Overview

I'm an observational astronomer studying how galaxies form and evolve over time. From our nearest neighbors in the Local Group to distant galaxies billions of light-years away, I am fortunate to have gotten the chance to observe with and use data from the best telescopes around and above our globe, including the Subaru Telescope, Hubble Space Telescope, and Magellan Telescopes. Take a look at the projects that I've worked on!

Graduate Work

My PhD research focuses on learning about the evolution of our nearest neighboring galaxies through the lens of their past and present satellite populations. Our local neighborhood contains many galaxies that are similar to the Milky Way in mass and shape but with very different disk structures, star formation histories, gas contents, and other structural and morphological properties. Cosmological simulations have shown that one of the reasons for this diversity may be past collisions and interactions with other galaxies, which grow galaxies over time, can alter their star and gas content and drive the delivery of smaller satellite galaxies and star clusters.

The biggest challenge in studying such interactions is that they occur on gigayear timescales, and so we cannot directly observe them in action. But we can study the aftermath of these interactions by looking at the stellar halo of a galaxy, which is a faint and diffuse envelope of stars, clusters, and dwarf galaxies around a galaxy that are accreted from past mergers. The halo acts as a fossil record of a galaxy's past interactions and can provide valuable insights into its merger history and satellite populations. While the stellar halos of galaxies are treasure troves of information, digging around them for these clues is difficult due to their extremely low surface brightness. The only way I can detect them is by resolving individual stars in these halos, which is only possible in nearby galaxies. My work uses photometric and spectroscopic techniques to probe these stellar halo populations and measure the scale and kinematics of past mergers and also characterize the present-day satellite populations of nearby galaxies.

The main questions that drive my thesis work are:
Stars icon How can we learn about the properties and impacts of bygone satellites of MW-mass galaxies?
Galaxy icon What are the properties of present-day satellites of MW-mass galaxies and how can they inform theories of galaxy formation and evolution?

A diagram illustrating the stellar halo of a galaxy and highlighting the research the author has done with it.

Galaxy iconSaying "hallo" to M94's stellar halo Camera icon

Three-panel image showing resolved stars in M94's stellar halo. Left panel shows RGB stars in the blue RGB branch and are colored mostly blue, while the middle panel shows RGB stars in the red RGB branch and are colored mostly red. Right panel is a color composite density plot of the stellar halo.
The resolved halo of M94. The two figures on the left show the metallicity distribution of stars in the blue and red RGB branches, respectively, showing that the metal-poor blue RGB stars form M94's stellar halo population. The rightmost figure is a faux color smoothed density plot where each bin is color-coded by stars in three different metallicity bins.
My first project in grad school focused on characterizing the merger history of the nearby (4.2 Mpc) spiral galaxy M94. This galaxy is unique because it hosts the largest pseudobulge in the Local Universe. Pseudobulges are dense, flat, inner regions of a galaxy that have retained some of their disk-like properties, including more rotational motion than random motion you'd expect to be associated with a classical bulge. Central structures such as bulges and pseudobulges are thought to be built up by some combination of secular processes and mergers, with pseudobulges more commonly associated with secular processes that happen over a long time period (gas infall, tidal and disk instabilities, bar formation, etc.), whereas classical bulges are more commonly associated with hierarchical evolution (mergers, clustering, ram pressure stripping, etc.). The goal of this project was to characterize the scale of M94's dominant merger and see which of these processes was more important in building up its giant pseudobulge.

A scatter plot of the bulge mass vs accreted halo mass for nearby Local Volume galaxies, color-coded and symbolized by whether the galaxy hosts a bulge or a pseudobulge. There is no evident correlation between the two properties.
A log-log plot of the bulge mass vs accreted halo mass for nearby Local Volume galaxies, color-coded and sybmolized by whether the galaxy hosts a bulge or a pseudobulge Galaxies with massive bulge do not necessarily have massive accreted halos, and vice versa. M94 is shown in magenta.
To do this, I used data from the Subaru Telescope's Hyper Suprime-Cam (HSC) to resolve individual stars in M94's stellar halo and identify its red giant branch (RGB) stars, which are the tracers of old, likely accreted stellar populations. I found that M94 hosts two distinct RGB populations: a redder, concentrated metal-rich population whose surface brightness profile shows a unbroken exponential profile that is a clear continuation of the outer disk, and a bluer, more diffuse metal-poor population associated with the halo. Integrating the halo profile, I found that M94 hosts one of the least massive and most metal-poor stellar halos among Local Volume galaxies. Its dominant merger partner was likely around or less than the mass of the SMC, indicating M94 likely had a low-mass accretion history and that its giant pseudobulge was likely built up by secular processes rather than mergers.

