Abstract: In the first part of this talk, I will present the first CMB constraints on direction-dependent cosmological birefringence, derived from WMAP-7 data. Cosmological birefringence is a postulated rotation of the linear polarization of photons that arises due to a Chern-Simons coupling of a new scalar field to electro- magnetism. In particular, it appears as a generic feature of simple quintessence models for Dark Energy, and therefore, should it be detected, would provide insight into the nature of cosmic acceleration. Prior work has sought this rotation, assuming the rotation angle to be uniform across the sky, by looking for the parity-violating TB and EB correlations a uniform rotation produces in the CMB temperature/polarization. However, if the scalar field that gives rise to cosmological birefringence has spatial fluctuations, then the rotation angle may vary across the sky. I will report the first constraint on the rotation-angle power spectrum for multipoles up to L = 512 and discuss implications for a scale-invariant model. In the second part of the talk, I will explain how the same formalism and analysis method can be used to probe patchy screening from the epoch of cosmic reionization, and present preliminary results. I will conclude by proposing a simple parametrization of patchy screening that can be used as a figure of merit for future CMB experiments.

 Abstract: We have identified a large population of dust-reddened quasars by matching radio sources detected in the FIRST survey to the 2MASS near-infrared catalog (F2M) and selecting sources with red optical-to-near-infrared colors. We followed up our candidates with optical and/or near-infrared spectroscopy and have identified ~120 dust-reddened quasars, defined as having at least one broad emission line in and a reddening of E(B−V) > 0.1. The sample spans a wide redshift range, 0.1 < z < 3 and reaches a reddening, E(B−V) = 1.5. When corrected for extinction, red quasars are the most luminous objects at every redshift and their fraction increases with luminosity. The properties of red quasars suggest that they are revealing an evolutionary phase where the heavily obscured quasar is emerging from its dusty environment prior to becoming a "normal" blue quasar. We compute the fraction of quasars that are in this red phase and determine that its duration is ~20% as long as the unobscured quasars phase: a few million years. I will also report on an expansion of this sample to fainter flux limits with UKIDSS, where we discover reddened quasars at z>2. And I will present future work on these object to better understand their role in the co-evolution of quasars and galaxies.

 Abstract: I will talk about two ongoing cosmological revolutions that are fueled by high angular resolution and high sensitivity observations of the cosmic microwave and infrared backgrounds: constraining the properties of the first galaxies in the Universe (the epoch of reionization) via the kinetic Sunyaev-Zel'dovich effect, and constraining astroparticle physics (neutrino masses) and gravitational physics (dark energy and matter) through gravitational lensing. I will report on the status and prospects of these endeavors, and discuss how they complement studies of the Ly-alpha forest and highly redshifted 21 cm radiation, as well as particle physics experiments.

 Abstract: Mapping the intensity of the Cosmic Microwave Background (CMB) has had a transformative impact on our understanding of cosmology, and has enabled precise measurement of several of the large-scale parameters that describe our universe. CMB polarization has the potential to significantly increase our understanding of the formation of gravitational structure, and to provide much-needed observational insight into the very early universe, probing physics at energy scales unavailable in any other setting. Extracting this information from CMB polarization requires a new generation of instruments with significantly increased raw sensitivity and refined mitigation of non-cosmological signals. A leading architecture for such instruments is that of high-throughput telescopes coupled to arrays of lenslet-coupled microfabricated bolometric detectors. As part of my Ph.D. research at UC Berkeley, I designed and fabricated the first array of lenslet-coupled detectors to observe the CMB, the focal plane array for the Polarbear instrument. With our collaborators, UC Berkeley and UC San Diego deployed POLARBEAR to Chile last year, and it is currently carrying out routine cosmological observations. We have also been developing a new detector technology: multi-band detectors which provide the higher sensitivity and spectral information that is necessary for the next generation of CMB polarization instruments. I will discuss POLARBEAR’s current CMB polarization observations, the detector array that is enabling these measurements, and the multi-band detectors that will be deployed on the next generation of CMB polarization instruments.

