Compatible potentials and orbits in synodic systems

For a given family of orbits f(x,y) = c ^* which can be traced by a material point of unit in an inertial frame it is known that all potentials V(x,y) giving rise to this family satisfy a homogeneous, linear in V(x,y), second order partial differential equation (Bozis,1984). The present paper offers an analogous equation in a synodic system Oxy, rotating with angular velocity . The new equation, which relates the synodic potential function (x,y), = –V(x, y) + % MathType!MTEF!2!1!+-% feaafiart1ev1aaatCvAUfeBSjuyZL2yd9gzLbvyNv2CaerbuLwBLn% hiov2DGi1BTfMBaeXatLxBI9gBaerbd9wDYLwzYbItLDharqqtubsr% 4rNCHbGeaGqiVu0Je9sqqrpepC0xbbL8F4rqqrFfpeea0xe9Lq-Jc9% vqaqpepm0xbba9pwe9Q8fs0-yqaqpepae9pg0FirpepeKkFr0xfr-x% fr-xb9adbaqaaeGaciGaaiaabeqaamaabaabaaGcbaWaaSqaaSqaai% aaigdaaeaacaaIYaaaaaaa!3780!\[\tfrac{1}{2}\]²(x ² + y ²) to the given family f(x,y) = c ^*, is again of the second order in (x,y) but nonlinear.As an application, some simple compatible pairs of functions (x,y) and f(x, y) are found, for appropriate values of , by adequately determining coefficients both in and f.     

Families of planar orbits generated by homogeneous potentials

The second order partial differential equation which relates the potentialV(x,y) to a family of planar orbitsf(x,y)=c generated by this potential is applied for the case of homogeneousV(x,y) of any degreem. It is shown that, if the functionf(x,y) is also homogeneous, there exists, for eachm, a monoparametric set of homogeneous potentials which are the solutions of an ordinary, linear differential equation of the second order. Iff(x,y) is not homogeneous, in general, there is not a homogeneous potential which can create the given family; only if =f _y /f _x satisfies two conditions, a homogeneous potential does exist and can be determined uniquely, apart from a multiplicative constant. Examples are offered for all cases.     

Lie transforms and the Hamiltonization of non-Hamiltonian systems

To develop the perturbation solution of the non-Hamiltonian system of differential equationsy=g(y, t; ), it is sufficient to obtain the perturbation solution of a Hamiltonian system represented by the HamiltonianK=Y·g(y, t; ) which is linear in the adjoint vectorY. This Hamiltonization allows the direct use of the perturbation methods already established for Hamiltonian systems. To demonstrate this fact, a Hamiltonian algorithm developed by this author and based on the Lie-Deprit transform is applied to the Hamiltonized system and is shown to be equivalent to the application of the non-Hamiltonian form of this same algorithm to the original non-Hamiltonian system.     

Theβ index in cool stars

If we follow recent work and in order to extend theuvby photometric calibrations to spectral types later than G0, we present an attempt to use the combineduvby and systems for stars in the range G5-K7 and luminosity classes V to III.The behaviour in the MK-, (b–y)– and cl– diagrams of the 200 stars, only good spectroscopic data being considered, suggests the usefulness of the index as an independent parameter for late-type Main-Sequence stars, following in a natural way the general trend defined by Crawford for F- and G-type stars.Paper presented at the 11th European Regional Astronomical Meetings of the IAU on New Windows to the Universe, held 3–8 July, 1989, Tenerife, Canary Islands, Spain     

Fourier techniques for an analysis of eclipsing binary light curves: III

The aim of the present paper will be to present a new approach to the light changes of eclipsing binary systems. The light changes have been expanded into the Fourier-Bessel and Dini series. The coefficients of these expansions which are the Hankel transforms of the order of the loss of light (1–l) have been expressed in terms of the eclipse elements. These discrete Hankel transformsH (j _m ) andH ( _m ) valid for only the positive real zerosj _m and _m of the functionsJ and [xJ(x)+J (x)] have been generalized for any positive value of andy. Thus, these general expressions for the Hankel transforms of the light curves which are valid for all types of eclipses, for any arbitrary degree of the adopted limb-darkening law and, moreover, for any positive value of the free parameters andy, may be used for the solution of the elements of eclipsing binary systems.     

