H31D-0407 0800h
Towards the field-scale experiments and numerical modeling of pesticides in tropical soils
Intensive use of pesticides in agriculture inevitably poses an increased threat to groundwater. Recent findings of pesticide residues in selected drinking water wells in Hawaii brings further attention to this problem since the primary source for potable water in Hawaii is groundwater from basal or dike-confined aquifers. A challenging research project was carried out at the University of Hawaii to elucidate potential impacts of selected pesticides on groundwater and to understand pesticide behavior in tropical soils. The major outcome of the project will be a recommendation to the Hawaii Department of Agriculture whether to restrict or approve these pesticide products entering Hawaii's agricultural market. Three sites on Oahu, one on Maui, and one on Kauai were selected for field evaluation of leaching. The soil types on Oahu are Wahiawa Oxisol (Poamoho), Molokai Oxisol (Kunia), and Waialua Vertisol (Waimanalo). The soil at Kula, Maui is an andisol (loam of Kula series) and that at Mana, Kauai is a Vertisol of Malama series. Three herbicides (S-metolachlor, imazaquin, sulfometuron methyl), one fungicide (trifloxystrobin), and one insecticide (imidacloprid) were used in our study. In addition, a commonly used herbicide (atrazine) and potassium bromide tracer were applied as reference chemicals. After spraying, the plots were covered with straw to decrease evaporation from bare soil surface and irrigated with aerial sprinklers for a period of 16 weeks. Disturbed soil samples from various depths were taken at regular intervals for pesticide analysis. Water flow dynamics was monitored with TDR probes and tensiometers installed at three depths. Weather data were acquired simultaneously. In-situ measurements of unsaturated hydraulic conductivity were done using a tension disc infiltrometer. Laboratory experiments of soil-water retention, as well as degradation, sorption, and column displacement experiments for the selected pesticides were conducted. Hence, comprehensive a database for mathematical modeling of the pesticide transport was obtained. Results so far indicate that S-metolachlor showed its low leachability and short decay half-life. On the other hand, imidacloprid exhibited its low sorption ability with higher leaching potential and longer half-life. Likewise, transport of bromide seems to be controlled by soil hydrologic properties and water application rates.
H31D-0408 0800h
Single-Rod Probes for Time Domain Reflectometry: Sensitivity and Calibration
Time domain reflectometry probes consisting of one conducting rod and a wave mode converter are an alternative configuration which overcomes some of the disadvantages of conventional probes. We examined four different single-rod probes and a two-rod probe regarding their sensitivity to a small and a large conductive scatterer in their vicinity. The single-rod probes were assembled combining a small/large wave mode converter with an uncoated/coated rod. We found that the volume sampled by single-rod probes is larger and more symmetric than in the case of a two-rod probe of equal size. A comparison of the mode converters showed a higher loss for the smaller converter but only a small difference concerning the spatial sensitivity. Coating the conducting rod with a high dielectric constant material reduces the spatial sensitivity. One of the single-rod probes and the two-rod probe were calibrated in a sand tank (particle size 0.08-0.2 mm) with volumetric water content up to $0.35$ m$^3$m$^{-3}$. The calibration showed only small differences between the single-rod and the two-rod probe regarding the measured bulk dielectric constant. Based on this study the single-rod probe is a promising new tool for improved time domain reflectometry measurement of soil moisture.
H31D-0409 0800h
Specification of infiltration and lateral boundary conditions for the site-scale saturated zone flow model near Yucca Mountain
The boundary conditions for the base-case and alternate conceptual site-scale saturated zone flow models for the area surrounding the Yucca Mountain repository are specified through analyses of the Death Valley Regional Flow System (DVRFS) models, the site-scale unsaturated zone (UZ) models, and recharge below Fortymile Wash. The base-case site-scale saturated zone flow model is supplied data that are extracted from the 1997 DVRFS, the 1997 UZ model, and from data relating to infiltration through Fortymile Wash. The alternate conceptual site-scale saturated zone flow model is supplied data extracted from the 2001 DVRFS, the 2003 UZ model, and again from Fortymile Wash. Corrected total infiltration rates for the base-case site-scale saturated zone flow model total 24.4~kg/s with 6.7~kg/s coming from below the 1997 UZ model footprint and 2.0~kg/s from flow infiltrating through Fortymile Wash. Infiltration into the alternate conceptual site-scale saturated zone flow model is 77.3~kg/s with 5.6~kg/s coming from below the 2003 UZ model, and 1.9~kg/s through Fortymile Wash. Lateral recharge flux target boundary conditions are extracted from the 1997 DVRFS for the base-case saturated zone site-scale model and from the 2001 DVRFS model for the alternate conceptual model saturated zone site-scale model and both distributions and magnitudes vary greatly. {\st This work was supported by the Yucca Mountain Site Characterization Office as part of the Civilian Radioactive Waste Program, which is managed by the U.S. Department of Energy, Yucca Mountain Site Characterization Project. Sandia is a multiprogram laboratory operated by Sandia Corporation, a Lockheed Martin Company, for the United States Department of Energy's National Nuclear Security Administration under contract DE-AC04-94AL85000.}
H31D-0410 0800h
Effects of Natural Drift Degradation on In-Drift Thermohydrological Conditions
Understanding thermohydrological processes at the potential high-level waste repository at Yucca Mountain, Nevada, is important for assessing its long-term performance. Detailed process models that provide the in-drift and near-field thermohydrological conditions are needed to estimate the composition of water that may contact the waste package and to evaluate the potential for corrosion of waste packages. Drift degradation could significantly influence the environment inside waste emplacement drifts. Degradation of the host rock may backfill portions of the repository, and drifts could potentially be backfilled within 1000 years after closure. This poster presents a two-dimensional detailed process model that incorporates the temporal variation of in-drift and drift wall geometry as a result of drift degradation. Model results indicate that radiation and convection dominate in-drift heat transfer until the drip shield is completely surrounded by rubble. Subsequently, the insulating effect of rubble causes an abrupt increase in the temperatures of the in-drift components and conduction through the rubble dominates the in-drift heat transfer. The heat generated by emplaced waste transports water vapor away from the drift creating a dryout zone and redistributing pore fluids within a potentially large volume of host rock. The likelihood of water seeping into the drift is strongly affected by the extent and duration of the dryout zone. Model results showing the temporal variability of the dryout zone, both in the host rock and the rubble pile, will be presented. Sensitivity analyses show that the in-drift thermohydrologic conditions are sensitive to the thermal properties of the rubble and the rate of drift degradation. This poster in an independent product of CNWRA and does not necessarily reflect the view or regulatory position of Nuclear Regulatory Commission.
H31D-0411 0800h
Three-Dimensional Model of Heat and Mass Transfer in Fractured Rocks to Estimate Environmental Conditions Along Heated Drifts
Temperature gradients along the thermally-perturbed drifts of the potential high-level waste repository at Yucca Mountain, Nevada, will drive natural convection and associated heat and mass transfer along drifts. A three-dimensional, dual-permeability, thermohydrological model of heat and mass transfer was used to estimate the magnitude of temperature gradients along a drift. Temperature conditions along heated drifts are needed to support estimates of repository-edge cooling and as input to computational fluid dynamics modeling of in-drift axial convection and the cold-trap process. Assumptions associated with abstracted heat transfer models and two-dimensional thermohydrological models weakly coupled to mountain-scale thermal models can readily be tested using the three-dimensional thermohydrological model. Although computationally expensive, the fully coupled three-dimensional thermohydrological model is able to incorporate lateral heat transfer, including host rock processes of conduction, convection in gas phase, advection in liquid phase, and latent-heat transfer. Results from the three-dimensional thermohydrological model showed that weakly coupling three-dimensional thermal and two-dimensional thermohydrological models lead to underestimates of temperatures and underestimates of temperature gradients over large portions of the drift. The representative host rock thermal conductivity needed for abstracted heat transfer models are overestimated using the weakly coupled models. If axial flow patterns over large portions of drifts are not impeded by the strong cross-sectional flow patterns imparted by the heat rising directly off the waste package, condensation from the cold-trap process will not be limited to the extreme ends of each drift. Based on the three-dimensional thermohydrological model, axial temperature gradients occur sooner over a larger portion of the drift, though high gradients nearest the edge of the potential repository are dampened. This abstract is an independent product of CNWRA and does not necessarily reflect the view or regulatory position of the Nuclear Regulatory Commission.
