NS31A-1561
A general self-potential model including redox effects, the influence of the Reynolds number, and multi-phase flow. Theory and applications
We determine the macroscopic transport properties of isotropic microporous media by volume-averaging the local Nernst-Planck and Navier-Stokes equations in non-isothermal conditions incorporating streaming, diffusional, and thermal effects. In porous media, the excess of charge, that counterbalance the charge deficiency of the surface of the minerals, is partitioned between the Gouy-Chapman and the Stern layer. Rather than using Poisson-Boltzmann distributions to describe the ionic concentrations in the pore space of the medium, we rely on Donnan distributions obtained by equating the chemical potentials of the water molecules and ions between a reservoir of ions and the pore space of the medium. The macroscopic Maxwell equations and the macroscopic linear constitutive transport equations are derived in the vicinity of equilibrium assuming that the porous material is deformable. In the vicinity of thermodynamic equilibrium, the cross-coupling phenomena of the macroscopic constitutive equations of transport follow Onsager reciprocity. In addition, all the material properties entering the constitutive equations only depend on two textural properties, the permeability and the electrical formation factor. Extension of this model is then performed incorporating three distinct additional effects (a) extension of the electrokinetic equations at high Reynolds numbers, in the inertial laminar flow regime ; (b) extension of the model to multi-phase flow conditions (under the assumptions that the two fluid phases are continuous), and (3) introduction of electro-redox theory and the development of a fundamental model of bio-battery with application to contaminant plumes. Applications will be discussed shortly regarding (1) fracturing, (2) ground water flow, (c) leakage in embankment dams, and (d) contaminant plumes rich in organic matter (http://www.andre-revil.com).
http://www.andre-revil.com
NS31A-1562
Quantifying the effects of unknown resistivity structure on self-potential data analysis
Self-potential data are the result of a source mechanism (e.g. electrokinetic, electrochemical, etc.) that generates an electric field in the earth. Without complete knowledge about the resistivity structure of the medium, one can not properly interpret the source properties of interest. In this study, we quantify how uncertainties in the resistivity structure can influence self-potential data analysis and source inversion results. Several synthetic forward and inverse examples are used to illustrate the sensitivity of measured data and inversion results to various unknown resistivity structures. Additionally, we show self-potential source inversion examples from field datasets using different resistivity assumptions, where the true resistivity structure is unknown.
NS31A-1563
Can we Retrieve the Seepage Velocity From Self-Potential Anomalies Associated With Leakages in Dams and Embankments?
Abstract: Leakage in dams and embankments are responsible for self-potential (SP) anomalies, which can be recorded using non-polarizable electrodes. A set of laboratory experiments using glass bead packs were made in order to understand the influence of the Reynolds number, Re, on the amplitude of the self-potential signals. At low Re numbers, the coupling coefficient decreases with the increase of Reynolds number. At high Re numbers, the coupling coefficient and the permeability, decreases with the increase of Re numbers. All these variations are described by a new model that we called the BCR model (Streaming potentials of granular media. Influence of the Dukhin and Reynolds numbers. Submitted in Journal of Geophysical Research). A finite element numerical simulation (FEMLAB Comsol Mutliphysics 3.2) is performed, using the BCR model, to see if there is a possibility to quantitatively estimate the seepage velocity associated with these self-potential anomalies. This simulation was done on a water filled basin with a preferential fluid flow associated with a gravel-filled pipe. The magnitude of the SP signals depends also on the conductivity of the ground water that can be also measured in the field. A sensitivity analysis shows that, in some conditions, the seepage velocity associated with leakage can be quantitatively estimated from the magnitude of the SP signals providing of the value of the conductivity of water.
NS31A-1564
Self-potential potential monitoring of the small catchment area of La Soutte: acquisition and modelling
Several theoretical studies, laboratory experiments, and field tests, have shown that Self-Potential can be used to characterize the dynamics of underground water flow. Indeed, water flow through porous materials generates electric currents through electrokinetic coupling. Most case studies reported in the literature consider either continuous measurements or repeated surveys in application to the modelling of geothermal flow through fractures, rainfall infiltration, or water-table imaging. We consider a long-term study that uses both continuous measurements and repeated surveys in order to monitor hydrological processes in a small catchment area in a mountainous environment. Since 2004, our group has started to investigate La Soutte (Vosges, France), a 6-hectare glade at 950 m altitude. The basement is shallow ( < 20 m) and outcrops at places, it consists of fractured granitic and volcanic rocks. Weather conditions change strongly throughout the year; thus the site undergoes contrasting natural stimulations by the rainfall, storms, aquifer level changes, temperature and solar radiation. On the one hand, repeated surveys using unpolarizable electrodes at 10-20 cm depth show a maximum horizontal SP variation of 10 to 30 mV. Numerical modelling using a coupled 2D hydro-electric code shows Darcy velocities of about 1e-11 m/s. An electrokinetic coupling coefficient C=1.5e-13 V/Pa has been measured on samples. On the other hand, continuous measurements emphasize time variations. We use them to study rainfall events and combined effects of vertical (1D infiltration) and horizontal flow (topographic gradient, especially near streams). SP data must be used with caution as they are affected to several sources. We present quantitative methods to appraise their quality and uncertainties: histograms of measurements for repeated surveys, effect of magnetic storms on monitoring electrodes,...
