Parameter and observation importance in modelling virus transport in saturated porous media-investigations in a homogenous system

This paper evaluates the importance of seven types of parameters to virus transport: hydraulic conductivity, porosity, dispersivity, sorption rate and distribution coefficient (representing physical-chemical filtration), and in-solution and adsorbed inactivation (representing virus inactivation). The first three parameters relate to subsurface transport in general while the last four, the sorption rate, distribution coefficient, and in-solution and adsorbed inactivation rates, represent the interaction of viruses with the porous medium and their ability to persist. The importance of four types of observations to estimate the virus-transport parameters are evaluated: hydraulic heads, flow, temporal moments of conservative-transport concentrations, and virus concentrations. The evaluations are conducted using one- and two-dimensional homogeneous simulations, designed from published field experiments, and recently developed sensitivity-analysis methods. Sensitivity to the transport-simulation time-step size is used to evaluate the importance of numerical solution difficulties. Results suggest that hydraulic conductivity, porosity, and sorption are most important to virus-transport predictions. Most observation types provide substantial information about hydraulic conductivity and porosity; only virus-concentration observations provide information about sorption and inactivation. The observations are not sufficient to estimate these important parameters uniquely. Even with all observation types, there is extreme parameter correlation between porosity and hydraulic conductivity and between the sorption rate and in-solution inactivation. Parameter estimation was accomplished by fixing values of porosity and in-solution inactivation.     

Implications of non-equilibrium transport in heterogeneous reactive barrier systems: evidence from laboratory denitrification experiments

Organic substrates in reactive barrier systems are often heterogeneous material mixtures with relatively large contrasts in hydraulic conductivity and porosity over short distances. These short-range variations in material properties imply that preferential flow paths and diffusion between regions of higher and lower hydraulic conductivity may be important for treatment efficiency. This paper presents the results of a laboratory column experiment where denitrification is investigated using a heterogeneous reactive substrate (sawdust mixed with sewage sludge). Displacement experiments with a non-reactive solute at three different flow rates are used to estimate transport parameters using a dual porosity non-equilibrium model. Parameter estimation from breakthrough curves produced relatively consistent values for the fraction of the porosity consisting of mobile water (β) and the mass transfer coefficient (α), with average values of 0.27 and 0.42 d(-1), respectively. The column system removes >95% of the influent nitrate at low and medium flow, but only 50-75% of the influent nitrate at high flow, suggesting that denitrification kinetics and diffusive mass transfer rates are limiting the degree of treatment at lower hydraulic residence times. Reactive barrier systems containing dual porosity media must therefore consider mass transfer times in their design; this is often most easily accommodated by adjusting flowpath length.     

Geostatistical analysis of soil hydrologic properties in a field plot

A 91-m transect was set up in an irrigated field near Las Cruces, New Mexico to obtain soil water tension and water content data to investigate their spatial variability. A total of 455 sampling points were monitored along a grid consisting of 91 stations placed 1 m apart by 5 depths per station (at 0.3, 0.6, 0.9, 1.2 and 1.5 m below the surface). Post-irrigation tension and wetness measurements were recorded over 45 days at 11 time periods. Soil water tension was measured with tensiometers using a hand-held pressure transducer. A neutron probe was used to obtain volumetric water content. Using the observed wetness and tension data, unsaturated hydraulic conductivity values were derived (using a cubic spline function to estimate the gradient), and an exponential model was used to fit the calculated conductivity-tension curves to obtain hydraulic conductivity parameter values. The spatial and temporal variability of wetness, tension, saturated hydraulic conductivity and pore-size distribution parameters, and texture at the 0.3-m depth were examined using geostatistical techniques.The exponential model was found to inadequately describe the hydraulic conductivity/tension relationship for the full range of tension, particularly in the tension range near saturation. The derived values of the saturated hydraulic conductivity parameter were much greater than expected and do not correspond to reasonable saturated hydraulic conductivity values.All of the soil parameters studied exhibited large spatial variability horizontally and vertically in the field. Ranges of dependence determined from semivariogram analysis over the 44-day drainage period are 3–32 m for wetness, 6–34 m for soil water tension, 5–35 m for natural log of saturated hydraulic conductivity parameters, 5–11 m for pore-size distribution parameter, and 8–24 m for percent sand, silt and clay at the 0.3 m depth. An alternate hole-effect model is suggested to describe the texture semivariograms.It was determined that the variance of volumetric water content generally increased at each depth over the measured time periods, which is consistent with certain past field studies and a stochastic analysis of unsaturated flow in heterogeneous soils. Future research is recommended relating soil texture to soil hydrologic parameters with the goal of predicting soil behavior with less extensive sampling schemes.     

