Optimizing expansions of the Earth’s gravity potential and its derivatives over ellipsoidal harmonics

A set of spherical harmonics is the most widely used representation of the Earth’s gravity potential. This series converges outside and on the surface of a reference sphere enveloping the Earth. However, the Earth’s surface is better approximated by the reference ellipsoid—a compressed ellipsoid of revolution that covers the entire Earth. The gravity potential can be expanded in a series over ellipsoidal harmonics on the surface of the reference ellipsoid and on the surface of other external confocal ellipsoids of revolution. In contrast to spherical harmonics, depending on the associated Legendre functions of the first kind, ellipsoidal harmonics depend also on the associated Legendre functions of the second kind. The latter contain the very slowly converging hypergeometric Gauss series. The number of series increases with increasing the order of their derivatives. In this work, we derived new series for the gravitational potential of the Earth and its derivatives over ellipsoidal harmonics. Starting from the first order derivative, all the series corresponding to higher order derivatives depend on the same two hypergeometric Gauss series. The latter converges considerably faster than that for the hypergeometric series previously used when computing the gravity potential and its derivatives.     

The Phobos gravitational field modeled on the basis of its topography

The parameters of the best-fitting ellipsoid have been derived using the latest spherical harmonics of the Phobos topography (Duxbury, 1989) by solution of non-linear overdetermined inverse problem. The lengths of the equatorial axes of the ellipsoid have been determined (a = 12.9 km, b = 11.4 km). They are nearly the same as established by Duxbury (ibid.) on the basis of the linearized relationship between the squared lengths of ellipsoidal axes and the topography coefficients C ₂₀ and C _22. The length of the polar axis (c = 9.1 km) differs of about 20% from Duxbury's value. Supposing mass homogeneity of Phobos, the Stokes parameters of the external gravitational field have been derived up to those of the sixth degree and order. The large irregularities in the Phobos figure cause the values of the Duxbury's potential coefficients be fairly inaccurate except the harmonics C ₂₀, C ₃₂, S ₄₃ and S ₅₁, i.e. linearized relationship between gravity and topography cannot be applied for Phobos. Finally, positions of the centre of figure and the directions of the principal axes of inertia have been established.     

Effect of topographic bias on geoid and reference ellipsoid of Venus, Mars, and the Moon

Since the continuation of an external gravity field inside topographic masses by a harmonic function results in topographic bias, geoid computation by means of global gravity models (GGMs) in terms of external-type series of spherical harmonics, at locations where the GGMs are evaluated inside the topographic masses, will be biased. Consequently, if the reference ellipsoid is defined based on the geoid, it will also be biased. In this paper, the effects of topographic bias on the geoid and reference ellipsoid of Venus, Mars, and the Moon are studied. Moreover, a thorough error analysis in the geoid and reference ellipsoid computation is presented, and it is shown that the estimated standard deviation (STD) of the geoid potential value, the geoidal heights, and the semimajor and semiminor axes of the reference ellipsoid are independent of the topographic bias. According to the results, the effects of topographic bias on the geoid potential value and the semimajor and semiminor axes of the reference ellipsoid in comparison with their estimated STDs are insignificant for Venus, Mars, and the Moon. Moreover, the effect of topographic bias on the geoidal heights of Venus as compared with the estimated STD of its geoidal heights is insignificant. However, the effects of topographic bias on the geoidal heights of Mars and the Moon can be significant, especially in high mountains such as the Tharsis volcanic region on Mars.     

