摘要
积雪不仅作为一种关键的淡水存储方式,也是全球气候系统的重要组成部分之一.目前,基于大地测量型接收机的GNSS-MR(Global Navigation Satellite System Multipath Reflectometry)技术遥测地表特征参数的研究已广泛开展.然而,先前大多的研究利用GPS (Global Positioning System)卫星探测积雪深度并获取了良好的结果,但是其仍存在观测数据单一、反演精度偏低等问题.为增加GNSS-MR技术所使用的数据源且GLONASS卫星数量接近GPS系统,本文提出了基于GNSS-MR算法,采用GLONASS卫星L1和L2载波低卫星高度角(小于25°)的信噪比数据测量了加拿大YEL2跟踪站2015年7月至2016年6月的逐日雪深.该反演值分别与GPS的结果和实测雪深对比,从多路径反射信号与雪深变化量的关系、反演精度以及相关性等多方面进行详细分析,验证了基于GLONASS卫星信号反演地表雪深的适用性和可靠性.结果显示:利用GLONASS卫星信噪比数据的反演雪深与实测雪深间具有高一致性.其中,GPS和GLONASS卫星L1载波的反演雪深无明显差异,两者的RMSE皆在4 cm左右;而GLONASS卫星L2-RMSE为2.6 cm,相关系数达到0.98,其雪深反演的效果明显优于前两种方法.因此,该结果表明采用GLONASS卫星反射信号探测地表雪深能进一步扩展GNSS-MR技术的应用.
Snow is not only a critical storage method of the fresh water resource, but also an important part of the global climate system. Nowadays, GNSS-MR(Global Navigation Satellite System Multipath Reflectometry) with a geodetic receiver have been widely developed and emerged as a new technique for monitoring land-surface characteristic parameter. However, most previous studies only take GPS(Global Positioning System) observations into consideration. Although some good results have been obtained, the drawbacks are still existed for the current GPS-MR model, e.g., insufficient observations and low detection accuracy. To increase more observations for GNSS-MR technique, and the quantity difference of the satellite between GLONASS and GPS is not obvious. In this paper, a new snow depth estimation method based GNSS-MR algorithm using the SNR observations of GLONASS L1-and L2-reflected signals with low elevation angle(less than 25°) to measure daily snow depth from July 2015 to June 2016 at YEL2 station, Canada. For these aspects of the relationship between the reflected signals and snow depth variations, the estimation accuracy and the correlation, the estimations from GLONASS are compared with GPS L1-derived snow depths and in situ measurements, respectively. These results validate the performance of the proposed method and show a high consistency between snow depths estimated using GLONASS signals and in situ measurements. Herein, an absence of any obvious differences for L1-derived snow depths from GPS and GLONASS, with a RMSE of about 4 cm. Additionally, there is a correlation coefficient of 0.98 and a RMSE of 2.6 cm for L2 SNR on snow depth estimation, which is superior to the above two methods. This results indicate that the applications of GNSS-MR technique can be furtherly broadened by using the SNR data of GLONASS.
作者
周威
黄良珂
刘立龙
陈军
李松青
ZHOU Wei;HUANG Liang-ke;LIU Li-long;CHEN Jun;LI Song-qing(College of Geomatic Engineering and Geoinformatics,Guilin University of Technology,Guangxi Guilin 541004,China;Guangxi Key Laboratory of Spatial Information and Geomatics,Guangxi Guilin 541004,China;GNSS Research Center,Wuhan University,Wuhan 430079,China;School of Geodesy and Geomatics,Wuhan University,Wuhan 430079,China)
出处
《地球物理学进展》
CSCD
北大核心
2019年第5期1842-1848,共7页
Progress in Geophysics
基金
国家自然科学基金(41664002,41704027)
广西自然科学基金(2017GXNSFDA198016,2017GXNSFBA198139)
广西“八桂学者”岗位专项
广西空间信息与测绘重点实验室项目(16-380-25-01,15-140-07-34)联合资助
