Meet the Researcher - Zoe de Beurs

Breakout Room: 21

ZoedeBeurs_HeadshotResearcher Name: Zoe de Beurs
Title of Research: Detecting Exoplanets Using Artificial Intelligence and Radial Velocity Measurements
Division Representing: Physics and Astronomy
Institution: University of Texas at Austin
Institution Location: Texas
Home State: Texas
District Number: 25
Advisor/Mentor: Andrew Vanderburg
Funding Source: Texas Institute for Discovery Education in Science (TIDES) Advanced Research Fellowship
Scholarships: 2020 Goldwater Scholar ($7,500) - Barry M. Goldwater scholarship and Excellence in Education Foundation; 2020 Distinguished Undergraduate Scholarship ($5,000) - Universities Space Research Association; 2020 NASA Texas Space Grant Consortium Scholar ($2,000) - NASA Space Grant Consortium 

Research Experience:  
As a senior at UT Austin triple majoring in physics, mathematics, and astronomy, I've conducted research at several high-caliber institutions. In 2017, I conducted quantum optics experiments at UT. I took these experimental skills to my 2018 Univ. of Chicago NSF internship, where I developed thermal models for our ultra-high energy particle detector. As a 2019 NSF intern at the Harvard & Smithsonian Center for Astrophysics, I created new methods for classifying X-ray binaries, leading to a first-author publication submitted to the Astrophysical Journal. Most importantly, I now work with Prof. Andrew Vanderburg using artificial intelligence (AI) to enhance the radial velocity (RV) exoplanet discovery method. Currently, noise from the host star hides the tiny RV signals of Earth-twins. We used AI to reduce this noise by half, leading to first-author publication submitted to the Astronomical Journal. Similar techniques could help detect Earth-twins and answer fundamental questions about our place in the Universe.  I plan to continue sharing these exciting discoveries through outreach and to make science more accessible - especially towards historically underrepresented groups. At UT, I advocated for inclusion in academia as President of the Natural Sciences Council. I've led teams that started a salary negotiation workshop focused on women in science, directed more than a dozen events promoting equity and inclusion, and advocated for reducing barriers to research. This is a critical time for our field to become more inclusive, and my experiences uniquely equip me to play a leading role in increasing access and equity in STEM.

Presentation Experience: 
I have shared my research through over thirty talks, posters, and outreach events for both academic and non-academic audiences. In academic settings, I have won seven awards for Best Talk/Poster and been invited to present at multiple high-profile venues, including the Harvard Exoplanet Seminar, the California-Harvard Astrostatistics Collaboration, and the Dean's Scholars Distinguished Lecture Series for an audience which included Nobel Laureate Rainer Weiss. I have also shared my own journey in science with general audiences in a variety of formats, including my video on high-energy particles from outer space, which led to a 1st Place Award at the UT Research Showdown [1]. Recently, I also gave an outreach talk at Astronomy on Tap ATX about how artificial intelligence can help us understand the Universe [2]. Speaking to both children and adults with a curiosity for science, regardless of their background, is something I aspire to continue doing throughout my career. This excitement for sharing science led to being chosen as one of five student representatives for the Academy of Undergraduate Researchers Across Texas (AURA), which allowed me to represent UT Austin in speaking to Texas State Officials. In particular, I had the opportunity to present my astrophysics research at Undergraduate Research Day at the Texas Capitol and visited several state legislators' offices to speak about the importance of undergraduate research. Through Posters on the Hill 2021, I hope to take this one step further and share the importance of science at the federal level. [1][2]

Significance of Research:       
The study of planets outside our solar system (exoplanets) - in particular those that are similar in mass and temperature to the Earth - is one of the greatest quests in astrophysics and has the ultimate goal of finding traces of life elsewhere in the Universe. Detecting the signals of Earth-twins requires detecting the motion of a star dozens of light years away that is moving slower than most turtles walk. To detect these signals, we use the radial velocity (RV) exoplanet discovery technique, which detects the gravitational pull of a planet on its host star and helps us measure a planet's mass and distance from the star. However, the RV method is limited by stellar activity noise introduced by the host star, which can mimic or hide planetary signals. In this work, we show that artificial intelligence can effectively remove stellar activity noise from RVs. We demonstrate our technique on simulated data, reducing the RV scatter from 82.0 cm/s to 3.1 cm/s, and on real solar observations from the HARPS-N Solar Telescope, reducing the RV scatter from 1.47 m/s to 0.78 m/s (a factor of ~ 1.9 improvement). In the future, similar techniques could remove activity signals from observations of stars outside our solar system and eventually help detect habitable-zone Earth-mass exoplanets around Sun-like stars. Discovering Earth-twins addresses fundamental questions about potential sources of extraterrestrial life and about our own place within the solar system, the galaxy, and the Universe.

Uniqueness of Research: 
Discovering Earth-like exoplanets addresses fundamental questions about our own place within the galaxy and potential for sources of life beyond the solar system. However, revealing the tiny signals of these exoplanets requires removing dominant noise sources. In this research, we use artificial intelligence to characterize and predict these noise signals, paving the way towards detecting an Earth 2.0.