H31D-0412 0800h
A New Versatile Design for Automated Tension Infiltration
Numerous studies in recent years have addressed in situ measurement of soil hydraulic retention and conduction parameters using tension infiltrometers. While convenient for field use and easy to set up, several problems exist with the long-standing design that utilizes an attached bubbling column to maintain supply tensions at the soil interface (Perroux and White,1988). In an attempt to better meet the demands of field work several improvements have been made to make the tool easier to use and more reliable. Elements of the development include an internal reference column for maintaining constant tension, high precision pneumatic valves computer-regulated to let air into the supply tower, and a durable stainless steel supply membrane designed to withstand harsh field conditions with bubbling pressure exceeding 40cm of tension. The instrument is fully controlled through a user-friendly control unit that enables the researcher to stipulate the conditions of each infiltration that must be run; a series of infiltrations at multiple tensions without having to adjust any vales or refill the supply tower; and the porous membrane is now durable enough to last for hundreds of measurements without replacement. Software was also developed for efficient data transfer and interpretation. Preliminary experiments with this novel design indicate its practicality in most geophysical settings, and further opportunities for advancement are identified.
H31D-0413 0800h
A new laboratory instrument for defining near-saturation wetting-drying and capillary conductivity
The moisture release curve is a fundamental descriptor of soil water movement. While tension table apparatus for defining drainage curves in the near-saturated region (0 to -100 cm H2O) have been available for many decades, there has been little advance in automating the process, particularly when measurements of wetting, drying and capillary conductivity are combined. We describe a new instrument, the Automated Moisture Release Apparatus (AMRA), that uses a precision water flow measurement coupled with an accurately controlled variable hydraulic head to exactly define the relationship between matric potential and volumetric water content? from 0 to -100 cm H2O. The new instrument automates the process without any disturbance to the soil core and and has fully programmable suction steps and equilibration times. We compare AMRA moisture release curve results with published Accusand tests for four different grain size mixtures and under different sample height and volume combinations. Our results are consistent with published results and our tests provide guidance on appropriate core lengths for media characterization. AMRA results from field samples from well-characterized soils at the HJ Andrews Experimental Forest in Oregon are also described. These experiments show both consistent drying behavior and drainable porosity decline with depth in the vadose zone. Overall, the instrument's precision, programmable platform, and combined conductivity and moisture content measurement features, make it a useful tool for estimating soil water storage properties/parameters for use in vadose zone hydrology.
H31D-0414 0800h
Investigation of water movement in the unsaturated zones using oxygen and hydrogen isotopes
In other to investigate movements of water through unsaturated soil zones of temperate climates, 17 lysimeters were installed at a test site of a volcanic island, Korea and oxygen and hydrogen isotopic compositions of soil waters were monitored for one year period. For comparison of isotopic compositions, monthly composite precipitations were also collected during the study period. No or little evaporation was observed in the soil waters. This indicates that transpiration is much more important than evaporation in hydrologic budget of the island. Oxygen and hydrogen isotopic compositions of soil waters showed a seasonal variation, indicating that the isotopic compositions of soil water directly reflected those of precipitation. A seasonal change for soil waters of 30 cm depth was faster than that for soil waters of 60-80 cm depth. Overall, dampening of variation of isotopic composition was pronounced with increasing depth. Comparing deuterium excess values of precipitations and soil waters, transit time of about 2 months was estimated for infiltration of water through the soil layer to depths of 30 cm. For the soil layer of to depths of 60-80 cm, transit time of about 4-5 months was estimated. The isotopic compositions of soil waters plotted between the local meteoric water line of dry season precipitation and that of rainy season precipitation, indicating that soil waters were recharged from the whole season precipitations in the study area. This relationship may provide a useful means of evaluation or seasonality of groundwater recharge.
H31D-0415 0800h
Are Advecting Processes in the Vadose Zone of the Albuquerque Basin Altering the Conductive Heat Transfer Signal From Surface Temperature Change ?
Temperature measurements ( T logs ) in the deep vadose zone ( about 60m to 120m depth ) of the Albuquerque Basin have been repeated over the past year at four piezometer nests. The measurements were made with a very fast time response thermistor, which allowed data to be taken every meter going down hole. This depth resolution of temperature data permits a rather detailed observation of the thermal regime in the vadose zone. At one site ( Lincoln Middle School ) the temperature profile below 20m clearly shows a conductive profile resulting from surface temperature change due to urbanization and nearby ( about 10m ) asphalt pavement. At the other three sites the cause of non-linearity in the T log is less certain. Temperature records suggest about 1 deg C increase in near surface air temperature over the past thirty years at the Albuquerque airport; although this data may also be affected by urbanization. The Tome and 98th Street sites are being approached by paved roads and urbanization. At the Tome site expressions representing horizontal advection are the statistically preferred fit to the T log from about 25m to 58m ( F statistic ). At the 98th Street site an expression representing a surface temperature step best fits the T log from 20m to about 75m; however, the temperature step (about 1 deg C to 2 deg C, 3 to 15 yr ago ) is variable between logs, and the profile of the T log with abrupt discontinuities may suggest other than just conductive heat transfer. The fourth piezometer nest at the Mesa del Sol site is the most remote of the sites considered, with as little nearby surface disturbance as might be expected for a drilling location. At depths between 30m and 70m the expressions representing surface temperature change, horizontal advection, and vertical advection, all fit the T log reasonably well. The temperature step expression suggests about 1 deg C to 1.8 deg C surface temperature increase about 13 yr to 28 yr ago. Deeper in the vadose zone, from about 60m to 120m, the expression representing horizontal advection is a generally preferred fit to the T log. Perhaps horizontal liquid flow through perched water zones, steps in the ground water table elevation caused by normal faulting, horizontal variations in barometric pressure effects at depth, and sub horizontal vapor flow moving around permeability barriers on its way to the surface, promote horizontal advection at the Tome and Mesa del Sol sites.
H31D-0416 0800h
Plutonium Particle Migration in the Shallow Vadose Zone: The Nevada Test Site as an Analog Site
The upper meter of the vadose zone in desert environments is the horizon where wastes have been released and human exposure is determined through dermal, inhalation, and food uptake pathways. This region is also characterized by numerous coupled processes that determine contaminant transport, including precipitation infiltration, evapotranspiration, daily and annual temperature cycling, dust resuspension, animal burrowing, and geochemical weathering reactions. While there is considerable interest in colloidal transport of minerals, pathogenic organisms, and contaminants in the vadose zone, there are limited field sites where the actual occurrence of contaminant migration can be quantified over the appropriate spatial and temporal scales of interest. At the US Department of Energy Nevada Test Site, there have been numerous releases of radionuclides since the 1950's that have become field-scale tracer tests. One series of tests was the four safety shots conducted in an alluvial valley of Area 11 in the 1950's. These experiments tested the ability of nuclear materials to survive chemical explosions without initiating fission reactions. Four above-ground tests were conducted and they released plutonium and uranium on the desert valley floor with only one of the tests undergoing some fission. Shortly after the tests, the sites were surveyed for radionuclide distribution on the land surface using aerial surveys and with depth. Additional studies were conducted in the 1970's to better understand the fate of plutonium in the desert that included studies of depth distribution and dust resuspension. More recently, plutonium particle distribution in the soil profile was detected using autoradiography. The results to date demonstrate the vertical migration of plutonium particles to depths in excess of 30 cm in this arid vadose zone. While plutonium migration at the Nevada Test Site has been and continues to be a concern, these field experiments have become analog sites for the release of radiological materials potentially important to consequence management investigations. In particular, these 50-year old experiments with long and detailed site investigations under relative undisturbed conditions offer insights into transport pathways that must be represented in simulation models that evaluate responses to radiological dispersal devices (RDDs). A compilation of the available site characterization data suggests additional experimental and modeling programs that can ultimately quantify the fate of contaminant particles released at the soil surface.
H31D-0417 0800h
Estimation of Moisture-Dependent Anisotropy in Effective Unsaturated Hydraulic Conductivity and Moisture-Retention Curve Using Spatial Moments of Moisture Plumes.
We have developed a new approach that uses spatial moments of three-dimensional snapshots of a moisture plume under transient flow conditions to estimate the three-dimensional effective unsaturated hydraulic conductivity tensor. The evolution of the spatial first moment in the vertical (z) direction of a moisture plume is used to determine the vertical velocity (Vz) of the center of the plume. This velocity is then related to the gravity term, dKz(theta)/d(theta), of the moisture-based Richards' equation and thus, the vertical unsaturated hydraulic conductivity can be derived. Afterwards, the rate of changes of the second spatial moments in the x, y, and z directions of the plume are related to the water diffusivity tensor (Dx, Dy, and Dz). Assuming that Dx=Kx/C, Dy=Ky/C, and Dz=Kz/C relations, where C is the moisture capacity term, we solve for Kx, Ky, Kz, and C. Application of the new approach to a field experiment yields an effective unsaturated hydraulic conductivity tensor that exhibits moisture-dependent anisotropy. The effective hydraulic conductivities appear to be consistent with laboratory-measured unsaturated hydraulic conductivity data from small core samples; they also reproduce general behavior of observed moisture plume at the site. Finally, potential applications of the approach to laboratory- and field-scale problems, in conjunction with recently advanced geophysical survey tools, are discussed.
