A discrete particle representation of hillslope hydrology: hypothesis testing in reproducing a tracer experiment at Gårdsjön,Sweden

Despite the long history of the continuum equation approach in hydrology, it is not a necessary approach to the formulation of a physically based representation of hillslope hydrology. The Multiple Interacting Pathways (MIPs) model is a discrete realization that allows hillslope response and transport to be simultaneously explored in a way that reflects the potential occurrence of preferential flows and lengths of pathways. The MIPs model uses random particle tracking methods to represent the flow of water within the subsurface alongside velocity distributions that acknowledge preferential flows and transition probability matrices, which control flow pathways. An initial realization of this model is presented here in application to a tracer experiment carried out in Gårdsjön, Sweden. The model is used as an exploratory tool, testing several hypotheses in relation to this experiment. Copyright © 2011 John Wiley & Sons, Ltd.     

Diffusive mass exchange of non‐reactive substances in dual‐porosity porous systems – column experiments under saturated conditions

Diffusive mass exchange into immobile water regions within heterogeneous porous aquifers influences the fate of solutes. The percentage of immobile water is often unidentified in natural aquifers though. Hence, the mathematical prediction of solute transport in such heterogeneous aquifers remains challenging. The objective of this study was to find a simple analytical model approach that allows quantifying properties of mobile and immobile water regions and the portion of immobile water in a porous system. Therefore, the Single Fissure Dispersion Model (SFDM), which takes into account diffusive mass exchange between mobile and immobile water zones, was applied to model transport in well‐defined saturated dual‐porosity column experiments. Direct and indirect model validation was performed by running experiments at different flow velocities and using conservative tracer with different molecular diffusion coefficients. In another column setup, immobile water regions were randomly distributed to test the model applicability and to determine the portion of immobile water. In all setups, the tracer concentration curves showed differences in normalized maximum peak concentration, tailing and mass recovery according to their diffusion coefficients. These findings were more pronounced at lower flow rates (larger flow times) indicating the dependency of diffusive mass exchange into immobile water regions on tracers' molecular diffusion coefficients. The SFDM simulated all data with high model efficiency. Successful model validation supported the physical meaning of fitted model parameters. This study showed that the SFDM, developed for fissured aquifers, is applicable in porous media and can be used to determine porosity and volume of regions with immobile water. Copyright © 2015 John Wiley & Sons, Ltd.     

Using high resolution tracer data to constrain water storage,flux and age estimates in a spatially distributed rainfall‐runoff model

Models simulating stream flow and conservative tracers can provide a representation of flow paths, storage distributions and mixing processes that is advantageous for many predictive purposes. Compared with models that only simulate stream flow, tracer data can be used to investigate the internal consistency of model behaviour and to gain insight into model performance. Here, we examine the strengths and weaknesses of a data‐driven, spatially distributed tracer‐aided rainfall‐runoff model. The model structure allowed us to assess the influence of landscape characteristics on the routing and mixing of water and tracers. The model was applied to a site in the Scottish Highlands with a unique tracer data set; ~4 years of daily isotope ratios in stream water and precipitation were available, as well as 2 years of weekly soil and ground water isotopes. The model structure was based on an empirically based, lumped tracer‐aided model previously developed for the catchment. The best model runs were selected from Monte Carlo simulations based on dual calibration criteria using objective functions for both stream isotopes and discharge at the outlet. Model performance for these criteria was reasonable (Nash–Sutcliffe efficiencies for discharge and isotope ratios were ~0.4–0.6). The model could generally reproduce the variable isotope signals in the soils of the steeper hill slopes where storage was low, and damped isotope responses in valley bottom cells with high storage. The model also allowed us to estimate the age distributions of internal stores, water fluxes and stream flow. Average stream water age was ~1.6 years, integrating older groundwater in the valley bottom and dynamic younger soil waters. By tracking water ages and simulating isotopes, the model captured the changes in connectivity driven by distributed storage dynamics. This has substantially improved the representation of spatio‐temporal process dynamics and gives a more robust framework for projecting environmental change impacts. Copyright © 2016 The Authors Hydrological Processes Published by John Wiley & Sons Ltd.     

