关键词: 水位监测, 卫星测高, 城市湖泊, 波形重跟踪, 武汉东湖

Abstract:

[Objectives] Satellite altimetry has become a crucial method for monitoring lake water levels, yet significant challenges remain in its application to small lakes, particularly in complex urban environments. Currently, limited studies explore the effectiveness of satellite altimetry for monitoring variations of urban lake water levels. Using East Lake in Wuhan as a case study, this study evaluates the quality of Jason-3 satellite altimetry data, aiming to validate the capability of satellite altimetry in monitoring urban lake water level variations. [Methods] Based on the Jason-3 Sensor Geophysical Data Record (SGDR) products from 2017 to 2022, this study used two key parameters—pulse peakiness and waveform width—to first analyze the altimetry waveform characteristics of East Lake. In addition to the original range observations and ICE-retracked ranges provided by SGDR products, this study applied the Offset Center of Gravity (OCOG) and threshold methods for waveform retracking. Among them, the threshold retracking method selected eight threshold levels ranging from 20% to 90% (in 10% increments) to analyze the retracking performance under different thresholds. A robust coarse elimination strategy based on the Median Absolute Deviation (MAD) was employed to eliminate outliers from the water level observation data, followed by the calculation of periodic average water levels to construct the lake water level time series. To evaluate the quality of water level data by different methods, the range and standard deviation of water levels in each period, as well as the number of invalid periods, were statistically analyzed. Additionally, the accuracy of the results using different methods was verified using the measured data from hydrological stations. Finally, meteorological data (precipitation, evaporation) and a water balance model were integrated to quantify the contributions of natural and anthropogenic factors to East Lake’s water level variations. [Results] (1) Statistical analysis of pulse peakiness and waveform width from the lake surface altimetry echoes revealed that approximately 50% of East Lake’s waveforms exhibited specular reflections with distinct sharp peaks, while about 30% displayed complex shapes containing two or more peaks. (2) The results of accuracy validation using the on-site measured data of water levels showed that the 50% threshold retracking method achieved optimal performance, with a root mean square error (RMSE) of 0.108 m and a correlation coefficient of 0.87. (3) Based on the 50% threshold retracking method, and using Jason-3 data, the water level time series of East Lake from September 2017 to February 2022 was established. The results demonstrated that the lake water level remained stable around 19.5 m during this period, with annual fluctuations <0.5 m, monthly variations <0.2 m, and no pronounced seasonal pattern. Although precipitation was the primary water source, water levels showed extremely low correlation with precipitation (R=0.007), and weak negative correlation with evaporation (R=-0.44). According to the analysis of water balance, artificial regulation played a key role in the water level variations of East Lake. [Conclusions] (1) Jason-3 satellite altimetry data can effectively monitor urban lake water level variations, but requires careful data processing, including waveform retracking and outlier elimination. (2) Despite complex waveforms over urban lakes, retracking methods significantly improve altimetry accuracy. Compared with waveform retracking methods such as OCOG, ICE, and threshold method, the 50% threshold method is more suitable for urban lakes.

Key words: water level monitoring, satellite altimetry, urban lake, waveform retracking, East Lake in Wuhan