250 63 295KB
English Pages 32 Year 2009
International Workshop on Induced Polarization in Near-Surface Geophysics
Bonn, Germany September 30 – October 1, 2009
On behalf of the Induced Polarization working group of the German Geophysical Society (DGG), the Department of Geodynamics/Geophysics of the Steinmann Institute, University of Bonn, holds the
International Workshop on Induced Polarization in Near-Surface Geophysics
Organization: Andreas Kemna, University of Bonn (Spokesman of the DGG IP working group) Adrián Flores-Orozco, University of Bonn
Co-sponsored by Transregional Collaborative Research Centre 32 “Patterns in soil-vegetation-atmosphere systems: monitoring, modelling and data assimilation” Zonge Engineering and Research Organization, Inc.
Program 08:30 - 09:00
Welcome and overview
09:00 - 09:45 09:45 - 10:30
Keynote talk Andrew Binley On the value of induced polarisation in hydrogeophysics Keynote talk André Revil Induced polarization: A mechanistic approach and its use to infer permeability
10:30 - 11:00
Coffee break
11:00 - 11:45 11:45 - 12:30
Keynote talk Lee Slater The spectre of field scale ‘spectral’ induced polarization? Keynote talk Ken Williams Getting a grip on electrode polarization: Insights from a biogeochemical perspective
12:30 - 14:00
Lunch, Coffee
14:00 - 14:15 14:15 - 17:00
Poster preparation Poster session
17:00 - 17:30
Definition of working groups for next morning
17:30 Food and drinks (sponsored by Zonge)
Thursday, October 1 09:00 - 10:30 10:30 - 11:00 11:00 - 12:30
Discussion in working groups Coffee break Discussion in working groups (continued)
12:30 - 14:00
Lunch, Coffee
14:00 - 16:00 16:00 - 16:30 16:30 - 17:30
Group reports, overall discussion Coffee break Conclusions
Participants Aktarakçi, Hasan
Advanced Geosciences Europe, S.L. Attwa, Mohamed LIAG, Hannover August, Anastasia Bonn Univ. Bastani, Mehrdad Uppsala Univ. Bergers, Rainer Cologne Univ. Binley, Andrew Lancaster Univ. Blascheck, Roland RWTH Aachen Börner, Frank DGFZ, Dresden Breede, Katrin FZJ, Jülich Carlson, Norman Zonge, Tucson Caterina, David Liège Univ. Rossi, Matteo Padua Univ. Dingoko, Odirile Council for Geoscience Drenth, Eize Oranjewoud El-Kaliouby, Hesham Lancaster Univ. Figula, Joana PBG Geophysical Exploration Company Flores-Orozco, Adrian Bonn Univ. Greenhalgh, Mark Aarhus Univ. Grissemann, Christoph Bundesanstalt für Geowissenschaften und Rohstoffe
[email protected]
Günther, Thomas Haaken, Klaus Haegel, Franz-Hubert Holland, Raphael Huisman, Sander Katayama, Hiroyuki
[email protected] [email protected] [email protected] [email protected] [email protected] [email protected]
Kavian, Mohsen Kemna, Andreas Klitzsch, Norbert Koch, Kristof Lagmanson, Markus Legaz, Aurelie Leroux, Virginie Lück, Erika Maj, Elzbieta
LIAG, Hannover Bonn Univ. FZJ, Jülich LIAG, Hannover FZJ, Jülich Japan Oil, Gas and Metals National Corporation TU Delft Bonn Univ. RWTH Aachen Lausanne Univ. Advanced Geosciences Aarhus Univ. Lund University Potsdam Univ. PBG Geophysical Exploration Company
[email protected] [email protected] [email protected] [email protected] [email protected] [email protected] [email protected] [email protected] [email protected] [email protected] [email protected] [email protected] [email protected] [email protected] [email protected] [email protected] [email protected] [email protected]
[email protected] [email protected] [email protected] [email protected] [email protected] [email protected] [email protected] [email protected] [email protected]
Martin, Roland Martin, Tina Münch, Martin Nguyen, Frederic Niederleithinger, Ernst Nover, Georg Oberdörster, Christoph Okay, Gonca Radic, Tino Revil, André Sada, Marek Sarma, V.S. Savvaidis, A. Schütze, Claudia Sito, Lucasz Slater, Lee Spangenberg, Ute Tanguy, Robert van Schoor, Michael Volkmann, Jan Weigand, Maximilian Williams, Ken Wojdyla, Marek Zadorozhnaya, Valeriya Zimmernmann, Egon Zisser, Norbert
Bonn Univ. BAM, Berlin Bonn Univ. Liège Univ. BAM, Berlin Bonn Univ. Bonn Univ. Univ. Pierre et Marie Curie, Paris Radic Instruments Colorado School of Mines PBG Geophysical Exploration Company NGRI, Hyderabad Uppsala Univ. Leipzig Univ. PBG Geophysical Exploration Company Rutgers Univ. Postdam Univ. Liège Univ. CSIR RWTH Aachen Bonn Univ. LBNL, Berkeley PBG Geophysical Exploration Company Council for Geoscience FZJ, Jülich Bonn Univ.
