Research

SPISEA: A Flexible Python Package for Simulating Clusters

SPISEA (Stellar Population Interface for Stellar Evolution and Atmospheres) is an open-source python package that generates simple stellar populations. It offers the user control over 13 input parameters, including the stellar evolution and atmosphere models, extinction law, differential extinction, initial mass function, and stellar multiplicity. In addition, the user can define the initial-final mass relation in order to produce compact stellar remnants (white dwarfs, neutron stars, and black holes). SPISEA has been used for modeling the IMF of star clusters (Hosek et al. 2019), measuring the extinction law (Hosek et al. 2018), predicting microlensing event rates (Lam et al. 2020), and calculating photometric transformations between filters in regions of high extinction and non-standard extinction laws (Gautam et al. 2019, Chen et al. 2019).

SPISEA is described in Hosek et al. (2020). It is available for download on GitHub with documentation on ReadtheDocs. Contributions are welcome!

Diagram of the SPISEA code. The open boxes represent inputs specified by the user when creating a cluster. The primary code outputs are the isochrone and cluster objects, which are represented by shaded boxes. Taken from Hosek et al. (2020).

The Arches and Quintuplet Star Clusters: Star Formation Near the Galactic Center

The Arches and Quintuplet clusters are young (2-5 million years old) and massive (~10,000 solar masses) star clusters near the center of our galaxy. They provide a unique window into star formation in the extreme environment at the Galactic Center. Of particular interest is the Initial Mass Function (IMF), which describes the distribution of stars formed as a function of mass during the star formation process.

Before we can understand the IMF of these clusters, we need to identify which stars actually belong to the cluster versus those in the foreground and background. This is challenging because the Arches and Quintuplet are heavily shrouded by thick patches of interstellar gas and dust which vary in intensity across the field. As a result, standard methods of identifying cluster stars based on their similar brightnesses and colors are not accurate. To overcome this, my collaborators and I identify cluster members via astrometry, or the precise measurement of stellar positions and proper motions. Since cluster members are gravitationally bound to each other, they have a common proper motion relative to the field stars.


The Arches Cluster

Using multiple Hubble Space Telescope observations of the Arches cluster taken over 6 years, we calculate cluster membership probabilities for stars down to ~2 solar masses with only 6% - 10% field contamination. In Hosek et al. 2015 we use this sample to measure the stellar radial density profile of the cluster out to 3 pc for the first time, revealing it to be significantly larger than previously expected. We also measure significant mass segregation, where high-mass stars are preferentially concentrated toward the center of the cluster, and fail to detect evidence for tidal tails for this stellar mass range.

Combining these measurements of the cluster structure with a custom extinction law derived for highly reddened stellar populations in the Galactic Plane (Hosek et al. 2018), we measure the IMF of the Arches cluster. We adopt a forward modeling approach to simultaneoulsy constrain the IMF with other cluster properties (e.g. age, distance, and total mass) while accounting for observational uncertainties, completeness, mass segregation, and stellar multiplicity. We find that the Arches either has a significant overabundance of high-mass stars relative to local star forming regions (i.e. a ``top-heavy'' IMF) or a dearth of low-mass stars (i.e. a ``bottom-light'' IMF). Though our data cannot yet distinguish between these two models, it is clear that the Arches IMF is quite unusual. Comparing the Arches to young massive clusters in the Galactic Disk, we find tentative evidence that this unusual IMF might be attributed to the extreme Galactic Center environment. However, further studies of these other clusters are required to confirm this (Hosek et al. 2019).

Left: The Arches Cluster, a young massive star cluster near the Galactic Center. In this HST WFC3-IR image, F153M is red, F139M is green, and F127M is blue. Right: The observed mass function of the Arches cluster (red points), compared to the best-fit IMF (red shaded region), assuming a single power-law model. There is a significant overabundance of high-mass stars compared to the local IMF (blue dotted line). Taken from Hosek et al. (2019).

