Earth’s Surface Mass Transport Recovered From Temporal Gravity Field And Its Applications

As the most prosperous and valuable technique for measuring gravity in the early21st century, satellite gravity measurements make good reparation of the traditional observations, which have greatly improved the precision of gravity field model, especially in middle and long wave. Besides, the availability of the time-variable gravity data from GRACE provides an effective approach to continuously monitor variations of gravity in middle and large scales, and even global scale. GRACE mission, regarded as the first approach to study the flow of mass in and on the surface of the earth, has been widely invoked to analyze the Earth systems, such as surface mass variations, water cycle and climate changes. The application studies were mostly based on two methods:the mass redictribution recovered from GRACE temporal gravity field and recovered from GRACE observations directly. The solutions from spherical harmonic coefficients of GRACE temporal gravity field models and mass variations from Mascon(mass consentration) method were dicussed in this thesis, and the results from these two methods were used to reveal the mass redistribution over several typical regions.The main work and contributions in this dissertation cover mainly:Methods about recovering earth’s surface mass transport, including the spherical harmonic coefficients solutions from GRACE temporal gravity models and inversing from GRACE satellite tracking data, as well as its applications about terrestrial water storages, glacial melting, sea level changes and solid earth deformations were discussed. And present the role and importance of study on the regional surface mass variation study using GRACE in earth science and related researches.Mass variations recovered from GRACE monthly gravity field models, referred to as the spherical harmonic coefficient solutions, were descried. And the results with different filters according to the errors of gravity field models were analyzed. Comparing the GRACE gravity models from different providers, it shows that model error of the updated gravity field models from all providers are deduced much further, and the results from each provider are consistent much better with each other than before. These provide a basis for choosing the filters and gravity models while using the solution of spherical harmonic coefficients. Discuss the earth surface mass variations from the spherical harmonic coefficients solutions about Antarctic ice sheet melting and coseismic gravity changes of the earthquake in Japan. Antarctic ice sheet mass changes from GRACE reveal that the rate of melting of the Antarctic ice sheet is80.0Gt/a and increasing and contributes0.22mm/a to the global sea-level rise. And the ice sheet mass in the East Antarctic tends to be in balance. There is prominent ice sheet mass loss in the Amundsen Gulf and Graham Land of West Antarctic, where the mass loss rates increase significantly from2006.In New Schwaben Land, the North Antarctic and Enderby Land, along the coast of East Antarctica, the ice sheet mass increases remarkably from April2009.Comparing the coseismic gravity changes from GRACE observations and simulation from the fault slip distribution of March2011Tohoku-Oki earthquake in Japan, the spatial distribution were consiste well with each other. And the maximum of coseismic gravity changes arrived-5.30μGal. The results revealed that the GRACE gravity field models could effectively detect the secular trend of surface mass variation and the sudden change of surface mass variations leading by short-term events.Based on the dynamic method, the mathematical formula is deduced to directly calculate the regional surface mass change by using satellite gravity observations according to the relationship between surface mass change and the gravity anomaly, that is the Mascon mathematical model to inverse regional surface mass change via GRACE gravity observations.The system flow of using the Mascon method to do simulation studies is also given.The key issues that should be considered during the implementation of the Mascon method are investigated. Through analyzing the sensitivity of the satellite observations to surface mass change, we find that strong correlations exist between the KBRR residual signal and the relative position of the satellite trajectory to the regional mass anomaly together with its magnitude, compared with the orbit observation itself. Besides, we discuss the influence of the background gravity field model on the inter-satellite distance variation rate, finding that there is large difference between the inter-satellite observations obtained from models with different precisions.It also has a significant impact on the stability and convergence speed of the gravity field resolution. Furthermore, the space-time constraint equation is established through taking into account the spatial and temporal correlation between mass blocks, which could provide an effective constraint to solve the Mascon parameters. Based on the basic principle and the mathematical model of the Mascon method, the validity of the Mascon theory and the correctness of our system flow are verified using simulated GRACE satellite observations. Through comparison between the Mascon parameters obtained from different parameter selections and the reference anomaly field, we find that taking into account signal leakage and the edge effect would impact the inversion of the regional surface mass change to some extent. However, this effect would be weaken by calculating an area that appropriately larger than the study area. Given that the correlation distance and the scale factor determine the weight between the time-space constraint equation and the normal equation of the observations, inversions are done through selecting different correlation distances and scaling factors. Take the Amazon basin as an example, the range of the calculation area should be larger than8degree of the study area, while the correlation distance and the scaling factor should be set as250km and0.001respectively would be appropriate in order to obtain a good inversion result. Besides, our statistical analyses also show that the Mascon method can not inverse surface mass change with equivalent water height of smaller than5cm.Calculation algorithm of Mascon method has been displayed, which revered regional surface mass variations from GARCE satellite observations. And the pre-processing of the GRACE Level1B data and the calibration of the accelerometer data from fixed reference gravity field were discussed. Take the Amazon River Basin for example, surface mass variation of July2009which express as Mascon parameter has been recovered. While comparing with the results from the solutions of the GARCE spherical harmonic coefficients, they were consisting well in spatial distribution. And both the monthly and10days results of surface mass variation could reflect the flood conditions over Amazon River Basin on July2009.urface mass variation feature over Tibet plateaus were analyzed. The hydrological effects and the post glacial rebound effect were discussed from the global hydrological model and the GIA model respectively. The surface mass variation of characteristics area, such as Asian High Mountain glacial melting, the water source of three rivers and the stretch region of the southeast Tibet plateaus, were investigated by comparing the GRACE results with multiple observations including hydrological data and models, GPS data and so on. The GRACE data not only detect the decrease of the terrestrial water storage caused by the great drought in2010sucsufully, but also explore the high mountain ice melting which never reflect by the moden hydrological models. Besides, GRACE gravity models can also calculate the surface deformation of hydrological loading, which provides a kind of real observations to deduct non-tectonic deformation from GPS.