Consistent interpretation of the results of through-, out-diffusion and tracer profile analysis for trace anion diffusion in compacted montmorillonite

Literature data for anion diffusion in compacted swelling clays contain systematic inconsistencies when the results of through-diffusion tests are compared with those of out-diffusion or tracer profile analysis. In the present work we investigated whether these inconsistencies can be explained by taking into account heterogeneities in the compacted samples; in particular increased porosities at the clay boundaries. Based on the combined results of out-diffusion, tracer profile analysis and the spatial distribution of the electrolyte anion in the clay, we conclude that the inconsistencies can indeed be resolved by taking into account a heterogeneous distribution of the total and the anion-accessible porosity. This, by definition, leads to a position dependence of the effective diffusion coefficient. Neglecting these effects results in a rather subordinate systematic error in the determination of effective diffusion coefficients of anions from through-diffusion tests with clay thicknesses in the centimetre range. However, stronger errors in terms of absolute values and conceptual interpretation may be introduced in out-diffusion tests and profile analyses of the diffused tracer. We recommend that anion diffusion tests should be accompanied by measurements of the total and anion-accessible porosity as a function of position in the direction of diffusion.     

Direct simulation of heterogeneous diffusion and inversion procedure applied to an out-diffusion experiment. Test case of Palmottu granite

An out-diffusion laboratory experiment using a non-reactive tracer was fitted using the Time Domain Diffusion (TDD) method. This rapid particle tracking method allows simulation of the heterogeneous diffusion based on pore-scale images and local values of diffusivities. The superimposed porosity and mineral 2D maps act as computation grids to condition diffusion pathways. We focused on a Palmottu granite sample, in which the connected pore space has a composite microstructure with cracks linking microporous minerals and is above the percolation threshold. Three main results were achieved: (i) When compared to the fitting obtained with one coefficient (best mean square residual R = 1.6 x 10(-2)), diffusion is shown to be suitably characterised with two coefficients related to cracks and microporous minerals (best R = 6.5 x 10(-4)), (ii) rather than imposing a local apparent diffusion coefficient D(a) independent of the local porosity Phi, a best fit is obtained by applying Archie's relationship D(a) = D(0) x G with G = Phi(m) to each pixel of the calculation grids (G is the geometry factor, D(0) is the diffusion coefficient in free fluid, and m is Archie's exponent), and (iii) the order of magnitude of the fitted diffusion coefficient or Archie's exponents (m=0 for microcracks and m=1.82 for microporous minerals) is physically realistic.     

Matrix diffusion coefficients in volcanic rocks at the Nevada test site: influence of matrix porosity, matrix permeability, and fracture coating minerals

Diffusion cell experiments were conducted to measure nonsorbing solute matrix diffusion coefficients in forty-seven different volcanic rock matrix samples from eight different locations (with multiple depth intervals represented at several locations) at the Nevada Test Site. The solutes used in the experiments included bromide, iodide, pentafluorobenzoate (PFBA), and tritiated water ((3)HHO). The porosity and saturated permeability of most of the diffusion cell samples were measured to evaluate the correlation of these two variables with tracer matrix diffusion coefficients divided by the free-water diffusion coefficient (D(m)/D*). To investigate the influence of fracture coating minerals on matrix diffusion, ten of the diffusion cells represented paired samples from the same depth interval in which one sample contained a fracture surface with mineral coatings and the other sample consisted of only pure matrix. The log of (D(m)/D*) was found to be positively correlated with both the matrix porosity and the log of matrix permeability. A multiple linear regression analysis indicated that both parameters contributed significantly to the regression at the 95% confidence level. However, the log of the matrix diffusion coefficient was more highly-correlated with the log of matrix permeability than with matrix porosity, which suggests that matrix diffusion coefficients, like matrix permeabilities, have a greater dependence on the interconnectedness of matrix porosity than on the matrix porosity itself. The regression equation for the volcanic rocks was found to provide satisfactory predictions of log(D(m)/D*) for other types of rocks with similar ranges of matrix porosity and permeability as the volcanic rocks, but it did a poorer job predicting log(D(m)/D*) for rocks with lower porosities and/or permeabilities. The presence of mineral coatings on fracture walls did not appear to have a significant effect on matrix diffusion in the ten paired diffusion cell experiments.     

