability of this clustering due to chance. The existence of a ninth planet with the above parameters that is anti-aligned to the ETNO clustering is the most plausible explanation for this.
Through PURM, I had the opportunity to work in Dr. Masao Sako’s lab in the Physics and Astronomy Department. My project was to analyze data from the Dark Energy Survey (DES), which has been taking images of the night sky for 5 years using the 570-megapixel Dark Energy Camera (DECam) mounted on the Blanco Telescope in Chile. Although intended to study dark energy, the data contains large numbers of moving objects, including TNOs. Thus, we can use DES for our own intents and purposes. We passed the data from DES into a “friends-of-friends” algorithm developed by Pedro Bernardinelli (a graduate student in Dr. Sako’s lab) that calls objects within one arcsecond of each other as “friends” and groups all friends of friends together to form chains. Having obtained chains of detections, we sorted them by change in time and angle to obtain chains that were suited for further analysis. This removed junk data such as groups of supernovae detections that vary in brightness but never move.
After obtaining the groups of detections, we ran them through an orbit fitter written by Prof. Gary Bernstein (Dept. of Physics and Astronomy). It takes in the position and MJD (Modified Julian Date) of each detection in any group and gives the orbital parameters (with errors) of the predicted TNO. After sifting these groups further, the groups with similar orbital parameters (meaning they are detections of the same TNO) were merged (provided they have low chi-square error on doing so). This resulted in a final data set that has all detections for a particular TNO in the same group.
At this stage, we went back to the roots and pulled up the DES images of the detections for each TNO and analyzed them to see if they were really moving or just artefacts of data processing. Once we verified that the data corresponds to real moving objects in the solar system, the orbital parameters of these newly found TNOs were then compared with the predicted values for planet nine.
This project taught me some valuable research skills such as scientific collaboration (I was communicating with grad students two floors above me and with Prof. David Gerdes from the University of Michigan) and result presentation (in the weekly meetings). In addition, I gained a lot of programming experience in Python and shell script, which will be very useful to me as I intend to be a physics as well as a computer engineering major. The most important thing that I learned was the overall process of scientific research and data analysis and I know I will continue to benefit from and build upon this knowledge in my career.
I am very excited to continue working with Dr. Sako and further pursue research in my time at Penn.