 |
||
|
AbstractsOral Presentations Galaxy Formation & Dynamics
Galaxy Formation & Dynamics INVITED SPEAKER: Fabio Governato University of Washington Galaxy Formation Ain’t Like Dusting Crops, Boy! Abstract is coming University of Toronto Tidal Interactions in the Local Group: M33 Flies by M31
The Pan-Andromeda Archaeological Survey (PAndAS) is producing an unprecedented view of the stellar halos of M31 and M33. The survey has discovered new dwarf satellites of M31, an abundance of stellar shells and streams that are undoubtedly the result of past tidal interactions and clear signs of the tidal distortion of the disk of M33. These new observations bolster the idea of a more complex dynamical evolution of the Local Group in the recent past. University of Vienna
Modelling Merging Galaxies Using MINGA State-of-the art simulations of interacting galaxies use complex multi-component methods and require enormous computational power. Therefore, one needs to know the correct initial conditions for remodeling observed systems in advance. Some of these parameters might be hidden in detailed observations, e.g. in HI data cubes. However, to perform an effective search in such a high dimensional parameter space, it is necessary to use fast simulations and sophisticated search strategies. Within our code MINGA we coupled a Genetic Algorithm with a restricted N-body code (Theis, 1999). Our GA is based on a version of PIKAIA (Charbonneau, 1995) and allows for cross-over mutation of the fittest models. The N-body code is a modified version of the approach suggested by Toomre & Toomre (1972), i.e. it allows for dark matter halos. The code allows to run a single simulation in about one CPU second on a modern machine. MINGA was successfully applied to recover the orbital and halo parameters of NGC4449 (Theis & Kohle, 2001) and M51 (Theis & Spinneker, 2003). Recently we also implemented dynamical friction in order to remodel later stages of interaction and mergers. By comparing the restricted N-body code with a self-consistent tree-code we found, that a mass- and distance-dependent Coulomb logarithm should be used together with the dynamical friction approach by Chandrasekhar (1943). Especially for high mass ratios between satellite and host galaxy, an orientation correction of the frictional force needs to be applied in order to account for density gradients of the halos.
David Williamson Saint Mary’s University Galaxy Formation – A Detailed Examination of Cloud-Cloud Interactions
The details of the processes involved in the formation of a galaxy from an collapsing cloud of gas that has collected within a dark matter halo are not well understood. The effects of phenomena such as viscosity are considered to be important but thus far have largely been treated by phenomenological models. Such models frequently invoke assumptions about the underlying geometry, something that changes rapidly during the formation of galaxies. Hence testing these predictions via full 3d hydrodynamic simulations is a key step in the validation of these models.
McMaster University Star Formation in Cosmological SPH Simulations of Galaxy Formation
In order to relate cosmological simulations of galaxies to observations, it is necessary to have an accurate physical representation of how stars form. Two developments are affecting the way star formation will be modeled. First, simulations run over the past decade have based star formation on the observations of Kennicutt (1998). These model the star formation rate as depending on gas mass and the gas free fall time together with an imposed density threshold. Recent, higher resolution observations (Kennicutt et al. 2007; Bigiel et al. 2008) have challenged this paradigm. Second, cosmological simulations are beginning to resolve gas below masses of 1e5 M_sun - moving into the regime of molecular clouds. Modelling cooling and collapse on these scales requires consideration of metal and molecular hydrogen cooling physics. We describe the implementation of these effects in the new McMaster Unbiassed Galaxy Survey (MUGS). McMaster University
IGM Enrichment with Feedback and Turbulent Diffusion We present a study of the chemical enrichment of the intergalactic medium (IGM) using a series of smooth particle hydrodynamic (SPH) simulations. We implemented a self-consistent metal cooling model and a turbulent diffusion model for metals and thermal energy. Our basic star formation and feedback model naturally results in wind generation that is sufficient to reproduce the observed cosmic star formation history (SFH) and the universal neutral hydrogen fraction. This challenges the view that special wind recipes are essential for correct star formation but leaves room for stronger feedback to play a role in the distribution of metals.
At z=0, about 40% of the baryons are in the warm-hot intergalactic medium (WHIM), but most metals (82%) are locked in stars. At higher redshift the fraction of metals in the IGM was higher due to more efficient enrichment. IGM metals primarily reside in the WHIM over the whole redshift range (differing from the momentum driven wind model of Oppenheimer and Davé 2006). The metallicity of the WHIM lies between 0.01 and 0.1 solar with a slight decrease at lower redshift.
The wind escape efficiency decreases with galaxy mass, but the efficiency of star formation increases and peaks around 1011 solar masses. This increases the metal content of the galactic winds. Beyond 1012 solar masses star formation declines following the classic Rees and Ostriker (1977) picture and the metal content of the objects saturates. Metal cooling transfers metals from the WHIM to bound gas, and increases the SFR. Metal diffusion allows winds to mix prior to escape, thus decreasing the IGM metal content in favour of gas within galactic halos and star forming gas. Diffusion significantly increases the amount of low metallicity gas and affects the density-metallicity relation.
Saint Mary’s University
Surface Brightness Profiles of Simulated Galaxies: Comparison to Stacked HUDF Images at z=6 We compare the surface brightness profiles of observed and simulated galaxies at z~6. The goal of the study is to use these comparisons to constrain feedback models in galaxy formation, such that they produce accurate depictions of internal galaxy properties such as size, dynamics and extinction. The observational data is taken from the stacked HUDF images analysed by Hathi et al. (2008), while our new simulations were performed with a parallel version of the HYDRA code. Detailed comparisons were enabled by the creation of a mock image pipeline which includes radiative transfer effects, such as full internal dust absorption & scattering. We show the effect of several feedback prescriptions on the average surface brightness profile slope and width, and discuss our attempts to explain the break from the best fit inner profiles at larger radii seen in the stacked observational composites.
