Gregory A. Newman

Direct reservoir parameter estimation using joint inversion of marine seismic AVA and CSEM data

A new joint inversion algorithm to directly estimate reservoir parameters is described. This algorithm combines seismic amplitude versus angle (AVA) and marine controlled source electromagnetic (CSEM) data. The rock-properties model needed to link the geophysical parameters to the reservoir parameters is described. Errors in the rock-properties model parameters, measured in percent, introduce errors of comparable size in the joint inversion reservoir parameter estimates. Tests of the concept on synthetic one-dimensional models demonstrate improved fluid saturation and porosity estimates for joint AVA-CSEM data inversion (compared to AVA or CSEM inversion alone). Comparing inversions of AVA, CSEM, and joint AVA-CSEM data over the North Sea Troll field, at a location with well control, shows that the joint inversion produces estimated gas saturation, oil saturation and porosity that is closest (as measured by the RMS difference, L1 norm of the difference, and net over the interval) to the logged values whereas CSEM inversion provides the closest estimates of water saturation.

A parallel finite-difference approach for 3D transient electromagnetic modeling with galvanic sources

A parallel finite-difference algorithm for the solution of diffusive, three-dimensional (3D) transient electromagnetic field simulations is presented. The purpose of the scheme is the simulation of both electric fields and the time derivative of magnetic fields generated by galvanic sources (grounded wires) over arbitrarily complicated distributions of conductivity and magnetic permeability. Using a staggered grid and a modified DuFort-Frankel method, the scheme steps Maxwells equations in time. Electric field initialization is done by a conjugate-gradient solution of a 3D Poisson problem, as is common in 3D resistivity modeling. Instead of calculating the initial magnetic field directly, its time derivative and curl are employed in order to advance the electric field in time. A divergence-free condition is enforced for both the magnetic-field time derivative and the total conduction-current density, providing accurate results at late times. In order to simulate large realistic earth models, the algorithm has been designed to run on parallel computer platforms. The upward continuation boundary condition for a stable solution in the infinitely resistive air layer involves a two-dimensional parallel fast Fourier transform. Example simulations are compared with analytical, integral-equation and spectral Lanczos decomposition solutions and demonstrate the accuracy of the scheme.

Imaging CSEM data in the presence of electrical anisotropy

IMAGING CSEM DATA IN THE PRESENCE OF ELECTRICAL ANISOTROPY Gregory A. Newman * , Michael Commer * and James J. Carazzone + Earth Sciences Division, Lawrence Berkeley National Laboratory, Berkeley California ExxonMobil Upstream Research Company, Houston Texas Email: ganewman@lbl.gov ABSTRACT Formation anisotropy should be incorporated into the analysis of controlled source electromagnetic (CSEM) data because failure to do so can produce serious artifacts in the resulting resistivity images for certain data configurations of interest. This finding is demonstrated in model and case studies. Sensitivity to horizontal resistivity will be strongest in the broadside electric field data where detectors are offset from the tow line. Sensitivity to the vertical resistivity is strongest for over flight data where the transmitting antenna passes directly over the detecting antenna. Consequently, consistent treatment of both over flight and broadside electric field measurements requires an anisotropic modeling assumption. To produce a consistent resistivity model for such data we develop and employ a 3D CSEM imaging algorithm that treats transverse anisotropy. The algorithm is based upon non-linear conjugate gradients and full wave equation modeling. It exploits parallel computing systems to effectively treat 3D imaging problems and CSEM data volumes of industrial size. Here we use it to demonstrate the anisotropic imaging process on model and field data sets from the North Sea and offshore Brazil. We also verify that isotropic imaging of over flight data alone produces an image generally consistent with the vertical resistivity. However, superior data fits are obtained when the same over flight data are analyzed assuming an anisotropic resistivity model.

