ALBERT

Description: 〈span〉〈div〉Abstract〈/div〉We have trained a supervised deep 3D convolutional neural network (CNN) on marine seismic images for poststack structural seismic image enhancement and noise attenuation. Rather than adding artificial noise to training inputs, the difference in noise levels between the training inputs and labels was created by shot density differences. This design enables the trained CNN to mimic the results and power of stacking to specifically target random and coherent migration artifacts while enhancing low-amplitude reflections. We used field seismic from multiple Gulf of Mexico surveys to train the CNN and the SEG Advanced Modeling (SEAM) phase I synthetic data to evaluate the trained network. The diverse geologic features in the training data are needed to avoid overfitting. The processed outputs of the trained neural network are much cleaner than the inputs, and they highlight geologic structures for easier interpretation. Different scales of geologic structures, from high-resolution faults and diffractors to deep subsalt sediments, are well-preserved by the deep neural network. The trained network can be applied on either prestack gathers or poststack images. The approach is easy to implement and straightforward to parameterize, and it has proven to be an effective and flexible production tool for post-migration data conditioning.〈/span〉

Description: Extensive ROV-based sampling and exploration of the seafloor was conducted along an eroded transform-parallel fault scarp on the northeastern side of the Guaymas Basin in the Gulf of California to observe the nature of fluids venting from the seafloor, measure the record left by methane-venting on the carbonates from this area, and determine the association with gas hydrate. One gas vent vigorous enough to generate a water-column gas plume traceable for over 800 m above the seafloor was found to emanate from a ∼10-cm-wide orifice on the eroded scarp face. Sediment temperature measurements and topography on a sub-bottom reflector recorded in a transform-parallel seismic reflection profile identified a subsurface thermal anomaly beneath the gas vent. Active chemosynthetic biological communities (CBCs) and extensive authigenic carbonates that coalesce into distinct chemoherm structures were encountered elsewhere along the eroded transform-parallel scarp. The carbon isotopic composition of methane bubbles flowing vigorously from the gas vent (−53.6±0.8‰ PDB) is comparable to methane found in sediment cores taken within the CBCs distributed along the scarp (−51.9±8.1‰ PDB). However, the δ13C value of the CO2 in the vent gas (+12.4±1.1‰ PDB) is very distinct from those for dissolved inorganic carbon (DIC) (−35.8‰ to −2.9‰ PDB) found elsewhere along the scarp, including underneath CBCs. The δ13C values of the carbonate-rich sediments and rocks exposed on the seafloor today also span an unusually large range (−40.9‰ to +12.9‰ PDB) and suggest two distinct populations of authigenic carbonate materials were sampled. Unconsolidated sediments and some carbonate rocks, which have lithologic evidence for near-seafloor formation, have negative δ13C values, while carbonate rocks that clearly formed in the subsurface have positive δ13C values (up to +23.0‰) close to that measured for CO2 in the vent gas. There appears to be two carbon sources for the authigenic carbonates: (1) deeply-sourced, isotopically heavy CO2 (∼+12‰); and (2) isotopically light DIC derived from local anaerobic oxidation of methane at the sulfate–methane interface in the shallow subsurface. Addition of isotopically light methane-derived carbon at the seafloor may completely mask the isotopically heavy CO2 signature (+12.4‰) in the underlying sediments. Thus, the authigenic carbonates may have formed from the same methane- and carbon dioxide-bearing fluid, but under different migration and alteration conditions, depending on how it migrated through the sediment column.

Description: As offshore petroleum exploration and development move into deeper water, industry must contend increasingly with gas hydrate, a solid compound that binds water and a low-molecular-weight gas (usually methane). Gas hydrate has been long studied in industry from an engineering viewpoint, due to its tendency to clog gas pipelines. However, hydrate also occurs naturally wherever there are high pressures, low temperatures, and sufficient concentrations of gas and water. These conditions prevail in two natural environments, both of which are sites of active exploration: permafrost regions and marine sediments on continental slopes. In this article we discuss seismic detection of gas hydrate in marine sediments. Gas hydrate in deepwater sediments poses both new opportunities and new hazards. An enormous quantity of natural gas, likely far exceeding the global inventory of conventional fossil fuels, is locked up worldwide in hydrates. Ex-traction of this unconventional resource presents unique exploration, engineering, and economic challenges, and several countries, including the United States, Japan, Canada, India, and Korea, have initiated joint industry-academic-governmental programs to begin studying those challenges. Hydrates also constitute a potential drilling hazard. Because hydrates are only stable in a restricted range of pressure and temperature, any activity that sufficiently raises temperature or lowers pressure could destabilize them, releasing potentially large volumes of gas and decreasing the shear strength of the host sediments. Assessment of the opportunities and hazards associated with hydrates requires reliable methods of detecting hydrate and accurate maps of their distribution and concentration. Hydrate may occur only within the upper few hundred meters of deepwater sediment, at any depth between the seafloor and the base of the stability zone, which is controlled by local pressure and temperature. Hydrate is occasionally exposed at the seafloor, where it can be detected either visually or acoustically by strong seismic reflection amplitudes or high backscatter …