Mgix emission lines in an active region spectrum obtained with the Solar EUV Rocket Telescope and Spectrograph (SERTS)

Theoretical electron-temperature-sensitive Mgix emission line ratios are presented forR _I =I(443.96 )/I(368.06 ),R ₂ =I(439.17 )/I(368.06 ),R ₃ =I(443.37 )/I(368.06 ),R ₄ =I(441.22 )/I(368.06 ), andR ₅ =I(448.28 )/I(368.06 ). A comparison of these with observational data for a solar active region, obtained during a rocket flight by the Solar EUV Rocket Telescope and Spectrograph (SERTS), reveals excellent agreement between theory and observation forR ₁ throughR ₄, with discrepancies that average only 9%. This provides experimental support for the accuracy of the atomic data adopted in the line ratio calculations, and also resolves discrepancies found previously when the theoretical results were compared with solar data from the S082A instrument on boardSkylab. However in the case ofR ₅, the theoretical and observed ratios differ by almost a factor of 2. This may be due to the measured intensity of the 448.28 line being seriously affected by instrumental effects, as it lies very close to the long wavelength edge of the SERTS spectral coverage (235.46–448.76 ).     

On the adelphic potentials compatible with a set of planar orbits

Given a planar potentialB=B(x, y), compatible with a monoparametric family of planar orbitsf(x, y)=c, we face the problem of producing potentialsA=A(x, y), adelphic toB(x, y), i.e. nontrivial potentials which have in common withB(x, y) the given set of orbits. We establish a linear, second order partial differential equation for a functionP(x, y) and we prove that, to any definite positive solution of this equation, there corresponds a potentialA(x, y) adelphic toB(x, y).     

Limits of stability for an area-preserving polynomial mapping

The Henon-Heiles mappingx=x+a(y–y ³), y=y–a(x–x³) has been studied, with the aim of finding where the unstable regions of the (x, y) plane are. When this mapping is put into the normal form, it is found to be a typical twist mapping. The criteria of Moser (1971) are used to obtain an upper limit to the size of a stable region around the origin, and this limit decreases to zero as the value of the parameter a increases toward 2.0. However, direct calculation fora=1.99 shows that there is a fairly large region insidex=0.412,y=0, from which escape from near the outer boundary requires at least 160 mappings. The region of high stability thus appears to be much larger than any region of absolute stability predicted by the KAM theorem.A general survey has been made of instability regions for the parameter valuea=1.0, this survey having been carried out to the extent which is allowed by a computer with 18-decimal-place accuracy. First, for all thex-axis fixed points (of the above mapping) deemed to be representative and significant, both the locations and variational matrix traces have been calculated. (The latter show whether the fixed point is elliptic or hyperbolic.) Ifn is the number of mappings andk is the number of circuits around the origin, then the listing (Table IV) is for fractionsk/n between 1/6 and 1/22, inclusive. (This covers the range 0x<0.96, withx=0 the fixed point forn=6,k=1).Escape toward infinity can be rapid, with less than 200 mappings necessary to reach the vicinity of then=1 fixed points (atx=±1,y=0 andx=0,y=±1) from outer regions of the (x, y) plane, such as for |x|>0.93,y=0. In this case, the unstable regions may be tongues encircling the origin. However, as the distance from the origin is decreased, the tongues can be replaced by exceedingly fine threads rapidly becoming less than say 10^–16 in thickness. Such a thread issues fromx=0.905468199,y=0 and requires of the order of 40 000 mappings to escape. It does so by spiralling about the origin and penetrating through several series of loops associated with various fixed points at successively greater (absolute) values ofx(y=0). The region between this thread and the origin is therefore highly stable. Practical stability of a region may be regarded as attained when the region is interior to a series of loops for which the trace of the variational matrix is close to 2.0. This occurs forn=53,k=4, with fixed point atx=0.819786,y=0 and Trace=2.0000 0004.If an invariant curve does in fact exist, then one must be able to show that the outward spiralling from a given series of loops is brought to a halt at some stage. This does not occur in the region where direct computation is possible, as we show in this article, and it remains to be seen under what conditions it can take place.     

