Eclipsing binaries (EBs) are systems in which two stars orbit and pass in front of each other. They are important astrophysical tools utilized in this project to directly measure the fundamental properties of stars, such as their masses, and radii. We create a sample of ~50 bright EB targets based on high-quality photometry from the Kepler Satellite and spectroscopy from the Apache Point Observatory Galactic Evolution Experiment (APOGEE). Here, we present mass, radius, and orbital solutions for a subset of these EBs. For each system, we extract radial velocities (RVs), the speed at which each star in the system is moving away from the observer, from the APOGEE spectra. Then, we combine the RV with Kepler light curve (LC) information to simultaneously model the system light and line-of-sight speed as a function of time. Because our model has high dimensions with 18 free parameters, we first solve for the LC and RV solutions separately, and then simultaneously solve them. We use a least-squares optimization method to determine the best fit solution and quantify the uncertainties in model parameters running Monte Carlo Markov Chain (MCMC) simulations. The model parameters that we find will further our understanding of fundamental stellar properties and evolution.

As a part of the Astronomy Undergraduate Engineering Group, we have further developed the software responsible for controlling the Manastash Ridge Observatory’s (MRO) auto-guiding system. Using the programming language Python, we have evolved the guider’s graphical user interface (GUI) that will be used to control the auto-guiding system so as to make it capable of communicating with the telescope to make adjustments, such as tracking, focus, and guider rotation. The guider software establishes basic functionality allowing us to take images, locate stars in them, and log the associated data, as well as use the stars in the field of view to dynamically correct tracking and focus errors. Essentially, we have reconciled two partially developed systems—the GUI and the guider software—to use the GUI as an accessible and streamlined control system for the guider. This system will provide users with an easy means of regulating data collection.

New hardware at the University of Washington’s resident astronomical research facility, Manastash Ridge Observatory, allows the opportunity to implement new software control systems. We’ve built these control systems, Bifrost and Evora, in the programming language Python to provide an intuitive well-documented replacement for former limited software. Bifrost is the telescope control computer, handling all aspects of physical telescope motion. The new auto-guider provides precise corrections to telescope motion and is controlled by the observer through Bifrost. Bifrost also provides the observer with a robust toolset such as logging, real-time airmass readout, and target trajectory plots. Additionally, scientific imaging is performed by the new Andor Camera. Evora interfaces with Andor, enabling a modern feature set of camera controls beyond simple exposures and temperature control. More advanced features include series imaging, camera scripting, and low level photo analysis. Together, Bifrost and Evora provide a feature-complete observing experience for the end-user. Most importantly, they provide a strong base for future projects through documentation and extensible design to enable the development of new software such as for the upcoming spectrograph.

With over 2,000 new discoveries the field of exoplanets is rapidly expanding. As the detection of new planets continues to increase, many seek to answer: What do these distant exoplanets look like? Current telescopes are not yet able to image the surfaces of exoplanets, but plans are underway for a future NASA telescope that will be capable of observing light reflected off terrestrial exoplanets using a coronagraph instrument to block out the starlight. Located at distances of many light years, these planets only appear as dots of light. Using the Virtual Planetary Laboratory’s (VPL) 3D Spectral Earth model, light reflected off the many unique surfaces and clouds on Earth were modeled and compressed into a single spatially-unresolved dot. By observing how the brightness of the planet changes as it rotates on its axis we were able to see different surfaces reflecting more or less light. A coronagraph-equipped telescope noise model was used to apply realistic observational noise to the spectra from the VPL 3D Spectral Earth Model. This allowed us to simulate data a future telescope would be capable of observing. After producing the data, periodograms were used to infer the planetary rotation rate, and Principal Component Analysis (PCA) helped us recognize how many unique surfaces an exoplanet has and approximately what color each surface is. Following the PCA, a more sophisticated surface mapping algorithm called the Surface Albedo Mapping Using RotAtional Inversion (SAMURAI) model was used to derive the underlying planetary surface covering fractions and their respective wavelength-dependent albedos that give rise to the time-dependent data that we observe. We show that many plausible configurations for the next-generation NASA telescope will be capable of mapping the area distributions and colors of oceans, land, and even vegetation, should these surfaces exist on any nearby exoplanets.