NS31A-1565
Fluid flow monitoring in oilfields using downhole measurements of electrokinetic potential
Permanently installed downhole sensors are increasingly being deployed to provide �real-time' reservoir data during hydrocarbon production, which helps to reduce uncertainty in the reservoir description and contributes to reservoir management decisions. Where wells are equipped with inflow control valves (so called �intelligent' wells), it is possible to develop a feedback loop between measurement and control to optimize production. We suggest that measurements of electrokinetic potential during production, using permanently installed downhole electrodes, could be used to detect water encroachment towards an intelligent oil well. Downhole electrodes mounted at the production well on the outside of insulated casing, have been successfully applied in subsurface resistivity surveys during oil production. Similar technology could be used to measure electrokinetic potential. Moreover, recent and ongoing work has changed our understanding of electrokinetic coupling under two-phase conditions. We present the results of numerical simulations of fluid movement during hydrocarbon production, using a new formulation which captures both the changing fluid distributions and the resulting electrical potentials. We suggest that encroaching water causes changes in electrokinetic potential at the production well which could be resolved above background electrical noise; indeed, changes in water saturation could be detected several 10's to 100's of metres away from the well. This contrasts with most other downhole monitoring techniques, which sample only the region immediately adjacent to the wellbore. Signal resolution is improved if the water has a relatively low salinity, and the pressure gradient into the well is large. However, significant uncertainties remain concerning the nature of electrokinetic coupling during the flow of oil and water, particularly in mixed and oil-wet reservoirs.
NS31A-1566
Glaciological challenges for IP, SP, and seismoelectrics in the 21st century
The near-surface geophysical community has so far paid little attention to glaciological problems, although many exciting and timely opportunities and challenges for IP, SP, and seismoelectric research urgently await consideration. I initially review current understanding of the low-frequency electrical properties of ice, firn, snow, and frozen and unfrozen glacial sediments, and present several recent and pertinent case studies demonstrating the usefulness of state-of-the-science IP, SP, and seismoelectric techniques in glaciology. Subsequently I outline future research opportunities and challenges which should exploit more rigorously state-of-the-science IP, SP, and seismoelectric techniques in the field and laboratory, and develop or adapt from other areas of electrical geophysics novel mathematical and statistical concepts for joint data inversion and integration, for glaciological purposes such as e.g. [a] ice core logging; [b] glacier dynamics and fracturing; [c] glacier hydrology; and [d] exploration of life on Mars. I conclude that lower-frequency electrical techniques such as IP, SP, and seismoelectrics have unduly been neglected in glaciology as compared with higher-frequency radar techniques over the past few decades, arguing for a revival of concerted research efforts into these techniques.
NS31A-1567
Induced Polarization of Glass Bead Packs With Varying Mean Grain Size, pH, and Salinities: a Review of Theoretical Models
The spectral induced-polarization (IP) response of rocks and soils is described by a complex conductivity, which depends on pore solution chemistry, texture, and electrochemical properties of the mineral water interface. The complex conductivity response of porous media in the frequency range from the milliHertz to 10 kHertz reveals relaxations in the power spectrum. These relaxation effects are related mainly to surface polarization processes due to the existence of the electrical double or triple layer at the mineral water interface. A deeper knowledge of these physical mechanisms is important for the correct interpretation of field data. We have previously developed an electrical triple layer model to describe the electrochemical properties of clay minerals. The triple layer model comprises the Stern and the diffuse layer and their dependence on chemistry of the pore water. The low frequency dispersion is caused by the influence of field-induced electrolyte concentration variations on the local electric currents in the double layer. We use the Dukhin-Shilov theory combined with a triple layer model to predict the effect of the distribution of grain size (for a silica bead packs), pH, and water composition upon the induced polarization spectrum. Comparison with laboratory data is also provided.