http://www.hwr.arizona.edu/yeh
H31D-0418 0800h
The Effects of Moisture on the Surface and Pore Areas of Kaolinite, Montmorillonite and Silica Gel
A key factor in the uptake and release of contaminants in soils is the amount of water adsorbed on the surfaces of the soil particles. It is hypothesized that the surface and pore areas of porous solids will decrease because of the presence of this adsorbed water layer. In our experiments, we incubated four different solids (2 silica gels, kaolinite, and montmorillonite) in three different controlled environments (relative humidity (RH) values of 16%, 76%, and 86%) at ambient temperatures for over 440 days. Our results indicate that moisture significantly affects the surface areas of all solids except kaolinite. The surface areas of one of the silica gel samples incubated at 16%, 76%, and 86% were 271.1 m$^{2}$/g, 237.5 m$^{2}$/g, and 218.1 m$^{2}$/g, respectively. Contrary to this, the surface areas of montmorrillonite increased with increasing RH because of the expansion of the clay layers. However, long-term incubation of montmorrillonite at 86% humidity resulted in a decrease in surface areas (surface areas of montmorrillonite at 76% RH increase from 2.95 to 7.10 m$^{2}$/g while at 86% RH, they increase to 6.03 m$^{2}$/g)), probably because more moisture was available at 86% RH to condense inside the pores and between the layers of the clay, reducing surface areas. This study is important because it will allow us to quantify the effects of moisture on the physical properties of particles and, hence improve our current understanding of contaminant distribution and fate in soils. These results are also important because they enhance our understanding of the swelling of clays such as montmorrillonite.
H31D-0419 0800h
Low-frequency dilatational wave propagation through unsaturated porous media containing two immiscible fluids
The quantitative description of elastic wave propagation in an elastic porous medium containing two immiscible fluids is one of the classic problems in the physics of flow through unsaturated porous materials. An analytical theory of the low-frequency behavior of dilatational waves propagating through such a porous medium is presented based on the Berryman-Thigpen-Chin (BTC) model, in which capillary pressure effects are neglected. We show that the BTC equations in the frequency domain can be transformed, at sufficiently low frequencies, into a dissipative wave equation (telegraph equation) and a propagating wave equation in the time domain. These partial differential equations describe two independent modes of dilatational wave motion that are analogous to the Biot fast and slow compressional waves in a single-fluid system. The equations can be solved analytically under a variety of initial and boundary conditions.
H31D-0420 0800h
Quaternary Geology, Unsaturated Soil-Moisture Measurements, and Evidence for Overland Flow in the Globe Piedmont, Mojave Desert, CA
We hypothesize that the evolution of desert surfaces by pavement development and pedogenesis strongly controls soil moisture properties including water infiltration rates and moisture retention. This leads to strong spatial variation in rainfall-runoff relationships and availability of moisture and nutrients for biota. Correlating these hydrologic properties with surficial geologic units based on pedogenesis may provide the only practical means of estimate runoff generation and soil-water-plant dynamics over broad areas. To illustrate these relationships, we installed and monitored a variety of sensors in undisturbed soils within different aged deposits across the arid Globe piedmont, northeast of Kelso, Mojave National Preserve, California. Since 2002 we have monitored rainfall, soil moisture, and unsaturated soil matric potential using weather stations, and depth-stratified heat-dissipation and dielectric-constant probes. Our studies focus on two adjacent alluvial fans (700 to 950 m altitude) that contain active washes draining bedrock highlands, and intrafan regions of desert pavement. In this region, old (Pleistocene) deposits have advanced desert pavements capping silt-rich, vesicular A (Av), argillic, and petrocalcic horizons. Old deposits support sparse vegetation cover and generally small canopy volumes. In contrast, coarse-textured, weakly developed soils of active and young (Holocene) deposits correlate with greater cover of certain plants such as creosote bush ({\it Larrea tridentata}). We estimated transient initial infiltration rates by measuring the time elapsed draining a given water volume in a bottomless bucket that served as an infiltration ring. Infiltration experiments and depth-stratified probe data indicate that young deposits with weak soils have higher infiltration rates than older deposits. We measured the lowest infiltration capacities (70-420 mm/hr) on Pleistocene fan deposits and on an active wash where the surface was capped by a muddy depositional crust. Young alluvial fan and colluvial deposits were highly conductive with initial rates of 1000 to 2700 mm/hr. In active washes, cm-thick depositional laminae caused significant lateral flow. Macropore flow was only locally observed. Experimental wetting fronts within active washes terminated within silty-sand lenses above coarse sand to gravel layers. In older deposits, lateral flow followed Av horizon boundaries. Field-measured infiltration rates were strongly time-dependent with initial values exceeding final values. This decrease may reflect wetting from initially dry conditions and the sponge-like influence of cyanobacteria, lichen, and clay. Storm-averaged rainfall intensities (~14 mm/hr) associated with a monsoonal storm in August 2003 did not exceed the lowest experimental infiltration capacities. Averaged over 5 minute intervals, peak rain intensities were 67 mm/hr, similar to the lowest measured infiltration rates of 70 mm/hr on older deposits. In addition, field observations reveal that intrafan overland flow, erosion, and sediment transport occurred during the storm downslope of desert pavement on older deposits. The correlation of peak rainfall intensities, experimental infiltration rates, and field observations of runoff support our hypothesis that advanced pedogenesis promotes runoff generation. Future infiltration experiments using shallow water ponded in a 1-m diameter ring will constrain how pedogenesis limits long-term infiltration and influences the location and timing of runoff, wetting front depths, and lateral flow.
H31D-0421 0800h
Geochemical Modeling of pH Neutralization of High Alkaline-Saline Waste Fluids in Unsaturated Sediments
Leakage of high alkaline-saline fluids, such as those stored in Hanford, a site of the U.S. Department of Energy (DOE) in Washington State, has raised attention of scientific community. These fluids have unique thermodynamic and physical properties. Chemical components in the fluids are incompletely dissociated, especially those containing divalent or polyvalent ions. A number of laboratory experiments through injecting synthetic high alkaline-saline fluids (up to 10M of sodium nitrate, pH $>$12) into the sediments sampled from the DOE Hanford site were conducted to study the reactive transport processes of the fluids in subsurface environments. The experimental results observed show that the composition of the high alkaline sodium nitrate fluids can be drastically changed due to fluid-rock interactions, and eventually lead to pH neutralization of the fluid in the plume front. The dominant fluid-rock interactions are cation exchanges (Na$^{+}$-K$^{+}$-Ca$^{+2}$-Mg$^{+2}$-H$^{+}$), precipitation of calcium and magnesium minerals, and dissolution of silica. In order to precisely model the reactive transport of these processes, a coupling of the Pitzer's ion-interaction geochemical model and a flow and transport model would be highly needed. The extended existing reactive geochemical transport code, BIO-CORE$^{2Dc}$, incorporating a comprehensive Pitzer ion-interaction model, is capable of predicting the experimental observations. In addition, the developed model was tested against two reported cases. In both cases, the measured mean ionic activity coefficients were well reproduced by our model, while the Debye-Hückel model, usually used to calculate aqueous species activities in dilute solutions, was unable to predict the experimental data. Finally, modeling study based on our laboratory column experiment was performed. Our simulation is able to capture the observed pH trends, changes in exchangeable cations such as Ca$^{+2}$, Mg$^{+2}$, and formation of secondary precipitation phases in the plume front.
H31D-0422 0800h
The Cambric Ditch at the Nevada Test Site as a Long-term Vadose Zone Test Bed
Atomic weapons testing at the Nevada Test Site introduced many tracers for quantifying hydrologic transport processes in arid climates. The particular experiment at the Cambric site in Frenchman Flat represents an ongoing 29-year field test that could never be repeated and continues to offer opportunities for vadose zone studies. The Cambric test released the energy yield equivalent of 0.75 kt of TNT when it was detonated 294 m below the land surface and 73 m below the water table in Frenchman Flat in May 1965. Beginning in 1975, groundwater was pumped steadily from a well located 91 m from the detonation point in order to elicit information on radionuclide migration. The pumping well effluent was monitored, discharged to an unlined ditch, and allowed to flow towards a dry lake 1.6 km away. Approximately one third of this flow was lost to infiltration. Over the next 16 years, pumped groundwater was shown to contain tritium, fission products (technetium-99, iodine-129) and activation products (chlorine-36), all of which can be used to trace water flow in the vadose zone. Bromide was also added as an additional tracer into the ditch. Multi-year records exist for water migration in the shallow vadose zone along with temperature profiles. Over the course of the pumping experiment, vegetation developed in and near the ditch, providing an additional pathway for water loss by transpiration and selective radionuclide transport. Significant water has not flowed in the ditch since 1991 and the site remains an ideal analog site for the studying drying in arid climates, the adaptability of vegetation under changing water conditions, and the use of helium-3 as a tracer of soil-atmosphere exchange and vadose zone - groundwater interactions. In addition, there is evidence that tritiated water and chlorine-36 have infiltrated over 200 meters vertically in the vadose zone and have become a source term for groundwater contamination. The Cambric Ditch remains as a field site ideally suited for further experimental investigations and the development and testing of transport models at spatial and temporal scales typical of contaminated, arid vadose zone sites. Data from this site has been useful in model development and additional field studies are suggested to further guide the application of models to contaminant migration in other deep vadose zones. Portions of this work were conducted under the auspices of the U.S. Department of Energy by the University of California, Lawrence Livermore National Laboratory under contract W-7405-Eng-48.
