This article investigates bypassing the inversion steps involved in a standard litho-type classification pipeline and performing the litho-type classification directly from imaged seismic data. We consider a set of deep learning methods that map the seismic data directly into litho-type classes, trained on two variants of synthetic seismic data: (i) one in which we image the seismic data using a local Radon transform to obtain angle gathers, (ii) and another in which we start from the subsurface-offset gathers, based on correlations over the seismic data. Our results indicate that this single-step approach provides a faster alternative to the established pipeline while being convincingly accurate. We observe that adding the background model as input to the deep network optimization is essential in correctly categorizing litho-types. Also, starting from the angle gathers obtained by imaging in the Radon domain is more informative than using the subsurface offset gathers as input.@en

Seismic waves from active experiments carry information regarding the subsurface in the form of reflected data that is recorded at the surface. This recorded data is subjected to sophisticated processing methods to estimate relevant parameters describing the geology of the subsurface. Traditionally the recorded data is used to create an image of the subsurface in terms of reflectivities, using seismic migration, which back-projects the data recorded at the surface into the earth. The resulting image can be interpreted in terms of structures and depositional patterns. There is another route that is followed to quantify the elastic properties of the subsurface by means of inversion of the recorded data. The essence of seismic inversion is to obtain the elastic properties of the earth’s subsurface from a finite set of noisy measurements, by forward modelling based on assumed properties and feed-back that projects the data mismatch onto model parameter space. Full-waveform inversion (FWI) is a special form of inversion that is gaining considerable attention in the last decade, which can be attributed to the advancement in the computational power available. However, several challenges remain for multi-parameter FWI to be successfully implemented on real size data problems in industry or academia at a scale fine enough to be useful in reservoir characterization.@en

Reservoir characterisation is a data driven process which involves the integration of different datasets to describe the subsurface. One of the difficulties of integrating geological data with the wave-equation based seismic inversion is that geological information is invariably interpreted as a layer-based model, whereas the wave-equation is defined and solved on a grid. Mapping a layer-based model space onto a grid-based space leads to highly non-Gaussian, multi-modal distribution functions, even when the layer-based properties have simple Gaussian distributions. In this paper an analytic method is presented that translates the prior layer-model based distributions to grid-based prior distributions. From the unconstrained seismic inversion result a Gaussian likelihood function is constructed and the method to find the maximum a posterior estimate (MAP) and its uncertainty is described. As geological prior information we use well data, a geological concept of the environment of deposition and structural seismic interpretation in the form of some horizons to guide the prior model in between wells. Given the prior model, a measure for the probability of the data is formulated. When this process is repeated for various prior scenarios, the probability of the scenario, given the data, can be calculated for every location.@en

Any target-oriented localised inversion scheme for reservoir elastic parameters is as good as the input dataset. Thus, the accuracy of the input dataset i.e. local reflection response or impulse response (virtual source-receiver response) is of utmost importance, especially when the target area is below a complex overburden. In these subsurface settings, the overburden internal multiples and associated transmission imprint obscure the local response, which in turn affects the estimated elastic parameters resolution. Here, we demonstrate a novel process called JMI-res, based on Joint Migration Inversion (JMI), to estimate the reservoir elastic parameters from the surface seismic elastic data for a complex subsurface scenario. In JMI-res, we first obtain the accurate local impulse
responses at the target depth level, while correctly accounting for overburden internal multiples and then we apply a localized inversion scheme on the estimated impulse responses to get the reservoir elastic parameters. Moreover,
the propagation velocity estimation is an integral part of JMI-res. In this paper, we show that JMI-res provides much more reliable local target impulse responses, thus yielding high-resolution elastic parameters, compared to
standard redatuming based on time reversal of recorded data, courtesy of proper handling of internal multiples in the redatuming step.@en

A previous geological and petrophysical model of the fluvio-deltaic Book Cliffs outcrops contained eight lithotypes, within each of which a number of lithologies were grouped. While this model was an adequate representation of the overall depositional architecture, for reservoir-geological purposes the potential reservoir and non-reservoir lithologies needed to be separated. Here, a new and more detailed geological model is presented in which more differentiation is put on the potential reservoir lithologies. This new model contains 12 lithologies with layers down to 1 m in thickness. Assuming a burial depth of 3 km and an average clay content, representative rock physical properties are assigned to lithologies based on published data. After the model thickness has been stretched by a factor of 4 in order to represent a more realistic reservoir, a full-waveform forward seismic response is modelled. These data are used as inputs into an iterative elastic wave-equation-based inversion scheme, with the goal to retrieve the rock properties and layer geometries. The results of this conceptual study show that sandstone units in the shoreface and distributary channels, which are potential reservoirs, are successfully identified. The recovery of medium parameters has a high resolution because the non-linear relationship between rock properties and the seismic data has been exploited. @en

A method is presented to link prior geological information, from wells and interpretation, to full waveform inversion at reservoir scale. The method converts the layer-based prior information to grid-based property probability density distributions that are highly non-Gaussian and is based on Bayes' Rule. The likelihood function for the unconstrained inversion is based on the Hessian of the inversion kernel and the minimum residual energy in the objective function. Good results have been obtained from a synthetic case study based on a very realistic outcrop model (Book Cliffs,Utah). Also results from a real data case study will be shown. @en

Inversion results from seismic data of a synthetic example based on the Cretaceous fluvio-deltaic Book Cliffs outcrops in Utah (USA) have been used to extract the reservoir parameters. The input data sets are compressibility and shear compliance which are from the full elastic wave-equation based inversion method. A fuzzy logic inference algorithm has been applied in which the lithology templates are based on well-logging data. The membership functions of the lithologies are constructed firstly. Then inversion results are used to predict the reservoir lithology. It is suggested that this classification method performs well because most of the time the same or similar lithologies have been predicted. However, this approach heavily depends on the input inversion results and therefore the full elastic wave-equation based inversion has been chosen.@en