Highly-reduced dynamic orbits and their use for global gravity field recovery: A simulation study for GOCE

The so-called highly reduced-dynamic (HRD) orbit determination strategy and its use for the determination of the Earth’s gravitational field are analyzed. We discuss the functional model for the generation of HRD orbits, which are a compromise of the two extreme cases of dynamic and purely geometrically determined kinematic orbits. For gravity field recovery the energy integral approach is applied, which is based on the law of energy conservation in a closed system. The potential of HRD orbits for gravity field determination is studied in the frame of a simulated test environment based on a realistic GOCE orbit configuration. The results are analyzed, assessed, and compared with the respective reference solutions based on a kinematic orbit scenario. The main advantage of HRD orbits is the fact that they contain orbit velocity information, thus avoiding numerical differentiation on the orbit positions. The error characteristics are usually much smoother, and the computation of gravity field solutions is more efficient, because less densely sampled orbit information is sufficient. On the other hand, the main drawback of HRD orbits is that they contain external gravity field information, and thus yield the danger to obtain gravity field results which are biased towards this prior information.     

Application of Gravity Gradients in the Process of GOCE Orbit Determination

The possibility of improving the Gravity Field and Steady-State Ocean Circulation Explorer (GOCE) mission satellite orbit using gravity gradient observations was investigated. The orbit improvement is performed by a dedicated software package, called the Orbital Computation System (OCS), which is based on the classical least squares method. The corrections to the initial satellite state vector components are estimated in an iterative process, using dynamic models describing gravitational perturbations. An important component implemented in the OCS package is the 8th order Cowell numerical integration procedure, which directly generates the satellite orbit. Taking into account the real and simulated GOCE gravity gradients, different variants of the solution of the orbit improvement process were obtained. The improved orbits were compared to the GOCE reference orbits (Precise Science Orbits for the GOCE satellite provided by the European Space Agency) using the root mean squares (RMS) of the differences between the satellite positions in these orbits. The comparison between the improved orbits and the reference orbits was performed with respect to the inertial reference frame (IRF) at J2000.0 epoch. The RMS values for the solutions based on the real gravity gradient measurements are at a level of hundreds of kilometers and more. This means that orbit improvement using the real gravity gradients is ineffective. However, all solutions using simulated gravity gradients have RMS values below the threshold determined by the RMS values for the computed orbits (without the improvement). The most promising results were achieved when short orbital arcs with lengths up to tens of minutes were improved. For these short arcs, the RMS values reach the level of centimeters, which is close to the accuracy of the Precise Science Orbit for the GOCE satellite. Additional research has provided requirements for efficient orbit improvement in terms of the accuracy and spectral content of the measured gravity gradients.     

Grim4-C1, -C2p: combination solutions of the global earth gravity field

On the basis of the GRIM4-S1 satellite-only Earth gravity model, being accomplished in a common effort by DGFI and GRGS, a combination solution, called GRIM4-C1, has been derivcd using 1° × 1° mean gravity anomalies and 1° × 1° Seasat altimeter derived mean geoid undulations. In the meantime improvements could be achieved by incorporating more tracking data (GEOSAT, SPOT2-DORIS) into the solution, resulting in the two new parallel versions, the satellite-only gravity model GRIM4-S2 and the combined solution GRIM4-C2p (preliminary). All GRIM4 Earth gravity models cover the spectral gravitational constituents complete up to degree and order 50.In this report the emphasis is on the discussion of the combined gravity models: combination and estimation techniques, capabilities for application in precise satellite orbit computation and accuracies in long wavelength geoid representation. It is shown that with the new generation of global gravity models general purpose satellite-only models are no longer inferior to combination solutions if applied to satellite orbit restitution.     