INVITED SPEAKER: Richard Klein Lawrence Livermore National Laboratories Feedback Effects in the Formation of High Mass and Low Mass Star Formation
The formation of massive stars remains one of the most significant unsolved problems in astrophysics, with implications for the formation of the elements and the structure and evolution of galaxies. It is these stars, with masses greater than 8-10 solar masses, that eventually explode as supernovae and produce most of the heavy elements in the universe, dominate the energy injection into the interstellar medium of galaxies and by injecting both heavy elements and energy into the surrounding medium, shape the evolution of galaxies. Despite the importance of massive star formation, relatively little is known about them theoretically as they pose a major theoretical challenge: How is it possible to sustain a sufficiently high mass accretion rate into a protostellar core despite the radiation pressure on the accreting envelope? I will discuss our work on the first 3D simulations of massive star formation. Using our high resolution 3D radiation-hydrodynamic adaptive mesh refinement code ORION with a v/c correct treatment of the radiation transport, we have investigated the formation of high mass stars from both smooth and turbulent initial conditions in the collapsing massive core. I discuss our work on identifying 2 new mechanisms that efficiently solve the problem of the Eddington barrier to high mass star formation; the presence of 3D Rayleigh Taylor instabilities in radiation driven bubbles present in the accreting envelope and the feedback due to protostellar outflows providing radiation an escape mechanism from the accreting envelope in addition to the feedback from protostellar radiation and its affect on stellar multiplicity. I will present predictions for upcoming EVLA and ALMA submillimeter observations. I also discuss the effects of radiative transfer on low mass star formation in a turbulent molecular cloud. I will compare the distribution of stellar masses, accretion rates, and temperatures in the cases with and without radiative transfer, and demonstrate that radiative feedback has a profound effect on accretion, multiplicity, and mass by reducing the number of stars formed and the total rate at which gas turns into stars. Calculations that omit radiative feedback from protostars significantly underestimate the gas temperature and the strength of this effect.
University of Rochester
AstroBEAR: The Opportunities, Costs and Challenges of Multi-Physics AMR In this talk I review new research using AstroBEAR, a parallel AMR multi-physics code (MHD, microphysics, self-gravity, diffusion) for astrophysics that has been used extensively for studies of high mach number flows. After a short overview of research applications (Outflow Driven Turbulence, Evolved Star Nebula, Laboratory Astrophysics, Accretion Disk Studies, Cosmic Ray Shocks) I report new results on the dynamics of heterogeneous flows. New studies of strongly magnetized radiative shocks interacting with extended clumpy media are presented as well as 3-D studies of shocks interacting with internally magnetized clumps (no field in the ambient medium). These results are presented to demonstrate code capabilities and to introduce/discuss strengths and weaknesses of numerical methods developed for AstroBEAR. The talk will also review the development history of the project to illustrate the challenges facing multi-year, multi-student/post-doc astrophysical code projects. As is well known, gone are the days when a single student constructed their own code for their thesis project. Paralleling developments in other domains of science, state-of-the-art astrophysical codes now require extensive collaborative efforts that span years and disciplines (astrophysics, applied math, computer science). The health of the field requires many players experimenting with a variety of approaches and in this talk I will raise issues related to how a smaller group such as that at the University of Rochester can manage challenges such as code complexity/propagation, student training, "active" documentation, and maintaining collaboration networks.
University of Rochester
Clumpy Flows: Quasi-Ballistic Model of HH Objects We present 3D hydrodynamic simulations of Herbig-Haro jets which yield a unique explanation of observed morphologies. Insight is actively being sought as to whether the heterogeneity of an HH object jet beam can be explained best by interactions of the jet with its environment or by the jet's launching mechanism. Both explanations have received extensive analytic and numerical attention. The present work focuses on nonuniformity in the initial launching which permits the imbedding of dense "clumps'' in an otherwise smooth jet beam. The clumps have variable size, velocity, and position within the jet beam. They are seen to have significant effects on the jet morphology, including disruption of the jet head and the formation of low-density "cavities'' in the jet beam which might otherwise be explained by, say, working surfaces in a pulsed jet. Recent epochal HST images are especially relevant to our investigation for a variety of reasons. For example, in a typical jet-pulsation model, there is little reason to suspect large variations in the velocity dispersions and chemical composition of successive working surfaces. The present model provides a natural explanation for such variations perpendicular to the jet beam at a "knot'' and allows a unique momentum transport mechanism relevant to overall jet morphologies. This work is supported by the Laboratory for Laser Energetics at the University of Rochester.
Jon Ramsey and David A. Clarke Saint Mary’s University Simulations of Protostellar Jets Extended to Observational Length Scales
Jets and outflows are an integral part of the current star formation paradigm, and are observed any place star formation is ongoing. The importance of these objects lies in their ability to remove angular momentum from matter accreting onto a central mass. While observations easily resolve these objects on scales down to several hundred AU, the launching and collimating mechanism (which is 10 AU in size) remains unresolved and is often obscured by a dusty disc. In recent years, there have been numerous simulations of the jet launching mechanism, but these studies do not extend to spatial scales resolvable by existing observations. With such a large disparity between scales, it is clear there is a disconnect between the simulation and observation of protostellar jets.
In this talk, I will present results of simulations with AZEuS, an adaptive mesh refinement version of the ZEUS-3D fluid code. AZEuS allows one to model simultaneously both the small jet launching region and the much larger observational length scales. I will highlight results made possible with adaptive mesh refinement, how the models compare to observed protostellar jets, and where this project is going in the near future.