On the theory of the oblateness of the Sun

In this paper we first emphasize why it is important to know the successive zonal harmonics of the Sun's figure with high accuracy: mainly fundamental astrometry, helioseismology, planetary motions and relativistic effects. Then we briefly comment why the Sun appears oblate, going back to primitive definitions in order to underline some discrepancies in theories and to emphasize again the relevant hypotheses. We propose a new theoretical approach entirely based on an expansion in terms of Legendre's functions, including the differential rotation of the Sun at the surface. This permits linking the two first spherical harmonic coefficients (J ₂ and J ₄) with the geometric parameters that can be measured on the Sun (equatorial and polar radii). We emphasize the difficulties in inferring gravitational oblateness from visual measurements of the geometric oblateness, and more generally a dynamical flattening. Results are given for different observed rotational laws. It is shown that the surface oblateness is surely upper bounded by 11 milliarcsecond. As a consequence of the observed surface and sub-surface differential rotation laws, we deduce a measure of the two first gravitational harmonics, the quadrupole and the octopole moment of the Sun: J ₂=−(6.13±2.52)×10⁻⁷ if all observed data are taken into account, and respectively, J ₂=−(6.84±3.75)×10⁻⁷ if only sunspot data are considered, and J ₂=−(3.49±1.86)×10⁻⁷ in the case of helioseismic data alone. The value deduced from all available data for the octopole is: J ₄=(2.8±2.1)×10⁻¹². These values are compared to some others found in the literature. Supplementary material to this paper is available in electronic form at http://dx.doi.org/10.1023/A:1005238718479     

Nonsingular expansions of the gravity potential and its derivatives at satellite altitudes in the ellipsoidal coordinate system

The series in ellipsoidal harmonics for derivatives of the Earth’s gravity potential are used only on the reference ellipsoid enveloping the Earth due to their very complex mathematical structure. In the current study, the series in ellipsoidal harmonics are constructed for first- and second-order derivatives of the potential at satellite altitudes; their structure is similar to the series on the reference ellipsoid. The point P is chosen at a random satellite altitude; then, the ellipsoid of revolution is described, which passes through this point and is confocal to the reference ellipsoid. An object-centered coordinate system with the origin at the point P is considered. Using a sequence of transformations, the nonsingular series in ellipsoidal harmonics is constructed for first and second derivatives of the potential in the object-centered coordinate system. These series can be applied to develop a model of the Earth’s potential, based on combined use of surface gravitational force measurements, data on the satellite orbital position, its acceleration, or measurements of the gravitational force gradients of the first and second order. The technique is applicable to any other planet of the Solar System.     

Gravity field of Mars from Mariner 9 tracking data

Further reduction of Doppler tracking data from Mariner 9 confirms our earlier conclusion that the gravity field of Mars is considerably rougher than the fields of either the Earth or the Moon. The largest positive gravity anomaly uncovered is in the Tharsis region which is also topographically high and geologically unusual. The best determined coefficients of the harmonic expansion of the gravitational potential are: J₂ = (1.96 ± 10.01) × 10^?3 ; C₂₂ = ?(5.1 ± 0.2) × 10^?5; and S₂₂ = (3.4 ± 0.2) × 10^?5. The other coefficients have not been well determined on an individual basis, but the ensemble yields a useful model for the gravity field for all longitudes in the vicinity of 23° South latitude which corresponds to the periapse position for the orbiter.The value obtained for the inverse mass of Mars (3 098 720 ± 70 M⊙^?1) is in good agreement with prior determinations from Mariner flyby trajectories. The direction found for the rotational pole of Mars, referred to the mean equinox and equator of 1950.0, is characterized by α = 317°.3 ± 0°.2, δ = 52°.7 ± 0°.2. This result is in excellent agreement with Sinclair's recent value, determined from earth-based observations of Mars' satellites, but differs by about 0°.5 from the previously accepted value. Other important physical constants that have either been refined or confirmed by the Mariner 9 data include: (i) the dynamical flattening, f = (5.24 ± 0.02) × 10^?3; (ii) the maximum principal moment of inertia, C = (0.375 ± 0.006) MR²; and (iii) the period of precession of Mars' pole, P ? (1.73 ± 0.03) × 10⁵ yr, corresponding to a rate of 7.4 sec of arc per yr.     

