首页 | 本学科首页   官方微博 | 高级检索  
     检索      

顾及多方向观测值权比反演地球重力场的动力积分法
引用本文:罗志才,周浩,钟波,李琼.顾及多方向观测值权比反演地球重力场的动力积分法[J].地球物理学报,2015,58(9):3061-3071.
作者姓名:罗志才  周浩  钟波  李琼
作者单位:1. 武汉大学测绘学院, 武汉 430079; 2. 武汉大学地球空间环境与大地测量教育部重点实验室, 武汉 430079; 3. 武汉大学测绘遥感信息工程国家重点实验室, 武汉 430079
基金项目:国家自然科学基金项目(41131067,41174020,41374023,41474019);地理信息工程国家重点实验室开放基金项目(SKLGIE2013-M-1-3);地球空间环境与大地测量教育部重点实验室开放基金项目(13-02-05);大地测量与地球动力学国家重点实验室开放基金项目(SKLGED2015-1-3-E)资助.
摘    要:考虑到不同坐标系下各个方向观测值对反演地球重力场的频谱贡献不同,建立了顾及多方向观测值权比的动力积分法,并利用该方法反演了高精度的GOCE HL-SST卫星重力场模型.首先,分析了不同坐标系下各个方向观测值与地球重力场信息的响应关系,其中惯性系(IRF)下X、Z方向的观测值分别对扇谐系数、带谐系数最为敏感,Z方向的解算精度在全频段均略高于X、Y方向;地固系(EFRF)下各个方向的独立解算精度均与能量守恒法的解算精度相当;局部指北坐标系(LNOF)下X、Z和Y三个方向的解算精度依次递减,且Y方向在47阶附近有明显"驼峰"现象.其次,比较了不同坐标系下顾及三个方向观测值权比的加权解算模型,其中加权联合解算模型精度在20至70阶次均明显优于等权解算模型,在带谐项和共振阶次精度提升明显,且LNOF下的加权联合解算精度要优于IRF和EFRF.最后,比较了GOCE和CHAMP卫星的模型解算精度,采用本文计算方法,仅利用2个月GOCE轨道观测值解算的模型精度优于包含更长观测时段信息的AIUB-CHAMP01S和EIGEN-CHAMP03S模型,且略优于ASU-GOCE-2months模型.

关 键 词:动力积分法  加权  GOCE  地球重力场  
收稿时间:2015-02-09

Dynamic integral approach based on weighted multi-direction observations for inversion of the earth's gravity field
LUO Zhi-Cai,ZHOU Hao,ZHONG Bo,LI Qiong.Dynamic integral approach based on weighted multi-direction observations for inversion of the earth's gravity field[J].Chinese Journal of Geophysics,2015,58(9):3061-3071.
Authors:LUO Zhi-Cai  ZHOU Hao  ZHONG Bo  LI Qiong
Institution:1. School of Geodesy and Geomatics, Wuhan University, Wuhan 430079, China; 2. Key Laboratory of Geospace Environment and Geodesy, Ministry of Education, Wuhan University, Wuhan 430079, China; 3. State Key Laboratory of Information Engineering in Surveying, Mapping and Remote Sensing, Wuhan University, Wuhan 430079, China
Abstract:The precise GPS high-low satellite-to-satellite tracking (HL-SST) data is the key supplement to recover the long-wavelength information of the earth's gravity field for the gravity field and steady-state ocean circulation explorer (GOCE) mission. The dynamic integral approach is one of the efficient techniques to determine the spherical harmonic coefficients (SHCs) from GOCE HL-SST observations. However, the traditional dynamic integral approach is based on the inertial reference frame (IRF), and the discrepancy of spectral contribution of multi-direction observations are not fully considered in the SHCs determination. To analyze the contribution of observations in different directions, we build a new dynamic integral approach which takes the weighted ratios of multi-direction observations into consideration. Meanwhile, a dynamic integral approach at different reference frames is also established, which is used to analyze the response relationship between the observations at different frames. The high precision GOCE HL-SST data are used to evaluate the new dynamic integral approach. In the SHCs determination, the conservative perturbing forces are modeled by precise background models, and the non-conservative perturbing forces are observed by onboard instruments. Using the new dynamic integral approach, the response relationship between multi-direction observations in different frames and earth gravity information is analyzed firstly. In the IRF, the observations from X and Z directions are most sensitive to sectorial and zonal coefficients, respectively, and the accuracy of the model recovered from Z direction is higher than those recovered from X and Y directions in the whole frequency band. In the earth-fixed reference frame (EFRF), all of the solutions recovered from individual components are of similar accuracy over all harmonic degrees, and approximately equal to the solution recovered from the energy balance approach (EBA). In the local north-oriented frame (LNOF), the accuracy of solutions decreases from X, Z to Y directions, and there is a hump peaking around degrees 47 in the solution recovered solely from Y direction. Secondly, in terms of the weighted solutions which take the different contribution of each component into consideration, their accuracies between 20 to 70 degrees are higher than the equal weighted ones, and the coefficients in the zonal and resonant area are improved obviously. At the north pole and south pole, the gravity anomalies derived from weighted solutions also present better performance than equal weighted solutions. Although the equal weighted solutions have a good consistency in different frames due to the rotation invariant features, the weighted solutions, which adequately take the relationships between all components, have separate accuracy curves, and they are decreased from IRF, EFRF to LNOF. Thirdly, we compare the gravity field solutions based on 2 months of GOCE HL-SST with those obtained from the challenge mini-satellite payload (CHAMP) mission. Because of considering the contribution of observations in different directions, the GOCE HL-SST solution determined with our dynamic integral approach is slightly better than ASU-GOCE-2months model. Although only using 2 months of GOCE orbits, our solutions are much better than AIUB-CHAMP01S and EIGEN-CHAMP03S models, which are determined from 1 year and 2.8 year of CHAMP data, respectively. From the comparison with equal weighted dynamic integral approach, the numerical results indicate the new dynamic integral approach, which fully considers the contribution of multi-directions, is more suitable for the SHCs determination than the traditional approach. The numerical results also demonstrate that the new dynamic integral approach performance in LNOF is superior to that in IRF and EFRF. Therefore, it is more preferable to adopt the weighted dynamic integral approach in LNOF for SHCs determinations with HL-SST data.
Keywords:Dynamic integral approach  Weighting  GOCE  Earth gravity field model
本文献已被 CNKI 等数据库收录!
点击此处可从《地球物理学报》浏览原始摘要信息
点击此处可从《地球物理学报》下载免费的PDF全文
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号