摘要
针对内陆湖泊水位序列构建及水位预测难度较大的问题,该文以鄱阳湖为例,首先通过高斯拟合提取ICESat-2激光测高数据的水面光子,进而计算湖泊水位并构建水位序列;其次,使用长短期记忆网络预测水位;最后结合降雨量数据对鄱阳湖水位变化情况进行分析。实验结果表明:构建的水位序列与验证数据集相关性R^(2)为0.79,相关系数为0.89。水位预测结果与水位序列的均方根误差为0.05。验证了ICESat-2数据在内陆湖泊高精度水位计算中的应用前景。
Aiming at the problem that it is difficult to construct the water level sequence and predict the water level of inland lakes,this paper takes Poyang Lake as an example.Firstly,the surface photons of ICESat-2 laser altimetry data are extracted by Gaussian fitting,and then the lake water level is calculated and the water level sequence is constructed.Secondly,the long short-term memory network is used to predict the water level.Finally,combined with rainfall data,the water level change of Poyang Lake was analyzed.The experimental results show that the correlation R^(2) between the water level sequence constructed in this paper and the validation data set is 0.79,and the correlation coefficient is 0.89.The root mean square error between the water level prediction result and the water level sequence is 0.05.The application prospect of ICESat-2 data in high-precision water level calculation of inland lakes is verified.
作者
田时岳
王昶
何佳洋
王旭
莫凡
温振
TIAN Shiyue;WANG Chang;HE Jiayang;WANG Xu;MO Fan;WEN Zhen(School of Civil Engineering,University of Science and Technology,Anshan,Liaoning 114051,China;School of Resources and Civil Engineering,Liaoning Institute of Science and Technology,Benxi,Liaoning 117004,China;Land Satellite Remote Sensing Application Center,Beijing 100048,China;School of Surveying and Spatial Information,Shandong University of Science and Technology,Qingdao,Shandong 266590,China)
出处
《测绘科学》
CSCD
北大核心
2023年第12期105-114,共10页
Science of Surveying and Mapping
基金
辽宁省教育厅面上项目(LJKMZ20220638)
辽宁科技学院博士科研启动金项目(2307B29)
自然资源部国土卫星遥感应用重点实验室基金项目(BN2302-6)
国家自然科学基金青年基金项目(42001416)
关键词
鄱阳湖
ICESat-2
水位序列
长短期记忆网络
水位预测
Poyang Lake
iCESat-2
water level sequence
long short-term memory network
water level prediction
