Numerical solutions of partial differential equations enable a broad range of scientific research. The Dedalus Project is a flexible, open-source, parallelized computational framework for solving general partial differential equations using spectral methods. Dedalus translates plain-text strings describing partial differential equations into efficient solvers. This paper details the numerical method that enables this translation, describes the design and implementation of the codebase, and illustrates its capabilities with a variety of example problems. The numerical method is a first-order generalized tau formulation that discretizes equations into banded matrices. This method is implemented with an object-oriented design. Classes for spectral bases and domains manage the discretization and automatic parallel distribution of variables. Discretized fields and mathematical operators are symbolically manipulated with a basic computer algebra system. Initial value, boundary value, and eigenvalue problems are efficiently solved using high-performance linear algebra, transform, and parallel communication libraries. Custom analysis outputs can also be specified in plain text and stored in self-describing portable formats. The performance of the code is evaluated with a parallel scaling benchmark and a comparison to a finite-volume code. The features and flexibility of the codebase are illustrated by solving several examples: the nonlinear Schrodinger equation on a graph, a supersonic magnetohydrodynamic vortex, quasigeostrophic flow, Stokes flow in a cylindrical annulus, normal modes of a radiative atmosphere, and diamagnetic levitation. The Dedalus code and the example problems are available online at http://dedalus-project.org/.

The magnetorotational instability (MRI) occurs when a weak magnetic field destabilises a rotating, electrically conducting fluid with inwardly increasing angular velocity. The MRI is essential to astrophysical disk theory where the shear is typically Keplerian. Internal shear layers in stars may also be MRI unstable, and they take a wide range of profiles, including near-critical. We show that the fastest growing modes of an ideal magnetofluid are three-dimensional provided the shear rate, $S$, is near the two-dimensional onset value, $S_c$. For a Keplerian shear, three-dimensional modes are unstable above $S\approx0.10S_c$, and dominate the two-dimensional modes until $S\approx2.05S_{c}$. These three-dimensional modes dominate for shear profiles relevant to stars and at magnetic Prandtl numbers relevant to liquid-metal laboratory experiments. Significant numbers of rapidly growing three-dimensional modes remain well past $2.05S_{c}$. These finding are significant in three ways. First, weakly nonlinear theory suggests that the MRI saturates by pushing the shear rate to its critical value. This can happen for systems, like stars and laboratory experiments, that can rearrange their angular velocity profiles. Second, the non-normal character and large transient growth of MRI modes should be important whenever three-dimensionality exists. Finally, three-dimensional growth suggests direct dynamo action driven from the linear instability.

A new gridding technique for the solution of partial differential equations in cubical geometry is presented. The method is based on volume penalization, allowing for the imposition of a cubical geometry inside of its circumscribing sphere. By choosing to embed the cube inside of the sphere, one obtains a discretization that is free of any sharp edges or corners. Taking full advantage of the simple geometry of the sphere, spectral bases based on spin-weighted spherical harmonics and Jacobi polynomials, which properly capture the regularity of scalar, vector and tensor components in spherical coordinates, can be applied to obtain moderately efficient and accurate numerical solutions of partial differential equations in the cube. This technique demonstrates the advantages of these bases over other methods for solving PDEs in spherical coordinates. We present results for a test case of incompressible hydrodynamics in cubical geometry: Rayleigh-B\'enard convection with fully Dirichlet boundary conditions. Analysis of the simulations provides what is, to our knowledge, the first result on the scaling of the heat flux with the thermal forcing for this type of convection in a cube in a sphere.

A physically realistic stellar wind model based on Alfv\'en wave dissipation has been used to simulate the wind from Barnard's Star and to estimate the conditions at the location of its recently discovered planetary companion. Such models require knowledge of the stellar surface magnetic field that is currently unknown for Barnard's Star. We circumvent this by considering the observed field distributions of three different stars that constitute admissible magnetic proxies of this object. Under these considerations, Barnard's Star b experiences less intense wind pressure than the much more close-in planet Proxima~b and the planets of the TRAPPIST-1 system. The milder wind conditions are more a result of its much greater orbital distance rather than in differences in the surface magnetic field strengths of the host stars. The dynamic pressure experienced by the planet is comparable to present-day Earth values, but it can undergo variations by factors of several during current sheet crossings in each orbit. The magnetospause standoff distance would be $\sim$\,$20 - 40$\,\% smaller than that of the Earth for an equivalent planetary magnetic field strength.

