Schwarzschild's Orbit Superposition Method Research Articles

Equilibrium dynamical models in the inner region of the Large Magellanic Cloud based on Gaia DR3 kinematics

The Large Magellanic Cloud (LMC) contains complex dynamics driven by both internal and external processes. The external forces are due to tidal interactions with the Small Magellanic Cloud and the Milky Way, while internally its dynamics mainly depend on the stellar, gas, and dark matter mass distributions. Despite this complexity, simple physical models often provide valuable insights into the primary driving factors. We used Gaia Data Release 3 (DR3) to explore how well equilibrium dynamical models based on the Jeans equations and the Schwarzschild orbit superposition method are able to describe the LMC's five-dimensional phase-space distribution and line-of-sight (LOS) velocity distribution, respectively. In the Schwarzschild model, we incorporated a triaxial bar component for the first time and derived the LMC's bar pattern speed. We fit comprehensive Jeans dynamical models to all Gaia DR3 stars with proper motion and LOS velocity measurements found in the footprint of the VISTA near-infrared survey of the Magellanic System using a discrete maximum likelihood approach. These models are very efficient at discriminating genuine LMC member stars from Milky Way foreground stars and background galaxies. They constrain the shape, orientation, and enclosed mass of the galaxy under the assumption of axisymmetry. We used the Jeans model results as a stepping stone to more complex two-component Schwarzschild models, which include an axisymmetric disc and a co-centric triaxial bar, which we fit to the LMC Gaia DR3 LOS velocity field using a $ minimisation approach. The Jeans models describe the rotation and velocity dispersion of the LMC disc well, and we find an inclination angle of $ circ circ $, line of nodes orientation of $ circ circ $, and an intrinsic thickness of the disc of $q_0^d = b a (minor to major axis ratio). However, bound to axisymmetry, these models fail to properly describe the kinematics in the central region of the galaxy dominated by the LMC bar. We used the derived disc orientation and the Gaia DR3 density image of the LMC to obtain the intrinsic shape of the bar. Using these two components as input to our Schwarzschild models, we performed orbit integration and weighting in a rotating reference frame fixed to the bar, deriving an independent measurement of the LMC bar pattern speed of $ Both the Jeans and Schwarzschild models predict the same enclosed mass distribution within a radius of $6.2$\,kpc of $ Msun.

TriOS Schwarzschild Orbit Modeling: Robustness of Parameter Inference for Masses and Shapes of Triaxial Galaxies with Supermassive Black Holes

Evidence for the majority of the supermassive black holes in the local Universe has been obtained dynamically from stellar motions with the Schwarzschild orbit superposition method. However, there have been only a handful of studies using simulated data to examine the ability of this method to reliably recover known input black hole masses M BH and other galaxy parameters. Here, we conduct a comprehensive assessment of the reliability of the triaxial Schwarzschild method at simultaneously determining M BH, stellar mass-to-light ratio M*/L, dark matter mass, and three intrinsic triaxial shape parameters of simulated galaxies. For each of 25 rounds of mock observations using simulated stellar kinematics and the TriOS code, we derive best-fitting parameters and confidence intervals after a full search in the 6D parameter space with our likelihood-based model inference scheme. The two key mass parameters, M BH and M*/L, are recovered within the 68% confidence interval, and other parameters are recovered between the 68% and 95% confidence intervals. The spatially varying velocity anisotropy of the stellar orbits is also well recovered. We explore whether the goodness-of-fit measure used for galaxy model selection in our pipeline is biased by variable complexity across the 6D parameter space. In our tests, adding a penalty term to the likelihood measure either makes little difference, or worsens the recovery in some cases.