Convergent Radial Tracing of Viral and Solute Transport in Gneiss Saprolite

Deeply weathered crystalline rock aquifer systems comprising unconsolidated saprolite and underlying fractured bedrock (saprock) underlie 40% of sub-Saharan Africa. The vulnerability of this aquifer system to contamination, particularly in rapidly urbanizing areas, remains poorly understood. In order to assess solute and viral transport in saprolite derived from Precambrian gneiss, forced-gradient tracer experiments using chloride and Escherichia coli phage ΦX174 were conducted in southeastern Uganda. The bacteriophage tracer was largely unrecovered; adsorption to the weathered crystalline rock matrix is inferred and enabled by the low pH (5.7) of site ground water and the bacteriophage's relatively high isoelectric point (pI = 6.6). Detection of the applied ΦX174 phage in the pumping well discharge at early times during the experiment traces showed, however, that average ground water flow velocities exceed that of the inert solute tracer, chloride. This latter finding is consistent with observations in other hydrogeological environments where statistically extreme sets of microscopic flow velocities are considered to transport low numbers of fecal pathogens and their proxies along a selected range of linked ground water pathways. Application of a radial advection-dispersion model with an exponentially decaying source term to the recovered chloride tracer estimates a dispersivity (α) of 0.8 ± 0.1 m over a distance of 4.15 m. Specific yield (S_y) is estimated to be 0.02 from volume balance calculations based on tracer experiments. As single-site observations, our estimates of saprolite S_y and α are tentative but provide a starting point for assessing the vulnerability of saprolite aquifers in sub-Saharan Africa to contamination and estimating quantitatively the impact of climate and abstraction on ground water storage.     

Integrating tracer experiments with modeling to assess runoff processes and water transit times

Representing runoff process complexity in a simple model structure remains a challenge in hydrology. We present an integrated approach to investigate runoff processes using a hillslope tracer experiment and modeling exercise to explore model parameterization, process representation, and transit times. A spatially-explicit model constrained by soil hydrologic properties, runoff, and applied tracer data was used to identify the dominant processes necessary to explain both water and solute flux from a steep hillslope. The tracer data allowed for the rejection of model parameter sets based on the calibration to runoff data alone, thus reducing model uncertainty. The additional calibration to tracer data, improved parameter identifiability and provided further insight to process controls on hillslope-scale water and solute flux. Transit time distributions developed using the model provided further insight to model structure such as subsurface volume, mixing assumptions, and the water table dynamics. Combining field experiments with the modeling exercise may lead to a more comprehensive assessment of runoff process representation in models.     

Mechanisms and pathways of lateral flow on aspen‐forested,Luvisolic soils,Western Boreal Plains,Alberta, Canada

Rainfall simulation experiments by Redding and Devito ( 2008 , Hydrological Processes 23: 4287–4300) on two adjacent plots of contrasting antecedent soil moisture storage on an aspen‐forested hillslope on the Boreal Plain showed that lateral flow generation occurred only once large soil storage capacity was saturated combined with a minimum event precipitation of 15–20 mm. This paper extends the results of Redding and Devito ( 2008 , Hydrological Processes 23: 4287–4300) with detailed analysis of pore pressure, soil moisture and tracer data from the rainfall simulation experiments, which is used to identify lateral flow generation mechanisms and flow pathways. Lateral flow was not generated until soils were wet into the fine textured C horizon. Lateral flow occurred dominantly through the clay‐rich Bt horizon by way of root channels. Lateral flow during the largest event was dominated by event water, and precipitation intensity was critical in lateral flow generation. Lateral flow was initiated as preferential flow near the soil surface into root channels, followed by development of a perched water table at depth, which also interacted with preferential flow pathways to move water laterally by the transmissivity feedback mechanism. The results indicate that lateral flow generated by rainfall on these hillslopes is uncommon because of the generally high available soil moisture storage capacity and the low probability of rainfall events of sufficient magnitude and intensity. Copyright © 2010 John Wiley & Sons, Ltd.     

