Water level extraction algorithm based on adaptive weighting and deviation matching of multi-source satellite altimetry data

doi:10.11947/j.AGCS.2024.20230360

Abstract

Abstract:

The extraction of precise water level information from satellite altimetry data is crucial for long-term monitoring of lake and reservoir levels. Using Qinghai Lake as a case study, a 20-year dataset is compiled by integrating altimetry data from four different satellites: Envisat, SARAL, Sentinel-3A, and Sentinel-3B. In this study, an innovative algorithm is proposed for the extraction of water levels from multi-source satellite altimetry data. This algorithm integrates adaptive weighting and deviation matching techniques to enhance the accuracy and reliability of water level extraction. Adaptive weighting involves the selection of suitable correction algorithm models based on various environmental conditions and the determination of unique weight parameters for each altimetry data source, thus standardizing the data. The deviation matching method quantifies qualitative data to maximize the precision of water level extraction. Additionally, an artificial intelligence framework is established to automate and integrate the water level extraction process, streamlining the workflow. Experimental results demonstrate that applying adaptive weighting to multi-source altimetry data characteristic values enables reasonable classification and exhibits strong correlations. This approach provides a robust foundation for generating high-precision, long-term water level records. When combined with the deviation matching method, the correlation between daily extracted water levels and actual measurements exceeds 0.9. By setting a correlation coefficient threshold of 0.8, reliable water level extraction for up to a 5-month duration in a single extraction is achievable. To address long-term water level extraction requirements, a methodology is introduced that combines single-day and multi-day extraction, resulting in the construction of 12 years of continuous high-precision water level records. The obtained results exhibit correlation coefficients exceeding 0.9, mean absolute error (MAE) values within the range of 1.5 cm to 2.0 cm, and root mean square error (RMSE) values ranging from 2.0 cm to 2.5 cm. This success underscores the practical value of the data processing algorithm and model in the context of water level extraction and prediction. In conclusion, this research demonstrates the feasibility and utility of combining artificial intelligence with satellite altimetry in constructing long-term, high-precision water level records for small-scale water bodies.

Key words: multi-source satellite altimetry, adaptive weighting, deviation matching, dataset construction, Qinghai Lake water level

CLC Number:

P228

Xukang XIE, Wei LI. Water level extraction algorithm based on adaptive weighting and deviation matching of multi-source satellite altimetry data[J]. Acta Geodaetica et Cartographica Sinica, 2024, 53(11): 2111-2124.