An axisymmetric diffusion experiment for the determination of diffusion and sorption coefficients of rock samples

Diffusion anisotropy is a critical property in predicting migration of substances in sedimentary formations with very low permeability. The diffusion anisotropy of sedimentary rocks has been evaluated mainly from laboratory diffusion experiments, in which the directional diffusivities are separately estimated by through-diffusion experiments using different rock samples, or concurrently by in-diffusion experiments in which only the tracer profile in a rock block is measured. To estimate the diffusion anisotropy from a single rock sample, this study proposes an axisymmetric diffusion test, in which tracer diffuses between a cylindrical rock sample and a surrounding solution reservoir. The tracer diffusion between the sample and reservoir can be monitored from the reservoir tracer concentrations, and the tracer profile could also be obtained after dismantling the sample. Semi-analytical solutions are derived for tracer concentrations in both the reservoir and sample, accounting for an anisotropic diffusion tensor of rank two as well as the dilution effects from sampling and replacement of reservoir solution. The transient and steady-state analyses were examined experimentally and numerically for different experimental configurations, but without the need for tracer profiling. These experimental configurations are tested for in- and out-diffusion experiments using Koetoi and Wakkanai mudstones and Shirahama sandstone, and are scrutinized by a numerical approach to identify favorable conditions for parameter estimation. The analysis reveals the difficulty in estimating diffusion anisotropy; test configurations are proposed for enhanced identifiability of diffusion anisotropy. Moreover, it is demonstrated that the axisymmetric diffusion test is efficient in obtaining the sorption parameter from both steady-state and transient data, and in determining the effective diffusion coefficient if isotropic diffusion is assumed. Moreover, measuring reservoir concentrations in an axisymmetric diffusion experiment coupled with tracer profiling may be a promising approach to estimate of diffusion anisotropy of sedimentary rocks.     

Influence of the mode of matrix porosity determination on matrix diffusion calculations

The theoretical basis for matrix diffusion in fractured rocks and the methodology for the determination of diffusion coefficients in the laboratory are well established. One significant problem, however, remains in that it is difficult to quantify the degree of sample disturbance affecting the geometrical, geophysical and hydraulic properties of the rock matrix. A new technique, with in situ rock impregnation with resin, for examining the diffusion-accessible rock matrix has been developed and successfully adopted to the rock matrix behind a water-conducting fracture in host crystalline rocks at Nagra's Grimsel Test Site in Switzerland and JNC's Kamaishi In Situ Test Site in Japan. In line with the results of a large number of natural analogue and laboratory studies, the existence of an in situ interconnected pore network was substantiated. Matrix porosities determined on the laboratory samples from both the sites are 1.5-3 times higher than in situ values, irrespective of the technique applied. On the Grimsel granodiorite matrix, matrix porosity existing in situ and artefacts of stress release and physical disturbance, induced by sampling and sample preparation, were clearly distinguished, allowing in situ porosity to be quantified. Laboratory work with conventional techniques tends to overestimate the porosity of the rock matrix, hence leading to an overestimation of in situ matrix diffusion. The implications of these differences to a repository performance assessment are assessed with a couple of examples from existing assessments, and recommendations for future approaches to the examination of in situ matrix porosity are made.     

Diffusion of tritiated water and 22Na+ through non-degraded hardened cement pastes

Diffusion experiments through hardened cement pastes (HCP) using tritiated water (HTO) and 22Na(+), considered to be conservative tracers, have been carried out in triplicates in a glove box under a controlled nitrogen atmosphere. Each experiment consisted of a through-diffusion test followed by an out-diffusion test. The experimental data were inversely modelled applying an automated Marquardt-Levenberg procedure. The analysis of the through-diffusion data allowed the extraction of values for the effective diffusion coefficients, D(e), and the rock capacity factor, alpha. Good agreement between measured and calculated tracer breakthrough curves was achieved using both a simple diffusion model without sorption and a diffusion/linear sorption model. The best-fit K(d)-values were found to be consistent with R(d)-values measured in previous batch-sorption experiments. The best-fit values from the through-diffusion tests were then used to predict the results of subsequent out-diffusion experiments. Good agreement between experimental data and predictions was achieved only for the case of linear sorption. Isotopic exchange can only partially account for both the amount of tracer taken up in the batch-sorption tests and the measured retardation in the diffusion experiments and, hence, additional mechanisms have to be invoked to explain the data.     