McMaster University
Radiative Transfer in 3D Simulations of Star And Planet Formation Radiative transfer (RT) plays a fundamental role in both star formation and gas-giant planet formation via gravitational fragmentation. In the former, recent numerical results have shown that the inclusion of RT effectively suppresses the production of low-mass fragments, thus altering the low-mass IMF as well as the overall star formation efficiency. In the latter, recent numerical, and analytic, results have shown that the likelihood of a protostellar disk to become gravitationally unstable and fragment, producing the possible progenitors of gas-giant planets, is highly dependent on the disk’s ability to radiatively cool. Thus, accurate numerical simulations of these processes must include RT.
In this light, I will discuss our implementation of RT in the flux-limited diffusion approximation in the Gasoline smoothed-particle hydrodynamics (SPH) code. This will include an overview of our algorithm, test cases, and preliminary results focusing on the stability of protostellar disks to gravitational fragmentation.
Saint Mary’s University
Numerical Hydrodynamics Modeling of Protostellar Disk Formation and Long-term Evolution I present a numerical hydrodynamics code that is suitable for studying the evolution of self-consistently formed protostellar disks on time scales of order 3 Myr. The code solves the basic equations of hydrodynamics in the thin-disk approximation on the polar grid and the disk vertical structure is followed approximately via the assumption of the local vertical hydrostatic equilibrium. The effects of turbulent viscosity are simulated using the usual alpha- prescription of Shakura & Sunyaev. The gravitational potential is found by solving the Poisson integral using the convolution theorem to speed up the summation. The disk thermodynamics includes radiative cooling from the disk surface, stellar and background irradiation, viscous and shock heating. Forming protoplanets are substituted with accreting point particles, the dynamics of which is followed using the N-body technique. I briefly discuss recent results and possible applications of the code.
University of Toronto
RMHD Simulations of Massive Star Cluster Feedback To understand the roles massive stars play in the inefficiency of star formation, driving turbulence in Giant Molecular Clouds, and puffing up star-burst galaxy discs, we need a better understanding of how the energy from these stars couples to the surrounding material in a variety of environments. I will present analytical results proving that the largely overlooked feedback from stars pre-supernova is significant and possibly critical in understanding the dynamics and evolution of material within galaxies. I will briefly discuss one-dimensional models confirming the analytical results and then conclude with preliminary results of three-dimensional radiative-magneto-hydrodynamic simulations containing several feedback mechanisms from stars.
University of Massachusetts
Astrochemical Evolution of Turbulent Giant Molecular Clouds Contemporary galactic star formation occurs predominantly within gravitationally unstable, cold, dense molecular gas within giant molecular clouds (GMCs). A body of work has highlighted the key role which turbulence plays in the dynamical evolution of GMCs, both by establishing their density and velocity structure, and also in heating the molecular gas. The ability to develop realistic astrochemical models of turbulent GMC models and compare these models against observation therefore promises to both expand and deepen our understanding of the physics of GMCs.
In this talk, we will discuss the development of novel simulation techniques which elucidate the astrochemical evolution of turbulent GMCs. Specifically, we have initiated a research program to develop computational methods which permit a detailed, spatially-resolved treatment of the thermal and chemical evolution in realistic, three-dimensional simulations of turbulent GMCs, with the long-term goal of studying the astrochemical evolution of an extensive suite of large-scale, three-dimensional, turbulent, magnetized GMC simulations. This research will stimulate and support the interpretation of observations obtained with current and future instruments, including those on Hubble, Spitzer, and the James Webb Space Telescope (JWST). By 2013, the JWST Mid-Infrared Instrument (MIRI) promises unprecedented observations which combine high-sensitivity with high-spatial and spectral resolution within GMCs between 5 and 28 microns. Our codes and databases will enable the computation of molecular abundances and synthetic spectra of realistic, three-dimensional, large-scale simulations of turbulent, magnetized giant molecular clouds and facilitate their comparison to observations. This program of research ultimately promises to elucidate essential aspects of the complex process of star formation, including the role of magnetic fields, supersonic turbulence, and dust grains.
Université Laval
The Effect of Gas-Dust Energetics on the Star Formation Process The origin of the initial mass function and the formation of massive stars have been puzzles in the field of star formation for many years. In order to get closer to solving these puzzles, we use simulations on the scale of a cluster to model the effect of young stars on the molecular gas out of which they form. The stars in our simulations heat the surrounding dust. The hot dust couples to the gas, which can also cool via molecular transitions. We study how young stars affect their environment and the subsequent birth of stars. We find that including dust-gas energetics has a significant and realistic effect on the mass function. In particular, we are able to reproduce the high-mass end of the Salpeter IMF. These simulations use a P3M-SPH algorithm with particle splitting, augmented by a semi-analytical treatment of dust heating by stars.
INVITED SPEAKER: Edward Thommes University of Guelph
From Gas Disks to Gas Giants: Simulating the Birth of Planetary Systems
The ensemble of now well over 300 discovered planetary systems displays a wide range of masses, orbits and in multiple systems, dynamical interactions. These represent the endpoint of a complex sequence of events, wherein an entire protostellar disk converts itself into a small number of planetary bodies. Here we present self-consistent numerical simulations of this process, which produce results in agreement with some of the key trends observed in the properties of the exoplanets. Though the typical formation history of a planetary system is highly stochastic, there are nevertheless clear correlations between a system's birth disk and the characteristics of the mature planetary system which ultimately grows from it. Analogues to our own Solar System are naturally accounted for in this picture, as the products of disks just slightly above the giant-planet-forming threshold. However, such outcomes are in the minority, and a "typical" planetary system tends to have significant eccentricities and shorter-period orbits, akin to the discovered exoplanets.
