共模误差对陆态网测站速度场估计的影响

Influence of Common Mode Error on The Estimation Velocity Field of CMONOC's Stations

本文基于52个陆态网连续运行观测站和中国境内及周边10个IGS站近5年的坐标时间序列,利用相关系数加权的叠加滤波法定量计算坐标序列中存在的共模误差,并使用CATS软件分别估计测站各分量剔除共模误差前后的坐标时间序列的速度及其精度,结果显示:在E、N、U分量上,相对于原始坐标时间序列,由剔除共模误差的坐标时间序列整理而得的测站各分量估计速度与陆态网发布速度偏差的均值分别减小了7.32%,1.87%,7.84%,同时各分量估计速度的精度均值也分别提高了33.43%,21.11%,59.26%,这表明在估计区域的测站速度场时,若不剔除原始坐标序列中存在的共模误差,测站的运动速度及其精度可能会被不恰当地估计,即利用剔除共模误差后的坐标时间序列而得结果理论上更加准确可靠。

This paper adopts the method of correlation coefficient weighted stacking filtering to estimate the common mode error with the five years coordinate sequences of 52 CMONOC’s continuous operation observation stations and 10 IGS stations in or around China. Then uses software CATS to estimate the stations’ velocity before and after removed the CME of each coordinate component. The results show that in the E, N, U component, compared to the original coordinate time series, the average bias between the estimated velocity of the coordinate time series which the common mode error is removed and the released velocity by CMONOC is reduced by 7.32%, 1.87% and 7.84% respectively, and the average estimation accuracy of each component increases by 33.43%, 21.11% and 59.26% respectively, which indicates that if the common mode error exists in the original coordinate sequences, the stations’ velocity and its accuracy may be estimated inappropriately, while the results of estimated velocity and it will be more accurate and reliable after removing the common mode error.