 Abstract: I will present recent results on characterizing the infrared spectral energy distributions (SEDs) of 191 mid-IR selected z~0.3-3.0 and L_IR~10^11-10^13Lsun galaxies, and study how their SEDs differ from those of local and high-z analogs. Infrared SEDs depend both on the power source (AGN or star-formation) and the dust distribution. Therefore, differences in the SEDs of high-z and local galaxies provide clues as to differences in their physical conditions. I will show that there is strong evolution in the SEDs between local and z~2 IR-luminous galaxies, as well as that there is a wide range of SEDs among high redshift IR-luminous sources. Lastly, I will discuss possible explanations for this SED evolution as revealed in the morphology (based on HST NICMOS images) of our sources, as well as based on theoretical SED models (based on GADGET+SUNRISE simulations).

 Abstract: It has been known for some time that the center of the Milky Way Galaxy would be the most luminous source of gamma-ray photons from dark matter annihilation. Our recent work shows that observations by the Fermi Gamma-ray Space Telescope reveal that Galactic Center (GC) has a source with large luminosity and high statistical significance that is consistent with extended emission from dark matter annihilation in three ways: (1) its spatial morphology is consistent with that expected from numerical simulations; 2) its luminosity is consistent with the expected thermal-production dark matter annihilation rate; and (3) its energy spectrum is consistent with annihilation into standard model quark channels for weak-scale dark matter particle masses. The very crowded region of the GC also harbors a high density of other astrophysical high-energy sources, including millisecond pulsars, that could mimic this signal. I will discuss the details of the observation, its several interpretations, its implications, and the potential future resolution of the nature of this source.

 Abstract: We use a large spectroscopic sample of L* galaxies at redshifts 1.5<z<3.4 with Keck, Hubble, and Spitzer observations, to study the average star formation history of galaxies at z~2. We first perform a detailed comparison between the bolometric SFRs of z~2 galaxies and those obtained from SED-fitting. This comparison suggests that exponentially-declining models are, in general, a poor representation of the star formation histories at z~2. Taking into account a number of systematic biases, we find a near unity relationship between SFR and stellar mass for z~2 galaxies. The median specific SFR of our sample, when placed in context with the median specific SFRs found for higher redshift dropout samples, suggests a scenario where SFRs rise roughly exponentially with time. We find that the net cold gas accretion rate, as inferred from the specific SFR and the K-S relation, is typically 2-3 times larger than the SFR at z>4. However, if we evolve to higher redshift the star formation histories and the masses of the halos that are expected to host L* galaxies at z~2, then we find that <10% of the baryons accreted onto typical halos at z>4 actually contribute to star formation at those epochs. These results highlight the relative inefficiency of star formation even at early cosmic times when galaxies were first assembling.

 Abstract: What a galaxy has done with its metals is a direct tracer of its history of star formation, gas flows, and feedback processes. I will show that a simple model combining empirical star formation histories with the local relation between stellar mass, gas-phase metallicities, and star formation rates reproduces the metal distributions of z=0 galaxies remarkably well. The bulk of metals released by supernovae and AGB stars, however, are no longer in galaxies, and are instead in the circumgalactic medium (CGM) or intergalactic medium. The COS-Halos survey has created a statistically-sampled map of the gaseous CGM of low redshift ~L* galaxies out to impact parameters ~150 kpc using the Cosmic Origins Spectrograph (COS) on the Hubble Space Telescope. Using this map, I will show that the masses of metals found in the cool (10^4 < T < 10^5 K) photoionized and the more highly ionized OVI-traced CGM could potentially account for all of the metals expelled from ~L* galaxies at z~0.

 Abstract: In analytic and semi-analytic models, it is generally assumed that the brightest or most massive galaxy in a dark matter halo is the central galaxy, and that it resides at rest at the center of the potential well. It is also assumed that the mass (and concentration) of a halo determines the properties and evolution of the galaxies it hosts, perhaps with some scatter in the relation between them. Using models, mock catalogs, and SDSS catalogs, I test these assumptions and argue that they break down in some regimes. This implies that a more nuanced view of the 'halo-galaxy connection' is required and that the interpretation of results of such models should be treated with caution. In particular, I show that the brightest galaxy in a halo is often a satellite galaxy, that halo mass only to some extent determines central (not satellite) galaxy properties, and that mass is not a dominant property even of dynamically relaxed halos.

