基于Sentinel-3A SRAL 2级产品的鄱阳湖水位评估与校准

Evaluation and calibration of Sentinel-3A SRAL Level 2 product over Poyang Lake

  • 摘要: Sentinel-3A卫星合成孔径雷达高度计(SRAL)因其时空分辨率优势在水位监测上应用潜力较大。基于2016—2018年Landsat-8 与Sentinel-2 光学遥感获取鄱阳湖星子站邻近水域,提取湖上Sentinel-3A SRAL 2级产品卫星测高点,提出一种卫星测高水位计算与校准方法,并结合实测水位进行评估。结果表明:Sentinel-3A SRAL 2级产品在鄱阳湖的过境数据有效率为64%,3—9月有连续覆盖数据,12月至次年2月受水位低或湖滩出露影响无有效数据;不同高程系统下的卫星观测水位与实测水位序列的一致性极显著,皮尔逊相关系数为0.999,在0.001水平上显著相关,实测水位变化量与卫星观测水位变化量的皮尔逊相关系数为1,二者的平均偏差为−0.175 m,标准差为0.084 m,其中降轨统计指标值优于升轨,枯水期则优于丰水期,以降轨枯水期指标值为最优:平均偏差、均方根误差、标准差分别为−0.082、0.107和0.076 m。以2016—2017年、2017—2018年、2016—2018年卫星测高水位与实测数据的平均偏差作为校准参数,校准水位的平均绝对偏差都为0.073 m,皮尔逊相关系数为1。研究验证了卫星测高数据计算和校准河湖水位方法的有效性,该类数据可应用于水文、气候变化研究与洪旱监测等。

     

    Abstract: The Sentinel-3 Satellite Synthetic Aperture Radar Altimeter (SRAL) data is one of the emerging data sources for monitoring water level. Based on the water body information of Poyang Lake extracted by Landsat-8 and Sentinel-2 optical remote images, the Sentinel-3A SRAL satellite measurements near Xingzi hydrological Station were extracted. According to the principle of satellite altimetry considering geophysical corrections, an outlier elimination rule and automatic calculation of each date satellite water level process were proposed, and measured water level was used to evaluate and calibrate the Sentinel-3A radar altimeter retrieved water level. Results show that the effective rate of Sentinel-3A satellite data in Poyang Lake is 64%, and valid data cover March to September. The Pearson correlation coefficient of satellite retrieved water level is greater than 0.99, and there is a significant correlation at the level of 0.001. The Pearson correlation coefficient between satellite water level change and measured water level change is 1, the average deviation is −0.175 m and STDEVA is 0.084. For ascending and descending orbits, the value of the statistics in descending orbit is better than those in ascending orbit. For high-water and low-water seasons, the values of the statistics in the low-water season are better than those in the high-water season, and the statistical index in the low-water season is the smallest: the average deviation is −0.082 m, RMSE is 0.107 m. The average absolute deviation of the calibration water level using the average deviation between the measured data and satellite altimeter water level in 2016-2017, 2016-2018, 2016-2018 as the calibration parameter both are 0.073 m, and R-value is 1. The water level calibration results based on different calibration parameters have little difference. The study has proved the effectiveness of the method of satellite altimetry data calculation and lake water level calibration, which is conducive to hydrology, climate change research, and monitoring of flood control and drought resistance.

     

/

返回文章
返回