Abstract: A method may include obtaining an input gather regarding a geological region of interest. The method may further include obtaining parameterization data regarding a seismic processing operation. The parameterization data may correspond to a first set of process parameter values that are different from a second set of process parameter values that are used to generate the input gather. The method may further include generating a predicted output gather using a machine-learning model, the input gather, and the parameterization data. The machine-learning model may include an encoder model and a decoder model. The method may further include generating a seismic image of the geological region of interest using the predicted output gather.

Abstract: Systems and methods include a method for deblending signal and noise data. A shot domain for actual sources, a receiver domain for virtual sources, and a receiver domain for actual sources are generated from blended shot data. A dictionary of signal atoms is generated. Each signal atom includes a small patch of seismic signal data gathered during a small time window using multiple neighboring traces. A dictionary of noise atoms is generated. Each noise atom includes a small patch of seismic noise data gathered during a small time window using multiple neighboring traces. A combined signal-and-noise dictionary is generated that contains the signal atoms and the noise atoms. A sparse reconstruction of receiver domain data is created from the combined signal-and-noise dictionary. The sparse reconstruction is split into deblended data and blending noise data based on atom usage to create deblended shot domain gathers for actual sources.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for echo-deblending using coincidence-filtering of offshore seismic data.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for training a seismic data de-blending model. In one aspect, a method comprises: obtaining a plurality of de-blending training examples, wherein each de-blending training example defines: (i) one or more blended seismic traces, and (ii) for each blended seismic trace, a corresponding plurality of target unblended seismic traces; using the de-blending training examples to train a de-blending model having a plurality of de-blending model parameters, comprising, for each de-blending training example: processing the one or more blended seismic traces of the training example using the de-blending model to generate an output which defines, for each of the one or more blended seismic traces of the training example, a corresponding plurality of estimated unblended seismic traces; and adjusting values of the plurality of de-blending model parameters.

Abstract: Systems and methods include a method for deblending signal and noise data. A shot domain for actual sources, a receiver domain for virtual sources, and a receiver domain for actual sources are generated from blended shot data. A dictionary of signal atoms is generated. Each signal atom includes a small patch of seismic signal data gathered during a small time window using multiple neighboring traces. A dictionary of noise atoms is generated. Each noise atom includes a small patch of seismic noise data gathered during a small time window using multiple neighboring traces. A combined signal-and-noise dictionary is generated that contains the signal atoms and the noise atoms. A sparse reconstruction of receiver domain data is created from the combined signal-and-noise dictionary. The sparse reconstruction is split into deblended data and blending noise data based on atom usage to create deblended shot domain gathers for actual sources.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for training a seismic data de-blending model. In one aspect, a method comprises: obtaining a plurality of de-blending training examples, wherein each de-blending training example defines: (i) one or more blended seismic traces, and (ii) for each blended seismic trace, a corresponding plurality of target unblended seismic traces; using the de-blending training examples to train a de-blending model having a plurality of de-blending model parameters, comprising, for each de-blending training example: processing the one or more blended seismic traces of the training example using the de-blending model to generate an output which defines, for each of the one or more blended seismic traces of the training example, a corresponding plurality of estimated unblended seismic traces; and adjusting values of the plurality of de-blending model parameters.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for echo-deblending using coincidence-filtering of offshore seismic data.

Abstract: Computational systems and methods for interpolating a pressure wavefield based on pressure wavefield and particle motion wavefield data are disclosed. The pressure and particle motion wavefields are sampled at a sample rate that is less than a lower bound for sample rates typically used to interpolate the pressure wavefield from the pressure wavefield samples alone. The particle motion wavefield can an acceleration wavefield or a time derivative of a velocity wavefield.

Abstract: One embodiment relates to a method of wavefield regularization for geophysical data acquisition of seismic geophysical data. Measured traces, are obtained from an array of sensors. For each grid point on a processing grid, best-fitting traces of the measured traces are found. Using the best-fitting traces, spectral amplitudes of down-going and up-going wavefields are computed. The down-going and up-going wavefields are subsequently transformed to an output grid in a space-time domain. Another embodiment relates to an apparatus for wavefield regularization of geophysical data acquisition. Other embodiments, aspects and features are also disclosed.

Abstract: Computational systems and methods for interpolating a pressure wavefield based on pressure wavefield and particle motion wavefield data are disclosed. The pressure and particle motion wavefields are sampled at a sample rate that is less than a lower bound for sample rates typically used to interpolate the pressure wavefield from the pressure wavefield samples alone. The particle motion wavefield can an acceleration wavefield or a time derivative of a velocity wavefield.

