The Carbon IV emission line in quasar spectra is well-known to show systematic shifting toward smaller wavelengths often attributed winds flowing off the accretion discs along the line of sight. We investigate variability in the shift of the C IV line in two ways, using spectra observed multiple times: through correlation of the shift with changes in the intensity of the rest of the quasar, as well as through the effects of this variability on host supermassive black hole mass estimates. Our results show weak to no correlation between changes in the shift with quasar light flux between observations, potentially indicating near-constant outflow speeds. We additionally investigate biases in existing SMBH mass estimates by comparing our multiple-observation estimated masses with masses derived from single observations. Our results indicate that the previously observed correlation between C IV shift and SMBH mass can be empirically corrected thanks to the lack of variability in C IV shifts, leading to less biased SMBH masses. This will help pave the way for much more accurate methods of calculating masses for extremely distant black holes, and helps constrain physical structure models of active galaxies.

We want to understand whether or not the underlying magnetic activity of a star plays a role in the "parking" of a planet as it migrates inward. Magnetic activity of a planet's host star is one among a bevy of factors that inform planetary migration, and may imprint itself on the final orbital location of a planet. In this framework, less active stars allow planets to reside nearer to them than their more active counterparts, which "brake" planets' inward migration sooner. NASA's Kepler mission collected an ideal data set to address this question. It detected thousands of extrasolar planets from the periodic dimming of their host stars as the planet crossed the face of the star. By using the data collected from the mission, we are initially attempting to find a relationship between the mass of the host star and the current location of its nearest planet. We employ stellar mass as a proxy for stellar activity for this preliminary treatment. We can infer some level of stellar activity from the mass of the star, but we plan further investigation to determine the extent of this relationship and how it may change over time. We will look at the brightness variations of the stars in the Kepler sample, over short and long timescales, in order to infer stellar activity in a more robust fashion. Then, we will jointly examine where the closest orbiting planet resides around this sample of stars. We will establish how this parking distance relates to stellar type and stellar activity. We aim to untangle the role of magnetic activity from the other factors determining planetary architecture. 

Exoplanets are planets outside our solar system. The discovery and characterization of exoplanet systems informs us of the types of worlds found in nature and whether they might support life. There have been an unprecedented number of exoplanet detections in recent years spurred in large part by the advanced observing capabilities of NASA’s Kepler Space Telescope. The Kepler Spacecraft continuously monitors a large, fixed set of stars and collects time series photometric data that are unparalleled in both quantity and precision. Plots of stellar brightness against time called lightcurves are the canonical representation of Kepler data. The goal of our research is to detect exoplanets from the characteristic features they produce in these lightcurves known as transits. A transit occurs when an orbiting planet partially eclipses its stellar host, reducing the brightness observed by the telescope and recorded in the lightcurve. To detect transits, we use the Quasi-Periodic Automated Transit Search Algorithm (QATS). As an automated tool, QATS is instrumental in reducing the effective size of the Kepler dataset. However, QATS is sensitive to variability of the target star and to systematic artifacts of the telescope just as it is to transits. Thus optimal detection efficiency requires removal of all non-transit signals prior to applying the algorithm. Determining the best way to do this is the present focus of our work because it bounds our ability to detect planets whose size, composition, and orbital parameters are conducive to life.

 “Trojan” asteroids, or asteroids trapped in stable gravitational positions preceding and trailing a planet in its orbit, accompany almost every planet of our solar system. These hunks of rock were captured into their current locations in the early stages of our solar system’s formation, and their presence hints at the dynamical history of bodies orbiting the Sun. However, we have no reason to assume that our own planets are alone in possessing Trojan asteroids. The Kepler Space Telescope mission, launched in 2009, has been instrumental in the recent search for exoplanets (planets orbiting stars other than the Sun). It identifies these planets from a tell-tale dip in starlight when a planet passes in front of its host star. It has identified thousands of new worlds to date. However, no studies have yet been made attempting to detect Trojan asteroids accompanying these exoplanets from the Kepler photometry data. We expect to be able to detect these asteroids using statistical model fitting to the Kepler data, similarly to how exoplanets themselves are detected. By fitting a model based on the properties of the particular exoplanet to the photometry data, it is possible to deduce the presence of Trojan asteroids and their properties. However, if we are unable to detect these Trojan asteroids, we will be able to set limiting constraints on the presence of asteroids in exoplanetary systems. Observations of these Trojan asteroids, or the lack thereof, would give insight to the evolution and migration models of these systems. These results will ultimately lead to a better understanding on how planetary systems form, through which processes they migrate, and how they compare to our own solar system.

