The Tierras Observatory: An Ultra-precise Time-series Photometer to Characterize Nearby Low-mass Stars and Their Terrestrial Exoplanets

Abstract

Although the study of exoplanets has seen dramatic advances in the past decade, the analogs of many of the denizens of our own Solar system remain beyond the grasp of our current observatories. Most of the known terrestrial exoplanets are significantly larger than the Earth, and it remains an open question whether Mars and Mercury-sized worlds are common. No exo-satellites or exo-rings have been discovered to-date. And, we've only begun to characterize the long term variability of stars and its impact on their attendant worlds. Time-series photometry is the path to significant progress on all these questions, but we are limited by the photometric precision of our observatories. In this thesis, I describe the design, construction, commissioning, on-sky performance, and first science results of the \textit{Tierras} Observatory, a refurbished 1.3-m ultra-precise fully-automated photometer designed to further our understanding of terrestrial exoplanets, exo-satellites, and the variability of mid-to-late M dwarf stars. The observatory re-utilizes the 28-year-old 2MASS North telescope, located at the Fred Lawrence Whipple Observatory atop Mount Hopkins, Arizona. \textit{Tierras} is designed to regularly achieve a photometric precision of 250 ppm on a time scale of both 10 minutes and across an observing season. I begin this thesis by detailing the five design choices that allow \textit{Tierras} to achieve 250 ppm precision per 10-min bins, and recounting their execution. \textit{First} is the optical and opto-mechanical design, alignment, bonding, and deployment of a four-lens focal reducer and field-flattener, which increases the field-of-view of the 1.3-m telescope from 0.19 degrees to 0.48 degrees on a side and allows us to accommodate an optimal number of comparison stars in the field and correct for atmospheric variation effects. \textit{Second} is a custom narrow bandpass filter (FWHM 40 nm) centered around 863.5 nm to minimize precipitable water vapor errors known to limit state-of-the-art ground-based photometers targeting M dwarf stars. \textit{Third} is a deep-depletion $4K \times 4K$ CCD with a quantum efficiency of 90% in the \textit{Tierras} bandpass, operating in shutter-less, frame-transfer mode to minimize readout time and readout noise, while maximizing our time gathering science data. The CCD is housed inside a custom dewar, and cooled to $-105^{\circ}$C via a cryo-cooler requiring a single annual refill. \textit{Fourth}, is a fully automated mode of operation, enabled by a fully refurbished telescope and dome that we retrofitted with absolute position encoders, a modern telescope control software, a new integrated (telescope and camera) control interface, and a fully operational observing robot. \textit{Fifth} is a custom set of baffles, designed and built in-house, which reduce sky background by two orders of magnitude. In the second part of this thesis, I summarize the \textit{Tierras} Observatory commissioning effort, present the results of our first year of science observations, and quantify our photometric precision as a function of bin size, and across the 2022-2023 observing season. I show that \textit{Tierras} is already delivering 250-ppm precision light curves (in 10 min bins) for M dwarf stars of varied spectral types, and outline current and future efforts to improve the instrument's long-term photometric stability further. I conclude with a description of the observational efforts we plan to undertake during the next 3-5 years. We will survey stars known to host a transiting planet to search for additional terrestrial planets too small or too cool to be detected by TESS or other ground-based facilities; monitor known exoplanets (both terrestrials and jovians) at longer orbital periods to search for satellites or systems of circumstellar rings; and undertake a long term monitoring campaign to determine hitherto unmeasured rotation periods of M dwarf stars in a volume-complete sample.