Long-Term Stability Of Geoidal Geopotential FromTopex/Poseidon Satellite Altimetry 1993–1999

The TOPEX/POSEIDON (T/P) satellite altimeter data from January 1, 1993to October 24, 1999 (cycles 11–261) was used for investigating thelong-term variations in the geoidal geopotential W₀ and/orin the geopotential scale factor R₀ = GM/W₀ (GM is theadopted geocentric gravitational constant). The mean valuesdetermined for the whole period covered are: W₀ =(62 636 856.161 ± 0.002) m² s^-2, R₀ =(6 363 672.5448 ± 0.0002) m. The actual accuracy is limited bythe altimeter calibration error (2–3 cm) and it isestimated to be about ± 0.5 m² s^-2 (± 5 cm).The yearly variations of the above mean values are at the formalerror level. No long-term trend in W₀, representing the oceanvolume change, was found for the seven years period 1993–9 on thebasis of T/P altimeter (AVISO) data. No sea surface topography modelwas used in the solution. This revised version was published online in July 2006 with corrections to the Cover Date.     

Mars high resolution gravity fields from MRO, Mars seasonal gravity, and other dynamical parameters

With 2 years of tracking data collection from the MRO spacecraft, there is noticeable improvement in the high frequency portion of the spherical harmonic Mars gravity field. The new JPL Mars gravity fields, MRO110B and MRO110B2, show resolution near degree 90. Additional years of MGS and Mars Odyssey tracking data result in improvement for the seasonal gravity changes which compares well to global circulation models and Odyssey neutron data and Mars rotation and precession (). Once atmospheric dust is accounted for in the spacecraft solar pressure model, solutions for Mars solar tide are consistent between data sets and show slightly larger values (k₂ = 0.164 ± 0.009, after correction for atmospheric tide) compared to previous results, further constraining core models. An additional 4 years of Mars range data improves the Mars ephemeris, determines 21 asteroid masses and bounds solar mass loss (dGM_Sun/dt < 1.6 × 10⁻¹³ GM_Sun year⁻¹).     

The early thermal evolution of Mars

Hf‐W isotopic systematics of Martian meteorites have provided evidence for the early accretion and rapid core formation of Mars. We present the results of numerical simulations performed to study the early thermal evolution and planetary scale differentiation of Mars. The simulations are confined to the initial 50 Myr (Ma) of the formation of solar system. The accretion energy produced during the growth of Mars and the decay energy due to the short‐lived radio‐nuclides ²⁶Al, ⁶⁰Fe, and the long‐lived nuclides, ⁴⁰K, ²³⁵U, ²³⁸U, and ²³²Th are incorporated as the heat sources for the thermal evolution of Mars. During the core‐mantle differentiation of Mars, the molten metallic blobs were numerically moved using Stoke's law toward the center with descent velocity that depends on the local acceleration due to gravity. Apart from the accretion and the radioactive heat energies, the gravitational energy produced during the differentiation of Mars and the associated heat transfer is also parametrically incorporated in the present work to make an assessment of its contribution to the early thermal evolution of Mars. We conclude that the accretion energy alone cannot produce widespread melting and differentiation of Mars even with an efficient consumption of the accretion energy. This makes ²⁶Al the prime source for the heating and planetary scale differentiation of Mars. We demonstrate a rapid accretion and core‐mantle differentiation of Mars within the initial ~1.5 Myr. This is consistent with the chronological records of Martian meteorites.     

Dimension of the Earth's General Ellipsoid

The problem of specifying the Earth's mean (general)ellipsoid is discussed. This problem has been greatly simplified in the era of satellite altimetry, especially thanks to the adopted geoidal geopotential value, W₀ = (62 636 856.0 ± 0.5) m² s^-2.Consequently, the semimajor axis a of the Earth's mean ellipsoid can be easily derived. However, an a priori condition must be posed first. Two such a priori conditions have been examined, namely an ellipsoid with the corresponding geopotential that fits best W₀ in the least squares sense and an ellipsoid that has the global geopotential average equal to W₀. It has been demonstrated that both a priori conditions yield ellipsoids of the same dimension, with a–values that are practically identical to the value corresponding to the Pizzetti theory of the level ellipsoid: a = (6 378 136.68 ± 0.06) m.     

