Methods, systems, and computer readable media for gridding of subsurface salt structures include determining a predetermined area lacks three dimensional seismic coverage, generating a two dimensional seismic top salt interpretation for the predetermined area, generating a bathymetry elevation of the predetermined area, determining that at least one two dimensional seismic line intersects a bathymetric feature of interest, and determining a correlation coefficient between the two dimensional seismic top salt interpretation and the bathymetry elevation. The method may further include determining the correlation coefficient is greater than a predetermined threshold value, and applying the bathymetry elevation as an additional control for gridding top of the subsurface salt structure. The step of gridding the top of the subsurface salt structure may further include applying at least one of kriging with external drift (KED), polygon-based approaches, regression-kriging, and other geostatistical methods.

A method including obtaining, by a computer processor, at least one key log in each of a set of training wells located, at least partially, within a hydrocarbon reservoir, identifying a target formation bounding surface in each of the set of training wells, and generating an initial depth surface for the target formation bounding surface from the target formation bounding surface in each of the set of training wells. The method further including, determining from the initial depth surface an initial depth estimate of the target formation bounding surface at a location of a current well, forming an objective function based, at least in part on a correlation between each key log in each of the set of training wells, and each corresponding key log in the current well, and optimizing the objective function by varying a depth shift between each of the set of training wells and the current well, to determine an optimum depth shift that produces an extremum of the objective function. The method still further including combining the initial depth estimate of the target formation bounding surface at the location of the current well with the optimum depth shift to produce a final depth estimate of the target formation bounding surface at the location of the current well.

Methods, systems, and computer readable media for gridding of subsurface salt structures include determining a predetermined area lacks three dimensional seismic coverage, generating a two dimensional seismic top salt interpretation for the predetermined area, generating a bathymetry elevation of the predetermined area, determining that at least one two dimensional seismic line intersects a bathymetric feature of interest, and determining a correlation coefficient between the two dimensional seismic top salt interpretation and the bathymetry elevation. The method may further include determining the correlation coefficient is greater than a predetermined threshold value, and applying the bathymetry elevation as an additional control for gridding top of the subsurface salt structure. The step of gridding the top of the subsurface salt structure may further include applying at least one of kriging with external drift (KED), polygon-based approaches, regression-kriging, and other geostatistical methods.

Geostatistical realizations are generated based on estimates, which are based on measurements for which one can compute corresponding variances. A shift field is generated by generating random values, which may be constrained according to a confidence restraint. A value of the shift field is applied to a standard deviation for an estimate calculated based on the corresponding variance for that estimate. The random values may be generated according to a gradient noise algorithm taking as an input a wavelength defining smoothness of an array of random values. The wavelength is decoupled from a grid spacing of the estimated values.

The subject matter of this specification can be embodied in, among other things, a method for geological modeling includes receiving a forward depositional model, determining a Latin Hypercube Sampling (LHS) stratigraphic model based on the projected forward depositional model, performing forward depositional modeling, transform the forward depositional model from time domain to stratigraphic-depth domain, determining one or more pseudo-wells based on the transformed model, determining a mismatch value based on the transformed forward depositional model and a collection of simulated physical value, and determining a kriging surrogate model based on the LHS stratigraphic model and the mismatch value.

Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, to generate generating geo-cellular models with improved lacunae. In one aspect, a method includes receiving a seismic dataset of a surveyed subsurface, and the seismic dataset includes seismic porosities in depth of the surveyed subsurface. Seismic porosities resampled into a three dimensional (3D) geological fine layer model grid. Seismic porosities at well locations are extracted using the 3D geological fine layer model grid. Log porosities and the seismic porosities are upscaled into coarse layers, and the coarse layers are identical for all the well locations. Match factors are determined based on differences between the upscaled log porosities and the downscaled seismic porosities. Co-krig the log porosities are correlated with the 3D geological fine layer model grid using the match factors as a soft constraint to produce a final 3D model.

A method for generating one or more reservoir models using machine learning is provided. Generating reservoir models is typically a time-intensive idiosyncratic process. However, machine learning may be used to generate one or more reservoir models that characterize the subsurface. The machine learning may use geological data, geological concepts, reservoir stratigraphic configurations, and one or more input geological models in order to generate the one or more reservoir models. As one example, a generative adversarial network (GAN) may be used as the machine learning methodology. The GAN includes two neural networks, including a generative network (which generates candidate reservoir models) and a discriminative network (which evaluates the candidate reservoir models), contest with each other in order to generate the reservoir models.

The present invention describes A computer implemented method for generating a validated geological model from three-dimensional (3D) seismic data and legacy data, the method comprising the steps of (a) receiving said legacy data from a first data source; (b) receiving said 3D seismic data from a second data source; (c) based on receiving said legacy data and said 3D seismic data, generating an adaptive geological model from said 3D seismic data, said adaptive geological model comprising at least one characteristic geological property; (d) generating at least one synthetic seismic data from at least a first region of interest of said adaptive geological model, the synthetic seismic data being adapted to determine a qualitative similarity value between at least said first region of interest of said adaptive geological model and a corresponding region of interest of said received 3D seismic data; (e) comparing said qualitative similarity value to a corresponding value within said legacy data; (f) adjusting said at least one characteristic geological property until said qualitative similarity value is within a predetermined threshold region of said corresponding value from said legacy data, and (g) automatically generating said validated geologic model including said at least one characteristic geological property that has been modified to be within said predetermined threshold region of said corresponding value from said legacy data.

In an exemplary embodiment, a method and system is disclosed for developing a subterranean geomechanics model of a complex geological environment. The method can include estimating a pore pressure field, a stress field, a geomechanics property field, and a geological structure field from a geological concept model; geostatistically interpolating vectors and tensors from the estimated fields; and combining the results from the estimated fields and the geostatistically interpolated vectors and tensors to derive a geostatistical geomechanical model of the geological environment.

A three-dimensional electrical resistivity tomography method and system belonging to the field of geological geophysical prospecting, the method including the steps of prospecting a region containing a geological anomaly with at least two prospecting modes respectively to acquire two-dimensional resistivity data of a corresponding detection plane; unifying coordinate systems of resistivity data points acquired in all prospecting modes, and extracting data points with the same coordinates; carrying out data fusion on extracted resistivity data at the same position by utilizing a principal component analysis method; and carrying out three-dimensional coordinate conversion on resistivity data acquired after fusion to form a three-dimensional model.