[email protected] [email protected] [email protected] [email protected] [email protected] [email protected] [email protected] [email protected] [email protected] [email protected] [email protected] [email protected] [email protected] [email protected] [email protected] [email protected] [email protected] [email protected] [email protected] [email protected] [email protected] [email protected] [email protected] [email protected] [email protected] [email protected]
Contributions Keynote talks Binley, Andrew Revil, André Slater, Lee Williams, Ken Posters Attwa, Mohamed Börner, Frank Breede, Katrin Carlson, Norman Flores-Orozco, Adrian Haegel, Franz-Hubert Kamenetsky, Felix Kavian, Mohsen Klitzsch, Norbert Koch, Kristof
On the value of induced polarisation in hydrogeophysics Induced polarization: A mechanistic approach and its use to infer permeability The spectre of field scale ‘spectral’ induced polarization? Getting a grip on electrode polarization
SIP field measurements for aquifer characterization at the LIAG testsite Schillerslage SIP for soil and groundwater protection Effects of saturation on spectral induced polarization of sand-clay mixtures Changes in IP effects over time resulting from accelerated degradation processes at a landfill Data rror quantification for SIP Imaging the distribution of moisture by electrical impedance tomography during the evaporation of water from sand IP and SPM effects in equations of electromagnetics Measured low-frequency effects of pore fluid salinity on complex electrical parameters of unconsolidated layered sand and sand-clay structures Influence of pore structure on SIP properties - microscale modelling and measurements Controlled changes in grain size and pore characteristics of proxies of alluvial sediments and their impact on the hydraulic conductivity and spectral induced polarization response
Leroux, Virginie
Measuring techniques in time-domain IP
Lück, Erika Martin, Tina
About the data quality of the rolling system 'GEOPHILUS ELECTRICUS' Complex resistivity on wood
Milde, Stefanie Niederleithinger, Ernst
Studies on the origin of the IP effect with gel-filled sandstone samples Multiparameter investigations on wood samples
Okay, Gonca
Structural characterization of the disturbed zone in the argillaceous Tournemire site with induced polarization method Boundary estimation between dissemination and massivity in minerals using physical model studies in induced polarisation (IP) Delay time vs depth of detection in induced polarisation (IP) - model and field studies Spectral induced polarisation (SIP) - a significant geophysical tool for mineral detection and discrimination Application of membrane polarization for studying internal structure of samples Mathematical modeling of membrane polarization caused by constrictivity of pores Evaluating the EIT96 measurement system using defined samples with known electrical properties Relationship between low-frequency electrical properties and hydraulic permeability of low-permeable sandstones
Sarma, V.S. Sarma, V.S. Sarma, V.S. Zadorozhnaya, Valeriya Zadorozhnaya, Valeriya Zimmernmann, Egon Zisser, Norbert
On the value of induced polarisation in hydrogeophysics Andrew Binley Lancaster University, UK
For the past two decades, or more, several researchers have highlighted the potential value of induced polarisation (IP) (and spectral induced polarisation - SIP) for the characterisation of hydrogeological properties. Since measured IP responses are, in some way, affected by properties of the pore-grain interface , it is intuitive to assume that fundamental properties controlling flow of fluids will be related to measured IP/SIP properties. Several experimental studies have revealed such links and confirmed the potential of IP and SIP in hydrogeophysics. Links between polarisation and surface area, electrical relaxation and pore throat size (or permeability), although still empirical, are included in such studies and show great promise. There is, however, evidence that other hydrological state variables influence IP and SIP properties: fluid saturation, fluid conductivity, fluid temperature, for example, all appear to influence polarisation of subsurface materials. Through a review of some experimental studies of IP/SIP - hydrogeological relationships we reanalyse the potential value of IP and SIP in hydrogeophysics. We also include an assessment of the translation of field measurements of IP and SIP measurements to images of hydrogeologic properties, since this remains the ultimate goal of several studies. This is not only constrained by the potential ambiguity of IP/SIP - hydrogeological relationships and their lack of universality but potentially seriously hampered by the challenges of inverting SIP data (for example, model non-uniqueness). By analysing the strengths and limitations of IP/SIP in hydrogeophysics we can highlight key areas of potential future research in this field.
Induced polarization: A mechanistic approach and its use to infer permeability André Revil Colorado School of mines
I will show how we can build a mechanistic theory to understand induced polarization in the frequency range 1 mHz to 1 MHz including various contributions to polarization (mainly the polarization of the Stern layer and the Maxwell Wagner polarization). In this modeling, I will show that surface conductivity of porous rocks has two contributions: the first is associated with the diffuse layer coating the grains and is frequency independent. The second contribution is associated with the Stern layer of weakly sorbed counterions on the mineral surface and is frequency dependent. In the frequency range 1 mHz-100 Hz, this second contribution is also associated with the main polarization mechanism observed by the spectral induced polarization method in granular media around few Hertz. At the macroscale, this Stern layer contribution can be related to the grain size distribution. Because surface conductivity at the macroscopic scale can be connected to an effective pore size that can be used to characterize permeability, a new relationship can be derived between this effective pore size, the formation factor, and the grain size distribution. Permeability, in turn, can be related to this effective pore size and the formation factor. This offers therefore a new relationship between the permeability, the formation factor, and the distribution of the relaxation times that can be inferred from induced polarization in the frequency domain. This new relationship is consistent with various formula derived in the literature in the limit where the grain size distribution is given by the delta function and agree fairly well with various experimental data. It helps also to provide an a theoretical foundation to the Cole-Cole model widely used to represent the behavior of the complex conductivity of granular media. This model seems to explain most (if not all) the literature data to date made with sands, and clean to clayey sandstones, clay-rocks and supensions of clays, and offer a universal way to estimate permeability under the assumption that the Maxwell-Wagner polarization and the polarization of the Stern layer can be separated.
The Spectre of Field Scale ‘Spectral’ Induced Polarization?
Lee Slater Department of Earth & Environmental Sciences, Rutgers-Newark, USA
The induced polarization (IP) geophysical method has recently reemerged as a promising technology for characterizing and monitoring the shallow earth. The growing interest in the technique reflects the sensitivity of the technique to interfacial properties and processes that are not readily captured with the more traditional geophysical methods of investigation. Whereas most geophysical methods (e.g. resistivity, seismic and ground penetrating radar) primarily sense changes in the volumetric properties of the earth, the induced polarization response is driven by the properties of interfaces in porous media. These interfaces exert a strong control on the hydraulic properties of soils and rocks, and are the location where biogeochemical transformations are concentrated. A number of datasets have highlighted the value of field-scale induced polarization measurements for mapping lithologic variability, estimating distributions of hydraulic conductivity and monitoring biogeochemical transformations. The interpretation of these field-scale datasets has been supported by robust empirical relationships between polarization magnitude, readily measured with time and frequency domain instruments, and pore geometric parameters such as the surface area normalized by pore volume (Spor). In contrast, although many laboratory studies now increasingly focus on extracting pore or grain size estimates from spectral induced polarization (SIP) datasets, successful field-scale applications of SIP remain practically non-existent, primarily due to the apparently insurmountable coupling problem at the high end of the spectrum and time at the low end. Using recent examples of laboratory and field-scale datasets, I weigh up the costs and benefits of field-scale ‘spectral induced polarization’ versus ‘induced polarization’ measurements. Using recent datasets, I show how field-scale IP can be used to visualize the hydrogeologic framework of a hyporheic corridor from rapidly acquired waterborne IP datasets collected along 3 km lines. In contrast I describe recent data from a low-noise field site that highlights the challenges of SIP interpretation even with meticulous, time-intensive data acquisition. I also present recent laboratory datasets demonstrating how the shape of SIP specta, from which pore or grain size distributions are predicted, is inherently dependent on fluid chemistry in coarse grained soils. From these considerations, I suggest that the ‘spectral’ part of induced polarization will remain beyond the reach of field scale relevance for many years to come.
