融合GPS与强震仪数据实时监测瞬时地壳形变

针对强震仪与GPS各自的优点与缺陷，以2016年意大利Mw 6.0级地震为例，探讨了利用Kalman滤波融合GPS与强震仪数据实时监测中等强度地震引起的瞬时地壳形变的可行性与优势。Kalman滤波获取的位移波形以GPS数据作为整体趋势，细节部分用强震仪数据进行描述，融合后的位移波形精度较高，能够准确和细致地描述地震引起的地表动态形变；与此同时，Kalman滤波位移波形也能够准确地探测中等强度地震引起的同震永久阶跃；Kalman滤波位移波形的功率谱在低频部分与GPS位移波形的功率谱密度相似，在高频部分与强震仪位移波形的功率谱密度相似，表明Kalman滤波结合了GPS与强震仪各自的优点；交叉小波分析表明，Kalman滤波获取的位移波形与GPS、强震仪数据呈现较强的正相关性，小波相似值均在0.8以上。以上结果表明，融合GPS与强震仪数据可以结合二者各自的优点，弥补各自的缺陷；融合GPS与强震仪数据可以有效探测中等强度地震引起的瞬时地壳形变。

Real-time monitoring of instantaneous crustal deformation by integrating GPS and strong seismometer data

In view of the merits and drawbacks of strong motion (SM) and GPS, taking the Mw 6.0 Italy earthquake as a subject, the feasibility of Kalman filter to combine GPS and SM data, and then to conduct the real-time monitoring of instantaneous crustal deformation caused by moderate earthquake using fusion is discussed. The displacement waveform obtained by Kalman filtering uses GPS data as the overall trend, and the details are described by the SM displacement data. The combined displacement waveform has higher precision and can accurately and meticulously describe the coseismic ground dynamic deformation caused by earthquakes. At the same time, the Kalman filter displacement waveform can accurately detect the coseismic permanent offsets caused by moderate earthquake. The power spectrum density (PSD) of combined displacement waveform is similar to that of GPS displacement waveform in the low-frequency regions, while it is similar to the PSD of SM displacement waveform in the high-frequency regions, indicating that Kalman filtering can combine the advantages of GPS and SM. The spectral index of the displacement waveform obtained by Kalman filtering is different from that of GPS and SM displacement waveforms, indicating that the fused displacement waveform can better describe the variation of the noise characteristics of displacement waveform data caused by the earthquake. Cross-wavelet analysis shows that the displacement waveform obtained by Kalman filter has strong positive correlation with GPS and SM data, and the cross wavelet semblance is above 0.8. The above results show that the combined displacement waveform can combine the advantages of GPS and SM to make up for their defects. The combination of GPS and SM can effectively detect the transient crustal deformation induced by the earthquake with moderate magnitude.