Multiple stellar populations in Schwarzschild modeling and the application to the Fornax dwarf

Dwarf spheroidal (dSph) galaxies are believed to be strongly dark matter dominated and thus are considered perfect objects to study dark matter distribution and test theories of structure formation. They possess resolved, multiple stellar populations that offer new possibilities for modeling. A promising tool for the dynamical modeling of these objects is the Schwarzschild orbit superposition method. In this work we extend our previous implementation of the scheme to include more than one population of stars and a more general form of the mass-to-light ratio function. We tested the improved approach on a nearly spherical, gas-free galaxy formed in the cosmological context from the Illustris simulation. We modeled the binned velocity moments for stars split into two populations by metallicity and demonstrate that in spite of larger sampling errors the increased number of constraints leads to significantly tighter confidence regions on the recovered density and velocity anisotropy profiles. We then applied the method to the Fornax dSph galaxy with stars similarly divided into two populations. In comparison with our earlier work, we find the anisotropy parameter to be slightly increasing, rather than decreasing, with radius and more strongly constrained. We are also able to infer anisotropy for each stellar population separately and find them to be significantly different.

Dynamical Modeling of Galaxies and Supermassive Black Holes: Axisymmetry in Triaxial Schwarzschild Orbit Superposition Models

Abstract We present a detailed analysis of the behavior of the triaxial Schwarzschild orbit superposition method near the axisymmetric limit. Orbit superposition modeling is the primary method used to determine dynamical masses of supermassive black holes (M BH) in nearby galaxies; however, prior studies have reported conflicting results when comparing the outcome from axisymmetric orbit codes with that from a triaxial orbit code in the axisymmetric limit. We show that in order to achieve (oblate) axisymmetry in a triaxial code, care needs to be taken to axisymmetrize the short-axis tube orbits and to exclude both the long-axis tube and box orbits from the orbit library. Using up to 12 Gauss–Hermite moments of the line-of-sight velocity distributions as constraints, we demonstrate the effects of orbit types on the best-fit M BH in orbit modeling of the massive elliptical galaxy NGC 1453 reported in Liepold et al. In addition, we verify the efficacy of our updated code on a mock galaxy data set. We identify a subset of slowly precessing quasi-planar orbits for which the typical integration times can be insufficient to fully capture the equilibrium orbital behavior in both axisymmetric and triaxial systems with central black holes. Further investigation is needed for a more reliable treatment of these orbits.

A New Implementation of the Schwarzschild Method for Constructing Observationally Driven Dynamical Models of Galaxies of All Morphological Types

Abstract We present Forstand, a new code for constructing dynamical models of galaxies with the Schwarzschild orbit-superposition method. These models are constrained by line-of-sight kinematic observations and applicable to galaxies of all morphological types, including disks and triaxial rotating bars. Our implementation has several novel and improved features, is computationally efficient, and has been made publicly available. Using mock data sets taken from N-body simulations, we demonstrate that the pattern speed of a bar can be recovered with an accuracy of 10–20%, regardless of orientation, if the 3D shape of the galaxy is known or inferred correctly.

Dark matter distribution and dynamics of dwarf spheroidal galaxies

AbstractI review the current status of dynamical modelling of dwarf spheroidal galaxies focusing on estimates of their dark matter content. Starting with the simplest methods using the velocity dispersion profiles I discuss the inherent issues of mass-anisotropy degeneracy and contamination by unbound stars. I then move on to methods of increasing complexity, aiming to break the degeneracy, up to recent applications of the Schwarzschild orbit superposition method. The dynamical modelling is placed in the context of possible scenarios for the formation of dwarf spheroidal galaxies, including the tidal stirring model and mergers of dwarf galaxies. The two scenarios are illustrated with examples from simulations: a comparison between the tidal evolution of dwarfs with cuspy and cored dark matter profiles and the formation of a dwarf spheroidal with prolate rotation.

Evaluating the ability of triaxial Schwarzschild modelling to estimate properties of galaxies from the Illustris simulation

