H13G-01 INVITED 13:40h
Using Time-Lapse Geophysical Estimates of Pressure Change to Image Permeability
Estimates of fluid pressure and saturation changes provided by time-lapse geophysical methods are sensitive to reservoir flow properties, such as permeability. As shown in this talk, given high-resolution estimates of pressure changes in the subsurface, one may define a linear equation for permeability variations. In effect, the inverse problem for permeability, which in most settings is decidedly non-linear, is mapped into a sequence of linear sub-problems. The resulting linear inverse problems may be solved efficiently using sparse-matrix techniques. Two applications demonstrate the feasibility of this approach. The first application utilizes a set of tilt and borehole pressure data from a field site at Raymond, California. The observations are associated with a well test in which fluid is withdrawn from a shallow fracture zone. I am able to image a high permeability, north-trending channel within the fracture zone. The second application is to a set of crosswell saturation and pressure estimates from a CO$_2$ flood at the Lost Hills field in California. The resulting permeability estimates agree with a permeability log in an adjacent well and are in accordance with water and CO$_2$ saturation changes imaged in the interwell region.
H13G-02 INVITED 13:55h
Characterization of structures and transport processes in a heterogeneous aquifer using electrical conductivity and induced polarization imaging
Transport processes in aquifers are strongly determined by the structure of the aquifer. Since the structure typically exhibits significant heterogeneity, conventional measurement techniques, such as local sediment or water sampling, are generally incapable of capturing the variability of structural parameters as well as the complexity of transport processes in heterogeneous aquifers. Among the variety of proposed geophysical methods for hydrologic system characterization at the field scale, in particular electrical imaging techniques are being considered powerful non- to minimally invasive tools to overcome this incapability. They can provide spatially and temporally highly resolved information on subsurface parameters which are closely linked to both structural and transport properties. The most promising approaches in this regard include, first, the time-lapse application of electrical conductivity imaging to characterize the spatio-temporal behavior of subsurface solute transport and, second, the application of induced polarization imaging for improved structural characterization. In the first case, transport properties, such as advection velocity and dispersivity, can be quantified by interpreting the time-lapse electrical imaging results by means of equivalent transport models. Complementarily, induced polarization imaging allows the direct assessment of pertinent structural properties such as characteristic pore size and hydraulic conductivity. Importantly, both approaches live on the availability and validity of certain, normally site-specific calibration relations, such as between solute concentration and associated bulk electrical conductivity change, or between internal surface area of the sediment and imaginary component of electrical conductivity. We investigated the value of both approaches by conducting cross-borehole field surveys at the Krauthausen test site, which bears a shallow heterogeneous aquifer. The direct structure characterization capabilities are examined by comparing electrical imaging results with structural information gained independently, for example from a series of cone penetration tests. To assess the transport characterization capabilities, different tracer experiments were conducted and monitored by time-lapse electrical imaging, the results of which are compared with results obtained from conventional multi-level sampling in monitoring wells. In addition, synthetic experiments have been conducted to critically elucidate the limitations of the methodological approach.