Diffusive flux and pore anisotropy in sedimentary rocks

Diffusion of dissolved contaminants into or from bedrock matrices can have a substantial impact on both the extent and longevity of dissolved contaminant plumes. For layered rocks, bedding orientation can have a significant impact on diffusion. A series of laboratory experiments was performed on minimally disturbed bedrock cores to measure the diffusive flux both parallel and normal to mineral bedding of four different anisotropic sedimentary rocks. Measured effective diffusion coefficients ranged from 4.9×10(-8) to 6.5×10(-7)cm(2)/s. Effective diffusion coefficients differed by as great as 10-folds when comparing diffusion normal versus parallel to bedding. Differences in the effective diffusion coefficients corresponded to differences in the "apparent" porosity in the orientation of diffusion (determined by determining the fraction of pore cross-sectional area measured using scanning electron microscopy), with the difference in apparent porosity between normal and parallel bedding orientations differing by greater than 2-folds for two of the rocks studied. Existing empirical models failed to provide accurate predictions of the effective diffusion coefficient in either bedding orientation for all four rock types studied, indicating that substantial uncertainty exists when attempting to predict diffusive flux through sedimentary rocks containing mineral bedding. A modified model based on the apparent porosity of the rocks provided a reasonable prediction of the experimental diffusion data.     

Diffusion of tritiated water and Na through non-degraded hardened cement pastes

Diffusion experiments through hardened cement pastes (HCP) using tritiated water (HTO) and ²²Na⁺, considered to be conservative tracers, have been carried out in triplicates in a glove box under a controlled nitrogen atmosphere. Each experiment consisted of a through-diffusion test followed by an out-diffusion test.The experimental data were inversely modelled applying an automated Marquardt–Levenberg procedure. The analysis of the through-diffusion data allowed the extraction of values for the effective diffusion coefficients, D_e, and the rock capacity factor, α. Good agreement between measured and calculated tracer breakthrough curves was achieved using both a simple diffusion model without sorption and a diffusion/linear sorption model. The best-fit K_d-values were found to be consistent with R_d-values measured in previous batch-sorption experiments.The best-fit values from the through-diffusion tests were then used to predict the results of subsequent out-diffusion experiments. Good agreement between experimental data and predictions was achieved only for the case of linear sorption.Isotopic exchange can only partially account for both the amount of tracer taken up in the batch-sorption tests and the measured retardation in the diffusion experiments and, hence, additional mechanisms have to be invoked to explain the data.     

Use of X-ray absorption imaging to examine heterogeneous diffusion in fractured crystalline rocks

Heterogeneous diffusion in different regions of a fractured granodiorite from Japan has been observed and measured through the use of X-ray absorption imaging. These regions include gouge-filled fractures, recrystallized fracture-filling material and hydrothermally altered matrix. With the X-ray absorption imaging technique, porosity, relative concentration, and relative mass of an iodine tracer were imaged in two dimensions with a sub-millimeter pixel size. Because portions of the samples analyzed have relatively low porosity values, imaging errors can potentially impact the results. For this reason, efforts were made to better understand and quantify this error. Based on the X-ray data, pore diffusion coefficients (Dp) for the different regions were estimated assuming a single diffusion rate and a lognormal multirate distribution of Dp. Results show Dp for the gouge-filled fractures are over an order of magnitude greater than those of the recrystallized fracture-filling material, which in turn is approximately two times greater than those for the altered matrix. The recrystallized fracture-filling material was found to exhibit the greatest degree of variability. The results of these experiments also provide evidence that diffusion from advective zones in fractures through the gouge-filled fractures and recrystallized fracture-filling material could increase the pore space available for matrix diffusion. This evidence is important for understanding the performance of potential nuclear waste repositories in crystalline rocks as diffusion is thought to be an important retardation mechanism for radionuclides.     

