arXiv subjects

Michael Joyce

Publications and source records attributed to Michael Joyce.

At least 55 records · Page 3Linked to original sources

arXiv2010

A dynamical classification of the range of pair interactions

We formalize a classification of pair interactions based on the convergence properties of the {\it forces} acting on particles as a function of system size. We do so by considering the behavior of the probability distribution function (PDF) P(F) of the force field F in a particle distribution in the limit that the size of the system is taken to infinity at constant particle density, i.e., in the "usual" thermodynamic limit. For a pair interaction potential V(r) with V(r) \rightarrow \infty) \sim 1/r^a defining a {\it bounded} pair force, we show that P(F) converges continuously to a well-defined and rapidly decreasing PDF if and only if the {\it pair force} is absolutely integrable, i.e., for a > d-1, where d is the spatial dimension. We refer to this case as {\it dynamically short-range}, because the dominant contribution to the force on a typical particle in this limit arises from particles in a finite neighborhood around it. For the {\it dynamically long-range} case, i.e., a \leq d-1, on the other hand, the dominant contribution to the force comes from the mean field due to the bulk, which becomes undefined in this limit. We discuss also how, for a \leq d-1 (and notably, for the case of gravity, a=d-2) P(F) may, in some cases, be defined in a weaker sense. This involves a regularization of the force summation which is generalization of the procedure employed to define gravitational forces in an infinite static homogeneous universe. We explain that the relevant classification in this context is, however, that which divides pair forces with a > d-2 (or a < d-2), for which the PDF of the {\it difference in forces} is defined (or not defined) in the infinite system limit, without any regularization. In the former case dynamics can, as for the (marginal) case of gravity, be defined consistently in an infinite uniform system.

cond-mat.stat-mech↗

arXiv2010

Relaxation to thermal equilibrium in the self-gravitating sheet model

We revisit the issue of relaxation to thermal equilibrium in the so-called "sheet model", i.e., particles in one dimension interacting by attractive forces independent of their separation. We show that this relaxation may be very clearly detected and characterized by following the evolution of order parameters defined by appropriately normalized moments of the phase space distribution which probe its entanglement in space and velocity coordinates. For a class of quasi-stationary states which result from the violent relaxation of rectangular waterbag initial conditions, characterized by their virial ratio R_0, we show that relaxation occurs on a time scale which (i) scales approximately linearly in the particle number N, and (ii) shows also a strong dependence on R_0, with quasi-stationary states from colder initial conditions relaxing much more rapidly. The temporal evolution of the order parameter may be well described by a stretched exponential function. We study finally the correlation of the relaxation times with the amplitude of fluctuations in the relaxing quasi-stationary states, as well as the relation between temporal and ensemble averages.

cond-mat.stat-mech↗

arXiv2010

Quasi-stationary states and the range of pair interactions

"Quasi-stationary" states are approximately time-independent out of equilibrium states which have been observed in a variety of systems of particles interacting by long-range interactions. We investigate here the conditions of their occurrence for a generic pair interaction V(r \rightarrow \infty) \sim 1/r^a with a > 0, in d>1 dimensions. We generalize analytic calculations known for gravity in d=3 to determine the scaling parametric dependences of their relaxation rates due to two body collisions, and report extensive numerical simulations testing their validity. Our results lead to the conclusion that, for a < d-1, the existence of quasi-stationary states is ensured by the large distance behavior of the interaction alone, while for a > d-1 it is conditioned on the short distance properties of the interaction, requiring the presence of a sufficiently large soft-core in the interaction potential.

cond-mat.stat-mech↗

arXiv2010

Gravitational force in an infinite one-dimensional Poisson distribution

We consider the statistical properties of the gravitational field F in an infinite one-dimensional homogeneous Poisson distribution of particles, using an exponential cut-off of the pair interaction to control and study the divergences which arise. Deriving an exact analytic expression for the probability density function (PDF) P(F), we show that it is badly defined in the limit in which the well known Holtzmark distribution is obtained in the analogous three-dimensional case. A well defined P(F) may, however, be obtained in the infinite range limit by an appropriate renormalization of the coupling strength, giving a Gaussian form. Calculating the spatial correlation properties we show that this latter procedure has a trivial physical meaning. Finally we calculate the PDF and correlation properties of differences of forces (at separate spatial points), which are well defined without any renormalization. We explain that the convergence of these quantities is in fact sufficient to allow a physically meaningful infinite system limit to be defined for the clustering dynamics from Poissonian initial conditions.