Theory of the motion of an artificial Earth satellite

An improved analytical solution is obtained for the motion of an artificial Earth satellite under the combined influences of gravity and atmospheric drag. The gravitational model includes zonal harmonics throughJ ₄, and the atmospheric model assumes a nonrotating spherical power density function. The differential equations are developed through second order under the assumption that the second zonal harmonic and the drag coefficient are both first-order terms, while the remaining zonal harmonics are of second order.Canonical transformations and the method of averaging are used to obtain transformations of variables which significantly simplify the transformed differential equations. A solution for these transformed equations is found; and this solution, in conjunction with the transformations cited above, gives equations for computing the six osculating orbital elements which describe the orbital motion of the satellite. The solution is valid for all eccentricities greater than 0 and less than 0.1 and all inclinations not near 0^o or the critical inclination. Approximately ninety percent of the satellites currently in orbit satisfy all these restrictions.     

Martian gravity field model and its time variations from MGS and Odyssey data

We present results of several years of research and data processing aimed at modelling the Mars gravity field and its longest wavelength time variations. The new solution includes tracking data from Mars Global Surveyor (MGS) from 1998 to 2006 (end of mission) and from Mars Odyssey from 2002 to the spring of 2008; this is the longest analyzed data set from these two orbiter missions as compared to previous works. The new model has been obtained by a team working in Europe, independently from the works of groups at NASA Jet Propulsion Laboratory (JPL) and Goddard Space Flight Center (GSFC), also with totally independent software. Observations consist in two and three-way Doppler measurements (also one way for MGS), and range tracking data collected by the Deep Space Network and have been processed in 4 day arcs, taking into account all disturbing forces of gravitational and non-gravitational origins; for each arc the state vector, drag and solar pressure model multiplying factors, and angular momentum dump parameters are adjusted. The static field (MGGM08A) is represented in spherical harmonics up to degree and order 95 and is very close to previously published models (in terms of spectral components and also over specific features); correlations with the global Mars topography are established and apparent depths of compensation by degree are derived. Lumped zonal harmonics of degree two and three are solved for every 10 days, exhibiting variations in line with previous results (including authors’ ones); the work also shows the difficulty of finding clean signatures (annual and semi-annual) for the zonal coefficient of second degree. The k₂ Love number is also derived from the ensemble of data, as well as from subsets of them; values between 0.110 and 0.130 are found, which are consistent with the existence of a Martian fluid core of significant radius.     

Analytical Solution of Coupled Perturbation of Tesseral Harmonic Terms of Mars's Non-Spherical Gravitational Potential

The non-spherical gravitational potential of the planet Mars is sig- nificantly different from that of the Earth. The magnitudes of Mars’ tesseral harmonic coefficients are basically ten times larger than the corresponding val- ues of the Earth. Especially, the magnitude of its second degree and order tesseral harmonic coefficient J_2,2 is nearly 40 times that of the Earth, and approaches to the one tenth of its second zonal harmonic coefficient J₂. For a low-orbit Mars probe, if the required accuracy of orbit prediction of 1-day arc length is within 500 m (equivalent to the order of magnitude of 10−4 standard unit), then the coupled terms of J2 with the tesseral harmonics, and even those of the tesseral harmonics themselves, which are negligible for the Earth satellites, should be considered when the analytical perturbation solution of its orbit is built. In this paper, the analytical solutions of the coupled terms are presented. The anal- ysis and numerical verification indicate that the effect of the above-mentioned coupled perturbation on the orbit may exceed 10⁻⁴ in the along-track direc- tion. The conclusion is that the solutions of Earth satellites cannot be simply used without any modification when dealing with the analytical perturbation solutions of Mars-orbiting satellites, and that the effect of the coupled terms of Mars's non-spherical gravitational potential discussed in this paper should be taken into consideration.     

