Processes Shaping the Formation and Evolution of Planetary Systems

The sample of over 5,000 planets discovered beyond our solar system (exoplanets) provide us a first opportunity to constrain the dominant physics shaping how planetary systems form, and how they evolve over billions of years. Our team, the Planetary Origins and Evolution Lab, does this through a combination of theoretical and computational work, and will be focusing on two promising directions. We are looking for collaborative and enthusiastic team members interested in committing to this fall and spring for academic credit, with the possibility of continuing on as a paid researcher in the summer. 

Atmospheric Mass Loss: It has recently become clear that, early in their lives, many planets orbiting close to their host star lose a significant fraction of their mass through intense stellar irradiation. Our group aims to develop a) simple physical models for how such atmospheric loss gravitationally torques planetary orbits, and b) computationally investigate signatures of such effects in the observed exoplanet sample.

Collisional Sculpting of Planetary Systems: There is also strong evidence that planetary systems typically form with additional planets and subsequently destabilize, leading to violent collisions and orbital rearrangements. In this picture, the exoplanet population has been “dynamically sculpted” to leave only orbital configurations that are long-term stable.

Our recent work has elucidated the chaotic dynamics that leads to such collisions (why do such tiny gravitational tugs between planets make any difference at all?) through a combination of theoretical developments and machine learning techniques. We are now looking to apply those machine learning models, both to a) test this dynamical sculpting hypothesis against the observed exoplanet sample, and b) to provide independent constraints on planetary and orbital properties.

What Will We Do? We will create populations of planetary systems and computationally model their orbits with the open-source orbital mechanics REBOUND package, in conjunction with the analytical and machine learning stability classifiers in the SPOCK package (all in Python). We will then quantitatively search for the predicted orbital signatures in the observed exoplanet sample. We will also develop simple physical models for the effects of atmospheric mass loss on planetary orbits.

Essay Prompt: What interests you about this research and what do you hope to get out of this research experience? Which projects (or subtasks) are you most interested in pursuing? Why?

 

Name of research group, project, or lab
Planetary Origins and Evolution Lab
Why join this research group or lab?

You'll be working as part of a supportive and collaborative team to answer fundamental questions on the physical processes that dominantly shape planetary systems. You may have the opportunity to work with research groups beyond Harvey Mudd (atmospheric mass loss projects likely with collaborators at Princeton, dynamical sculpting likely with collaborators at Caltech). 

Representative publication
Logistics Information:
Project categories
Computer Science
Physics
Astronomy
Machine Learning
Numerical Modeling
Student ranks applicable
First-year
Sophomore
Junior
Senior
Student qualifications

This research is accessible for all academic levels and we definitely encourage first and second year students to apply! Ph24 and CS 5 (concurrently is fine), and a love for planets and stackoverflow are all that is required. Possibly useful courses for more advanced students include Ph111 and CS 144. We are most eager to find curious, collaborative team members who are passionate about answering big questions.

Time commitment
Fall - Part Time
Spring - Part Time
Compensation
Academic Credit
Number of openings
3
Techniques learned

Skills developed through the various projects above include: Computational modeling, Python programming, designing an effective numerical experiment, data mining and analysis, application of machine learning models (benefits and pitfalls!), physical intuition and order-of-magnitude calculations, interacting with the Linux terminal, communication and writing skills.

Contact Information:
Mentor
Daniel Tamayo
dtamayo@hmc.edu
Professor
Name of project director or principal investigator
Daniel Tamayo
Email address of project director or principal investigator
dtamayo@hmc.edu
3 sp. | 18 appl.
Hours per week
Fall - Part Time (+1)
Fall - Part TimeSpring - Part Time
Project categories
Machine Learning (+4)
Computer SciencePhysicsAstronomyMachine LearningNumerical Modeling