A Parallel Finite-Difference Approach for Three-Dimensional Transient Electromagnetic Modeling With Galvanic Sources

A parallel finite-difference algorithm for the solution of diffusive, three-dimensional (3D) transient electromagnetic field simulations is presented. The purpose of the scheme is the simulation of both electric fields and the time derivative of magnetic fields generated by galvanic sources (grounded wires) over arbitrarily complicated distributions of conductivity and magnetic permeability. Using a staggered grid and a modified DuFort-Frankel method, the scheme steps Maxwells equations in time. Electric field initialization is done by a conjugate-gradient solution of a 3D Poisson problem, as is common in 3D resistivity modeling. Instead of calculating the initial magnetic field directly, its time derivative and curl are employed in order to advance the electric field in time. A divergence-free condition is enforced for both the magnetic-field time derivative and the total conduction-current density, providing accurate results at late times. In order to simulate large realistic earth models, the algorithm has been designed to run on parallel computer platforms. The upward continuation boundary condition for a stable solution in the infinitely resistive air layer involves a two-dimensional parallel fast Fourier transform. Example simulations are compared with analytical, integral-equation and spectral Lanczos decomposition solutions and demonstrate the accuracy of the scheme.

Solution accelerators for large scale 3D electromagnetic inverse problems

We provide a framework for preconditioning nonlinear 3D electromagnetic inverse scattering problems using nonlinear conjugate gradient (NLCG) and limited memory (LM) quasi-Newton methods. Key to our approach is the use of an approximate adjoint method that allows for an economical approximation of the Hessian that is updated at each inversion iteration. Using this approximate Hessian as a preconditoner, we show that the preconditioned NLCG iteration converges significantly faster than the non-preconditioned iteration, as well as converging to a data misfit level below that observed for the non-preconditioned method. Similar conclusions are also observed for the LM iteration; preconditioned with the approximate Hessian, the LM iteration converges faster than the non-preconditioned version. At this time, however, we see little difference between the convergence performance of the preconditioned LM scheme and the preconditioned NLCG scheme. A possible reason for this outcome is the behavior of the line search within the LM iteration. It was anticipated that, near convergence, a step size of one would be approached, but what was observed, instead, were step lengths that were nowhere near one. We provide some insights into the reasons for this behavior and suggest further research that may improve the performance of the LM methods.

Massively parallel electrical-conductivity imaging of hydrocarbons using the IBM Blue Gene/L supercomputer

Large-scale controlled source electromagnetic (CSEM) three-dimensional (3D) geophysical imaging is now receiving considerable attention for electrical conductivity mapping of potential offshore oil and gas reservoirs. To cope with the typically large computational requirements of the 3D CSEM imaging problem, our strategies exploit computational parallelism and optimized finite-difference meshing. We report on an imaging experiment, utilizing 32,768 tasks/processors on the IBM Watson Research Blue Gene/L (BG/L) supercomputer. Over a 24-hour period, we were able to image a large scale marine CSEM field data set that previously required over four months of computing time on distributed clusters utilizing 1024 tasks on an Infiniband fabric. The total initial data misfit could be decreased by 67 percent within 72 completed inversion iterations, indicating an electrically resistive region in the southern survey area below a depth of 1500 m below the seafloor. The major part of the residual misfit stems from transmitter parallel receiver components that have an offset from the transmitter sail line (broadside configuration). Modeling confirms that improved broadside data fits can be achieved by considering anisotropic electrical conductivities. While delivering a satisfactory gross scale image for the depths of interest, the experiment provides important evidence for the necessity of discriminating between horizontal and vertical conductivities for maximally consistent 3D CSEM inversions.

3D modeling of a deepwater EM exploration survey

Analysis of a current offshore prospect employed 3D numerical modeling of a controlled-source electromagnetic (CSEM) exploration system. The analysis considers the sensitivity of data presentations to assumptions about the background model. The numerical simulations show that false anomalies and significant distortion to anomaly magnitude can be caused by normalization of the observed electric fields by reference fields calculated from an incorrect or oversimplified background model. Bathymetry effects on the measured electric fields, if not accounted for, can produce anomalies as large as those of target sands. The maximum sensitivity to oil-water contacts or other strong lateral variations within the modeled channel sands is achieved by profiling along the length of the channel. Profiles run offset from a simulated oil-water contact by as little as 2 km show a response below the expected noise levels. Good background models can be constructed by taking advantage of the magnetotelluric data recorded by m...