H31D-0423 0800h
The Efficacy of Thermodynamics in Development of Governing Equations and Constitutive Relations for Saline Solutions in Variably Saturated Porous Media
Relations are presented that describe the behavior of water with high salt concentrations in variably saturated porous media for isothermal systems. Equations were derived using classic equilibrium thermodynamics of closed systems. Resulting corrections for vapor pressure, liquid pressure, and Darcy coefficients are presented, and an extension of the relations to non-isothermal systems is proposed. Next, the governing equations obtained using general process thermodynamics for continuous systems are presented. A discussion of the usual use of these equations follows, along with some general observations. Historically, the principles of Onsager have been employed to develop a linearly coupled set of forces and fluxes. Except for the reciprocity relations, which allow the determination of fewer coefficients, little is gained by this standard approach. A similar set of governing equations results from simply assuming every flux may be driven by a gradient in any of the potentials, but this requires estimation of every coefficient separately. On the other hand, the use of Onsager relations requires a unique separation of the reversible and irreversible terms in the energy equation, which is not always a trivial derivation. Either method guarantees a sufficient number of coefficients requiring determination that within the experimental uncertainty, it may be difficult to tell if the physics is really captured, or if there just exists sufficient freedom of the parameters to fit the data. Results of recent research indicates that for the case of water movement in the vicinity of highly concentrated salts, the dilute approximations extended by use of a more general chemical activity term is sufficient for modeling the constitutive relations, except in very dry, fine textured sediments.
H31D-0424 0800h
Effects of Salinity and Drought Stresses on Root Water Uptake
In vadose zone numerical simulation models, root water extraction is typically accounted for by introducing a sink term into the Richards equation. Various forms of the sink term have been proposed to simulate the reduction in water extraction that occurs when soil salinity and/or drought conditions exist in the root zone. While this representation of root water uptake is a fixture of modern simulation models, there has been relatively little work demonstrating agreement between the models and measured water uptake data. In this work, we compare HYDURS-1D model simulations with drainage and root water uptake data collected for forage crops grown in lysimeters using a range of salinity and drought treatments. We found good agreement between the model and the data using a single set of salinity and water stress parameters, a noteworthy result given the broad range of experimental conditions considered (irrigation waters with electrical conductivities as high as 28 dS/m). On the other hand, the required salinity and water stress parameters did not correspond to published salt tolerance data for these crops, suggesting that the near term prospects for using this model in a purely predictive capacity (i.e., without detailed crop- and site-specific calibration) are limited.
H31D-0425 0800h
In-Situ Determination of Specific Yield Using Soil Moisture and Water Level Changes in the Riparian Zone of the Arkansas River, Kansas.
Specific yield (S$_{y}$ - drainable porosity) is an important parameter for numerous hydrogeologic investigations, but few reliable methods for its in-situ determination exist. S$_{y}$ is commonly estimated by saturating and draining soil cores, using soil water characteristics in parametric models, or from pumping tests. However, these methods can introduce large uncertainties due to core disturbance, inappropriate models, and the impact of heterogeneity, respectively. At a site in the riparian zone of the Arkansas River in south-central Kansas, the amount and variability of groundwater consumption by phreatophytes is being assessed using water-table fluctuations. S$_{y}$ is an important control on the magnitude of these fluctuations, so estimates of its value and its variability in time and space are needed. An in-situ method based on soil-moisture and water-level changes (Skaggs et al., 1978) was used for this purpose because it can provide information at the scale appropriate for such an investigation. Neutron access tubes (0 - 1.82 m in depth) were installed at four locations in the riparian zone to periodically gather soil moisture information. Water level data in the shallow alluvial aquifer were collected every 15 minutes in six wells using pressure transducers calibrated with manual measurements. The method of Skaggs et al. assumes that conditions in the unsaturated zone are at equilibrium, i.e. soil moisture changes are only a product of the rise or fall of the water table. One period for which this assumption appears reasonable was in September and October of 2003 when the water table dropped more than 0.61 m in four weeks. The S$_{y}$ estimates for this period displayed relatively little variability between sites (0.19-0.21). Total porosity for the depth intervals through which the water table fell varied from 0.28-0.36, so the S$_{y}$ estimates appear reasonable. S$_{y}$ estimates from a pumping test, however, displayed much greater variability (0.15-0.31), which is most likely a product of heterogeneity in hydraulic conductivity. Results from a deeper set of neutron access tubes (0-3.05 m in depth) installed in the summer of 2004 will also be presented.
H31D-0426 0800h
Shape of gas flow paths causes power law tailing
In soil and/or groundwater remediation, we often see prolonged tailings: continuous outflow of low concentration pollutants for very long time, and in many cases power low behavior of late-time time-concentration curves. We considered that this kind of tailing can be caused by the shape of the gaseous flow introduced in saturated/unsaturated porous media. When gas is introduced to porous media, like air-sparging or soil vapor extraction, the shape of the gas flow path would be tree-like, or to some extent "fractal." So, there would be a distribution of the distance that a solute would have to travel by diffusion before getting to a gas/water interface, and we might expect that the distribution of this "diffusion distance" would be power-law-like. In order to see if tailing can be caused by this mechanism, simple column experiments were carried out. A column, 64 mm in inner diameter and 240 mm in height, was prepared and was packed with 1mm diameter glass beads. Nitrogen gas containing 5 % CO2 and 5% He was supplied from the bottom of the column, and after the water in the column is approximately saturated with CO2, the sparging gas was changed to pure nitrogen. The CO2 and He concentrations in the effluent gas was monitored and recorded. As the result, we saw tailing: the double-log plots of the concentration vs. time relationship was practically linear, and the absolute value of the slope in the double-log charts were 1.28, 0.95 and 0.83 according to the gas flow rates of 40, 80 and 120 ml/min, respectively. Slope less than 1.00 showed that these tailings cannot be explained by Freundlich-type adsorption behavior. Model analysis showed that this power low time-concentration behavior with the slope of approximately -1.0 can be explained by the power law distribution of diffusion distance \textit{a} with PDF p(\textit{a}) proportional to \textit{a}^{-1}.
H31D-0427 0800h
Hypothesis Testing in Hydrologic Restoration: Soil Moisture, Root Density, and Plant Community Associations in Northeastern Sierra Nevada, California, Meadows
In order to test the qualitative impression of success of an innovative hydrologic restoration technique called `plug and pond', we investigated the linkages between meadow rehydration and changes in soil moisture, root density, and plant community composition in northeastern Sierra Nevada, California meadows. Conceptually, plug and pond restorations reconnect stream flows to floodplains and remnant channels, returning meadows to a pre-incised-channel moisture regime. We developed hypotheses to 1) determine if plant community composition can serve as an indicator of soil moisture and root density in northeastern Sierra meadows, and therefore, serve as an indicator of post- restoration soil moisture and root density changes, and 2) quantify year-to-year post-restoration variation of soil moisture and root density in Carmen Valley, Knuthsen Meadow, Tahoe National Forest (TNF). In the summers of 2003 and 2004, we gathered data at a six-inch depth on soil moisture, fine root density, and very fine root density, and identified plant species aboveground, at a series of randomly selected sites. We applied a detrended correspondence analysis (DCA) to the identified plant species in order to create groups of `wet', `intermediate', and `dry' plant communities in twelve TNF reference meadows. Using ANOVA, we found a statistically significant difference in soil moisture and root density between the wet plant community and a combined intermediate/dry community. Regression analysis suggests that 40.5% of the variation in very fine root density is accounted for by soil moisture. Preliminary results from data gathered in 2004 suggests that the qualitative division of Knuthsen Meadow into an `upper, wet' section and a `lower, drier' section is supported by soil moisture means of 13% and 9%, respectively. In the year-to-year analysis, we expect to find that hydrologic restoration has lead to an increase in soil moisture and root density in Knuthsen Meadow. Finally, we anticipate that this study will provide resource managers with quantitative tools that can be applied in conjunction with other adaptive management procedures to achieve project goals.
