| Compared with satellite-borne and airborne gravity measurements,terrestrial time-varying gravity measurements(TGMs)are performed closer to the sources of the mass change and have a relatively higher accuracy.Therefore,TGMs can contribute to mitigation and risk assessment by improving our understanding of past and present changes in ice mass,groundwater resources,volcanoes,and earthquakes.TGMs can ascertain the gravity field through single,repeated,or continuous gravity measurements by placing gravimeters at fixed measurement points on the ground.However,the measurement of gravity is affected by many factors,such as the instrument error,fieldsource monitoring capability of the gravity campaigns,and hydrological and environmental artifacts.A better understanding and quantification of these effects important for the geophysical interpretation of gravity measurements.Based on the modified Bayesian gravity adjustment(MBGA)and Bayesian equivalent source inversion(BESI)methods,this study involved terrestrial gravity observations in South China and along the eastern margin of the Tibetan Plateau from three aspects:the analysis of instrumental errors including drift and calibration factors,evaluation of the field source monitoring ability of gravity campaigns,and interpretation of gravity field source models.The main contents of this study are as follows.(1)The instrumental errors of the TGMs in South China from 2015 to 2018 were quantitatively analyzed using the MBGA method.The results show that the four relative gravimeters used in the South China gravity network have nonlinear drift rates,and there are clear differences among them.For example,FJ-1316 has a maximum drift rate of 0.8 ×10-5m/s2/day.The results of the MBGA are slightly better than those of the classical gravity adjustment(CGA)method in terms of residual gravity differences(GD)and the standard deviation of residual GDs.The estimated scale factors show that the average value of the differences between the calibrated scale factors and estimated scale factors(obtained from the MBGA method)of FJ-1316 and FJ-814 in the Fujian gravity campaign was 5.89-6.28 ×10-4.In addition,the average value of the differences between the calibrated scale factors and estimated scale factors of GD-232 and GD-369 of the Guangdong gravity campaign was 4.2-6.23 × 10-5.This indicates that the input scale factors of the Fujian gravity campaign are not very accurate.A cross-validation analysis was performed to verify the accuracy of the gravity values(GVs)obtained using the MBGA method.The GV deviations are the differences between the results of the absolute gravimetry and gravity values obtained from the MBGA and CGA.GV deviations obtained from the MBGA using estimated scale factors were smaller,by approximately 20-60×10-8 m/s2,than those obtained using the CGA method with a calibrated scale factor.With the increase in absolute points used for adjustment,the differences in GV deviations between the CGA and MBGA methods were gradually reduced,and the accuracy of GVs at the edge of the campaign notably improved.The results show that the uncertainty caused by the scale factor was larger than that caused by the nonlinear drift for the relative gravimeters used in the South China survey network.Importantly,the magnitude can reach 20×10-8 m/s2 for gravity changes.The use of the MBGA obtains nonlinear drift rates and reliable scale factors and produces better gravity values.(2)To determine the field source monitoring capability and gravity field characteristics of the South China region,a gravity model was constructed based on spherical tesseroid units of three sizes.Gravity field source modeling experiments and noise tests were performed to assess the quality of the gravity field recovery using different modeling methods.The variation in the observed gravitational field was calculated and recovered from 2015 to 2017 using the least-squares collocation method.The results show that the least-squares collocation method is better than the kriging and minimum curvature methods in recovering the gravity field,especially at the edge of the gravity network.Furthermore,the least-squares collocation method is robust in terms of noise resistance.From the forward modeling and inversion results,the spatial resolution capability of the South China gravity survey network was approximately 55 km,and the equivalent observable signal was approximately 30×10-8 m/s2.The cumulative temporal and spatial characteristics of the gravity field were consistent between 2015 and 2017.Moreover,negative gravity anomalies continued to increase and accumulate at different time scales,and there were local periodic changes in positive and negative fluctuations.