Galen Bergsten is a graduate student at the University of Arizona's Lunar and Planetary Laboratory. They work with Prof. Ilaria Pascucci to study exoplanet demographics, with a focus on the occurrence rate of Earth-sized habitable zone planets.


I study populations of exoplanets to learn more about how they form and evolve, and to aid our search for life by understanding the frequency of Earth-like planets.

Occurrence of Earth Analogues

A comparison plot of recent estimates for the occurrence of Earth-sized habitable zone planets around M dwarfs and FGK stars. The left side of the figure compares results of recent studies for M dwarfs. Our new estimate is more uncertain than these previous works, sharing a 1-sigma upper bound of roughly 40% but now with a lower bound of roughly 1% (instead of previous bounds around 20%), and a median value of 14%. Text at the bottom reads that, for M dwarfs, this occurrence rate is less observationally constrained than previously believed. The right side of the figure includes comparisons to recent studies of Sun-like stars. While there is a range of plotted Sun-like estimates spanning from 5% to 85%, all of these values overlap with the new M dwarf estimate within 1 sigma. Text at the bottom restates that Kepler offers no evidence for higher Earth-sized habitable zone planet occurrence rates around M versus FGK stars. A plot showing various literature estimates for Gamma_Earth, the occurrence of Earth-sized habitable zone planets (normalized by the log-period and log-radius dimensions of the habitable zone integral). Three colored bands highlight different types of works: works considering atmospheric loss effects or an intrinsic population of rocky planets; works considering the statistical treatment of reliability, and works (so far, only Bergsten et al. 2022) considering both at once. Some earlier papers consider none of these. A vertical line indicates the occurrence rate used for the decadal mission concepts at ~33%, while our newest result is closer to ~15% for the full FGK sample. Individual results for different mass bins are also included but offer no clear trend with stellar mass.

Earth-sized habitable zone planets are hard to find, which makes it difficult to estimate how common planets like our own might be. We can predict the frequency of Earth-like planets using population models fit to Kepler data, incorporating per-star sensitivity metrics, per-candidate reliability, and considerations of physical processes like atmospheric evolution. For Sun-like stars, we estimate an occurrence rate of \(\eta_\oplus = 9^{+3}_{-2}\%\), or one in ten stars like our Sun hosting a planet like our own.

A comparison plot of occurrence estimates calculated in instellation using different methodologies from Bergsten et al. (2023), including original values from Dressing and Charbonneau (2015). All of the updated estimates have larger uncertainties, with the lowest estimates accounting for both reliability and parameter uncertainties.

The picture is less clear for Earth analogues orbiting M dwarfs (or stars between one-half and one-tenth of a Solar mass). Kepler did not originally target many M dwarfs, and Gaia-informed revisions of stellar properties reduced the current sample even further. As a result, our updated occurrence estimate of \(\eta_\oplus = 9^{+18}_{-8}\%\) (or \(14^{+25}_{-13}\%\) in an optimistic case) for M dwarfs is highly uncertain. Thus, the Kepler sample does not offer evidence that M dwarfs should have more Earth-sized planets in their habitable zones (compared to Sun-like stars). Other surveys like TESS and K2 are well-suited to provide additional insight on this topic!

Atmospheric Evolution of Small Planets

Top: A cartoon population of exoplanets arranged in a grid of orbital period and planet radius, each with some atmospheric envelope drawn around an internal core. There are two rows of super-Earths surrounded by a box indicating super-Earth-sized objects, and three rows of sub-Neptunes. Bottom: Similar to the top panel, but now some of the shortest-period planets have lost or shrunken atmospheres. The super-Earth box now includes some shurnken sub-Neptunes at the shortest periods, while the planets farther out retain their atmospheres and are larger. An arrow labeled 'time' connects the two panels, indicating the bottom panel is an older population.

Young small planets close to their stars can lose their atmospheres and shrink over time. The planets we observed with Kepler are mostly old and have already undergone this atmospheric evolution, making it difficult to infer the size distribution of planets at the time of their formation. But these loss mechanisms are inefficient for planets farther away from their stars, meaning the long-period planets can tell us about the primordial populations.

A plot showing data and modeled results for the fraction of small planets contributed by super-Earths as a function of orbital period. Values at 100 days are around 40%, with a label suggesting these super-Earths are all intrinsically rocky. Values at 2 days are around 80%; a horizontal line connecting the short and long periods shows the difference of 40%, implying that 40% of small planets are former sub-Neptunes at the shortest periods. The curve spanning from 2 to 100 days follows a decreasing log curve which appears to flatten out on either side.

We find that 40% of small planets are likely intrinsically rocky super-Earths at the time of formation. At short orbital periods, atmospheric loss processes add an extra 40% of super-Earth sized objects from the remnant cores of former sub-Neptunes. The latter is stellar mass dependent, with heavier (brighter) stars stripping more of their planets out to farther separations.

About Me

I joined the University of Arizona as a PhD student in August 2020, and recently passed my candidacy exams in May 2023. From 2016-2020, I was an undergraduate studying physics/astronomy and biology at the University of Utah, where I worked with Dr. Gail Zasowski to study stellar populations and the interstellar medium. I have general interests in quantitative astrobiology, the search for planets/systems like our own, and the accessibility of astronomy (both in academia and at large).

Outside of academia, I love cooking, spending time at botanical gardens, reading manga and playing indie games. For more on what I get up to outside of work, check out my Side Gigs page.