cond-mat.stat-mech↗

arXiv2009

1-d gravity in infinite point distributions

The dynamics of infinite, asymptotically uniform, distributions of self-gravitating particles in one spatial dimension provides a simple toy model for the analogous three dimensional problem. We focus here on a limitation of such models as treated so far in the literature: the force, as it has been specified, is well defined in infinite point distributions only if there is a centre of symmetry (i.e. the definition requires explicitly the breaking of statistical translational invariance). The problem arises because naive background subtraction (due to expansion, or by "Jeans' swindle" for the static case), applied as in three dimensions, leaves an unregulated contribution to the force due to surface mass fluctuations. Following a discussion by Kiessling, we show that the problem may be resolved by defining the force in infinite point distributions as the limit of an exponentially screened pair interaction. We show that this prescription gives a well defined (finite) force acting on particles in a class of perturbed infinite lattices, which are the point processes relevant to cosmological N-body simulations. For identical particles the dynamics of the simplest toy model is equivalent to that of an infinite set of points with inverted harmonic oscillator potentials which bounce elastically when they collide. We discuss previous results in the literature, and present new results for the specific case of this simplest (static) model starting from "shuffled lattice" initial conditions. These show qualitative properties (notably its "self-similarity") of the evolution very similar to those in the analogous simulations in three dimensions, which in turn resemble those in the expanding universe.

cond-mat.stat-mech↗

arXiv2008

Towards quantitative control on discreteness error in the non-linear regime of cosmological N body simulations

The effects of discreteness arising from the use of the N-body method on the accuracy of simulations of cosmological structure formation are not currently well understood. After a discussion of how the relevant discretisation parameters introduced should be extrapolated to recover the Vlasov-Poisson limit, we study numerically, and with analytical methods we have developed recently, the central issue of how finite particle density affects the precision of results. In particular we focus on the power spectrum at wavenumbers around and above the Nyquist wavenumber, in simulations in which the force resolution is taken smaller than the initial interparticle spacing. Using simulations of identical theoretical initial conditions sampled on four different "pre-initial" configurations (three different Bravais lattices, and a glass) we obtain a {\it lower bound} on the real discreteness error. With the guidance of our analytical results, we establish with confidence that the measured dispersion is not contaminated either by finite box size effects or by subtle numerical effects. Our results show notably that, at wavenumbers {\it below} the Nyquist wavenumber, the dispersion increases monotonically in time throughout the simulation, while the same is true above the Nyquist wavenumber once non-linearity sets in. For normalizations typical of cosmological simulations, we find lower bounds on errors at the Nyquist wavenumber of order of a percent, and larger above this scale. The only way this error may be reduced below these levels at these scales, and indeed convergence to the physical limit firmly established, is by extrapolation, at fixed values of the other relevant parameters, to the regime in which the mean comoving interparticle distance becomes less than the force smoothing scale.

astro-ph↗

arXiv2008

Cosmological simulations of structure formation and the Vlasov equation

In cosmology numerical simulations of structure formation are now of central importance, as they are the sole instrument for providing detailed predictions of current cosmological models for a whole class of important constraining observations. These simulations are essentially molecular dynamics simulations of N (>> 1), now up to of order several billion) particles interacting through their self-gravity. While their aim is to produce the Vlasov limit, which describes the underlying (``cold dark matter'') models, the degree to which they actually do produce this limit is currently understood, at best, only very qualitatively, and there is an acknowledged need for ``a theory of discreteness errors''. In this talk I will describe, for non-cosmologists, both the simulations and the underlying theoretical models, and will then focus on the issue of discreteness, describing some recent progress in addressing this question quantitatively.

