The problem of small divisors in planetary motion

Small divisors caused by certain linear combinations of frequencies appear in all analytical planetary theories. With the exception of the deep resonance between Neptune and Pluto, they can be removed at the expense of introducing secular and mixed secular terms, limiting the domain in which the solution is valid. Because of them classical solutions are known not to converge uniformly; Poincaré referred to them as asymptotic. The KAM theory shows that if one is far enough from exact commensurability and has small enough planetary masses, expansions exist which will converge to quasi-periodic orbits. Solutions showing very small divisors are excluded from this region of convergence. The question of whether they are intrinsic to the problem or are just manifestations of the method of solution is not settled. Problems with a single commensurabily that can be isolated from the rest of the Hamiltonian may have solutions with no small divisors. The problem of two or more commensurabilities remains unsolved.     

First order planetary perturbations with elliptic functions

The differential equations of planetary theory are solved analytically to first order for the two-dimensional case, using only Jacobian elliptic functions and the elliptic integrals of the first and second kind. This choice of functions leads to several new features potentially of importance for planetary theory. The first of these is that the solutions do not require the expansion of the reciprocal of the distance between two planets, even for those variables which depend on two angular arguments. A second result is that the solution is free from small divisors with the exception of two special resonances. In fact, not only are the solutions for resonant orbits free from small divisors, the perturbations for all variables are expressible in closed form. A subset of the resonant orbits maintains this form and in addition has the remarkable feature that the first order perturbations are purely periodic; they contain no secular terms. A solution for the 13 resonance case is given as an example.     

ITERATIVE SOLUTION TO PERTURBED KEPLER PROBLEM VIA KUSTAANHEIMO–STIEFEL EQUATION

The main purpose of this paper is to supply a proof of formulas for constructing a perturbative solution to the perturbed Kepler problem by utilizing quaternion algebra of the Kustaanheimo–Stiefel formulation. The main advantage of this approach is a removal, from the corresponding solution, of fast oscillations (in the case of conservative forces) and small divisors (in the case of time-dependent forces). This revised version was published online in July 2006 with corrections to the Cover Date.     

Asymptotic solutions in the many-body problem

A three-body problem is considered in which two masses, forming a close binary, orbit a comparatively distant mass. An asymptotic solution of this problem is presented, where the small parameter is related to the distance separating the binary and the remaining mass. Accepting certain model constraints, this solution is accurate within a constant errorO(¹¹) and uniformly valid for time intervalsO(^–3). Two specific examples are chosen to verify the literal solution: one relating to the Sun-Earth-Moon configuration of the solar system, the other to an idealized stellar system where the three masses are in the ratio 20:1:1. In both cases close agreement is found when the analytical solution is compared with an equivalent numerically-generated orbit.     

Asymptotic solutions in the many-body problem

A small particle moves in the vicinity of two masses, forming a close binary, in orbit about a distant mass. Unique, uniformly valid solutions of this four-body problem are found for motion near both equilateral triangle points of the binary system in terms of a small parameter , where the primaries move in accordance with a uniformly-valid three-body solution. Accuracy is maintained within a constant errorO(⁸), and the solutions are uniformly valid as tends to zero for time intervalsO(^–3). Orbital position errors nearL ₄ andL ₅ of the Earth-Moon system are found to be less than 5% when numerically-generated periodic solutions are used as a standard of comparison.     

The Rotation of a Non-Rigid, Non-Symmetrical Earth II: Free Nutations and Dissipative Effects

The study of the rotation of a non-rigid, non-symmetrical Earth with a heterogeneous and stratified liquid core was recently accomplished by González and Getino (1997) through the Hamiltonian formalism. In this work that model is extended by including the effect of the dissipation arising from the mantle–core interaction due to the viscous and electromagnetic coupling. A canonical transformation to a new set of non-singular variables is performed, in order to avoid small divisors in the system of equations. Numerical estimations of the effect of the dissipation are given in form of tables and graphics, and the significance of this effect is discussed. This revised version was published online in July 2006 with corrections to the Cover Date.     

Application of the Method of Matched Asymptotic Expansions to the radial oscillations of a magnetic cylinder

This study deals with the singular character of the perturbation introduced into the eigenvalue problem of the linear and adiabatic oscillations of a gaseous configuration by a magnetic field that is non-zero on the boundary surface of the configuration. This singular character implies that a regular perturbation scheme cannot yield uniformly valid expansion for the eigenfunctions.This investigation considers the application of the Method of Matched Asymptotic Expansions (M.M.A.E.) to the latter singular perturbation problem in order to obtain uniformly valid expansions for the eigenfunctions and first-order expressions for the eigenfrequencies. As an illustrative example, the M.M.A.E. is applied to the eigenvalue problem of the linear, radial, and adiabatic oscillations of a homogeneous cylindrical plasma with a constant longitudinal magnetic field.     

