School of Physics and Astronomy

 Using a set of simulations of MRI turbulence performed using the codes ZEUS (finite difference) and RAMSES (finite volume), I will describe various methods to quantify numerical dissipation in these codes. I will show that the results is not as simple as one might naively guess and discuss the consequences when applied to the problem of angular momentum transport in accretion disks.

Star clusters in the Milky Way contain up to hundreds of thousands of stars held together by gravitational interactions. As the vast majority of stars is formed in clusters, clusters provide crucial information on the star formation process. N-body simulations are frequently used to constrain the past and future evolution of star clusters. In these presentations we provide an overview of the different N-body techniques and their advantages and disadvantages. We discuss the birth, life, and death of a star cluster, and present several related applications.

I will present simulations of the collapse of initially cold and clumpy star clusters. From these initial conditions I will show that dynamical mass segregation almost always results in only a crossing time. I will talk about how the setting-up of initial conditions is vital, and the interpretion of results often very problematic.

 The magnetisation of the ISM displays an important initial condition for star formation. Magnetic fields are recognised to play a significant role in the fragmentation and rotational braking of protostellar cores. They are also a possible candidate for the regulation of star formation, and might thus explain the low observed SF efficiency. Moreover, magnetic fields influence the intermittent structure of the turbulent cascade, whose inertial range slope is a key parameter in understanding the initial core mass function.

Since the early works of Parker in the 1950s, it is well known that turbulent flows which lack certain symmetries can produce mean magnetic fields out of unordered motion. In the 1960s this phenomenon has been mathematically formalised in the framework of so-called mean-field MHD by Krause and Rädler. Based on this approach, the field amplification in the Sun, in convective stars, and even in galaxies can be studied by means of dynamo models. Within those large eddy simulations, one only considers the mean-flow, while the effect of the turbulence is modelled via the so-called alpha parameter. Recently, it has become possible to derive this closure parameter from direct simulations, e.g. of the turbulent ISM, by means of a passive tracer-field method. Here I give a short introduction on how this method is implemented numerically.

We present two implementations of ionizing radiations in the tree-SPH code VINE: iVINE and VINERY. iVINE is a highly efficient parallelized approach which allows for high resolution simulations under the assumption of plane-parallel irradiation of the simulated area. VINERY, the combination of VINE and SPHRAY (Altay et al. 2008), is a Monte Carlo approach with a very efficient ray-shooting algorithm which allows for the simulation of point sources as well. In addition, we present some first  results. We show that radiation from hot stars penetrates the ISM, efficiently heats cold low density gas and amplifies over-densities seeded by the initial turbulence. The formation of observed pillar-like structures in star forming regions (e.g. in M16) can be explained by this scenario. At the tip of the pillars gravitational collapse can be induced, eventually leading to the formation of low mass stars.  Detailed analysis of the evolution of the turbulent spectra shows that UV-radiation indeed provides an excellent mechanism to sustain and even drive turbulence in the ISM. A wide range parameter study enables us to derive an analytic approach to determine the density and shape of the resulting pillars.

Emission lines encode information on the geometry and dynamics of the gas and dust as it flows around and onto the protostar. In order to extract this information it is necessary to compute synthetic profiles and compare them with observations. This is a challenging problem, since it involves frequency-dependent radiative-transfer in a moving medium where departures from spherical symmetry are significant. In this talk I will review the techniques used to solve the rate equations in atoms and molecules and hence calculate line profiles in cores, discs, and magnetospheric accretion streams.