The University of Western Ontario
Interstellar Meteoroids and the Gravitational Slingshot Interstellar meteoroids, solid particles originating outside our Solar System, provide an intriguing potential source of astronomical information. However the identification of a particle as interstellar is not trivial, particularly in the absence of isotopic information. The detection of interstellar meteoroids within Earth's atmosphere or by dust detectors on spacecraft has traditionally relied on a presumed hyperbolic excess velocity. A video meteor station newly-constructed near London Ontario was designed to make velocity measurements of sufficiently high accuracy to distinguish true hyperbolic excess velocities, and thus detect and study interstellar meteoroids if they are present. Yet, hyperbolic meteoroid velocities can be produced within the Solar System itself in small quantities by processes such as the gravitational slingshot. Thus the study of interstellar meteoroids requires an understanding of the internally-produced population of hyperbolic meteoroids in the Solar System. We will present a theoretical study of the population of hyperbolic meteoroids expected at the Earth from the gravitational slingshot effect of the various planets.
David L. Clark and Paul Wiegert University of Western Ontario
Searching for Meteoroid Images in Sky Surveys The Fireball Retrieval on Survey Telescopic Image (FROSTI) project seeks to locate meteoroids on pre-existing sky survey images. Fireball detection systems such as the University of Western Ontario's ASGARD system provide fireball state vector information that may be used to determine a pre-contact trajectory. This trajectory is utilized to search databases of sky survey image descriptions in order to identify serendipitous observations of the impactor within the minutes or hours prior to its entering the atmosphere. Commonly used analytic methods for meteoroid orbit determination proved to be insufficient in modeling meteoroid approach, so a RADAU based gravitational integrator has been developed. Software has also been developed to represent the description of an arbitrary survey image in a consistent survey independent fashion, with survey specific plug-ins periodically updating a centralized image description catalogue. This periodic update includes pre-processing of each image description to support an innovative image search strategy that easily accounts for arbitrary object and observer position and motion. The search algorithm also deals with meteoroid positional probability clouds calculated from initial meteor observation error bars.
INVITED SPEKER: Robert Deupree Saint Mary’s University
Structure, Evolution, and Oscillations of Rotating Star The structure and evolution of rotating stars is one of the more challenging problems in stellar astrophysics. Given a rotation law, the corresponding structure can be computed in a relatively straightforward manner. However, determining the evolution of the rotation law is a significantly greater challenge which requires consideration of the mechanisms which introduce hydrodynamic and secular flow. While this has been done in 1D for many years, it is based on conjectures about the nature of these flows and their consequences and the results only provide some assurance that mixing of processed material to the stellar surface does occur. The present work attempts to follow the angular momentum redistribution during evolution by implicitly solving the conservation equations in 2D. Results during the early evolution show that some mixing does occur beyond the formal boundary of the convective core and that the rotation rate increases from that of the initial distribution close to the rotation axis.
Tohoku University
Magnetic Effects on Stellar Oscillations
Relatively cool magnetic Ap stars show short-period (~6 - 20 min) oscillations. The oscillations are high radial-order p-modes modified by a few kilo Gauss magnetic fields. In the presence of a magnetic field, not only the oscillation frequencies are shifted but also the oscillation amplitude is concentrated toward the magnetic poles. In addition, slow waves are generated, which carries away the oscillation energy. I will talk about such properties and how to calculate frequencies and eigenfunctions of p-modes of magnetized stars.
Chris Cameron, J. M. Matthews, M. S. Cunha, D. B. Guenther Saint Mary’s University
Asteroseismic Tuning of the Magnetic Ap Star HR 1217 HR 1217 is a well known rapidly oscillating Ap (roAp) star that pulsates in high-overtone, near equally spaced, (magneto-) acoustic modes with a period of about 6 minutes. Two global (ground-based) photometric campaigns and 30 days of data from the Canadian MOST satellite yield a number of periodicities that have the potential to probe the internal structure, including the magnetic field, of this star for the first time. We present a grid of more than 50,000 stellar pulsation models having a luminosity and effective temperature range appropriate for HR 1217 that includes a large range of magnetic field strengths (1 - 10 kG). This is the largest grid of stellar pulsation models of any Ap star to date and is critical in the interpretation of the MOST data. We present the properties of the model grid and describe the method used to match these models to the frequencies observed in HR 1217.
Joyce Ann Guzik and David S. Dearborn Los Alamos National Laboratory and Lawrence Livermore National Laboratories
3D Solar Model Using the Djehuty Code We are exploring the utility of the Djehuty code for gaining insight into the dynamics of the solar interior. Djehuty is a Arbitrary Lagrangian Eulerian (ALE) explicit hydrodynamics code developed at Lawrence Livermore National Laboratory that is capable of modeling stars in three dimensions, with emphasis on the stellar interior. Djehuty physics packages include the Eggleton EOS, OPAL and Alexander opacities, radiative and convective energy transport, and thermonuclear energy generation with a full reaction network. Here we present first results exploring solar models in 3D with Djehuty. The interior structure of the model is generated with a 1D evolution code. The model is run in Djehuty until it relaxes and small numerical perturbations dissipate, and convective motions start up and are regularized. For a nonrotating model, we will examine conditions at the convection zone base and the extent of convective overshoot. We will also examine the effect on convection dynamics and angular momentum distribution when slow rotation is introduced. We plan to also determine whether gravity waves are generated by the convection and how deeply they penetrate into the core, and their potential for inducing mixing, affecting angular momentum transport, or modifying the temperature gradient. Finally we will see whether we can detect in the simulations any longer-period oscillations, such as the radial fundamental or low-degree, low-order p-modes. Click here for Dr. Guzik's poster abstract
University of Tennessee Neutrino Powered Explosions: Computing the Death of Massive Stars
At the end of a massive star's evolution, its degenerate iron core collapses forming a proto-neutron star surrounded by an accretion shock. Heating of the layer between the shock and proto-NS by neutrinos in the next several hundred ms drives the shock outward and expels the outer layers of the star as a core-collapse supernova. I will present the latest in 1-, 2-, and 3-D numerical models of stellar collapse and explosion computed with codes using up-to-date neutrino transport and opacities and other physical inputs. These models show that robust explosions are generated from a wide range of progenitors (12-25 solar masses) when multi-dimensional models with multi-group neutrino transport with detailed input physics are computed. I will also discuss numerical issues relating to the present and future of such calculations.
