利用GNSS反射信号监测海面高度变化——基于法国BRST站2019~2021年数据  

Change of Sea Surface Height Monitored by GNSS Reflected Signals-Based on Data from BRST Station in France from 2019 to 2021

作　　者：郭斐[1] 李佰瀚 张治宇 刘万科[1] GUO Fei;LI Bai-han;ZHANG Zhi-yu;LIU Wan-ke(School of Geodesy and Geomatics,Wuhan University,Wuhan 430079,Hubei,China)

机构地区：[1]武汉大学测绘学院,湖北武汉430079

出　　处：《地球科学与环境学报》2023年第3期548-558,共11页Journal of Earth Sciences and Environment

基　　金：国家重点研发计划项目(2022YFB3903902);湖北省杰出青年科学基金项目(2021CFA039)。

摘　　要：GNSS反射测量(GNSS-R)技术凭借其数据来源广泛、低成本、高时空分辨率等优势,在地表与海洋环境监测等方面展现巨大潜力,已成为海面高度(SSH)反演的重要技术途径。现有研究大多聚焦于3~6个月内的短期GPS潮位反演,难以反映海面高度的季节性变化及年际特征,且在动态海面改正时仅考虑了垂向速度的影响,忽视了海面波动的垂向加速度,导致低潮位与高潮位的反演精度较差。基于此,以法国某一岸基跟踪站——BRST站为例,利用其连续3年的BDS/GPS/GLONASS/Galileo四系统反射信号,通过Lomb-Scargle谱分析和二阶动态潮位改正模型,采取稳健回归策略反演海面高度,并将最终结果与验潮站观测值进行对比,分析潮位变化趋势。结果表明:GNSS-R技术反演结果与验潮站观测值具有较好的一致性,反演精度有逐年提升的趋势,均方根误差(RMSE)为7.57 cm,相关系数为0.935;海面高度的季节性变化特征明显,秋、冬季平均海面高度偏高,夏季平均海面高度偏低,且海面高度的季节性变化与温度的季节性变化存在着相反的趋势;M2、S2、K1、O1、N2、K2、P1、Q1、M4等9个分潮的振幅差为0.06~6.76 cm,其平均绝对误差(MAE)为1.60 cm,迟角差为0.03°~6.96°,其平均绝对误差为2.45°,在频域上进一步验证了GNSS-R技术监测海面高度变化的可靠性。Global navigation satellite system reflectometry(GNSS-R)has shown great potential in surface and marine environment parameters monitoring,with the advantages of extensive data sources,low cost,high spatial and temporal resolution,and has become an important approach for sea surface height(SSH)inversion.The existing research mainly focuses on the short-term GPS sea surface inversion within 3-6 months,which is difficult to reflect the seasonal variation and interannual characteristics of SSH.Meanwhile,only the influence of vertical velocity is considered in the dynamic sea surface correction,and the vertical acceleration of sea surface fluctuation is ignored,resulting in the poor estimation accuracy of low and high tide levels.For the above issues,a shore-based tracking station BRST in France was taken as an example,and the reflected signals of 3 consecutive years from BDS/GPS/GLONASS/Galileo system were used to estimate SSH.Based on Lomb-Scargle spectrum analysis and second-order dynamic tide level correction model,a robust regression strategy was adopted.The estimation results were compared with tide gauge records to analyze the trend of sea level change.The results show that the inversion results of GNSS-R technology are in good agreement with tide gauge records,and the inversion accuracy has a tendency to improve year by year,with a root mean square error(RMSE)of 7.57 cm and a correlation coefficient of 0.935 for the three-year daily average;the seasonal variation of SSH is obvious,the average SSH is high in autumn and winter,and low in summer,and the seasonal variation of SSH is opposite to that of temperature;the differences of amplitude for nine tidal constituents including M2,S2,K1,O1,N2,K2,P1,Q1 and M4,range from 0.06 cm to 6.76 cm with the mean absolute error(MAE)of 1.60 cm,and the differences of phase range from 0.03°to 6.96°with the MAE of 2.45°.It is further verified that the reliability of GNSS-R technology in monitoring SSH change in the frequency domain.

关 键 词：GNSS反射测量 海面高度 海平面变化 信噪比 反射信号 动态潮位改正 分潮提取 遥感 

分 类 号：P228[天文地球—大地测量学与测量工程]