Adjoint tomography of South America based on 3D spectral-element seismic wave simulations

• Grant number: 18/04918-6
• Support type: Scholarships abroad - Research Internship - Doctorate (Direct)
• Effective date (Start): August 15, 2018
• Effective date (End): February 14, 2019
• Field of knowledge: Physical Sciences and Mathematics - Geosciences
• Principal Investigator: Marcelo Sousa de Assumpção
• Grantee: Caio Henrique Ciardelli
• Supervisor abroad: Ebru Bozdag
• Home Institution: Instituto de Astronomia, Geofísica e Ciências Atmosféricas (IAG). Universidade de São Paulo (USP). São Paulo , SP, Brazil
• Local de pesquisa : Colorado School of Mines, United States
• Associated to the scholarship: 16/03120-5 - Full-waveform tomography of South America with spectral-element and adjoint methods, BP.DD

Abstract

The Ph.D. project "Full-waveform Tomography in South America and Surrounding Oceans with Spectral Elements and Adjoint Methods" is currently being developed at the IAG-USP under FAPESP grant 2016/03120-5 and, as predicted in the initial project, its third year would be accomplished abroad, in the US. Full-waveform Tomography (FWT) using adjoint methods is the state-of-the-art method in seismology to obtain seismic waves velocity models for the Earth's interior. The key idea is to extract the full information of the seismograms instead of using only the P and S-waves travel-times, as usual in other methodologies. Using more information from the data reduces the ambiguity and non-uniqueness, common in all geophysical methods, and allows a higher precision and accuracy in the model. FWT is complex and requires a deep knowledge of the physics governing seismic waves propagation, high performance computing, computational modeling and advanced inversion techniques. During the last year of the under-graduation, the student developed his coursework in 1D and 2D isotropic seismic waves modeling, to prepare for the direct Ph.D. project in FWT, in which not only the 3D anisotropic seismic waves forward modeling is being carried out but also the inverse problem using adjoint inversion to obtain a new velocity model for the South-American continent and surrounding oceans. This new model will allow a better understanding of the upper mantle structure beneath South America as well as improving future focal mechanisms determinations. The project is part of the FAPESP Pantanal-Chaco-Paraná Basins Thematic Project (grant 2013/24215-6).The proposed one-year internship is planned for two stages: six months in Colorado School of Mines to complete the tomography inversion, working with Dr. Ebru Bozdag, and six months in Northwestern University to interpret the results in terms of geodynamics, working with Dr. Suzan van der Lee. (AU)