H44A-01 INVITED
Experimental Evidence That Electric Permittivity is a Measure of Interfacial Area in Unconsolidated Sand
An experimental tool is presented with which the capillary pressure and the electric permittivity of a porous medium can be measured simultaneously. The equipment is designed to conduct measurements for the unconsolidated sand-distilled water-gas system for pressures up to 20 bar and constant temperature conditions. The bulk phase pressures of the gas and the water are measured at the top and bottom of the sample holder. Also the pressure difference between these points is measured, which is taken as the capillary pressure. The sample holder is a parallel plate capacitor with stainless steel plates, which also serves as support for the sample. The measured total impedance is directly related to the effective value of the complex dielectric constant of the sample. Using this tool, the relation between capillary pressure and electric behavior has been investigated by measuring them simultaneously as function of the water saturation. The excess electric coefficient is postulated to be a measure for the interfacial area. Consequently, our set-up measures simultaneously the capillary pressure and the interfacial area. Drainage and imbibition experiments have been conducted for sand-distilled water-gas systems. The main capillary cycles and the scanning curves show hysteresis with the drainage curves showing higher values than the imbibition curves. The 100 kHz permittivity data also show hysteresis between drainage and imbibition. Furthermore non-monotonic behavior is observed, which is analogous to the interfacial characteristics obtained from network and micro-pore models. The permittivity behavior is attributed to polarization of the gas-water and water-solid interfaces. The permittivity hysteresis is provoked by the different phase distributions and geometries. Our results show that the capillary pressure is a unique function of the permittivity and water saturation, and therefore this work provides further evidence that the permittivity is a measure for the interfacial area.
H44A-02
Inversion for the statistical structure of subsurface water content from ground-penetrating radar reflection data: Initial results and interpretation
The distribution of subsurface water content can be an excellent indicator of soil texture, which strongly influences the unsaturated hydraulic properties controlling vadose zone contaminant transport. Characterizing the heterogeneity in subsurface water content for use in numerical transport models, however, is an extremely difficult task as conventional hydrological measurement techniques do not offer the combined high spatial resolution and coverage required for accurate simulations. A number of recent studies have shown that ground-penetrating radar (GPR) reflection images may contain useful information regarding the statistical structure of subsurface water content. Comparisons of the horizontal correlation structures of radar images and those obtained from water content measurements have shown that, in some cases, the statistical characteristics are remarkably similar. However, a key issue in these studies is that a reflection GPR image is primarily related to changes in subsurface water content, and not the water content distribution directly. As a result, statistics gathered on the reflection image have a very complex relationship with the statistics of the underlying water content distribution, this relationship depending on a number of factors including the frequency of the GPR antennas used. In this work, we attempt to address the above issue by posing the estimation of the statistical structure of water content from reflection GPR data as an inverse problem. Using a simple convolution model for a radar image, we first derive a forward model relating the statistical structure of a radar image to that of the underlying water content distribution. We then use this forward model to invert for the spatial statistics of the water content distribution, given the spatial statistics of the GPR reflection image as data. We do this within a framework of uncertainty, such that realistic statistical bounds can be placed on the information that is inferred. In other words, we attempt to address the question "what can we infer about the water-content statistical structure, given the GPR data?", rather than "what is the water content statistical structure?". Results of applying our estimation technique to simple synthetic models are positive, and give us hope that reflection GPR data can be used in practice to better constrain knowledge of the nature of subsurface water content heterogeneity. If successful, this type of approach could also be used with seismic reflection data to infer the statistical nature of velocity heterogeneities.
H44A-03
Joint Application of TDR, GPR and Inverse Hydraulic Modeling to Infer Field Scale Hydraulic Properties
Estimating field scale hydraulic properties is still a challenge in hydrology. Most classical methods require undisturbed soil samples that have to be excavated during time consuming and labour intensive field work which is often followed by tedious measurements of hydraulic properties in the laboratory. Since these methods can only be applied with a limited number of samples, often only a few point measurements need to be used to characterize field scale hydraulic properties while layer geometry has to be derived from interpolation of these values and additional drilling. The combination of geophysical measurement techniques and hydraulic modeling offers an attractive alternative to bridge the gap between i) few accurate point measurements that are used to infer local hydraulic properties and ii) spatial mapping of the respective layers over large scales. We use a time series of water contents measured in a soil profile with time domain reflectometry to estimate hydraulic properties of the different soil layers with a 1D hydraulic inverse model. Here, hydraulic properties are estimated from \it in situ \normalfont measurements under natural boundary conditions without the need of excavating undisturbed samples and lab measurements. Water fluxes across the soil-atmosphere boundary, which provide the upper boundary condition for the hydraulic modeling, are measured by an automatic weather station. Assuming homogeneous soil layers and neglecting small scale heterogeneity, multi-channel ground-penetrating radar is used to map the spatial geometry of the different soil layers and to extrapolate the hydraulic properties determined for the soil profile to the field scale. Provided that continuous reflections are present, the method is applicable at scales of hundreds of meters to kilometers.
