Search

We present a synthetic multi‐scale, multi‐physics dataset constructed from the Kimberlina 1.2 CO2 reservoir model based on a potential CO2 storage site in the Southern San Joaquin Basin of California. Among 300 models, one selected reservoir‐simulation scenario produces hydrologic‐state models at the onset and after 20 years of CO2 injection. Subsequently, these models were transformed into geophysical properties, including P‐ and S‐wave seismic velocities, saturated density where the saturating fluid can be a combination of brine and supercritical CO2, and electrical resistivity using established empirical petrophysical relationships. From these 3D distributions of geophysical properties, we have generated synthetic time‐lapse seismic, gravity and electromagnetic responses with acquisition geometries that mimic realistic monitoring surveys and are achievable in actual field situations. We have also created a series of synthetic well logs of CO2 saturation, acoustic velocity, density and induction resistivity in the injection well and three monitoring wells. These were constructed by combining the low‐frequency trend of the geophysical models with the high‐frequency variations of actual well logs collected at the potential storage site. In addition, to better calibrate our datasets, measurements of permeability and pore connectivity have been made on cores of Vedder Sandstone, which forms the primary reservoir unit. These measurements provide the range of scales in the otherwise synthetic dataset to be as close to a real‐world situation as possible. This dataset consisting of the reservoir models, geophysical models, simulated time‐lapse geophysical responses and well logs forms a multi‐scale, multi‐physics testbed for designing and testing geophysical CO2 monitoring systems as well as for imaging and characterization algorithms. The suite of numerical models and data have been made publicly available for downloading on the National Energy Technology Laboratory's (NETL) Energy Data Exchange (EDX) website. [ABSTRACT FROM AUTHOR]

Multiphysics inversion exploits different types of geophysical data that often complement each other and aims to improve overall imaging resolution and reduce uncertainties in geophysical interpretation. Despite the advantages, traditional multiphysics inversion is challenging because it requires a large amount of computational time and intensive human interactions for preprocessing data and finding trade‐off parameters. These issues make it nearly impossible for traditional multiphysics inversion to be applied as a real‐time monitoring tool for geological carbon storage. In this paper, we present a deep learning (DL) multiphysics network for imaging CO2 saturation in real time. The multiphysics network consists of three encoders for analysing seismic, electromagnetic and gravity data and shares one decoder for combining imaging capabilities of the different geophysical data for better predicting CO2 saturation. The network is trained on pairs of CO2 label models and multiphysics data so that it can directly image CO2 saturation. We use the bootstrap aggregating method to enhance the imaging accuracy and estimate uncertainties associated with CO2 saturation images. Using realistic CO2 label models and multiphysics data derived from the Kimberlina CO2 storage model, we evaluate the performance of the deep learning multiphysics network and compare its imaging results to those from the deep learning single‐physics networks. Our modelling experiments show that the deep learning multiphysics network for seismic, electromagnetic, and gravity data not only improves the imaging accuracy but also reduces uncertainties associated with CO2 saturation images. Our results also suggest that the deep learning multiphysics network for the non‐seismic data (i.e., electromagnetic and gravity) can be used as an effective low‐cost monitoring tool in between regular seismic monitoring. [ABSTRACT FROM AUTHOR]

Deep‐learning inversion has recently drawn attention in geological carbon storage research due to its potential of imaging and monitoring carbon storage in real time, significantly improving efficiency and safety of carbon storage operations. We present a deep‐learning full waveform inversion method that after the neural network has been trained can image CO2 saturation and its uncertainty in real time. Our deep‐learning inversion method is based on the U‐Net architecture with the neural network trained on pairs of synthetic seismic data and CO2 saturation models. Accordingly, our training establishes a mapping relationship between seismic data and CO2 saturation models and once fully trained directly estimates CO2 saturation as a function of subsurface location. We further quantify uncertainties of CO2 saturation estimates using the Monte Carlo dropout method and a bootstrap aggregating method. For this proof‐of‐concept study, the CO2 training models and data are derived from the Kimberlina 1.2 model, a hypothetical 3D geological carbon storage model that is constructed based on various geological and hydrological data from the Southern San Joaquin Basin, California. We perform deep‐learning inversion experiments using noise‐free and noisy training and test data sets and compare the results. Our modelling experiments show that (1) the deep‐learning inversion can estimate 2D distributions of CO2 fairly well even in the presence of Gaussian random noise and (2) both CO2 saturation imaging and uncertainty quantification can be done in real time. Our results suggest that the deep‐learning inversion method can serve as a robust real‐time monitoring tool for geological carbon storage and/or other time‐varying reservoir/aquifer properties that result from injection, extraction, and/or other subsurface transport phenomena. [ABSTRACT FROM AUTHOR]