Getting a grip on electrode polarization: Insights from a biogeochemical perspective Kenneth H. Williams Lawrence Berkeley National Laboratory
The use of geophysical approaches for characterizing and monitoring stimulated microbial processes has seen increased interest over the past several years spawning the sub-discipline of "biogeophysics." Of these approaches, the spectral induced polarization (SIP) technique has shown great promise as a means for delineating subsurface microbial activity at both laboratory and field scales. While laboratory SIP studies have attempted to resolve a wide variety of processes, field SIP studies to date have focussed on imaging subsurface microbial processes related to biomineralization (i.e. mineral precipitation induced by biological activity) and the accumulation of electroactive end products. This talk will present the results of SIP field studies conducted at the U.S. Department of Energy's Rifle, Colorado field study site, with an emphasis on biogeochemical processes accompanying microbial iron and sulfate reduction. The iterative and/or concomitant stimulation of both processes through organic carbon addition induces the precipitation of conductive mineral phases, such as FeS, and the accumulation of electroactive ions, such as Fe(II) and hydrogen sulfide. As will be shown, the accumulation of both mineral and aqueous species is capable of generating significant SIP phase anomalies (10 to 60 mrad) readily detectable with commercially available instrumentation. Careful experimental design appears capable of yielding high quality field SIP datasets over a wide range of frequencies (0.06 to 256 Hz), with such large bandwidth likely critical for discriminating the predominant metabolic process. The field spectral results have been corroborated by laboratory studies where constraints were placed upon the timing and location of the biogeochemical processes in order to isolate the mechanisms underlying the SIP response. The source generating mechanisms will be discussed within the context of electrochemical phenomena, with an emphasis on heterogeneous charge transfer processes mediated by a variety of semi-conductive mineral phases and reduced aqueous compounds. In short, a new look at an old mechanism - electrode polarization - will be undertaken, with biogeochemical processes providing the central dogmatic theme.
SIP field measurements for aquifer characterization at the LIAG test site, Schillerslage M. Attwa, R. Holland and T. Günther Leibniz Institute for Applied Geophysics, LIAG
Spectral induced polarization surveys, in the form of 2D, can provide additional information or even more detailed information about the lithological characterization of subsurface layers than resistivity measurements. In the new LIAG test site Schillerslage, five 2D SIP profiles were measured to image and characterize the near surface geology, verified by borehole data. The structure of quaternary sediments, overlaying cretaceous marls and clays, is imaged by SIP data: Two sandy aquifers are separated by a fine grained, 3m thick till layer in a depth of about 13m under surface, but varying in depth und thickness. While differences in water saturation as well as the cretaceous sediments are imaged by resistivity data, the till layer and additionally thin peat layers in the upper aquifer are detected by phase shifting. The higher resistivity magnitude and lower phase values of the upper aquifer reflect that it consists of coarse sand and gravel, while the lower aquifer is mainly of medium and carbonate bearing sand. The results are to be verified by further drilling, pumping tests, hydraulic tomography and laboratory SIP measurements especially to assess hydraulic properties.
SIP for soil and groundwater protection F. Börner, B. Weihnacht Dresden Groundwater Research Centre, Meraner Str. 10, 01217 Dresden, Germany
A detailed knowledge of geohydraulic properties of soil zone and aquifers is necessary to enable a sustainable use of groundwater ressources. Against this background we studied the changes of geophysical parameters on a large scale experimental facility and on a 2 m high soil wall of a test site. The unsaturated materials investigated range from coarse-grain gravel to sandy loam. Spectral induced polarization as well as ground-penetrating radar and ultrasound transmission were applied as geophysical methods. The measured parameters were used to calculate soil parameters such as porosity, water content, density and grain surface area necessary to obtain geohydraulic parameters such as hydraulic conductivity, field capacity and retention parameters. Soil samples were taken and analysed regarding porosity, apparent density, true density and internal surface. The objective of the multiphase displacement experiments made in this study was the determination of residual water saturation and residual gas saturation in heterogeneous sandstones. In order to obtain representative parameters, large scale samples have been used. During the water imbibition process the water composition and hence the water salinity changes in space and time. In that case the water saturation determination is still possible using SIP or a combination of geophysical methods. The comparison between petrophysical data from the laboratory and from geophysical measurements showed good correlations for the majority of the data.
Effects of saturation on spectral induced polarization of sand-clay mixtures K. Breede1, O. Esser1, E. Zimmermann1, J.A. Huisman1, A. Kemna2 2
1 Forschungszentrum Jülich, ICG-4-Agrosphere University of Bonn, Dept. of Geodynamics and Geophysics
Groundwater is a vulnerable resource that is endangered by pollutants and contaminants. Soil is an important protective buffer for groundwater and, therefore, the understanding of flow and transport processes in soils is very important. However, the prediction capabilities of unsaturated flow and transport models in the vadose zone are often limited due to an insufficient knowledge about the structural and textural heterogeneity of the soil. To obtain more information about soil structure, texture and heterogeneity, as well as hydraulic parameters, non-invasive electrical methods may be employed in field-scale studies. To investigate the potential of the approach, a laboratory measurement setup was developed which allows combined electrical and hydraulic measurements. The latter are conducted via a multi-step outflow device. Various pressure steps are applied to a saturated sample and the outflow is recorded. When equilibrium is reached, spectral induced polarization (SIP) measurements are conducted before the next pressure step is applied. The electrical measurements are carried out with a high-accuracy impedance spectrometer. Combined electrical and hydraulic measurements were conducted on packed sand-clay mixtures. The measured resistivity magnitude and phase spectra and their dependence on water content are clearly different for each mixture. For pure sand, the phase values increase with decreasing water content over the entire frequency range and a phase peak is formed for higher pressure steps (i.e., lower water content). The increasing phase is due to the increasing resistivity and an associated increasing chargeability of the sample. The phase spectrum of a sand-clay mixture with 5 % clay shows the same behavior like the pure sand; however, the shift of the phase peak to higher frequencies is much stronger. This shift suggests that relaxation time and length become smaller with decreasing water content and thus that smaller pores are active at lower saturation states. The sand-clay mixture with 20 % clay shows a significantly different behavior. At full saturation, the phase spectrum exhibits a weak peak at about 0.2 Hz. With decreasing saturation, first a distinct phase peak is formed at about 0.02 Hz; then phase values decrease, and simultaneously a shift to higher frequencies occurs. The phase values are also much smaller than for the other mixtures. The different behavior of the sand-clay mixture with 20 % clay could result from the higher clay content. Overall polarizability does not increase steadily with clay mineral concentration, but reaches a maximum and then decreases again. It seems that the latter occurs for the 20 % sand-clay mixture, where phase values are smaller than for the 5 % sand-clay mixture. In the near future the conducted measurements will be used to investigate relationships between electrical and hydraulic properties.
