Methods and systems for determining an image of a subterranean region of interest are disclosed. The method includes obtaining a seismic dataset and a geological dip model for the subterranean region of interest and determining a set of input seismic gathers from the seismic dataset. The method further includes determining a central seismic gather and a set of neighboring seismic gathers in a vicinity of the central seismic gather from the set of seismic gathers, determining a set of dip-corrected neighboring seismic gathers based, at least in part, on the set of neighboring seismic gathers and a geological dip from the geological dip model, and determining a noise-attenuated central seismic gather by combining the dip-corrected neighboring seismic gathers and the central seismic gather. The method still further includes forming the image of the subterranean region of interest based, at least in part, on the noise-attenuated central seismic gather.

Systems and methods for determining shear failure of a rock formation are disclosed. The method includes receiving, by a processor, a plurality of parameters related to physical properties of the rock formation, applying the plurality of parameters to a predetermined failure criterion, and determining shear failure of the rock formation based on the failure criterion. In some embodiments the failure criterion is a modified Hoek-Brown failure criterion that takes into consideration an intermediate principal stress, and the difference between normal stresses and an average confining stress.

Geologic modeling methods and systems disclosed herein employ an improved simulation gridding technique. For example, an illustrative geologic modeling method may comprise: obtaining a geologic model representing a faulted subsurface region in physical space; mapping the physical space geologic model to a design space model representing an unfaulted subsurface region; gridding the design space model to obtain a design space mesh; partitioning cells in the design space mesh with faults mapped from the physical space geologic model, thereby obtaining a partitioned design space mesh; mapping the partitioned design space mesh to the physical space to obtain a physical space simulation mesh; and outputting the physical space simulation mesh.

Systems, methods, and computer-readable media for using full waveform inversion for imaging surveyed mediums are provided. The full waveform inversion uses a Sifrian functional to fully leverage Hessian information and update a model by augmenting and assembling data derived from the Sifrian functional when equilibrated. The Sifrian inversion produces high resolution images of the surveyed medium typically only seen with full Hessian inversions and can produce such images without requiring supercomputer computation power or extremely long computation time.

Disclosed are methods, systems, and computer-readable medium to perform operations including: generating, using a source wavelet and a current velocity model, modeled seismic data of the subterranean formation; applying a pre-condition to a seismic data residual calculated using the modeled seismic data and acquired seismic data from the subterranean formation; generating a velocity update using the source wavelet and the pre-conditioned seismic data residual; updating, using the velocity update, the current velocity model to generate an updated velocity model; determining that the current velocity model satisfies a predetermined condition; and responsively determining that the updated velocity model is the velocity model of the subterranean formation.

This disclosure describes a system and method for generating a subsurface model representing lithological characteristics and attributes of the subsurface of a celestial body or planet. By automatically ingesting data from many sources, a machine learning system can infer information about the characteristics of regions of the subsurface and build a model representing the subsurface rock properties. In some cases, this can provide information about a region using inferred data, where no direct measurements have been taken. Remote sensing data, such as aerial or satellite imagery, gravimetric data, magnetic field data, electromagnetic data, and other information can be readily collected or is already available at scale. Lithological attributes and characteristics present in available geoscience data can be correlated with related remote sensing data using a machine learning model, which can then infer lithological attributes and characteristics for regions where remote sensing data is available, but geoscience data is not.

A method to model a subterranean formation of a field. The method includes obtaining a seismic volume comprising a plurality of seismic traces of the subterranean formation of the field, computing, based on the seismic volume, a seismically-derived value of a structural attribute representing a structural characteristic of the subterranean formation, computing, based on a structural model, a structurally-derived value of the structural attribute, the structural model comprising a plurality of structural layers of the of the subterranean formation, comparing the seismically-derived value and the structurally-derived value to generate a difference value representing a discrepancy of modeling the structural attribute at a corresponding location in the subterranean formation, and generating a seismic interpretation result based on the difference value and the corresponding location.

Examples described herein provide a computer-implemented method that includes analyzing a first dataset by applying the first dataset to a first model to generate a first result. The method further includes analyzing a second dataset by applying the second dataset to a second model to generate a second result. The method further includes performing validation on the first model and the second model by comparing the first result to the second result. The method further includes, responsive to determining that the first result and the second result match, modifying an operational action of a surface assembly based on at least one of the first result or the second result. The method further includes, responsive to determining that the first result and the second result do not match, updating at least one of the first model or the second model.

To generate dispersion curves for acoustic waves in a radially layered system, a matrix M containing solutions to the wave equation subject to the boundary conditions of the system is constructed. The reciprocal condition number (RCN) of the matrix M is determined as a function of acoustic wave frequency and slowness. The local minima of the RCN in the frequency-slowness plane produces the dispersion curves corresponding to allowable acoustic modes in the system. A sensitivity analysis which identifies the dispersion curves dependent on a selected parameter. The dispersion curves independent of the perturbed parameters are eliminated by perturbing the modeling parameters and generating the RCN of the perturbed matrix M and then subtracting the RCN values of the unperturbed matrix M, leaving the dispersion curves that exhibit dependence on the selected parameter.

A method for waveform inversion, the method including receiving observed data d, wherein the observed data d is recorded with sensors and is indicative of a subsurface of the earth; calculating estimated data p, based on a model m of the subsurface; calculating, using a trained neural network, a misfit function J.sub.ML; and calculating an updated model m.sub.t+1 of the subsurface, based on an application of the misfit function J.sub.ML to the observed data d and the estimated data p.