星载光子点云数据冰川高程变化监测误差改正方法研究

Study on error correction method for glacier elevation change monitoring using spaceborne photon point cloud data

针对利用ICESat-2星载光子点云数据进行冰川高程变化监测时,不同时期数据差异及地形因素会导致冰川高程变化量估算误差较大的问题,该文以青藏高原唐古拉山脉冰川为例,利用2018—2022年ICESat-2光子点云和SRTM数据,提出了组合坡度、标准差及点间距离3个指标的ICESat-2光子点云优选策略,减小了不同时期光子数据差异引起的冰川高程变化量估算误差;进而构建了基于分段线性拟合函数的冰川高程变化量估算误差改正模型,改正了因地形坡度引起的误差;估算了唐古拉山脉冰川2000—2022年平均高程变化速率和2018—2022年高程逐年变化量。结果表明,青藏高原唐古拉山脉冰川2000—2022年平均高程变化速率为-0.41±0.43 m/a,与国家青藏高原科学数据中心提供的亚洲高山区冰川高程变化数据集及前人基于花杆雪坑实测数据所得冰川高程变化速率相近。通过分析高程变化与气象数据的响应关系,验证了该文提取的区域冰川高程时-空变化特点与理论趋势相符。

Aiming at the problem that the difference of data quality in different acquision periods and topographic factors causing a large error in the estimation of glacier elevation change when using the satellite-borne photon point cloud time series data of ICESat-2 to monitor the glacier elevation change,this paper proposed the optimization strategy of ICESat-2 photon point cloud,which combined three indicators of slope,standard deviation,and distance between points,thus reducing the difference of photon data in different periods using ICESat-2 photon point cloud and SRTM from 2018 to 2022 in the study area of the glacier of Tanggula Mountains in Tibetan Plateau.Then,the error correction model of glacier elevation change estimation based on piecewise linear fitting function was constructed,and the error caused by terrain slope was corrected.The average elevation change rate of glaciers in Tanggula Mountains from 2000 to 2022 and the annual elevation change from 2018 to 2022 were estimated.The results showed that the average elevation change rate of glaciers in Tanggula Mountains of Tibetan Plateau from 2000 to 2022 was-0.41±0.43 m/a,which was close to the glacier elevation change dataset provided by the National Tibetan Plateau Science Data Center and the glacier elevation change rate measured result obtained from the previous scholars based on the method of stakes/firn.By analyzing the response relationship between elevation change and meteorological data,it was verified that the temporal-spatial variation characteristics of regional glacier elevation extracted in this paper were consistent with the theoretical trend.