Using a thermal tracer to estimate overland and rill flow velocities

Flow velocity is a basic hydraulic property of surface flows and its precise calculation is necessary for process based hydrological models, such as soil erosion and rill development models, as well as for modelling sediment and solute transport by runoff. This study presents a technique based on infrared thermography to visualize very shallow flows and allow a quantitative measurement of overland flow and rill flow velocities. Laboratory experiments were conducted to compare the traditional dye tracer technique with this new thermal tracer technique by injecting a combined tracer (heated dye) into shallow flowing surface water. The leading edge tracer velocities estimated by means of infrared video and by the usual real imaging video were compared. The results show that thermal tracers can be used to estimate both overland and rill flow velocities, since measurements are similar to those resulting from using dye tracers. The main advantage of using thermography was the higher visibility of the leading edge of the injected tracer compared with the real image videos. Copyright © 2013 John Wiley & Sons, Ltd.     

Multitracer test approach to characterize reactive transport in karst aquifers

A method to estimate reactive transport parameters as well as geometric conduit parameters from a multitracer test in a karst aquifer is provided. For this purpose, a calibration strategy was developed applying the two-region nonequilibrium model CXTFIT. The ambiguity of the model calibration was reduced by first calibrating the model with respect to conservative tracer breakthrough and later transferring conservative transport parameters to the reactive model calibration. The reactive transport parameters were only allowed to be within a defined sensible range to get reasonable calibration values. This calibration strategy was applied to breakthrough curves obtained from a large-scale multitracer test, which was performed in a karst aquifer of the Swabian Alb, Germany. The multitracer test was conducted by the simultaneous injection of uranine, sulforhodamine G, and tinopal CBS-X. The model succeeds to represent the tracer breakthrough curves (TBCs) of uranine and sulforhodamine G and verifies that tracer-rock interactions preferably occur in the immobile fluid region, although the fraction of this region amounts to only 3.5% of the total water. However, the model failed to account for the long tailing observed in the TBC of tinopal CBS-X. Sensitivity analyses reveal that model results for the conservative tracer transport are most sensitive to average velocity and volume fraction of the mobile fluid region, while dispersion and mass transfer coefficients are least influential. Consequently, reactive tracer calibration allows the determination of sorption sites in the mobile and immobile fluid region at small retardation coefficients.     

An integral-balance nonlinear model to simulate changes in soil moisture,groundwater and surface runoff dynamics at the hillslope scale

We present a system of ordinary differential equations (ODEs) capable of reproducing simultaneously the aggregated behavior of changes in water storage in the hillslope surface, the unsaturated and the saturated soil layers and the channel that drains the hillslope. The system of equations can be viewed as a two-state integral-balance model for soil moisture and groundwater dynamics. Development of the model was motivated by the need for landscape representation through hillslopes and channels organized following stream drainage network topology. Such a representation, with the basic discretization unit of a hillslope, allows ODEs-based simulation of the water transport in a basin. This, in turn, admits the use of highly efficient numerical solvers that enable space–time scaling studies. The goal of this paper is to investigate whether a nonlinear ODE system can effectively replicate observations of water storage in the unsaturated and saturated layers of the soil. Our first finding is that a previously proposed ODE hillslope model, based on readily available data, is capable of reproducing streamflow fluctuations but fails to reproduce the interactions between the surface and subsurface components at the hillslope scale. However, the more complex ODE model that we present in this paper achieves this goal. In our model, fluxes in the soil are described using a Taylor expansion of the underlying storage flux relationship. We tested the model using data collected in the Shale Hills watershed, a 7.9-ha forested site in central Pennsylvania, during an artificial drainage experiment in August 1974 where soil moisture in the unsaturated zone, groundwater dynamics and surface runoff were monitored. The ODE model can be used as an alternative to spatially explicit hillslope models, based on systems of partial differential equations, which require more computational power to resolve fluxes at the hillslope scale. Therefore, it is appropriate to be coupled to runoff routing models to investigate the effect of runoff and its uncertainty propagation across scales. However, this improved performance comes at the expense of introducing two additional parameters that have no obvious physical interpretation. We discuss the implications of this for hydrologic studies across scales.     

