A Caltech Library Service

The Orbits of Young Extrasolar Planets as Formation Probes


Blunt, Sarah (2023) The Orbits of Young Extrasolar Planets as Formation Probes. Dissertation (Ph.D.), California Institute of Technology. doi:10.7907/2j2b-f649.


In this thesis, I have translated a few flavors of exoplanet timeseries measurements into constraints on exoplanet orbital parameters, and used these to make inferences about planet formation. I begin by introducing the two main observational techniques I used to perform these analyses: optical interferometry and stellar radial velocity monitoring. I then discuss some of the big open questions of exoplanet formation, particularly the mechanism for forming giant planets close to and far from their stars, where core accretion is thought to be too inefficient to form giant planet cores in time for them to accrete atmospheres.

High cadence radial velocity monitoring enables advances in our understanding of stellar activity, the fundamental stumbling block in the path to discovering and characterizing planets like the Earth. In my second thesis chapter, I present an argument that previously published RV-derived activity models of the PMS star V1298~Tau suffer from overfitting, casting doubt on published mass estimates of the young planets in the system which necessitated rapid contraction after formation, in tension with formation theory. I walk through several potential explanations for this overfitting, pointing out that the star has a strong differential rotation signal which is not included in published model fits, and encourage broader use of cross validation techniques in stellar activity model evaluation.

Optical interferometry, particularly using the VLTI/GRAVITY instrument, enables astrometry measurements that are orders of magnitude more precise than contemporary coronographic instruments, which translates to precise orbital parameters. In my third thesis chapter, I present and analyze two new VLTI/GRAVITY astrometric measurements of a young, widely separated planet and use them to make a preliminary argument that the planet's eccentricity is low or moderate. This sets an upper limit on the time (relative to disk dispersal) that the planet attained its current wide separation, and downweights the possibility of scattering after disk dispersal.

In the next two chapters, I showcase my contributions to two widely used open-source orbit-fitting software toolkits: orbitize! and radvel. I highlight two major new features of orbitize! that are available in the main code base as of the release of version 2: jointly fitting radial velocity measurements and jointly fitting absolute astrometry measurements. In the radvel chapter, I motivate and describe an updated Gaussian Process regression model for stellar activity modeling that reduces the potential for overfitting.

Item Type:Thesis (Dissertation (Ph.D.))
Subject Keywords:Exoplanets, planet formation
Degree Grantor:California Institute of Technology
Division:Physics, Mathematics and Astronomy
Major Option:Astrophysics
Thesis Availability:Public (worldwide access)
Research Advisor(s):
  • Howard, Andrew W.
Thesis Committee:
  • Knutson, Heather A. (chair)
  • Stevenson, David John
  • Ravi, Vikram
  • Mawet, Dimitri
  • Howard, Andrew W.
  • Wang, Jason
Defense Date:10 May 2023
Record Number:CaltechTHESIS:06032023-031445650
Persistent URL:
Blunt, Sarah0000-0002-3199-2888
Default Usage Policy:No commercial reproduction, distribution, display or performance rights in this work are provided.
ID Code:16076
Deposited By: Sarah Blunt
Deposited On:06 Jun 2023 15:07
Last Modified:13 Jun 2023 18:39

Thesis Files

[img] PDF - Final Version
See Usage Policy.


Repository Staff Only: item control page