(3)An Ms7.0 earthquake struck Jiuzhaigou(China)on August 8,2017.The epicenter was on the eastern margin of the Tibetan Plateau,an area covered by a dense time-varying gravity observation network.Using data from seven repeated high-precision hybrid gravity observation surveys from 2014 to 2017,gravity values before the occurrence of the Jiuzhaigou Ms7.0 earthquake were obtained,with a field source resolution better than 75 km based on the MBGA method.The comparison revealed that the scale factor obtained using the MBGA method was more accurate than that derived from either the CGA or BGA method,thereby effectively reducing the gravity difference residuals to 20 ×10-8m/s2.These results prove the validity of using the MBGA method,over the CGA and BGA methods,to reduce instrumental errors of the scale factor and nonlinear drifts.The effectiveness and robustness of the BESI method were assessed using synthetic tests with different resolutions and noise levels.The results showed that the field source resolution of the gravity network in the ETP was generally better than 75 km.The results also indicated that the field source parameters can be recovered effectively in the presence of Gaussian noise,with a standard deviation of 12 ×10-8 m/s2.,as is usual in the gravity observation environment.The range of apparent density variations obtained using the equivalent inversion method was approximately 0.59‰-0.74‰of the average density of the crust.A major gravity change occurred from the southwest to the northeast of the eastern Tibetan Plateau,approximately 2 years before the earthquake.Additionally,a substantial gravity gradient zone was consistent with the tectonic trend that gradually appeared within the focal area of the Jiuzhaigou earthquake during 2015-2016.Factors that might cause such regional gravitational changes(e.g.,vertical crustal defomation and variations in near-surface water distributions)were studied.The results suggest that the gravity effects contributed by these known factors were not enough to produce gravity changes as large as those observed,which might be related to the process of fluid material redistribution in the crust.Analysis of the gravity field changes in three study areas around the epicenter of the Jiuzhaigou earthquake,using the BESI method,revealed a substantial decrease in gravity change in the western focal area of the Jiuzhaigou earthquake after 2015.This decrease in gravity change was highly correlated with a low-velocity zone,which might indicate that crustal fluid migrated from the region southwest of the epicenter 2 years before the earthquake occurred.The gradient zone of apparent density variation coincided with the boundary area of the active crust,suggesting that the variation in the gravity field source was dominated by the tectonic structure of the ETP.(4)The MBGA method was used to process the terrestrial gravity observation data of the northeastern margin of the Tibetan Plateau,from 2012 to 2015,at an annual scale.The results show that the 2016 Menyuan Ms6.4 earthquake occurred in a high-gradient zone with significant gravity change and was located at the turning point of the zero curve.The BESI method was used to measure the field source resolution at the eastern margin of the Tibetan Plateau.The results show that the field source resolution of the network is approximately 100 km,and the observable gravity anomaly is approximately±80×10-8 m/s2.The field source with a 1° resolution can be equivalent to 0.740.81 ‰ of the apparent density variation in the equivalent layer with a depth of 10 km and a thickness of 1 km.The GPS velocity field shows that the overall horizontal movement of the northeastern margin of the Tibetan Plateau is northeast and eastward,with a magnitude of 2.6-11.8 mm/yr,and the velocity gradually decreases from southwest to northeast.The Lenglongling Fault Zone has strong activity characteristics.The magnitude of the GPS motion velocity on both sides of the fault zone was significantly different,and the strain accumulation near the Menyuan epicenter was significant.The dominant distribution of the GPS main strain features is mainly in the northeast direction of compressive strain,which is generally characterized by compressive deformation,and the surface strain rate is mainly characterized by contraction and extrusion.The epicenter of the 2016 Menyuan Ms6.4 earthquake was located near the peak of the high shear strain rate and high surface strain rate.Meanwhile,the maximum shear strain rate was about 1.4×10-8 mm/yr,and the surface shrinkage rate was 0.9×10-8 mm/yr.Furthermore,the gravity variation rate caused by nearsurface water,such as near-soil water and snow water,was approximately 0.05×10-8 m/s2/yr.The maximum gravity variation caused by rainfall was 6×10-8 m/s2,and the maximum vertical surface deformation rate was 1.17×10-8 m/s2.The variation in time-varying gravity caused by the hydrologic factors and vertical surface deformation was approximately 8×10-8 m/s2.The 3D velocity structure shows that there is a low-velocity layer 30-40 km below the epicenter of the Menyuan earthquake,and there may be fluid and fluid migration within the crust.The results of electromagnetic sounding show that the Lenglongling Fault is under the northeast-pushing action of the Tibetan Plateau,and there are significant physical differences between the north and south sides of the fault.The variation in the gravity field before the Menyuan Ms6.4 earthquake,as recorded by terrestrial gravity observations,may be related to the upward migration of low-velocity layers in the deep crust. |
PDF Full Text Request