A computer-implemented method is described for seismic facies identification including receiving a seismic dataset representative of a subsurface volume of interest; applying a model conditioned by petrophysical classifications to the seismic dataset to identify seismic facies and generate a classified seismic image; and identifying geologic features based on the classified seismic image. The method generates seismic facies probability volumes.

The present disclosure relates to a method for determination of a real subsoil geological formation. In at least one embodiment, the method includes receiving a model representing the real subsoil. The model includes a shore line and a fluvial formation connected to the shore line at a point of the model. The shore line divides the model into a marine zone and a continental zone. The method further includes determining a delta zone in the model based on the shore line and based on the fluvial formation, determining a stochastic trajectory in said delta zone, and determining a lobe formation in said delta zone based on the determined stochastic trajectory based on a stochastic process.

A method and apparatus for modeling a subsurface region, including: obtaining a training set of geologically plausible models for the subsurface region; training an autoencoder with the training set; extracting a decoder from the trained autoencoder, wherein the decoder comprises a geologic-model-generating function; using the decoder within a data-fitting objective function to replace output-space variables of the decoder with latent-space variables, wherein a dimensionality of the output-space variables is greater than a dimensionality of the latent-space variables; and performing an inversion by identifying one or more minima of the data-fitting objective function to generate a set of prospective latent-space models for the subsurface region; and using the decoder to convert each of the prospective latent-space models to a respective output-space model. A method and apparatus for making one or more hydrocarbon management decisions based on the estimated uncertainty.

A method for generating a stochastic emulation model is disclosed. In an embodiment, the method uses a computer (100) having a processor (110) configured to execute commands stored on a non-transitory memory element (120), the non-transitory memory element (120) having an emulation module (124). The method includes generating, by the emulation module (124), a stochastic emulation model (3) from the output of a deterministic emulation model (η).

A method for simulating the evolution of a geological gridded model of an area comprising a plurality of cells, over a predetermined period of time T, comprising: a) assigning a water depth to each cell, b) determining, for each cell, a direction and velocity of a water current, c) introducing a number of particles in the model, d) transporting each particle based on the direction and velocity of the current, and e) updating the model according to the transport of each particle, wherein the current is a tidal current, comprising alternate phases of rising tide current and falling tide current, simulating the evolution of the model over the period of time T comprises iterating steps a. to e. a number of times equal to 2k, each iteration of steps a. to e. corresponds to simulating the evolution of the model over a period of time T/2k, each iteration of step b. is performed the rising tide or falling tide current, and two successive iterations of step b. are performed for each one of the rising tide current and falling tide current.

Implementations of the present disclosure provide techniques for stochastic stratigraphic correlation that produce quantitative estimation of uncertainty by generating multiple viable correlations. In contrast to deterministic models where the output of the model is fully determined by the parameter values and the initial conditions, stochastic models possess some inherent randomness. Using stochastic models, the same set of parameter values and initial conditions will lead to an ensemble of different outputs. The described techniques include introducing probabilistic sampling into dynamic time warping processes. In some implementations, the techniques allow for a user to interactively update correlation markers and their associated uncertainties. The techniques can adhere to these markers and produce multiple viable solutions.

Some implementations provide a method including: accessing measurement data that characterize one or more features of a reservoir, wherein the measurement data are from more than well locations of the reservoir and from a range of depths inside the reservoir; detecting portions of the measurement data that characterize the one or more features with a statistical metric that is below a pre-determined threshold; based on removing the portions of the measurement data, identifying a plurality of layers along the range of depths of the reservoir; within each layer of the plurality of layers, grouping the measurements data among a plurality of clusters, each corresponding to a flow unit (FU) and determined by a machine learning algorithm; generating a three-dimensional (3D) permeability model of the reservoir based on the FU of each layer and a saturation height function; and simulating a performance of the reservoir based on the 3D permeability model.

Analysis and display of source-to-sink information according to some aspects includes grouping geochronological data associated with a sediment sample into optimized subpopulations within a reference population and target populations, and producing Gaussian functions for the reference population and the target populations using the subpopulations as a priori constraints. The Gaussian functions describe a distribution of zircons. The subpopulations within the reference population and the target populations are compared based on at least one statistical attribute from the Gaussian functions to identify areas of sediment provenance, and the areas of sediment provenance are displayed in various ways, for example, on a paleographic map as of an age of deposition of the sediment sample. A sink-to-sink analysis can also be performed to identify dissimilarities between samples.

The present disclosure describes a computer-implemented method that includes: receiving a seismic dataset of a surveyed subsurface of a reservoir, the seismic dataset comprising observed pressure and production data of the reservoir as well as a set of geological and geo-mechanical parameters representing physical features of the surveyed subsurface; generating multiple realizations of a discrete fracture network (DFN) based on a subset of the set of geological and geo-mechanical parameters; selecting, from the multiple realizations, one or more realizations based on a parameter with a value under a 10% quantile of a full range of likely values; performing a forward simulation for the reservoir based on the selected one or more realizations and the observed pressure and production data; determining that a misfit of the forward simulation is below a threshold based on evaluating an objective function; and producing a model of the reservoir based on the forward simulation.

Methods for coordinate-related despiking of hydrocarbon reservoir data include receiving, by a computer system, multiple datapoints of a geomechanical property of a hydrocarbon reservoir modeled by a three-dimensional (3D) grid. Each datapoint corresponds to 3D coordinates of the 3D grid. For each datapoint, the computer system aggregates the datapoint with a noise component generated using the 3D coordinates corresponding to the datapoint. The computer system determines that the aggregated datapoint is unique to the multiple datapoints. The computer system performs a transform on the datapoints for Gaussian simulation. A display device of the computer system generates a graphical representation of the geomechanical property of the hydrocarbon reservoir based on the Gaussian simulation of the transformed datapoints.