University of Twente Student Theses
Joint inversion of satellite gravity and seismic surface wave data: A synthetic test and its application for Botswana crust and upper most mantle velocity and density modeling.
Assefa, Getnet.A (2022) Joint inversion of satellite gravity and seismic surface wave data: A synthetic test and its application for Botswana crust and upper most mantle velocity and density modeling.
|
PDF
6MB |
| Abstract: | The joint inversion of geophysical observations has recently become an active research area to characterize the subsurface's physical properties. They are defined as the simultaneous optimization of two or more objective functions to estimate a model that describes all data sets simultaneously. Due to the constrained optimization of the subsurface model from different independent data sets, joint inversion can help to reduce the intrinsic non-uniqueness of the geophysical inverse problem. As a result, by merging different geophysical methods into a single inversion scheme, joint inversion approaches aim to reduce the number of acceptable models that fits different data set. Therefore, recently with the widespread geophysical data set available, it has become a topic of widespread active research area in large-scale lithospheric modelling or small-scale exploration studies by integrating different geophysical data sets in a joint inversion scheme. In this research, a new methodology developed by Fadel. (2018) that efficiently integrates gravity field and its gradients with seismic surface wave data is tested and implemented for real data. The new methodology is based on parallel versions of the tesseroid forward solver for the gravity total field, and the 6-gradient components, and 3D surface wave dispersion forward solver. The Levenberg–Marquardt damped least-squares method is used for the joint inversion of the multiple data components to retrieve the directly-coupled shear-wave velocity and density structure of the subsurface using Brochure velocity-density relationships. However, the required computational cost of the approach was the main drawback of the method to apply for large scale regional modeling. That was due to the approximation of the Jacobian for the surface wave forward solver iterating through each cell of the 3D cube. Therefore, recent improvements were introduced to the method efficiency to approximate the Jacobian based on a 1D algorithm that operates over the 2D grid of the 3D model, which significantly improves the computational cost of the method. In this research, the new developments were first tested using two phases of synthetic regional case studies and then followed by real data application to test the method on a real case study and to improve our knowledge about the tectonics of Botswana. The synthetic tests were conducted in three different scenarios of data integration to understand the added values of each data to the inversion. The results from the two phases of the synthetic tests prove that the performance of the method in terms of computational cost is greatly improved by approximately 80 % after improvements and show that shear wave velocity and density are retrieved accurately when the two data sets are integrated than the surface wave only and gravity only inversion, respectively. Furthermore, the added value of the unique GOCE gradient data for subsurface density modelling is analysed. As expected, the results suggest that the GOCE gravity gradients have added value to improve the resolving power of the inversion, especially for crustal to uppermost mantle depth ranges. After that depth range, it does not show an added value in the retrieved density and velocity model. Finally, the joint inversion approach is applied using real data measurements of gravity and seismic surface wave data to study the crust and upper mantle velocity and density structure of Botswana. The results of the real data case study highlight the limitations of the method and also show the first joint inversion velocity and density structure of Botswana which can be significantly improved in future studies. |
| Item Type: | Essay (Master) |
| Faculty: | ITC: Faculty of Geo-information Science and Earth Observation |
| Subject: | 38 earth sciences |
| Programme: | Geoinformation Science and Earth Observation MSc (75014) |
| Link to this item: | https://purl.utwente.nl/essays/93804 |
| Export this item as: | BibTeX EndNote HTML Citation Reference Manager |
Show download statistics for this publication
Show download statistics for this publicationRepository Staff Only: item control page