 Abstract: The coming decade is going to be very exciting for gravitational lensing science with large surveys like the Dark Energy Survey (DES), the Subaru Hyper-suprime Cam (HSC), the Large Synaptic Survey Telescope (LSST) and the space based Euclid mission returning data with unprecedented precision. Lensing directly probes the distribution of matter through the bending of light due to gravity. Massive galaxy clusters produce lensing of background sources on all scales ranging from weak distortions, change of number density due to magnification to giant arcs and multiple images near their cores. Lensing masses are important in calibrating the mass-observable (X-ray temperature or SZ scaling) relation. I will highlight an ongoing cluster lensing study involving a sample of 20 supermassive clusters where we are using the information from shear and magnification to calibrate the normalization and scatter in the mass-temperature relationship of clusters. Such a calibration will be a valuable input to deriving cosmological constraints from cluster number counts. I will also discuss systematics involved using magnification for mass calibrations from stacked cluster analysis using data from surveys.

 Abstract: Over the last decade hydrodynamical simulations of galaxy formation advanced significantly. I will present recent results using a new hydro scheme based on a hybrid approach combining classical Lagrangian and Eulerian methods. Specifically, I will focus on differences between these classical schemes and the new hydro method demonstrating that the global gas structure, the global star formation history, properties of galaxies and the IGM gas supply to galaxies are significantly changed compared to previous calculations. Most dramatically, galaxy gas accretion is altered leading to a significant increase of hot mode accretion compared to previous predictions.

 Abstract: Over the past decade, one of the most concerning issues with the LCDM model has been the "missing satellites" problem: although LCDM unavoidably predicts that the Milky Way should host a vast population of dark matter satellites, only a dozen or so satellite galaxies had been observed. While the discovery of a new population of ultra-faint dwarf galaxies has been taken as an encouraging sign of reconciliation between theory and observation, improved understanding of the structure of Milky Way satellites has revealed an additional problem: the most massive dark matter satellites of the Milky Way predicted in LCDM are too dense to host any of the Milky Way's bright dwarf galaxies. I will discuss various ways to explain this surprising result, which include the possibility that galaxy formation is markedly different in low-mass dark matter halos than in larger systems, or that the nature of dark matter differs from the LCDM standard of a cold, weakly interacting massive particle. I will also explore connections to another potential "crisis" for the LCDM model in the Milky Way, the missing baryons problem.

 Abstract: We present a robust, self-consistent, semi-empirical model which matches observed galaxy stellar masses and star formation rates over nearly all of cosmic history. Key results include constraints on star formation rates as a function of halo mass, average star formation histories for galaxies from z=0 to z=8, and the buildup of the intracluster light as a function of halo mass and time. We also examine the connections between halo growth and galaxy growth, and we find that the universe's star formation history since z~4 can be explained by a star formation efficiency which is strongly dependent on mass but only weakly dependent on time. We discuss the implications of this latter finding, especially in terms of the impact on black hole/supernova feedback models and the impact on the shape of the cosmic star formation rate for 0<z<8.

 Abstract: Early-type galaxies are generally considered to be old and passive systems, having no current star formation (SF). In this talk I show how UV observations have revealed that a significant fraction of field galaxies on the red sequence exhibit moderate levels of galaxy-scale SF, up to a solar mass per year. Are these recent arrivals on the red sequence that retained some SF and may be responsible for its purported buildup since z~1? Alternatively, are these galaxies rejuvenated, in which case is the SF episodic, related to merging events, or continuous? Our HST and ground-based observations of a sample of low redshift UV-excess early types shows that the extended SF is only found in galaxies with old disks (S0s), but not in true ellipticals, regardless of the galaxy mass. The dichotomy suggests that SF quenching may have different effectiveness in the two Hubble types. Regular morphology of SF regions in our sample is indicative of long-lived low-level accretion of gas from the IGM. Minor gas-rich mergers do not appear to be the dominant drivers of SF in early types. These new observations point towards a picture in which significant fraction of field S0s continue to grow their outer disks, while ellipticals remain truly passive. A view of recent buildup of the red sequence through post-merger quenching is disfavored.