cond-mat.stat-mech↗

arXiv2008

Infinite self-gravitating systems and cosmological structure formation

The usual thermodynamic limit for systems of classical self-gravitating point particles becomes well defined, as a {\it dynamical} problem, using a simple physical prescription for the calculation of the force, equivalent to the so-called ``Jeans' swindle''. The relation of the resulting intrinsically out of equilibrium problem, of particles evolving from prescribed uniform initial conditions in an infinite space, to the one studied in current cosmological models (in an expanding universe) is explained. We then describe results of a numerical study of the dynamical evolution of such a system, starting from a simple class of infinite ``shuffled lattice'' initial conditions. The clustering, which develops in time starting from scales around the grid scale, is qualitatively very similar to that seen in cosmological simulations, which begin from lattices with applied correlated displacements and incorporate an expanding spatial background. From very soon after the formation of the first non-linear structures, a spatio-temporal scaling relation describes well the evolution of the two-point correlations. At larger times the dynamics of these correlations converges to what is termed ``self-similar'' evolution in cosmology, in which the time dependence in the scaling relation is specified entirely by that of the linearized fluid theory. We show how this statistical mechanical ``toy model'' can be useful in addressing various questions about these systems which are relevant in cosmology. Some of these questions are closely analagous to those currently studied in the literature on long range interactions, notably the relation of the evolution of the particle system to that in the Vlasov limit and the nature of approximately quasi-stationary states.

cond-mat.stat-mech↗

arXiv2008

Gravitational dynamics of an infinite shuffled lattice: early time evolution and universality of non-linear correlations

In two recent articles a detailed study has been presented of the out of equilibrium dynamics of an infinite system of self-gravitating points initially located on a randomly perturbed lattice. In this article we extend the treatment of the early time phase during which strong non-linear correlations first develop, prior to the onset of ``self-similar'' scaling in the two point correlation function. We establish more directly, using appropriate modifications of the numerical integration, that the development of these correlations can be well described by an approximation of the evolution in two phases: a first perturbative phase in which particles' displacements are small compared to the lattice spacing, and a subsequent phase in which particles interact only with their nearest neighbor. For the range of initial amplitudes considered we show that the first phase can be well approximated as a transformation of the perturbed lattice configuration into a Poisson distribution at the relevant scales. This appears to explain the ``universality'' of the spatial dependence of the asymptotic non-linear clustering observed from both shuffled lattice and Poisson initial conditions.

cond-mat.stat-mech↗

arXiv2007

Tilings of space and superhomogeneous point processes

We consider the construction of point processes from tilings, with equal volume tiles, of d-dimensional Euclidean space. We show that one can generate, with simple algorithms ascribing one or more points to each tile, point processes which are "superhomogeneous'' (or "hyperuniform''), i.e., for which the structure factor S(k) vanishes when the wavenumber k tends to zero. The exponent of the leading small-k behavior depends in a simple manner on the nature of the correlation properties of the specific tiling and on the conservation of the mass moments of the tiles. Assigning one point to the center of mass of each tile gives the exponent γ=4 for any tiling in which the shapes and orientations of the tiles are short-range correlated. Smaller exponents, in the range 4-d<γ<4 (and thus always superhomogeneous for d\leq 4), may be obtained in the case that the latter quantities have long-range correlations. Assigning more than one point to each tile in an appropriate way, we show that one can obtain arbitrarily higher exponents in both cases. We illustrate our results with explicit constructions using known deterministic tilings, as well as some simple stochastic tilings for which we can calculate S(k) exactly. Our results provide, we believe, the first explicit analytical construction of point processes with γ> 4. Applications to condensed matter physics, and also to cosmology, are briefly discussed.

cond-mat.stat-mech↗

arXiv2007

Two-point correlation properties of stochastic "cloud processes''