H13G-03 14:10h
Monitoring a Field-Scale Biostimulation Pilot Project Using Cross-Hole Radar and Borehole Geophysical Methods
The U.S. Geological Survey (USGS) conducted geophysical investigations in support of a field-scale biostimulation pilot project at the Anoka County Riverfront Park (ACP), located downgradient of the Naval Industrial Reserve Ordnance Plant, in Fridley, Minnesota. The objective of the pilot project, conducted by the U.S. Naval Facilities Engineering Command, is to assess the applicability of subsurface injection of vegetable-oil emulsion (VOE) to promote microbial degradation of chlorinated hydrocarbons. Naturally occurring microbes, which use the VOE as substrate, ultimately break down chlorinated hydrocarbons into chloride, carbon dioxide, and water through oxidation-reduction reactions. To monitor movement of the VOE and changes in water chemistry resulting from VOE advection, dissolution, and (or) enhanced biological activity, the USGS acquired cross-hole zero-offset radar profiles; radar travel-time tomography data; and a suite of borehole geophysical logs, including electromagnetic (EM) induction conductivity. Data were collected during 5 site visits over 1.5 years. Preliminary results of these experiments have been reported elsewhere; this paper reports on the final analysis and combined interpretation of multiple data types, including application of petrophysical models to radar zero-offset profiles and tomograms to yield estimates of VOE saturation and changes in total-dissolved solids downgradient of the VOE injection zones. Comparison of pre- and post-injection datasets provides insight into the spatial and temporal distributions of both VOE and ground water with altered chemistry-information critical to understanding and verifying biodegradation of chlorinated hydrocarbons at the site. Cross-hole radar zero-offset slowness profiles and tomograms indicate the VOE remained close to the injection wells. Downgradient of the injection zones, radar amplitude profiles and EM logs indicate bulk formation electrical conductivity changes after VOE injection, which is diagnostic of changing water chemistry. At several monitoring wells, the field data indicate that the plume of ground water with altered chemistry would not be detected by direct sampling given the locations of monitoring-well screens; hence the geophysical data provide a context for interpretation of water samples and supplementary information useful for evaluation of the biostimulation effort. The study results demonstrate the utility of geophysical monitoring for engineered remediation operations.
H13G-04 14:25h
Analysis of spatial moments from a nonreactive tracer using electrical resistivity tomography
The use of geophysical data for hydrogeologic characterization has traditionally been qualitative. In this study, we use modified moment analysis of the 3D electrical resistivity tomograms through time to estimate the mass, center of mass, and spatial variance of a 2.2 g/L sodium-chloride tracer injected over 9 hours into saturated sands and gravels at the Massachusetts Military Reservation in Cape Cod, Massachusetts. Sixty ERT data sets were collected over 20 days, and each data set was inverted to produce a 3D map that images the plume. The inverted tomograms provide valuable insights into field-scale tracer migration behavior, but standard tomographic inversion and application of Archie's Law to convert bulk electrical conductivities to solute concentration results in underestimation of tracer mass and greater temporal spreading than observed in field data of concentration breakthrough at the pumping well. The ERT inversions also display greater apparent dispersion than tracer plumes estimated by 3D advective-dispersive simulation. This behavior is attributed to 1) reduced measurement sensitivity to electrical resistivity values with distance from the electrodes and, 2) differential smoothing from tomographic inversion. Despite these issues, the center of mass calculated from the ERT inversions coincides with that estimated by migration of the simulated tracer plume.
H13G-05 14:40h
Towards the Quantitative Characterization of Fractures Using GPR
GPR methods have been used successfully in hydrogeophysical studies for the detection of fractures. However, GPR studies have provided very limited quantitative information about the physical parameters that control fracture hydraulic properties, such as fracture aperture and aperture variation, channeling along the fracture plane, fracture surface roughness, and fluid content of fractures. In this study we examine the polarization properties of electromagnetic wavefields and develop analytical forward models of the GPR response of discrete fractures in geologic environments. We consider different geologic environments containing fractures of varying aperture, orientation, and fluid content. We simulate analytically the response of these fractures to transmitting and reflecting electromagnetic waves of varying polarization and frequency. We find that for a range of signal frequencies and angles of incidence, varying polarization GPR signals display characteristic amplitude and phase responses that allow for the determination of fracture fluid content and fracture aperture. Our findings indicate that quantitative characterization of fracture hydraulic properties can be achieved by considering the amplitude and phase response of varying polarization GPR wavefields.