H31D-0428 0800h
Artificial neural network application to the solute transport through unsaturated zone
For effective groundwater management, the amount of solutes transported from surface to groundwater table through the unsaturated zone should be estimated. Recently, artificial neural network (ANN) receives much attention as a tool for analyzing solute transport through the unsaturated zone. The ANN is considered to be a versatile tool for approximating complex functions without taking into account the complicated physical mechanisms of the unsaturated zone. In this study we develop an ANN model to evaluate the amount of solute reaching to the groundwater table when a certain amount of surface input is given. Using the developed model we investigate the applicability of the ANN to solute transport through the unsaturated zone using numerical and laboratory experiments. For numerical tests, 2624 training data are obtained using HYDRUS2D. We group the data into two training sets : one for arrival time of solutes and the other for solute mass with respect to time. For the latter, the data are trained in the form of normalized mass, cumulative mass, and logarithmic mass, respectively to examine the influence of data form on training. The optimal neural network architecture is determined through a numerical case study with the number of nodes in the hidden layer, momentum, learning rate and the range of initial weights. Then we estimate the solute transport for various inputs. Results show that ANN model can estimate the solute transport efficiently when sufficient training sets are available. Also, the choice of network parameters and formation of training set patterns are important for estimation of solute transport. For validation, the developed ANN model is applied to soil column tests data. The results are compared with the estimation using conventional convection-dispersion equation (CDE).
H31D-0429 0800h
``Phytomonitoring'': A Screening Tool For Detection Of Subsurface VOC Contamination
In highly urbanized areas, characterizing the distribution of subsurface contamination is complicated by the dense urban underground and surface infrastructure. Drilling and monitoring in such settings requires extensive and complex coordination, and, for widespread problems extending over many square kilometers, high monetary outlays. When we encountered extensive groundwater contamination of the Central Coastal Plain aquifer of Israel underlying the Tel Aviv metropolis by volatile organic contaminants (VOCs), such problems rapidly became insurmountable. In this 200 km2 region, the average thickness of the vadose zone is about 30 m, and of the underlying freshwater saturated aquifer, about 130 m. As the ground surface is mainly impermeable due to urban development, volatile VOC vapors may build up to high levels in the vadose zone, and travel throughout long distances. As such, we were highly motivated to find an alternative, non-invasive, inexpensive means of scanning the vadose zone for VOCs, which would provide the basis for more traditional surveys at the predetermined sites. Specifically, we asked ourselves if it would be possible to exploit the VOC uptake ability of trees and shrubs for detecting subsurface vadose zone contamination below an urban environment. Preliminary laboratory tests showed that trees take up VOCs from unsaturated sediments. Further, VOCs were found in trees sampled at a former industrial site where the subsurface is heavily contaminated with chlorinated solvents (i.e., trichloroethene TCE; tetrachloroethene PCE), at the heart of our study area. As such, trees at numerous sites in the Tel Aviv area were sampled and analyzed for VOC content in tree trunk cores. Compounds detected in the tree cores include benzene, toluene, ethylbenzene, xylenes, TCE, PCE, and 1,1,1-trichloroethane. A good correlation between subsurface contamination and positive detection of contaminant VOCs in tree cores was found. The results support the idea that phytomonitoring can be implemented for preliminary screening of potentially contaminated sites in extensive urban areas where monitoring the unsaturated zone can prove problematic, time-consuming and expensive.
H31D-0430 0800h
Hygrometric Measurement of Soil Water Potential
Knowledge of soil water potential as a function of water content is required to make unsaturated flow and transport predictions. Although numerous methods are available to measure soil water potential, they are largely difficult and time consuming procedures. The goal of the research is to develop a hygrometric method that will perform satisfactorily with minimal required hardware or technician time. The volume of a drop of saline water will change due to evaporation or condensation until its salinity, and hence osmotic potential, is equal to the water potential in the adjacent gas phase. This relationship is exploited by our method to measure soil moisture potential. To begin, a drop of KCl solution with known mass and KCL concentration is placed adjacent to a soil sample with known water content inside a hermetically sealed container. The mass of the KCl drop is recorded over time with an electronic balance. As thermodynamic equilibrium is achieved, the mass of water within the KCl drop changes until its osmotic potential is equal to the capillary potential of water within the soil sample. After the mass of the KCl drop reaches equilibrium, the KCl concentration is calculated, which enables direct determination of the water potential within the soil sample. Unlike transient hygrometric measurements of water potential using psychrometers, no calibration is required.
H31D-0431 0800h
Microenvironmental and Seasonal Variations in the Soil-Water Content of the Unsaturated Zone of a Sand Dune System at Pinery Provincial Park
The volumetric water contents of 1-2 m deep sand profiles were obtained monthly across the 250 m transition that contains the succession from unvegetated dune sands to C3 and C4 grass-dominated dunes to oak-savannah stabilized dunes at Pinery Provincial Park, south-western Ontario, Canada. The physical properties of the homogeneous sands are virtually identical at each profile. This consistency allows one to examine the relative influence of topography, degree of soil formation, vegetation type and abundance, forest canopy development, and seasonality on soil moisture contents in this rare, fragile and endangered ecosystem. The soils were all well drained, despite the frequent rain events that characterize this temperate-humid region. However, a less mobile fraction of water was retained in the soil for at least several months, even following large rain events. Soil water contents were too low to encourage preferential downslope flow; topography generally had little influence on soil moisture contents. The presence and type of vegetation had the greatest influence on soil moisture contents, both across the dune succession and vertically within the profiles. Grass roots facilitated deeper infiltration of precipitation relative to unvegetated dune soils. During the growing season, soil moisture contents were lowered significantly in the rooting zones of both the forested and dune-grass dominated profiles. The greatest differences in moisture content between forested and unforested areas occurred in the shallowest soil layers. Despite interception of rain by the oak canopy, water contents were higher in the forest soils. This behaviour reflects greater moisture retention by the organic-rich A-horizon and leaf litter layer typical of these more mature soils, and less evaporation from the cooler forest floor. Seasonal trends were dominated by the extent of soil water removal via transpiration, although these patterns were sometimes obscured by large precipitation events occurring a few days prior to sampling. Inter-site differences in moisture content were accentuated in the warmest summer months, when soils were subjected to the competing influences of increased evapotranspiration and periodic wetting events.
H31D-0432 0800h
Vadose Zone Characterization and Monitoring Beneath Waste Disposal Pits Using Horizontal Boreholes
Vadose zone characterization and monitoring immediately below landfills using horizontal boreholes is an emerging technology. However, this topic has received little attention in the peer-reviewed literature. The value of this approach is that activities are conducted below the waste, providing clear and rapid verification of containment. Here we report on two studies that examined the utility of horizontal boreholes for environmental characterization and monitoring under radioactive waste disposal pits. Both studies used core sample analyses to determine the presence of various radionuclides, organics, or metals. At one borehole site, water content and pore-water chloride concentrations were also used to interpret vadose zone behavior. At another site, we examined the feasibility of using flexible membrane liners in uncased boreholes for periodic monitoring. For this demonstration, these retrievable liners were air-injected into boreholes on multiple occasions carrying different combinations of environmental surveillance equipment. Instrument packages included a neutron logging device to measure volumetric water at regular intervals, high-absorbency collectors that wicked available water from borehole walls, or vent tubes that were used to measure air permeability and collect air samples. The flexible and retrievable liner system was an effective way to monitor water content and collect air permeability data. The high-absorbency collectors were efficient at extracting liquid water for contaminant analyses even at volumetric water contents below 10 percent, and revealed vapor-phase tritium migration at one disposal pit. Both demonstration studies proved that effective characterization and periodic monitoring in horizontal boreholes is both feasible and adaptable to many waste disposal problems and locations.
H31D-0433 0800h
Unraveling the Fate and Transport of SrEDTA-2 and Sr+2 in Hanford Sediments
Accelerated migration of strontium-90 has been observed in the vadose zone beneath the Hanford tank farm. The goal of this paper is to provide an improved understanding of the hydrogeochemical processes that contribute to strontium transport in the far-field Hanford vadose zone. Laboratory scale batch, saturated packed column experiments, and an unsaturated transport experiment in an undisturbed core were conducted to quantify geochemical and hydrological processes controlling Sr+2 and SrEDTA-2 sorption to Hanford flood deposits. After experimentation, the undisturbed core was disassembled and samples were collected from different bedding units as a function of depth. Sequential extractions were then performed on the samples. It has been suggested that organic chelates such as EDTA may be responsible for the accelerated transport of strontium due to the formation of stable anionic complexes. Duplicate batch and column experiments performed with Sr+2 and SrEDTA-2 suggested that the SrEDTA-2 complex was not stable in the presence of soil and rapid dissociation allowed strontium to be transported as a divalent cation. Batch experiments indicated a decrease in sorption with increasing rock:water ratios, whereas saturated packed column experiments indicated equal retardation in columns of different lengths. This difference between the batch and column experiments is primarily due to the difference between equilibrium conditions where dissolution of cations may compete for sorption sites versus flowing conditions where any dissolved cations are flushed through the system minimizing competition for sorption sites. Unsaturated transport in the undisturbed core resulted in significant Sr+2 retardation despite the presence of physical nonequilibrium. Core disassembly and sequential extractions revealed the mass wetness distribution and reactive mineral phases associated with strontium in the core. Overall, results indicated that strontium will most likely be transported through the Hanford far-field vadose zone as a divalent cation.