We evaluate the capabilities of Schwarzschild's orbit-superposition method by applying it to galaxies from the large scale, high resolution Illustris simulation. Nine early-type galaxies with a range of triaxiality are selected, and we create mock integral field unit data for five line-of-sight projections of each galaxy. Each of the 45 mock data sets is taken as an independent observed galaxy. Using van den Bosch's 2008 triaxial Schwarzschild implementation, we assess model estimates of various galaxy properties, covering mass profiles, intrinsic shapes, stellar orbit distributions and velocity anisotropies. Total mass within $\overline{R_{\rm e}}$ is recovered well with average deviations within $\pm15$ percent. Stellar mass is underestimated by $\sim24$ percent and dark matter overestimated by $\sim38$ percent (assuming an NFW dark matter profile and allowing for degeneracy between stellar mass and dark matter mass). Using a gNFW profile, these values improve to $\sim13$ percent for stellar mass and $\sim18$ percent for dark matter. Axis ratio estimates show a moderate bias of $\Delta (b/a)=0.07$ and $\Delta (c/a)=0.14$ ($a\ge b\ge c$). Distributions of the orbit circularities $\lambda_z$ and $\lambda_x$, representing rotation about the minor and major axes, are well reconstructed. Separating orbits into thermal categories, our models match the average fractions of these categories to within $10$ percent. Velocity anisotropy is well estimated with values matching in the inner regions but becoming slightly radially biased in the outer regions. Overall, the galaxy property estimates we obtained using Schwarzschild modelling are not implausible and are representative of the simulated galaxies we modelled.

Schwarzschild dynamical model of the Fornax dwarf spheroidal galaxy

We present a full dynamical model of the Fornax dwarf spheroidal galaxy obtained with the spherically symmetric Schwarzschild orbit superposition method applied to the largest kinematic data set presently available. We modelled the total mass content of the dwarf with the mass-to-light ratio $\Upsilon$ varying with radius and found that Fornax is with high probability embedded in a massive and extended dark matter halo. We estimated the total mass contained within 1$\,$kpc to be $M(<1\,$kpc$)=1.25^{+0.06}_{-0.13} \times 10^8\,$M$_{\odot}$. The data are consistent with the constant mass-to-light ratio, i.e. the mass-follows-light model, only at 3$\sigma$ level, but still require a high value of $\Upsilon \approx 11.2\,$M$_{\odot}/$L$_{\odot}$. Our results are in general agreement with previous estimates of the dynamical mass profile of Fornax. As the Schwarzschild method does not require any assumptions on the orbital anisotropy of the stars, we obtained a profile of the anisotropy parameter $\beta$ as an output of our modelling. The derived anisotropy is close to zero in the centre of the galaxy and decreases with radius, but remains consistent with isotropic orbits at all radii at 1$\sigma$ confidence level.

The effect of non-sphericity on mass and anisotropy measurements in dSph galaxies with Schwarzschild method

In our previous work we confirmed the reliability of the spherically symmetric Schwarzschild orbit-superposition method to recover the mass and velocity anisotropy profiles of spherical dwarf galaxies. Here we investigate the effect of its application to intrinsically non-spherical objects. For this purpose we use a model of a dwarf spheroidal galaxy formed in a numerical simulation of a major merger of two disky dwarfs. The shape of the stellar component of the merger remnant is axisymmetric and prolate which allows us to identify and measure the bias caused by observing the spheroidal galaxy along different directions, especially the longest and shortest principal axis. The modelling is based on mock data generated from the remnant that are observationally available for dwarfs: projected positions and line-of-sight velocities of the stars. In order to obtain a reliable tool while keeping the number of parameters low we parametrize the total mass distribution as a radius-dependent mass-to-light ratio with just two free parameters we aim to constrain. Our study shows that if the total density profile is known, the true, radially increasing anisotropy profile can be well recovered for the observations along the longest axis whereas the data along the shortest axis lead to the inference an incorrect, isotropic model. On the other hand, if the density profile is derived from the method as well, the anisotropy is always underestimated but the total mass profile is well recovered for the data along the shortest axis whereas for the longest axis the mass content is overestimated.

Recovering the mass profile and orbit anisotropy of mock dwarf galaxies with Schwarzschild modelling

We present a new study concerning the application of the Schwarzschild orbit superposition method to model spherical galaxies. The method aims to recover the mass and the orbit anisotropy parameter profiles of the objects using measurements of positions and line-of-sight velocities usually available for resolved stellar populations of dwarf galaxies in the Local Group. To test the reliability of the method, we used different sets of mock data extracted from four numerical realizations of dark matter haloes. The models shared the same density profile but differed in anisotropy profiles, covering a wide range of possibilities, from constant to increasing and decreasing with radius. The tests were done in two steps, first assuming that the mass profile of the dwarf is known and employing the method to retrieve the anisotropy only, and then varying also the mass distribution. We used two kinds of data samples: unrealistically large ones based on over 270 000 particles from the numerical realizations and small ones matching the amount of data available for the Fornax dwarf. For the large data samples we recover both the mass and the anisotropy profiles with very high accuracy. For the realistically small ones we also find a reasonably good agreement between the fitted and the input anisotropies, however the total density profiles can be significantly biased as a result of their oversensitivity to the available data. Our results therefore provide convincing evidence in favour of the applicability of the Schwarzschild method to break the mass-anisotropy degeneracy in dwarf galaxies.