Changes in IP effects over time resulting from accelerated degradation processes at a landfill N. R. Carlson, C. M. Mayerle Zonge Engineering & Research Organization, Inc.
In 1999, 95 lines of IP data were used to provide a detailed map of the Rio Nuevo South landfill in Tucson, Arizona, USA as part of a downtown re-vitalization project for the City of Tucson, and the survey results were then verified with 14 boreholes. In preparation for new construction, an enhanced biodegradation pilot project was tested on a small part of landfill (15m X 15m), and follow-up IP data suggested that over time, the enhanced biodegradation reduced the measured IP effect. After the successful pilot project, which increased the speed of the waste decay by up to 100 times in the dry, desert conditions, the degradation program was expanded to include the majority of the landfill beginning in 2004. In 2009, after five years of enhanced degradation, we repeated two lines of IP from the 1999 survey, using the same survey parameters and same type of equipment, in order to verify the change in the IP effect that we had seen in the pilot project test. Although resistivity values decreased on one line but showed an increase on the second line, both lines showed a significant decrease in the IP data by as much as 50% in some sections of the lines. The survey results indicate that IP data may be an extremely valuable tool as a non-intrusive monitoring method when landfill degradation is being accelerated by enhanced decay processes.
Measuring techniques in time-domain IP Torleif Dahlin and Virginie Leroux Engineering Geology, LTH
Measuring induced polarisation in time-domain is an interesting approach in practice since it can be done with only slightly modified resistivity equipment. The measurement procedure is effective and yields good and useful results in many cases. 2D and 3D surveying is possible to a reasonable cost and resistivity data at least is almost always obtained. However, even if it has been shown that stainless steel electrodes can be used instead of non-polarisable electrodes, still problems arise when the grounding resistance is high and on very conductive grounds. The noise level can then become sufficiently high to overwhelm the relatively small induced potentials. In the first case, it is necessary to separate the current and potential cables. Examples are shown of the improvement of data quality with that procedure. On very conductive grounds inductive coupling arises in the earlier part of the decay curve. Although it bears information in itself, its separation from the induced polarisation response is not straightforward. Decay curves have to be evaluated before proceeding to standard inversion. Negative decay is not to be rejected indiscriminately, since it can occur as a result of the geometry and of the sensitivity pattern of the particular array used, as shown by example. In any case, recording the full decay curve with high resolution instruments is essential for future developments, and preferably the full waveform should be recorded. It constitutes the first step towards a possible extraction of part of the spectral information from the timedomain induced polarisation. Could this be achieved, it would open up for more detailed interpretation which could prove very useful in near-surface applications.
Data error quantification in SIP imaging Adrian Flores-Orozco1, Andreas Kemna1, Egon Zimmermann2 1
Department of Geodynamics and Geophysics, University of Bonn Institute for Electronics (ZEL), Forschungszentrum Jülich GmbH
2
Although the use of induced polarization (IP) imaging in near-surface geophysical studies is increasing, particularly for hydrogeological and environmental applications, the issue of IP data error analysis prior to inversion has not yet received much attention. Electrical impedance (EI) data sets were collected on a 2-D experimental tank for different scenarios under controlled conditions, using a new measurement system with high phase accuracy. Data errors were evaluated in terms of discrepancy between normal and reciprocal measurements, and the dependency of resistance and phase errors on resistance. In addition, the phase error on the absolute phase value was investigated. In agreement with previous studies, the normal-reciprocal resistance error increased linearly with resistance. The phase error decreased exponentially with increasing resistance, i.e., with increasing signal strength. This behavior can be described with a power law relationship between the resistances and the error in the phase measured. Based on these findings, a model was proposed to describe phase errors in EI data sets. With the purpose of validation, the proposed analysis and error quantification has been performed for data sets collected at the field scale. The data were collected with different measuring devices for diverse applications and under different conditions. However, the analysis of the raw data show consistency with the tank-data, and a power law relationship is given between phase errors and resistances measured. The proposed model was implemented in a complex resistivity inversion scheme, and inversions were performed with and without the new error description. An improved image quality was obtained with the new error model, demonstrating the effectiveness of this new approach.
Imaging the Distribution of Moisture by Electrical Impedance Tomography during the Evaporation of Water from Sand
Franz-Hubert Haegel1, Egon Zimmermann2, Harry Vereecken1 1
Forschungszentrum Jülich, 52425 Jülich, Germany Institut für Chemie und Dynamik der Geosphäre, ICG-4 Agrosphäre 2 Zentralinstitut für Elektronik
Among the different processes involved in the distribution of water in soil, evaporation exhibits the most complex behaviour and is far from being understood quantitatively in detail. Therefore electrical impedance tomography (EIT) with 20 current injection and 48 voltage electrodes was tested here as an additional method to monitor the evaporation of tap water from sand at 20 °C in a long-term laboratory experiment. The results obtained by EIT show that the evaporation rate was high during the first days, but then slowed dramatically down in accordance with findings of other groups. The real part σ’ of the complex conductivity shows an increase in the centre of the container which is due to an increase of electrolyte concentration and wall effects. The imaginary part of conductivity σ’’ yields additional information on the water saturation in the container. Although the signals are highly disturbed by electrode effects due to the loss of contact during advancing desaturation, distinct patterns of σ’’ which can be qualitatively explained are found at different locations during the experiment.