H31D-0434 0800h
Xenon Isotope Releases from Buried Transuranic Waste
Xenon is an inert rare gas produced as a fission product in nuclear reactors and through spontaneous fission of some transuranic isotopes. Thus, xenon will be released from buried transuranic waste. Two complementary methods are used to measure xenon isotopes: radiometric analysis for short-lived radioxenon isotopes and mass spectrometry for detection of stable xenon isotopes. Initial measurements near disposal facilities at the U.S. Department of Energy's Hanford Site show radioxenon and stable xenon isotopic signatures that are indicative of transuranic waste. Radioxenon analysis has greater sensitivity due to the lower background concentrations and indicates spontaneous fission due to the short half life of the isotopes. Stable isotope ratios may be used to distinguish irradiated fuel sources from pure spontaneous fission sources and are not as dependent on rapid release from the waste form. The release rate is dependent on the type of waste and container integrity and is the greatest unknown in application of this technique. Numerical multi-phase transport modeling of burial grounds at the Idaho National Engineering and Environmental Laboratory indicates that, under generalized conditions, the radioxenon isotopes will diffuse away from the waste and be found in the soil cap and adjacent to the burial ground at levels many orders of magnitude above the detection limit.
H31D-0435 0800h
Hydraulic functions of variably saturated porous media: Upscaling from single-pore to sample-scale model
Extrapolation of hydraulic functions ($P_c(S) and K(S)$) to oven dry conditions is of interest for many practical problems such as for example modeling of fingering in porous media. Experimental soil water-retention data are normally obtained for high and moderate values of the saturation and lacking at saturations below or close to residual. We develop a model which predicts hydraulic functions down to oven dryness by using pore-scale modeling. Following the approach by Tuller and Or [1, 2], the pore space is represented by an array of connected polygonal unit cells. Three steady-state and laminar flow regimes in the unit cell are considered to derive the hydraulic functions: 1) flow in completely filled pores, 2) corner flow in partially filled pores, and 3) film flow on solid surfaces. Unlike the Tuller and Or model we take into account an influence of the adjacent cells. The state of the unit cell (either filled with liquid or yielding corner and film flow) depends on the pressure in that cell and is affected by the state of the adjacent cells as well. Thus, this model may be treated as a combination of the Mualem model [3] and the the Tuller and Or model. This general model allows us to take into account hysteretic properties of the $P_c-S$ relation and calculate the main drainage curve, the main wetting curve, and scanning curves for all ranges of saturation even below residual. We fit our model with available experimental data, obtained with typical laboratory silica sand in the high to moderate saturation range for the drainage cycle. All calculated hydraulic functions (hydraulic conductivity, main wetting curve, and scanning drainage curves) match experimental data without any additional assumptions. We regard this fact as a validation of the proposed model. \begin{enumerate} \item Tuller, M., Or D, Dudley LM. Adsorption and capillary condensation in porous media: Liquid retention and interfacial configurations in angular pores. Water Resour. Res., 1999, 35:1949-1964. \item Tuller, M., Or D. Unsaturated hydraulic conductivity of structured porous media: A review of liquid configuration-based models. Vadose Zone J., 2002, 1:14-37. \item Mualem Y. A conceptual model of hysteresis. Water Resour. Res., 1974, 10:514-520. \end{enumerate}
H31D-0436 0800h
Ground saturation following various harvest patterns: implications of assessing a hydrologic model versus statistical analysis
We assessed groundwater response to various harvest intensities in southeast Alaska. Observations of peak water table heights in 56 groundwater-monitoring wells at two study locations \(Hanus and Portage\) suggest that soil saturation levels on hillslopes differed significantly with harvest intensity only at Portage Bay following 25, 75, and 100% harvest. Before the forest at Portage Bay was cut by 100%, the average rainfall needed for 50% saturation of the soil was 54 mm, but after clearcutting soils reached 50% saturation with only 21 mm. We hypothesize that Hanus Bay sites lacked statistically significant differences in levels of saturation following harvest due to several factors: 1\) location, size, and landform of areas contributing to wells, and 2\) number, species, size, and density of trees in relationship to wells for pre and post-harvesting conditions. To test these hypotheses, we developed a simple hydrologic model to predict saturation. Model results were tested with existing records of saturation for various levels of rainfall to estimate which factors influenced observed patterns.
H31D-0437 0800h
Climatic controls on diffuse groundwater recharge in semiarid environments
Although there is no diffuse groundwater recharge at many semiarid sites, evidence for diffuse recharge exists at some locations where mean annual precipitation, P*, is much less than mean annual potential evapotranspiration, PET*, particularly where soils are coarse and hydroclimatic conditions are variable. We investigate the climatic controls on diffuse recharge using a one-dimensional, liquid-only flow model. The model is driven by a stochastic parameterization of climate that includes storm size distribution and seasonality of precipitation (P) and potential evapotranspiration (PET), constrained by data from 536 weather stations in the southwestern U.S. Storm size distribution and seasonality determine the frequency and duration of intervals when P exceeds PET, which controls the flux of water past the root zone. For coarse soils, climates with large, infrequent storms yield recharge when P*/PET* exceeds 0.4, compared to 0.7 for a typical storm size distribution. Recharge through fine soils is insensitive to storm size and occurs when P*/PET* exceeds 0.8. Seasonality has a stronger influence on recharge than storm size distribution, and the effects are similar for coarse and fine soils. Recharge is relatively insensitive to rainfall seasonality. In contrast, the typical PET annual cycle lowers P*/PET* of the recharge threshold by 0.3. The relative timing of P and PET maxima is critical: recharge occurs at P*/PET* values that are lower by 0.2 when the rainy season occurs during winter instead of summer. Over the range of climate and soils examined, P*/PET* values at the recharge threshold varied from 0.2-0.7. Therefore, P* and PET* are insufficient to predict where recharge will occur.
H31D-0438 0800h
Water and Solute Mass Transport in Soils Developed on glacial Drift: A Br Tracer Investigation Using Instrumented Soil Monoliths at an Agricultural Long Term Ecological Research Site (Kellogg Biological Station, Hickory Corners, Southern Michigan)
Hydrologic processes control the residence time of water in the soil column. This is of central importance in understanding mineral weathering rates in terms of reaction kinetics and solute transport. In order to better quantify the coupling between water and solute mass transport and to better define controls on carbonate and aluminosilicates weathering rates, we have conducted bromide-tracer introduction experiments at four replicate soil monoliths (4 m$^{3}$ volume) instrumented and managed by the KBS-LTER. Monolith soils are developed on the pitted outwash plain of the morainic system left by the last retreat of the Wisconsin glaciation, around 12,000 years ago. Soil profiles from the monolith sections extend to 200 cm and they were sampled and characterized texturally and mineralogically. Quartz and feldspar are dominant throughout the soil profile, while carbonates and hornblende occur only in deeper soil horizons. The four replicate monoliths are instrumented with gas and soil water sampling devices (Prenart tension lysimeters) at various depths. The monoliths also have a large capacity tray at the bottom, which permits collection of water for weight and chemical determinations. A bromide tracer solution (as lithium bromide) was applied to coincide as closely as possible with a major snowmelt event (2/27/04). The saturated and unsaturated transport of bromide through the four monoliths was followed as a function of time and soil profile depth for the duration of the snowmelt as well as intermittent rain events. Because the soil was saturated at the time of bromide application, the bromide solution is expected to move rapidly through macropores, followed by slower movement into micropores. The unsaturated transport of bromide is largely controlled by the intensity and duration of the rains if it is dominated by piston flow as opposed to preferential channel flow. In general, the tracer moved through the shallow soils very quickly, which is shown by early sharp peaks in bromide concentrations. Transport of bromide into deeper soil horizons, however, differs markedly among the four monoliths. Even within a given monolith, waters sampled at the same depth by different tension lysimeters show a very different pattern of bromide transport over time with some lyimeters suggesting piston flow, while others in the same monolith suggest preferential channel flow. These differences are likely driven by heterogeneous soil textures. The water recovered from the monolith trays over the first three months of the study period is between 80 and 90 percent of the total precipitation recorded at the LTER site. This recovery is reasonable given the fact that temperature was low and crops were not yet actively growing. The recovery of bromide is different among the monoliths and in general is less than 50 percent, which means more than 50 percent of tracer is still in the soils even after three months. Residence time of water has been calculated after some assumptions on the breakthrough curve. The water mass transport constraints imposed by the bromide tracer study will be utilized in concert with additional data on soil water geochemistry.