Chris Geroux and Robert Deupree Saint Mary’s University Interior Mass as the Radial Independent Variable in Nonlinear Stellar Pulsation Calculations
Nonlinear hydrodynamic computations of classical variable stars such as RR Lyrae and Delta Cephei stars was one of early great triumphs of (then) high performance computing in astrophysics. One of the last remaining obstacles in this area is the interaction of convection and stellar pulsation. Previous 2D calculations showed how convection quenched pulsation at the red edge of the RR Lyrae gap, and we build on these by extending nonlinear hydrodynamic calculations to 3D. We will use large eddy simulations to model the convective flow. The 2D calculations found that a radial coordinate moving with the horizontal average of the radial velocities at a given radial interface produces a migration of material through the zoning over a large number of periods, and the 2D calculations could not compute full amplitude solutions with adequate resolution near the hydrogen ionization region. We propose to remove this difficulty by computing a radial grid velocity so that the mass between spherical shells is conserved The method of solution and adiabatic test cases are presented.
Saint Mary’s University APEMoST - A Software Package for Bayesian Astronomical Data Analysis The mounting complexity of theoretical models, as well as the increasing quality of observations from new and improved instruments, have recently resulted in a paradigm shift in astronomical data analysis. Rather than rules of thumb and conservative upper error estimates, quantitative, statistically sound methods of comparing observational data and theoretical models are needed. Meanwhile, increasing amounts of high quality data have resulted in a growing demand for automated data analysis. We introduce APEMoST (Automated Parameter Estimation and Model Selection Toolkit), an easily adaptable and highly scalable open-source software package that fulfils these requirements. It performs Bayesian parameter estimation and model comparison using Markov Chain Monte Carlo methods. We explain the underlying concepts of the program and show several applications.
INVITED SPEAKER: Evan Scannapieco Arizona State University The Challenge of Turbulent Mixing in Computational Structure Formation The range of physical scales involved in cosmological structure formation is enormous. Galaxies form out of regions initially spanning several megaparsecs, but are strongly influenced by feedback from stars and black holes, which acts on scales smaller than an AU. The result is a complex, turbulent distribution spanning so many orders of magnitude that it is unlikely to be directly simulated in our lifetimes. Thus a full understanding of structure formation will require the development of subgrid models, which capture the key aspects of chaotic motions that act on unresolved scales that are nevertheless much larger than a particle mean free path. I will focus on two types of cosmological structures that are strongly impacted by unresolved turbulent motions -- cool core galaxy clusters and dwarf starbursting galaxies -- and present simple models that point to new directions for capturing their key features. Developing and refining such tools will play a crucial role in computational astrophysics for years to come. College of Charleston Where Can We Push the Envelope of Simulations of Black Hole Accretion Disks?
In this talk I will discuss recent upgrades to the Cosmos++ computational astrophysics code, including the addition of a fully conservative GRMHD formulation based on the HARM scheme and a constrained transport option for the magnetic fields. This has been coupled with a previously implemented radiative cooling package. Finally, all of this is built to work on any logically Cartesian grid, where the physical grid can take a variety of shapes including cylindrical, spherical-polar, or the so-called cubed-sphere. We find the last option to be particularly useful in our simulations of black-hole accretion disks with arbitrary inclinations between the disk and hole. These tools bring us to the verge of the next step of realism in black-hole accretion disk simulations: general relativistic radiation MHD. Indiana University
Computing Gravitational Waves from Extreme-Mass-Ratio Binary Black Hole Inspirals The inspiral of stellar-mass compact objects into supermassive black holes is expected to be a strong source of gravitational waves (GWs), and should be detected at high signal-to-noise ratios by the planned LISA space-based gravitational-wave observatory. As a step towards calculating the expected GW signal from such a source, here I consider the problem of calculating the "self-force'' on the small body (modelled as a point particle of mass μ M orbiting in a perturbed Schwarzschild or Kerr spacetime of mass M, with μ << 1) due to gravitational radiation reaction. I use the Barack-Ori mode-sum regularization procedure, which involves a spherical harmonic decomposition of the Lorenz-gauge metric perturbation induced by the particle into (ℓ,m) modes, then for each mode the numerical solution of a linear wave equation on the background spacetime with a moving Dirac δ -function source term, then a final regularized sum over the modes. Here I focus on the use of Berger-Oliger adaptive mesh refinement (AMR) in the numerical solution, using characteristic (double-null) coordinates. I describe a new algorithm for Berger-Oliger characteristic AMR, which is relatively simple to implement, highly efficient in both memory usage and computation time, and easily accommodates both 2nd and 4th order finite differencing. I present results from this algorithm for the self-force on a scalar-field particle in a circular orbit in Schwarzschild spacetime. Click here for Dr. Thornburg's poster abstract Tom Jones, Pete Mendygral, David Porter University of Minnesota Seam O'Neill University of Maryland AGN Interactions with Cluster Media: MHD Simulations with Relativistic Electron Transport
AGNs are now widely thought to play important roles in the energy budgets of galaxy cluster media. Evaluation of those roles depends essentially on an understanding of how AGN outflows interact with the ICM, including energy and entropy deposition, and such important processes as the generation of turbulence and magnetic fields. We are engaged in an extensive computational effort to simulate in 3D MHD both FRI and FRII radio galaxy dynamics in realistic settings. Bi-directional jets are injected in cluster cores from stationary AGNs and from supersonically moving AGNs closer to cluster peripheries. The simulations include an Eulerian treatment of the kinetic equation for passive populations of relativistic, "CR'' electrons, allowing for realistic "synthetic observations'' of nonthermal emissions associated with the AGN activity. We also are able to test observational "AGN calorimetry'' estimates using cluster thermal X-ray cavities. This work is supported by the NSF and by the University of Minnesota Supercomputing Institute.