Linear analysis of the light curves of eclipsing variables

In this paper, general model analysis of the form c _{i i} () for the fractional loss of lightf() exhibited by a close binary system will be considered in the sense of the least-squares criterion. A general recursive method will be established for constructing the normal equations for the most useful functions _i (), and an economical storage of the method on a digital computer, with its computational steps, will also be given. Moreover, a full recursive computational algorithm for the least-squares approximation will also be established. By means of this algorithm, all the solution vectors, the variance for different orders of fit and the corresponding variance-covariance matrix could be computed once and for all and, moreover, recursively. The economical storage of the algorithm and its computational steps will be given. Finally, some of the practical difficulties encountered in the application of the least-squares criterion will be analysed, and some techniques for detecting and controlling these difficulties are also given. Numerical examples on the Algol system using a Fourier cosine series of the form c _i cos [(j - 1) /] will be given for illustration.     

Motion of rigid bodies in a set of redundant variables

In this article we study the conditions for obtaining canonical transformationsy=f(x) of the phase space, wherey(y ₁,y ₂,...,y _2n ) andx(x ₁,x ₂,...,x _2m ) in such a way that the number of variables is increased. In particular, this study is applied to the rotational motion in functions of the Eulerian parameters (q ₀,q ₁,q ₂,q ₃) and their conjugate momenta (Q ₀,Q ₁,Q ₂,Q ₃) or in functions of complex variables (z ₁,z ₂,z ₃,z ₄) and their conjugate momenta (Z ₁,Z ₂,Z ₃,Z ₄) defined by means of the previous variables. Finally, our article include some properties on the rotational motion of a rigid body moving about a fixed point.     

Special Families of Orbits in the Direct Problem of Dynamics

The direct problem of dynamics in two dimensions is modeled by a nonlinear second-order partial differential equation, which is therefore difficult to be solved. The task may be made easier by adding some constraints on the unknown function = f _y /f _x , where f(x, y) = c is the monoparametric family of orbits traced in the xy Cartesian plane by a material point of unit mass, under the action of a given potential V(x, y). If the function is supposed to verify a linear first-order partial differential equation, for potentials V satisfying a differential condition, can be found as a common solution of certain polynomial equations.The various situations which can appear are discussed and are then illustrated by some examples, for which the energy on the members of the family, as well as the region where the motion takes place, are determined. One example is dedicated to a Hénon—Heiles type potential, while another one gives rise to families of isothermal curves (a special case of orthogonal families). The connection between the inverse/direct problem of dynamics and the possibility of detecting integrability of a given potential is briefly discussed.This revised version was published online in October 2005 with corrections to the Cover Date.     

Internal temperature of isothermal neutron star core matter

Two new equations of state obtained for matter constituting isothermal neutron star core by using isentropic nature of matter for the equalities =₂ and =₃ (where 's are usual adiabatic exponents) have been utilised to discuss the internal temperature of core. The temperature of matter has been obtained asT=T _a(P+E)/. Variation ofT/T _a(t) with energy density has been discussed for these new equations of state and some standard equations of state for nuclear matter.     

Chemical evolution of the Galaxy at the initial rapid-collapse phase

Equations for the chemical evolution of the Galaxy are derived, accounting for (i) the dynamical evolution of the Galaxy (i.e. the collapse of the proto-galaxy), and (ii) either a variable mass-spectrum in the birth-rate stellar function of the type (m, t)=(t)(m, t), or a constant mass-spectrum with variable lower mass limit for star birth:m _mf=m_mf(Z). Simple equations are adopted for the collapse of the proto-galaxy, accounting for the experimental data (i.e. axial ratio and major semi-axis) relative to the halo and to the disk, and best fitted for a rapid collapse; gas density is assumed to be always uniform. Numerical computations of several cases show that there is qualitative agreement with the experimental data relative to theZ(t) function when: (i) the mass-spectrum is nearly constant in time: (m, t)(m)=m ^–2.35; (ii) the efficiency (t) is sufficiently high; moreover, the super metallic effect (SME) takes place for greater than a given value (1.5); (iii) the shorter the collapse timeT _c, the more rapid is the initial increase of metallicity, the asymptotic value being left nearly unaltered. The theoretical present-day values of gas density and metallicity so obtained differ from the experimental values by a factor of 2 or 3. Leaving aside other possible explanations, such a discrepancy is within the range of the uncertainties concerning the amount of gas returned back into space by the decay of the stars. Our theoretical results are not in complete agreement with the observed data bearing on theN _n(Z) function (N _n is the number of stars whose Main-Sequence lifetime is not less than the age of the Galaxy), while a hypothesis of star formation with different efficiencies in different zones of the Galaxy, and successive stellar mixing from zone to zone, is not inconsistent with such data.     