H31D-0439 0800h
Seasonal variation in tracer movement in a forested experimental plot using manual and automated sampling techniques
In recent years, implications associated with groundwater contamination have increased the efforts of researchers studying solute transport through the unsaturated soil zone near the ground surface (vadose zone). Success in tracking the movement of water, solutes, and the development of vadose zone hydrologic models requires high-quality field data. However, near continuous, spatially distributed soil moisture and matric potential data sets are rare because conventional soil parameter instrumentation is point-based and labor intensive. An automated vadose zone monitoring system (AVM) was developed to complement a set of manually monitored instrument arrays in an effort to address the quality and quantity of data collected in the vadose zone. Tracer (tritium) movement was evaluated for winter and summer irrigation applications on a forested field plot on the Atlantic Coastal Plain in South Carolina. Tritiated water was applied in two pulse events through an irrigation system and breakthrough data were measured from soil cores, suction lysimeters and soil vapor wells in the field. Measured breakthrough data for both winter and summer tracer applications were compared to solute transport solutions and modeled using numerical modeling software. The data from automated and manual sampling systems were used to evaluate the results of a one dimension hydrologic model that predicted the movement of water and a tracer (tritium) movement associated with winter and summer irrigation events.
H31D-0440 0800h
Monitoring Soil Moisture in Saline Soils using Neutron Probe, Time Domain Reflectometry, and Heat Dissipation Sensor Measurements
Knowledge of spatial and temporal variability of soil moisture content (SMC) is important for understanding of land-atmosphere interactions, groundwater recharge, and water balance. Different measurement methods have contrasting strengths and weaknesses. Traditional neutron probe measurements cannot be automated and are time-intensive. However, there are widespread problems with using automated time domain reflectometry (TDR) for monitoring SMC due to high soil salinity/electrical conductivity. The objective of this study was to show how these limitations can be overcome by using multiple methods. Neutron probe access tubes, TDR probes (coated and uncoated), and heat dissipation sensors (HDS) were installed at an engineered field laboratory in a semiarid setting. The texture of the soils was sandy clay loam, including 0.3 m of uncompacted topsoil with low salinity and non-swelling clays underlain by 1.0 to 1.7 m of compacted subsoil with high salinity and swelling clays. A neutron probe was used to manually measure SMC profiles at 20 locations at approximately monthly intervals over a 3.5 yr period. During a 4 to 5 yr overlapping period, daily automated measurements were made at 8 locations of apparent dielectric constant (Ka) and bulk electrical conductivity (EC) profiles using TDR (128 probes) and matric potential profiles using HDS (54 sensors). TDR measurements in the high salinity soils were effectively calibrated in situ using neutron probe measurements. Modeled estimates of spatial average water content were generally within 0.01 m3/m3. A similar approach was used to combine neutron probe, TDR, and HDS measurements to generate in situ soil water retention functions. These functions were then used to estimate SMC from matric potential measurements. These approaches allowed SMC to be monitored in high salinity swelling soils and provided much higher resolution time series than were obtained from the limited neutron probe measurements.
H31D-0441 0800h
Spatial variability of ground-water recharge in selected principal aquifers of the eastern United States
Over 500 vadose-zone sediment cores were collected as part of a regional study of ground water recharge to aquifers comprising glacial deposits and the Floridan, Coastal Lowlands, Piedmont and Blue Ridge, and North Atlantic Coastal Plain aquifer systems. Study objectives were to compare methods for estimating recharge; to compare and contrast recharge estimates for selected principal aquifers in the eastern U.S.; and to identify landscape factors that significantly influence recharge. We evaluated the Darcian, water-budget, water table-fluctuation, and tracer methods for estimating recharge. Sediment cores were analyzed for particle size distribution, moisture content, bulk density, organic matter, and selected anions. Direct measurements of unsaturated hydraulic conductivity (K) were made on a small number of cores by the steady-state centrifuge method. For all cores, K was derived from pedotransfer functions based on texture and bulk density data. Darcian water fluxes were calculated assuming either nonuniform matric potential or a unit gradient (q about equals K). Unit gradient estimates of recharge represent homogeneous sediments and thick layers in heterogeneous systems. Preliminary results indicate that the point estimates of recharge vary considerably within principal aquifers, and that median recharge is highest in glacial deposits in the northeastern U.S. Median recharge estimated by the Darcian method was similar to estimates derived from a base-flow index. Overall, there was good correspondence between unit-gradient and non-unit-gradient estimates of recharge, indicating that matric forces were not dominant in sampled sediment layers. Unit-gradient recharge was strongly related to moisture content and sediment texture. For sands, hydraulic conductivities derived from pedotransfer functions compared favorably with those measured by the centrifuge method. The pedotransfer method, however, overpredicted K for a silty sample with high moisture content. It is possible that the pedotransfer method is biased towards sandy samples because the empirical model on which it is based is calibrated to predominantly coarse-textured soils.
H31D-0442 0800h
Water Budget and Modeling of Stream Channel Infiltration of Coalbed Methane Co-Produced Water at a Storage Impoundment Site, Powder River Basin, Wyoming
Rapid coalbed methane (CBM) development in the Powder River Basin, Wyoming, has resulted in a dramatic increase in the number of producing wells, with as many as 40,000 new wells projected to drilled during the next decade. CBM development involves the co-production of large volumes of coalbed water, which is most commonly discharged to impoundments. Little is known about the potentially significant effects that this co-produced water may have on shallow aquifers and water budgets. Since many of the impoundments in the Powder River Basin are in-channel and supplement surface water flow, it is important to understand what factors influence stream channel infiltration. Modeling of cross-sectional infiltration in a stream channel was undertaken using the U. S. Geological Survey's SUTRA finite-element code, simulating fluid movement from the stream channel, through the unsaturated zone, and into the shallow aquifer. Soil type, saturated hydraulic conductivity, and degree of anisotropy within the soil layer were analyzed to better understand the effects that each have on channel infiltration. At a small study site containing two in-channel infiltration impoundments in the Beaver Creek drainage (a tributary to the Powder River), water budgets have been determined from late July, 2003 to the present (excluding the winter months). Calculated infiltration rates of 0.04 cfs/mile in a 0.8 m wide stream channel compared well with modeled regional soils. Slug tests were utilized to determine aquifer properties in the underlying alluvium/weathered bedrock. A two-layer SUTRA model is presented representing the upper soil layer and underlying alluvium/weathered bedrock and compared to observed changes in groundwater levels in the stream sections affected by introduced CBM water.
H31D-0443 0800h
Thermal Neutron Radiography of Deuteriated Water in Soils
As for x-rays, neutron radiography is a noninvasive imaging technique based on the attenuation of thermal neutrons by the object in question, described by Beer's law. However, neutron imaging is complementary to x-rays, as it is especially well suited for materials containing hydrogen atoms and mostly other low atomic weight attenuating materials. Although neutron attenuation techniques are routinely used in engineering, relatively little is known about its application to soils. We will present results demonstrating the tremendous potential of using neutron attenuation techniques to measure spatial and temporal distribution of water in soils at the 50 micrometer spatial resolution. The neutron source is a Mark II Triga Reactor at McClellan Nuclear Radiation Center (MNRC) in Sacramento, CA. The reactor runs at 1.8 MW and emits a poly-energetic neutron beam, including the thermal range. Unfortunately beam hardening and backscattering are a major source of uncertainty. Recent laboratory experiments conducted at MNRC suggest that beam hardening is considerably reduced when using deuteriated water, because its cross-section for attenuation of thermal neutrons is much lower than for regular water.
H31D-0444 0800h
Underground Corrosion after 32 Years: A Study of Fate and Transport
In 1970, the National Bureau of Standards (NBS) initiated a comprehensive long-term corrosion test. The NBS, now known as the National Institute of Standards and Technology (NIST), buried over 1000 specimens-consisting of different stainless steel types, specialty alloys, composite configurations, and multiple material forms and treatment conditions-at six distinctive soil-type sites throughout the United States. Researchers from Department of Energy (DOE) National Laboratories at the Idaho Engineering and Environmental Laboratory and Savannah River recovered and analyzed one set of specimens after 34 years of burial. Objectives were to enhance understanding of subsurface corrosion and near-field contaminant transport. The specimens were buried at a US Coast Guard base near Wildwood, NJ. They were originally buried in dry, sand (poorly-graded sand with an average saturated hydraulic conductivity of 4.7 X 10-3 cm/sec) but upon recovery, were found to rest about 2 feet below the water table. In 1970, the sand was vegetated with beach grasses, now it supports abundant vines, shrubs, and small-to-medium diameter trees. Background soil and water samples were collected for chemical and microbial analysis. Soil pH ranged from 5.71 to 6.64, while water pH ranged between 4.23 and 5.55. Soluble chloride ranged between 16 and 59 mg/L. Additional samples were collected to determine whether corrosion product transport had occurred in the soil or water and how far. Sensitized Type 301 and Type 304 stainless steel plates and U-bend specimens sustained the greatest observable corrosion with adhering corrosion product. This abstract addresses the initial investigations.