Applied tracer tests in fractured rock: Can we predict natural gradient solute transport more accurately than fracture and matrix parameters?

Applied tracer tests provide a means to estimate aquifer parameters in fractured rock. The traditional approach to analysing these tests has been using a single fracture model to find the parameter values that generate the best fit to the measured breakthrough curve. In many cases, the ultimate aim is to predict solute transport under the natural gradient. Usually, no confidence limits are placed on parameter values and the impact of parameter errors on predictions of solute transport is not discussed. The assumption inherent in this approach is that the parameters determined under forced conditions will enable prediction of solute transport under the natural gradient. This paper considers the parameter and prediction uncertainty that might arise from analysis of breakthrough curves obtained from forced gradient applied tracer tests. By adding noise to an exact solution for transport in a single fracture in a porous matrix we create multiple realisations of an initial breakthrough curve. A least squares fitting routine is used to obtain a fit to each realisation, yielding a range of parameter values rather than a single set of absolute values. The suite of parameters is then used to make predictions of solute transport under lower hydraulic gradients and the uncertainty of estimated parameters and subsequent predictions of solute transport is compared. The results of this study show that predictions of breakthrough curve characteristics (first inflection point time, peak arrival time and peak concentration) for groundwater flow speeds with orders of magnitude smaller than that at which a test is conducted can sometimes be determined even more accurately than the fracture and matrix parameters.     

Investigation of hydrogeologic processes in a dipping layer structure: 1. The flow barrier effect

Numerical simulations of a field experiment were performed in order to study the processes contributing to lateral diversion of water flow in the unsaturated zone. The experimental site is a glacial delta with dipping layers of alternating finer and coarser sand. Model soil physical parameters for each of the stratigraphic layers were estimated from soil grain size distributions. Anisotropy of the hydraulic conductivity within the layers was critical in order to reproduce the flow pattern observed in the experiment. Capillary and hydraulic barriers were of minor importance for the observed lateral diversion.     

Characterization of subsoil heterogeneity, estimation of grain size distribution and hydraulic conductivity at the Krauthausen test site using Cone Penetration Test

A Cone Penetration Test (CPT) survey with a high spatial resolution was performed in order to investigate the stratigraphy as well as the spatial variability of various soil properties of the Krauthausen test site. Analyses of the CPT measurements showed the subsurface to be dominated by a planar layered structure. Variogram analysis of the various CPT parameters disclosed that within each layer the soil properties have an anisotropic spatial correlation structure. A correlation analysis of the measured CPT data and co-located grain size distributions from soil samples was performed. Since the correlation coefficients were greater equal to 0.7, a reliable empirical relationship between the data sets could be developed. Based on this empirical relationship grain size distributions were estimated at CPT locations. The statistical processing of estimated and measured grain size distributions with respect to their spatial correlation structure disclosed good agreement between the data sets. The estimated grain size distributions from CPT data were used to estimate the hydraulic conductivity in the aquifer. The results provide detailed information of the spatial heterogeneity of the hydraulic conductivity at Krauthausen test site. The validation of these results, using a prior investigation of hydraulic conductivity statistics, suggests the CPT a fast and inexpensive tool for the estimation of three dimensional hydraulic conductivity fields with sufficient accuracy.     

Incorporating layer- and local-scale heterogeneities in numerical simulation of unsaturated flow and tracer transport

This study characterizes layer- and local-scale heterogeneities in hydraulic parameters (i.e., matrix permeability and porosity) and investigates the relative effect of layer- and local-scale heterogeneities on the uncertainty assessment of unsaturated flow and tracer transport in the unsaturated zone of Yucca Mountain, USA. The layer-scale heterogeneity is specific to hydrogeologic layers with layerwise properties, while the local-scale heterogeneity refers to the spatial variation of hydraulic properties within a layer. A Monte Carlo method is used to estimate mean, variance, and 5th, and 95th percentiles for the quantities of interest (e.g., matrix saturation and normalized cumulative mass arrival). Model simulations of unsaturated flow are evaluated by comparing the simulated and observed matrix saturations. Local-scale heterogeneity is examined by comparing the results of this study with those of the previous study that only considers layer-scale heterogeneity. We find that local-scale heterogeneity significantly increases predictive uncertainty in the percolation fluxes and tracer plumes, whereas the mean predictions are only slightly affected by the local-scale heterogeneity. The mean travel time of the conservative and reactive tracers to the water table in the early stage increases significantly due to the local-scale heterogeneity, while the influence of local-scale heterogeneity on travel time gradually decreases over time. Layer-scale heterogeneity is more important than local-scale heterogeneity for simulating overall tracer travel time, suggesting that it would be more cost-effective to reduce the layer-scale parameter uncertainty in order to reduce predictive uncertainty in tracer transport.     

