Measurement Characteristics Of GNSS Combined With MEMS Accelerometer For Seismic Monitoring

Earthquake is one of the most harmful natural disasters and can cause huge casualties and property losses.Earthquake is currently difficult to predict in short-term,but theoretically,earthquake monitoring is an effective way to provide earthquake early warning messages after an earthquake occurs.In traditional seismic monitoring,broadband seismometers in the near-field often clip in the case of strong vibrations,and strong-motions are affected by instrument tilt and rotation under strong vibrations,which causes baseline drift and other problems,making it difficult to recover accurate seismic displacement.High-rate GNSS can directly calculate the station displacement in the earth-centered,earth-fixed coordinate system,which can obtain low-frequency signals or even permanent displacement,and it has no range limitation and baseline drift problems.Using space earth observation technology like GNSS for seismic monitoring has also become an important method in recent years.However,the noise level of GNSS displacements is several orders of magnitude higher than that observed by seismometers or strong-motions,and GNSS cannot capture small-scale ground deformations.In view of the complementary advantages of GNSS and strong-motions for seismic monitoring,the combination of high-rate GNSS and strong-motions has become an trend in recent years.However,strong-motion networks and GNSS network s are deployed separately in C hina,and there are few collocated GNSS and strong-motion stations.Based on the existing GNSS stations of CMONOC network,this paper uses low-cost MEMS accelerometers and installs them on the GNSSstations,forming a collocated GNSS and MEMS accelerometer seismic monitoring network in Yunnan Province.Then the measurement characteristics of GNSS and MEMS accelerometer of the combined network are analyzed.The main contents and results are as follows:1.Measurement characteristics analysis and data processing of GNSS and MEMS accelerometers of the established seismic monitoring network are conducted.The quality analysis of the GNSS data in 2018-2019 of the network was carried out,from the aspects of data efficiency,signal-to-noise ratio,and pseudo-range multipath.The results show that the network GNSS data efficiency is high.Among them,the data quality of the YNHZ station is not good,and the cycle slips occur frequently,mainly caused by the bad observation environment;the signal-to-noise on the L₂ of the YNJP station is relatively poor,too.Allan variance analysis was used to process the MEMS accelerometer data.The results show that the main noise types in the MEMS accelerometer are Gaussian white noise and flicker noise.Using the open source software PRIDE PPP-AR,the GPS data of each station in the 180 days of year 2018to 139 days of year 2019 was processed for single-day static solutions and epoch-wise kinematic solutions individually.Static PPP-AR results show that the horizontal precision is aroud 2?3mm.During this period,the crustal movement caused the stations to move eastward by more than 4 cm and south by about 1 cm,and caused vertical displacements variation of up to 3 cm.The fix-rate of ambiguity resolution was more than 95%.The kinematic solutions show that most of the other stations can achieve the positioning accuracy of the horizontal RMS value of 1 to 3 cm and the vertical RMS value of 3 to 5 cm in a single day,except for the poor data quality of the YNHZ station and the abnormality of some days of the YNJP station.The result shows that GNSS can obtain accurate station coordinates.2.Shaketable test for earthquake monitoring verification using GNSS and strong-motion.The shake table test was carried out,and the accelerometer was tested by simulating vibration signals of different frequencies,and its low-frequency measurement characteristics were analyzed from the aspects of signal power spectral density and integral displacement.The results showed that the strong-motion could not recover the low-frequency displacement signal,especially signals below 0.02 Hz.The data fusion of GNSS and strong-motion can obtain accurate low-frequency?as low as 0.005 Hz?displacement signals in the time and frequency domains.Afterwards,the data of the collocated GPS and strong-motion station LKTA/LTZ of the 2016 M_W7.8 Kaik?ura earthquake in New Zealand were processed to obtain more accurate broadband velocity and displacement.The data fusion displacement shows that a permanent displacement of 7.5cm is generated in the east direction,and the low-frequency displacement signal is retained,which is more accurate and reliable than the acceleration integration of strong-motion data.3.Establishment of GNSS and MEMS accelerometer collocated monitoring network and analysis of earthquake events recorded.The process of establishing the GNSS/MEMS accelerometer seismic monitoring network is introduced,including the process of site selection,instrument parameters,installation and configuration.After that,based on the actual earthquake events recorded by the network,the velocity,displacement,and frequency of the GNSS/MEMS accelerometer data were analyzed in the time and frequency domains.The results show that the seismogeodetic velocity of GNSS/MEMS accelerometer is still sensitive enough to detect the arrival time of P wave and recognize the arrival of S wave.Due to the solution accuracy of 1Hz GNSS,it is difficult to detect weak displacement signals with seismogeodetic solutions.It shows that the seismogeodetic solutions.of GNSS and MEMS accelerometers can retain the advantages of high sampling rate,high sensitivity,and high precision of MEMS accelerometers in high frequency bands,while retaining low frequency and even permanent GNSS displacement in low frequency bands.,so that accurate broadband displacement can be obtained.The whole results can provide practical reference for the establishment of GNSS/MEMS accelerometer network in China,and verify the effectiveness of the GNSS/MEMS accelerometer seismic monitoring network.