Measured low-frequency effects of pore fluid salinity on complex electrical parameters of unconsolidated layered sand and sand-clay structures
M. Kavian1, E.C. Slob1, W.A. Mulder2 1
2
Delft University of Technology Delft University of Technology and Shell International E&P
We conducted a series of laboratory measurements of the electrical properties of samples of saline water-saturated unconsolidated sand and sand-clay layered structures as a function of frequency, water saturation, and salinity. The ultimate objective is to determine if the effect of heterogeneities at scales much smaller than the skin depth can be captured by introducing an effective frequency-dependent, possibly anisotropic electric permittivity and resistivity, whose behaviour can be described by simple functions. We employ the parallel-plate capacitor technique to measure the complex impedance for frequencies between 210 Hz and 3 MHz. We performed main drainage and secondary imbibition cycles at atmospheric pressure and temperatures between 21°C and 22°C. We found hysteresis in the electric permittivity and conductivity, caused by the redistribution of the water and air phases. The hysteretic effect becomes more pronounced at higher concentrations of salt. We also found that the sand grain size does not affect the permittivity of dry sand. For the saturated sand, the situation changes considerably and the coarser grain size leads to a larger polarization effect. An explanation of this phenomenon is that for the coarse-grain size sand, the capillary pressure is lower than for the fine-grain size sand, allowing the water molecules to respond more easily to the external electric field. For a fixed concentration of salt and by increasing the frequency, the electric permittivity of the sand structure decreases faster than for the sandclay structure. In this case, the drainage and imbibitions curves for sand structures lie above the related curves of sand-clay structure only for the lower frequencies. For a fixed low frequency, the measured effective permittivity increases with increasing salinity for both the sand and sand-clay structures and the hysteretic curves for the sand structure always lie above those for the sand-clay structure. The situation reverses when higher frequencies are used. Then, the hysteretic curves for the sand-clay structure lie above those for the sand structure.
Influence of pore structure on SIP Properties – Microscale Modelling and Measurements
Norbert Klitzsch, Jan Volkmann, Eugen Wiens, Oliver Mohnke Applied Geophysics and Geothermal Energy, E.ON Energy Research Center, RWTH Aachen University, Germany
In the last two decades many empirical relations were proposed which relate the frequency dependent electrical properties of water saturated rocks to structural properties such as pore radius, and inner surface area, or to hydraulic conductivity. Unfortunately, these relations are not universal; they apply only for particular rock types and water compositions. In order to quantify the influence of inner rock structure (as well as of electrochemical water and rock properties) on the SIP (spectral induced polarisation) properties we model the charge transport processes at the pore space (and calculate the SIP properties therefrom). We have applied the membrane polarisation model as proposed by Marshall and Madden (1959) to simulate the SIP properties of 3-D pore models. We extended the model by accounting for the electrical double layer EDL as origin of the IP effect. The EDL is implemented as boundary effect by reduced counterion mobilities close to the inner surface. To validate the modelling results we compare them with SIP measurements on artificial samples (porous silica glasses). The samples are characterized by narrow pore radius distributions and well defined surface properties. SIP measurements were conducted in a frequency range from 1 mHz to 1 MHz using four and two electrode configurations for low and high frequencies, respectively. Furthermore, X-ray computer tomography (CT), NMR saturation recovery (T1), and NMR CPMG (T2) measurements have been carried out on these samples to characterise the inner pore structure. Reference Marshall, D.J. und Madden, T.K. 1959. Induced polarization, a study of its causes. Geophysics 24 (4): 790–816.
Controlled changes in grain size and pore characteristics of proxies of alluvial sediments and their impact on the hydraulic conductivity and spectral induced polarization response K. Koch1, A. Kemna2, and K. Holliger1 1
Institute of Geophysics, University of Lausanne, Switzerland Department of Geodynamics and Geophysics, University of Bonn, Germany
2
Understanding the influence of pore space characteristics on the hydraulic conductivity and spectral induced polarization (SIP) response is critical for establishing relationships between the electrical and hydrological properties of surficial sedimentary deposits. Here, we present the results of laboratory SIP measurements on saturated quartz samples with granulometric characteristics ranging from fine sand to fine gravel. We alter the pore characteristics using three principal methods: (i) variation of the grain sizes, (ii) changing the degree of compaction, and (iii) changing the level of sorting. We then examine how these changes affect the SIP response, the hydraulic conductivity and specific surface area of the sand samples. It was shown that controlling factors like specific surface area or pore size provides good results when predicting hydraulic conductivity of comparable sample types, but the interdependencies remain not sufficiently understood in order to allow for a more wideranging correlation of electric and hydraulic properties. In general, the results indicate a clear connection between the applied changes in pore characteristics and the SIP response. In particular, we observe a systematic correlation between the hydraulic conductivity and the relaxation time of the Cole-Cole model describing the observed SIP effect for the whole range of considered grain sizes. The results do, however, also indicate that the detailed nature of these relations depends strongly on variations on the pore characteristics, such as for example the degree of compaction. The results of this study underline the complexity of the origin of the SIP signal and, as a consequence, the difficulty to relate it to a single structural factor of a studied sample.
About the data quality of the rolling system ‚GEOPHILUS ELECTRICUS‘
E. Lueck1, U. Spangenberg1 J. Ruehlmann2 1 University of Potsdam Institute of Vegetable and Ornamental Crops, Großbeeren
2
Geoelectrical soil mapping has become widely accepted to image the spatial heterogeneity of soil properties. Much work has been done to test the applicability of existing geoelectrical methods and to develop measurement systems applicable for large areas. Within this framework, the soil sensor GEOPHILUS ELECTRICUS was developed. It uses rolling electrodes (constructed by J. Bigus and I. Hausschild – IGZ Grossbeeren) in combination with a SIP-instrument developed by T. Radic (Radic research, Berlin). The system is capable of measuring complex conductivity (amplitude and phase shift) in a frequency range between 1 mHz und 1 kHz. Four frequencies and five channels can be measured simultaneously. Different experiments were done to test and to improve the data quality. To evaluate the features different components have to be considered: -
the accuracy of the SIP-instrument the electrodes the soil conditions like moisture, roughness and compactness the quality of the positioning system
As expected, conductivity data are much more stable than the measurements of the phase angle and the near surface information (channel 1) have a lower noise level than the deeper channels. Instrumental limitations (Imax = 250 mA and input voltage < 2V) restrict the multichannel application for high conductive soils (about 20 Ohm m) as well as for high resistive soils (greater than 1000 Ohm m). In many cases, the conductivity maps were quite similar for all frequencies whereas the quality of phase shift data showed a strongly frequency dependency. Good results were obtained using a signal frequency of about 200 Hz. This corresponds well with frequencies used by other rolling systems. The signal frequency of the ARP system (Geocarta, France) is 220 Hz and VERIS 3100 (Veris technologies, USA) uses 150 Hz.