Applying Schwarzschild's orbit superposition method to barred or non-barred disc galaxies

We present an implementation of the Schwarzschild orbit superposition method which can be used for constructing self-consistent equilibrium models of barred or non-barred disc galaxies, or of elliptical galaxies with figure rotation. This is a further development of the publicly available code SMILE; its main improvements include a new efficient representation of an arbitrary gravitational potential using two-dimensional spline interpolation of Fourier coefficients in the meridional plane, as well as the ability to deal with rotation of the density profile and with multicomponent mass models. We compare several published methods for constructing composite axisymmetric disc--bulge--halo models and demonstrate that our code produces the models that are closest to equilibrium. We also apply it to create models of triaxial elliptical galaxies with cuspy density profiles and figure rotation, and find that such models can be found and are stable over many dynamical times in a wide range of pattern speeds and angular momenta, covering both slow- and fast-rotator classes. We then attempt to create models of strongly barred disc galaxies, using an analytic three-component potential, and find that it is not possible to make a stable dynamically self-consistent model for this density profile. Finally, we take snapshots of two N-body simulations of barred disc galaxies embedded in nearly-spherical haloes, and construct equilibrium models using only information on the density profile of the snapshots. We demonstrate that such reconstructed models are in near-stationary state, in contrast with the original N-body simulations, one of which displayed significant secular evolution.

Orbit-based dynamical models of the Sculptor dSph galaxy

We have developed spherically symmetric dynamical models of dwarf spheroidal (dSph) galaxies using Schwarzschild's orbit superposition method. This type of modelling yields constraints both on the total mass distribution (e.g. enclosed mass and scale radius) and on the orbital structure of the system (e.g. velocity anisotropy). This method is thus less prone to biases introduced by assumptions in comparison to the more commonly used Jeans modelling, and it allows us to put reliable constraints on their dark matter content. Here we present our results for the Sculptor dSph galaxy, after testing our methods on mock data sets. We fit both the second and fourth velocity moment profiles to break the mass–anisotropy degeneracy. For an Navarro, Frenk & White (NFW) dark matter halo profile, we find that the mass of Sculptor within 1 kpc is M1 kpc = (1.03 ± 0.07) × 108 M⊙, and that its velocity anisotropy profile is tangentially biased and nearly constant for radii beyond ∼100 pc. The preferred concentration (c ∼ 15) is low for its dark matter mass but consistent within the scatter found in N-body cosmological simulations. When we let the value of the central logarithmic slope α vary, we find that the best-fitting model has α = 0, although an NFW cusp or shallower is consistent at the 1σ confidence level. On the other hand, very cuspy density profiles with logarithmic central slopes α < −1.5 are strongly disfavoured for Sculptor.

Generalized Schwarzschild’s method

We describe a new finite element method (FEM) to construct continuous equilibrium distribution functions of stellar systems. The method is a generalization of Schwarzschild's orbit superposition method from the space of discrete functions to continuous ones. In contrast to Schwarzschild's method, FEM produces a continuous distribution function (DF) and satisfies the intra element continuity and Jeans equations. The method employs two finite-element meshes, one in configuration space and one in action space. The DF is represented by its values at the nodes of the action-space mesh and by interpolating functions inside the elements. The Galerkin projection of all equations that involve the DF leads to a linear system of equations, which can be solved for the nodal values of the DF using linear or quadratic programming, or other optimization methods. We illustrate the superior performance of FEM by constructing ergodic and anisotropic equilibrium DFs for spherical stellar systems (Hernquist models). We also show that explicitly constraining the DF by the Jeans equations leads to smoother and/or more accurate solutions with both Schwarzschild's method and FEM.