When comparing with other Local Volume galaxies, it becomes apparent that bulge mass and accreted halo mass are not correlated — galaxies such as M94, M81, and M31 all have similar large bulges/pseudobulges, but have very different accreted halo masses. And while M94 and M101 have similarly quiet merger histories, their bulges differ by almost an order of magnitude in mass.

Galaxy iconDiverse dwarfs of M81 Camera icon

Scatter plot of the radius and luminosity of local volume dwarf galaxies, color-coded by their host galaxy.
The V-band absolute magnitude vs azimuthally-averaged half-light radius of Local Volume dwarf galaxies and globular clusters. The four M81 dwarfs covered in my work are black circles with errorbars. Though these are the faintest M81 dwarfs yet found, they span almost an order of magnitude in size.
Stellar halos are also home to faint satellite dwarf galaxies, who can be delivered during mergers. While they might not seem as interesting as their giant and beautiful Milky Way-mass hosts, these ultra-faint dwarf galaxies (UFDs) are some of the smallest, oldest, most metal-poor, least chemically enriched, and most dark matter-dominated galaxies in the universe, making them great probes of small scale cosmology, dark matter, and galaxy formation models. This also makes them incredibly difficult to find, especially outside of the Local Group where our dwarf satellite census is mostly complete but might not be representative of all satellite populations around Milky Way-mass galaxies.

Harnessing the power of Hubble's Snapshot imaging program, I characterized four faint dwarfs in the M81 (D=3.6 Mpc) group. Most faint dwarfs don't have much diffuse light, so resolved stellar populations of RGB stars are again needed to identify and study them. Using HST/ACS data, I fit for the structural parameters, such as the centroid, ellipticity, radius, and position angle, of these dwarfs using a maximum likelihood MCMC approach and also measured their density profiles, photometric metallicities, and luminosities.

My analysis found that although these systems are old (~13 Gyr), faint (Mv ≥ -7.9), metal-poor ([M/H] < -1.5), and show no signs of recent star formation, they are quite diverse in their properties. One of them, D1006+69, is one of the most centrally-concentrated galaxies with a Sérsic index of n~5. Most UFDs are fit with a Sérsic index of n ~ 1 (e.g. an exponential profile), making D1006+69 an outlier. Another, D1009+68, is quite elliptical with an ϵ ~ 0.6, while DWJ0954+6821 is the particular compact for its magnitude and the smallest M81 satellite. They are all the faintest M81 dwarfs to date, with two of them being some of the lowest surface brightness galaxies for their magnitude. All four of these systems were previously discovered by detecting overdensities of stellar sources in ground-based Subaru Hyper Suprime-Cam data, and are now confirmed to be dwarf galaxies. My analysis showed a difference between some of the recovered structural parameters when compared to the discovery data, highlighting the important synergy between ground-based discovery and space-based follow-up.

Most intriguingly, despite the high ellipticity of D1009+68 and the extended nature of D1006+69, none of these four dwarfs show any signatures of tidal disruption or stripping, highlighting that ellipticity is not necessarily set by tides. D1006+69 in particular is intriguing because other studies have found UFDs with similarly extended natures or stars at large radii, which opens up the possibility that these dwarfs host their own accreted stellar halos. Understanding how to best fit faint dwarfs and whether we need to start expanding to multi-component models to better encompass these extended structures is an important question for future studies of these systems.

Galaxy iconMeasuring the halo kinematics of NGC 253 Spectrum icon

While resolved star photometry is crucial for understanding the properties of a galaxy's main merger partner, spectroscopy of these stars is necessary to turn a static snapshot into a dynamic view of the satellite infall. Measuring the kinematics of a galaxy's stellar halo can provide insights into the angular momentum of an accreted satellite, its orbital parameters, the time of merger, and the possibility of detecting a rotational signautre in the halo. A measurement of stellar halo rotation has only been done for three galaxies to date, the Milky Way, Andromeda, and NGC 4945, and my current work aims to add NGC 253 to this list.

NGC 253 is a nearby (D=3.5 Mpc) starburst spiral galaxy that is the brightest member of the Sculptor group. It also has a faint shell/shelf structure in its halo, which is likely from a recent merger. I used the Magellan/M2FS multi-object fiber-fed spectrograph (pictured right) to obtain spectra of ~200 asymptotic giant branch (AGB) stars in the stellar halo of NGC 253, centered on its shell structure, over the course of four good nights split between 2023 and 2024. The goal of this project is to measure the kinematics of some of the faintest halo stars ever attempted (J ~21.6 / F814W ~23), adding a 4th ever measurement of a halo rotation velocity, and understand the orbit and angular momentum content of NGC 253's dominant merger.

