Impact Cratering in the Solar System
Asteroid and comet impacts can reveal important geological aspects of the planets and moons they occur on. Recently, I estimated the amount of material that would be ejected by impacts on Venus if it once hosted a thin atmosphere, and the prospects of recovering these rocks as meteorites on the Moon (Cabot et al. 2020). We compared secular trajectories to N-body simulations of meteorites originating from Venus and determined petrological considerations (i.e. oxygen isotope ratios and metal abundance ratios) that would assist with identification.
N-body trajectories of meteorites ejected from Venus from a hypervelocity impact.
In Levine et al. (2021), I helped assess the production of nitrogen ice fragments from collisions analogous to Kuiper Belt Objects (KBOs) striking Pluto. Desch & Jackson (2021) proposed such collisions as responsible for forming 'Oumuamua and similar Interstellar Objects (ISOs). We found that impacts are generally of insufficient speed to eject lightly-shocked nitrogen fragments. The ejecta produced by Grady-Kipp fragmentation are also considerably smaller than current estimates for 'Oumuamua's diameter.
Simulated KBO Impact on Pluto (left) using iSALE 2D (e.g. Collins 2014) against predictions from shock wave-interference theory (Melosh 1984). Figure from Levine et al. (2021).
of Exoplanet Atmospheres
I have made novel detections of HCN in the atmospheres of HD 189733b and HD 209458b (Cabot et al. 2019, Hawker et al. 2018) using the CRIRES infrared spectrograph. HCN forms in high temperature gas with C/O > 1, making it a tracer of hot Jupiter formation conditions (Madhusudhan 2012). Specifically, the presence of HCN favors planetary migration from past the water and carbon dioxide snowlines, compared to in situ formation.
Cross-correlation functions (CCFs) showing detections of four atomic/ionic species in WASP-121b. Figure from Ben-Yami et al. (2021).
The optical regime probes species connected to the thermal structure in hot Jupiter atmospheres. Metals, such as iron, titanium, and vanadium, as well oxides and hydrides containing them, are considered to be potential species responsible for thermal inversions. I detected atmospheric absorption from iron in archival HARPS observations of WASP-121b (Cabot et al. 2020). After generating additional model templates, cross-correlation revealed additional species Fe II, V I and Cr I --- the latter two representing the first detections in an exoplanet (Ben-Yami et al. 2020).
I also led the cross-correlation analysis that made novel detections in the atmosphere of MASCARA-2b using EXPRES (Hoeijmakers et al. 2019). This represented the first use of EXPRES for atmospheric characterization. I used EXPRES again to confirm a new exoplanet TOI-1518b via Doppler Tomography, and detect neutral iron in its atmosphere (Cabot et al. 2021). I have also investigated the robustness of telluric correction in the infrared (Cabot et al. 2019) and optical (Langeveld et al. 2021) regimes.
Radial Velocity Planet Search with EXPRES
As part of the Fischer exoplanet group at Yale, I am closely involved with the first science results from the Extreme Precision Spectrometer (EXPRES), a high-resolution optical spectrograph installed at Lowell Discovery Telescope (AZ). I implemented a Gaussian Process regression framework for modeling stellar activity variations in a case study of HD 101501 (Cabot et al. 2021). Significant RV variations were linked to surface features on the star. I quantified the necessary observing cadence to detect low-mass exoplanets, which motivated a revised EXPRES observing schedule.
The surface of Epsilon Eridani at various stages of a full rotation. The spots visible here were responsible for the bulk of RV variations reported in Roettenbacher et al. (2021)
In Roettenbacher et al. (2021), specific features on the surface of Epsilon Eridani (obtained from lightcurve inversion and interferometry) were linked to RV variations in a contemporaneous EXPRES dataset. I used a stellar disk model in conjunction with the surface images to model the expected RV signature of the spots on the rotating surface. The model predictions reduced the RV scatter by approximately 4 m/s.