SHARCNET
AMR Simulations of Protogalactic Clumps I will talk about our latest attempts to model clumpy interstellar medium in z~3 star-forming galaxies using adaptive mesh refinement to resolve large molecular clouds while simulating the dynamics of entire galactic disks. One of our goals is to reproduce high-velocity galactic winds observed in LBGs and DLAs without suppressing cooling which is a method frequently used to get efficient feedback. I will also discuss how these models can be incorporated into cosmological simulations. Queen’s University
Using Scale-Free Cosmologies to Explore the Small-Scale Universe In the Cold Dark Matter paradigm, structures form in a hierarchical fashion with small-scale perturbations collapsing first to form halos that then grow via mergers to form larger ones. However, in the ultra small-scale limit, dark matter structures form nearly simultaneously across a wide range of scales and hierarchical clustering no longer provides a theoretical guide and credible simulations become computational daunting. We explore this small-scale limit using a series of simulations of scale-free cosmological models, i.e., Einstein-de Sitter cosmologies where the initial power spectrum is a power-law. Our simulations are larger than previous scale-free simulations by a factor of 3 or more. By using the index of the power spectrum as a proxy for scale, we effectively examine various scales in the Universe. We examine the substructure of dark matter halos and the nonlinear evolution of the power spectrum.
Rob Thacker, R., Saez, D., Fullana, M., Arnau, J., & Couchman, H. M. P. Saint Mary’s University
CMB Weak Lensing at High ℓ
Accurate calculation of the impact of foregrounds on the high ℓ CMB multipoles requires N-body simulations and ray-tracing schemes with both high spatial and temporal resolution. To this end we have developed a new code that combines a gravitational Adaptive Particle-Particle, Particle-Mesh (AP3M) solver with a weak lensing evaluation routine. The lensing deviations are evaluated using all the N-body steps in the simulation, while periodicity effects are avoided using so-called "preferred directions". We compare our work with other approaches based on multiple planes, as well as to our former work with Particle-Mesh simulations. We have also systematically evaluated the impact of numerous numerical parameters (e.g. particle softening) on the robustness of the high ℓ signal. For a reaslistic cosmological model the power [ℓ (ℓ +1)C ℓ /2π]1/2 takes on values of a few μK in this interval, which suggests that a future detection is feasible and may explain the excess power at high ℓ in the BIMA and CBI observations.
Saint Mary’s University
Matching the Pulsation Frequencies of Pre-main Sequence Delta-Scuti Stars to Their Theoretical Counterparts X² tests are used to compare the observed pulsation frequencies of pre-main-sequence (PMS) δ-Scuti variables to the theoretical pulsation frequencies computed from a dense grid of YREC7 stellar-evolution models. Such comparisons identify the models within the grid to which the observed stars pulsations frequencies can best be matched. Families of solutions are typical (e.g. a range of age and mass), but by combining pulsation observations with other observations (such as luminosity and effective temperature) many of these solutions are eliminated, thus significantly constraining the implied physical characteristics of the observed star. In this way, PMS models are matched against real data to a higher precision than was previously possible without this technique, providing better tests of PMS evolutionary theory than in the past.
Université Laval
Realistic Numerical Models of Galactic Outflows Galactic outflows play and important role in the evolution of galaxies and the intergalactic medium IGM. These outflows transport energy and metal-enriched gas into the IGM, which impact the formation of the next generation of galaxies. Galactic outflows are often included in cosmological simulations using semi-analytical models. However, these models are often quite simplistic, especially in their prediction of the metal content and composition of the outflow. We use the population-synthesis code Starburst99 to simulate in detail the evolution of galaxies. This code predicts the amount and composition of gas released by Type Ia and Type II supernovae, stellar winds, and WR stars, as well as the amount of energy that will be powering the galactic outflows. This enables us to make detailed predictions on the chemical composition of the outflows and its evolution with time.
David A. Clarke, Jon Ramsey, Alexander B. Men'shchikov Saint Mary’s University
AZEuS: Numerical Methods for a Staggered AMR MHD Code We show how the ideas of AMR (Adaptive Mesh Refinement), originally designed for zone-centred Godunov-type fluid solvers, have been modified for a code built upon a staggered mesh architecture, such as ZEUS-3D. In addition to the staggered mesh, we required our design to pass waves (e.g., shocks, rarefactions, contacts) across grid interfaces with minimal disruption so that problems employing nested grids as well as adaptive ones may be explored. The staggered mesh has precipitated a redesign of the refluxing step for face-centred momentum, while the need to pass waves cleanly across grid interfaces has led us to apply higher-order interpolation schemes for constructing grids and boundary values. Demonstrations of these design features will be presented in this poster.
Yale University
Stellar Convection Simulations at Yale We summarize recent and on going research on 3D RHD numerical simulations of the outer layers of the Sun and sun-like stars. Our aim is develop a physically more realistic description of convection in stellar models than the mixing length theory. Solar and stellar seismology provide sensitive observational tests of the improved stellar models.
Joel Tanner, Sarbani Basu, Frank Robinson and Pierre Demarque Yale University Comparing Properties of Surface Convection in Stars with Different Masses and Evolutionary States.
By comparing convective properties computed from simulations derived from a series of G and K type stellar models, we find marked differences between simulations and mixing length theory. These differences increase with the strength of the turbulence, as measured by the turbulent pressure. Two important differences are in the run of the superadiabaticity versus depth and in the size of the convective eddies. Provided the turbulence is weak (i.e. the ratio of turbulent pressure to gas pressure is a few percent) the estimate of eddy size used in a stellar model eddy is quite close to the mean size of the simulated eddies. However, in hotter and more evolved stars, the strength of the turbulence increases and the mixing length estimate becomes a very poor approximation to the actual eddy size.
