GPS时间序列中同震和震后形变的自动识别和估计

Automatic Detection and Estimation of Coseismic and Postseismic Deformation in GPS Time Series

地震发生时,附近GPS观测站会捕捉到同震和震后形变,同震和震后形变对地学研究和维持动态参考框架具有重要作用,识别和估计同震和震后形变也是GPS时间序列处理的重要内容。提出综合检校法识别和估计GPS时间序列中的同震和震后形变,首先利用该方法对GPS站建模,综合考虑多源的同震阶跃及兼容性和均方根RMS减小率来判断测站是否受到地震影响,然后用试错法搜索最优衰减常数τ后,利用最小二乘估计包含同震和震后形变的所有参数,最后以真实震例验证了该方法的有效性。

Once an earthquake occurs,nearby GPS stations are able to capture the coseismic and postseismic deformation which is critic for the geoscience research and maintenance of the dynamitic reference frame,thus to identify and estimate them is also an important part of the GPS time series analysis.The comprehensive inspection method to automatic detect and estimate coseismic and postseismic deformation is proposed and employed here,which models the GPS time series and comprehensively considers coseismic offsets and their coherence from different sources,and RMS improvement to decide if the station is affected by an earthquake.If it is,trail-and-error is used to search the best decay time and then all parameters including coseismic and postseismic deformations are estimated through linear least squares method.Earthquake examples account for the effectiveness of the proposed method.