Contaminant transport in a municipal drinking water supply: a steady-state approach to estimate rate and uncertainty

Contaminant transport is generally considered to be a key factor when assessing and classifying the environmental risk of polluted areas. In the study presented here, a steady-state approach was applied to obtain estimates of the transit time and concentration of the pesticide metabolite BAM (2,6-dichlorobenzoamide) at a site where it is contaminating a municipal drinking water supply. A Monte Carlo simulation technique was used to quantify the uncertainty of the results and to evaluate the sensitivity of the used parameters. The adopted approach yielded an estimated median transit time of 10 y for the BAM transport from the polluted site to the water supply. Soil organic carbon content in the unsaturated zone and the hydraulic conductivity in the saturated zone explained 44% and 23% of the uncertainty in the transit time estimate, respectively. The sensitivity analysis showed that the dilution factor due to regional groundwater flow and the soil organic carbon content at the polluted site explained 53% and 31% of the uncertainty of concentration estimates, respectively. In conclusion, the adopted steady-state approach can be used to obtain reliable first estimates of transit time and concentration, but to improve concentration predictions of degrading contaminants, a dynamic model is probably required.     

Modelling the migration of contaminants through variably saturated dual-porosity, dual-permeability chalk

In the Hesbaye region in Belgium, tracer tests performed in variably saturated fissured chalk rocks presented very contrasting results in terms of transit times, according to artificially controlled water recharge conditions prevailing during the experiments. Under intense recharge conditions, tracers migrated across the partially or fully saturated fissure network, at high velocity in accordance with the high hydraulic conductivity and low effective porosity (fracture porosity). At the same time, a portion of the tracer was temporarily retarded in the almost immobile water located in the matrix. Under natural infiltration conditions, the fissure network remained inactive. Tracers migrated downward through the matrix, at low velocity in relation with the low hydraulic conductivity and the large porosity of the matrix. Based on these observations, Brouyère et al. (2004a) [Brouyère, S., Dassargues, A., Hallet, V., 2004a. Migration of contaminants through the unsaturated zone overlying the Hesbaye chalky aquifer in Belgium: a field investigation, J. Contam. Hydrol., 72 (1-4), 135-164, doi: 10.1016/j.conhyd.2003.10.009] proposed a conceptual model in order to explain the migration of solutes in variably saturated, dual-porosity, dual-permeability chalk. Here, mathematical and numerical modelling of tracer and contaminant migration in variably saturated fissured chalk is presented, considering the aforementioned conceptual model. A new mathematical formulation is proposed to represent the unsaturated properties of the fissured chalk in a more dynamic and appropriate way. At the same time, the rock water content is partitioned between mobile and immobile water phases, as a function of the water saturation of the chalk rock. The groundwater flow and contaminant transport in the variably saturated chalk is solved using the control volume finite element method. Modelling the field tracer experiments performed in the variably saturated chalk shows the adequacy and usefulness of the new conceptual, mathematical and numerical model.     

3-D numerical evaluation of density effects on tracer tests

In this paper we present numerical simulations carried out to assess the importance of density-dependent flow on tracer plume development. The scenario considered in the study is characterized by a short-term tracer injection phase into a fully penetrating well and a natural hydraulic gradient. The scenario is thought to be typical for tracer tests conducted in the field. Using a reference case as a starting point, different model parameters were changed in order to determine their importance to density effects. The study is based on a three-dimensional model domain. Results were interpreted using concentration contours and a first moment analysis. Tracer injections of 0.036 kg per meter of saturated aquifer thickness do not cause significant density effects assuming hydraulic gradients of at least 0.1%. Higher tracer input masses, as used for geoelectrical investigations, may lead to buoyancy-induced flow in the early phase of a tracer test which in turn impacts further plume development. This also holds true for shallow aquifers. Results of simulations with different tracer injection rates and durations imply that the tracer input scenario has a negligible effect on density flow. Employing model cases with different realizations of a log conductivity random field, it could be shown that small variations of hydraulic conductivity in the vicinity of the tracer injection well have a major control on the local tracer distribution but do not mask effects of buoyancy-induced flow.     