The Large Synoptic Survey Telescope (LSST) will carry out an unprecedented imaging survey of half the sky. With about 1000 observations over a 10-year period, these data will enable faint time-domain astronomy. The measured properties of newly discovered and known astrometric and photometric transients will be publicly reported within 60 sec. Several hundred million variable objects will be detected and due to this large number their fast classification, that is a prerequisite for timely follow up with other facilities, will have to be undertaken using machine learning methods. Before selecting a subset of the most interesting variable objects for follow up observations, known types of variable stars, which will dominate the sample, will have to be efficiently recognized and classified. This project assesses the extent to which machine learning methods are able to recognize known classes of variable stars, and how this performance scales with the number of data points. A sample of 7,000 variable stars, manually classified by domain experts, served as our training sample. This dataset, collected by the LINEAR survey, contains all the major types of variable stars and has sufficiently precise and well-sampled observations for a detailed light curve analysis. Fourier series expansions allowed us to quantify these light curves, and asses various classification methods in the subsequent best-fit Fourier coefficient space. We explored this space using Support Vector Machines, K-nearest neighbor classifier and decision tree methods, and compared the performance of these methods via their Receiver Operating Characteristic curves.

Molecular hydrogen is an important component of star and galaxy formation, but, unfortunately, is difficult to observe and measure directly. Neutral chlorine, however, is easier to detect and can be used to trace molecular hydrogen, because ionized chlorine rapidly reacts with molecular hydrogen and becomes neutral. While it is well established that there is a relationship between neutral chlorine and molecular hydrogen, there has been little research done using modern observations to determine the specifics of this relationship and to apply it to distant absorption systems. By comparing the amounts of molecular hydrogen and neutral chlorine (as well as atomic hydrogen and ionized chlorine) along the sightlines of various stars in the Milky Way, we hope to produce a more accurate chemical model of the interstellar clouds that contain these gases and to improve on the method for using chlorine to determine the molecular content of distant galaxies. At this stage in our research, we have discovered that the fraction of hydrogen in molecular form in the dense cores of these interstellar gas clouds is significantly larger than the ratio used in current chemical models. Also, while previous research has found that the slope of the correlation between neutral chlorine and molecular hydrogen is near one, our research, which includes a wider variety of sight lines, is revealing that this slope is actually significantly smaller. As we continue with our research, our next goal will be to repeat the process we have used on Milky Way stars on stars in the Large and Small Magellanic Clouds and to see if our findings close to home hold true as we move farther away.

Until the launch of NASA’s Kepler mission in 2009, the existence of rocky, Earth-like planets orbiting stars other than our Sun was uncertain. In the 5 years since its launch, the mission identified a wellspring of thousands of planets. We now know that planets around other stars are the rule, rather than the exception. One of the most interesting and surprising discoveries emergent from Kepler is that planets are not most common around Sun-like stars. Rather, they reside most often around small stars known as M dwarfs. These stars are cool, with effective stellar temperatures of 2000-4000 K, and are significantly smaller than our sun. Despite their diminutive stature, they are also the common stars in the universe, and the hardest to characterize. We understand the planets orbiting these stars only as a well as we understand the stars themselves. To this end, for our project we seek to refine our knowledge of the temperatures and radii of a select group of M dwarf host stars of interesting exoplanets identified with the Kepler Space Telescope. We employ dozens of near-infrared spectra gathered at the 3.5 meter telescope at Apache Point Observatory in New Mexico to study these planet hosts. We will leverage features in the near-infrared portion of these stars' spectra to understand their physical properties. The radius and temperature of the star will then tell us more about the possibly terrestrial planets they are hosting, and whether they reside in their stars’ habitable zones.

 Previously, details on the angular diameter, age, reddening, and distance from the sun have been published for 60 open star clusters. This previous research mentioned the metallicity of these clusters, but no results were published with metallicity values. According to the Big Bang theory, the early universe contained mainly Hydrogen with small amounts of Helium. For heavier elements to form, fusion inside a stellar core must take place. When these stars die they then expel heavier elements, called ‘metals’ by astronomers, into space. These metals will then be present for the next generation of stars that forms. Therefore, if astronomers can derive the metallicity of a star they can figure out the relative age of that star. The University of Washington League of Astronomers club research group wrote a proposal to observe at the Dominion Astrophysical Observatory (DAO) in Victoria, Canada and received eight nights of time on DAO's 1.8 m telescope. We have obtained images of many of the target clusters using Strömgren filters. Observations using these filters will allow us to calculate the metallicity of stars within each of our target clusters. Ultimately, by determining the metallicity of a range of open star clusters within the plane of the Milky Way, we can help constrain the galaxy's metallicity gradient which gives us a better understanding of galactic formation and chemical evolution models.