The host complex of GRB 020903 is one of only a few long-duration gamma ray burst (GRB) environments where spatially-resolved observations are possible. It may also be the only known GRB host consisting of multiple interacting components, as well as an active galactic nucleus. We were granted 4.5 hours of observing time on the Gemini Multi-Object Spectrograph (South) to obtain spatially resolved spectra of the GRB 020903 host complex. Using long-slit observations at two different position angles we were able to obtain optical spectra of the four main regions of the GRB host, with a spectral range of 3600 - 9000 Å. From this data we discern the redshift of each region, concluding that they do not comprise a single interacting system as previously assumed. Two regions, including the GRB host region, are at an approximate redshift of 0.251 as expected. The other two are at a redshift of 0.662. We also measure the metallicity, star formation rate, and young stellar population age of each region to create a spatially-resolved map of these parameters for the larger host complex. Based on the distribution of these characteristics we determine whether the localized GRB explosion site is representative of the host complex as a whole, or localized in a metal-poor or strongly star-forming region.

Supernovae (SNe) are the spectacularly violent deaths of evolved young massive stars, which expel a shock wave into the intergalactic medium that in turn can spark star formation and disperse heavy elements into the galaxy. While a SN event can be classified by its spectral signature definitively, determining the nature of a SN progenitor depends upon chance photometry taken prior to the event. We circumvent this challenge by turning to the study of SN host environments and their surrounding interstellar medium within the unique and rare population of galaxies that have hosted three or more SN events within the last century. This subpopulation offers the opportunity to study the locations and environmental properties of stellar populations prone to supernovae production. Using moderate-resolution slit spectra spanning 3500-7000 Angstroms taken with the Apache Point Observatory 3.5m telescope DIS spectrograph, our goal is to map the metallicity, ionization parameter, and star formation rates using emission line diagnostic ratios for all of the SN host sites across each SN-rich galaxy. Dubbed the “Fireworks Galaxy” at a distance of 6.9 ± 2.4 Mpc, NGC 6946 has uniquely produced nine core-collapse supernovae (CCSNe) within the last century, which specifically occur when massive stars develop an iron core under gravitational collapse. We present the spatially-resolved metallicity and star formation rates (SFRs) of NGC 6946, tracing nine spectrophotometric slits centered on each CCSN host site across the galaxy. Future work includes stellar population synthesis modelling to determine the host galaxies’ stellar populations, ages, and SFR histories for NGC 6946 and, eventually, the other nine SN-rich host galaxies in our sample.

The Kepler Space Telescope was designed to detect exoplanets by measuring the change in light received as target planets pass between their host stars and Earth. After the failure of two Kepler reaction wheels, the pointing capability required for high-accuracy transit detection was lost. To reduce the noise generated by unstable pointing, the space telescope was aligned to be held stable by the photon pressure from the sun, and the telescope entered a second phase of observation called K2. Although this reconfiguration still detects transits, the precision has decreased resulting in fewer exoplanet discoveries than Kepler. The EVEREST (EPIC Variability Extraction and Removal for Exoplanet Science Targets) pipeline was developed to remove instrumental noise from K2 data and increase the detection capabilities of the telescope. Due to the failed reaction wheels, Kepler targets have motion relative to the detector. EVEREST uses a detrending method called pixel level decorrelation (PLD), which uses the flux measured in each pixel to account for the motion of the star. One possible source of noise is variation in the quantum efficiency between pixels as the star moves across the detector. This includes inter-pixel variation globally on the detector and intra-pixel variation from the center of pixels to their edges. My research characterizes the sensitivity variation of the detector and its impact on the light curves of Kepler targets. I am developing methods using python to simulate a K2 star with a planet, modelling the sensitivity variation, and using this model to test the power of PLD. Preliminary results indicate that defining apertures to limit total flux results in higher accuracy light curves. My research also aims to apply the results to the EVEREST pipeline, as the combination of PLD and sensitivity variation modeling could result in the highest accuracy publicly available K2 data to date.