Evaluation of groundwater flow patterns around a dual-screened groundwater circulation well

Dual-screened groundwater circulation wells (GCWs) can be used to remove contaminant mass and to mix reagents in situ. GCWs are so named because they force water in a circular pattern between injection and extraction screens. The radial extent, flux and direction of the effective flow of this circulation cell are difficult to measure or predict. The objective of this study is to develop a robust protocol for assessing GCW performance. To accomplish this, groundwater flow patterns surrounding a GCW are assessed using a suite of tools and data, including: hydraulic head, in situ flow velocity, measured hydraulic conductivity data from core samples, chemical tracer tests, contaminant distribution data, and numerical flow and transport models. The hydraulic head data show patterns that are consistent with pumping on a dual-screened well, however, many of the observed changes are smaller than expected. In situ thermal perturbation flow sensors successfully measured horizontal flow, but vertical flow could not be determined with sufficient accuracy to be useful in mapping flow patterns. Two types of chemical tracer tests were utilized at the site and showed that much of the flow occurs within a few meters of the GCW. Flow patterns were also assessed based on changes in contaminant (trichloroethylene, TCE) concentrations over time. The TCE data clearly showed treated water moving away from the GCW at shallow and intermediate depths, but the circulation of that water back to the well, except very close to the well, was less clear. Detailed vertical and horizontal hydraulic conductivities were measured on 0.3 m-long sections from a continuous core from the GCW installation borehole. The measured vertical and horizontal hydraulic conductivity data were used to construct numerical flow and transport models, the results of which were compared to the head, velocity and concentration data. Taken together, the field data and modeling present a fairly consistent picture of flow and transport around the GCW. However, the time and expense associated with conducting all of those tests would be prohibitive for most sites. As a consequence, a sequential protocol for GCW characterization is presented here in which the number of tools used can be adjusted to meet the needs of individual sites. While not perfect, we believe that this approach represents the most efficient means for evaluating GCW performance.     

3-D numerical evaluation of density effects on tracer tests

In this paper we present numerical simulations carried out to assess the importance of density-dependent flow on tracer plume development. The scenario considered in the study is characterized by a short-term tracer injection phase into a fully penetrating well and a natural hydraulic gradient. The scenario is thought to be typical for tracer tests conducted in the field. Using a reference case as a starting point, different model parameters were changed in order to determine their importance to density effects. The study is based on a three-dimensional model domain. Results were interpreted using concentration contours and a first moment analysis. Tracer injections of 0.036 kg per meter of saturated aquifer thickness do not cause significant density effects assuming hydraulic gradients of at least 0.1%. Higher tracer input masses, as used for geoelectrical investigations, may lead to buoyancy-induced flow in the early phase of a tracer test which in turn impacts further plume development. This also holds true for shallow aquifers. Results of simulations with different tracer injection rates and durations imply that the tracer input scenario has a negligible effect on density flow. Employing model cases with different realizations of a log conductivity random field, it could be shown that small variations of hydraulic conductivity in the vicinity of the tracer injection well have a major control on the local tracer distribution but do not mask effects of buoyancy-induced flow.     

Effects of biofilm growth on flow and transport through a glass parallel plate fracture

The effects of biofilm growth on flow and solute transport through a sandblasted glass parallel plate fracture was investigated. The fracture was inoculated using soil microorganisms. Glucose, oxygen and other nutrients were supplied to support growth. The biomass initially formed discrete clusters attached to the glass surfaces, but over time formed a continuous biofilm. From dye tracer tests conducted during biofilm growth, it was observed that channels and low-permeability zones dominated transport. The hydraulic conductivity of the fracture showed a sigmoidal decrease with time. The hydraulic conductivity was reduced by a factor of 0.033, from 18 to 0.6 cm/s, corresponding to a 72% decrease in the hydraulic aperture, from 500 to 140 microm. In contrast, the mass balance aperture, determined from fluoride tracer tests, remained relatively constant, indicating that the impact of biomass growth on effective fracture porosity was much less than the effect on hydraulic conductivity. Analyses of pre-biofilm tracer tests revealed that both Taylor dispersion and macrodispersion were influencing transport. During biofilm growth, only macrodispersion was dominant. The macrodispersion coefficient alpha(macro) was found to increase logarithmically with hydraulic conductivity reduction.     