Tracer diffusion coefficients in sedimentary rocks: correlation to porosity and hydraulic conductivity

Matrix diffusion is an important transport process in geologic materials of low hydraulic conductivity. For predicting the fate and transport of contaminants, a detailed understanding of the diffusion processes in natural porous media is essential. In this study, diffusive tracer transport (iodide) was investigated in a variety of geologically different limestone and sandstone rocks. Porosity, structural and mineralogical composition, hydraulic conductivity, and other rock properties were determined. The effective diffusion coefficients were measured using the time-lag method. The results of the diffusion experiments indicate that there is a close relationship between total porosity and the effective diffusion coefficient of a rock (analogous to Archie's Law). Consequently, the tortousity factor can be expressed as a function of total porosity. The relationship fits best for thicker samples (> 1.0 cm) with high porosities (> 20%), because of the reduced influence of heterogeneity in larger samples. In general, these correlations appear to be a simple way to determine tortuosity and the effective diffusion coefficient from easy to determine rock porosity values.     

The effect of a biofilm on solute diffusion in fractured porous media

At sites in fractured rock where contamination has been exposed to the rock matrix for extended periods of time, the amount of contaminant mass residing in the matrix can be considerable. Even though it may be possible to diminish concentrations by the advection of clean water through the fracture features, back diffusion from mass held in the matrix will lead to a continuing source of contamination. In such an event, the development of a biofilm (a thin film of microbial mass) on the wall of the fractures may act to limit or prevent the back diffusion process. The objective of this preliminary study is to explore the influence imparted by the presence of a biofilm on the process of matrix diffusion. The investigation was conducted using radial diffusion cells constructed from rock core in which biofilm growth was stimulated in a central reservoir. Once biofilms were developed, forward diffusion experiments were conducted in which a conservative solute migrated from the central reservoir into the intact rock sample. Diffusion experiments were performed in a total of 11 diffusion cell pairs where biofilm growth was stimulated in one member of the pair and inhibited in the other. The effect of the presence of a biofilm on tracer diffusion was determined by comparison of the diffusion curves produced by each cell pair. A semi-analytical model that accounts for the presence of a biofilm was used to investigate the effect of the biofilm on mass transfer due to changes in the effective porosity, effective diffusion coefficient, and the depth of penetration of the biofilm into the intact rock. The results show that the biofilm acted to plug the rock matrix, rather than forming a discrete layer on the reservoir surface. The reduction in effective porosity due to the biofilm ranged from 6% to 52% with the majority of the samples in the 30% to 50% range. Based on the present results, with more efficient biofilm stimulation, it is reasonable to assume that a more complete plugging of the microcrack porosity might be possible, leaving a much thicker and efficient barrier than could be achieved via a surface biofilm.     

Contribution of the diffusive exchange method to the characterization of pore-water in consolidated argillaceous rocks

Consolidated argillaceous rocks, potential host rocks for nuclear waste disposal, display particular properties (low water content, very small pore size) that require specific methods, especially for the characterization of their pore-water. The diffusive exchange method was applied to the claystone of Tournemire by taking into account the effect of swelling or micro-cracks induced by drilling. The good agreement of diffusion parameters with those obtained from the through-diffusion method showed that the extent of this disturbance was low. This method led to stable isotope content values in pore-water in good agreement with those measured in the fracture fluids located in the vicinity, validating the method. On the contrary, the vacuum distillation technique gave a systematic depletion in heavy isotopes of water, probably induced by an incomplete extraction of pore-water. For the chloride tracer, the determination of the accessible porosity on each sample allowed the estimate of chloride concentration of pore-water to be improved, by taking into account all the petrophysical heterogeneities.     