Quantification of vertical water fluxes in the vadose zone using particle‐size distribution and pedology‐based approaches to model soil heterogeneities

One‐dimensional flow simulations were conducted at four locations of the shallow alluvial aquifer of the upper Rhine River (at the Erstein polder) to quantify the time‐dependent moisture distribution, the water flux and the water volume infiltrated in the unsaturated zone as a function of soil heterogeneities during a five‐day‐long flooding event. Three methods of estimating the hydraulic parameters of soil in the vadose zone were tested. They are based on the following: (1) experimental data, (2) soil particle‐size distribution and (3) pedology information on soils. Water fluxes calculated from modelling approaches 2 and 3 were compared with those of the experiment‐based values and the effect of these differences on the arrival time and velocity of water at the water table were analysed. Major differences in water fluxes were found among the methods of estimating the hydrodynamic parameters. At the Terrace location, the groundwater recharge predicted using soil data from methods 1 and 2 are approximately 4500 and 2400 mm, respectively. Flow simulations using soil data and the experiment‐based method show the highest velocities of infiltrating water at the soil surface and largest volume of groundwater infiltration but result in the lowest centres of the moisture content mass. The results obtained using soil data based on the pedological method are similar to those calculated using soil parameters based on the particle‐size distribution of extracted soil samples. Water pressure profiles calculated on Terrace and Channel location, 3 and 7 days after the inundation event agreed reasonably well with those observed when using hydrodynamic parameters from the experiment‐based method. However, the flow model using the pedology‐based parameters largely underestimates the time needed to achieve hydrostatic conditions of the soil water profile once water flooding at the soil surface stops. This can be mainly attributed to the low values of estimated van Genuchten parameter α. Copyright © 2012 John Wiley & Sons, Ltd.     

Tracer‐based evidence of heterogeneity in subsurface flow and storage within a boreal hillslope

Runoff from boreal hillslopes is often affected by distinct soil boundaries, including the frozen boundary and the organic‐mineral boundary (OMB), where highly porous and hydraulically conductive organic material overlies fine‐grained mineral soils. Viewed from the surface, ground cover appears as a patchwork on sub‐meter scales, with thick, moss mats interspersed with lichen‐covered, silty soils with gravel inclusions. We conducted a decameter‐scale subsurface tracer test on a boreal forest hillslope in interior Alaska to quantify locations and mechanisms of transport and storage in these soils, focusing on the OMB. A sodium bromide tracer was added as a slug addition to a pit and sampled at 40 down‐gradient wells, screened primarily at the OMB and within a 7 × 12 m well field. We maintained an elevated head in the injection pit for 8.5 hr to simulate a storm. Tracer breakthrough velocities ranged from <0.12 to 0.93 m hr^?1, with the highest velocities in lichen‐covered soils. After 12 hr and cessation of the elevated head, the tracer coalesced and was only detected in thick mosses at a trough in the OMB. By 24 hr, approximately 17% of the tracer mass could be accounted for. The majority of the mass loss occurred between 4 and 12 hr, while the tracer was in contact with lichen‐covered soils, which is consistent with tracer transport into deeper flow paths via preferential flow through discrete gravelly areas. Slow breakthroughs suggest that storage and exchange also occurred in shallow soils, likely related to saturation and drainage in fine‐grained mineral soils caused by the elevated hydraulic head. These findings highlight the complex nature of storage and transmission of water and solutes from boreal hillslopes to streams and are particularly relevant given rapid changes to boreal environments related to climate change, thawing permafrost and increasing fire severity.     