This work is currently in progress. The biggest challenge is that due to the faint nature of these halo stars, the spectra are very sky-dominated and extremely low S/N, so I have had to develop robust sky-subtraction methods combining existing algorithms (ESO'sskycorr) with principle component analysis (PCA) to isolate the signal from the stars.

A girl with glasses and bundled in a scarf, fleece, and long purple pants stands on a stepstool behind a big bundle of fiber optic cables connected from a plate to a spectrograph. She is plugging one of the fibers in.

Undergraduate Work

Before diving into resolved stellar populations of faint stellar halos, my undergraduate research harnessed the power of nature's magnifying glasses to study distant, high redshift gravitationally lensed galaxies. Strong gravitational lensing is a phenomenon that occurs when a massive foreground galaxy or galaxy cluster bends the light from a more distant background object, magnifying and distorting its image. This effect can be used to study the properties of distant galaxies and quasars that would otherwise be too faint to observe and put constaints on dark matter distributions and cosmological parameters.

A gif showing a background of galaxies get warped and distorted by a foreground galaxy, which is acting as a gravitational lens.
A simulation of how a massive foreground object lenses backgruond galaxies around it. Movie by Frank Summers (STScI)

Masses of Two Lensed Galaxies

A 6 panel image showing black-and-white HST images, GALFIT models, and residuals for two lensed galaxies, SGAS J1723 and SGAS J2340.
HST images, GALFIT models, and residuals for SGAS J1723 (top row) and SGAS J2340 (bottom row). The SGAS J1723 HST image is a stack of the F390W and F775W images, while the SGAS J2340 image is the F814W filter.
My first major research project was measuring the stellar mass of two lensed galaxies, SGAS J1723 and SGAS J2340, using multi-band HST WFC3/IR and Spitzer IRAC imaging data. This was part of a larger photometric and spectroscopic study to explore the spatial variations in emission line ratios of these two systems, which can tell us about their metallicity, ionization state, and reddening.

I used GALFIT, a 2D parametric fitting code, to create detailed, multi-band morphological models of the lensed galaxies and calculate the photometry of each lensed source, clump-by-clump. Using this photometry as a constraint, I then used spectral energy distribution (SED) fitting to measure the stellar mass of these galaxies using the MCMC-based SPS code, Prospector. I found that both galaxies are low-mass (~109 M☉) and very young.

COOL-LAMPS logo. The second 'O' is in the shape of a strongly-lensed galaxy arc.

A color image of a strongly lensed galaxy arc, with a Magellan/LDSS3 spectrum overlaid.
An RGB image of the lensed galaxy arc at z~5 our class found! Overlaid is the Magellan/LDSS3 spectrum of the brightest knot in the arc. The raw spectrum is plotted in light blue, a smoothed spectrum is in orange, and a reference spectrum to determine the redshift is in green. Rest-frame UV stellar and nebular absorption lines are indicated by the red lines.
My last year of undergrad (2019-2020), I was lucky enough to be a student in the first cohort of a new research class called the Field Course in Astronomy and Astrophysics. This class, led by Prof. Mike Gladders, was designed as a Course-based Undergraduate Research Experience (CURE) and focused giving us background knowledge, data analysis, research, and academic writing skills, and hands-on experience with observing through the theme of finding and characterizing strongly lensed galaxies and quasars across redshifts. The class resulted in the creation of the ChicagoO Optically selected Lenses - Located At the Margins of Public Surveys (COOL-LAMPS) collaboration, which has had five other cohorts of students since our inaugual class and has resulted in 8 student publications to date.

One of the bright lenses that our class found and wrote the first COOL-LAMPS paper on, COOL-J1241+2219, was found in DECaLS DR8 in a visual lens search. Though our intially-planned in-person observing run to Magellan to follow-up candidate lenses was cancelled due to the COVID-19 pandemic, we were still able to get data remotely with the Magellan/Baade/FOURSTAR and Magellan/Clay/PISCO imagers, and spectroscopy with the Magellan/Clay/LDSS-3C and the Gemini-North/GNIRS spectrographs. The data we obtained allowed us to confirm CJ1241+2219 as the brightest galaxy known at z>5. Because I already had experience doing lens characterization, I was able to provide mentorship to other students in the class on how to use GALFIT to model the giant arc and Prospector to do SED modeling.