中文摘要：

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,计算时间较直接最小二乘法至少降低1000倍;第四,基于卫星轨道误差1 cm和卫星重力梯度误差3×10^-12 /s^2,在250阶处反演累计大地水准面和重力异常的精度分别为9.295 cm和0.204 mGal. 第五,论证了基于国际GRACE和GOCE卫星计划反演高精度和高空间分辨率地球重力场的互补性.

英文摘要：

The GOCE Earth＇s gravitational field complete up to degree and order 250 is recovered based on the time-space domain method associated with Kaula regularization in order to study the influences of satellite gravity gradiometry on the accuracy of medium-high frequency Earth＇s gravitational field recovery. The simulated results show： Firstly, the time-space domain method is an effective way to accurately and rapidly determine the high-degree Earth＇s gravitational field; Secondly, the Kaula regularization is one of the key processes to reduce ill condition of normal matrix; Thirdly, the large-scale linear system of equations is solved quickly using the improved pre-conditioned conjugate-gradient iterative approach, and the computing time can be reduced at least 1000 times as compared to the direct least-squares approach; Fourthly, at the degree 250, cumulative geoid height and gravity anomaly errors are 9.295 cm and 0.204 mGal with orbital error 1 cm and gravity gradient error 3×10^-12/s^2, respectively. Finally, the complementarity of high-accuracy and high-resolution Earth＇s gravitational field recovery between international GRACE and GOCE missions is demonstrated.