 Abstract: Gamma rays probe the most energetic processes in the universe and are a promising tool to search for signatures of new physics. One current mystery in high-energy astrophysics is the origin of the diffuse gamma-ray background. The contribution of undetected sources is expected to induce small-scale anisotropies in this emission, and these may provide a means of identifying and constraining the properties of its contributors. I will review the results of the first anisotropy analysis of the diffuse gamma-ray background measured by the Fermi Large Area Telescope, and highlight the new constraints this measurement has placed on high-energy source populations. I will focus in particular on implications for blazar population models and for a signal from the annihilation or decay of dark matter particles. I will conclude by discussing new multi-wavelength techniques using anisotropy to identify contributors to diffuse emission.

 Abstract: The formation of stars is a fundamental astrophysical process; and yet we still debate whether it varies with environment. Milky Way young star clusters range in mass over four orders of magnitude; but, the best-studied star forming regions, such as Taurus and Orion, represent only a small range of initial conditions at the low mass end. Young star clusters with masses greater than 10,000 solar masses are promising targets for determining whether the initial mass function (IMF) that results from the star formation process is universal or depends on environment. I present results from a Keck adaptive optics and HST study of several massive young star clusters in the Milky Way, including around the supermassive black hole at the Galactic Center. Precise IMFs are constructed by using high-precision astrometry and spectroscopy to distinguish individual cluster stars. We find a moderately top-heavy IMF in the Galactic center and compare this to similarly massive clusters in the Milky Way disk, such as Westerlund 1, and to the "universal" IMF found locally.

 Abstract: The favored dark energy plus cold dark matter (LCDM) model of cosmology predicts that the Milky Way should be surrounded by thousands of dark matter satellite clumps, in great excess of the observed count of Galactic dwarf satellite galaxies. This mismatch is known as the "missing satellite problem". Recent discoveries in theSloan Digital Sky Survey have revealed a new population of ultra-faint dwarf satellites, motivating excitement within the community that some "missing" LCDM satellites are finally being found. Unfortunately for the theory, the situation is not quite so rosy once the dynamical masses of the known satellites are considered. Specifically, the majority of the most massive dark matter satellites predicted to exist are too dense to host any of the bright satellite galaxies of the Milky Way. This poses a serious puzzle theoretically: either galaxy formation becomes effectively stochastic on scales smaller than ~0.1L* or the central densities of dark matter subhalos are significantly lower than predicted in dissipationless simulations. I discuss some possible solutions to this puzzle from the standpoint of baryonic physics and non-standard dark matter physics.

 Abstract: After a decade of near consensus that planets grow bottom-up from colliding dust particles (“core accretion”), observations in 2008 revealed a new population of extremely massive planets on wide orbits that posed significant problems for the core accretion model. Mathematical models of growing planets demonstrate that such planets must instead form in the direct collapse of a protostellar disk (“disk instability”). With at least two modes of planet growth operating in the Galaxy, we now want to establish the relative importance of each mode in building the overall planet population. I first discuss the possibility that magnetically dead zones allow disk instability to form planets on short-period orbits, not just wide orbits. I then describe a new strategy for discovering objects that formed by disk instability. The picture emerging from my group’s recent research is that disk instability is a rare formation pathway for both planets and stars.

 Abstract: Both balloon-borne and ground-based experiments located in Antarctica allow us to use our universe as a laboratory for fundamental physics. I will first discuss experiments that search for the highest energy neutrinos and cosmic rays to probe the nature of the ultra-high energy universe in a unique way and test our understanding of particle physics at energies much greater than those achievable at particle colliders. The best limit to date on the flux of ultra-high energy neutrinos comes from the ANITA experiment, a balloon-borne radio telescope designed to detect coherent radio Cherenkov emission from cosmogenic ultra-high energy neutrinos. The future of ultra-high energy neutrino detection lies with ground-based radio arrays, which would represent an enormous leap in sensitivity. I will then discuss experiments that image the Cosmic Microwave Background (CMB) to search for a signature of Inflation, the predicted superluminal expansion of the universe during the first moments after the Big Bang. The Keck Array, currently beginning its third observing season at the South Pole, is a suite of five microwave polarimeters that observe the CMB at degree angular scales to specifically search for this signature of Inflation. The Keck Array is well-positioned to be the first experiment to actually detect this signature of Inflation, testing models of Inflation at the GUT scale. In the next decade, precision measurements of CMB polarization promise to probe a wide variety of fundamental physics, including the energy scale of Inflation, dark energy in the early universe, and the sum of the neutrino masses.