Los Alamos National Laboratory 1D Nonlinear Hydrodynamic Models of Mira Variables: Modeling the Interior/Atmosphere Boundary
Very different kinds of models are needed to handle stellar interiors vs. stellar atmospheres and winds. In the latter, non-local energy exchange, non-LTE and departures from radiative and chemical equilibrium must be taken into account. Nonlinear behavior of the interior provides complex inner boundary conditions to the atmosphere models. There is likely also to be feedback from the atmosphere on the pulsations at the photosphere, modifying the wave function and perhaps even playing a role in mode selection. We use the 1D Lagrangian nonlinear hydrodynamics code Dynstar (Ostlie and Cox 1992) to model the pulsations of Mira variables. We examine the pulsation driving region and photosphere to generate more realistic driving functions than the usual sinusoidal 'piston' boundary condition for the Bowen atmosphere code (Bowen 1988; Wang and Willson 2009) used to study pulsation-driven mass loss. We plan to use the Bowen code results to create improved outer boundary conditions for the Dynstar code to investigate whether feedback from the dynamical atmosphere modifies the pulsation behavior.
University of Rochester
Disk Formation and Structure in AstroBEAR: Initial Tests In the later phases of cloud core collapse there are a number of shocks that can form as material rains down from an ambient cloud onto on a pre-formed gaseous disk. These shocks are not well-resolved in observations, however indirect evidence of them are seen in the interpretation of spectra covering a range of temperatures and densities. Using the hydrodynamic AMR numerical code AstroBEAR, we plan to investigate these phenomena at a range of length scales. Our code is adapted to cylindrical coordinates with angular momentum; so-called "2.5D". In progress towards this goal, and in support of the development of AstroBEAR in 2.5D, we have investigated the natural formation of a hydrostatically-supported disk from cloud collapse. We present here the results of various explored configurations of dynamical medium combined with different sink cell methods.
Falk Herwig, Paul Woodward, Marco Pignatari, Chris Fryer, Gabriel Rockefeller, David Porter, Michael Bennett, Raphael Hirschi University of Victoria University of Victoria
The convective-reactive flash in Sakurai’s object
We present our multi-physics, multi-scale investigation of nuclear combustion in stellar evolution. We use global stellar evolution simulations to calculate the convective-reactive phases of stars, as for example in thermonuclear He-shell flashes on compact objects which entrain unprocessed H from stable layers above the emerging convection zone. A well observed (20 heavy elements with highly non-solar abundance distributions, e.g. [hs/ls]<-2) prototype of this real-time stellar evolution phase is Sakurai’s object (V4334 Sagittarii), which serves as a validation case for our simulations. We have performed detailed nucleosynthesis simulations using the multi-zone NuGrid codes. These simulations - not surprisingly - show that one-dimensional simulations do not correctly describe the mixing properties of nuclear combustion (Damköhler number around unity). Therefore we have performed full spherical 3D hydrodynamic simulations of He-shell flash convection to investigate the entrainment and mixing of H-rich material into the 12C-rich convection zone. We show that the mixing properties of these hydrodynamic simulations appear to be qualitatively consistent with the expectations from the comprehensive nucleosynthesis simulations. Université Laval
The Build-up of Metallicity in Clusters We combine a P3M-SPH algorithm for cosmological simulations with a new technique for resolving subgrid phenomena associated with galaxies. We introduce a special kind of particles, called "galaxy objects," or GALOBs, to represent galaxies. GALOBs form dynamically during the simulation, and have the ability to produce metals and to generate galactic outflows, leading to the metal-enrichment of the intergalactic medium. Using this algorithm, we simulate the formation of clusters of galaxies in a LCDM universe. We build merger trees to determine the merger history of each massive cluster, and follow the build-up of the metallicity in the intracluster medium.
Viktor Khalack, LeBlanc F., Behr B.B., Wade G.A., Bohlender D.A. Université de Moncton
Detection of Vertical Stratification of Iron Abundance in BHB Stars Most of the Blue Horizontal Branch (BHB) stars in Globular Clusters appear to have enhanced abundances of metals, and depleted helium abundance in comparison with the mean cluster composition. It is commonly believed that these abundance peculiarities appear as a result of atomic diffusion in the stellar atmosphere. To test this hypothesis, we have undertaken an abundance stratification analysis of the atmospheres of six BHB stars, based on recently-acquired HIRES spectra. From our numerical simulations, we have found that two stars in M13 (WF4-3085 and WF2-2541) and two stars in M15 (B267and B279) show clear signatures of vertical stratification of iron. Iron abundance increases towards the lower atmosphere. The other two stars in our sample (WF4-3485 in M13 and B84 in M15) do not show any significant variations of iron abundance with atmospheric depth. Our results support the idea that atomic diffusion can dominate other hydrodynamical processes in the atmospheres of BHB stars.
Viktor Khalack, LeBlanc F., Wade G.A. Université de Moncton
Modelling the Incomplete Paschen-Back Effect for SiII Lines in Spectra of Magnetic CP Stars We present calculations of Si II absorption lines under the influence of the incomplete (partial) Paschen-Back effect, illustrating the changes of atomic level splitting and component strengths with the magnitude and direction of the magnetic field, relative to those obtained for normal Zeeman effect. Such lines are promiment in spectra of some magnetic chemically peculiar (CP) stars. It is shown that the incomplete Paschen-Back effect can lead to noticeable line shifts which strongly depend on total multiplet strength and on magnetic field strength and configuration. A few examples of the simulation of silicon line profiles in the spectrum of HD~94660 (a magnetic CP star with a comparatively strong magnetic field) assuming both Zeeman and Paschen-Back splitting are presented.