A technique for estimating one-dimensional diffusion coefficients in low-permeability sedimentary rock using X-ray radiography: comparison with through-diffusion measurements

The measurement of diffusive properties of low-permeability rocks is of interest to the nuclear power industry, which is considering the option of deep geologic repositories for management of radioactive waste. We present a simple, non-destructive, constant source in-diffusion method for estimating one-dimensional pore diffusion coefficients (D(p)) in geologic materials based on X-ray radiography. Changes in X-ray absorption coefficient (Deltamicro) are used to quantify changes in relative concentration (C/C(0)) of an X-ray attenuating iodide tracer as the tracer solution diffuses through the rock pores. Estimated values of D(p) are then obtained by fitting an analytical solution to the measured concentration profiles over time. Measurements on samples before and after saturation with iodide can also be used to determine iodide-accessible porosity (phi(I)). To evaluate the radiography method, results were compared with traditional steady-state through-diffusion measurements on two rock types: shale and limestone. Values of D(p) of (4.8+/-2.5)x10(-11) m(2).s(-1) (mean+/-standard deviation) were measured for samples of Queenston Formation shale and (2.6+/-1.0)x10(-11) m(2).s(-1) for samples of Cobourg Formation limestone using the radiography method. The range of results for each rock type agree well with D(p) values of (4.6+/-2.0)x10(-11) m(2).s(-1) for shale and (3.5+/-1.8)x10(-11) m(2).s(-1) for limestone, calculated from through-diffusion experiments on adjacent rock samples. Low porosity (0.01 to 0.03) and heterogeneous distribution of porosity in the Cobourg Formation may be responsible for the slightly poorer agreement between radiography and through-diffusion results for limestones. Mean values of phi(I) for shales (0.060) and limestones (0.028) were close to mean porosity measurements made on bulk samples by the independent water loss technique (0.062 and 0.020 for shales and limestones, respectively). Radiography measurements offer the advantage of time-saving for diffusion experiments because the experiment does not require steady-state conditions and also allows for visualization of the small-scale heterogeneities in diffusive properties within rocks at the mm to cm scale.     

Measurement of effective air diffusion coefficients for trichloroethene in undisturbed soil cores

In this study, we measure effective diffusion coefficients for trichloroethene in undisturbed soil samples taken from Picatinny Arsenal, New Jersey. The measured effective diffusion coefficients ranged from 0.0053 to 0.0609 cm2/s over a range of air-filled porosity of 0.23-0.49. The experimental data were compared to several previously published relations that predict diffusion coefficients as a function of air-filled porosity and porosity. A multiple linear regression analysis was developed to determine if a modification of the exponents in Millington's [Science 130 (1959) 100] relation would better fit the experimental data. The literature relations appeared to generally underpredict the effective diffusion coefficient for the soil cores studied in this work. Inclusion of a particle-size distribution parameter, d10, did not significantly improve the fit of the linear regression equation. The effective diffusion coefficient and porosity data were used to recalculate estimates of diffusive flux through the subsurface made in a previous study performed at the field site. It was determined that the method of calculation used in the previous study resulted in an underprediction of diffusive flux from the subsurface. We conclude that although Millington's [Science 130 (1959) 100] relation works well to predict effective diffusion coefficients in homogeneous soils with relatively uniform particle-size distributions, it may be inaccurate for many natural soils with heterogeneous structure and/or non-uniform particle-size distributions.     

A comparison of hydraulic and transport parameters measured in low-permeability fractured media

Data from 90 tracer experiments performed in low-permeability fractured media have been studied to explore correlations among parameters controlling flow and transport. The original data had been interpreted by different authors using different models, which prevents direct comparison of their estimated parameters. In order to produce comparable parameters, the data have been reexamined using simple models (homogeneous domain, steady-state flow regime, single porosity). Specifically, hydraulic conductivity has been derived as the ratio of water flux to head gradient and apparent porosity as the ratio of water velocity to water flux; the former estimated from both first and peak arrival times. Hydraulic conductivity and porosity correlate along a straight line of slope 1:3 in log scale. While the regression is too noisy to be of predictive use, it lends some support to the use of a generalized cubic law. The fact that correlation for first arrival time porosity (0.77) is larger than for peak arrival porosity (0.62) suggests that first arrival is controlled by the same flow paths as hydraulic conductivity. Apparent porosity derived from peak arrival time is found to grow with travel time along a line of 0.55 slope (again log scale). The correlation coefficient ranges between 0.73 and 0.80 (depending on the data set) for hard rocks. The fact that this correlation is maintained when varying the flow rate at a given site leads us to suggest that it is caused by diffusion mechanisms. This conclusion is further supported by the increase of apparent porosity with the matrix porosity of the rock on which the experiments were performed.     