 Abstract: We know that most stars, if not all, were once surrounded by protoplanetary disks. How these young disks evolve into planetary systems is a fundamental question in astronomy. It is widely accepted that dust grain growth is the first step in creating the planetesimals that amalgamate into planets. Theoretical simulations then predict that a young, forming planet will clear the material around itself, leaving behind an observational signature in the form of a gap in the disk. I have identified a new class of disks that have gaps using Spitzer and ground-based infrared observations. These gapped disks are the strongest evidence for disk clearing by planets, making these disks the precursors of the exoplanet systems that have been detected to date. I will present this evidence and outline current and future research with Herschel, ALMA, and the next generation of telescopes, aiming to discover the earliest signs of planet formation in large populations of young disks. These observations will provide a far more complete and quantitative view of protoplanetary systems in the first million years after a star is formed.

 Abstract: The dwarf galaxies in the Local Group are a rich playground for exploring galactic and chemical evolution. I will show that resolved stellar spectroscopy of these galaxies with Keck/DEIMOS has revolutionized our perception of how galaxies process gas and generate metals. In particular, galaxies of all morphological types--contrary to common wisdom--obey a universal stellar mass-metallicity relationship. I will include some of the smallest galaxies, including Segue 2, with a luminosity of only 900 L_sun. I will show that it is the least massive galaxy known, with a stellar velocity dispersion less than 2 km/s. Despite its tiny mass, Segue 2's chemical composition identifies it as a galaxy, not an unusual globular cluster, like NGC 2419.

 Abstract: In the past decade, photometric and spectroscopic galaxy surveys have started to explore the Universe at redshifts z=1-3, the epoch during which galaxies and black holes appear to grow most rapidly. It was found that the galaxy population at this early epoch is quite diverse, and ranges from starburst galaxies with large and irregular morphologies to very small galaxies with quiescent stellar populations. However, despite this tremendous progress, these studies are hampered by the small sizes of spectroscopic galaxy samples, whereas much larger photometric samples lack the required spectroscopic information. I will discuss two observing programs which compromise between these two techniques: the 3D-HST survey, a grism survey with HST/WFC3 and the NEWFIRM Medium-Band Survey. I will show how these studies have contributed to our understanding of the formation histories of massive galaxies. Furthermore, I will present new stellar kinematic measurements of distant galaxies obtained using deep near-infrared spectroscopy, and discuss the implications for the structural evolution of galaxies.

 Abstract: For a particle physicist, stellar interiors represent extremely hermetic detectors, sensitive to very rare processes. For example, it is well known that neutrino emission, despite the weakness of the relevant interaction (in fact, because of it!), dominates the advanced stages of stellar evolution. The same argument extends to new physics scenarios with light, weakly interacting particles, of which the axion is a prime example. But how does one know what type of stars to look at? As will be shown in the talk, the answer to this question is less than trivial. In particular, massive stars, previously thought to be useless for particle physics, turn out to be the most sensitive known probes of the axion-photon coupling. I will also mention what massive stars can tell us about the neutrino magnetic moments.

 Abstract: The gaseous environments of galaxies are a crucial but poorly-constrained component of galaxy formation and evolution. The circumgalactic medium (CGM) is the principal reservoir for future gas accretion, and its kinematics and metal enrichment provide vital constraints on the physical properties of galaxy-scale outflows. Characterization of this gas therefore presents a unique window into baryonic flows that are expected to profoundly influence galaxy evolution. I will present results from the Keck Baryonic Structure Survey (KBSS), a unique spectroscopic survey designed to explore the connection between galaxies and intergalactic baryons. The KBSS is optimized to trace the cosmic peak of star formation (z~2-3), combining Keck/HIRES spectra of 15 hyperluminous QSOs with densely-sampled galaxy redshift surveys surrounding each QSO sightline. I will characterize the physical properties of the CGM gas through the spatial distribution, column densities, kinematics, and absorber line widths of ~6000 HI absorbers surrounding ~900 foreground star-forming galaxies within 50 kpc to 3 Mpc of a sightline. These measurements provide clear evidence of gas inflow and outflow as well as accretion shocks or hot outflows from these forming galaxies. I will compare these observations with recent theoretical predictions, highlighting discrepancies that suggest our theoretical picture of gas flows into and out of galaxies is still incomplete.

