My interests lie in the modeling and characterization of extrasolar planets, from small, terrestrial planets like our own Earth to large, Jupiter-like planets orbiting close to their host stars!
Traditional methods of exoplanet characterization that only make use of reflected flux lack the ability to fully distinguish between different physical features of the target, such as cloud layers, hazes, or surface features. Polarimetry, however, is a powerful and more sensitive technique that has this ability, as it measures light as a vector (by the orientation of the electric field) rather than just a scalar intensity. It is therefore extremely sensitive to the composition and structure of the planetary atmosphere and surface, being affected by properties such as the mixing ratios of atmospheric gases, cloud optical thickness, cloud top pressure, cloud particle size, and surface albedo.
We can use spectropolarimetry - the measurement of the polarization of light across a spectrum - to assess the habitability of terrestrial planets, breaking degeneracies between habitable and non-habitable planetary scenarios.
Earthshine is sunlight that is scattered or reflected by the dayside of the Earth and then reflected back to the planet by the nightside of the Moon, where it can then be measured by ground-based facilities. These observations allow us to view the disk-integrated reflected signal of the Earth, thus presenting a view of the Earth-as-an-exoplanet!
Comparison between exoplanet-Earth models generated by the two different radiative transfer codes (solid blue and dashed gold lines) with the original earthshine observations (dotted black line) and the earthshine observations corrected for lunar depolarization (dashed-dotted red line). Also included in this plot are data points from two additional exoplanet-Earth models, but now including water clouds with either medium (purple squares) or large (green triangles) cloud droplets.
Primary areas of research: spectropolarimetric radiative transfer modeling, model intercomparison, and fitting data to observations
In this project we benchmarked two independent polarization-enabled radiative-transfer codes against each other and against unique linear spectropolarimetric observations of the earthshine. We found that while the results from the two codes generally agreed with each other, there was a phase dependency between the compared models. Additionally, with our current assumptions, the models from both codes underestimated the level of polarization of the earthshine. We also reported an interesting discrepancy between our models and the observed 1.27 µm O2 feature in the earthshine, and provided an analysis of potential methods for matching this feature. Our results suggested that only having access to the 11.27 µm O2 feature coupled with a lack of observations of the O2 A and B bands could result in a mischaracterization of an Earth-like atmosphere. Providing these assessments is vital to aid the community in the search for life beyond the solar system.
For more information, please check out my paper in The Astrophysical Journal at the following link:
Primary areas of research: spectropolarimetric radiative transfer modeling, early Earth analysis, and observational constraints
Research into determining the habitability of terrestrial exoplanets has been primarily focused on comparisons to modern-day Earth. In reality, Earth's atmosphere and surface have undergone significant evolution since its formation. Additionally, current characterization strategies of Earth-like exoplanets focus only on the unpolarized flux from these worlds. Better understanding the changes in the reflected light spectrum of the Earth throughout its evolution, as well as analyzing its polarization, will be crucial for analyzing its habitability and providing comparison templates to potentially habitable exoplanets. In this project we generated spectropolarimetric models of the reflected light from the Earth, as functions of both wavelength and planetary phase angle, at six epochs across all four of the planet's geologic eons. We found that the changing surface albedos and atmospheric gas concentrations across the different epochs allowed the habitable and non-habitable scenarios to be distinguished, and diagnostic features of clouds and hazes are more prominent in the polarized signals. We showed that common simplifications for exoplanet modeling, including Mie scattering and Lambertian reflection, affected the resulting planetary signals and led to non-physical features. Finally, our results suggested that pushing the HWO planet-to-star flux contrast limit down to 1 x 10^(-13) could allow for the characterization in both unpolarized and polarized light of an Earth-like planet at any stage in its history.
For more information, please check out my paper in The Astrophysical Journal at the following link:
Unpolarized (top row) and polarized (bottom row) contrast ratios for our six Earth Through Time epochs, with their atmospheres containing patchy clouds and hazes where applicable, scaled to their corresponding solar spectra through time. These contrast ratios assume the planets are in circular edge-on orbits and are observed at either quadrature (left column) or α = 40 degrees (right column). Our results suggest that an Earth-like planet at any point in its history, from a young and hot Hadean-like planet to a current Modern-like planet, could be characterized around a Sun-like star in unpolarized (polarized) light if the contrast ratio capability of future instruments could be pushed to a lower limit of 1 x 10^(-12) (1 x 10^(-13)).