A long-standing goal of exoplanet astronomy is to understand the physics that determines whether or not a planet could support life. An important, but poorly-understood, aspect of planetary habitability is the role of geophysics. Here on Earth, the decay of radiogenic isotopes heats the interior and drives plate tectonics that support geochemical cycles that maintain the stability of our climate over geologic time scales. Whether such cycles are common or rare among exoplanets is an open question. A further complication is tidal heating — while not significant on Earth, it may provide a large source of heat for potentially habitable planets around low-mass stars. To explore these processes, we simulate the evolution of Earth-like exoplanets over a range of radiogenic and tidal powers. We find that, depending on the radiogenic abundances and magnitude of tidal heating, planets may be unable to support a magnetic field or experience extreme volcanism, both of which could be detrimental to life. These results suggest the geophyisical evolution of exoplanets may be a determining factor in the habitability of some exoplanets, such as Proxima Centauri b and members of the Trappist-1 system.

Open clusters are groups of gravitationally bound stars formed from the same molecular gas cloud. These clusters and their stellar metallicity enhance our knowledge of stellar evolution, stellar cluster characteristics, and the chemical evolution of the Milky Way (MW) Galactic disk. Stars that are actual members of the cluster must be identified within each cluster’s observed field of view. We present OCMAP, an Open-source Customizable Membership Analysis Program developed in python for use within the astronomy data science community. Cluster membership is determined using simulated star fields and stellar photometry magnitudes across multiple filters. OCMAP matches stars across images, plots color-magnitude diagrams (CMDs), standardizes stellar magnitudes, and calculates the local and field star density on a CMD for each prospective member in order to determine a probability of membership for each star. OCMAP was developed to ultimately determine the membership and metallicity of 42 open clusters observed with the Apache Point Observatory 3.5m telescope and Dominion Astrophysical Observatory 1.83m Plaskett telescope using Sloan and Strömgren filters, respectively. We tested OCMAP on 8 of the 42 open clusters we have observed, and now present the resulting members-only Sloan [u-g] vs. [g-r] and Strömgren [m1] vs. [b-y] metallicity plots, as well as Sloan [g] vs. [g-r] and Strömgren [y] vs. [b-y] color-magnitude diagrams for each open cluster.

As part of an ongoing effort to maintain the current high standard of astronomical instrumentation at the Apache Point Observatory in New Mexico, the Astronomy Undergraduate Engineering Group is pursuing a series of instrument and facility upgrades. One such upgrade includes new optical slits for the Dual Instrument Spectrograph (DIS). This instrument is a “variable slit” aperture that has a series of fixed width slits that mount into a rotating wheel. With DIS, the user can rapidly switch the slits as needed for their observing program, which is a powerful feature of the spectrograph. DIS is used in long slit spectroscopy from observing the rotation curves of galaxies to the expansion of optically thin nebulae. The slits being replaced were machined from stainless steel, giving undesirable width and edge uniformity. To eliminate these issues, we have constructed the new series of DIS slits using microelectromechanical systems (MEMS) processing at the Washington Nanofabrication Facility in order to get slit widths with precision on a sub-micron scale and edge uniformity less than 1%. The MEMS processing involved laser lithography, wet etching silicon wafers with KOH, and thin film deposition of aluminum. Full characterization of the finished product was done with microscopy and Fraunhofer laser diffraction. This fabrication process will provide unprecedented spectroscopic accuracy, enabling astronomers from dozens of member institutions to gather data faster and more accurately than ever. The MEMS processing manufacturing technique also has the potential to be used at other sites as it allows a way to make a wide variety of slit configurations more efficiently and at a lower cost than ever before.

The University of Washington’s Manastash Ridge Observatory, a facility utilized by undergraduates to train students in current observational astronomy methods, is being upgraded with a new spectrograph system. Our student-led upgrade is a redesign of all aspects of the existing spectrograph and calibration system. We have purchased a new camera and redesigned a pickoff system that will allow the telescope to switch between spectroscopy and direct imaging. The necessary focal reducing and transfer optics are simple, off-the-shelf items, while the pickoff mirror and slit viewer optics will be made by us. The new spectrograph will use fiber optic cables to take light from multiple locations in the image and extract several spectra simultaneously. This will allow for the spectra of the target, background sky, and a calibration source, which consists of mercury and neon lamps, to be directly compared. Our new method simplifies the process of data reduction and reduces any systematic errors that may occur from one exposure to the next. The light from the dedicated calibration fiber will provide a consistent reference with known emission lines from one exposure to the next, and provide the baseline needed to scale wavelength dispersion per pixel. Once completed, the spectrograph will allow University of Washington students to perform spectroscopy during observation, which is relevant to both undergraduate astronomy course work and modern astronomy research.