The Quintuplet Cluster

We conduct a similar analysis of the Quintuplet cluster, using Hubble Space Telescope proper motions to calculate cluster membership probabilities for stars down to ~1.8 solar masses. In Rui et al. 2019, we measure the stellar radial density profile of the cluster, finding that it extends to a radius of 3.2 pc without any signs of tidal truncation. Interestingly, we do not observe significant evidence of mass segregation, indicating that the Quintuplet is dynamically younger than the Arches despite being the older of the two clusters. An IMF analysis of the Quintuplet cluster is ongoing.


Extragalactic Spectroscopy of Blue Supergiants

Blue supergiants (BSGs) are the advanced evolutionary stage of stars 12-40 times the mass of the sun. They are so bright that we can obtain resolved low-resolution spectra of these objects in nearby galaxies out to 10 Mpc (~32 million light years) away, far beyond the distance where resolved spectroscopy of other stars is possible. Blue supergiants are thus a useful tool for studying these galaxies.

Blue supergiants in dwarf galaxy NGC 3109, identified by blue and red circles. These stars are bright enough that we can obtain resolved low-resolution spectra of them despite the fact that the galaxy is 1.27 Mpc (~4 million light years) away. The other bright objects in the image are not associated with the galaxy.

By comparing the spectra of blue supergiants to stellar atmosphere models, we can derive two important pieces of information: how many metals (elements heavier than helium, in astronomer-speak) are present in the star and its distance from Earth. The metal content, or metallicity, of a star is important because it reflects the metallicity of its surrounding environment, which in turn yields information about the evolution of the galaxy at large. From spectroscopy we can also determine the surface temperature and gravity of the star, which can be converted into a luminosity (or intrinsic brightness) through the Flux-Weighted Gravity-Luminosity Relation (FGLR). By comparing how bright the star appears on Earth to how bright it actually is, we can determine the star's (and thus the galaxy's) distance. For more information about the FGLR and its advantages as a distance indicator, see Kudritzki et al (2008) and Kudritzki et al (2012).

Working with Rolf Kudrtizki, Fabio Bresolin and collaborators, I did quantitative spectroscopy on 12 late-B and early-A type blue supergiants in dwarf galaxy NGC 3109, providing an independent measure of the galaxy's metallicity and distance. This work is part of an ongoing effort to study Local-Group and nearby galaxies using blue supergiants, important because galaxy metallicities are difficult to measure and because it provides a crucial check of distances determined using other methods such as Cepheid Variables. For details see Hosek et al. (2014).

The observed Flux-Weighted Gravity-Luminosity Relation (FGLR) of blue supergiant stars. If we can determine a BSG's surface temperature T and gravity g (combined in the variable g_f = g / (T * 10^-4)^4, in cgs units), this relation allows us to determine the star's absolute magnitude, which we can use to calculate a distance.

Undergraduate Research

Outburst Dust Production of Comets

The intrinsic brightness of a comet is an indicator of how active it is, or how much material it is ejecting into space. In general, comets become brighter as they get closer to the sun and warm up, which causes water and other volatiles to sublimate from its surface. Every so often, a comet will dramatically brighten and then fade back to its original brightness over the span of a few days. These events are called outbursts, and are thought to be triggered by a sudden massive release of dust into space. Because they are so unpredictable, outbursts are difficult to study in their entirety.

Comet 29P/Schwassmann-Wachmann 1 (SW1) is an example of a comet that is known to outburst frequently. Working with NASA Marshall Space Flight Center Meteoroid Environment Office, I monitored he brightness of the comet during its 2011 and 2012 apparitions. We observed two outbursts over this time, and because of the frequency of our observations, we were able to record the events from nearly beginning to end. This allowed us to roughly calculate the total mass of dust produced in each event and estimate that a significant fraction (greater than 50%) of the yearly dust production of SW1 comes from outburst activity. This study demonstrates that outbursts can be a significant source of cometary dust, and so these events must be monitored for in Earth-crossing comets to improve meteor shower forcasts. See Hosek et al. (2013) for details.

The May 2011 outburst of SW1. Left: Image of the comet before (left) and after (right) the observed outburst. Right: Afρ, a proxy for dust production, versus time during the May 2011 outburst. Note the rapid increase in dust production followed by the prolonged decay. This outburst was observed over a span of 9 days.