Figures/Tables 16

Fig.1

Tab.1

Fig.2

Fig.3

Fig.4

Tab.2

Tab.3

Fig.5

Fig.6

Tab.4

Tab.5

Fig.7

Fig.8

Tab.6

Fig.9

Fig.10

References 31

[1]	姜丽光, 刘俊, 张星星. 基于卫星雷达测高技术的湖库动态监测理论、方法和研究进展[J]. 遥感学报, 2022, 26(1): 104-114.
	JIANG Liguang, LIU Jun, ZHANG Xingxing. Monitoring lakes and reservoirs using satellite radar altimetry: theory, methods, and progresses[J]. National Remote Sensing Bulletin, 2022, 26(1): 104-114.
[2]	JIANG Liguang, NIELSEN K, ANDERSEN O B, et al. CryoSat-2 radar altimetry for monitoring freshwater resources of China[J]. Remote Sensing of Environment, 2017, 200: 125-139.
[3]	DU Bin, LI Jiancheng, JIN Taoyong, et al. Synthesis analysis of swot karin-derived water surface heights and local cross-calibration of the baseline roll knowledge error over Lake Baikal[J]. Earth and Space Science, 2021, 8(11): e2021EA001990.
[4]	HOU Jiawei, VAN DIJK A I J M, RENZULLO L J, et al. GloLakes: water storage dynamics for 27 000 lakes globally from 1984 to present derived from satellite altimetry and optical imaging[J]. Earth System Science Data, 2024, 16(1): 201-218.
[5]	JIANG Liguang, NIELSEN K, ANDERSEN O B. Improvements in mountain lake monitoring from satellite altimetry over the past 30 years-lessons learned from Tibetan Lakes[J]. Remote Sensing of Environment, 2023, 295: 113702.
[6]	JIANG Liguang, ZHAO Yanan, NIELSEN K, et al. Near real-time altimetry for river monitoring-a global assessment of Sentinel-3[J]. Environmental Research Letters, 2023, 18(7): 074017.
[7]	ZWALLY H J, ROBBINS J W. New insights to antarctic glacial history and earth dynamics from integration of satellite gravimetry and altimetry measurements of mass changes[C]//Proceedings of 2019 AGU Fall Meeting Abstracts. San Francisco: [s.n.], 2019: C21E-1503.
[8]	ZWALLY H J, ROBBINS J W, LUTHCKE S B, et al. Mass balance of the Antarctic ice sheet 1992—2016: reconciling results from GRACE gravimetry with ICESat, ERS1/2 and EnviSat altimetry[J]. Journal of Glaciology, 2021, 67(263): 533-559.
[9]	YANG Lei, LIU Min, LIU Na, et al. Recovering bathymetry from satellite altimetry-derived gravity by fully connected deep neural network[J]. IEEE Geoscience and Remote Sensing Letter, 2023, 20: 1-5.
[10]	ZHU Chengcheng, YANG Lei, BIAN Hongwei, et al. Recovering gravity from satellite altimetry data using deep learning network[J]. IEEE Transactions on Geoscience and Remote Sensing, 2023, 61: 1-11.
[11]	ABDALLA S, KOLAHCHI A A, ABLAIN M, et al. Altimetry for the future: building on 25 years of progress[J]. Advances in Space Research, 2021, 68(2): 319-363.
[12]	FABLET R, VERRON J, MOURRE B, et al. Improving mesoscale altimetric data from a multitracer convolutional processing of standard satellite-derived products[J]. IEEE Transactions on Geoscience and Remote Sensing, 2018, 56(5): 2518-2525.
[13]	汪海洪, 罗志才, 杨元德, 等. 基于波形分类的近海卫星测高数据自适应重跟踪方法[J]. 测绘学报, 2012, 41(5): 729-734.
	WANG Haihong, LUO Zhicai, YANG Yuande, et al. An adaptive retracking method for coastal altimeter data based on waveform classification[J]. Acta Geodaetica et Cartographica Sinica, 2012, 41(5): 729-734.
[14]	BAO Lifeng, LU Yang, WANG Yong. Improved retracking algorithm for oceanic altimeter waveforms[J]. Progress in Natural Science, 2009, 19(2): 195-203.
[15]	SOTIROPOULOU K F, VAVATSIKOS A P. A decision-making framework for spatial multicriteria suitability analysis using promethee ii and k nearest neighbor machine learning models[J]. Journal of Geovisualization and Spatial Analysis, 2023, 7(2): 20.
[16]	金涛勇, 李建成, 姜卫平, 等. 基于多源卫星测高数据的新一代全球平均海面高模型[J]. 测绘学报, 2011, 40(6): 723-729.
	JIN Taoyong, LI Jiancheng, JIANG Weiping, et al. The new generation of global mean sea surface height model based on multi-altimetric data[J]. Acta Geodaetica et Cartographica Sinica, 2011, 40(6): 723-729.
[17]	李大炜, 李建成, 金涛勇, 等. 利用多代卫星测高资料监测1993～2011年全球海平面变化[J]. 武汉大学学报(信息科学版), 2012, 37(12): 1421-1424.
	LI Dawei, LI Jiancheng, JIN Taoyong, et al. Monitoring global sea level change from 1993 to 2011 using topex and jason altimeter missions[J], Geomatics and Information Science of Wuhan University, 2012, 37(12): 1421-1424.
[18]	YAO Fangfang, LIVNEH B, RAJAGOPALAN B, et al. Satellites reveal widespread decline in global lake water storage[J]. Science, 2023, 380(6646): 743-749.
[19]	姜卫平, 褚永海, 李建成, 等. 利用ENVISAT测高数据监测青海湖水位变化[J]. 武汉大学学报(信息科学版), 2008, 33(1): 64-67.
	JIANG Weiping, CHU Yonghai, LI Jiancheng, et al. Water level variation of Qinghai Lake from altimeteric data[J]. Geomatics and Information Science of Wuhan University, 2008, 33(1): 64-67.
[20]	HUANG Qi, LONG Di, DU Mingda, et al. An improved approach to monitoring Brahmaputra River water levels using retracked altimetry data[J]. Remote Sensing of Environment, 2018, 211: 112-128.
[21]	OKADA S, OHZEKI M, TAGUCHI S. Efficient partition of integer optimization problems with one-hot encoding[J]. Scientific Reports, 2019, 9(1): 13036.
[22]	褚永海, 李建成, 金涛勇, 等. T/P雷达高度计后向散射系数陆地表面观测应用[J]. 大地测量与地球动力学, 2009, 29(3): 104-108.
	CHU Yonghai, LI Jiancheng, JIN Taoyong, et al. Application of T/P altimeter backscatter data to land surface observation[J]. Journal of Geodesy and Geodynamics, 2009, 29(3): 104-108.
[23]	孙明智, 刘新, 汪海洪, 等. 多源卫星测高数据监测拉昂错1992年—2020年水位变化[J]. 遥感学报, 2022, 26(1): 126-137.
	SUN Mingzhi, LIU Xin, WANG Haihong, et al. Monitoring lake level change in La-ang Co from 1992 to 2020 using multi-altimeter data[J]. National Remote Sensing Bulletin, 2022, 26(1): 126-137.
[24]	郭金运, 孙佳龙, 常晓涛, 等. TOPEX/Poseidon卫星监测博斯腾湖水位变及其与NINO3 SST的相关性分析[J]. 测绘学报, 2010, 39(3): 221-226.
	GUO Jinyun, SUN Jialong, CHANG Xiaotao, et al. Water level variation of bosten lake monitored with TOPEX/Poseidon and its correlation with NIN03 SST[J]. Acta Geodaetica et Cartographica Sinica, 2010, 39(3): 221-226.
[25]	THARA D K, PREMASUDHA B G, XIONG Fan. Auto-detection of epileptic seizure events using deep neural network with different feature scaling techniques[J]. Pattern Recognition Letters, 2019, 128: 544-550.
[26]	NI Sen, JIA Pengfei, XU Yang, et al. Prediction of CO concentration in different conditions based on Gaussian-TCN[J]. Sensors and Actuators B: Chemical, 2023, 376(Part B): 133010.
[27]	GU Jiuxiang, WANG Zhenhua, JASON K, et al. Recent advances in convolutional neural networks[J]. Pattern recognition, 2018, 77: 354-377.
[28]	VASWANI A, SHAZEER N, PARMAR N, et al. Attention is all you need[J]. Advances in Neural Information Processing Systems, 2017, 30: 18-27.
[29]	ZHOU Haoyi, ZHANG Shanghang, PENG Jieqi, et al. Informer: beyond efficient Transformer for long sequence time-series forecasting[C]//Proceedings of 2021 AAAI Conference on Artificial Intelligence. San Francisco: [s.n.], 2021: 11106-11115.
[30]	LI Zewen, LIU Fan, YANG Wenjie, et al. A survey of convolutional neural networks: analysis, applications, and prospects[J]. IEEE Transactions on Neural Networks and Learning Systems, 2021, 33(12): 6999-7019.
[31]	HOCHREITER S, SCHMIDHUBER J. Long short-term memory[J]. Neural Computation, 1997, 9(8): 1735-1780.

