Search

Correct lithofacies interpretation sourced from wireline log data is an essential source of prior information for joint seismic inversion for facies and impedances, among other applications. However, this information is difficult to interpret or extract manually due to the multivariate and high dimensionality of wireline logs. Facies inference is also challenging for traditional clustering-based approaches because pervasive compaction trends affect a number of petrophysical measurements simultaneously. Another common pitfall in automated clustering approaches is the inability to account for underlying diagenetic processes that correlate with depth. Here, we address these challenges by introducing a rock-physics machine learning toolkit for joint litho-fluid facies classification. The litho-fluid types are inferred from the borehole data within the objective framework of a maximum-likelihood approach for latent facies variables and rock-physics model parameters, explicitly accounting for compaction and depth effects. The inference boils down to an expectation-maximization (EM) algorithm with strong spatial coupling. Each litho-fluid type is associated with an instance of a particular rock-physics model with a unique set of fitting parameters, constrained to a physically reasonable range. These fitting parameters in turn are inferred using bound-constrained optimization as part of the EM algorithm. Outputs produced by the toolkit can be used directly to specify the necessary prior information for seismic inversion, including per-facies rock-physics models and facies proportions. We present an example application of the tool to real borehole data from the North West Shelf of Australia to illustrate the method and discuss its characteristic features in depth.

The ultrasonic measurements technique is well established to measure the elastic properties of rocks in the laboratory for seismic and well-log data interpretation. The key components of every laboratory ultrasonic setup are piezoelectric transducers, which generate and register elastic waves in rock samples. The elastic properties of rocks are determined through the velocities of elastic waves, which are measured by the times of the waves' travel from the source to the receiver transducer. Transducers can be specifically designed to generate P-waves (P-transducers) or S-waves (S-transducers). In limited studies, the measurement of P-wave velocities with S-transducers is mentioned. Such measurement is possible due to specific aspects of the operation of S-transducers. Namely, S-transducers are known to emit parasitic low-energy P-waves, which travel faster than high-energy S-waves and hence can be registered. However, no justification or elaboration of this method of measuring P-wave velocities was reported. To fill this gap, we first compare P-wave velocities measured with S-transducers against P-wave velocities measured with P-transducers in different rocks and materials. We show that the discrepancy between velocities measured with the two methods in homogeneous materials is less than 1% and can be up to 4% for natural rocks. Second, we numerically simulate the operation of S-transducers, show that parasitic P-waves have a dipole structure, and explain how the receiver transducer can register this compressional dipole. Finally, we use laser doppler interferometry to measure the displacement of the free surface of a sample caused by elastic waves emitted by the source S-transducer. We observed the dipole structure of the sample's surface displacement upon P-wave arrival on the surface.

Shales have always been a difficult target for drilling of deviated and horizontal wells. In the presence of azimuthal stress fields, inclined boreholes in smectite-rich shales exhibit geomechanical instabilities and can result in borehole failure. The complex geology of the major gas fields in the Northern Carnarvon Basin on the North West Shelf of Australia makes it necessary to drill deviated wells through the smectite-rich shale seal extending more than 1 km in thickness. Predicting the mineralogical composition of shales in the area is therefore crucial for the success of drilling operations related to hydrocarbon exploration and production. Here we introduce a novel workflow that combines seismic data, well logs, and laboratory measurements to rapidly infer smectite content in shale. The workflow is applied to the Duyfken 3D seismic survey in the central part of the Northern Carnarvon Basin. The results of our quantitative interpretation are verified against the laboratory X-ray diffraction measurements from the test well that was not used for interpretation, and they match the test data well within the determined uncertainty bounds.

Multiple potential shale-gas plays have been identified in Australia. The total estimates of recoverable dry and wet shale gas exceed 157 tcf. Technological advances and more informed estimations of total organic carbon (TOC) abundance and distribution might further increase the assessments of recoverable shale gas. TOC content is known only at the depths where laboratory measurements on recovered core samples are performed. However, reliable estimation of potential resources can be based only on information about vertical and lateral distribution in organic matter throughout the prospective gas-shale reservoir. This information commonly is obtained from conventional wireline logs such as gamma ray, density, transit time, and resistivity. Methods routinely used for these estimations were developed for organic-rich shales from Northern American basins, where organic matter of marine Type I and Type II typically is observed. In prospective Australian basins, organic matter ranges from marine Type I to terrestrial Type III. This fact, along with different depositional history of Australian reservoirs, might lead to quite different logging-tool responses. Thus, the relations between the amount of TOC and estimation from logs might be quite different.

Maximizing the recovery of known hydrocarbon reserves is one of the biggest challenges facing the petroleum industry today. Optimal production strategies require accurate monitoring of production-induced changes of reservoir saturation and pressure over the life of the field. Time-lapse seismic technology is increasingly used to map these changes in space and time. However, until now, interpretation of time-lapse seismic data has been mostly qualitative. In order to allow accurate estimation of the saturation, it is necessary to know the quantitative relationship between fluid saturation and seismic characteristics (elastic moduli, velocity dispersion, and attenuation). The problem of calculating acoustic properties of rocks saturated with a mixture of two fluids has attracted considerable interest (Gist, 1994; Mavko and Nolen-Hoeksema, 1994; Knight et al., 1998. For a comprehensive review of theoretical and experimental studies of the patchy saturation problem see Toms et al., 2006).

Despite decades of research, current understanding of elastic properties of shales is insufficient as it is based on a limited number of observations caused by the time-consuming nature of testing resulting from their low permeability. Though it is well known that shales are highly anisotropic and assumed to be transversely isotropic (TI) media, few laboratory experiments have been carried out for measuring the five elastic constants that define TI media on well-preserved shales. Many previous measurements were made without control of pore pressure, which is crucial for the determination of shale elastic properties.