Recovery of the internal orbital structure of galaxies

We construct axisymmetric and triaxial galaxy models with a phase-space distribution function that depends on linear combinations of the three exact integrals of motion for a separable potential. These Abel models, first introduced by Dejonghe & Laurent and subsequently extended by Mathieu & Dejonghe, are the axisymmetric and triaxial generalizations of the well-known spherical Osipkov–Merritt models. We show that the density and higher order velocity moments, as well as the line-of-sight velocity distribution of these models can be calculated efficiently and that they capture much of the rich internal dynamics of early-type galaxies. We build a triaxial and oblate axisymmetric galaxy model with projected kinematics that mimic the two-dimensional kinematic observations that are obtained with integral-field spectrographs such as SAURON. We fit the simulated observations with axisymmetric and triaxial dynamical models constructed with our numerical implementation of Schwarzschild orbit-superposition method. We find that Schwarzschild method is able to recover the internal dynamics and three-integral distribution function of realistic models of early-type galaxies.

The supermassive black hole in NGC 4486a detected with SINFONI at the Very Large Telescope

The near-infrared (IR) integral field spectrograph SINFONI at the European Southern Observatory Very Large Telescope opens a new window for the study of central supermassive black holes. With a near-IR spatial resolution similar to Hubble Space Telescope optical and the ability to penetrate dust, it provides the possibility to explore the low-mass end of the M•‐σ relation ( σ< 120 km s −1 ) where so far very few black hole masses were measured with stellar dynamics. With SINFONI, we observed the central region of the low-luminosity elliptical galaxy NGC 4486a at a spatial resolution of ≈0.1 arcsec in the K band. The stellar kinematics were measured with a maximum penalized likelihood method considering the region around the CO absorption band heads. We determined a black hole mass of M• = (1.25 +0.75 −0.79) × 10 7 M� (90 per cent confidence limit) using the Schwarzschild orbit superposition method including the full two-dimensional spatial information. This mass agrees with the predictions of the M•‐σ relation, strengthening its validity at the lower σ end.

The dynamical distance and intrinsic structure of the globular clusterωCentauri

We determine the dynamical distance D, inclination i, mass-to-light ratio M/L and the intrinsic orbital structure of the globular cluster ω Cen, by fitting axisymmetric dynamical models to the ground-based proper motions of van Leeuwen et al. and line-of-sight velocities from four independent data-sets. We bring the kinematic measurements onto a common coordinate system, and select on cluster membership and on measurement error. This provides a homogeneous data-set of 2295 stars with proper motions accurate to 0.20 mas yr -1 and 2163 stars with line-of-sight velocities accurate to 2 km s -1 , covering a radial range out to about half the tidal radius. We correct the observed velocities for perspective rotation caused by the space motion of the cluster, and show that the residual solid-body rotation component in the proper motions (caused by relative rotation of the photographic plates from which they were derived) can be taken out without any modelling other than assuming axisymmetry. This also provides a tight constraint on I) tan i. The corrected mean velocity fields are consistent with regular rotation, and the velocity dispersion fields display significant deviations from isotropy. We model ω Cen with an axisymmetric implementation of Schwarzschild's orbit superposition method, which accurately fits the surface brightness distribution, makes no assumptions about the degree of velocity anisotropy in the cluster, and allows for radial variations in M/L. We bin the individual measurements on the plane of the sky to search efficiently through the parameter space of the models. Tests on an analytic model demonstrate that this approach is capable of measuring the cluster distance to an accuracy of about 6 per cent. Application to w Cen reveals no dynamical evidence for a significant radial dependence of M/L, in harmony with the relatively long relaxation time of the cluster. The best-fit dynamical model has a stellar V-band mass-to-light ratio M/L V = 2.5 ±0.1 M ○. /L ○. and an inclination i = 50° ± 4°, which corresponds to an average intrinsic axial ratio of 0.78 ± 0.03. The best-fit dynamical distance D = 4.8 ± 0.3 kpc (distance modulus 13.75 ± 0.13 mag) is significantly larger than obtained by means of simple spherical or constant-anisotropy axisymmetric dynamical models, and is consistent with the canonical value 5.0 ± 0.2 kpc obtained by photometric methods. The total mass of the cluster is (2.5 ± 0.3) x 10 6 M ○. . The best-fit model is close to isotropic inside a radius of about 10 arcmin and becomes increasingly tangentially anisotropic in the outer region, which displays significant mean rotation. This phase-space structure may well be caused by the effects of the tidal field of the Milky Way. The cluster contains a separate disk-like component in the radial range between 1 and 3 arcmin, contributing about 4% to the total mass.