We report time-series CCD I-band photometry for the pre-main-sequence cluster NGC 6530, located within the Lagoon Nebula. The data were obtained with the 4Kx4K imager on the SMARTS 1.0m telescope at CTIO on 36 nights over the summers of 2005 and 2006. In total we have light curves for ~50,000 stars in an area ~1 deg2, with a sampling cadence of ~1 hour. The stars in our sample have masses in the range 0.25-4.0 M(sun), assuming a distance of 1.25 kpc to the cluster. Our goals are to look for stars with rotation periods and to identify eclipsing binary candidates. Here we present light curves of photometrically variable stars and potential eclipsing binary star systems. These light curves were derived in conjunction with ACCRE (Advanced Computing Center for Research and Education), the Vanderbilt University supercomputing center. We will test these data against current stellar evolution theories that predict specific relationships between a star’s mass, radius, temperature, and luminosity as a function of age. These relationships are important, as they are the basis for determining the fundamental properties of stars. Our measurements of these properties will test the basic theories that are central to the determination of stellar ages and masses. We will also examine the distribution of stellar rotation periods. This is essential to understanding how the angular momenta of stars evolve with time, a problem that remains one of the longest outstanding questions in star formation research.

Ecole normale superieure, Observatoire de Paris

Kapteyn Astronomical Institute, University of Groningen

Smoothed Particle Hydrodynamics (SPH) is one of the principle numerical algorithms used to study astrophysical fluid dynamics problems. We will review the fundamentals of SPH in detail including the derivation of the SPH equations, methods of including artificial dissipation and extending SPH to include self-gravity.  We will discuss various caveats that should be considered including the resolution requirements and sources of error in SPH simulations.  Finally, we briefly discuss algorithms and optimizations that can significantly speed up SPH simulations, such as trees, block time-stepping and sink particles.

If the spin axes of stars in young clusters are randomly orientated in space, then it becomes possible, with measurements of period and projected equatorial velocity, to obtain accurate estimates of distance, mean age and age dispersion of the cluster (Jeffries 2007a,b). A new technique is presented which can be used to test this assumption. The inclination of a set of stars, found from their projected equatorial velocity, period and angular diameter is analyzed to determine whether their spin axes are randomly distributed in space or if there is a preferred direction of rotation. The method is used to assess the degree of alignment of stars in the Pleiades and Alpha-Persei clusters. Both clusters show results consistent with a uniform distribution. Using period and Vsini data to estimate distance to the Pleiades gives a result consistent with main sequence fitting and higher than the value found from Hipparcos measurements.

 The thermodynamic behavior of the star-forming gas plays a crucial role in fragmentation and determines the stellar mass function as well as the dynamic properties of the formed stars. This holds for star formation in molecular clouds in the solar neighborhood as well as for the formation of the very first stars in the early universe. In this talk, I will discuss results from various numerical simulations of star cluster formation and show how the thermodynamic state of the turbulent interstellar medium influences the collapse of the star-forming gas.

I will review the current status of a number of observational tests of evolutionary models, covering the period of time between the emergence of a star as a class II object and its approach to the main sequence. Topics will include the mass-radius relationship; age spreads in young clusters and lithium depletion.

Eta Carinae is a unique massive binary star system in our Galaxy, presenting some basic undetermined parameters and open questions. Its 5.54 year periodicity in a very eccentric orbit is observed in all wavelengths. We identify the important physical processes that we expect to take place at periastron passage, and use these to explain observations. In particular, we show that mass accretion onto the companion is expected to occur, and might explain some basic observations. For example, the accreted mass possesses enough angular momentum to form a thick disk, or a belt, around the secondary, and shut down the secondary wind for about ten weeks. After periastron the belt dissipates as its mass is blown away by the reestablished secondary wind. We also study the wind collision in detail, and come up with straightforward explanations to some key observations, like Doppler shifts, X-ray absorption, and the absorption of the He I 10830 lines.

We describe a method for incorporating single-fluid ambipolar diffusion in thestrong coupling approximation into a multidimensional magnetohydrodynamic code based on the total variation diminishing |scheme. Contributions from am-bipolar diffusion terms are included by explicit finite difference operators in a fully unsplit way, maintaining second order accuracy. The divergence-free condition of magnetic fields is exactly ensured at all times by a flux-interpolated constrained transport scheme. The super time stepping method is used to accelerate the timestep in high resolution calculations and/or in strong ambipolar diffusion. We perform two test problems, the steady-state oblique C-type shocks and the decay of Alfven waves, confirming the accuracy and robustness of our numerical approach. Results from the simulations of the compressible MHD turbulence with ambipolar diffusion show the flexibility of our method as well as its  ability to follow complex MHD flows in the presence of ambipolar diffusion.These simulations show that the dissipation rate of MHD turbulence is strongly affected by the strength of ambipolar diffusion.