Analytical and numerical manifolds in a symplectic 4-D map

We study analytically the orbits along the asymptotic manifolds from a complex unstable periodic orbit in a symplectic 4-D Froeschlé map. The orbits are given as convergent series. We compare the analytic results by truncating the series at various orders with the corresponding numerical results and we find agreement along a more extended length, as the order of truncation increases. The agreement is improved when the parameters approach those of the stability domain. Along the manifolds no terms with small divisors appear in the series. The same result is found if we use a parametrization method along the asymptotic curves. In the case of orbits starting close to the manifolds small divisors appear, but the orbits remain close to the manifolds for an extended period of time. If the parameters of the map are close to the stable domain the orbits recede and approach the origin several times and remain confined in a certain volume around the origin for a long time before escaping to large distances. For special sets of parameters we see resonance phenomena and the orbits take particular forms near every resonance.     

Solution of the Sitnikov Problem

A new analytic expression for the position of the infinitesimal body in the elliptic Sitnikov problem is presented. This solution is valid for small bounded oscillations in cases of moderate primary eccentricities. We first linearize the problem and obtain solution to this Hill's type equation. After that the lowest order nonlinear force is added to the problem. The final solution to the equation with nonlinear force included is obtained through first the use of a Courant and Snyder transformation followed by the Lindstedt–Poincaré perturbation method and again an application of Courant and Snyder transformation. The solution thus obtained is compared with existing solutions, and satisfactory agreement is found.     

Analytical treatment of air drag and earth oblateness effects upon an artificial satellite

A development of an analytical solution for the motion of an artificial Earth satellite subject to the combined effects of Earth gravity and air drag is presented. The atmospheric model takes into account a linear variation of the density scale height with altitude, the rotation and the oblateness of the atmosphere. The perturbation theory is based upon Lie transforms. The secular and long-periodic terms as well as the short-periodic effects are included in the theory which is valid for small to moderate eccentricities and for all values of the inclination.Belgian National Fund for Scientific Research     

Can the Einstein-de Sitter model adequately describe the Universe at any epoch?

The characteristics of Friedmann model universes containing decoupled matter and radiation are investigated to establish the range in redshift over which the frequently assumed Einstein-de Sitter approximation is valid. This approximation is found to have decidedly doubtful value for small, yet entirely plausible, values of the material density parameter.     

A new analytic approach to the Sitnikov problem

A new analytic approach to the solution of the Sitnikov Problem is introduced. It is valid for bounded small amplitude solutions (z _max = 0.20) (in dimensionless variables) and eccentricities of the primary bodies in the interval (–0.4 < e < 0.4). First solutions are searched for the limiting case of very small amplitudes for which it is possible to linearize the problem. The solution for this linear equation with a time dependent periodic coefficient is written up to the third order in the primaries eccentricity. After that the lowest order nonlinear amplitude contribution (being of order z ³) is dealt with as perturbation to the linear solution. We first introduce a transformation which reduces the linear part to a harmonic oscillator type equation. Then two near integrals for the nonlinear problem are derived in action angle notation and an analytic expression for the solution z(t) is derived from them. The so found analytic solution is compared to results obtained from numeric integration of the exact equation of motion and is found to be in very good agreement. CERN SL/AP     

On current sheet approximations in models of eruptive flares

We consider an approximation sometimes used for current sheets in flux-rope models of eruptive flares. This approximation is based on a linear expansion of the background field in the vicinity of the current sheet, and it is valid when the length of the current sheet is small compared to the scale length of the coronal magnetic field. However, we find that flux-rope models which use this approximation predict the occurrence of an eruption due to a loss of ideal-MHD equilibrium even when the corresponding exact solution shows that no such eruption occurs. Determination of whether a loss of equilibrium exists can only be obtained by including higher order terms in the expansion of the field or by using the exact solution.     

Contraction of satellite orbits using KS uniformly regular canonical elements in an oblate diurnally varying atmosphere

A new non-singular analytical theory for the motion of near Earth satellite orbits with the air drag effect is developed in terms of the Kustaanheimo and Stiefel (KS) uniformly regular canonical elements, by assuming the atmosphere to be oblate diurnally varying with constant density scale height. The series expansions include up to third-order terms in eccentricity and c (a small parameter dependent on the flattening of the atmosphere). Only two of the nine equations are solved analytically to compute the state vector and change in energy at the end of each revolution, due to symmetry in the equations of motion. Numerical comparisons of the important orbital parameters semimajor axis and eccentricity up to 1000 revolutions, obtained with the present solution, with the third-order analytical theories of Swinerd and Boulton and in terms of the KS elements, with respect to the numerically integrated values, show the superiority of the present solution over the other two theories over a wide range of eccentricity, perigee height and inclination.     