CRAAG
Transmission and Reflection of Compressive Waves
The p-modes are oscillations observed at the sun surface, which permit to probe its interior structure. These modes interact with the superficial layers of the sun. The convection zone of the sun is the source of all disturbances and thus of the oscillating wave energy .This later is transmitted through the upper layers of the sun atmosphere up to the solar corona. It is, therefore, imperative to study the p-modes propagation through these layers, knowing that in magnetized medium, these modes are confined under sun surface. In this work, we explore interaction of acoustic waves with a simple structure composed of two regions: i) non-magnetic region, ii) magnetic region with flow. The reflected and transmitted waves and the energy balance are studied and examinated, in the aim of determining the negative energy and the unstable modes.
McMaster University
The Turbulent ISM Interstellar clouds form the nursery from which new stars are born, and it is the way in which these clouds are created and evolve that guides the growth of stellar populations. The explanation of the formation of these clouds has evolved greatly in past decades, from the pressure balanced models of Field, Goldsmith, & Habing (1969) and the work of McKee and Ostriker (1977) to the current view of a turbulent interstellar medium in which the clouds are not in pressure equilibrium with their surroundings. The effects of turbulence on star formation occur both within giant molecular clouds and via affects on the GMCs themselves. Turbulence provides both large scale velocities that can concentrate gas in regions of high densities at shocks, and also a turbulent pressure that may work against gravitational collapse. Simulations of turbulence in the ISM let us probe how these processes play out to yield the ISM we observe. Driving turbulence in a stratified isothermal atmosphere shows that the structure of the gas is well represented by a set of log-normal PDFs at any given plane of constant height. The pressure support from the turbulent motions of the gas keeps the atmosphere in hydrostatic equilibrium, with equal contributions to the total pressure from the thermal and turbulent components. Saint Mary's University Bridging the Gap: Synthetic Radio Observations of Numerical Simulations of Extragalactic Jets The standard paradigm for radio galaxies is based on high-speed plasma jets, formed in active galactic nuclei, which then penetrate into the surrounding intergalactic medium creating giant lobes of luminous material. These lobes then emit in the radio due to synchrotron radiation from high-energy electrons immersed in weak magnetic fields. Modern computational resources have allowed increasingly sophisticated magnetohydrodynamical (MHD) simulations of these plasma flows; however, simulating the emission from these jet models, and thus bridging the gap between theory and observation, remains a difficult task. I will present a semi-empirical model of synchrotron emission that I have incorporated into full three-dimensional MHD jet simulations. From these models I generate synthetic radio maps that can then be compared to actual observations. This thesis presents the results of this radio mapping procedure. By synthetically observing a source whose detailed structure is known beforehand, one can hope to gain insights into what real observations are telling us about these types of jets.
Université Laval
Disk-Disk Galaxy Mergers at High Redshift : Substructure Formation We are presenting here our latest numerical simulations of high-redshift gas-rich galaxy mergers. Using a state-of-art Tree-SPH algorithm called GCD+ (Kawata & Gibson 2003), which includes star formation, supernovae explosions and feedback, and chemical enrichment, we performed a series of simulations of gas-rich mergers, resulting in the formation of remnants with disk morphology. These simulations reveal the presence, around the remnants, of gravitationally bound substructures that were formed during the mergers. By studying the chemical enrichment patterns of the stars located inside these substructures, we can constrain the processes which may be responsible for the observed bimodal distribution of globular cluster at the present age.
Saint Mary’s University
Fitting Cool Star SEDs with PHOENIX Models Burnashev (1985) have presented a unique, uniformly re-calibrated data set of absolute spectrophotometry of hundreds of stars, obtained at observatories in the former Soviet Union, that covers the visible and near UV bands. We are mining this data set for high quality spectral energy distributions (SEDs) of late-type stars (spectral class late F to late K) of solar and 1/3 solar metallicity that sample the cool side of the HR diagram. We have calculated a grid of about 200 spherical atmospheric models and synthetic SEDs of yellow, orange, and red dwarfs, sub-giants, and giants in local thermodynamic equilibrium (LTE) using version 15 of the PHOENIX code. We plan to map out discrepancies between the latest LTE (and, eventually, Non-LTE) models and observed SEDs as a function of stellar parameters and metallicity. One main goal is to determine how the notorious discrepancy in the near UV band varies with effective temperature, surface gravity, and metallicity in the hope that we can characterize sources of missing near UV opacity. Indiana University
Gravitational Waves from Extreme-Mass-Ratio Binary Black Hole Coalescences
The inspiral of stellar-mass compact objects into supermassive black holes (BHs) is expected to be a strong source of gravitational waves (GWs), and should be detected at high signal-to-noise ratios by the planned LISA space-based gravitational-wave observatory. I briefly outline the LISA mission and some of the exciting contributions it can make towards astrophysics and cosmology. I then outline the current state of efforts towards calculating the expected GW signals from various binary-BH sources, and present details of one step towards such a calculation. I consider the problem of calculating the "self-force'' on the small BH as it orbits the large BH (here modelled as a perturbed Schwarzschild or Kerr spacetime) due to gravitational radiation reaction. I use the Barack-Ori mode-sum regularization procedure, which involves a spherical harmonic decomposition of the Lorenz-gauge metric perturbation induced by the particle into $(\ell,m)$ modes, then for each mode the numerical solution of a linear wave equation on the background spacetime with a moving Dirac $\delta$-function source term, then a final regularized sum over the modes. Here I focus on the use of Berger-Oliger adaptive mesh refinement (AMR) in the numerical solution, using characteristic (double-null) coordinates. I describe a new algorithm for Berger-Oliger characteristic AMR, which is relatively simple to implement, highly efficient in both memory usage and computation time, and easily accomodates both 2nd and 4th order finite differencing. I presents results from this algorithm for the self-force on a scalar-field particle in a circular orbit in Schwarzschild spacetime.
|