H13G-06 14:55h
Flow-Channeling in Fractured Bedrock: Ground Penetrating Radar Investigations and Interpretation
Recent hydrogeophysical field experiments at a test site in Altona Flats, New York, USA, indicate that surface ground penetrating radar (GPR) can be used to quantitatively examine subhorizontal bedrock bedding plane partitions (fractures) at depth. Quantifiable factors include (1) fracture aperture and (2) variability in fluid salinity within the fracture. To analyze fracture aperture over the 10m x 10m test area, GPR data were collected at stations located every 20 cm in both the x and y direction using 50, 100, and 200 MHz antennas at eight different polarizations each (for a total of 24 "passes" over all the stations). The polarizations at each station included setting the GPR antennas in "parallel broadside" and "parallel endfire" configurations in both the x and y direction (for four total polarizations) and in "cross polarized" orientation at 90 degree intervals of rotation (for the second four polarizations). Thus, a significant amount of complimentary reflection information from the fracture is available for analysis. The innovation therefore includes integrating multiple polarizations of GPR data in 3D to characterize a bedrock fracture. To analyze variability in fluid salinity, the fracture-aperture data were used as background, and a series of additional GPR profiles (generally at 1-m spacing) were collected during various tracer tests. The difference in amplitude between the two datasets therefore can be correlated to changes in fracture fluid salinity because amplitude variations related to aperture are removed. Although other researchers have tracked tracer using GPR, those studies have not been done on horizontal fractures using surface GPR in 3D. The resulting innovation includes estimating tracer geometry (i.e., channeling), velocity, and concentration by integrating fracture aperture information (from GPR) with tracer-related GPR data.
http://www.geophysics.buffalo.edu
H13G-07 15:10h
Flow-Channeling in Fractured Bedrock: Combining GPR, Tracer, and Hydraulic Data
Channeling of flow in a fracture plane has long been hypothesized, but never observed in map view. In a series of recent experiments, Ground Penetrating Radar (GPR) reflections were used to track the migration of saline tracer through a single saturated horizontal fracture in sandstone bedrock near Altona New York. Reflection amplitudes increased significantly in the presence of saline tracer. Tracer movement was mapped at multiple points in time over an approximately 100 square meter grid. Both forced and natural gradient transport experiments were conducted. In forced gradient experiments, saline tracer swept an area that was much smaller than would be predicted for a homogeneous conductivity field. Tracer also moved in a very heterogeneous manner under the influence of a natural hydraulic gradient. Hydraulic response in wells and tracer breakthrough compare favorably with the distribution of GPR-mapped saline fluid. These preliminary results suggest that the concept of channeled flow in rock fractures is valid, but specific channeling behavior is highly dependent upon hydraulic forcing. This combination of GPR, tracer, and hydraulic information may help develop better methods for measuring effective porosity through fractured bedrock.
http://www.geology.buffalo.edu/Research/hydro/
H13G-08 15:25h
Relationship between horizontal flow measurements in wells and flow in the surrounding formation
Two computational models are developed to estimate and visualize the flow field through a wellbore to evaluate the utility and accuracy of a scanning colloidal borescope flowmeter (SCBFM). The SCBFM images colloidal particles traveling through a well, and measures the direction and velocity of movement. The first model calculates the flow that would arise in a fractured medium where the majority of flow is through a single fracture into an unscreened well. For comparison to this model, an SCBFM was deployed in a sand tank flow simulator to measure the flow field within a PVC well screen having a single slot (representing a fracture) aligned with the flow direction. In this case, the SCBFM and model agree with respect to the flow direction and to within a factor of three for the velocity near the center of the well. For a well drilled in a fractured rock medium with no casing and with proper calibration, the SCBFM could be used to identify and quantify flowing features. The second model numerically estimates the flow field within a multi-slot well screen in an attempt to evaluate the potential usefulness of employing the SCBFM in a screened well to estimate flow speed and direction in the surrounding porous medium. Results indicate that when the slots are aligned with the principal flow direction, the SCBFM can be used to infer the local ground-water flow speed and direction. However, if the slots are not aligned with the flow (a situation difficult to evaluate a priori), the SCBFM can only provide order of magnitude velocity measurements and direction measurements with an uncertainty of approximately ±25°. Controlled laboratory experiments and detailed modeling are strongly recommended to calibrate any device intended to measure flow within a well. {\st Sandia is a multiprogram laboratory operated by Sandia Corporation, a Lockheed Martin Company, for the United States Department of Energy's National Nuclear Security Administration under contract DE-AC04-94AL85000.}