Solute transport in crystalline rocks at Äspö — I: Geological basis and model calibration

Water-conducting faults and fractures were studied in the granite-hosted Äspö Hard Rock Laboratory (SE Sweden). On a scale of decametres and larger, steeply dipping faults dominate and contain a variety of different fault rocks (mylonites, cataclasites, fault gouges). On a smaller scale, somewhat less regular fracture patterns were found. Conceptual models of the fault and fracture geometries and of the properties of rock types adjacent to fractures were derived and used as input for the modelling of in situ dipole tracer tests that were conducted in the framework of the Tracer Retention Understanding Experiment (TRUE-1) on a scale of metres. After the identification of all relevant transport and retardation processes, blind predictions of the breakthroughs of conservative to moderately sorbing tracers were calculated and then compared with the experimental data. This paper provides the geological basis and model calibration, while the predictive and inverse modelling work is the topic of the companion paper [J. Contam. Hydrol. 61 (2003) 175].The TRUE-1 experimental volume is highly fractured and contains the same types of fault rocks and alterations as on the decametric scale. The experimental flow field was modelled on the basis of a 2D-streamtube formalism with an underlying homogeneous and isotropic transmissivity field. Tracer transport was modelled using the dual porosity medium approach, which is linked to the flow model by the flow porosity. Given the substantial pumping rates in the extraction borehole, the transport domain has a maximum width of a few centimetres only. It is concluded that both the uncertainty with regard to the length of individual fractures and the detailed geometry of the network along the flowpath between injection and extraction boreholes are not critical because flow is largely one-dimensional, whether through a single fracture or a network. Process identification and model calibration were based on a single uranine breakthrough (test PDT3), which clearly showed that matrix diffusion had to be included in the model even over the short experimental time scales, evidenced by a characteristic shape of the trailing edge of the breakthrough curve. Using the geological information and therefore considering limited matrix diffusion into a thin fault gouge horizon resulted in a good fit to the experiment. On the other hand, fresh granite was found not to interact noticeably with the tracers over the time scales of the experiments.While fracture-filling gouge materials are very efficient in retarding tracers over short periods of time (hours–days), their volume is very small and, with time progressing, retardation will be dominated by altered wall rock and, finally, by fresh granite. In such rocks, both porosity (and therefore the effective diffusion coefficient) and sorption K_ds are more than one order of magnitude smaller compared to fault gouge, thus indicating that long-term retardation is expected to occur but to be less pronounced.     

Modeling solute diffusion in the presence of pore-scale heterogeneity: method development and an application to the Culebra dolomite member of the Rustler Formation, New Mexico, USA

Previous studies have revealed the presence of pore-scale variability in diffusivity in the Culebra (dolomite) member of the Rustler Formation, NM. In this study, eight laboratory-scale diffusion experiments on five Culebra samples were analyzed using a methodology for modeling solute diffusion through porous media in the presence of multiple matrix diffusivities, Dp. A lognormal distribution of Dp is assumed within each of the lab samples. The estimated standard deviation (sigma d) of ln(Dp) within each sample ranges from 0 to 1, with most values lying between 0.5 and 1. The variability over all samples leads to a combined sigma d in the range of 1.0-1.2, which is consistent with the distribution of independently determined formation factor measurements for similar Culebra samples. A comparison of our estimation results to other rock properties suggests that, at the lab-scale, the geometric mean of Dp increases with bulk porosity and the quantity of macroscopic features such as vugs and fractures. However, sigma d appears to be determined by variability within such macroscopic features and/or by micropore-scale heterogeneity. In addition, comparison of these experiments to those at larger spatial scales suggests that increasing sample volume results in an increase in sigma d.     