We study how the two-point density correlation properties of a point particle distribution are modified when each particle is divided, by a stochastic process, into an equal number of identical "daughter" particles. We consider generically that there may be non-trivial correlations in the displacement fields describing the positions of the different daughters of the same "mother" particle, and then treat separately the cases in which there are, or are not, correlations also between the displacements of daughters belonging to different mothers. For both cases exact formulae are derived relating the structure factor (power spectrum) of the daughter distribution to that of the mother. These results can be considered as a generalization of the analogous equations obtained in ref. [1] (cond-mat/0409594) for the case of stochastic displacement fields applied to particle distributions. An application of the present results is that they give explicit algorithms for generating, starting from regular lattice arrays, stochastic particle distributions with an arbitrarily high degree of large-scale uniformity.

cond-mat.stat-mech↗

arXiv2007

Quantification of discreteness effects in cosmological N-body simulations: II. Evolution up to shell crossing

We apply a recently developed perturbative formalism which describes the evolution under their self-gravity of particles displaced from a perfect lattice to quantify precisely, up to shell crossing, the effects of discreteness in dissipationless cosmological N-body simulations. We give simple expressions, explicitly dependent on the particle density, for the evolution of power in each mode as a function of red-shift. For typical starting red-shifts the effect of finite particle number is to {\it slow down} slightly the growth of power compared to that in the fluid limit (e.g. by about ten percent at half the Nyquist frequency), and to induce also dispersion in the growth as a function of direction at a comparable level. In the limit that the initial red-shift tends to infinity, at fixed particle density, the evolution in fact diverges from that in the fluid limit (described by the Zeldovich approximation). Contrary to widely held belief, this means that a simulation started at a red-shift much higher than the red-shift of shell crossing actually gives a worse, rather than a better, result. We also study how these effects are modified when there is a small-scale regularization of the gravitational force. We show that such a smoothing may reduce the anisotropy of the discreteness effects, but it then {\it increases} their average effect. This behaviour illustrates the fact that the discreteness effects described here are distinct from those usually considered in this context, due to two-body collisions. Indeed the characteristic time for divergence from the collisionless limit is proportional to $N^{2/3}$, rather than $N/ \log N$ in the latter case.

astro-ph↗

arXiv2007

The nef cone volume of generalized Del Pezzo surfaces

We compute a naturally defined measure of the size of the nef cone of a Del Pezzo surface. The resulting number appears in a conjecture of Manin on the asymptotic behavior of the number of rational points of bounded height on the surface. The nef cone volume of a Del Pezzo surface Y with (-2)-curves defined over an algebraically closed field is equal to the nef cone volume of a smooth Del Pezzo surface of the same degree divided by the order of the Weyl group of a simply-laced root system associated to the configuration of (-2)-curves on Y. When Y is defined over a non-closed field of characteristic 0, a similar result holds, except that the associated root system is no longer necessarily simply-laced.

math.AG↗

arXiv2007

Quantification of discreteness effects in cosmological N-body simulations: I. Initial Conditions

The relation between the results of cosmological N-body simulations, and the continuum theoretical models they simulate, is currently not understood in a way which allows a quantification of N dependent effects. In this first of a series of papers on this issue, we consider the quantification of such effects in the initial conditions of such simulations. A general formalism developed in [1] allows us to write down an exact expression for the power spectrum of the point distributions generated by the standard algorithm for generating such initial conditions. Expanded perturbatively in the amplitude of the input (i.e. theoretical, continuum) power spectrum, we obtain at linear order the input power spectrum, plus two terms which arise from discreteness and contribute at large wavenumbers. For cosmological type power spectra, one obtains as expected, the input spectrum for wavenumbers k smaller than that characteristic of the discreteness. The comparison of real space correlation properties is more subtle because the discreteness corrections are not as strongly localised in real space. For cosmological type spectra the theoretical mass variance in spheres and two point correlation function are well approximated {\it above} a finite distance. For typical initial amplitudes this distance is a few times the inter-particle distance, but it diverges as this amplitude (or, equivalently, the initial red-shift of the cosmological simulation) goes to zero, at fixed particle density. We discuss briefly the physical significance of these discreteness terms in the initial conditions, in particular with respect to the definition of the continuum limit of N-body simulations.