H44A-04
Multi-Channel Ground Penetrating Radar: A Fast, Non-Invasive Tool to Detect Reflector Depth and Average Water Content Simultaneously
Many hydrologic applications require information from subsurface structure and water content distribution over distances of some hundreds of meters to kilometers. An accurate monitoring tool therefor would provide i) input data for and ii) allow the validation of large scale hydrologic numerical simulations. It would also close the gap between local measurements and large scale remote sensing applications. A promising tool is surface ground penetrating radar (GPR). The drawback of standard common offset applications was, that observed reflections depend on both reflector depth and propagation velocity. Hence, assumptions of either depth or propagation velocity had to be made. We overcome this problem using multi-channel GPR. This is a transportable unit, which can be considered as a moving mini common midpoint measurement. Its operation speed is roughly the same as a normal common offset measurement. The evaluation is done by numerical inversion of reflected wave travel times, which must be extracted from the different radargrams measured at different antenna separations. A synthetic and an experimental data set will be presented. The examples show, that multi-channel GPR leads to rather exact estimates of the subsurface structure and water content. Errors of this technique stem from air gaps below the antenna and from uncertainties in the extraction of the travel times, caused by interferences and electromagnetic noise, for instance. Whereas the air gaps, originating from soil roughness or antenna design, just lead to noise-like fluctuation in the results, interferences may lead to qualitative miss-interpretations.
H44A-05
Relating GPR Signal Response to Fracture Fluid Salinity
Predicting flow and transport in fractured bedrock remains a challenging problem. The spatial heterogeneity of flow (channeling) is commonly observed at seepage faces, but in-situ behavior is poorly understood. Time-lapse GPR amplitude differencing has been used to detect saline tracers in fractures and to identify groundwater flowpaths qualitatively. However, GPR observations of saline tracers in fractures have not been correlated quantitatively to meaningful hydrologic parameters, such as fracture aperture and fluid electrical conductivity. We investigate the relationships between GPR signal amplitude, phase, and frequency with fracture aperture and fluid salinity. Our research shows characteristic and quantifiable GPR responses that can be correlated to fracture properties. We use analytical modeling, numerical simulations, and field observations at a fractured sandstone aquifer in order to examine these relationships. We monitor the response of multifrequency GPR reflected signals from a water-saturated horizontal fracture at 7 m depth that we inject with saline tracers of increasing electrical conductivity. GPR response is simulated analytically and numerically. Although the fracture is a thin layer with millimeter scale aperture, increasing fluid conductivity results in greater EM signal wavenumber magnitude, which in turn decreases the signal wavelength and improves its thin layer resolution capabilities. These GPR signal responses from a thin layer are more pronounced in lower frequency signals. As shown by modeling, field data exhibit increasing signal amplitude and increasing phase lag as a function of increasing fluid electrical conductivity. The GPR amplitude and phase responses are detectable in the field and predictable by EM theory and modeling, therefore, they can be related to fracture aperture and fluid salinity for hydrologic investigations of fractured rock flow and transport properties.
H44A-06
Ground Penetrating Radar and Electrical Resistivity to Delineate Volcaniclastic Aquifers, La Laguneta, El Salvador
The village of La Laguneta in south-central El Salvador relies solely on groundwater for its freshwater needs. La Laguneta is located in a ca. 0.5 km2 basin on the flank of the inactive volcano El Vicente. The basin's principal aquifer consists of Quaternary siliceous pyroclastics and subordinate epiclastic volcanics underlain by a bedrock unit of Tertiary epiclastic volcanics and pyroclastics with intercalations of basic to intermediate lavas. Water depths in wells in the basin range from 10 to 25 m, with poor yields (<4 L/min) during the dry season when groundwater levels decline by as much as 10 m. Ground penetrating radar (GPR) and electrical resistivity (ER) were collected to determine the thickness of the principal aquifer unit across the basin. The 50 MHz GPR survey consisted of 11 2-D transects with 2 m trace spacing. The ER dipole-dipole survey consisted of 5 2-D transects with 5 m electrode spacing. Four ER transects coincided with GPR transect locations. Two wells, located adjacent to 3 GPR transects and 2 ER transects, provided ground truth. The GPR data provided good images to depths of 10 m at ca. 2 m resolution, revealing sedimentary onlap and fluvial cut and fill structures. This is underlain by a less-reflective and/or noisy interval, with some stronger horizontal reflectors visible to >30 m, that is interpreted as a volcanic/volcaniclastic aquitard beneath the fluvial sequence. A bedrock surface contour map based on the GPR data indicates that the present-day basin is longitudinally divided into two subsurface, partially-closed sub-basins with depths of ca. 13 m. The ER results are less conclusive but generally confirm the GPR interpretations. Many of the wells in the basin were drilled into the basement high between the two sub-basins explaining their low productivity. A well located in the northeast sub-basin documents the presence of a shallow water table at 4 m below ground surface. Assuming a horizontal water table and low seepage rate into the underlying volcanics, the center of each sub-basin should have the greatest saturated interval and best potential for a productive well.