The Rossby number is a crucial parameter describing the degree of rotational constraint on the convective dynamics in stars and planets. However, it is not an input to computational models of convection but must be measured ex post facto. Here, we report the discovery of a new quantity, the Predictive Rossby number, which is both tightly correlated with the Rossby number and specified in terms of common inputs to numerical models. The Predictive Rossby number can be specified independent of Rayleigh number, allowing suites of numerical solutions to separate the degree of rotational constraint from the strength of the driving of convection. We examine the scaling of convective transport in terms of the Nusselt number and the degree of turbulence in terms of the Reynolds number of the flow, and we find scaling laws nearly identical to those in nonrotational convection at low Rossby number when the Predictive Rossby number is held constant. Finally, we describe the boundary layers as a function of increasing turbulence at constant Rossby number.

In this paper we examine the interaction of convection, rotation and mean flows in a thermal annulus. In this system mean flows are driven by correlations induced by rotation leading to non-trivial Reynolds stresses. The mean flows act back on the convective turbulence acting as a barrier to transport. For this system we demonstrate that the Generalised Quasilinear Approximation (GQL) (Marston et al 2016) may provide a much better approximation to the complicated full nonlinear dynamics than the widely used Quasilinear Approximation (QL). This result will enable the construction of more accurate statistical theories for the description of geophysical and astrophysical flows.

High Peclet number, turbulent convection is a classic system with a large timescale separation between flow speeds and the thermal relaxation time. In this paper, we present a method of fast-forwarding through the long thermal relaxation of convective simulations, and we test the validity of this method. This accelerated evolution (AE) method involves measuring the dynamics of convection early in a simulation and using its characteristics to adjust the mean thermodynamic profile within the domain towards its evolved state. We study Rayleigh-B\'enard convection as a test case for AE. Evolved flow properties of AE solutions are measured to be within a few percent of solutions which are reached through standard evolution (SE) over a full thermal timescale. At the highest values of the Rayleigh number at which we compare SE and AE, we find that AE solutions require roughly an order of magnitude fewer computing hours to evolve than SE solutions.

This paper presents a method for the accurate and efficient computations on scalar, vector and tensor fields in three-dimensional spherical polar coordinates. The methods uses spin-weighted spherical harmonics in the angular directions and rescaled Jacobi polynomials in the radial direction. For the 2-sphere, spin-weighted harmonics allow for automating calculations in a fashion as similar to Fourier series as possible. Derivative operators act as wavenumber multiplication on a set of spectral coefficients. After transforming the angular directions, a set of orthogonal tensor rotations put the radially dependent spectral coefficients into individual spaces each obeying a particular regularity condition at the origin. These regularity spaces have remarkably simple properties under standard vector-calculus operations, such as \textit{grad} and \textit{div}. We use a hierarchy of rescaled Jacobi polynomials for a basis on these regularity spaces. It is possible to select the Jacobi-polynomial parameters such that all relevant operators act in a minimally banded way. Altogether, the geometric structure allows for the accurate and efficient solution of general partial differential equations in the unit ball.

We present a simulation code which can solve broad ranges of partial differential equations in a full sphere. The code expands tensorial variables in a spectral series of spin-weighted spherical harmonics in the angular directions and a scaled Jacobi polynomial basis in the radial direction, as described in Part-I. Nonlinear terms are calculated by transforming from the coefficients in the spectral series to the value of each quantity on the physical grid, where it is easy to calculate products and perform other local operations. The expansion makes it straightforward to solve equations in tensor form (i.e., without decomposition into scalars). We propose and study several unit tests which demonstrate the code can accurately solve linear problems, implement boundary conditions, and transform between spectral and physical space. We then run a series of benchmark problems proposed in Marti et al (2014), implementing the hydrodynamic and magnetohydrodynamic equations. We are able to calculate more accurate solutions than reported in Marti et al 2014 by running at higher spatial resolution and using a higher-order timestepping scheme. We find the rotating convection and convective dynamo benchmark problems depend sensitively on details of timestepping and data analysis. We also demonstrate that in low resolution simulations of the dynamo problem, small changes in a numerical scheme can lead to large changes in the solution. To aid future comparison to these benchmarks, we include the source code used to generate the data, as well as the data and analysis scripts used to generate the figures.