PanOptes is a low-cost, wide-field robotic survey telescope that uses two DSLR camera bodies to take images. These camera bodies house CMOS chips, which are already used in astronomy today. The use of off-the-shelf hardware keeps the cost down while still taking valuable science data. The ultimate goal of the project is to ultimately track the brightness of stars in a field in order to find transiting exoplanets, and the current capacity of the project is responding to gamma ray bursts. A major part of this project will be constructing a data pipeline that is capable of reducing several gigabytes of data per night of data, and extracting brightness values for each stellar body in the image. From this, brightness values over time will be tracked, allowing for planet transits as well as variable objects to be monitored. This low budget, high value student driven instrument will be tied into a wider network of small robotic telescopes to further increase the scientific value of the project.

Quasi-stellar objects (also known as quasars/QSOs) are the most luminous type of all known astronomical objects and can be seen over vast cosmological distances (often parameterized in terms of the degree to which light is “redshifted” due to the expansion of the universe). Quasars are characterized by an intrinsic astronomical spectrum that features a very bright continuum, which makes them useful as standard “background sources” to assess foreground material, such as the circumgalactic and intergalactic media. These are the warm-hot gaseous regions of space between and surrounding galaxies that are not directly measurable or observable due to their incredibly diffuse nature. However, as the quasar photons interact with this gas, the intrinsic spectra of quasars is distorted as atomic elements (such as Hydrogen, Oxygen and Iron) absorb some of this light. In my work, I use a Python Graphical User Interface to display the entire processed spectral region of interest and manually scroll through redshifts to detect and identify possible absorption features. I identify hundreds of features as foreground gas at distinct redshifts and subsequently correlate it with the redshift distribution of galaxies nearby in projection along the line of sight for a quasar. I am focusing on QSO PG1407+265, a quasar with a high signal to noise spectrum from a Hubble survey taken with the Cosmic Origins Spectrograph (COS). From that spectrum I have produced absorption models that I am analyzing in conjunction with ground-based galaxy spectra in order to compare host galaxy and circumgalactic metallicities. This particular QSO has a galaxy and absorption system rich sightline that will ultimately serve as a powerful diagnostic of the cosmic baryon cycle, the fundamental driver for galaxy evolution.

Analyzing complicated, extended source emissions in radio frequencies allows for going beyond typical point source models used for Epoch of Reionization analysis. Our research uses data from the Murchison Widefield Array, which is a radio telescope observing regions of the southern hemisphere to detect the faint signal from hydrogen during the formation of the first stars and galaxies. Extended source emissions can include celestial bodies ranging from active galactic nuclei emitting jets of synchrotron emission to the first galaxies that populated our universe. We anticipated the extended sources to show structural changes from day to day and in different frequencies, due to normalization errors, ionospheric disturbances, and physics of the frequency-dependent emission.We made histograms of the radio data of the extended source objects to create images in different frequencies and from different days to get a more accurate view of the extended sources. This process was also used to create difference images for direct comparison between frequencies and observations. These images can be incorporated into a formula that removes unwanted bright objects from our view of the Epoch of Reionization. We did, in fact, observe expected changes from day to day and in different frequencies, as well as observing changes that could be due to atmospheric disturbances or other potential sources of error. This has been taken into account along with the potential for improvements to normalization for the sake of comparison. The images created prove that we are statistically sampling at a higher resolution than expected due to the brightness of the extended source objects, which has important implications on our ability to work towards imaging and understanding the point just beyond the Epoch of Reionization, and from there, the early Universe.