The processes involved in the formation of massive stars is not nearly as well understood as those for lower mass stars.  For instance, stable circumstellar disks have been observed towards a large number of low mass protostars, while the short formation timescales and large average distances to massive star forming regions make it much more challenging to observe such phenomena.  We do not yet have the resolution and sensitivity required to observe disks around all but the closest massive star forming regions (this will require the capabilities of ALMA), but here we present evidence for rotating ionized and molecular gas in a few massive star forming regions.  With the Submillimeter Array and Very Large Array, we consistently see what appears to be Keplerian rotation of the hot and warm gas in and around these ultracompact HII regions.  This suggests that we may be able to see smaller disk like structures at higher resolution.

University of Leicester (UK)/University of Milan (Italy)

Recent studies have given emphasis to the idea that molecular cloud cores'  environment can influence the evolution of the cores themselves. The  present study applies a wavelet analysis to a numerical simulation of  an evolving molecular cloud. Such an analysis allows to filter out the  large-scale emission from images, leaving small-scale emission only.  We found that, by discriminating emission from large-scale structures, cores that indeed proceed to collapse appear to be gravitationally unbound in the analysis. This result stresses the importance of the environment inside molecular clouds and suggests that obtaining cores? masses via a wavelet analysis may result in an underestimation of their actual mass.

Vortices are believed to greatly help the formation of km sized planetesimals by collecting dust particles in their centers. However, vortex dynamics is commonly studied in non-self-gravitating disks. The main goal here is to examine the effects of disk self-gravity on the vortex dynamics via numerical simulations. In the self-gravitating case, when quasi-steady gravitoturbulent state is reached, vortices appear as transient structures undergoing recurring phases of formation, growth to sizes comparable to a local Jeans scale, and eventual shearing and destruction due to gravitational instability. Each phase lasts over 2 orbital or less periods. Vortices and density waves appear to be coupled implying that, in general, one should consider both vortex and density wave modes for a proper understanding of self-gravitating disk dynamics. Although this study is in the context of protoplanetary discs, the results can be applied to any self-gravitating disc. For example, AGN disc out of which stars form.

One of the fundamental astrobiology questions is how life has formed in our Solar System. In this context the formation and stability of abiotic organic molecules such as CH4, formic acid and amino acids, is important for understanding how organic material has formed and survived shocks and energetic particle impact from winds in the early Solar System. Shock waves have been suggested as a plausible scenario to create chondrules, small meteoritic components that have been completely molten by energetic events such as shocks and high velocity particle impacts. We study here the formation and destruction of certain gas-phase molecules such as methane and water during such shock events and compare the chemical timescales with the timescales for shocks arising from gravitational instabilities in a protosolar nebula.

I will present results of experiments testing two routes toward massive binary formation.  The first method, capture of a passing star by a massive protostellar disk, is tested with SPH simulations of star-disk encounters.  I will discuss how these results are incorporated into an N-body code.  Secondly, I will present experiments wherein we extract accretion information from a large-scale SPH simulation, and re-simulate accretion onto a proto-binary at much higher resolution.

 We study the evolution of a single supernova remnant and a dense shell triggered by the explosion under low-metallicity  environments (10^{-4}< Z/Zsun<10^{-2}),  using one-dimensional hydrodynamics simulations with non-equilibrium radiative cooling. We find that regardless of the metallicity, the mean shell temperature can decrease enough if the shell acquires a large amount of the surrounding medium. Such a cooled and dense shell is expected to become gravitationally unstable and fragment into stars.  We constrain the ambient density, explosion energy, and metallicity where the shell fragments through gravitational instability by utilizing a linear stability analysis of an expanding  and decelerating shell suggested by Elmegreen.  We find that the conditions for the shell fragmentation depend heavily on the ambient density and explosion energy, while hardly on the metallicity.

