The sensitivity of gravity observables for crust-mantle structure.

Abstract

This MSc thesis comprises the development of an extensive software tool for the computation of gravity quantities. The program is based on Synthetic Earth Gravity Models presented in previous literature studies. The program uses a discrete integration approach of Newton’s law for gravitation making use of point masses. Benchmark and resolution test results show the software tool to be correct. The final goal of this thesis is to test the significance of several gravity observables within numerical Earth modelling. Possible Earth density distributions are derived using the S40RTS tomographic seismic S-wave velocity mantle model and the UU-P07 tomographic P-wave model combined with a density model of the crust. The scaling of the velocity anomalies to density anomalies is done using various simple scaling factors. The results of the different scaling factors will be highlighted. Several simplifications and assumptions are made for gravity predictions. Only gravitation is used and flattening and rotation are neglected, an accurate crust model is absent including a terrain model or isostatic compensation. Scaling P-velocities is different from scaling S-velocities which creates additional uncertainty in the gravity predictions. Overall there have been varying successes. The theory and implementation of the theory is properly done but the results of the real Earth density distributions are not yet sufficient for good conclusions.