Disequilibrium chemical processes have a large effect upon the spectra of substellar objects. To study these effects, dynamical disequilibrium has been parameterized using the quench and eddy diffusion approximations, but little work has been done to explore how these approximations perform under realistic planetary conditions in different dynamical regimes. As a first step in addressing this problem, we study the localized, small scale convective dynamics of planetary atmospheres by direct numerical simulation of fully compressible hydrodynamics with reactive tracers using the Dedalus code. Using polytropically-stratified, plane parallel atmospheres in 2- and 3-D, we explore the quenching behavior of different abstract chemical species as a function of the dynamical conditions of the atmosphere as parameterized by the Rayleigh number. We find that in both 2- and 3-D, chemical species quench deeper than would be predicted based on simple mixing length arguments. Instead, it is necessary to employ length scales based on the chemical equilibrium profile of the reacting species in order to predict quench points and perform chemical kinetics modeling in 1-D. Based on the results of our simulations, we provide a new length scale, derived from the chemical scale height, which can be used to perform these calculations. This length scale is simple to calculate from known chemical data and makes reasonable predictions for our dynamical simulations.

As the Science Mission Directorate contemplates establishing an open code policy, we consider it timely to share our experiences as the developers of the open-source partial differential equation solver Dedalus. Dedalus is a flexible framework for solving partial differential equations. Its development team primarily uses it for studying stellar and planetary astrophysics. Dedalus was developed originally for astrophysical fluid dynamics (AFD), though it has found a much broader user base, including applied mathematicians, plasma physicists, and oceanographers. Here, we will focus on issues related to open-source software from the perspective of AFD. We use the term AFD with the understanding that astrophysics simulations are inherently multi-physics: fluid dynamics coupled with some combination of gravitational dynamics, radiation transfer, relativity, and magnetic fields. In practice, a few well-known open-source simulation packages represent a large fraction of published work in the field. However, we will argue that an open-code policy should encompass not just these large simulation codes, but also the input files and analysis scripts. It is our interest that NASA adopt an open-code policy because without it, reproducibility in computational science is needlessly hampered.

We conduct a global, weakly nonlinear analysis of the magnetorotational instability (MRI) in a Taylor-Couette flow. This is a multiscale perturbative treatment of the nonideal, axisymmetric MRI near threshold, subject to realistic radial boundary conditions and cylindrical geometry. We analyze both the standard MRI, initialized by a constant vertical background magnetic field, and the helical MRI, with an azimuthal background field component. This is the first weakly nonlinear analysis of the MRI in a global Taylor-Couette geometry, as well as the first weakly nonlinear analysis of the helical MRI. We find that the evolution of the amplitude of the standard MRI is described by a real Ginzburg-Landau equation (GLE), while the amplitude of the helical MRI takes the form of a complex GLE. This suggests that the saturated state of the helical MRI may itself be unstable on long spatial and temporal scales.

The magnetorotational instability (MRI) is a fundamental process of accretion disk physics, but its saturation mechanism remains poorly understood despite considerable theoretical and computational effort. We present a multiple scales analysis of the non-ideal MRI in the weakly nonlinear regime -- that is, when the most unstable MRI mode has a growth rate asymptotically approaching zero from above. Here, we develop our theory in a local, Cartesian channel. Our results confirm the finding by Umurhan et al. (2007) that the perturbation amplitude follows a Ginzburg-Landau equation. We further find that the Ginzburg-Landau equation will arise for the local MRI system with shear-periodic boundary conditions when the effects of ambipolar diffusion are considered. A detailed force balance for the saturated azimuthal velocity and vertical magnetic field demonstrates that even when diffusive effects are important, the bulk flow saturates via the combined processes of reducing the background shear and rearranging and strengthening the background vertical magnetic field. We directly simulate the Ginzburg-Landau amplitude evolution for our system and demonstrate the pattern formation our model predicts on long length and time scales. We compare the weakly nonlinear theory results to a direct numerical simulation of the MRI in a thin-gap Taylor Couette flow.

It has been proposed that mixing induced by convective overshoot can disrupt the inward propagation of carbon deflagrations in super-asymptotic giant branch stars. To test this theory, we study an idealized model of convectively bounded carbon flames with 3D hydrodynamic simulations of the Boussinesq equations using the pseudospectral code Dedalus. Because the flame propagation timescale is much longer than the convection timescale, we approximate the flame as fixed in space, and only consider its effects on the buoyancy of the fluid. By evolving a passive scalar field, we derive a {\it turbulent} chemical diffusivity produced by the convection as a function of height, $D_{\rm t}(z)$. Convection can stall a flame if the chemical mixing timescale, set by the turbulent chemical diffusivity, $D_{\rm t}$, is shorter than the flame propagation timescale, set by the thermal diffusivity, $\kappa$, i.e., when $D_{\rm t}>\kappa$. However, we find $D_{\rm t}<\kappa$ for most of the flame because convective plumes are not dense enough to penetrate into the flame. Extrapolating to realistic stellar conditions, this implies that convective mixing cannot stall a carbon flame and that "hybrid carbon-oxygen-neon" white dwarfs are not a typical product of stellar evolution.

