The Transit of HAT-P-3b

The Transit of HAT-P-3b
By: Lucas Ahn
Date: Wed, Jul 27, 2022

As someone who is interested in astronomy but never had access to complex astronomical equipment, I was happy to have the opportunity to detect an exoplanet using the RASC Robotic Telescope situated in California. The goal of this project was to detect an exoplanet using the transit method, an exoplanet detection method used by astronomers that involves monitoring the host star’s brightness over a long period of time and comparing the star’s brightness to other stars. The star likely hosts an exoplanet if the star’s brightness dips in a consistent pattern. The project was done in 4 stages which included selection, imaging, processing, and analysis.

I first selected an exoplanet using Swarthmore College’s Exoplanet Transit Finder. I originally worked with Jenna Hinds to select two different exoplanets to study. I was given a new outreach coordinator named Samantha Jewett who told me that my original two selections didn’t work. After two more failed selections, they told me they were able to capture data for HAT-P-3b. The data included the original images along with bias, dark, and flat frames. The original data drifted, there were bright patches that made stars appear brighter than they should, and a meridian flip occurred later in the image sequence due to the telescope being flipped 180 degrees to continue tracking the star.

Unflipped Raw Data
Flipped Raw Data














After receiving the raw data, I downloaded AstrolmageJ and was ready for the processing phase of the project. Since I was inexperienced with the tool, I closely watched the video tutorials so I didn’t mess up. I first created a master bias frame using the data reduction facility in AstrolmageJ and then used it to remove the bias from the flat frames. I noticed that white lines appeared in my processed flat frames.

Processed Flat Frame


In need of clarification, I asked Samantha if this will affect my data and she said that it’s most likely hot and dead pixels and will be removed by the dark frames. I took a break for a few months to focus on school and when I revisited the project earlier this summer, I noticed AstrolmageJ froze when I imported data. My coordinator sent me different solutions to try out and in the end, I reinstalled AstrolmageJ. With the issue fixed, I created a master flat frame and a master dark frame and applied them to my original data to equalize the brightness in each frame so I could compare stars to one another.

Next, I fixed the meridian flip by identifying the frame the flip took place in, then placing the frame and all the other frames after it into one folder. I then placed the image sequence in AstrolmageJ, flipped the sequence 180 degrees, and saved the images as .fits into a new folder. I then imported the unflipped images and saved them as .fits so that every image file was the same file type. Lastly, I stabilize my images using the “align stack with WCS or apertures” button. There was only one aperture present, but after some time, I clicked a button with a red circle and two other apertures appeared, allowing me to stabilize my images. Once they were stabilized, I saved them into a new folder.





Processed Image

Lastly, I moved on to analysis. I performed multi-aperture photometry which is the act of tracking the host star’s light over the entire image sequence and comparing the host star’s brightness to other stars of similar size. Some of the stars had abnormalities in their brightness, and thus I had to restart the process a few times to choose the right stars. Once the right stars were picked, I plotted each star. I noticed that my target star had a significant dip in its brightness which indicated an exoplanet. I wanted more data, so I added a line of best fit. The line of best fit gave new information including the planet’s size (0.97 Jupiter radii), its transit duration (1hr 42mins 34sec), and its orbital inclination (87.9 degrees). Its transit duration is a bit off compared to the data found in the Swarthmore College’s Exoplanet Transit Finder which is around 2hrs 5mins based on the ingress and egress times, though its calculated size is comparable to the data found by NASA which places the planet at around 0.94 Jupiter radii. Other notable information included the star’s density which is around 4.96036g/cm^3.







Data Set













The Robotic Telescope Project was one of the most interesting projects I’ve done outside school. After completing this project, I learned more about exoplanet detection, developed problem-solving skills, and did interesting scientific research that will benefit my university application. Though at times the project was a bit frustrating and confusing, the amount of aid given, whether that be from outreach coordinators or video tutorials, making it a truly memorable and fun experience. And though the data I found was a bit off compared to the data found by NASA and other professional astronomers, I’m very grateful to have participated in this project, and I highly recommend this project to those who’re interested in astronomy either as a hobby or as a possible career path.