Development of a Python Library for Processing Seismic Time Series

• Autor(a) principal: Almeida, Eduardo Rodrigues
• Data de Publicação: 2021
• Tipo de documento: Dissertação
• Idioma: eng
• Título da fonte: Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)
• Texto Completo: http://hdl.handle.net/10400.6/11823
• topic: Library; Methods; Python; Seismic Signal; Seismology; Signal Processing; Synthetic Data; Time Series; Domínio/Área Científica::Engenharia e Tecnologia::Engenharia Eletrotécnica, Eletrónica e Informática

Earthquakes occur around the world every day. This natural phenomena can result in enormous destruction and loss of life. However, at the same time, it is the primary source for studying Earth, the active planet. The seismic waves generated by earthquakes propagate deep into the Earth, carrying considerable information about the Earth’s structure, from the shallow depths in the crust to the core. The information transferred by seismic waves needs advanced signal processing and inversion tools to be converted into useful information about the Earths inner structures, from local to global scales. The ever­evolving interest for investigating more accurately the terrestrial system led to the development of advanced signal processing algorithms to extract optimal information from the recorded seismic waveforms. These algorithms use advanced numerical modeling to extract optimal information from the different seismic phases generated by earthquakes. The development of algorithms from a mathematical­physical point of view is of great interest; on the other hand, developing a platform for their implementation is also significant. This research aims to build a bridge between the development of purely theoretical ideas in seismology and their functional implementation. In this dissertation SeisPolPy, a high quality Python­based library for processing seismic waveforms is developed. It consists of the latest polarization analysis and filter algorithms to extract different seismic phases in the recorded seismograms. The algorithms range from the most common algorithms in the literature to a newly developed method, sparsity­promoting time­frequency filtering. In addition, the focus of the work is on the generation of high­quality synthetic seismic data for testing and evaluating the algorithms. SeisPolPy library, aims to provide seismology community a tool for separation of seismic phases by using high­resolution polarization analysis and filtering techniques. The research work is carried out within the framework of the Seismicity and HAzards of the sub­saharian Atlantic Margin (SHAZAM) project that requires high quality algorithms able to process the limited seismic data available in the Gulf of Guinea, the study area of the SHAZAM project.