Abstract: One embodiment relates to a method of wavefield regularization for geophysical data acquisition of seismic geophysical data. Measured traces, are obtained from an array of sensors. For each grid point on a processing grid, best-fitting traces of the measured traces are found. Using the best-fitting traces, spectral amplitudes of down-going and up-going wavefields are computed. The down-going and up-going wavefields are subsequently transformed to an output grid in a space-time domain. Another embodiment relates to an apparatus for wavefield regularization of geophysical data acquisition. Other embodiments, aspects and features are also disclosed.

Abstract: A best fitting trace in seismic data is determined for a desired trace to be reconstructed. A dip-based correction is calculated per trace and per sample for differences in azimuth, common midpoint coordinates, and offset between the best fitting trace and the desired trace. The dip-based correction is applied to the best fitting trace to reconstruct the desired trace for 3D surface-related multiple prediction.

Abstract: An operator is applied in a frequency domain to multiple source seismic data acquired in a marine survey, with multiple sources substantially horizontally collocated and at different depths, to generate an estimate of a primary source wavefield, the operator constructed to compensate for time delays between and for different depths of the primary and secondary sources. The primary source wavefield estimate is adaptively subtracted from the multiple source seismic data to generate secondary source wavefield estimates. Another operator is applied in the frequency domain to the secondary source wavefield estimates to generate another primary source wavefield estimate, the operator constructed to compensate for time delays between and for different depths of the primary and secondary sources. The primary and secondary source estimates are used, after further iterative refinements to remove noise, to generate upgoing and downgoing wavefield components of the multiple source seismic data.

Abstract: Dips are calculated for a series of sets of adjacent multiple contribution traces, from seismic data representative of subsurface formations, in the vicinity of a boundary of an aperture of a multiple contribution gather, the seismic data acquired by deploying a plurality of seismic sensors proximate an area of the earth's subsurface to be evaluated, the seismic sensors generating at least one of an electrical and optical signal in response to seismic energy. The boundary of the aperture of the multiple contribution gather is recursively extended, based on the calculated dips. Multiple contribution traces in the multiple contribution gather with the extended aperture boundary are stacked to generate predicted multiple traces. The predicted multiple traces are subtracted from the seismic data to generate surface-related multiple attenuated data useful for imaging the subsurface formations.

Abstract: A best fitting trace in seismic data is determined for a desired trace to be reconstructed. An azimuth correction is calculated for the azimuth difference between the best fitting trace and the desired trace. The azimuth correction is applied to the best fitting trace to reconstruct the desired trace for 3D surface-related multiple prediction.

Abstract: An operator is applied in a frequency domain to multiple source seismic data acquired in a marine survey, with multiple sources substantially horizontally collocated and at different depths, to generate an estimate of a primary source wavefield, the operator constructed to compensate for time delays between and for different depths of the primary and secondary sources. The primary source wavefield estimate is adaptively subtracted from the multiple source seismic data to generate secondary source wavefield estimates. Another operator is applied in the frequency domain to the secondary source wavefield estimates to generate another primary source wavefield estimate, the operator constructed to compensate for time delays between and for different depths of the primary and secondary sources. The primary and secondary source estimates are used, after further iterative refinements to remove noise, to generate upgoing and downgoing wavefield components of the multiple source seismic data.

Abstract: A best fitting trace in seismic data is determined for a desired trace to be reconstructed. An azimuth correction is calculated for the azimuth difference between the best fitting trace and the desired trace. The azimuth correction is applied to the best fitting trace to reconstruct the desired trace for 3D surface-related multiple prediction.

Abstract: A best fitting trace in seismic data is determined for a desired trace to be reconstructed. A dip-based correction is calculated per trace and per sample for differences in azimuth, common midpoint coordinates, and offset between the best fitting trace and the desired trace. The dip-based correction is applied to the best fitting trace to reconstruct the desired trace for 3D surface-related multiple prediction.

Abstract: Dips are calculated for a series of sets of adjacent multiple contribution traces, from seismic data representative of subsurface formations, in the vicinity of a boundary of an aperture of a multiple contribution gather, the seismic data acquired by deploying a plurality of seismic sensors proximate an area of the earth's subsurface to be evaluated, the seismic sensors generating at least one of an electrical and optical signal in response to seismic energy. The boundary of the aperture of the multiple contribution gather is recursively extended, based on the calculated dips. Multiple contribution traces in the multiple contribution gather with the extended aperture boundary are stacked to generate predicted multiple traces. The predicted multiple traces are subtracted from the seismic data to generate surface-related multiple attenuated data useful for imaging the subsurface formations.