Multiattribute probabilistic neural network for near-surface field engineering application

Unconfined compressive strength (UCS) is an important rock parameter required in the engineering design of structures built on top or within the interior of rock formations. In a site investigation project, UCS is typically obtained discretely (through point-to-point measurement) and interpolated. This method is less than optimal to resolve meter-scale UCS variations of heterogenous rock such as carbonate formations in which property changes occur within data spacing. We investigate the geotechnical application of multiattribute analysis based on near-surface reflection seismic data to probe rock formations for their strength attributes at meter-scale variability. Two Late Jurassic outcrops located in central Saudi Arabia serve as testing sites: the Hanifa Formation in Wadi Birk and the Jubaila Formation in Wadi Laban. The study uses core and 2D seismic profiles acquired in both sites, from which we constrain UCS, acoustic velocity, density, and gamma-ray values. A positive linear correlation between UCS and acoustic impedance along the core indicates that seismic attributes can be utilized as a method to laterally extrapolate the UCS away from the core location. Seismic colored inversion serves as input for neural network multiattribute analysis and is validated with a blind test. Results from data at both outcrop sites indicate a high degree of consistency with an absolute UCS error of approximately 5%. We also demonstrate the applicability of predicted UCS profiles to interpret mechanical stratigraphy and map lateral UCS heterogeneities. These findings provide a less expensive alternative to constrain UCS from limited core data on a field-scale site engineering project.

Case study: Seismic resolution and reservoir characterization of thin sands using multiattribute analysis and bandwidth extension in the Daqing field, China

The Daqing field, located in the Songliao Basin in northeastern China, is the largest oil field in China. Most production in the Daqing field comes from seismically thin sand bodies with thicknesses between 1 and 15 m. Thus, it is not usually possible to resolve Daqing reservoirs using only conventional seismic data. We have evaluated the effectiveness of seismic multiattribute analysis of bandwidth extended data in resolving and making inferences about these thin layers. Multiattribute analysis uses statistical methods or neural networks to find relationships between well data and seismic attributes to predict some physical property of the earth. This multiattribute analysis was applied separately to conventional seismic data and seismic data that were spectrally broadened using sparse-layer inversion because this inversion method usually increases the vertical resolution of the seismic. Porosity volumes were generated using target porosity logs and conventional seismic attributes, and isofrequency volumes were obtained by spectral decomposition. The resulting resolution, statistical significance, and accuracy in the determination of layer properties were higher for the predictions made using the spectrally broadened volume.

Multiattribute analysis for channel element discrimination in the Taranaki Basin, offshore New Zealand

A detailed study of Pliocene channel systems within the Taranaki Basin was undertaken from the Parihaka 3D seismic volume to improve our understanding of the Plio-Pleistocene channel elements in terms of structure, channel evolution, and lithology. Seismic picking parameters were chosen based on the lateral resolution for optimal mapping of the channels. Individual and multiattribute studies were performed on single, combined, and complex channel systems with the goal of identifying channel features and discriminating between shale- and sand-rich regions of the channels. For this target and data set, the variance attribute provided key insights into channel features, such as the edge of the channel, meander scrolls, and point bars. Root-mean-square amplitude and sweetness performed equally well in lithology identification, and, combined with variance, it aided in identifying sand-rich channels, as well as small individual channels that could provide sediment pathways into the deepwater Taranaki Basin. Depending on the complexity of the channel system, different attribute analyses had varying success with each system. Therefore, it is important to combine various attributes to discriminate channel elements as fully as possible. The lithologies of individual channels and their elements can be determined using seismic attributes, although it becomes increasingly difficult to discriminate small-scale features within the channel as the complexity of the channel system increases. Chronostratigraphic studies using stratal slicing techniques provided insight into the evolution of the channel system through time, demonstrating an overall sand-rich base of the channel, with a shallower shale-rich lithology at the top of the channel fill.

A simple geologic risk-tailored 3D controlled-source electromagnetic multiattribute analysis and quantitative interpretation approach

Three-dimensional surveying is the method of choice in marine controlled-source electromagnetic (CSEM) exploration for hydrocarbons in frontier regions, but robust interpretation of the typically large-size field data faces significant challenges, including how to determine the correct resistivity, depth, and lateral limits of hydrocarbon-saturated reservoirs in the presence of heterogeneous host rocks or anisotropy and how to relate CSEM information to the key elements of geologic prospect evaluation (the presence of source rocks, migration and charge, reservoir rock, trap, and seal). We have developed a simple geologic risk-tailored approach for multiattribute analysis and first-pass interpretation of CSEM data in frontier exploration in which little prior information is available. First, geometric normalization of electric field amplitudes at each receiver location yields “phase-consistent” sounding curves that directly represent subsurface electrical structure (and can indicate reservoir rock presence). It enables accurate determination of seafloor resistivity (whose areal variation and direct correlation with seepage-induced geochemical and seismic shallow-gas anomalies can indicate the presence of a working petroleum system). Edge-detection attributes are then used to determine the geographical position and boundary shape of anomalous 3D resistive bodies (the trap presence and structural closure). Keeping these known parameters fixed, the most likely burial depth and resistivities of the sought 3D bodies are found using a simple line search technique involving rigorous 3D modeling and the results are validated and optimized post facto using seismic depth constraints to locally improve the prediction of the size and resistivities of hydrocarbon-charged or water-bearing sections crucial for prospect derisking, reserve estimation, and well placement.

Investigation of seismic attributes, depositional environments, and hydrocarbon sweet-spot distribution in the Serbin field, Taylor Formation, Southeast Texas

We have conducted seismic-attribute analysis at the Serbin field — in an area straddling Lee, Fayette, and Bastrop Counties and covering approximately [Formula: see text] (approximately [Formula: see text]) — using new, reprocessed, 3D seismic data to provide additional understanding of depositional environments and better predict the distribution of hydrocarbon sweet spots. We converted the 3D seismic volume into a log-lithology volume and integrated core data to examine the distribution of lithology and interpret depositional environments. By conducting multiattribute analysis, we predicted resistivity (deep-induction log) volume and generated a resistivity map to identify hydrocarbon sweet spots. Our results show that reservoir sandstones in the Serbin field are storm-dominated, shelf-sand deposits. Although individual sandstone beds are lenticular and discontinuous, they collectively constitute a sheet-like geometry, trending northeast to southwest. On the basis of resistivity maps and rock property versus seismic-amplitude crossplots, we differentiated reservoirs in the lower Taylor Formation into two zones: (1) a northwest, high-resistivity, high-acoustic impedance zone and (2) a southeast, low-resistivity, low-acoustic impedance zone. The results also indicated that hydrocarbon sweet spots in the Serbin field are characterized by high resistivity and high impedance. Furthermore, the log-lithology method, although fast and effective, is limited because it cannot take into account sandstone zones having low acoustic impedance.