Spectral methods are an efficient way to solve partial differential equations on domains possessing certain symmetries. The utility of a method depends strongly on the choice of spectral basis. In this paper we describe a set of bases built out of Jacobi polynomials, and associated operators for solving scalar, vector, and tensor partial differential equations in polar coordinates on a unit disk. By construction, the bases satisfy regularity conditions at r=0 for any tensorial field. The coordinate singularity in a disk is a prototypical case for many coordinate singularities. The work presented here extends to other geometries. The operators represent covariant derivatives, multiplication by azimuthally symmetric functions, and the tensorial relationship between fields. These arise naturally from relations between classical orthogonal polynomials, and form a Heisenberg algebra. Other past work uses more specific polynomial bases for solving equations in polar coordinates. The main innovation in this paper is to use a larger set of possible bases to achieve maximum bandedness of linear operations. We provide a series of applications of the methods, illustrating their ease-of-use and accuracy.

We present a highly-parallel multi-frequency hybrid radiation hydrodynamics algorithm that combines a spatially-adaptive long characteristics method for the radiation field from point sources with a moment method that handles the diffuse radiation field produced by a volume-filling fluid. Our Hybrid Adaptive Ray-Moment Method (HARM$^2$) operates on patch-based adaptive grids, is compatible with asynchronous time stepping, and works with any moment method. In comparison to previous long characteristics methods, we have greatly improved the parallel performance of the adaptive long-characteristics method by developing a new completely asynchronous and non-blocking communication algorithm. As a result of this improvement, our implementation achieves near-perfect scaling up to $\mathcal{O}(10^3)$ processors on distributed memory machines. We present a series of tests to demonstrate the accuracy and performance of the method.

The nonlinear evolution of the Kelvin-Helmholtz instability is a popular test for code verification. To date, most Kelvin-Helmholtz problems discussed in the literature are ill-posed: they do not converge to any single solution with increasing resolution. This precludes comparisons among different codes and severely limits the utility of the Kelvin-Helmholtz instability as a test problem. The lack of a reference solution has led various authors to assert the accuracy of their simulations based on ad-hoc proxies, e.g., the existence of small-scale structures. This paper proposes well-posed Kelvin-Helmholtz problems with smooth initial conditions and explicit diffusion. We show that in many cases numerical errors/noise can seed spurious small-scale structure in Kelvin-Helmholtz problems. We demonstrate convergence to a reference solution using both Athena, a Godunov code, and Dedalus, a pseudo-spectral code. Problems with constant initial density throughout the domain are relatively straightforward for both codes. However, problems with an initial density jump (which are the norm in astrophysical systems) exhibit rich behavior and are more computationally challenging. In the latter case, Athena simulations are prone to an instability of the inner rolled-up vortex; this instability is seeded by grid-scale errors introduced by the algorithm, and disappears as resolution increases. Both Athena and Dedalus exhibit late-time chaos. Inviscid simulations are riddled with extremely vigorous secondary instabilities which induce more mixing than simulations with explicit diffusion. Our results highlight the importance of running well-posed test problems with demonstrated convergence to a reference solution. To facilitate future comparisons, we include the resolved, converged solutions to the Kelvin-Helmholtz problems in this paper in machine-readable form.

Water's density maximum at 4C makes it well suited to study internal gravity wave excitation by convection: an increasing temperature profile is unstable to convection below 4C, but stably stratified above 4C. We present numerical simulations of a water-like fluid near its density maximum in a two dimensional domain. We successfully model the damping of waves in the simulations using linear theory, provided we do not take the weak damping limit typically used in the literature. In order to isolate the physical mechanism exciting internal waves, we use the novel spectral code Dedalus to run several simplified model simulations of our more detailed simulation. We use data from the full simulation as source terms in two simplified models of internal wave excitation by convection: bulk excitation by convective Reynolds stresses, and interface forcing via the mechanical oscillator effect. We find excellent agreement between the waves generated in the full simulation and the simplified simulation implementing the bulk excitation mechanism. The interface forcing simulations over excite high frequency waves because they assume the excitation is by the "impulsive" penetration of plumes, which spreads energy to high frequencies. However, we find the real excitation is instead by the "sweeping" motion of plumes parallel to the interface. Our results imply that the bulk excitation mechanism is a very accurate heuristic for internal wave generation by convection.