In the past decade, rapid advances in distributed optical fibre sensing technologies have made it possible to record various geophysical data (e.g. strain, temperature and pressure) continuously in both time and space along the fibre, providing an unprecedented quantity and spatial density of data compared to traditional geophysical measurements as well as reducing data acquisition cost. To date, no distributed fibre‐based electromagnetic field sensing system has been implemented although electromagnetic sensing could have a broad range of applications to geophysical imaging and monitoring in borehole environments. The goal of this paper is to provide a theoretical feasibility study regarding the design and use of an electromagnetic sensing optical fibre for geophysical applications. First, we present the sensitivity analysis of a 'hypothetical' optical fibre coated with polyvinylidene fluoride, a polymer that provides relatively high piezoelectric properties, yet unlike ceramics, is flexible. Using a two‐dimensional electromagnetic modelling algorithm, we simulate the earth electric‐field‐to‐fibre‐strain transfer function and estimate the theoretical sensitivity of the optical fibre to electric fields. Given the state‐of‐the‐art distributed acoustic sensing strain sensitivities in the picometres strain range, our numerical modelling analysis suggests that a perfectly coupled polyvinylidene fluoride–coated optical fibre can measure electric field values in the mV/m to V/m amplitude range. We then apply a cylindrically symmetric modelling algorithm to simulate numerical models demonstrating the applicability of such a fibre in an oilfield environment. Scenarios investigated employ an electric field source and suggest that the measurements can be used to distinguish the oil versus water ratio with a fibre mounted inside a producing steel cased oil well as well as distinguishing between brine and hydrocarbon filled reservoir zones with a fibre located outside of the casing. [ABSTRACT FROM AUTHOR]

Cross-borehole ground penetrating radar (XBGPR) is used in monitoring a long-term vadose zone infiltration experiment at a test site in Socorro, NM in order to examine contaminant transport in the vadose zone. XBGPR attenuation tomography is conducted in order to test the ability of using images of electromagnetic attenuation for hydrogeologic investigations. The results of four pre-infiltration attenuation inversions shows standard deviations below 0.1 Np/m, and demonstrate the consistency of the XBGPR tomography technique for making time-lapse observations. Correlation to the core records indicates that XBGPR attenuation tomograms provide high-resolution images of clay distribution in the vadose zone. Water infiltration at the ground surface was initiated in February 1999 at a constant rate of 2.7 cm/day, and continued at this rate throughout the data collection experiment. Time-lapse attenuation tomograms show that attenuation increases by approximately 0.3 Np/m during the water infiltration, and indicate a snowplow effect may be occurring where salts are dissolved by the water and concentrated at the front of the plume. Seasonal temperature changes may also cause changes in electromagnetic attenuation images, and masking the evidence of water infiltration. Thus caution must be taken when using time-lapse attenuation images to interpret the movement of a water plume during a long-term experiment as temperature changes. [Copyright &y& Elsevier]

This paper presents recent advances and case histories of two non-traditional electromagnetic (EM) geophysical techniques for oil reservoir characterization and production/process monitoring. The crosswell electromagnetic induction method has seen rapid advancement since its inception in the early 1990s and, to this date, numerous surveys have been completed in active oil fields as well as other sites. Here, one example is given where a two-dimensional (2D) electrical conductivity image derived from crosswell data is employed for reservoir characterization, and two other examples presented where the method provided valuable insight into the two- and three-dimensional progress of water flood operations. The second, less mature method involves the development of a new generation single hole induction logging tool. With the Geothermal Borehole Induction Logging Tool (Geo-BILT), all three components of the magnetic field at multiple offsets from a three-component source are measured. The resulting data are inverted to produce 3D estimates of the electrical conductivity distribution surrounding the borehole. For this technology, we first demonstrate the concept using synthetic data and then show some preliminary field results. [Copyright &y& Elsevier]

Examines the formulation of three-dimensional (3-D) inverse solution for the magnetotelluric (MT) problem. Usage of finite difference methods; Interpretation of 3-D MT data; Reduction in the number of forward modelling applications; Complexity and realism of the images.

Microseismic observations during unconventional reservoir stimulation are typically seen as a proxy for clusters of hydraulic fractures and the extent of the stimulated reservoir. Such straightforward interpretation is often misleading and fails to provide a physically reasonable image of the fracturing process. This paper demonstrates the application of a physics-based machine learning algorithm which enables a rapid and accurate fracture mapping from the microseismic data. Our training and validation data set relies on a history-matched geomechanical modelling workflow implemented in GEOS software for the Hydraulic Fracturing Test Site 1 (HFTS-1) project. For this study we augmented the simulated fracture growth through geostatistical modelling of induced seismicity, so that the synthetic microseismic catalogue matches the main statistical properties of the field observations. We formulated the problem of mapping the actual fracture in the clutter of events to parallel common video segmentation workflows: several past video frames (microseismic density snapshots) are passed through a deep convolutional network to classify whether a given voxel is associated with a fracture or intact rock. We found that for accurate fracture mapping, the network's input and architecture must be augmented to incorporate the fluid injection parameters (pressure, rate, concentration of proppant, and location of the perforation within the cluster). The error rate for the network reached as little as 10 per cent of the fracture area, while a conventional microseismic interpretation approach yielded ∼300 per cent. Our approach also yields must faster predictions than conventional methods (minutes instead of weeks), and could enable engineers to make rapid decisions regarding engineering parameters (pumping rate, viscosity) in real time during stimulation. [ABSTRACT FROM AUTHOR]