Complex Resistivity on Wood Tina Martin Federal Institute for Materials Research and Testing, Division VIII.2 Non-destructive Damage Assessment and Environmental Measurement Methods, Unter den Eichen 87, 12205 Berlin, Germany
Complex Resistivity (CR) can not be used only to investigate geological materials but also for detection of damage in wood and standing trees. CR was used here to study the effect of anisotropy, water saturation level and fungi-infection on wood samples. Our laboratory results indicated that the strong anisotropy of wood affects not only the resistivity but also the phase. The water saturation level has also a big influence on the complex resistivity. With increasing water saturation level (S) the resistivity decrease. The phase and the imaginary part of the conductivity first increase and then decrease with increasing S. Many experiments with fungi-infected wood demonstrated that the fungi has a very big influence on the CR. At the onset of the fungi-infection, the resistivity decreases because the fungi behaves like a water-saturated tissue. Later on, when the fungi has destroyed every usable material, the resistivity increase very rapidly. Also, the phase is affected by the fungi. With increasing fungi-induced wood decomposition the phase decreases significantly. In other words, while the resistivity is highly dependent on the water content, the phase is very sensitive to any changes in the wood cell structures. The results from the laboratory measurments were portable to tomographic field measurements on standing trees. The CR method was first applied on healthy trees to study the general electrical behaviour. It could be seen that, the CR is strongly influenced by the season and the age of the tree. Reasonably good results were obtained for frequencies less than 0.1 Hz. Fungi-infected trees were also investigated. It was possible to detect the fungiinfected zone within the tree, but mainly in the resistivity measurements. The resistivity showed a decrease in infected areas but the phase changed very subtle.The reason for that was probably the suboptimal electrode coupling.
Studies on the origin of the IP-effect with gel-filled sandstone samples
Stefanie Milde and Andeas Hördt Institute of Geophysics and Extraterrestrial Physics, TU Braunschweig
The origin of the induced polarization (IP) in the pore space of sediments is not completely understood. Existing theories may be separated into two groups. In the first, only an electrical double layer around the sediment grains is required, with the grain size as one important geometrical parameter. The second class of theories is based on a coupling between pores of different sizes and ion mobilities to produce a frequency-dependent conductivity. We have carried out complex electrical conductivity measurements on gel-filled sandstone samples with different gel concentrations and fluid salinities. The idea is to reduce the ion mobility in the pore space, allowing to test hypotheses resulting from the different theories. The conductivity spectra from the gel-filled sandstone samples are distinctly different from water-filled samples. The phase shifts decrease, the spectrum is flattened, and the maximum moves towards higher frequencies. In terms of Cole-Cole parameters, the gel decreases the chargeabilities and decay times. We conclude that the geometry of the pore space, and the mobilities in the large pores, are important factors for the generation of the IP-effect. The reduction of the chargeabilities by a reduction of mobilities is consistent with one particular quantitative theory based on the pore space geometry.
Multiparameter investigations on wood samples
Ernst Niederleithinger and Tina Martin Federal Institute for Materials Research and Testing, Division VIII.2 Non-destructive Damage Assessment and Environmental Measurement Methods, Unter den Eichen 87, 12205 Berlin, Germany
To evaluate the stability of standing trees a new instrument using a bending method is under development at a private company. The work is accompanied by an extensive laboratory programby BAM for calibration and validation not only of the bending device, but also for electrical and acoustical tomography. Sound oak, poplar, beech and linden trees are under investigation. They are checked by bending and tomography and then cut down. Samples with a cross-section of 2 by 2 cm² are taken from various heights, sides and depths of the trunk. Moisture, density, E-modulus (bending and compression) and strength are determined. In addition, p-wave velocity as well as complex resistivity (1.2 mHz – 45 kHz) are measured. Main purposes is a) to validate the acoustic and resistivity tomographic images taken sat the standing trees and b) to evaluate correlations between the physical and mechanical parameters. First results show a reasonable correlation between tomographic images and lab investigations, even if the well known anisotropy is not accounted for. The velocity of elastic waves correlates quite well with the E-modulus for specific tree species. As previous investigations have shown, the resistivity values should primarily correlate with moisture content and decay. But as the moisture variation is narrow and most samples are sound, the results have so far limited significance. Nevertheless, species and position specific variations can be seen.
Structural Characterization of the disturbed zone in the Tournemire site with induced polarization method
Okay, G.(1),(2), Ghorbani A.(3), Cosenza P.(4), Cabrera J.(2), Camerlynck C.(1), Florsch N.(5), Revil A.(6),(7) (1) UMR 7619 Sisyphe, Paris 6 University, Paris, France (2) Institute of Radioprotection and Nuclear Safety, Fontenay aux Roses, France (3) Department of Mining and metallurgical engineering, Yazd University, Yazd, Iran (4) HydrASA, Poitiers University, Poitiers, France (5) Geodeş Intitute of Research and Development, Cap Town, France (6) Colorado School of Mines, Department of Geophysics, Golden, CO 80401, USA (7) Laboratoire de Géophysique Interne et Tectophysique, CNRS, Savoie University, Le Bourget-du-Lac, France
Radioprotection and Nuclear Safety Institute (IRSN) selected experimental Tournemire site (Aveyron, France) in order to study the confinement properties of the argillaceous media and the properties of the Excavation Disturbed Zone (EDZ). In this site, in February and June 2009, electrical tomography (2D) measurements have been performed at the eastern (excavated in 1996) and the northern gallery (recently excavated in 2008) to study and compare properties of the disturbed zone around galleries. A longitudinal profile at the ground, including 48 electrodes, in a separation of 20 cm, was chosen at each gallery. Microdrillings are made to install electrodes. Wet bentonite was injected into these micro-drillings to ensure electrical contact between the argillite and electrodes. Non-polarizing Cu/CuSO4 electrodes were used during injection and potential measurements. Measurements were carried out with a Syscal Pro device (IRIS instruments). Measured apparent resistivity and chargeability values were inverted with Res2Dinv software. Inverted data show a strong correlation between important chargeability anomalies and calcite-filled tectonic fractures. However, neo-formed fractures (with no filling), associated with a mechanical damage, didn’t show any chargeability anomalies. Calcareous nodules in the clay matrix were also an important source of the chargeability anomalies. Furthermore, resistivity sections had been able to characterize desaturated part of the rock which is about 50 cm deep for the eastern gallery and it was 20 cm deep for the northern gallery. At this stage, our preliminary results show that the impact of a recent excavation is less developed for the northern gallery than eastern gallery.