Classical Sweet-Parker models of reconnection predict that reconnection rates depend inversely on the resistivity, usually parameterized using the dimensionless Lundquist number ($\Lund$). We describe magnetohydrodynamic (MHD) simulations using a static, nested grid that show the development of a three-dimensional instability in the plane of a current sheet between reversing field lines without a guide field. The instability leads to rapid reconnection of magnetic field lines at a rate independent of $\Lund$ over at least the range $3.2\times 10^3 \lesssim \Lund \lesssim 3.2 \times 10^5$ resolved by the simulations. We find that this instability occurs even for cases with $\Lund \lesssim 10^4$ that in our models appear stable to the recently described, two-dimensional, plasmoid instability. Our results suggest that three-dimensional, MHD processes alone produce fast (resistivity independent) reconnection without recourse to kinetic effects or external turbulence. The unstable reconnection layers provide a self-consistent environment in which the extensively studied turbulent reconnection process can occur.

Early in the study of viscous accretion disks it was realized that energy transfers from distant sources must be important, not least because the flow at the disk midplane in the bulk of the disk is likely outwards, out of the gravitational potential well. If the source of the viscosity is powered by accretion, such as in the case of the magneto-rotational instability, such distant energy sources must lie in the innermost regions of the disk, where accretion occurs even at the midplane. We argue here that modulations in this energy supply can alter the accretion rate on dynamical, rather than far longer viscous, time scales. This means that both the steady state value of and fluctuations in the inner disk's accretion rate, depending on the details of the inner boundary condition and occurring on the inner disk's rapid evolution time, can affect the outer disk. This is particularly interesting because observations have shown that disk accretion is not steady (e.g.~EX Lupi type objects). We also note that the power supplied to shearing boxes is set by the boxes themselves rather than the physical energy fluxes in a global disk. That is, their saturated magnetic field is not subject to the full set of energy constraints present in an actual disk. Our analysis suggests that large scale radial transport of energy has a critical impact on the evolution and variability of accretion disks.

Thermal conduction is an important energy transfer and damping mechanism in astrophysical flows. Fourier's law - the heat flux is proportional to the negative temperature gradient, leading to temperature diffusion - is a well-known empirical model of thermal conduction. However, entropy diffusion has emerged as an alternative thermal conduction model, despite not ensuring the monotonicity of entropy. This paper investigates the differences between temperature and entropy diffusion for both linear internal gravity waves and weakly nonlinear convection. In addition to simulating the two thermal conduction models with the fully compressible Navier-Stokes equations, we also study their effects in the reduced, "sound-proof" anelastic and pseudo-incompressible equations. We find that in the linear and weakly nonlinear regimes, temperature and entropy diffusion give quantitatively similar results, although there are some larger errors in the pseudo-incompressible equations with temperature diffusion due to inaccuracies in the equation of state. Extrapolating our weakly nonlinear results, we speculate that differences between temperature and entropy diffusion might become more important for strongly turbulent convection.

This paper describes the open-source code Enzo, which uses block-structured adaptive mesh refinement to provide high spatial and temporal resolution for modeling astrophysical fluid flows. The code is Cartesian, can be run in 1, 2, and 3 dimensions, and supports a wide variety of physics including hydrodynamics, ideal and non-ideal magnetohydrodynamics, N-body dynamics (and, more broadly, self-gravity of fluids and particles), primordial gas chemistry, optically-thin radiative cooling of primordial and metal-enriched plasmas (as well as some optically-thick cooling models), radiation transport, cosmological expansion, and models for star formation and feedback in a cosmological context. In addition to explaining the algorithms implemented, we present solutions for a wide range of test problems, demonstrate the code's parallel performance, and discuss the Enzo collaboration's code development methodology.

We study the buildup of magnetic fields during the formation of Population III star-forming regions, by conducting cosmological simulations from realistic initial conditions and varying the Jeans resolution. To investigate this in detail, we start simulations from identical initial conditions, mandating 16, 32 and 64 zones per Jeans length, and studied the variation in their magnetic field amplification. We find that, while compression results in some amplification, turbulent velocity fluctuations driven by the collapse can further amplify an initially weak seed field via dynamo action, provided there is sufficient numerical resolution to capture vortical motions (we find this requirement to be 64 zones per Jeans length, slightly larger than, but consistent with previous work run with more idealized collapse scenarios). We explore saturation of amplification of the magnetic field, which could potentially become dynamically important in subsequent, fully-resolved calculations. We have also identified a relatively surprising phenomena that is purely hydrodynamic: the higher-resolved simulations possess substantially different characteristics, including higher infall-velocity, increased temperatures inside 1000 AU, and decreased molecular hydrogen content in the innermost region. Furthermore, we find that disk formation is suppressed in higher-resolution calculations, at least at the times that we can follow the calculation. We discuss the effect this may have on the buildup of disks over the accretion history of the first clump to form as well as the potential for gravitational instabilities to develop and induce fragmentation.