A Review of High-Performance Computational Strategies for Modeling and Imaging of Electromagnetic Induction Data

Many geoscientific applications exploit electrostatic and electromagnetic fields to interrogate and map subsurface electrical resistivity—an important geophysical attribute for characterizing mineral, energy, and water resources. In complex three-dimensional geologies, where many of these resources remain to be found, resistivity mapping requires large-scale modeling and imaging capabilities, as well as the ability to treat significant data volumes, which can easily overwhelm single-core and modest multicore computing hardware. To treat such problems requires large-scale parallel computational resources, necessary for reducing the time to solution to a time frame acceptable to the exploration process. The recognition that significant parallel computing processes must be brought to bear on these problems gives rise to choices that must be made in parallel computing hardware and software. In this review, some of these choices are presented, along with the resulting trade-offs. We also discuss future trends in high-performance computing and the anticipated impact on electromagnetic (EM) geophysics. Topics discussed in this review article include a survey of parallel computing platforms, graphics processing units to multicore CPUs with a fast interconnect, along with effective parallel solvers and associated solver libraries effective for inductive EM modeling and imaging.

An accelerated time domain finite difference simulation scheme for three‐dimensional transient electromagnetic modeling using geometric multigrid concepts

[1]xa0The fact that the transient electromagnetic (TEM) field is smoothed gradually in space with time allows for a reduced spatial sampling rate of the EM field. On the basis of concepts known from multigrid methods, we have developed a restriction operator in order to map the EM field and the material properties from a fine to a coarser finite difference mesh during a forward field simulation with an explicit time-stepping scheme. Two advantages follow. First, the grid size can be reduced. Field restriction involves reducing the number of grid nodes by a factor of 2 for each Cartesian direction. Second, as can be seen from the Courant-Friedrichs-Levy condition, the larger grid spacing allows for proportionally larger time step sizes. After field restriction, a material averaging scheme is employed in order to calculate the underlying effective medium on the coarse simulation grid. Example results show a factor of up to 5 decrease in solution run time, compared to a scheme that uses a constant grid. Key to the accuracy of the approach is knowledge of the proper time range to restrict the fields. An adequate criterion to decide during run time when to restrict involves an error measure for the locations of interest between the fields on the fine mesh and the restricted fields.

MPiTOUGH2: A parallel parameter estimation framework for hydrological and hydrogeophysical applications

We present a parallel joint hydrogeophysical parameter estimation framework specifically relevant for a class of inverse modeling applications where a large number of simulations of multi-phase, multi-component flow and transport through porous media impose exceedingly large computing demands. A modified Levenberg-Marquardt minimization algorithm provides for a robust and efficient calibration of complex models. The optimization framework is based on the parameter estimation and uncertainty analysis tool iTOUGH2, which we have parallelized using the Message Passing Interface in order to address the main computational burden of assessing parameter sensitivities. An underlying layer of hydrological and geophysical forward simulation operators use domain decomposition and parallel iterative Krylov solver techniques. The geophysical forward simulation operators originate from parallel algorithms for electrical and electromagnetic data types that have proven successful in solving large-scale imaging problems arising in geothermal as well as oil and gas exploration applications. We have pursued a consequent merge of the hydrological optimization framework with the geophysical component in order to maximize the efficiencies of the Message Passing Interface. The method offers new possibilities by combining hydrological data with geophysical measurements that involve, for example, time-harmonic electromagnetic fields. We first show improved model resolution capabilities on a synthetic joint inversion example where controlled-source electromagnetic observations are combined with hydrological data simulated from a conservative tracer injection experiment. Next, the method is applied to a 3-D joint inversion of field data from a CO2 injection experiment, where the required multi-phase, multi-component flow and transport simulations are highly computationally demanding. Overall improved data fits are achieved for both CO2 gas mole fractions and observed relative changes in electrical conductivity derived from geophysical measurements.