Some space gravity formulas and the dimensions and the mass of the Earth

Summary Adopting thePizzetti-Somigliana method and using elliptic integrals we have obtained closed formulas for the space gravity field in which one of the equipotential surfaces is a triaxial ellipsoid. The same formulas are also obtained in first approximation of the equatorial flattening avoiding the use of the elliptic integrals. Using data from satellites and Earth gravity data the gravitational and geometric bulge of the Earth's equator are computed. On the basis of these results and on the basis of recent gravity data taken around the equator between the longitudes 50° to 100° E, 155° to 180° E, and 145° to 180° W, we question the advantage of using a triaxial gravity formula and a triaxial ellipsoid in geodesy. Closed formulas for the space field in which a biaxial ellipsoid is an equipotential surface are also derived in polar coordinates and its parameters are specialized to give the international gravity formula values on the international ellipsoid. The possibility to compute the Earth's dimensions from the present Earth gravity data is the discussed and the value ofMG=(3.98603×10²⁰ cm³ sec^–2) (M mass of the Earth,G gravitational constant) is computed. The agreement of this value with others computed from the mean distance Earth-Moon is discussed. The Legendre polinomials series expansion of the gravitational potential is also added. In this series the coefficients of the polinomials are closed formulas in terms of the flattening andMG.Publication Number 327, and Istituto di Geodesia e Geofisica of Università di Trieste.     

基于卫星轨道扰动理论的重力反演算法

为了更充分利用低轨重力卫星的高精度观测数据,根据卫星轨道的扰动理论,导出了应用卫星轨道与星间距离观测值联合反演地球重力场模型的算法.该算法的实质是将牛顿运动方程在卫星轨道处进行展开,转化为第二类Volterra积分方程,并采用基于移动窗口的9次多项式内插公式进行数值求解.给出了该算法的观测方程,用QR分解法消去局部参数矩阵,最后采用预条件共轭梯度法求解法方程.利用GRACE卫星2008-01-01~2008-08-01时间段内的轨道及星间距离观测数据,解算了120阶次的地球重力场模型SWJTU-GRACE01S,该模型在120阶处的阶方差为1.58×10^-8,大地水准面差距累计误差为22.29 cm,与美国GPS水准网比较的标准差为0.793 m,结果表明:SWJTU-GRACE01S模型精度介于EIGEN-GRACE01S与EIGEN-GRACE02S模型之间,从而验证了该算法的有效性.     

Zero initial partial derivatives of satellite orbits with respect to force parameters violate the physics of motion of celestial bodies

Satellite orbits have been routinely used to produce models of the Earth’s gravity field. In connection with such productions, the partial derivatives of a satellite orbit with respect to the force parameters to be determined, namely, the unknown harmonic coefficients of the gravitational model, have been first computed by setting the initial values of partial derivatives to zero. In this note, we first design some simple mathematical examples to show that setting the initial values of partial derivatives to zero is generally erroneous mathematically. We then prove that it is prohibited physically. In other words, setting the initial values of partial derivatives to zero violates the physics of motion of celestial bodies. Supported by a Grant-in-Aid for Scientific Research (Grant No. B19340129)     

Gravity field contribution analysis of GOCE gravitational gradient components

A gravity field model is computed from the four accurate gravitational gradient components of GOCE (Gravity field and steady-state Ocean Circulation Explorer), combined with the analysis of the kinematic orbits, and some moderate constraint (or stabilization) in the polar areas where no observation from GOCE is available due to the orbit geometry. The normal matrix of each component is computed individually in order to study its contribution to the combined solution. The results show that the contribution of V_zz is the largest, with an average value of 32.74% of the total solution; the second and the third largest are V_zz and V_yy, with average values of 28.04% and 26.08%, respectively; the component V_xz contributes 11.81%. Validation with external data shows that each component has its characteristic value and that the information content of the component V_xz is not negligible and should be included for gravity field recovery. The orbit part as derived from high-low satellite-to-satellite tracking (SST-hl) to the GPS contributes mostly to the coefficients below degree and order (d/o) 20, and to non-zonal coefficients from d/o 20 to 80. The mean value of the contribution of the polar stabilization is the smallest with a value of 0.22%, nevertheless it is important. In addition to the contribution analysis in terms of the normal matrices, each individual component of the gradiometer has been combined with SST and polar stabilization, to give a set of single component gravity field models. These partially combined solutions are compared to the fully combined solution in terms of geoid differences. They show that the partially combined solution with V_zz is closest to the complete solution. Even closer is a combination with V_xx and V_yy. In addition to the GOCE-only solution, a GOCE-GRACE (Gravity Recovery And Climate Experiment) combined gravity field model is derived and the information content of GOCE and an available set of normal equations of GRACE are investigated. Results show that, as expected, GRACE dominates the solution below degree 90 and GOCE above degree 140.     