Soil pipe flow tracer experiments: 2. Application of a streamflow transient storage zone model

Soil pipes are important subsurface flow pathways in many soil erosion phenomena. However, limited research has been performed on quantifying and characterizing their flow and transport characteristics. The objectives of this research were to determine the applicability of a streamflow model with transient storage in deriving flow and transport characteristics of soil pipes. Tracer data from pulse inputs were collected in four different soil pipes after a fluorescein dye was injected in the upstream end of each soil pipe network in three branches (west, middle, and east) of a main catchment and a back catchment in Goodwin Creek Experimental Watershed in Mississippi. Multiple sampling stations were positioned along each soil pipe network. The transient storage zone model OTIS‐P was executed inversely to estimate transport parameters by soil pipe reach such as the soil pipe cross‐sectional area (A), soil storage zone cross‐sectional area (A_s), and exchange rate between the soil pipe and the soil storage zone (α_s). Model convergence was achieved, and simulated breakthrough curves of the reaches were in good agreement with actual tracer data for eight of the nine reaches of the three branches of the Main Catchment and five of the seven reaches of the Back Catchment soil pipe. Simulation parameters for the soil pipe networks were similar to the range of values reported for flow and transport characteristics commonly observed in streams. Inversely, estimated soil pipe flow velocities were higher with increased tortuosity, which led to a smaller cross‐sectional areas predicted for the soil pipe flowpaths, while other parameters were not sensitive to tortuosity. In general, application of One‐Dimensional Transport with Inflow and Storage‐P to this unique soil pipe condition suggested larger transient storage (A_s and α_s) compared with most stream systems. This was hypothesized to be because of relatively higher ratio of the wetted perimeter to flow area in the soil pipe, the hydraulic roughness of the soil pipe, potential retention in collapsed portions of the pipe, and interaction with smaller preferential flow systems. Copyright © 2015 John Wiley & Sons, Ltd.     

Changes in water table level influence solute transport in uniform porous media

Changes in the water table level result in variable water saturation and variable hydrological fluxes at the interface between the unsaturated and saturated zone. This may influence the transport and fate of contaminants in the subsurface. The objective of this study was to examine the impact of a decreasing and an increasing water table on solute transport. We conducted tracer experiments at downward flow conditions in laboratory columns filled with two different uniform porous media under static and transient flow conditions either increasing or decreasing the water table. Tracer breakthrough curves were simulated using a mobile–immobile transport model. The resulting transport parameters were compared to identify dominant transport processes. Changes in the water table level affected dispersivities and mobile water fractions depending on the direction of water table movement and the grain size of the porous media. In fine glass beads, the water flow velocity was similar to the decline rate of the water table, and the mobile water fraction was decreased compared with steady‐state saturated conditions. However, immobile water was negligible. In coarse glass beads, water flow was faster because of fingered flow in the unsaturated part, and the mobile water fraction was smaller than in the fine material. Here, a rising water table led to an even smaller mobile water fraction and increased solute spreading because of diffusive interaction with immobile water. We conclude that changes of the water table need to be considered to correctly simulate transport in the subsurface at the transition of the unsaturated–saturated zone. Copyright © 2014 John Wiley & Sons, Ltd.     

