联合星载GPS和KBRR星间测速数据反演GRACE时变重力场模型

A model of the time-variable gravity field inverted from combined GRACE on-board GPS and KBR range rate data

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.

Recovering the GRACE time-variable gravity field in high precision remains a challenging task of satellite gravity. To solve this problem, we use Gravity Recovery and Climate Experiment （GRACE） on-board GPS observations and K-band range-rate （KBRR） measurements to achieve the orbit of GRACE satellite and monthly temporal gravity field solutions to degree/ order 60 at the same time. The results are discussed in the aspects of orbit determination accuracy, GRACE on-board GPS data phase residuals, K-band range-rate measurements residuals and the precision of temporal gravity field solutions. The results show that： （1） compared with reduced dynamic orbit provided by NASArs Jet Propulsion Laboratory （JPL）, three-dimensional position accuracy of our GRACE orbit solution is better than 5 cm. （2） The residuals of on board GPS phase observations are about 5~8 mm, and the residuals of KBRR are about 0.18~0.30/~m ~ s 1 （3） Comparing the time-varying signal of our GRACE monthly solution with the RL05 models of Centre for Space Research （CSR） and JPL as well as the GeoForschungsZentrum （GFZ） RL05 model, our solutions agree with all of them in terms of the spatial distribution. Mass change signals in the Amazon basin and the Yangtze basin calculated by our model are compared with that of other models. The average correlation coefficients between ShangHai Astronomical observatory （SHA） and the aforementioned three time-variable gravity field models are all greater than 0.9. It is evident that the algorithm presented in this paper has the ability to recover the GRACE time-variable gravity field model with a high precision which achieves the satellite orbits and the gravity field model at the same time. It can provide an important technical support for inversion of the Chinese satellite gravity field. s