利用动力学方法解算GRACE时变重力场研究

本文利用动力学方法建立GRACE(Gravity Recovery And Climate Experiment)K波段距离变率(KBRR)观测、轨道观测与重力场系数的观测方程,通过GRACE Level 1B观测数据,成功解算出全球月时变重力场模型——IGG时变重力场模型,并将2008—2009年的解算结果与GRACE三大数据处理机构美国德克萨斯大学空间中心CSR(Center for Space Research)、美国宇航局喷气推进实验室JPL(Jet Propulsion Laboratory)和德国地学研究中心GFZ(GeoForschungs Zentrum)发布的最新全球时变重力场模型进行详细对比分析.结果表明:IGG结果在全球质量异常、中国及周边地区质量异常的趋势变化、全球质量异常均方差、2~60每阶位系数差值以及亚马逊流域和撒哈拉沙漠等典型区域平均质量异常等方面与CSR、JPL和GFZ解算的RL05结果较为一致.其中,IGG解算结果在2~20阶与CSR、GFZ和JPL最新解算结果基本一致,20~40阶IGG解算结果与GFZ、JPL单位最新解算结果较为接近,大于40阶IGG结果介于CSR与GFZ、JPL之间;亚马逊流域平均质量异常周年振幅IGG、CSR、GFZ和JPL获取到的结果分别为17.6±1.1cm、18.9±1.2cm、17.8±0.9cm和18.9±1.0cm等效水柱高.利用撒哈拉沙漠地区的平均质量异常做反演精度评定,IGG、CSR、GFZ和JPL的时变重力场获取到的平均质量异常均方差分别为1.1cm、0.9cm、0.8cm和1.2cm,表明IGG解算结果与CSR、GFZ和JPL最新发布的RL05结果在同一精度水平.     

Impact of GOCE Level 1b data reprocessing on GOCE-only and combined gravity field models

The reprocessing of Gravity field and steady-state Ocean Circulation Explorer (GOCE) Level 1b gradiometer and star tracker data applying upgraded processing methods leads to improved gravity gradient and attitude products. The impact of these enhanced products on GOCE-only and combined GOCE+GRACE (Gravity Recovery and Climate Experiment) gravity field models is analyzed in detail, based on a two-months data period of Nov. and Dec. 2009, and applying a rigorous gravity field solution of full normal equations. Gravity field models that are based only on GOCE gradiometer data benefit most, especially in the low to medium degree range of the harmonic spectrum, but also for specific groups of harmonic coefficients around order 16 and its integer multiples, related to the satellite’s revolution frequency. However, due to the fact that also (near-)sectorial coefficients are significantly improved up to high degrees (which is caused mainly by an enhanced second derivative in Y direction of the gravitational potential — V_YY), also combined gravity field models, including either GOCE orbit information or GRACE data, show improvements of more than 10% compared to the use of original gravity gradient data. Finally, the resulting gradiometry-only, GOCE-only and GOCE+GRACE global gravity field models have been externally validated by independent GPS/levelling observations in selected regions. In conclusion, it can be expected that several applications will benefit from the better quality of data and resulting GOCE and combined gravity field models.     

Low-degree gravity change from GPS data of COSMIC and GRACE satellite missions

This paper demonstrates estimation of time-varying gravity harmonic coefficients from GPS data of COSMIC and GRACE satellite missions. The kinematic orbits of COSMIC and GRACE are determined to the cm-level accuracy. The NASA Goddard's GEODYN II software is used to model the orbit dynamics of COSMIC and GRACE, including the effect of a static gravity field. The surface forces are estimated per one orbital period. Residual orbits generated from kinematic and reference orbits serve as observables to determine the harmonic coefficients in the weighted-constraint least-squares. The monthly COSMIC and GRACE GPS data from September 2006 to December 2007 (16 months) are processed to estimate harmonic coefficients to degree 5. The geoid variations from the GPS and CSR RL04 (GRACE) solutions show consistent patterns over space and time, especially in regions of active hydrological changes. The monthly GPS-derived second zonal coefficient closely resembles the SLR-derived and CSR RL04 values, and third and fourth zonal coefficients resemble the CSR RL04 values.     