With the James Clerk Maxwell Telescope (JCMT) Gould Belt Legacy Survey, we will map almost all of the well-known star-formation regions within 0.5 kpc with the Submillimetre Common User Bolometer Array 2 (SCUBA2). Most of these regions are associated with a ring of star formation, known as the Gould Belt. We will produce a flux-limited snapshot of nearby star formation over almost 700 deg2 of sky.  The resulting images will yield the first catalogue of prestellar and protostellar sources selected by submillimetre continuum emission. We will also obtain maps of a large sample of prestellar and protostellar sources in three CO isotopologues using the Heterodyne Array Receiver Program (HARP). Finally, we will map the brightest hundred sources with the SCUBA2 polarimeter (POL-2), producing the first statistically significant set of polarization maps in the submillimetre.

Jan Palous

 Astronomical Institute, Academy of Sciences of the Czech Republic

Recent work on constraining the initial binary fraction and separation distributions of M-dwarf binary systems in Orion-like clusters using N-body  simulations is presented. In addition, I describe current work to explain the observed characteristics of brown dwarf-brown dwarf binaries and M-dwarf-brown dwarf binary systems. I also detail the input physics and model set-up required to conduct efficient N-body simulations of such clusters.

 One of the most fundamental questions to date in modern astrophysics is how stars and planets form. Protoplanetary disk evolution and planet formation are closely entangled. Therefore, to understand their formation mechanisms, one first needs to develop a comprehensive picture of the physical and chemical conditions in protoplanetary disks. In this talk, I will present recent work in combining a thermal-chemical model of a typical T Tauri star disk with a molecular line radiative transfer program to investigate the diagnostic potential of the far-infrared lines of water. I discuss the observability of pure rotational water lines with the Herschel Space Observatory.

Smoothed Particle Hydrodynamics (SPH) is a highly versatile code that is very suitable for modelling systems with complex geometries.  In particular it has been very successful in simulations of the collapse of molecular clouds to form stars and the evolution of planet forming discs around young stars.  It has, however, often been criticised for not necessarily modelling all the physical processes particularly well.  In this talk I will discuss why this is indeed important and give some examples of recent work, using SPH that includes detailed thermodynamics and radiation transfer. 

We implement a method to account for radiative transfer effects (based on Stamatellos, Whitworth, Bisbas, Goodwin, 2007, A&A, 475, 37) in the hydrodynamics code FLASH. The method estimates a mean column density and a mean optical depth which then regulates the heating and cooling of each grid cell. We test the implementation on protostellar collapse scenarios. The results obtained are in good agreement with the literature.

Pre- and protostellar cores are highly obscured. As a result the  collapse phase of protostellar cores as well as the early embedded  stages of forming stars are only poorly understood. The missing pieces in the puzzle of low-mass star and disk formation  can only be replaced by ambitious numerical studies.

We present and compare high-resolution, 3D SPH simulations of  isolated molecular cloud core collapse conducted with VINE.  We model the initial cores as slightly supercritical Bonnor-Ebert spheres.  The observed amount of core angular momentum may be explained by regular  rotation or by sub-/transonic turbulence. Depending on each specific case one or the other model is in better agreement with observed velocity gradient  maps. Within this framework, the density and temperature evolution of the  forming star+disk system are in the center of this study.

We address the questions: 1) Are disks, which were formed from turbulent cores comparable to disks from  rigidly rotating cores concerning size and mass? 2) Is a disk's temperature evolution strongly influenced by the distribution  of the infalling gas? This would lead to a modified temperature structure in  an early evolutionary state of turbulent disk formation. 3) Are the fragmentation properties of subsonically turbulent cores essentially different from the properties of cores in rigid rotation?

We also discuss the influence of different treatments  of the SPH artificial viscosity on the disk fragmentation properties and on the redistribution of angular momentum during the simulation.