Dynamical modelling of stars and gas in NGC 2974: determination of mass-to-light ratio, inclination and orbital structure using the Schwarzschild method

We study the large-scale stellar and gaseous kinematics of the E4 galaxy NGC 2974, based on panoramic integral-field data obtained with SAURON. We quantify the velocity maps with Fourier methods (kinemetry), and show that the large-scale kinematics is largely consistent with axisymmetry. We construct general axisymmetric dynamical models for the stellar motions using the Schwarzschild orbit-superposition method, and compare the inferred inclination and mass-to-light ratio with the values obtained by modelling the gas kinematics. Both approaches give consistent results. However, we find that the stellar models provide fairly weak constraints on the inclination. The intrinsic orbital distribution of NGC 2974, which we infer from our model, is characterized by a large-scale stellar component of high angular momentum. We create semi-analytical test models, resembling NGC 2974, to study the ability of the Schwarzschild modelling technique to recover the given input parameters (mass-to-light ratio and inclination) and the distribution function. We also test the influence of a limited spatial coverage on the recovery of the distribution function (i.e. the orbital structure). We find that the models can accurately recover the input mass-to-light ratio, but we confirm that even with perfect input kinematics the inclination is only marginally constrained. This suggests a possible degeneracy in the determination of the inclination, but further investigations are needed to clarify this issue. For a given potential, we find that the analytical distribution function of our test model is well recovered by the three-integral model within the spatial region constrained by integral-field kinematics.

Two-integral Schwarzschild models

We describe a practical method for constructing axisymmetric two-integral galaxy models [with distribution functions of the form f(E,Lz), in which E is the orbital energy, and Lz is the vertical component of the angular momentum], based on Schwarzschild's orbit-superposition method. Other f(E,Lz)-methods are mostly based on solving the Jeans equations or on finding the distribution function directly from the density, which often places restrictions on the shape of the galaxy. Here, no assumptions are made and any axisymmetric density distribution is possible. The observables are calculated (semi-)analytically, so that our method is faster than most previous, fully numerical implementations. Various aspects are tested extensively, the results of which apply directly to three-integral Schwarzschild methods. We show that a given distribution function can be reproduced with high accuracy and investigate the behaviour of the parameter that is used to measure the goodness-of-fit. Furthermore, we show that the method correctly identifies the range of cusp slopes for which axisymmetric two-integral models with a central black hole do not exist.

Probing a regular orbit with spectral dynamics

We have extended the spectral dynamics formalism introduced by Binney & Spergel, and have implemented a semi-analytic method to represent regular orbits in any potential, making full use of their regularity. We use the spectral analysis code of Carpintero & Aguilar to determine the nature of an orbit (irregular, regular, resonant, periodic) from a short-time numerical integration. If the orbit is regular, we approximate it by a truncated Fourier time series of a few tens of terms per coordinate. Switching to a description in action-angle variables, this corresponds to a reconstruction of the underlying invariant torus. We then relate the uniform distribution of a regular orbit on its torus to the non-uniform distribution in the space of observables by a simple Jacobian transformation between the two sets of coordinates. This allows us to compute, in a cell-independent way, all the physical quantities needed in the study of the orbit, including the density and the line-of-sight velocity distribution, with much increased accuracy. The resulting flexibility in the determination of the orbital properties, and the drastic reduction of storage space for the orbit library, provide a significant improvement in the practical application of Schwarzschild's orbit superposition method for constructing galaxy models. We test and apply our method to two-dimensional orbits in elongated discs, and to the meridional motion in axisymmetric potentials, and show that for a given accuracy, the spectral dynamics formalism requires an order of magnitude fewer computations than the more traditional approaches.