Characterization of deepwater reservoirs in a frontier basin in the Brazilian equatorial margin: from seismic processing to machine learning approach

• Ano de defesa: 2020
• Autor(a) principal: Oliveira, Karen Maria Leopoldino
• Orientador(a): Castelo Branco, Raimundo Mariano Gomes
• Tipo de documento: Tese
• Tipo de acesso: Acesso aberto
• Idioma: por
• Palavras-chave em Português: Margem Equatorial Brasileira; Exploração de petróleo; Atributos sísmicos
• Link de acesso: http://www.repositorio.ufc.br/handle/riufc/53612

In recent years, the Brazilian Equatorial Margin has drawn attention due to its similarity to areas with new hydrocarbon discoveries in the African conjugated margin, and in French Guiana. However, studies on the tectonic regimes associated with transform margins and their evolution, structures, and petroleum potential are still lacking due to the geological complexity of this region. To address this knowledge gap, research has been done to better understand the geological structures, as well as to identify potential hydrocarbon accumulations in the deepwater Ceará Basin. To achieve this, we performed an integrated interpretation of a large 3D and 2D seismic data, new exploratory borehole data, as well as older well data with revised biostratigraphy. This data analysis refines the basin architecture and the Cretaceous-Paleogene tectonic evolution, including implications for hydrocarbon prospectivity in the Ceará Basin deepwater. The analysis also identifies potential hydrocarbon accumulations in turbiditic reservoirs and presents new insights about the dimensions of the underlying rift features situated in the continental slope. The results reveal a high potential for drift sequences in deepwater where the Late Albian-Early Cenomanian-Turonian sediments reach thicknesses of approximately 3048 to 4894 m. Moreover, this research shows evidence of Cretaceous to Paleogene magmatism, indicated by the well-imaged volcanoes and associated sills in the seismic data. The variety of stratigraphic and structural features developed through the Cretaceous history of the Mundaú sub-basin offers a variety of potential hydrocarbon traps and plays in a number of rift and post-rift sequences. In addition, a seismic attributes analysis and unsupervised machine learning approach were able to produce relatively high-resolution images and map the 3D geometry of ancient geomorphology across different stratigraphic levels from Albian to Turonian interval. A better understanding of the seismic geomorphology and seismic facies analysis provided valuable insights into an underexplored basin, and offer the best potential for deepwater stratigraphic traps. This approach may be used on similar frontier or emerging hydrocarbon basins to help de-risking the petroleum exploration.