Solution of a transfer equation by a Modified Spherical Harmonic Method in a thin anisotropically scattering atmosphere

Wan, Wilson and Sen (1986) have examined the scope of Modified Spherical Harmonic Method in a plane medium scattering anisotropically. They have used the phase functionp(µ, µ) = 1 +aµµ. In this paper, the Transfer Equation has been solved by the Modified Spherical Harmonic Method using the phase functionp(µ, µ) = 1 + ₁ P ₁(µ)P ₁(µ) + ₂)P ₂(µ)P ₂(µ) and a few sets of numerical solution have been predicted for three different cases.     

Simultaneous measurements of the polarization in Hα and D3 prominence emissions

Simultaneous observations of the polarization of radiations H and D₃ have been performed in 1979 at Pic du Midi on 60 solar prominences (300 pairs of measurements). For the essential part of the sample, the degree of polarization p(D₃) does not depend strongly upon the intensity E(D₃); on the contrary p(H) decreases steeply when E(H) goes beyond the brightness which corresponds to (H) = 1 (Figure 4); the deviation of the polarization direction (H) does not show such a strong dependance upon E(H) (Figure 6). A crude estimate of the possible role of multiple scattering in an optically thick prominence is in general agreement with observational data but a detailed comparison will be possible only when a complete theory of the Hanle effect in non optically thin layers will be available.Therefore, H polarization measurements are presently useful mainly when (H) < 1 and, in connexion with D₃ measurements (Figure 8), they should make possible to determine the three components of the prominence magnetic vector. Prospects are given for the observation of other lines (H; 10 830 Å) which do not suffer heavy self-absorption effects and would be well-suited to this type of study.     

Correlations between statistical population indices and spectral characteristics of carbon stars

The coefficients of correlation between spectroscopic data published by Yamashita (1967) and others for carbon stars and the statistical population indices calculated for these stars at the Toru Observatory are calculated (Table II). The intensity estimates of Cai 4227 Å, Nai D lines, the C¹³/C¹² ratio, then=n(Li/Ca) index, as well as CN and probably C₂ bands are higher in population I carbon stars. The CH(G) band and probably hydrogen (H, H, H) lines as well as Baii 4554 and 4934 Å lines are stronger in population II carbon stars. The photoelectric colour indices, corrected for interstellar reddening do not show significant population effects. They can be used as spectral type equivalents. For a population criterion the CH/CN intensity ratio is proposed.     

Comparison between some Hα and X-ray flares

In the present paper, H-evolutive curves of chromospheric events are compared with flux evolutive curves of X-ray events observed at the same time in different spectral regions. A correspondence between the emissions E(I _H/I _chr)'s at higher and higher H-intensity levels, and the X-ray fluxes F()'s in harder and harder -ranges is shown. Further, the present observations seem to indicate the existence of a single triggering mechanism during the flash-phase of a flare. It is also shown that these results may be in agreement with Brown's model for chromospheric flares.     

La transformation de Lyapunov de l'equation de Hill et son interpretation dynamique

Résumé La transformation de Lyapunov transforme une équation de Hill en une autre qui occupe la même place dans la classification de Yakubovich.Soit (C) une solution périodique d'un système conservatif à deux degrés de liberté. D'après le principe de moindre action de Maupertuis (C) est l'image d'une géodésique ().Nous montrons que les équations aux variations au voisinage de (C) et de () sont réductibles à deux équations de Hill qui se correspondent par une transformation de Lyapunov.