 At typical interstellar distances, the separation between a moon and its planet is too small to resolve; the separate spectra of the two objects combine and appear like one object. At most wavelengths the planet will outshine its moon, heavily skewing the center of light towards the planet. However at certain wavelengths a moon may outshine its planet, skewing that center of light towards the moon instead. Spectro-astrometric methods would find exomoons by detecting a shift in the center of light at a moon-dominated wavelength versus a planet-dominated wavelength. Once a moon is detected, tracking the motion of the center of light would also provide information about the masses and separation of the moon and planet. This research project is examining two systems: the Moon around the Earth, and an Earth-like moon around a warm Jupiter. Modeled spectra of the three objects were then used to model the detection of the two types of systems at a distance of 10 parsecs. For the Earth-Moon system, the strongest Moon dominated wavelength was found to be 2.83 microns. For the Jovian-Earth system, the strongest Earth dominated wavelength was found to be 2.44 microns. Further work will explore the signal to noise ratios and thus determine how large of a telescope is necessary to feasibly use this technique in the future.

Our understanding of our own galaxy and those at great distances hinges on how well we can model the complex processes by which stars form and evolve. The best laboratories in which to test our predictions are star clusters - groups of stars which by assumption formed at the same time. The nearby Andromeda galaxy (M31) is a fantastic source of clusters with similar chemical properties to our galaxy, at a variety of ages, and close enough to be resolved into individual bright stars by space telescopes. However, distinguishing cluster members from other field stars is a difficult problem. We have developed a method to assign a cluster membership probability to a given star by comparison with the density profile and properties of the cluster. We apply this method to a sample of well-characterized clusters observed by the Panchromatic Hubble Andromeda Treasury (PHAT) program, which has provided Hubble-based photometry of unprecedented depth in the M31 disk. As a first application, we study a subset of clusters associated with evolved stars that are known to be Cepheid variables - a class of pulsating supergiant star to which stellar models can provide a precise age. We demonstrate the accuracy of the method and characterize this newly identified population of M31 cluster Cepheids. We extend this method to the entire PHAT cluster sample, seeking to establish cluster membership of bright supergiant, horizontal branch, and AGB stars therein. The unprecedented size and quality of this sample of M31 clusters will allow us to test existing models of these important stages of stellar evolution.

Giant low surface brightness galaxies (GLSB) are a class of spiral galaxies with larger and fainter disks than typical spirals. Previous theoretical work suggests that their formation history is distinct from that of normal spiral galaxies. Two main theoretical models for GLSB formation have been proposed. One possibility is that they begin as ordinary spirals and undergo a face-on impact with a small satellite, creating a compression wave (i.e., a collisional ring) which excites star formation throughout the neutral hydrogen disk and results in a large, faint stellar disk. A competing possibility is that GLSBs begin as elliptical galaxies or normal spirals and acquire disks through tidal disruption and accretion of a satellite galaxy. Preliminary investigation of these objects has shown that they possess bulges which are morphologically and photometrically similar to typical elliptical galaxies, suggesting that the GLSB disks could be the remnants of a satellite accretion event. Using the Apache Point Observatory, we have obtained longslit spectra of the central bulges of two GLSBs, Malin 2 and UGC 6614. Combining these data with spatially resolved spectra of the disks taken at the McDonald Observatory allows us to compare the disk and bulge kinematics and distinguish between the possible formation scenarios. Our analysis reveals that disk and bulge kinematics are closely correlated in both targets, which is consistent with both possibilities but may favor the collisional ring galaxy scenario.

M dwarfs are small stars with masses between 10 and 50% that of the Sun. They are also the most common, making up about 70% of the stars in our galaxy, and are noted for their high levels of magnetic activity. Stellar flares are a dramatic manifestation of this activity, and occur frequently on M dwarfs. Flares are explosive events resulting from rearranging magnetic fields on the stellar surface. We used data from the Kepler satellite to search for flares from a binary M dwarf system, named GJ 1245. With more than 8 months of continuous imaging, taken every 1-minute, we detected over 3,000 flare events from these twin stars. We conducted a by-eye search of this data to validate our automatic flare-finding code. We present the energy distribution and morphology for flares on these stars, and compare them to other well studied flare stars from Kepler. The long term goal is to train the automatic flare-finding code to run without the need for human validation, and thus to study tens of thousands of stars. By studying the statistical properties of flares on many M dwarfs, we will further our understanding of stellar magnetic activity and how flares are affected by the internal magnetic dynamo.