Migration of clay fines can be a concern when less saline fluids are injected into brine‐saturated sandstone formations containing clays. If the salinity near fluid injection wells decreases below a critical value, the clay fines near the injection may detach, start migrating, and finally clog the pores. This effect can cause permeability decline near the well and may rapidly reduce the well injectivity. The focus of this work is on evaluating the impacts of clay fines migration on permeability decline in the field, using a numerical model and pressure buildup data collected during successive variable‐rate water injections in a deep sandstone reservoir. The numerical model accounts for the mixing of low‐salinity water with native brine and the migration of clay fines with the detachment and pore‐clogging processes. The model interpretation of the pressure buildup data implies that the observed reduction in well injectivity is mainly associated with the clay fines migration and related pore clogging near the well. The model reasonably well represents the pressure buildup data during the injections. Our simulations demonstrate that the permeability near the well can rapidly decline within the first hour of injection. The measured pressure buildup in post‐injection periods appears to decay more rapidly, compared to the simulation results of the model that assume irreversible permeability damage. This raises the question whether the permeability damage may be partly reversible near the well by backflow of brine after the injection of low‐salinity water. Article impact statement: The article shows clay fines detachment causes a rapid decline of near‐well permeability and this may be partly reversible by brine backflow. [ABSTRACT FROM AUTHOR]

SUMMARY An iterative finite element time-domain (FETD) method has been developed for simulating transient electromagnetic fields in 3-D diffusive earth media and has been verified through comparisons with analytic and finite-difference time-domain solutions. The adaptive time step doubling (ATSD) method plays an important role in reducing solution run-time by allowing large time steps in late time when high-frequency electric fields are increasingly attenuated in the Earth. We demonstrate that for the ATSD method to work effectively, the conductivity of the air and a drop tolerance of a preconditioner should be carefully selected. The conductivity of the air should not be too large for accurate simulations but also not too small for avoiding ill conditioning that results in error amplification in the ATSD method. A proper drop tolerance keeps the eigenvalues of the preconditioned FETD matrices clustered when a time step size is successively doubled during the ATSD processes, resulting in the convergence of iterative solutions with the reasonable number of iterations. A rule of thumb for determining the conductivity of the air and the drop tolerance has been presented. We also present the simultaneous multiple-sources modelling (SM2) approach. The SM2 approach simultaneously advances the electric fields excited by multiple individual sources in a single time stepping loop. This approach allows multiple sources to share the same preconditioner in the time stepping loop and improves the simulation efficiency per a survey line. [ABSTRACT FROM AUTHOR]

ABSTRACT The recent use of marine electromagnetic technology for exploration geophysics has primarily focused on applying the controlled source electromagnetic method for hydrocarbon mapping. However, this technology also has potential for structural mapping applications, particularly when the relative higher frequency controlled source electromagnetic data are combined with the lower frequencies of naturally occurring magnetotelluric data. This paper reports on an extensive test using data from 84 marine controlled source electromagnetic and magnetotelluric stations for imaging volcanic sections and underlying sediments on a 128-km-long profile. The profile extends across the trough between the Faroe and Shetland Islands in the North Sea. Here, we focus on how 2.5D inversion can best recover the volcanic and sedimentary sections. A synthetic test carried out with 3D anisotropic model responses shows that vertically transverse isotropy 2.5D inversion using controlled source electromagnetic and magnetotelluric data provides the most accurate prediction of the resistivity in both volcanic and sedimentary sections. We find the 2.5D inversion works well despite moderate 3D structure in the synthetic model. Triaxial inversion using the combination of controlled source electromagnetic and magnetotelluric data provided a constant resistivity contour that most closely matched the true base of the volcanic flows. For the field survey data, triaxial inversion of controlled source electromagnetic and magnetotelluric data provides the best overall tie to well logs with vertically transverse isotropy inversion of controlled source electromagnetic and magnetotelluric data a close second. Vertical transverse isotropy inversion of controlled source electromagnetic and magnetotelluric data provided the best interpreted base of the volcanic horizon when compared with our best seismic interpretation. The structural boundaries estimated by the 20-Ω·m contour of the vertical resistivity obtained by vertical transverse isotropy inversion of controlled source electromagnetic and magnetotelluric data gives a maximum geometric location error of 11% with a mean error of 1.2% compared with the interpreted base of the volcanic horizon. Both the model study and field data interpretation indicate that marine electromagnetic technology has the potential to discriminate between low-resistivity prospective siliciclastic sediments and higher resistivity non-prospective volcaniclastic sediments beneath the volcanic section. [ABSTRACT FROM AUTHOR]