---

The Lyapunov transformation of Hill's equation and his dynamic interpretation

The Lyapunov transformation carries Hill's equationÿ+F(t)y=0,F(t+T)=F(t) into another one which belongs to the same class in Yakubovich's classification.Let (C) be a closed trajectory of a Lagrangian conservative system with two degrees of freedom. By the Principle of Least action, we know that (C) is the image of a geodesic () of a certain two-dimensional surface ().We show that the two Hill equations associated with (C) and () are related by a certain Lyapunov transformation.

---

---

Paper presented at the 1981 Oberwolfach Conference on Mathematical Methods in Celestial Mechanics.     

Mean free paths of energetic particles at very large heliodistances (Pioneer 11 at 20 AU)

Pioneer 11 magnetic field data at 20 AU are analysed by the computational method of Moussas, Quenby, and Webb (1975), Moussas and Quenby (1978), and Moussas, Quenby, and Valdes-Galicia (1982a, b) to obtain the parallel mean free path , and the diffusion coefficient parallel to the magnetic field line K . This method is the most appropriate for the mean free path calculation at large heliodistances since the alternative method which is based on fitting of energetic particle intensities cannot be easily and accurately be used because the association of energetic particles with their parent flares is not precise. The results show that the mean free path has values between 0.85 and 0.98 AU, linearly increasing with energy according to (T_kinetic) = + MT, where = 0.846 AU and M = 4.44 × 10 ^–5 AU MeV^–1 for energies between 10 MeV and 3 GeV for protons. These values of the parallel mean free path are much larger than the values estimated by previous studies up to 6 AU. The diffusion coefficient dependence upon energy follows a relation which simply reflects an almost constant mean free path and a linear dependence on the velocity of the particle, so that at 20 AU heliodistance K (T _kin) = K _{, 1 MeV}(T _kin)T _kinetic , with = 1/2. The distance dependence of the parallel diffusion mean free path follows a power law, (R) = _{, 1 AU} R , where is 1 ± 0.1. While the parallel diffusion coefficient obeys a power-law relation with heliodistance R, K (R, T _kin) = K _{, 1 AU}(T _kin)R , with = 1 ± 0.1. The radial diffusion coefficient of cosmic rays is not expected to strongly depend upon the parallel diffusion coefficient because the nominal magnetic field at these large heliodistances (20 AU) is almost perpendicular to the radial direction and the contribution of the diffusion coefficient perpendicular to the magnetic field is expected to play a dominant role. However, the actual garden hose angle varies drastically and for long time periods and hence the contribution of the diffusion parallel to the field may continue to be important for the small scale structure of intensity gradients.     

The mass-ratio distribution of spectroscopic binary stars

In order to determine the mass-ratio distribution of spectroscopic binary stars, the selection effects that govern the observations of this class of binary systems are investigated. The selection effects are modelled numerically and analytically. The results of the models are compared to the data inThe Eighth Catalogue of the Orbital Elements of Spectroscopic Binary Stars (DAO8) compiled by Battenet al. (1989). The investigations involve binary systems with Main-Sequence primary components only, in order to avoid confusion of evolutionary and selection effects.For single-lined spectroscopic binaries (SBI) it is found that the mass ratios (q=M _sec/M _prim) in general adhere to a distribution _q q ^-2 forq>q ₀, withq ₀=0.3. The observations are consistent with a distribution that is flat forq<q ₀. The turn-over value varies fromq ₀=0.3 for systems with B-type primaries, toq ₀=0.55 for systems with K-type primaries. The semi-major axesa ₁ are distributed according to _a (a ₁)a ₁ ^-a with an average value of _a =1.3. The power varies from _a =1.7 for systems with B-type primaries to _a =0 for systems with K-type primaries. The eccentricitiese of the orbits of SBI systems are distributed according to _e (e)e ^-1.For double-lined spectroscopic binary stars (SBII) it is found that the shape of theq-distribution, as derived from observations, is almost entirely determined by selection effects. It is shown that the distribution is compatible with theq-distribution found for SBI systems. A sub-sample, consisting of the SBII systems from DAO8 with magnitudesm _V 5 ^m , is less hampered by selection effects, and shows the same shape of theq-distribution as the SBI systems, at theq-interval (0.67, 1).It is estimated that 19–45% of the stars in the solar neighbourhood are spectroscopic binary systems.