Most of the atomic matter in the universe lies in the diffuse ionized gas in the inter and circumgalactic medium, cycling in and out of galaxies over billions of years. Mapping this cosmic baryon cycle is the key to understanding galaxy properties and evolution. Because we cannot observe gas by its own emission, we use a technique called quasar absorption line spectroscopy to study the gas. Quasars serve as bright background sources, and as light passes through gas clouds and galaxy halos we can analyze absorption features in the spectra to determine elements that are present in the gas. These absorption features also give us information about the distance to each gaseous system. When an absorption system with a high density of elements is identified in proximity to a known galaxy, it is likely that the system is galactic halo gas that is cycling in and out of the galaxy. I present my analyses of three quasar spectra at redshifts z = 0.789, 0.88, and 0.99, with a list of identified gaseous systems that are potentially associated with known galaxies. The study of these systems will help us gain an understanding of how this cosmic baryon cycle has given rise to the visible structures in the universe

The circumgalactic medium (CGM) is gas that is constantly being thrown out and being recycled back into galaxies. Learning more about the composition and location of this gas can give information about its origin and fate. However, this gas is exceedingly difficult to observe in emission, as it is very diffuse and ionized, and therefore must be observed in absorption at ultraviolet wavelengths. Fortunately, the Hubble Space Telescope (HST) Cosmic Origins Spectrograph (COS) installed in 2009 allows for very sensitive measurements of CGM UV absorption lines. As light from very bright quasi-stellar objects (QSOs) passes through the universe on its way to Earth, it encounters a galaxy’s gaseous CGM resulting in absorption lines from ionized gas in the QSO spectra. My research involves an analysis of many QSO spectra from HST/COS. I identify ionized elements seen in absorption at distinct redshifts along the QSO line of sight. The goal is to identify everything in the spectrum, and in doing so isolate the systems of gas that contain strong Lyman Series lines from Hydrogen and metals (elements heavier than hydrogen) with a corresponding redshift from the Milky Way. These redshifts can be used to reference various databases to actually look for the host galaxy halo at some projected distance from the QSO absorber system. With this information, the distance between the galaxy and the sightline of the QSO can be calculated giving a lower limit for the extent of its CGM. Additionally, the composition of the galaxy can be compared to that of the gas in its CGM. Ultimately, these data will further our understanding of the dynamic processes that drive galaxy formation.

At low obliquity, planets like Earth will tend to form permanent ice sheets at the poles, if anywhere. However, at high obliquity, the poles receive more sunlight on average than the tropics, and so ice sheets might be expected to form at the equator. We have used a simple energy balance model for investigating the orbital, atmospheric, and obliquity parameters that allow for the formation of permanent ice at the equator. We find there is a narrow range of semi-major axis and obliquity values that allow for such “ice-belts”. First, we experimented with the range of stability of ice-belt formation by increasing the semi-major axis to the edge of the habitable zone, increasing CO2 levels to counteract the decrease in solar flux. This led to a very small range of conditions that allow for stable ice-belts. To further investigate the cause of this, we decreased the luminosity of the host star while increasing CO2 levels (to simulate the control temperatures). This stabilizes the otherwise unstable system, although the range of ice-belt formation is still narrow. Further, we find that this equatorial ice formation often leads to instability due to the ice-albedo feedback. This work suggests that it should not be common for Earth-like worlds at high obliquity to maintain large ice sheets.

Little is understood about how supermassive black holes affect their host galaxies in terms of growth and shape. Using a large simulation called Romulus25, we are able to study how galaxy properties correlate with black hole properties throughout time. From this simulation we draw a sample of 40 massive galaxies and analyze their properties to understand how they are affected by the growth of supermassive black holes. We examine galaxies with a variety of star formation histories and investigate how the black hole is involved in determining these histories. We seperate the galaxies into three categories: 1) Galaxies with on going star formation 2) Galaxies that stop forming stars 3.) Galaxies that stop and then restart star formation. We show how the black hole activity changes between categories. These results may lead to solving the mystery: what causes the star formation in the galaxies to quench, (and in some cases) un-quench? 