We use three-dimensional hydrodynamical simulations to study the rapid infall phase of the common envelope interaction of a red giant branch star of mass equal to 0.88 \msun and a companion star of mass ranging from 0.9 down to 0.1 \msun. We first compare the results obtained using two different numerical techniques with different resolutions, and find overall very good agreement. We then compare the outcomes of those simulations with observed systems thought to have gone through a common envelope. The simulations fail to reproduce those systems in the sense that most of the envelope of the donor remains bound at the end of the simulations and the final orbital separations between the donor's remnant and the companion, ranging from 26.8 down to 5.9 \rsun, are larger than the ones observed. We suggest that this discrepancy vouches for recombination playing an essential role in the ejection of the envelope and/or significant shrinkage of the orbit happening in the subsequent phase.

The magnetorotational instability (MRI) may dominate outward transport of angular momentum in accretion disks, allowing material to fall onto the central object. Previous work has established that the MRI can drive a mean-field dynamo, possibly leading to a self-sustaining accretion system. Recently, however, simulations of the scaling of the angular momentum transport parameter $\alphaSS$ with the magnetic Prandtl number $\Prandtl$ have cast doubt on the ability of the MRI to transport astrophysically relevant amounts of angular momentum in real disk systems. Here, we use simulations including explicit physical viscosity and resistivity to show that when vertical stratification is included, mean field dynamo action operates, driving the system to a configuration in which the magnetic field is not fully helical. This relaxes the constraints on the generated field provided by magnetic helicity conservation, allowing the generation of a mean field on timescales independent of the resistivity. Our models demonstrate the existence of a critical magnetic Reynolds number $\Rmagc$, below which transport becomes strongly $\Prandtl$-dependent and chaotic, but above which the transport is steady and $\Prandtl$-independent. Prior simulations showing $\Prandtl$-dependence had $\Rmag < \Rmagc$. We conjecture that this steady regime is possible because the mean field dynamo is not helicity-limited and thus does not depend on the details of the helicity ejection process. Scaling to realistic astrophysical parameters suggests that disks around both protostars and stellar mass black holes have $\Rmag >> \Rmagc$. Thus, we suggest that the strong $\Prandtl$ dependence seen in recent simulations does not occur in real systems.

The analysis of complex multiphysics astrophysical simulations presents a unique and rapidly growing set of challenges: reproducibility, parallelization, and vast increases in data size and complexity chief among them. In order to meet these challenges, and in order to open up new avenues for collaboration between users of multiple simulation platforms, we present yt (available at http://yt.enzotools.org/), an open source, community-developed astrophysical analysis and visualization toolkit. Analysis and visualization with yt are oriented around physically relevant quantities rather than quantities native to astrophysical simulation codes. While originally designed for handling Enzo's structure adaptive mesh refinement (AMR) data, yt has been extended to work with several different simulation methods and simulation codes including Orion, RAMSES, and FLASH. We report on its methods for reading, handling, and visualizing data, including projections, multivariate volume rendering, multi-dimensional histograms, halo finding, light cone generation and topologically-connected isocontour identification. Furthermore, we discuss the underlying algorithms yt uses for processing and visualizing data, and its mechanisms for parallelization of analysis tasks.

The formation of planetesimals via gravitational instability of the dust layer in a protoplanetary disks demands that there be local patches where dust is concentrated by a factor of $\sim$ a few $\times 10^3$ over the background value. Vortices in protoplanetary disks may concentrate dust to these values allowing them to be the nurseries of planetesimals. The concentration of dust in the cores of vortices increases the dust-gas ratio of the core compared to the background disk, creating a "heavy vortex." In this work, we show that these vortices are subject to an instability which we have called the heavy-core instability. Using Floquet theory, we show that this instability occurs in elliptical protoplanetary vortices when the gas-dust density of the core of the vortex is heavier than the ambient gas-dust density by a few tens of percent. The heavy-core instability grows very rapidly, with a growth timescale of a few vortex rotation periods. While the nonlinear evolution of this instability remains unknown, it will likely increase the velocity dispersion of the dust layer in the vortex because instability sets in well before sufficient dust can gather to form a protoplanetary seed. This instability may thus preclude vortices from being sites of planetesimal formation.