Solutions and periodicity of satellite relative motion under even zonal harmonics perturbations

Finding relative satellite orbits that guarantee long-term bounded relative motion is important for cluster flight, wherein a group of satellites remain within bounded distances while applying very few formationkeeping maneuvers. However, most existing astrodynamical approaches utilize mean orbital elements for detecting bounded relative orbits, and therefore cannot guarantee long-term boundedness under realistic gravitational models. The main purpose of the present paper is to develop analytical methods for designing long-term bounded relative orbits under high-order gravitational perturbations. The key underlying observation is that in the presence of arbitrarily high-order even zonal harmonics perturbations, the dynamics are superintegrable for equatorial orbits. When only J ₂ is considered, the current paper offers a closed-form solution for the relative motion in the equatorial plane using elliptic integrals. Moreover, necessary and sufficient periodicity conditions for the relative motion are determined. The proposed methodology for the J ₂-perturbed relative motion is then extended to non-equatorial orbits and to the case of any high-order even zonal harmonics (J _2n , n ≥ 1). Numerical simulations show how the suggested methodology can be implemented for designing bounded relative quasiperiodic orbits in the presence of the complete zonal part of the gravitational potential.     

Oxygen and carbon isotope ratios in the martian atmosphere

Oxygen and carbon isotope ratios in the martian CO₂ are key values to study evolution of volatiles on Mars. The major problems in spectroscopic determinations of these ratios on Mars are uncertainties associated with: (1) equivalent widths of the observed absorption lines, (2) line strengths in spectroscopic databases, and (3) thermal structure of the martian atmosphere during the observation. We have made special efforts to reduce all these uncertainties. We observed Mars using the Fourier Transform Spectrometer at the Canada–France–Hawaii Telescope. While the oxygen and carbon isotope ratios on Mars were byproducts in the previous observations, our observation was specifically aimed at these isotope ratios. We covered a range of 6022 to 6308 cm⁻¹ with the highest resolving power of ν/δν=3.5×10⁵ and a signal-to-noise ratio of 180 in the middle of the spectrum. The chosen spectral range involves 475 lines of the main isotope, 184 lines of ¹³CO₂, 181 lines of CO¹⁸O, and 119 lines of CO¹⁷O. (Lines with strengths exceeding 10⁻²⁷ cm at 218 K are considered here.) Due to the high spectral resolution, most of the lines are not blended. Uncertainties of retrieved isotope abundances are in inverse proportion to resolving power, signal-to-noise ratio, and square root of the number of lines. Laboratory studies of the CO₂ isotope spectra in the range of our observation achieved an accuracy of 1% in the line strengths. Detailed observations of temperature profiles using MGS/TES and data on temperature variations with local time from two GCMs are used to simulate each absorption line at various heights in each part of the instrument field of view and then sum up the results. Thermal radiation of Mars' surface and atmosphere is negligible in the chosen spectral range, and this reduces errors associated with uncertainties in the thermal structure on Mars. Using a combination of all these factors, the highest accuracy has been achieved in measuring the CO₂ isotope ratios: ¹³C/¹²C = 0.978 ± 0.020 and ¹⁸O/¹⁶O = 1.018 ± 0.018 times the terrestrial standards. Heavy isotopes in the atmosphere are enriched by nonthermal escape and sputtering, and depleted by fractionation with solid-phase reservoirs. The retrieved ratios show that isotope fractionation between CO₂ and oxygen and carbon reservoirs in the solid phase is almost balanced by nonthermal escape and sputtering of O and C from Mars.     

Constraining the gravitational redshift of a neutron star from the Σ-meson well depth

The effect of the Σ-meson well depth on the gravitational redshift is examined within the framework of relativistic mean field theory for the baryon octet system. It is found that, for a stable neutron star, the gravitational redshift increases with the central energy density increase or with the mass increase but decreases as the radius increases. Considering a change of U_S^(N)U_{\Sigma}^{(N)} from −30 MeV to 30 MeV, for a stable neutron star the gravitational redshift near to the maximum mass increases. In addition, it is also found that the growth of the U_S^(N)U_{\Sigma}^{(N)} makes the gravitational redshift as a function of M _max/R increase, the higher the U_S^(N)U_{\Sigma}^{(N)} the less the change in the gravitational redshift.     