A multi-directional tracer test in the fractured Chalk aquifer of E. Yorkshire, UK

A multi-borehole radial tracer test has been conducted in the confined Chalk aquifer of E. Yorkshire, UK. Three different tracer dyes were injected into three injection boreholes and a central borehole, 25 m from the injection boreholes, was pumped at 330 m(3)/d for 8 days. The breakthrough curves show that initial breakthrough and peak times were fairly similar for all dyes but that recoveries varied markedly from 9 to 57%. The breakthrough curves show a steep rise to a peak and long tail, typical of dual porosity aquifers. The breakthrough curves were simulated using a 1D dual porosity model. Model input parameters were constrained to acceptable ranges determined from estimations of matrix porosity and diffusion coefficient, fracture spacing, initial breakthrough times and bulk transmissivity of the aquifer. The model gave equivalent hydraulic apertures for fractures in the range 363-384 microm, dispersivities of 1 to 5 m and matrix block sizes of 6 to 9 cm. Modelling suggests that matrix block size is the primary controlling parameter for solute transport in the aquifer, particularly for recovery. The observed breakthrough curves suggest results from single injection-borehole tracer tests in the Chalk may give initial breakthrough and peak times reasonably representative of the aquifer but that recovery is highly variable and sensitive to injection and abstraction borehole location. Consideration of aquifer heterogeneity suggests that high recoveries may be indicative of a high flow pathway adjacent, but not necessarily connected, to the injection and abstraction boreholes whereas low recoveries may indicate more distributed flow through many fractures of similar aperture.     

Solute transport in crystalline rocks at Aspö--I: geological basis and model calibration

Water-conducting faults and fractures were studied in the granite-hosted Asp? Hard Rock Laboratory (SE Sweden). On a scale of decametres and larger, steeply dipping faults dominate and contain a variety of different fault rocks (mylonites, cataclasites, fault gouges). On a smaller scale, somewhat less regular fracture patterns were found. Conceptual models of the fault and fracture geometries and of the properties of rock types adjacent to fractures were derived and used as input for the modelling of in situ dipole tracer tests that were conducted in the framework of the Tracer Retention Understanding Experiment (TRUE-1) on a scale of metres. After the identification of all relevant transport and retardation processes, blind predictions of the breakthroughs of conservative to moderately sorbing tracers were calculated and then compared with the experimental data. This paper provides the geological basis and model calibration, while the predictive and inverse modelling work is the topic of the companion paper [J. Contam. Hydrol. 61 (2003) 175]. The TRUE-1 experimental volume is highly fractured and contains the same types of fault rocks and alterations as on the decametric scale. The experimental flow field was modelled on the basis of a 2D-streamtube formalism with an underlying homogeneous and isotropic transmissivity field. Tracer transport was modelled using the dual porosity medium approach, which is linked to the flow model by the flow porosity. Given the substantial pumping rates in the extraction borehole, the transport domain has a maximum width of a few centimetres only. It is concluded that both the uncertainty with regard to the length of individual fractures and the detailed geometry of the network along the flowpath between injection and extraction boreholes are not critical because flow is largely one-dimensional, whether through a single fracture or a network. Process identification and model calibration were based on a single uranine breakthrough (test PDT3), which clearly showed that matrix diffusion had to be included in the model even over the short experimental time scales, evidenced by a characteristic shape of the trailing edge of the breakthrough curve. Using the geological information and therefore considering limited matrix diffusion into a thin fault gouge horizon resulted in a good fit to the experiment. On the other hand, fresh granite was found not to interact noticeably with the tracers over the time scales of the experiments. While fracture-filling gouge materials are very efficient in retarding tracers over short periods of time (hours-days), their volume is very small and, with time progressing, retardation will be dominated by altered wall rock and, finally, by fresh granite. In such rocks, both porosity (and therefore the effective diffusion coefficient) and sorption K(d)s are more than one order of magnitude smaller compared to fault gouge, thus indicating that long-term retardation is expected to occur but to be less pronounced.