 Abstract: The Pleiades is the best studied open cluster in the sky. It is one of the primary open clusters used to define the "Zero Age Main Sequence", and hence it serves as a cornerstone for programs which use main-sequence fitting to derive distances. This role is called into question by the "Pleiades distance controversy" - the distance to the Pleiades from Hipparcos of about 120 pc is significantly different from the distance of 133 pc derived from other techniques. Although this amounts to a 10% difference in the distance, the resultant discrepancies as propagated into the Pleiades HR-diagram, and the necessary revisions of physical models to obtain agreement with the Hipparcos result, are quite significant. To resolve this issue we are carrying out a VLBI program to derive a new, independent trigonometric parallax distance to the Pleiades.

 Abstract: Our understanding of galaxy evolution centers around questions of how gas gets into galaxies, how it participates in star formation and black hole growth, and how it is returned to its galactic surroundings via feedback. I will present observational results on the relationship between gas that forms stars and gas that accretes onto supermassive black holes, and the nature of feedback that is capable of removing gas from galaxies. These results have important implications for how radiation, momentum, and kinetic energy from stars and black holes regulate the cold gas supply in galaxies. I will also discuss prospects for characterizing the physical properties of gaseous outflows and inflows using existing and future facilities.

 Abstract: Understanding the physics of galaxy formation and evolution is a primary focus of current theoretical and observational studies in astrophysics. I will first review the successes and limitations of galaxy formation models in a LambdaCDM universe. I will then discuss the physics of feedback from stars and black holes, which presently limits the predictive power of models. In particular, I will present new physical models of feedback-regulated star formation in galaxies and of black hole-driven galactic winds. I will conclude by summarizing ongoing efforts to significantly improve the physical realism of cosmological simulations of galaxy formation by explicitly resolving feedback processes.

 Abstract: First Stars are the first astrophysical objects born in the primordial environment in the universe as early as at redshift~40. The growth of First Stars has been simulated extensively by small-box, high-resolution hydrodynamic simulations and it was shown that they grow out of hydrogen-molecule cooling, are relatively massive (>~40-300 Msun), form very top-heavy initial mass functions, and produce copious amount of hydrogen- and helium-ionizing photons. While they are thus believed to have marked the ending of the Dark Ages and the beginning of the epoch of cosmic reionization (EOR), their impact has been usually neglected in large-box numerical simulations of EOR mainly due to the computational inefficiency to cover the wide dynamical range requied. I present a recent result which overcomes this difficulty and covers the full dynamic range of radiation sources in a large-box (~160 Mpc) simulation, which shows a significant impact of the First Stars on EOR and its observables. As for near-future possibility of observing them, I present forecast on CMB E-mode polarization observation by Planck due by 2014 based on the simulation result, and show that the existence of First Stars can be probed at 1-2 sigma confidence level. I also present forecast on 21-cm observations from EOR aimed by pathfinders of Square Kilometre Array (SKA) and SKA itself.

 Abstract: The populations of planets revealed by the Kepler space telescope have provided several new clues to how planets form throughout the Galaxy. The vast majority of exoplanetary systems discovered thus far are more compact than the Solar System and preferentially contain planets with radii 1-4 times the Earth; a size range absent from our family of familiar planets. Our group based at Caltech has undertaken a comprehensive program to understand the planet populations around the smallest and most numerous stars in the Galaxy using data from the Kepler space telescope augmented with ground based followup bservations. I will share some of our latest results including observational constraints on planet formation mechanisms, and a statistical analysis of planet frequency. I will then turn to address the implications of these findings, and present a new robotic, cost effective observatory under development called Minerva that will discover some of the nearest exoplanetary systems.