基于B spline和正则化算法的低轨卫星轨道平滑

本文提出了一个利用纯几何轨道和力模型的新算法来计算精确且相对平滑的卫星轨道. 该法将一个纯几何轨道表达为一个B spline的线性组合，线性组合的系数可以由最小二乘法估计获得. 力模型通过计算加速度来附加约束. 为了平衡几何轨道的点位误差和加速度的不精确，一个基于“广义交互确认（GCV，generalized cross validation）”的正则化算法运用其中. 由于B spline的本地控制性，该方法的计算效率相当高. 本文的数值分析表明了该法的有效性. 模拟计算的结论是：带加速度约束较不带加速度约束的平滑效果好. 力模型越精确，平滑的轨道就越精确. 三个月的CHAMP实测轨道数据处理结果表明，平滑后的轨道改进了重力场模型.     

基于下一代四星转轮式编队系统精确和快速反演FSCF地球重力场

第一,由于重力卫星编队轨道的稳定性设计是建立下一代高精度和高空间分辨率地球重力场模型的关键,因此为保证下一代四星转轮式编队系统的稳定性,轨道根数的最优设计如下:(1)轨道半长轴a、轨道偏心率e、轨道倾角i和升交点赤经Ω保持不变;(2)每对卫星的近地点幅角ω和平近点角M分别相差180°;(3)初始近地点辐角ω设置于赤道处,初始平近点角M设计于极点处;(4)卫星编队系统椭圆轨道的半长轴和半短轴之比为2:1. 第二,基于下一代四星转轮式编队系统,利用星间速度插值法,通过相关系数(激光干涉测量系统的星间速度0.85、GPS接收机的轨道位置和轨道速度0.95、星载加速度计的非保守力0.90)、观测时间30天和采样间隔10 s,反演了120阶FSCF-1/2/3/4(Four-Satellite Cartwheel Formation)地球重力场,在120阶处累计大地水准面精度为1.162×10^-4 m,较目前GRACE地球重力场精度至少提高一个数量级. 第三,下一代四星转轮式编队系统具有低轨道高度、高精度测量、全张量观测、弱混频效应和强时变信号的优点.     

联合星载GPS和KBRR星间测速数据反演GRACE时变重力场模型

高精度GRACE卫星时变重力场反演一直是卫星重力测量中的难题.为了恢复高精度的时变地球重力场模型,本文联合GRACE卫星的星载GPS和KBR星间测速观测数据,在对GRACE卫星进行精密定轨的同时,解算出60阶月平均地球重力场模型.通过对GRACE卫星的定轨精度、星载GPS相位和KBR星间测速数据的拟合残差以及时变地球重力场模型解算精度等分析,表明:(1)与美国宇航局喷气推进实验室(JPL)发布的约化动力学精密轨道相比,本文确定GRACE卫星轨道三维位置误差小于5 cm.(2)星载GPS相位数据拟合残差为5~8 mm,KBR星间测速数据拟合残差为0.18~0.30μm·s~(-1).(3)解算的月平均重力场模型与美国德克萨斯大学空间研究中心(CSR)、德国地学研究中心(GFZ)和JPL发布的RL05模型精度接近,时变信号在全球范围内具有很好的空间分布一致性.通过计算亚马逊流域和长江流域的水储量变化,本文与上述三个机构的计算结果无明显差异,且相关系数均达0.9以上.可见,本文建立的卫星轨道与重力场同解算法具有反演高精度GRACE时变重力场能力,为我国卫星重力场反演提供了重要的技术支持.     

The Earth's gravity field

The Earth's gravity field can be determined from gravity measurements made on the surface of the Earth, and through the analysis of the motion of Earth satellites. Gravity data can be used to solve the boundary value problem of gravimetric geodesy in various ways, from the classical formulation using a geoid to the concept of a reference surface interior to the masses of the Earth to a statistical method. We now have gravity information for 1⁰ data blocks over 46% of the Earth's surface and more than several million point measurements available.Satellite observations such as range, range-rate, and optical data have been analyzed to determine potential coefficients used to describe the Earth's gravitational potential field. Coefficients, in a spherical harmonic expansion to degree 12, can be determined from satellite data alone, and to at least degree 20 when the satellite data is combined with surface gravity material. Recent solutions for potential coefficients agree well to degree 4, but with increasing disagreement at higher degrees.     