Upscaling Point Velocity Measurements to Characterize a Glacial Outwash Aquifer

Small‐scale point velocity probe (PVP)‐derived velocities were compared to conventional large‐scale velocity estimates from Darcy calculations and tracer tests, and the possibility of upscaling PVP data to match the other velocity estimates was evaluated. Hydraulic conductivity was estimated from grain‐size data derived from cores, and single‐well response testing or slug tests of onsite wells. Horizontal hydraulic gradients were calculated using 3‐point estimators from all of the wells within an extensive monitoring network, as well as by representing the water table as a single best fit plane through the entire network. Velocities determined from PVP testing were generally consistent in magnitude with those from depth specific data collected from multilevel monitoring locations in the tracer test, and similar in horizontal flow direction to the average hydraulic gradient. However, scaling up velocity estimates based on PVP measurements for comparison with site‐wide Darcy‐based velocities revealed issues that challenge the use of Darcy calculations as a generally applicable standard for comparison. The Darcy calculations were shown to underestimate the groundwater velocities determined both by the PVPs and large‐scale tracer testing, in a depth‐specific sense and as a site‐wide average. Some of this discrepancy is attributable to the selective placement of the PVPs in the aquifer. Nevertheless, this result has important implications for the design of in situ treatment systems. It is concluded that Darcy estimations of velocity should be supplemented with independent assessments for these kinds of applications.     

Estimation of water saturation dependence of dispersion in unsaturated porous media: experiments and modelling analysis

In the dispersion theory, a linear relationship has been verified between the coefficient of hydrodynamic dispersion and water velocity, both in saturated and in unsaturated porous media. But for unsaturated soils the variability of flow directions and microscopic velocities can be larger than in saturated soils because of the lower degree of water saturation. This leads to an increased dispersion. Therefore, relationships between water content and relative water velocity fluctuations and water content together with the coefficient of dispersivity in unsaturated porous media respectively have been investigated systematically by displacement experiments in glass beads and coarse-textured sandy soil columns. The breakthrough curves (BTCs) of chloride showed that an increase of solute mixing with a decrease of water content was caused by an increase of flow velocity fluctuations for different pathways. In order to explain the observed tailing effect in unsaturated flow, two mathematical models were used to fit theoretically derived nonlinear functions of water content dependent dispersivities for both porous media. The close agreement between the observed and computed results suggests that the theoretical model of hydrodynamic dispersion can be extended to transport in unsaturated porous media, providing that BTCs of the effluent water are used to estimate representative dispersivity parameters of soils.     

Initial results from a distributed,physically based model of glacier hydrology

This paper describes the development and testing of a distributed, physically based model of glacier hydrology. The model is used to investigate the behaviour of the hydrological system of Haut Glacier d'Arolla, Valais, Switzerland. The model has an hourly time-step and three main components: a surface energy balance submodel, a surface flow routing submodel and a subglacial hydrology submodel. The energy balance submodel is used to calculate meltwater production over the entire glacier surface. The surface routing submodel routes meltwater over the glacier surface from where it is produced to where it either enters the subglacial hydrological system via moulins or runs off the glacier surface. The subglacial hydrology submodel calculates water flow in a network of conduits, which can evolve over the course of a melt season simulation in response to changing meltwater inputs. The main model inputs are a digital elevation model of the glacier surface and its surrounding topography, start-of-season snow depth distribution data and meteorological data. Model performance is evaluated by comparing predictions with field measurements of proglacial stream discharge, subglacial water pressure (measured in a borehole drilled to the glacier bed) and water velocities inferred from dye tracer tests. The model performs best in comparison with the measured proglacial stream discharges, but some of the substantial features of the other two records are also reproduced. In particular, the model results show the high amplitude water pressure cycles observed in the borehole in the mid-melt season and the complex velocity/discharge hysteresis cycles observed in dye tracer tests. The results show that to model outflow hydrographs from glacierized catchments effectively, it is necessary to simulate spatial and temporal variations in surface melt rates, the delaying effect of the surface snowpack and the configuration of the subglacial drainage system itself. The model's ability to predict detailed spatial and temporal patterns of subglacial water pressures and velocities should make it a valuable tool for aiding the understanding of glacier dynamics and hydrochemistry. © 1998 John Wiley & Sons, Ltd.     