Design of aquifer remediation systems: (1) describing hydraulic structure and NAPL architecture using tracers

Aquifer heterogeneity (structure) and NAPL distribution (architecture) are described based on tracer data. An inverse modelling approach that estimates the hydraulic structure and NAPL architecture based on a Lagrangian stochastic model where the hydraulic structure is described by one or more populations of lognormally distributed travel times and the NAPL architecture is selected from eight possible assumed distributions. Optimization of the model parameters for each tested realization is based on the minimization of the sum of the square residuals between the log of measured tracer data and model predictions for the same temporal observation. For a given NAPL architecture the error is reduced with each added population. Model selection was based on a fitness which penalized models for increasing complexity. The technique is demonstrated under a range of hydrologic and contaminant settings using data from three small field-scale tracer tests: the first implementation at an LNAPL site using a line-drive flow pattern, the second at a DNAPL site with an inverted five-spot flow pattern, and the third at the same DNAPL site using a vertical circulation flow pattern. The Lagrangian model was capable of accurately duplicating experimentally derived tracer breakthrough curves, with a correlation coefficient of 0.97 or better. Furthermore, the model estimate of the NAPL volume is similar to the estimates based on moment analysis of field data.     

Potential for CSO treatment with horizontal flow constructed wetlands: influence of hydraulic load,plant presence and loading frequency

This study aimed at analysing the performance of horizontal subsurface flow constructed wetlands (CWs) to treat combined sewer overflow (CSO). Four horizontal subsurface flow CWs, organized in two groups (A and B) each with a planted (Phragmites australis) and a non-planted bed, were loaded with simulated CSO, with group B receiving twice the hydraulic load of group A. Beds were monitored for pH, dissolved oxygen, conductivity, redox potential, chemical oxygen demand (COD), total suspended solids (TSS) and enterococci. Porosity variations were also estimated. Monitoring was conducted during spring and wintertime, with regular and irregular loading frequencies. Results showed an average treatment efficiency of 90–100 % for TSS, 60–90 % for COD and 2–6 log for enterococci. Removal rates were especially relevant in the first 24 h for COD and TSS. TSS and enterococci removal did not exhibit the influence of macrophytes or the applied hydraulic load while COD’s removal efficiency was lower in the higher load group and in planted beds.

     

Modelling tritium and phosphorus transport by preferential flow in structured soil

Subsurface solute transport through structured soil is studied by model interpretation of experimental breakthrough curves from tritium and phosphorus tracer tests in three intact soil monoliths. Similar geochemical conditions, with nearly neutral pH, were maintained in all the experiments. Observed transport differences for the same tracer are thus mainly due to differences in the physical transport process between the different monoliths. The modelling is based on a probabilistic Lagrangian approach that decouples physical and chemical mass transfer and transformation processes from pure and stochastic advection. Thereby, it enables explicit quantification of the physical transport process through preferential flow paths, honouring all independently available experimental information. Modelling of the tritium breakthrough curves yields a probability density function of non-reactive solute travel time that is coupled with a reaction model for linear, non-equilibrium sorption–desorption to describe the phosphorus transport. The tritium model results indicate that significant preferential flow occurs in all the experimental soil monoliths, ranging from 60–100% of the total water flow moving through only 25–40% of the total water content. In agreement with the fact that geochemical conditions were similar in all experiments, phosphorus model results yield consistent first-order kinetic parameter values for the sorption–desorption process in two of the three soil monoliths; phosphorus transport through the third monolith cannot be modelled because the apparent mean transport rate of phosphorus is anomalously rapid relative to the non-adsorptive tritium transport. The occurrence of preferential flow alters the whole shape of the phosphorus breakthrough curve, not least the peak mass flux and concentration values, and increases the transported phosphorus mass by 2–3 times relative to the estimated mass transport without preferential flow in the two modelled monoliths.     