Second-order theory of the rotation of an artificial satellite

This work presents the expansion of the second-order of an analytical theory of the attitude evolution of an artificial satellite perturbed by given torques. The first-order of the theory has already been presented by the author in Celestial Mechanics39 (1986) 309–327. It is a theory that is valid under very general conditions including slow rotation and inequal axes of inertia. The present theory is suitable for any internal or external disturbing forces producing the torques. A formal solution is expanded in the second-order according to powers of a small parameter characteristic of the order of magnitude of the disturbing torques. These torques are expanded in Fourier series and the theory applies whatever is the length of these series. The coefficients of the solution are given by an iterative formation law. The comparison of the results with a numerical integration based upon a HIPPARCOS model shows that the second order has brought an improvement to the theory by at least one order of magnitude over the results of the first order.     

Models of young powerful radio sources

Observations of compact symmetric double sources suggest that these objects with physical scales of order tens of parsecs to kiloparsecs are young radio active galactic nuclei. There is, in general, a striking similarity between the structures of these compact objects and the structures of large evolved radio galaxies although the latter are two to three orders of magnitude larger. This has led to the use of self-similar models of the evolution of radio sources as a framework for interpreting observational data. However, the assumptions on which the self-similar models are based become increasingly less valid on the physical scales which are probed by the observations of the smallest sources. In this paper, the dynamics of sources on these small scales is examined and a model developed which extends earlier work in a self-consistent way to small physical scales. The limit of applicability of the models is identified as is the transition from an early evolutionary phase to the self-similar phase of expansion.     

MHD free-convection flow past an accelerated vertical porous plate in a rotating fluid

The three-dimensional flow of an electrically conducting and incompressible viscous fluid past a uniformly accelerated infinite vertical porous plate is studied in a rotating fluid. The flow is assumed to be at small magnetic Reynolds number so that the induced magnetic field is neglected. An exact solution has been obtained by defining a complex velocity with the help of the Laplace transform method for the Prandtl number equal to unity. The effects of rotation, magnetic and free-convection parameters are discussed for the whole problem. Also, the skin-friction components on the plate are discussed.     

Progress in general planetary theory

Celestial Mechanics and Dynamical Astronomy - When on searches for a planetary theory valid over 1 million years, one can leave in the solution the short period terms whose amplitude are small, and...     

Analytic continuation of periodic orbits from equilibria of two-degree-of-freedom systems in rotating coordinates

A new theory is formulated for the analytic continuation of periodic (and aperiodic) orbits from equilibrium solutions of a two-degree-of-freedom dynamical system in rotating coordinates:% MathType!MTEF!2!1!+-% feaafeart1ev1aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn% hiov2DGi1BTfMBaeXafv3ySLgzGmvETj2BSbqefm0B1jxALjhiov2D% aebbfv3ySLgzGueE0jxyaibaiiYdd9qrFfea0dXdf9vqai-hEir8Ve% ea0de9qq-hbrpepeea0db9q8as0-LqLs-Jirpepeea0-as0Fb9pgea% 0lrP0xe9Fve9Fve9qapdbaqaaeGacaGaaiaabeqaamaabaabcaGcba% acbiGab8xDayaacaGaa8xlaiaa-jdacaWFUbGaeqyXduNaa8xpaiaa% -zfadaWgaaWcbaGccaWF4baaleqaaOGaaiilaiqbew8a1zaacaGaey% 4kaSIaaGOmaiaad6gacaWG1bGaeyypa0Jaa8NvamaaBaaaleaakiaa% -LhaaSqabaGccaGGSaGabmiEayaacaGaeyypa0JaamyDaiaacYcace% WG5bGbaiaacqGH9aqpcqaHfpqDaaa!54CD!\[\dot u - 2n\upsilon = V_x ,\dot \upsilon + 2nu = V_y ,\dot x = u,\dot y = \upsilon \]Away from resonance, a family of nonlinear, normal-mode orbits defines an autonomous velocity field u(x, y), u(x, y) represented by convergent algebraic-series expansions in the two position variables. This approach is useful for determining the global structure of solution curves and nonlinear stability of normal modes using Liapunov's direct method. At resonance, the series coefficients are time dependent because stationary modes are incompatible with the equations of motion. By eliminating small divisors, explicit time dependence provides a natural transition from non-resonance to resonance cases within the same theory.     

First-order theory of weakly eccentric orbital motion

Starting from the analytical theory of perturbed circular motions presented in Celestial Mechanics (Bois, 1994), this paper presents an extended resolution valid also for small eccentricity orbits. The solution is of the first order of a small parameter characterizing the magnitude of disturbing forces. The solution has the form of Fourier series with the coefficients given by iterative formation laws. The solution is free from singularities due to small eccentricity or inclination. As an example of numerical application the equatorial artificial satellite orbits are analyzed. For some high satellite orbits with small eccentricity the difference between the numerical integration and the analytical model does not exceed few centimeters per one revolution.On leave from Astronomical Observatory of A. Mickiewicz University, Soneczna 36, PL60-286 Pozna, Poland.