astro-ph↗

arXiv2006

Gravitational Dynamics of an Infinite Shuffled Lattice of Particles

We study, using numerical simulations, the dynamical evolution of self-gravitating point particles in static euclidean space, starting from a simple class of infinite ``shuffled lattice'' initial conditions. These are obtained by applying independently to each particle on an infinite perfect lattice a small random displacement, and are characterized by a power spectrum (structure factor) of density fluctuations which is quadratic in the wave number k, at small k. For a specified form of the probability distribution function of the ``shuffling'' applied to each particle, and zero initial velocities, these initial configurations are characterized by a single relevant parameter: the variance $δ^2$ of the ``shuffling'' normalized in units of the lattice spacing $\ell$. The clustering, which develops in time starting from scales around $\ell$, is qualitatively very similar to that seen in cosmological simulations, which begin from lattices with applied correlated displacements and incorporate an expanding spatial background. From very soon after the formation of the first non-linear structures, a spatio-temporal scaling relation describes well the evolution of the two-point correlations. At larger times the dynamics of these correlations converges to what is termed ``self-similar'' evolution in cosmology, in which the time dependence in the scaling relation is specified entirely by that of the linearized fluid theory. Comparing simulations with different $δ$, different resolution, but identical large scale fluctuations, we are able to identify and study features of the dynamics of the system in the transient phase leading to this behavior. In this phase, the discrete nature of the system explicitly plays an essential role.

cond-mat.stat-mech↗

arXiv2006

An alternative to grids and glasses: Quaquaversal pre-initial conditions for N-body simulations

N-body simulations sample their initial conditions on an initial particle distribution, which for cosmological simulations is usually a glass or grid, whilst a Poisson distribution is used for galaxy models, spherical collapse etc. These pre-initial conditions have inherent correlations, noise due to discreteness and preferential alignments, whilst the glass distribution is poorly defined and computationally expensive to construct. We present a novel particle distribution which can be useful as a pre-initial condition for N-body simulations, using a simple construction based on a ``quaquaversal'' tiling of space. This distribution has little preferred orientation (i.e. is statistically isotropic), has a rapidly vanishing large scale power-spectrum (P(k) ~ k^4), and is trivial to create. It should be particularly useful for warm dark matter and cold collapse simulations.

astro-ph↗

arXiv2004

Universality of power law correlations in gravitational clustering

We present an analysis of different sets of gravitational N-body simulations, all describing the dynamics of discrete particles with a small initial velocity dispersion. They encompass very different initial particle configurations, different numerical algorithms for the computation of the force, with or without the space expansion of cosmological models. Despite these differences we find in all cases that the non-linear clustering which results is essentially the same, with a well-defined simple power-law behaviour in the two-point correlations in the range from a few times the lower cut-off in the gravitational force to the scale at which fluctuations are of order one. We argue, presenting quantitative evidence, that this apparently universal behaviour can be understood by the domination of the small scale contribution to the gravitational force, coming initially from nearest neighbor particles.

astro-ph↗

arXiv2003

Bias and the power spectrum beyond the turn-over

Threshold biasing of a Gaussian random field gives a linear amplification of the reduced two point correlation function at large distances. We show that for standard cosmological models this does not translate into a linear amplification of the power spectrum (PS) at small k. For standard CDM type models this means that the``turn-over'' at small k of the original PS disappears in the PS of the biased field for the physically relevant range of the threshold parameter. In real space this difference is manifest in the asymptotic behaviour of the normalised mass variance in spheres of radius R, which changes from the ``super-homogeneous'' behaviour to a Poisson-like behaviour.This qualitative change results from the intrinsic stochasticity of the threshold sampling. While our quantitative results are specific to the simplest threshold biasing model, we argue that our qualitative conclusions should be valid generically for any biasing mechanism involving a scale-dependent amplification of the correlation function. One implication is that the real-space correlation function will be a better instrument to probe for the underlying Harrison Zeldovich spectrum in the distribution of visible matter, as the characteristic asymptotic negative power-law ξ(r) \sim -r^{-4} tail is undistorted by biasing.

astro-ph↗