On the computation of the spherical harmonic terms needed during the numerical integration of the orbital motion of an artificial satellite

A method is presented for the accurate and efficient computation of the forces and their first derivatives arising from any number of zonal and tesseral terms in the Earth's gravitational potential. The basic formulae are recurrence relations between some solid spherical harmonics,V _n,m, associated with the standard polynomial ones.     

Changes in the Sun’s mass and gravitational constant estimated using modern observations of planets and spacecraft

More than 635 thousand positional observations of planets and spacecraft of various types (mostly radiotechnical ones, 1961–2010) were used to estimate possible changes in the gravitational constant, Sun’s mass, and semi-major axes of planetary orbits, as well as the associated value of the astronomical unit. The observations were analyzed based on the EPM2010 ephemerides constructed at the Institute of Applied Astronomy (Russian Academy of Sciences) in a post-Newtonian approximation as a result of simultanious numerical integration of the equations of motion of nine major planets, the Sun, the Moon, asteroids, and trans-Neptunian objects. The heliocentric gravitational constant GM _⊙ was found to vary with a rate of (GṀ _⊙/GM _⊙ = (−5.0 ± 4.1)) × 10⁻¹⁴ per year (at the 3σ level). The positive secular changes in the semimajor axes ȧ _i /a _i were found for Mercury, Venus, Mars, Jupiter, and Saturn provided by high-precision observations. These changes also correspond to the decrease in the heliocentric gravitational constant. The changing of GM _⊙, itself is probably caused by the loss of the mass M _⊙ of the Sun due to its radiation and solar wind; these effects are partly compensated by the material falling onto the Sun. Allowing for the maximum bounds on the possible change in the Sun’s mass M _⊙, it has been found from the change obtained in GM _⊙ that the annual change Ġ/G of the gravitational constant G falls within the interval −4.2 × 10⁻¹⁴ < ȧ/G < +7.5 × 10⁻¹⁴ with a 95% probability. The astronomical unit (AU) is connected by its definition only with the heliocentric gravitational constant. The decrease of GM _⊙ obtained in this paper should correspond to a secular decrease in the AU. It is shown, however, that the modern level of accuracy does not allow us to determine a change in the AU. The attained posibility of determining changes in GM _⊙ using high-accuracy observations encourages us to have a relation between GM _⊙ and the AU fixed for a certain moment in time, since it is inconvenient to have a time-dependent length for the AU.     

Electron Temperatures and Flow Speeds of the Low Solar Corona: MACS Results from the Total Solar Eclipse of 29 March 2006 in Libya

An experiment was conducted in conjunction with the total solar eclipse on 29 March 2006 in Libya to measure both the electron temperature and its flow speed simultaneously at multiple locations in the low solar corona by measuring the visible K-coronal spectrum. Coronal model spectra incorporating the effects of electron temperature and its flow speed were matched with the measured K-coronal spectra to interpret the observations. Results show electron temperatures of (1.10±0.05) MK, (0.70±0.08) MK, and (0.98±0.12) MK, at 1.1 R _⊙ from Sun center in the solar north, east and west, respectively, and (0.93±0.12) MK, at 1.2 R _⊙ from Sun center in the solar west. The corresponding outflow speeds obtained from the spectral fit are (103±92) km s⁻¹, (0+10) km s⁻¹, (0+10) km s⁻¹, and (0+10) km s⁻¹. Since the observations were taken only at 1.1 R _⊙ and 1.2 R _⊙ from Sun center, these speeds, consistent with zero outflow, are in agreement with expectations and provide additional confirmation that the spectral fitting method is working. The electron temperature at 1.1 R _⊙ from Sun center is larger at the north (polar region) than the east and west (equatorial region).