Boundary estimation between dissemination and massivity in minerals using physical model studies in induced polarization (IP).
V. S. Sarma National Geophysical Research Institute, Council of Scientific and Industrial Research
In nature, Mineral ores can be massive or disseminated. Hitherto, studies have been carried out by Collett (1959), Bertin (1976) and others in this direction with the experiments over natural samples and a clear-cut definition to describe the nature of an ore is still awaited .In the present study ,physical model experiments have been carried out over synthetic samples. The advantage of experimentation over artificial samples is that we can have better control in varying the parameters like the grain size, volume content ,electrolyte concentrations etc .The samples are made in the laboratory out of cement and graphite to study the effect of increase in the volume percentage of graphite on the IP phenomena. In order to arrive at a reasonably good definition which is acceptable to all, laboratory studies were carried out using resistivity and induced polarisation over the synthetic samples. It is seen that the chargeability increases gradually as the graphite content increases to 18% by volume and then rapidly increases with a slope of 80 degrees from the horizontal when the graphite content increases to 25% indicating that further increase of graphite content from the 25% will not help substantially to increase the chargeability. Concurrently, resistivity response falls off with the increase in the graphite content. It is seen that resistivity decreases and chargeability increases linearly but slowly with the increase in the graphite content up to 25% by volume and then their trend becomes significantly high. This allows us to define a boundary between dissemination and massivity at 25% volume content of graphite. Almost a similar result is obtained from model tank experiments on cement-graphite sheets with varying graphite content. This model experimentation is carried out with Twoelectrode array over cement-graphite sheet submerged in host medium ,water .It is seen in this lab studies also that the apparent chargeability increases slowly and gradually till the volume content of the graphite reaches nearly 25% and then starts rapidly increasing with the further increase in the graphite content. The above two studies have indicated to define a boundary between dissemination and massivity at 25% volume content of graphite.
Delay time vs. depth of detection in induced polarization (IP) – model and field studies
V. S. Sarma National Geophysical Research Institute, Council of Scientific & Industrial Research
Apparent Time-Domain Induced Polarisation (TDIP) model profiles with different electrode arrays viz., two-electrode, three-electrode, dipole-dipole and wenner -sampled at delay times ( t ) = 20, 40 and 80 ms are obtained over a metallic sheet (110 x10 x 1 cm) submerged in host medium, water, contained in the Model tank. The top of the sheet is 1.0 cm below the water level. The primary current is sent as a series of alternate positive the negative square pulses with charging time of 400 ms and off-time of 800 ms. The thickness of the sheet is taken as unity and all other linear dimensions, like the distance of the electrode system from the centre of the target sheet, the depth of the target, etc are all expressed in terms of the thickness of the sheet. The IP-profiles show clearly that the amplitude of the anomaly increases with the decrease of delay time (t) for any electrode array. The shorter the delay time, greater is the amplitude of anomaly obtained with any of the arrays, and so, the depth of detection increases. It may, however, be noted that too much decrease of delay time may lead to electromagnetic coupling effects on IP measurements. Field studies are also carried out at Jonnagiri Village ,Kurnool District ,Andhra Pradesh ,India using Scintrex ( TSQ-3) Transmitter and Huntec (MK III) receiver .The apparent IP Modified pseudo-depth sections were constructed over the traverses at delay times of 90,180,360 and 720 ms. The charging current was sent in the form of square pulses with 2 s on-time and 2 s off-time by means of transmitter. Two interesting features were noted. (i)The maximum anomaly contours of both IP and resistivity fall, by and large, on the cross-section of the sulphide mineralization which is ultimately struck by the boreholes. (ii)The resistivity low ( pa=100 ohm-m ) and IP high ( M1=5.35% for td= 90ms.) just fall on the target cross-section. The IP modified pseudo-depth sections (M1,M2,M3 and M4) at different delay times exhibit that the maximum anomaly contours fall at the same coordinates in each of the sections. But the amplitude of IP anomaly decreases with an increase in delay time. This observation was found in all the depth sections with different electrode configurations. The above Model and Field studies mean that while the depth of investigation is independent of delay time where as the depth of detection is a function of it. The shorter the delay time, the greater the amplitude of IP anomaly and hence the larger the depth of detection. This implies that the depth of detection of a target with the IP method is much less than that with the resistivity method regardless of the electrode array used.
Spectral induced polarisation (SIP) – a significant geophysical tool for mineral detection and discrimination
V. S. Sarma1, R. Rajesh1 and P. Rajendra Prasad2 1
National Geophysical Research Institute, Council of Scientific and Industrial Research 2 Andhra University
Scale model studies have been carried out applying Spectral Induced Polarization (SIP) method using different electrode configurations viz., two-electrode,three-electrode,wenner and dipole-dipole for studying mineral detection and their discrimination possibilities, for example the minerals like graphite and sulphide etc. Field data has been acquired in Time Domain IP method over a sulphide area in Aldahalli village, Hassan district, Karnataka state, India using the above conventional electrode configurations. The data was later transformed to frequency domain IP.It was found that the frequency of the maximum phase shift (fm) obtained with the different electrode arrays lies in the range of variation from 4 to 8 Hz. Our laboratory model results right over cement – graphite targets with two-electrode configuration showed a range of variation from 0.4 to 0.7 Hz for different electrode spacings and target depths- almost one-tenth of the frequency values measured over sulphide deposit. This decade difference of ranges of the peaking frequency values – one over sulphide and the other over graphite, makes it, therefore, possible to their detection and further discrimination too. Modified pseudo-depth sections constructed over the identified mineralised zones also supported the observation. Our subsequent field studies carried out over a known graphite area at Josiar-alangam,Madurai district ,Tamilnadu India also supported the above laboratory findings. The large variation between the two ranges-one for sulphides and the other for graphite-makes it, therefore ,possible for their detection and discrimination.
