联合小波和方差分量估计方法分析中国IGS测站时间序列变化特征

Coordinate Time Series of IGS Station in China Based on the Combination of Wavelet Spectral and Variance Component Estimation

利用小波谱分析了中国区域IGS（International GNSS Service）站坐标时间序列中周期信号成分及其随时间的变化,根据小波谱中小波系数的能量在时域上的变化对时间序列进行分段,然后采用方差-协方差验后估计法分析了白噪声＋闪烁噪声组合模型下测站在不同时间段的运动特征及变化规律。结果表明,中国区域的IGS站运动特征在不同时间段内是变化的,其中URUM站垂直速度变化大于4 mm,BJFS以及CHAN站水平速度变化大于2 mm,其余站速度变化均在2 mm以内;周年、半周年信号的振幅也均有差异,其中BJFS站垂直分量周年振幅变化最大,超过5mm;此外,强度较弱的其他周期项存在于分段时间序列中,但其对速度、周期振幅以及噪声分量估计结果影响较小;对时间序列进行分段估计,所得的参数不同,其相差程度可反映出不同时间段内的板块运动等地球物理因素变化的差异。

Wavelet spectral analysis is employed in this paper to determine the frequency components and their variations over time in coordinating time series of IGS stations in China.Then least-squares variance component estimation（LS-VCE）is adopted to analyze motion characteristics of stations in China in different periods under the assumed model of white and flicker noise.The results indicate that motion characteristics of IGS station in China are seasonally varied in different periods.The variation of the vertical velocity of URUM station is greater than 4 mm,while velocity changes more than 2 mm for horizontal component of BJFS and CHAN.And the rest are no more than 2 mm.There are also differences in the amplitude of the annual and semi-annual signals.Among them,BJFS has the largest change in vertical component,which is more than 5 mm;In addition,weaker periodic items are discovered in some segment of the time series.Though,due to their little impact on the result of parameter estimation,they are discarded along the process.Moreover,the fact that the estimations of the parameters for one station are not consistent in different segments of the time series may reflect patterns of geophysical effects on certain stations over time such as plate motion.