Reflected linear polarization of Venus-as-an-exoplanet for five observational bands taken from 2021 - 2023 by the PICSARR instrument. Changes in the polarization of Venus across different sets of observations indicate changing atmospheric conditions on the planet!
An example Chi-squared analysis using a two-parameter gamma distribution and a refractive index for an 85% H2SO4, 15% H2O mixture for the cloud particles.
Primary areas of research: polarimetric photometry, cloud modeling, and fitting data to observations
When an outside observer would be looking at our solar system in their search for life, their prime targets would be Venus or the Earth. As we cannot yet observe an exo-Venus or exo-Earth to see what signals from these planets would look like, we instead use our own solar system planets as practice.
Using new polarized observations of Venus-as-an-exoplanet across different broadband wavelength bands taken by collaborators using the PICSARR instrument (Bailey et al. 2024, in prep), we can detect changes in the reflected polarization from the planet over time, thereby indicating seasonal changes. In an attempt to determine the cloud parameters that best fit these observations, we are performing a parameter scan of polarized radiative transfer models using different combinations of cloud particle size distributions, cloud particle refractive indices, and cloud top altitudes. Our results suggest that, depending on which part of the phase curve we are fitting our models to, we get different results for the best fit cloud parameters, again confirming our hypothesis of changing weather in the Venusian atmosphere.
This paper, led by Dr. Kimberly Bott, will be submitted in Spring 2026.
Primary areas of research: spectropolarimetric radiative transfer modeling and observational constraints
A look across planets in our solar system reveals that they all have one thing in common: heterogeneity. Yet most models of terrestrial exoplanets to date treat the planets as homogeneous objects, or if they include heterogeneity, it is constant in time. Ignoring variable features such as diurnal rotation or seasonal changes can lead to errors in the characterizations of these worlds.
In this project we run polarized radiative transfer models for multiple terrestrial exoplanet archetypes including simple ocean (or desert) planets with 1 to 4 randomly distributed continents (or water bodies); early and Modern Earth; and modern Venus. We incorporate different levels of heterogeneities in these models including diurnal rotation, varying surface configurations, and seasonal weather patterns such as atmospheric circulation above different surfaces. We find that the contributions from the different atmospheric and surface parameters result in asymmetric phase curves and variable spectra that allow us to begin to resolve degeneracies between habitable and non-habitable scenarios. Additionally, our results show that the polarization appears to be more sensitive than flux to heterogeneities such as patchy clouds/hazes and continents moving into and out-of-view.
For more information, please check out my paper in The Astrophysical Journal at the following link:
A cloud-free ocean planet with a single sandy supercontinent containing a microbial coast rotating in and out of view.
Diurnal rotation of the above planet observed at ~2.7 hour increments. The different surfaces rotating in and out of view clearly affect the resulting disk-integrated polarization, with larger variations occuring at longer wavelengths.
Current space missions including CHEOPS and JWST, as well as upcoming missions such as PLATO, will help diagnose cloud properties and global climate regimes on gas giant exoplanets with unprecedented detail and in 3D. Understanding the chemistry and temperature variations on these worlds requires constraints on the kinetic gas-phase chemistry (e.g., CH4) and photochemistry (e.g., SO2) as well as the planet’s interior temperature. Thus, a full 3D cloudy atmosphere model is needed with coherent observational constraints.
In our work, we use an iterative coupling between a 3D global circulation model (GCM), which produces 3D temperature and gas abundance profiles assuming chemical equilibrium, and a kinetic cloud formation model that takes into account nucleation, surface growth, gravitational settling, mixing, element conservation, and equilibrium gas-phase chemistry. We can then run these atmospheric and cloud results through the petitRADTRANS radiative transfer software to produce synthetic transmission spectra and phase curves for various gas giant exoplanets to analyze their 3D climates and cloud evolutions!
Stay tuned for more details and results from some of my current projects!