An integrated approach based 2D seismic profiles and biostratigraphic data was preformed to provide preliminary results on the tectonic evolution of the sedimentary basins of Abu Dhabi. Three major stratigraphic sequences were identified; Lower Permian-Jurassic rifted-margin, intermediate Cretaceous passive-margin and upper Late Cretaceous-Tertiary active-margin sequences. The tectonic subsidence and uplift analysis by backstripping suggest two rifting phases followed by lithospheric flexure that resulted in formation of the UAE foreland basin. This study provides initial results of an ongoing research investigating tectono-stratigraphy of the basins underlying the oilfields of Abu Dhabi Emirate.
The Ferrar large igneous province was a volcanic event which resulted in the emplacement of a large amount of igneous rocks. It occurred during the Jurassic about 180 million years ago. The petrography of the rocks that were emplaced during this event remain largely understudied. This paper studies the petrography of the Catamaran core which originates from Tasmania, Australia. The core samples will undergo rigorous full rock sample and thin section analysis to determine the petrology. These analyses suggested that the samples were largely basalts. These basalts showcased features such as quartz amygdoles and calcareous lenses indicating hydrothermal alterations and impurities in the magmatic melt.
Loss of containment events are a significant concern for the upstream petroleum sector. The early detection of anthropogenic activities could limit the spread of contamination and immediately mitigate the spills. Microbes are ubiquitous and, as the relative abundance of key microbial community members shift with environmental perturbations (such as oil contamination), microbial communities are useful for environmental monitoring. However, the challenge lies in the lack of tools that could identify features of these microbial communities for this application. Machine learning (ML) approaches offer a promising means to address this challenge. The aim of this study is to establish a means for early oil contamination detection through the implementation of ML, specifically focusing on establishing which ML input (amplicon sequence variants or k-mer representation) data would best serve predictive ML framework for microbial feature detection to inform for oil contamination.
The application of Deep Learning (DL) in metagenomic studies is rapidly increasing due to its identification and predictive ability. This paper looks at utilizing a multilane capsule network (CapsNet) for early recognition of oil perturbation which could be used for environmental monitoring. Comparison between CapsNet and classic machine learning techniques (random forest, support vector machine, and multi-layer perceptron) would be conducted to establish which model best suits the aim of the study. Ideally, a tool that is cost-efficient, fast, and accurate is preferred as immediate identification of such perturbations is essential for rapid mitigation of spills to prevent further damage and harm. Thus, these are the aspects that would be evaluated when comparing different approaches.
