基于多源卫星测高数据的青海湖水位变化  

Monitoring the water level changes of Qinghai Lake based on multi-source altimetry data

作　　者：毋梦艳 陈鹏[1,2] 李祖峰 杨新越 WU Mengyan;CHEN Peng;LI Zufeng;YANG Xinyue(College of Geomatics,Xi’an University of Science and Technology,Xi’an 710054,China;State Key Laboratory of Geodesy and Earth’s Dynamics,Innovation Academy for Precision Measurement Science and Technolgoy,CAS,Wuhan 430077,China;PowerChina Northwest Engineering,Co.,Ltd.,Xi’an 710065,China)

机构地区：[1]西安科技大学测绘科学与技术学院,陕西西安710054 [2]中国科学院精密测量科学与技术创新研究院大地测量与地球动力学国家重点实验室,湖北武汉430077 [3]中国电建集团西北勘测设计研究院有限公司,陕西西安710065

出　　处：《西安科技大学学报》2025年第1期191-201,共11页Journal of Xi’an University of Science and Technology

基　　金：国家自然科学基金项目(42374037);西安科技大学优秀青年科技基金项目(2018YQ2-10)。

摘　　要：为了更准确地获取各地的水位变化,需要建立时间和空间分辨率更高的水位监测方法。首先,利用Envisat、Cryosat-2和Sentinel-3A这3颗测高卫星分别提取青海湖2002—2010年、2011—2015年和2016—2020年的水位信息,构建统一基准的水体水位的时间序列;然后,结合青海湖的实测水位,并使用均方根误差(RMSE)和相关系数(R)作为精度评估指标;最后,验证3颗雷达测高卫星在青海湖水位反演的精度,基于卡尔曼(Kalman)滤波融合多源测高数据获取了青海湖2002—2020年的水位时间序列。结果表明:青海湖的水位呈逐年上涨趋势,最快以0.36 m/a的趋势在升高;Envisat、Cryosat-2和Sentinel-3A在青海湖的反演水位与实测水位的RMSE分别为0.54,0.13,0.14 m,相关系数R分别为0.36,0.89和0.97;此基础上,使用Kalman滤波获取的多源数据融合反演水位的RMSE和R分别为0.20 m和0.98,较卫星反演水位RMSE降低了17.10%,R提高了5.10%。Kalman滤波的多源测高数据融合反演水位有效弥补了单个卫星的时间分辨率低的缺点,精度较卫星反演水位显著提高,为更多内陆水体水位的变化建立高时空分辨率的水位时间序列奠定了基础。In order to obtain more accurate water level changes in various places,water level monitoring methods with higher temporal and spatial resolution are required.Firstly,the water level information of Qinghai Lake from 2002—2010,2011—2015 and 2016—2020 was extracted by using three altimetry satellites,Envisat,Cryosat-2 and Sentinel-3A,respectively,and a time series of water water level with a unified benchmark was constructed.Then combined with the measured water level of Qinghai Lake,the root mean square error(RMSE)and correlation coefficient(R)were used as accuracy evaluation indicators.Finally,the accuracy of the inversion of the water level of the three radar altimetry satellites in Qinghai Lake was verified.The results show that the water level of Qinghai Lake is increasing year by year,and the fastest trend is 0.36 m/a.The RMSE of the inversion and measured water levels of Envisat,Cryosat-2 and Sentinel-3A in Qinghai Lake are 0.54 m,0.13 m and 0.14 m,respectively,and the correlation coefficients R are 0.36,0.89 and 0.97.Based on Kalman filter fusion multi-source altimetry data,the water level time series of Qinghai Lake from 2002 to 2020 was obtained.The RMSE and R of multi-source data fusion inversion water level obtained by Kalman filter were 0.20 m and 0.98,respectively,which were 17.10% lower and R increased by 5.10% compared with the satellite inversion water level RMSE.In general,the multi-source altimetry data fusion inversion water level of Kalman filtered could effectively compensate for the shortcomings of low temporal resolution of a single satellite,and significantly improve the accuracy of water level inversion compared with satellite,laying the foundation for the establishment of water level time series with high temporal and spatial resolution for more inland water level changes.

关 键 词：卫星测高 多源数据融合 KALMAN滤波 青海湖 水位变化 

分 类 号：P228.3[天文地球—大地测量学与测量工程]