Application of membrane polarization for studying internal structure of samples
V. Hallbauer-Zadorozhnaya, L. Maré, O. Dingoko Council for Geosciences
The new algorithm was tested on laboratory measurements of selected samples of Dwyka shale, shale with dropped stones, mudstone, tillite, hematite, lava, and manganese ore. All samples have been measured twice: electrodes were applied along the samples’ x and z axes so as to distinguish any possible anisotropy. The parameters of porosity have been laboratory measured and used for calculation. Mathematical modeling of all samples has been performed for 3 different electrical current strengths to reach one reliable model of the pores’ space. A model of pore spaces for a sample can only be accepted if theoretical and laboratory data fit for all three currents. Usually 45-65 cells (relatively large pore contacted with narrower capillaries) are used to describe a model of pore spaces for each sample; computing time of any model takes 2-3 seconds. Three important parameters characterize the samples: range of pore radii, prevalent pore radius and relative amount of pores of total pore space. Having quite different ranges of pores the prevalent pores’ radii of the various rocks are different. The range of pore radii in the sample of shale is 2 µm up to 18 µm, but the prevalent pore size is only about 3.5-4.1 µm for both directions. They are anisotropic sedimentary rocks, due to platey clay minerals. Shale with drop stones contains the largest pores whilst lava contained the smallest. Mathematical modeling allows us to obtain information about the anisotropy of a pore space. Shale and manganese ore demonstrated the maximal anisotropy along different directions of current flow. Hematite, lava, mudstone, and tillite have no anisotropy. Mathematical modeling provided reliable information of pores’ space of rocks, their anisotropy and provided a direction for the further study of permeability and transportation, especially of contaminant compounds.
Mathematical modelling of membrane polarization caused by constrictivity of pores
V. Hallbauer-Zadorozhnaya Council for Geosciences
Mathematical modeling of a little known model of IP referred to as “induced polarization caused by constrictivity of pores” was developed. Polarization occurs in all types of rocks if surface areas and transfer numbers are different for connected pores. The IP processes for the time on is described by the homogeneous equation with specified linear boundary conditions. In the time off the process of concentration distribution can be described by the inhomogeneous diffusion equation with specified initial condition and ions’ exchange coefficients describing boundary conditions. The duration of the polarization process depends mainly on three parameters viz: pore radii of connected capillaries, transfer numbers, and the amplitude of applied electrical current. If a large pore is connected to a narrower pore the concentration at the contact can drop to zero. It means during the polarization process some contacts between pores of different transfer numbers will be blocked and the electrical current will flow through the remaining channels. Excess/loss of concentration occurring at the boundary between pores with different size and transfer numbers and decrease of number of non-blocked pores channels influence measured resistivity. In this case a rupture of the electrical circuit occurs. Obviously if transfer numbers of connected large capillaries do not differ, a much longer polarization process can be recorded. Firstly it is related to porous rocks containing mainly large pores where the influence of DEL can be neglected. Therefore membrane IP can be regarded as the successive blockage of inter-pore connections due to the excess/loss distribution of ions during current flow. During time-on a voltage is present due to flowing current Ucurr (t) and excess concentration Uexcess (t) at the contacts. However during the time-on only the excess of concentration Uexcess (t) is involved in the diffusion process which intends to level the ion concentration along the pores. However the process of concentration levelling at time off is much longer due to the slow process of diffusion of ions occurring without applied electrical current. This model is used in the interpretation of induced polarization data and laboratory measurements of petrophysical properties of sediments.
Evaluating the EIT96 measurement system using defined samples with known electrical properties
E. Zimmermann1, A. Kemna2, J. A. Huisman3, J. Berwix1 and W Glaas1 1
Central Institute for Electronics (ZEL), Forschungszentrum Jülich GmbH 2 Department of Geodynamics and Geophysics, University of Bonn 3 Agrosphere (ICG-4), Forschungszentrum Jülich GmbH
Spectral electrical impedance tomography (EIT) is a powerful method for non-invasive characterization of soils and sediments. With this method, the diagnostic capabilities increase with the spectral bandwidth and the phase accuracy of the measured impedances. Typically, a bandwidth of 1 mHz to 1 kHz with a high phase accuracy in the order of 1 mrad is needed for reliable characterisation. To meet these requirements, a new spectral EIT data acquisition system for laboratory applications was developed, which operates in the frequency range from 1 mHz to 45 kHz. This system uses a new acquisition method based on current injection swapping, which leads to noticeable improved phase images at higher frequencies. In this contribution, we will present spectral imaging results (2D and 3D) for defined targets (e.g., water and a water-filled container with metallic and biological objects). The electrical properties of the targets are known from independent measurements with an impedance spectrometer (ZEL-SIP04). The results obtained using spectral EIT will be compared with this independent information. Additionally, the main error sources in EIT measurements will be discussed and procedures to correct these errors will be proposed.
Relationship between low-frequency electrical properties and hydraulic permeability of low-permeable sandstones
Norbert Zisser1, Andreas Kemna2 and Georg Nover1 1 Petrophysics, University of Bonn Department of Geodynamics and Geophysics, University of Bonn
2
Relationships between low-frequency electrical properties and hydraulic permeability of rocks have been in the focus of geophysical investigations for a long time, as they offer a possibility for in-situ and non-invasive permeability estimation. Empirical relationships between formation factor (F), induced polarization (IP), spectral induced polarization (SIP), and permeability (K) have been established. We examine the hydraulic and low-frequency electrical properties as well as the anisotropy of these properties of low-permeable microfractured sandstones. F-K and IP-K relations were found to be of low value for the determination of permeability for the studied samples, whereas a strong link between SIP (10 mHz – 100 Hz) and K exists. The SIP response was transformed into a relaxation time distribution. A strong positive correlation between median relaxation time (τ50) of the distribution and K is observed, suggesting that τ is a measure of the characteristic hydraulic length scale. From a comparison of our results with published τ-K relationships it becomes evident that this relationship is formation specific, requiring a separate calibration for each formation. Nevertheless τ-K relationships offer a great potential for in-situ permeability determination, as they seem to be applicable for very different lithologies (e.g., for consolidated and unconsolidated sediments, shaly sands, and micro-fractured rocks).