CHAMP型卫星定轨顾及非线性改正的轨道扰动方程

本文针对CHAMP型卫星建立了顾及非线性改正的轨道扰动方程定轨理论与方法.首先从卫星运动的二阶微分方程出发,引入了正常引力位以及相应的参考轨道,然后分别推导了线性化轨道扰动方程与顾及非线性改正的轨道扰动方程,同时说明了建立的线性化轨道扰动方程与目前处理CHAMP卫星数据的动力学定轨方法是等价的.其次分别对线性化轨道扰动方程与顾及非线性改正的轨道扰动方程的精度进行了估计,在卫星定位精度为3cm与非惯性力测量精度为3×10~(-10)m·s~(-2)的前提下证明了下列结论:当参考轨道与实际轨道之间的距离ρ≤4.7m时线性化轨道扰动方程的精度能达到非惯性力的测量精度以及当ρ≤4.14×10~3m时顾及非线性改正的轨道扰动方程能达到非惯性力的测量精度.由此便可得出结论:相对于线性化轨道扰动方程,顾及非线性改正的轨道扰动方程具有更高的精度,且适合在更长的时间弧段上建立关于引力场位系数的法方程组,特别是针对CHAMP卫星计划进行的模拟计算也完全验证了该结论.最后利用叠加原理,给出了顾及非线性改正的轨道扰动方程的求解方法.此外,还针对GRACE卫星计划利用顾及非线性改正的轨道扰动方程进行了恢复引力场的模拟计算,结果表明:分段建立位系数的法方程组时子弧段分别取值2h、1d、6 d对恢复引力场的结果几乎不产生影响,这表明在处理GRACE数据时能够以6d的弧长来建立法方程组.     

Determination of the parameters of a selenocentric reference system and the deflections of the vertical at the lunar surface

Summary Stokes' constants and, the selenocentric constant, and the angular velocity of the rotation of the Moon define the shape of the external equiselenopotential surfaces, generalized in dependence on the degree N of the harmonics preserved. The scale factor for lengths was computed on the basis of absolute gravity measurement made by the first lunarlanding mission Apollo11 at the landing site[1] under the assumption of a sufficient accuracy of the Stokes' constants used[2, 15]. Anyway, the numerical solution here is only to be considered as an example of the application of the outlined theoretical method, inclusive of the parameters of the lunar reference system, which will be made considerably more accurate when gravity measurements at more points of the lunar surface are available.Presented at the XVth IUGG General Assembly, Moscow, July 30 – August 14, 1971.     

The Use of Resonant Orbits in Satellite Geodesy: A Review

Dynamic resonance, arising from commensurate (orbital or rotational) periods of satellites or planets with each other, has been a strong force in the development of the solar system. The repetition of conditions over the commensurate periods can result in amplified long-term changes in the positions of the bodies involved. Such resonant phenomena driven by the commensurability between the mean motion of certain artificial Earth satellites and the Earth’s rotation originally contributed to the evaluation and assessment of the Stokes parameters (harmonic geopotential coefficients) that specify the Earth’s gravitational field. The technique constrains linear combinations of the harmonic coefficients that are of relevant resonant order (lumped coefficients). The attraction of the method eventually dwindled, but the very accurate orbits of CHAMP and GRACE have recently led to more general insights for commensurate orbits applied to satellite geodesy involving the best resolution for all coefficients, not just resonant ones. From the GRACE mission, we learnt how to explain and predict temporary decreases in the resolution and accuracy of derived geopotential parameters, due to passages through low-order commensurabilities, which lead to low-density ground-track patterns. For GOCE we suggest how to change a repeat orbit height slightly, to achieve the best feasible recovery of the field parameters derived from on-board gradiometric measurements by direct inversion from the measurements to the harmonic geopotential coefficients, not by the way of lumped coefficients. For orbiters of Mars, we have suggestions which orbits should be avoided. The slow rotation of Venus results in dense ground-tracks and excellent gravitational recovery for almost all orbiters.     

