Estimation of Terrestrial Water Storage Variations in Sichuan-Yunnan Region from GPS Observations Using Independent Component Analysis
Estimation of Terrestrial Water Storage Variations in Sichuan-Yunnan Region from GPS Observations Using Independent Component Analysis
复制标题
DOI：
10.3390/rs14020282
复制
发表时间：
2022-01
期刊：
影响因子：
--
通讯作者：
Bingchen Liu;Wenkun Yu;W. Dai;Xuemin Xing;Cuilin Kuang
中科院分区：
文献类型：
--
作者：
Bingchen Liu;Wenkun Yu;W. Dai;Xuemin Xing;Cuilin Kuang
GPS可以用来测量地球表面质量载荷变化引起的陆地运动。提出了一种基于独立分量分析的反演方法，利用垂直GPS坐标时间序列估计中国川滇地区陆地蓄水量的变化。将独立分量分析方法应用于川滇地区中国地壳运动观测网GPS测站垂直坐标时间序列的水文形变信号提取。然后将这些垂直形变信号反转为TWS变化。基于重力恢复和气候实验(GRACE)数据和水文模型进行了对比实验，以进行验证。结果表明，由GPS(ICA)形变估计的TWS变化与由GRACE数据和水文模型得到的川滇地区的水量变化高度相关。川滇地区GPS垂直观测高估了TWS的变化。这些异常可能是由于大气负荷改正模型不准确或地形变率高的地区的对流层残留误差造成的，可以通过ICA预处理来消除。
GPS can be used to measure land motions induced by mass loading variations on the Earth’s surface. This paper presents an independent component analysis (ICA)-based inversion method that uses vertical GPS coordinate time series to estimate the change of terrestrial water storage (TWS) in the Sichuan-Yunnan region in China. The ICA method was applied to extract the hydrological deformation signals from the vertical coordinate time series of GPS stations in the Sichuan-Yunnan region from the Crustal Movement Observation Network of China (CMONC). These vertical deformation signals were then inverted to TWS variations. Comparative experiments were conducted based on Gravity Recovery and Climate Experiment (GRACE) data and a hydrological model for validation. The results demonstrate that the TWS changes estimated from GPS(ICA) deformations are highly correlated with the water variations derived from the GRACE data and hydrological model in Sichuan-Yunnan region. The TWS variations are overestimated by the vertical GPS observations the northwestern Sichuan-Yunnan region. The anomalies are likely caused by inaccurate atmospheric loading correction models or residual tropospheric errors in the region with high topographic variability and can be reduced by ICA preprocessing.