From the Sun to the Stars: A Solar Calibrator for the Keck Planet Finder and New Frontiers in Exoplanet Obliquities

Citation

Rubenzahl, Ryan Asa (2024) From the Sun to the Stars: A Solar Calibrator for the Keck Planet Finder and New Frontiers in Exoplanet Obliquities. Dissertation (Ph.D.), California Institute of Technology. doi:10.7907/sgbv-5841. https://resolver.caltech.edu/CaltechTHESIS:06022024-013442559

Abstract

The galactic census is underway. In the thirty years since the discovery of 51-Pegasi b, the first extrasolar planet discovered orbiting a main sequence star, over 5,600 more have been tallied. The known exoplanet population is diverse, yet no extrasolar system observed to date resembles our own. The radial velocity (RV) technique, which works by measuring the reflex motion of a star from a perturbing planet, remains the most capable method for discovering exo-Earths. An exo-Earth would accelerate its star up to 9 cm/s, Doppler-shifting stellar absorption lines across the detector of a modern spectrograph by 1/10,000th the width of a typical CCD pixel. At this level of precision, every component of the instrument becomes critical to the overall stability. Yet, despite many instruments reaching <30 cm/s precision (such as the Keck Planet Finder; KPF), exoplanet discovery has stalled around the 1 m/s level. The primary limitation is now correlated noise introduced by physical processes on the stellar surface, dubbed "stellar activity," which manifests RV variability up to many m/s on timescales from minutes to decades.

This thesis has two primary themes. The first is concerned with addressing the stellar activity problem and improving RV instrument performance. Both are well-probed using "Sun-as-a-star" observations, as the Sun is the only star in the universe with all orbiting planets accounted for and its surface resolved at all timescales, wavelengths, and spatial scales. Chapter 3 presents the Solar Calibrator (SoCal), an autonomous system that feeds stable, disc-integrated sunlight to KPF at the W. M. Keck Observatory. With SoCal, KPF acquires 200–800 daily high-resolution (R = 98,000) optical (445-–870 nm) solar spectra up to a signal-to-noise of 2400, providing a rich and unmatched dataset for developing novel methods for mitigating stellar activity. We also leveraged SoCal to discover, diagnose, and fix a detector issue in KPF, and to develop and optimize the data reduction pipeline. We compared SoCal RVs to solar RVs from the NEID solar feed and found excellent agreement on intra-day timescales at the single-measurement photon-noise level (30—40 cm/s).

The second theme of this thesis is the precise characterization of extrasolar planets in extremely close-in orbits. These most extreme exoplanets often constrain planet formation theories the most. Chapter 2 presents the discovery and characterization of TOI-1347 b, the most massive rocky ultra-short-period exoplanet discovered to date. We found tentative evidence for a high mean-molecular-weight atmosphere on the planet, which orbits its star in just 20 hours. An atmosphere on such a highly irradiated world would be unusual, but not impossible, though JWST follow-up measurements are needed to confirm. Chapters 4, 5, 6, and 7 probe the mysterious formation pathways of hot Jupiters from four unique angles. Archeological clues to their dynamical histories remain in their present-day stellar obliquity, the angle between the star’s rotation axis and the planet’s orbital plane. The first is WASP-107 b, a super-Neptune that must have migrated to explain its ultra-low density and escaping atmosphere. We measured WASP-107 b to be on a polar orbit, an indicator of a history of dynamics with its outer planetary companion WASP-107 c. The second is KELT-18 b, an ultra-hot Jupiter we also found to be on a polar orbit. The mutually misaligned stellar companion in the system may be to blame. The third, Kepler-1656 b, is a highly eccentric sub-Saturn that could plausibly be undergoing migration kick-started by its outer planetary companion. Its orbit may be aligned, atypical of the traditional picture of high-eccentricity migration. The fourth, Kepler-1658 b, is actively experiencing tidal orbital decay around an evolved star, two aspects that strongly constrain orbital realignment timescales. The system either retains its primordial configuration or constrains tidal efficiencies. Unfortunately, our transit observations are contaminated by a massive starspot, which preclude the direct measurement of the obliquity. This chapter instead explores new methods for directly modeling starspots in EPRV spectra.

Thesis Files