利用运动学轨道提高GRACE时变重力场解算

基于变分方程法,本文利用GARCE高精度K波段星间测速数据KBRR,结合德国格拉茨大学发布的运动学轨道和GFZ发布的简动力学轨道作为两种伪观测值,分别解算了2005-2010年60阶全球时变重力场模型Hust-IGG01与Hust-IGG02.通过与GRACE官方机构发布的模型和其他国际主流权威模型进行对比,发现基于运动学轨道结合KBRR解算的模型Hust-IGGO1优于基于简动力学轨道结合KBRR解算的模型Hust-IGG02:在重力场系数C_(20)时间序列的统计数据上,Hust-IGG01比Hust-IGG02更接近SLR结果,在如C_(60)、C_(70)、C_(80)以及C_(90)等重力场低阶项上的数学统计均更接近CSR RL05;Hust-IGG01的重力场系数误差分布和GFZ RL05在同一水平,而Hust-IGG02的误差估计过于乐观;Hust-IGG02在主要质量变化区域上存在5%~10%信号低估,而Hust-IGG01能完全达到国际主流机构利用GPS观测数据的解算水平,Hust-IGG01与官方机构CSR、JPL和GFZ最新模型在格陵兰岛的冰川消融年际趋势分别是-125.4、-125.4、-127.3、-124.3 Gt·a~(-1),在亚马逊流域的平均等效水高周年振幅分别是17.56、17.40、17.46、17.22 cm,在撒哈拉沙漠的平均等效水高均方差分别是0.87、0.77、1.10、0.87 cm;另外在Hust-IGG01的实际应用上,本文分析了全球32个主要流域质量变化的年际趋势、周年振幅和半周年振幅三种信号模式,统计结果显示Hust-IGG01与CSR RL05结果基本吻合.     

基于时空域混合法利用Kaula正则化精确和快速解算GOCE地球重力场

为了研究卫星重力梯度技术对中高频地球重力场反演精度的影响,本文基于时空域混合法,利用Kaula正则化反演了250阶GOCE地球重力场.模拟结果表明:第一,时空域混合法是精确和快速求解高阶地球重力场的有效方法;第二,Kaula正则化是降低正规阵病态性的重要方法;第三,基于改进的预处理共轭梯度迭代法可快速求解大型线性方程组...     

Ellipsoidal vertical deflections and ellipsoidal gravity disturbance: Case studies

First, we present three different definitions of the vertical which relate to (i) astronomical longitude and astronomical latitude as spherical coordinates in gravity space, (ii) Gauss surface normal coordinates (also called geodetic coordinates) of type ellipsoidal longitude and ellipsoidal latitude and (iii) Jacobi ellipsoidal coordinates of type spheroidal longitude and spheroidal latitude in geometry space. Up to terms of second order those vertical deflections agree to each other. Vertical deflections and gravity disturbances relate to a reference gravity potential. In order to refer the horizontal and vertical components of the disturbing gravity field to a reference gravity field, which is physically meaningful, we have chosen the Somigliana-Pizzetti gravity potential as well as its gradient. Second, we give a new closed-form representation of Somigliana-Pizzetti gravity, accurate to the sub Nano Gal level. Third, we represent the gravitational disturbing potential in terms of Jacobi ellipsoidal harmonics. As soon as we take reference to a normal potential of Somigliana-Pizzetti type, the ellipsoidal harmonics of degree/order (0,0), (1,0), (1, − 1), (1,1) and (2,0) are eliminated from the gravitational disturbing potential. Fourth, we compute in all detail the gradient of the gravitational disturbing potential, in particular in orthonormal ellipsoidal vector harmonics. Proper weighting functions for orthonormality on the International Reference Ellipsoid are constructed and tabulated. In this way, we finally arrive at an ellipsoidal harmonic representation of vertical deflections and gravity disturbances. Fifth, for an ellipsoidal harmonic Gravity Earth Model (SEGEN: http://www.uni-stuttgart.de/gi/research/paper/coefficients/coefficients.zip) up to degree/order 360/360 we compute the global maps of ellipsoidal vertical deflections and ellipsoidal gravity disturbances which transfer a great amount of geophysical information in a properly chosen equiareal ellipsoidal map projection.