The large-scale configuration of the cosmos has long been a mystery to astronomers. My research project aims to shed light on this mystery. The overarching goal for my research project is to further understand the distribution of matter on 10⁵ –10⁷ light-year distances by comparing numerical simulations to actual observations of the universe. This comparison tests the current theoretical parameters of the physics governing the largest scales of the cosmos. I used simulated data from models made by Professor Matthew McQuinn and observations of quasar spectra from the Very Large Telescope (VLT). From the VLT data, I analyzed quasar spectra for absorption lines at a hydrogen electronic transition called the Lyman-alpha line. The IGM absorbs amply at this transition, creating absorption lines throughout the quasars’ spectra. I processed this data by programming Python scripts to compare flux probability distribution functions (PDFs) of the Lyman-alpha absorption in the observed quasar spectra to the simulated quasar spectra. Through this comparison, PDFs of observations fell within one standard deviation of the simulated PDFs. Although at higher redshifts (further back in cosmological time), there was considerably more variance. In order to further constrain the observational flux PDFs, I aim to eliminate systematic and statistical uncertainties that may make the observations spuriously differ from models. The salient errors include fitting precise continua (spectra without absorption) and removing metal (predominantly carbon-4) contamination from the Lyman-alpha forest. I used Principal Component Analysis to predict the continuum spectrum but found that a fit by eye was more effective. Through analysis of absorption probabilities outside the Lyman-alpha forest, I statistically removed carbon-4 from the data inside the Lyman-alpha forest. Ensuring the most precise flux PDF measurements are used is critical to the comparison of observations to simulations and to the constraints of parameters governing the universe.

Through analyzing the properties of the host galaxies to Supermassive Black Hole (SMBH) mergers, we will be able to determine what phase of galaxy evolution these mergers track. We do this by analyzing ROMULUS25, a large cosmological simulation of the universe, from the Big Bang to present day. The data from ROMULUS25 includes the evolution of thousands of galaxies and is able to resolve the internal structure of galaxies. Most galaxies have SMBHs in their core, and when galaxies merge, it is believed that their SMBHs merge as well. Project LISA, as well as other future projects, will be able to detect gravitational waves coming from these mergers. With simulations like ROMULUS25, we will be able to better predict which galaxies are likely to host these SMBH mergers, which will give context to future gravitational wave studies with respect to galaxy evolution theory.

Supernovae enrich the interstellar medium, allowing it to create fascinating and exciting objects such as stars, planets, and people. However, supernovae are difficult to study because, most of the time, we do not observe these supernovae directly, since the whole event can occur in a very short time. Fortunately, they leave behind bright remnants that can give us information about the supernova and even the stars that existed before. We have observed one such supernova remnant in the nearby spiral galaxy NGC 300, which is approximately three million light years away from Earth. We have acquired multi-wavelength data of this source from Chandra X-ray Observatory, XMM-Newton X-ray Observatory, and Hubble Space Telescope. Using this variety of telescopes, we can understand how this supernova remnant behaves in both the x-ray and optical energy bands which gives us a more complete picture of our source. From our Hubble data, we gain detailed measurements of the size and shape of the source. From our Chandra and XMM-Newton data, we can shed light on the properties of this supernova remnant by performing a spectral analysis on the x-ray data. This spectral analysis gives us information about the temperature and metallicity of the source, as well as the density of the surrounding gas. With these quantities, we can better understand this supernova remnant and supernova remnants like it which, in turn, let us better understand the life of the star that created this source.

Our goal is to quantify the fraction of H-α emitting B stars in M31. A small subset of massive stars emits strongly in H-α, which indicates the presence of hotter gas than is typically associated with such stars. These stars’ line emission may be related to stellar rotation, but it is unclear what causes the rotation or if there are other possible causes for the line emission. In order to shed light on this problem, we’ve begun a project to identify a homogeneous sample of these stars in M31. Our data are comprised of imaging from six fields in M31. Using resolved photometry of over two million stars, we have found 50 candidates. Using a ratio of these candidates to ordinary stars of the same temperature that are not emitting in the H-α, we can put observational constraints on models of the formation of stars in this mass range, as any viable model must be able to reproduce the observational statistics of these rare stars. Therefore, despite their rarity, they provide a key diagnostic of stellar evolution models fundamental to many fields of astrophysics.