NS31A-1568
The dependence of induced polarization on natural iron concentration in wetland soils
Spectral Induced polarization (SIP) measurements in the frequency range 0.1-1000 Hz were conducted on clay and peat marsh soils, obtained from a contaminated freshwater weltand in New Jersey, that were subsequently analyzed for heavy metal concentrations, moisture content, organic matter, porosity, specific surface area, and pore fluid conductivity. A Cole-Cole relaxation model was fit to each of the samples and the model parameters analyzed in terms of the measured physiochemical properties. A linear relationship between the normalized chargeability (mn) and the estimated surface area to pore volume (Sp) is found when the iron content (ranging from 0.25 to 1.63 percent by volume between samples) is accounted for as a polarizable element of the soil. In fact, the dependence of mn on volumetric Fe concentration per unit volume of the bulk soil is described by a linear relationship with a correlation coefficient of 0.94. As the Fe concentration of soils is a critical biogeochemical parameter, these results suggest that SIP measurements may provide a hitherto unrecognized approach to probing soil geochemistry, iron cycling and anaerobic microbial activity. Furthermore, our results yield new insight into the physiochemical controls on SIP in natural, unconsolidated soils.
NS31A-1569
Comparing Arrays for Induced Polarisation Surveys in a Saline Environment
Conventional arrays for resistivity and induced polarisation (IP) surveys were developed before multi-channel switching systems and modern inversion algorithms allowed more complex and flexible data acquisition geometries. Though some work has been done exploring the optimisation of resistivity arrays (Furman et al. 2003a, 2003b & Stummer et al. 2004), less has been done for IP surveys (White et al. 2003). Using a ten channel recording system, several IP surveys were conducted using Wenner, dipole-dipole and non- conventional arrays with the aim of exploring which array type gave the best data quality, reproducibility and coverage. In addition, a four second injection interval was compared with a two second interval. The surveys were carried out at a legacy oil refinery site that has weathered hydrocarbons near the surface and extremely saline conditions. Preliminary results show that the reciprocal measurements with the Wenner array are more reproducible than those with the dipole-dipole array. As well, the decay curves for the four second injection interval follow theoretical decay curves more closely than those for the two second interval. Optimising survey array design is needed to meet the challenges posed by a growing interest in the use of induced polarisation for environmental site characterisation.
NS31A-1570
Electrical Resistivity and Induced Polarization as Tools for Mapping Near-Surface Weathered Hydrocarbon Bodies
Induced polarization (IP) and electrical resistivity (ER) surveys were conducted at two legacy oil refinery sites in central Alberta during the fall of 2005 and summer of 2006. The first site had areas of near surface weathered hydrocarbons, and the second site had areas of saline soil and groundwater and areas of weathered hydrocarbons. The conductive saline environment of the second site posed significant challenges for the IP investigation. At the both sites, ER imaged weathered hydrocarbon bodies as resistive anomalies above background values. IP inversions yielded high values of chargeability at the boundaries of the resistive hydrocarbon bodies. At the second site, a resistivity high was associated with a change in subsurface lithology. Unlike the anomalies resulting from weathered hydrocarbons, there was no corresponding increase in chargeability associated with the high resistivity due to the lithology change. The results from these two surveys indicate that a combination of ER and IP is an effective tool for both mapping weathered hydrocarbons and in differentiating between resistivity increases associated with near surface hydrocarbons and resistivity increases due to subsurface lithology.
NS31A-1571
Field Studies of the Electrical Properties of Permeable Reactive Barriers for Monitoring Barrier Aging
Permeable reactive barriers (PRB) are a promising technology for the remediation of groundwater containing a range of organic and inorganic contaminants. Although there are number of different types of reactive barriers, some of the most important are constructed from granular zero valent iron (ZVI). One challenge in the large- scale, long-term implementation of PRBs is to monitor the change in barrier properties over time. For example, mineral precipitates can reduce the effectiveness of the barrier by either insulating the reaction surfaces of the ZVI particles and/or by filling the pore space in the barrier and thus reducing its hydraulic permeability. Previous research has shown that resistivity and induced polarization (IP) measurements are sensitive to corrosion and precipitation due to redox reactions between ions in solution and the ZVI mineral surface. New field studies, supported by additional laboratory studies appear to confirm this work. Resisitivity and IP surveys were conducted at a total of seven barriers at four different sites: the Denver Federal Center; Monticello, Utah; the Kansas City, Missouri Department of Energy site, and the Asarco Smelter Site in East Helena, Montana. These surveys used combinations of surface and borehole surveys to characterized barriers. The surveys are repeated at approximately six-month intervals to provide information on temporal changes. In addition, surveys at the Kansas City barrier followed up on earlier research by providing several years of historical data and a new barrier at East Helena Montana has been instrumented with an autonomous monitoring system allowing continuous monitoring of the barrier electrical properties. Results show an increase in both real and imaginary conductivity as barriers age. For new barriers, the conductivity of ZVI is typically a few tens of mS/m, only modestly higher than that of the background sediments surrounding the barrier. For heavily altered barriers such as the Monticello, Utah barrier, the conductivity is typically tens of S/m, a thousand times higher the unaltered barriers. Field values of chargeability (measured using a 1 Hz primary waveform and an integration window centered at 40 ms) also tend to increase from roughly a 100 mV/V at the East Helena Barrier to about 300 mV/V at Monticello. Other sites tend to be intermediate between these extremes.