H31D-0445 0800h
Coupled Hydrology and Uranium Geochemistry of the Hanford Caliche Layer
Subsurface disposal and dissemination of uranium has occurred at the US Department of Energy Hanford Reservation in Richland, WA. A caliche layer forms the approximate boundary between the vadose and saturated zone in places, and uranium contamination is observed. The cemented, calcium-rich caliche is geologically distinct from the proximal unconsolidated sediments, which suggests that the transport of uranium may differ from the surrounding subsurface environment. Further, it has recently been shown that calcium exerts a major influence upon uranium speciation at neutral pH. Our objectives were to quantify the coupled hydrology and geochemistry of uranium transport, considering the high calcium content within the caliche. The hydrology of the layer was characterized by performing multiple nonreactive tracer experiments on two undisturbed cores under saturated and unsaturated conditions. Physical nonequilibrium was minimal, as evidenced by the co-elution of the nonreactive tracers, which contrasts with results from the surrounding unconsolidated sediments. The geochemistry was quantified by performing kinetic batch, equilibrium isotherms, and repacked miscible displacement experiments, and it was observed to be influenced by the presence of natural and added Ca. X-Ray Absorption Spectroscopy will be utilized to quantify the geochemistry of the adsorbed and/or precipitated complex. Coupled uranium transport processes were investigated under unsaturated conditions in the undisturbed cores, and less retardation was observed in comparison to repacked and batch experiments. This was attributed to differences in reactive surface area in undisturbed versus disturbed rock samples. Our results suggest uranium transport in the Hanford subsurface will be influenced by the distinct hydrological and geochemical characteristics of the caliche layer which contrast with that of the surrounding sediments.
H31D-0446 0800h
In Situ Gaseous Treatment and Reoxidation Studies - Implications for Vadose Zone Remediation
The {\it In Situ} Gaseous Reduction (ISGR) approach has been developed by Pacific Northwest National Laboratory (PNNL) for vadose zone soil remediation. This technology uses diluted hydrogen sulfide gas (H$_{2}$S) as a reductant for immobilization of contaminants that show substantially lower mobility in their reduced forms, e.g., Tc, U, and Cr. The ISGR approach can be used to immobilize or stabilize pre-existing contaminants in the vadose zone by direct H$_{2}$S treatment. Alternatively, a permeable reactive barrier (PRB) could be generated by pumping a gaseous mixture of hydrogen sulfide diluted in nitrogen through an interval in the vadose zone to produce a layer of reduced sediment. Recent work directed towards development of the ISGR technology has focused on collection of laboratory data needed to support design and installation of a subsurface barrier. This has included conducting small and large scale column tests to determine the reaction characteristics between sediment and hydrogen sulfide gas mixtures and the rates of these reactions. During gaseous treatment, most of the hydrogen sulfide component is consumed in soil and results in the generation of FeS. This constituent is the active agent that maintains reducing conditions in the barrier needed for contaminant immobilization. Another aspect of current testing activities involves determining the reducing capacity of treated soil during reoxidation. This information is needed to assess the effectiveness and lifetime of the barrier. Initial testing activities have also been undertaken to assess the rate and extent of immobilization of Cr, Tc, and U and to evaluate the potential for remobilization of these contaminants as barrier reoxidation occurs. Barrier design and performance behavior is being supported by analytical and numerical modeling activities that utilize laboratory data regarding sediment treatment, roxidation, and contaminant immobilization processes as it becomes available.
H31D-0447 0800h
Sensitivity of Groundwater Recharge to Variations of Climate, Soils, and Vegetation Based on Unsaturated-Flow Modeling
Understanding the relative importance of climate, vegetation, and soils in controlling groundwater recharge is important for estimating effects of climate variability and land use/land cover change on recharge/water resources. The purpose of this study was to evaluate the sensitivity of groundwater recharge to variations in climate, soil type, and vegetation type using unsaturated-flow modeling and to further assess the sensitivity of the model to variations in parameterization and design. More detailed sensitivity analyses were conducted at representative sites to evaluate the impact of vegetation parameters (percent bare area, leaf area index, and root depth), and other parameters on simulated recharge by generally increasing and decreasing parameters by 50%. One dimensional unsaturated flow models were developed for 13 sites (county areas) that represented a range of climate (arid - humid), vegetation (shrub, grass, forest, crops), and soil (fine to coarse grained, monolithic and layered) conditions based on readily available online data from Texas, US. Spatially and temporally averaged recharge rates are more appropriate for water resources management than point estimates at a single time; therefore, simulated long-term (1961 - 1990) average recharge for each combination of vegetation type and soil profile layering was regionalized to the site area (county) for this analysis using Geographic Information Systems (GIS). Simulated 30 yr average recharge rates for nonvegetated monolithic sand profiles ranged from 54 mm/yr in arid regions to 720 mm/yr in humid regions, correlating positively with precipitation (r=0.99). Adding vegetation to the monolithic sand profiles had a similar effect on recharge as soil profile layering; both reduced recharge relative to that in monolithic sand profiles by factors ranging from 2 to 10. Vegetation and soil-profile layering both resulted in high local variability in recharge within each site (county); however, spatially-weighted long-term average recharge rates were much less variable and were positively correlated with precipitation (r=0.79 to 0.96). The most realistic simulations included vegetation and layered soil profiles, which resulted in a range of recharge rates from 0 to 114 mm/yr and correlated with precipitation (r=0.95). Including bare area up to 50% greatly increased recharge. Decreasing LAI and root depth also had great impacts on recharge. Simulated recharge was more sensitive to PET rates in the humid setting relative to the arid setting. These detailed sensitivity analysis provide more in depth understanding of specific controls on recharge. The general analysis indicates that strong relationships between climate, vegetation and simulated recharge should be useful in assessing potential impacts of climate variability and land use change on recharge.
H31D-0448 0800h
Estimating the Permeability of Carbonate Rocks using Image Analysis and Effective Medium Theory
A methodology was recently developed to estimate the permeability of sedimentary rocks from two-dimensional pore images [Lock et al., J. Appl. Phys., 2002]. The only data required from the images are the areas and perimeters of the individual pores. The hydraulic conductivities of the individual pores are estimated from their areas and perimeters using the hydraulic radius approximation. Stereological correction factors are applied to determine the true cross-sectional shapes from the images, and to determine the true number density of pores per unit area. A constriction factor accounts for the variation of the cross-sectional area along the tube length. The pores are assumed to be arranged in a cubic lattice, after which the effective-medium equation of Kirkpatrick is used to estimate the effective conductance of the pores. Finally, the permeability is estimated from the effective pore conductance and the number density of pores. When applied to several data sets of sandstones, having permeabilities in the range of 20-1400 mD, the permeability estimates were always within roughly a factor of two of the values measured in the laboratory. This methodology is now being applied to a set carbonate rocks, having permeabilities in the range of 0.5 to 25 mD. Carbonates generally have more complex and heterogeneous pore structures than do sandstones. Nevertheless, our preliminary results how that, for rocks that do not contain appreciable amounts of vugs that are unconnected to the main conducting pore space, the method again yields permeabilities within a factor of two of the measured values. However, when applied to vuggy carbonates, the predictions may be too high by several orders of magnitude. The error in these cases arises from including isolated vugs in our calculation of the effective pore conductance. These vugs are easily identified by eye. However, as our aim has been to develop a rapid, objective permeability estimation method that requires little if any subjective intervention by the user, we are currently trying to develop automated ways of identifying the unconnected vugs, so as to remove them from the calculations.
H31D-0449 0800h
A new Approach for Interconnections Between Shrinkage Cracks and Soil Matrix
The shrinkage geometry factor connects the changes of subsidence and the total crack volume at the shrinkage of a soil layer matrix. An available approach for estimating the factor does not account for the crack volume in samples, stretching a shrinking soil layer, and a possible change of the factor with water content. At the same time the concept of the shrinkage geometry factor is used for estimating the crack volume and water flow in shrink-swell soils. For these applications the accuracy of the value of the shrinkage geometry factor is very important. A recently suggested new approach for the factor estimation enables one to account for the water content dependence of the factor and the impact of crack presence in a sample. The present work develops this approach also accounting for the stretching of a shrinking soil layer. Corrected values of the shrinking geometry factor are described in terms of the shrinkage curves of: a layer that is composed of unconnected solid cubes from the available approach, a stretched layer with cracks, a sample with cracks, and the soil matrix without cracks. This correction leads to changes in the soil subsidence and total crack volume compared to those obtained from the available approach. The consideration is based on two physical conditions: the soil matrix volume without cracks is the same for the sample and the layer; and, the matrix deformations of an unlimited shrinking soil layer are longitudinal those of a thin quasi-elastic plate. Two available experimental examples are considered to illustrate the approach to be developed. The former relates to clay paste samples. The latter relates to an aggregate clay soil in situ and in samples. Results show that the shrinkage geometry factor from the available approach and that obtained after correction significantly differ. That is, the impacts of cracks in soil samples and of soil layer stretching at shrinkage should be taken into account when estimating the shrinkage geometry factor value. The results also show the appreciable change of the corrected shrinkage geometry factor with water content. They can be useful in discussing the possibility of transferring hydraulic properties and flow features of a swell-shrink soil that are observed on or calculated for soil samples, to the case of the soil in field conditions.