针对卫星测高数据质量在湖库地区不稳定问题, 提出一种基于数据质量评价、筛选提取水位的方法。使用 Jason-3 卫星测高数据, 选取数据质量不稳定的洪泽湖地区为例进行实验。结果表明, 该方法在改善数据精度方面明显优于传统方法, 提取的测高水位与实测水位间相关系数从传统方法的 0.11 提高到 0.59, 均方根误差也从 2 m 减少到 0.5 m, 使得 Jason-3 数据用于湖库水位监测时具有较高的可信度。此外, 对于那些数据质量不好的周期, 提取的水位精度普遍不高, 基于数据质量评价结果将它们舍去, 可进一步提高了整体的监测精度, 相关系数可提高到 0.9, 均方根误差减少到 0.19 m, 这对于为无资料湖库构建精确的库容曲线具有重要意义。

This paper proposes a method based on data quality evaluation and extraction of water level to improve the steadiness of dat a quality of satellite altimetry in lakes and reservoirs area. We used Jason-3 satellite altimetry data and performed a case study in Hongze Lake area where the altimetric data quality was unsteady. Results showed that the accuracy of this method was obviously better than that of the traditional methods. The correlation coefficient between the satellite-derived water level and the gauged water level increased from 0.11 to 0.59, and the root mean square error was reduced from 2 m to 0.5 m, making the Jason-3 data more reliable for water level monitoring of lakes and reservoirs. In addition, for those periods with poor data quality, since the water level accuracy is gener ally low, they can be discarded based on the results of data quality evaluation. This can further improve the overall monitoring accuracy. The correlation coefficient can be increased to 0.9, and the root mean square error was reduced to 0.19 m. This is of great significance to building a precise capacity curve for the ungauged lakes and reservoirs.

国家重点研发计划项目 (2017YFC0405803) ; 中国水利水电科学研究院基本科研业务费专项 (WR0145B012017; WR0145B272016)