Monte Carlo analysis of field water flow comparing uni- and bimodal effective hydraulic parameters for structured soil

Soil structure critically affects the hydrological behaviour of soils. In this paper, we examined the impact of areal heterogeneity of hydraulic properties of a structured soil on soil ensemble behaviour for various soil water flow processes with different top boundary conditions (redistribution and drainage plus evaporation and infiltration). Using a numerical solution of the Richards' equation in a stochastic framework, the ensemble characteristics and flow dynamics were studied for drying and wetting processes observed during a time interval of ten days when a series of relatively intense rainfall events occurred. The effects of using unimodal and bimodal interpretative models of hydraulic properties on the ensemble hydrological behaviour of the soil were illustrated by comparing predictions to mean water contents measured over time in several sites at field scale. Although the differences between unimodal and bimodal fitting are not significant in terms of goodness of fit, the differences in process predictions are considerable with the bimodal soil simulating water content measurements much better than unimodal soil. We also investigated the relative contribution of the soil variability of each parameter on the variance of the water contents obtained as the main output of the stochastic simulations. The variability of the structural parameter, weighting the two pore space fractions in the bimodal interpretative model, has the largest contribution to water content variance. The contribution of each parameter depends only partly on the coefficient of variation, much more on the sensitivity of the model to the parameters and on the flow process being observed. We observed that the contribution of the retention parameters to uncertainty increases during drainage processes; the opposite occurs with the hydraulic conductivity parameters.     

Modeling porosity reductions caused by mineral fouling in continuous-wall permeable reactive barriers

A study was conducted to assess key factors to include when modeling porosity reductions caused by mineral fouling in permeable reactive barriers (PRBs) containing granular zero valent iron. The public domain codes MODFLOW and RT3D were used and a geochemical algorithm was developed for RT3D to simulate geochemical reactions occurring in PRBs. Results of simulations conducted with the model show that the largest porosity reductions occur between the entrance and mid-plane of the PRB as a result of precipitation of carbonate minerals and that smaller porosity reductions occur between the mid-plane and exit face due to precipitation of ferrous hydroxide. These findings are consistent with field and laboratory observations, as well as modeling predictions made by others. Parametric studies were conducted to identify the most important variables to include in a model evaluating porosity reduction. These studies showed that three minerals (CaCO3, FeCO3, and Fe(OH)2 (am)) account for more than 99% of the porosity reductions that were predicted. The porosity reduction is sensitive to influent concentrations of HCO3-, Ca2+, CO3(2-), and dissolved oxygen, the anaerobic iron corrosion rate, and the rates of CaCO3 and FeCO3 formation. The predictions also show that porosity reductions in PRBs can be spatially variable and mineral forming ions penetrate deeper into the PRB as a result of flow heterogeneities, which reflects the balance between the rate of mass transport and geochemical reaction rates. Level of aquifer heterogeneity and the contrast in hydraulic conductivity between the aquifer and PRB are the most important hydraulic variables affecting porosity reduction. Spatial continuity of aquifer hydraulic conductivity is less significant.     

Estimation of local scale dispersion from local breakthrough curves during a tracer test in a heterogeneous aquifer: the Lagrangian approach

The local scale dispersion tensor, Dd, is a controlling parameter for the dilution of concentrations in a solute plume that is displaced by groundwater flow in a heterogeneous aquifer. In this paper, we estimate the local scale dispersion from time series or breakthrough curves, BTCs, of Br concentrations that were measured at several points in a fluvial aquifer during a natural gradient tracer test at Krauthausen. Locally measured BTCs were characterized by equivalent convection dispersion parameters: equivalent velocity, v(eq)(x) and expected equivalent dispersivity, [lambda(eq)(x)]. A Lagrangian framework was used to approximately predict these equivalent parameters in terms of the spatial covariance of log(e) transformed conductivity and the local scale dispersion coefficient. The approximate Lagrangian theory illustrates that [lambda(eq)(x)] increases with increasing travel distance and is much larger than the local scale dispersivity, lambda(d). A sensitivity analysis indicates that [lambda(eq)(x)] is predominantly determined by the transverse component of the local scale dispersion and by the correlation scale of the hydraulic conductivity in the transverse to flow direction whereas it is relatively insensitive to the longitudinal component of the local scale dispersion. By comparing predicted [lambda(eq)(x)] for a range of Dd values with [lambda(eq)(x)] obtained from locally measured BTCs, the transverse component of Dd, DdT, was estimated. The estimated transverse local scale dispersivity, lambda(dT) = DdT/U1 (U1 = mean advection velocity) is in the order of 10(1)-10(2) mm, which is relatively large but realistic for the fluvial gravel sediments at Krauthausen.