We investigate fluid motions near the midplane of vertically stratified accretion disks with highly resistive midplanes. In such disks, the magnetorotational instability drives turbulence in thin layers surrounding a resistive, stable dead zone. The turbulent layers in turn drive motions in the dead zone. We examine the properties of these motions using three-dimensional, stratified, local, shearing-box, non-ideal, magnetohydrodynamical simulations. Although the turbulence in the active zones provides a source of vorticity to the midplane, no evidence for coherent vortices is found in our simulations. It appears that this is because of strong vertical oscillations in the dead zone. By analyzing time series of azimuthally-averaged flow quantities, we identify an axisymmetric wave mode particular to models with dead zones. This mode is reduced in amplitude, but not suppressed entirely, by changing the equation of state from isothermal to ideal. These waves are too low-frequency to affect sedimentation of dust to the midplane, but may have significance for the gravitational stability of the resulting midplane dust layers.

Numerical simulations including magnetic fields have become important in many fields of astrophysics. Evolution of magnetic fields by the constrained transport algorithm preserves magnetic divergence to machine precision, and thus represents one preferred method for the inclusion of magnetic fields in simulations. We show that constrained transport can be implemented with volume-centered fields and hyperresistivity on a high-order finite difference stencil. Additionally, the finite-difference coefficients can be tuned to enhance high-wavenumber resolution. Similar techniques can be used for the interpolations required for dealiasing corrections at high wavenumber. Together, these measures yield an algorithm with a wavenumber resolution that approaches the theoretical maximum achieved by spectral algorithms. Because this algorithm uses finite differences instead of fast Fourier transforms, it runs faster and isn't restricted to periodic boundary conditions. Also, since the finite differences are spatially local, this algorithm is easily scalable to thousands of processors. We demonstrate that, for low-Mach-number turbulence, the results agree well with a high-order, non-constrained-transport scheme with Poisson divergence cleaning.

The initial stages of planet formation in circumstellar gas discs proceed via dust grains that collide and build up larger and larger bodies (Safronov 1969). How this process continues from metre-sized boulders to kilometre-scale planetesimals is a major unsolved problem (Dominik et al. 2007): boulders stick together poorly (Benz 2000), and spiral into the protostar in a few hundred orbits due to a head wind from the slower rotating gas (Weidenschilling 1977). Gravitational collapse of the solid component has been suggested to overcome this barrier (Safronov 1969, Goldreich & Ward 1973, Youdin & Shu 2002). Even low levels of turbulence, however, inhibit sedimentation of solids to a sufficiently dense midplane layer (Weidenschilling & Cuzzi 1993, Dominik et al. 2007), but turbulence must be present to explain observed gas accretion in protostellar discs (Hartmann 1998). Here we report the discovery of efficient gravitational collapse of boulders in locally overdense regions in the midplane. The boulders concentrate initially in transient high pressures in the turbulent gas (Johansen, Klahr, & Henning 2006), and these concentrations are augmented a further order of magnitude by a streaming instability (Youdin & Goodman 2005, Johansen, Henning, & Klahr 2006, Johansen & Youdin 2007) driven by the relative flow of gas and solids. We find that gravitationally bound clusters form with masses comparable to dwarf planets and containing a distribution of boulder sizes. Gravitational collapse happens much faster than radial drift, offering a possible path to planetesimal formation in accreting circumstellar discs.

This document contains refereed supplementary information for the paper ``Rapid planetesimal formation in turbulent circumstellar discs''. It contains 15 sections (\S1.1 -- \S1.15) that address a number of subjects related to the main paper. We describe in detail the Poisson solver used to find the self-potential of the solid particles, including a linear and a non-linear test problem (\S1.3). Dissipative collisions remove energy from the motion of the particles by collisional cooling (\S1.4), an effect that allows gravitational collapse to occur in somewhat less massive discs (\S1.7). A resolution study of the gravitational collapse of the boulders is presented in \S1.6. We find that gravitational collapse can occur in progressively less massive discs as the grid resolution is increased, likely due to the decreased smoothing of the particle-mesh self-gravity solver with increasing resolution. In \S1.10 we show that it is in good agreement with the Goldreich & Ward (1973) stability analysis to form several-hundred-km-sized bodies, when the analysis is applied to 5 AU and to regions of increased boulder column density. \S11 is devoted to the measurement of random speeds and collision speeds between boulders. We find good agreement between our measurements and analytical theory for the random speeds, but the measured collision speeds are 3 times lower than expected from analytical theory. Higher resolution studies, and an improved analytical theory of collision speeds that takes into account epicyclic motion, will be needed to determine whether collision speeds have converged. In \S1.12 we present models with no magnetic fields. The boulder layer still exhibits strong clumping, due to the streaming instability, if the global solids-to-gas ratio is increased by a factor 3. Gravitational collapse occurs as readily as in magnetised discs.