数据来源	数据频率/Hz	时间尺度	模型修正
Sentinel-3A	20	2016-03-09—2021-12-22	多种重跟踪算法
Sentinel-3B	20	2018-11-28—2021-12-09	多种重跟踪算法
Envisat	20	2010-01-28—2010-10-22	多种重跟踪算法
Envisat	20	2010-11-05—2012-03-29	多种重跟踪算法
Envisat	20	2002-07-04—2009-12-27	模拟实测数据
SARAL	40	2013-03-14—2016-07-02	多种重跟踪算法
模拟数据	无	2012-04-10—2013-03-10	模拟测高数据

主成分	初始特征值			提取载荷平方和			旋转载荷平方和总计
主成分	总计	方差/（%）	累积/（%）	总计	方差/（%）	累积/（%）	旋转载荷平方和总计
1	8.948	35.791	35.791	8.948	35.791	35.791	8.486
2	4.557	18.228	54.019	4.557	18.228	54.019	4.856
3	3.382	13.526	67.545	3.382	13.526	67.545	3.104
4	1.532	6.127	73.673	1.532	6.127	73.673	1.690
5	1.352	5.407	79.080	1.352	5.407	79.080	1.531
6	1.169	4.674	83.754	1.169	4.674	83.754	1.272

主成分	参数	归属	特征	载荷	选取
	peakiness_1_plrm_ku	波峰	限制类因素	0.959	是
	peakiness_2_c	波峰	限制类因素	0.953	是
	peakiness_1_c	波峰	限制类因素	0.953	是
	tb_365	表面亮温	外界因素	0.928	是
	tb_238	表面亮温	外界因素	0.925	是
1	peakiness_1_plrm_ku	波峰	限制类因素	0.922	是
	sig0_ocog_c	后向散射系数	限制类因素	0.921	是
	sig0_ice_plrm_ku	后向散射系数	限制类因素	0.827	否
	sig0_ice_c	后向散射系数	限制类因素	0.718	否
	mod_dry	干对流层	传播修正	0.625	是
	sig0_ocog_ku	后向散射系数	限制类因素	0.606	否
	data	日期	时间特征	0.952	是
	cycle	周期	识别特征	0.951	是
2	in_situ	实测水位值	标签值	0.920	是
2	num	时序	识别特征	0.789	是
	sig0_ocog_ku	后向散射系数	限制类因素	-0.694	否
	sig0_ice_sheet_ku	后向散射系数	限制类因素	-0.743	否
	geoid	大地水准面	限制类因素	0.988	是
3	lon	经度	限制类因素	0.986	是
	lat	纬度	限制类因素	-0.993	否
4	iono_cor_gim	电离层	传播修正	0.892	是
5	ASOE	独热编码	识别特征	0.970	是
6	mod_wet	湿对流层	传播修正	0.653	是

单天算法	时间间隔/d	MAE/cm	RMSE/cm	R²
Ice1	Pass间隔天数	14.38	15.29	0.99
Ocean	Pass间隔天数	34.21	37.29	0.95
Sea-ice	Pass间隔天数	67.89	44.54	0.95
本文算法	1	1.95	2.19	0.91

测高卫星	卫星周期/d	Pass轨迹号	Pass间隔天数/d
Envisat	35	94、479、552	3、13、16、19、22、32
SARAL	35	21、94、479、552	3、13、16、19
S3A	27	135、664	8、19、27
S3B	27	249、664	12、15、27