A comparative modeling study of a dual tracer experiment in a large lysimeter under atmospheric conditions

In this paper, five model approaches with different physical and mathematical concepts varying in their model complexity and requirements were applied to identify the transport processes in the unsaturated zone. The applicability of these model approaches were compared and evaluated investigating two tracer breakthrough curves (bromide, deuterium) in a cropped, free-draining lysimeter experiment under natural atmospheric boundary conditions. The data set consisted of time series of water balance, depth resolved water contents, pressure heads and resident concentrations measured during 800 days. The tracer transport parameters were determined using a simple stochastic (stream tube model), three lumped parameter (constant water content model, multi-flow dispersion model, variable flow dispersion model) and a transient model approach. All of them were able to fit the tracer breakthrough curves. The identified transport parameters of each model approach were compared. Despite the differing physical and mathematical concepts the resulting parameters (mean water contents, mean water flux, dispersivities) of the five model approaches were all in the same range. The results indicate that the flow processes are also describable assuming steady state conditions. Homogeneous matrix flow is dominant and a small pore volume with enhanced flow velocities near saturation was identified with variable saturation flow and transport approach. The multi-flow dispersion model also identified preferential flow and additionally suggested a third less mobile flow component. Due to high fitting accuracy and parameter similarity all model approaches indicated reliable results.     

Estimation of water velocities in boreholes using a new passive flowmeter

Abstract

Preferential flow pathways in a fractured aquifer may yield abrupt reductions of the water velocity in a well. We propose a new device for measuring low (5–13 cm d^-1) velocities in wells originating from fractures at different depths. The presented flowmeter has been applied in a well in the Bari (southern Italy) fractured aquifer. In the same well, the horizontal flowmeter velocity (9.6 cm d^-1) at 0.5 m depth was compared with velocity (8 cm d^-1) derived from a field tracer test, providing a value 16.5% higher. Moreover, the flowmeter measurements at 1.5 m depth gave a horizontal velocity of 7.2 cm d^-1, which is 11% less than water flow velocity estimated from the field test. The new flowmeter implements the tracer point-dilution method in a plastic (PVC) pipe by causing the water flow to pass through an artificial filter. Laboratory calibration tests have confirmed the good performance of the proposed flowmeter technique, even for water flow up to 300 cm d^-1. The flowmeter was sensitive to 0.1 cm d^-1, with a detection limit of 1.5 cm d^-1, i.e. half the measurable flow velocity of existing flowmeters in wells.

Editor D. Koutsoyiannis; Associate editor S. Grimaldi     

Line-source multi-tracer test for assessing high groundwater velocity

Segmented line-source multi-tracer injection is suggested as an effective method for assessing groundwater velocities and flow directions in subsurfaces characterized by high water flux. Modifying the common techniques of injecting a tracer into a well became necessary after point-source natural and forced gradient tracer tests ended with no reliable information on the local groundwater flow. The tracer's line-source increases the likelihood of success of the test and could provide additional information regarding the lateral heterogeneity of the aquifer. In a field experiment conducted in the northwestern part on the Dead Sea coast, tracers were injected into an 8-m-long line injection system perpendicular to the assumed flow direction. The injection system was divided into four separate segments with four different tracers. An array of five boreholes located within a 10 × 10 m area downstream was used for monitoring the tracers' transport. Two dye tracers (uranine and Na naphthionate) were injected in a long pulse of several hours into two of the injection pipe segments. Two other tracers (Rhenium oxide and Gd-DTPA) were instantaneously injected into the other two segments. The tracers were detected 0.7 to 2.3 h after injection in four of the five observation wells, located 2.3 to 10 m away from the injection system. The groundwater velocity was determined to be ～80 to 170 m/d, based on the recoveries of the tracers. The groundwater flow direction was derived based on the arrival of the tracers and was found to be quite consistent with the apparent direction of the hydraulic gradient.