Migration of protoplanets in their gaseous host disks may be largely responsible for the observed orbital distribution of extrasolar planets. Recent simulations have shown that the magnetorotational turbulence thought to drive accretion in protoplanetary disks can affect migration by turning it into an orbital random walk. However, these simulations neglected the disk's ionization structure. Low ionization fraction near the midplane of the disk can decouple the magnetic field from the gas, forming a dead zone with reduced or no turbulence. Here, to understand the effect of dead zones on protoplanetary migration, we perform numerical simulations of a small region of a stratified disk with magnetorotational turbulence confined to thin active layers above and below the midplane. Turbulence in the active layers exerts decreased, but still measurable, gravitational torques on a protoplanet located at the disk midplane. We find a decrease of two orders of magnitude in the diffusion coefficient for dead zones with dead-to-active surface density ratios approaching realistic values in protoplanetary disks. This torque arises primarily from density fluctuations within a distance of one scale height of the protoplanet. Turbulent torques have correlation times of only $\sim 0.3$ orbital periods and apparently time-stationary distributions. These properties are encouraging signs that stochastic methods can be used to determine the orbital evolution of populations of protoplanets under turbulent migration. Our results indicate that dead zones may be dynamically distinct regions for protoplanetary migration.

The nature of ultracompact H II regions (UCHRs) remains poorly determined. In particular, they are about an order of magnitude more common than would be expected if they formed around young massive stars and lasted for one dynamical time, around 10^4 yr. We here perform three-dimensional numerical simulations of the expansion of an H II region into self-gravitating, radiatively cooled gas, both with and without supersonic turbulent flows. In the laminar case, we find that H II region expansion in a collapsing core produces nearly spherical shells, even if the ionizing source is off-center in the core. This agrees with analytic models of blast waves in power-law media. In the turbulent case, we find that the H II region does not disrupt the central collapsing region, but rather sweeps up a shell of gas in which further collapse occurs. Although this does not constitute triggering, as the swept-up gas would eventually have collapsed anyway, it does expose the collapsing regions to ionizing radiation. We suggest that these regions of secondary collapse, which will not all themselves form massive stars, may form the bulk of observed UCHRs. As the larger shell will take over 10^5 years to complete its evolution, this could solve the timescale problem. Our suggestion is supported by the ubiquitous observation of more diffuse emission surrounding UCHRs.

We investigate the question of whether ambipolar diffusion (ion-neutral drift) determines the smallest length and mass scale on which structure forms in a turbulent molecular cloud. We simulate magnetized turbulence in a mostly neutral, uniformly driven, turbulent medium, using a three-dimensional, two-fluid, magnetohydrodynamics (MHD) code modified from Zeus-MP. We find that substantial structure persists below the ambipolar diffusion scale because of the propagation of compressive slow MHD waves at smaller scales. Contrary to simple scaling arguments, ambipolar diffusion thus does not suppress structure below its characteristic dissipation scale as would be expected for a classical diffusive process. We have found this to be true for the magnetic energy, velocity, and density. Correspondingly, ambipolar diffusion leaves the clump mass spectrum unchanged. Ambipolar diffusion appears unable to set a characteristic scale for gravitational collapse and star formation in turbulent molecular clouds.

The observed shock wave positions and expansion in Cas A can be interpreted in a model of supernova interaction with a freely expanding stellar wind with a mass loss rate of ~3e-5 Msun/yr for a wind velocity of 10 km/s. The wind was probably still being lost at the time of the supernova, which may have been of Type IIn or IIb. The wind may play a role in the formation of very fast knots observed in Cas A. In this model, the quasi-stationary flocculi (QSFs) represent clumps in the wind, with a density contrast of several 1000 compared to the smooth wind. The outer, unshocked clumpy wind is photoionized by radiation from the supernova, and is observed as a patchy HII region around Cas A. This gas has a lower density than the QSFs and is heated by nonradiative shocks driven by the blast wave. Denser clumps have recombined and are observed as HI compact absorption features towards Cas A.