H44A-07
Use of gravity and drawdown information to estimate hydraulic properties during unconfined aquifer testing
High-resolution gravimetry measured at the ground surface may offer a cost-effective method to augment aquifer tests in unconfined aquifers without the expense of installing additional monitoring wells or piezometers. However, evaluating the usefulness of gravity, used either alone or together with piezometric measurements, to help constrain hydraulic parameters through aquifer testing is difficult because observations of drawdown in a monitoring well and gravity measurements respond to different physical states of the system at different spatial scales. Measurements of drawdown provide a direct measure of the energy potential local to the observation location, while measurements of gravity change indirectly measure changes in water storage and depend on the spatially-weighted average of the subsurface density change integrated over a large measurement volume. Hydrogeophysical numerical experiments are conducted to examine the value of drawdown and gravity responses for constraining aquifer test analyses based on synthetic measurements at nine ground-surface locations after seven days of pumping. We confirm that the conditions conducive to this use of gravity measurements are those that lead to a sufficiently large change in gravity signal: high drainable porosity of the aquifer, low depth to the water table, and small distance from the pumping well to a gravity measurement location. However, we further demonstrate that the ability of a gravimeter to detect a signal above the noise in the data is not a sufficient condition to guarantee utility of the measurement method to constrain hydraulic parameter estimation. Specifically, the inference of hydraulic properties requires consideration of parameter interaction and parameter sensitivity to instrument responses in addition to instrument resolution and signal detectability. For conditions amenable to gravity monitoring, combined use of drawdown and gravity data resulted in unbiased and precise estimates of both hydraulic conductivity and specific yield. This offers improvement over monitoring drawdown alone, which often provides poor estimates of specific yield.
H44A-08
Predictive Models for Hydrodynamic Coupling Coefficients in Clay Media.
In charged and low permeability media (e.g. clay media) the classical Darcy's law does not describe accurately the water movement. A generalized Darcy's law, one of the coupled fluxes equations, has to be used. The identification of the coupling parameters, in clay-rocks, is crucial in order to estimate the water flow. Here, we will only focus on the electrochemical-hydraulic coupling coefficients : the intrinsic permeability k and the osmotic permeability kc. These hydrodynamic coupling coefficients can be estimated using two approaches: \indent(i) theoretical models : \indent• porosity/intrinsic permeability relationships, defined for a clay medium, are used to estimate the intrinsic permeability. \indent• an electrochemical model is used to estimate the osmotic coupling coefficient. The electrical model, a triple layer model, is implemented to simulate the interactions between the charged surfaces of the clay minerals and the pore solution. \indent(ii) experiments : \indent• at the sample scale. \indent• at the field scale. The measurement of these parameters is generally challenging either at the sample or at the field scale. For this reason, predictive models can be useful. The purpose of this study is to give reference values for the two coupling parameters under consideration here, using to the petrophysical properties of the medium. Different models to estimate these coupling coefficients are tested : (i) the intrinsic permeability is estimated with a pretrophysical model. In this model, the intrinsic permeability depends on the effective pore radius and the electrical formation factor. (ii) the osmotic coupling coefficient is estimated with the model developed by Revil and Leroy (2004). The comparison between three different models with the available data shows that these data are more closely reproduced using this model. Some reference values are provided for several type of clays as a fonction of some readily measurable or estimable parameters or variables, such as the porosity, the solute concentration. \begin{footnotesize} Revil, A. and Leroy, P. (2004). \newblock JGR, 109(B03208):10.1029/2003JB002755. \end{footnotesize}