H13F-0979
Using geostatistical, artificial neural network and inverse models to estimate 3D soil type and hydraulic property distributions in a deep volume of unsaturated soil
We use geostatistical and neural network analyses to estimate the three-dimensional distributions of soil types and hydraulic properties in a relatively large volume of vadose zone underlying the Maricopa Agriculture Center near Phoenix, Arizona. Geostatistical analysis of soil texture and bulk density data from the site are analyzed geostatistically to reveal the underlying stratigraphy as well as finer features of their three-dimensional variability in space. Such fine features are revealed by cokriging soil texture (as primary variable) and water content measured prior to large-scale long-term infiltration experiments (as secondary variable). Resultant estimates of soil texture and bulk density data across the site are then used as input into a neural network model to produce estimates of soil hydraulic parameter (saturated and residual water content, saturated hydraulic conductivity, van Genuchten parameters alpha and n) distributions across the site in three dimensions. We compare these estimates with laboratory-measured values of these same hydraulic parameters and with estimates obtained by inversion using a three-dimensional flow simulator.
H13F-0980
Unsaturated Hydraulic Conductivity for Layered Heterogeneous Soils
This study investigates the effective hydraulic conductivities of unsaturated soils for one-dimensional structured heterogeneity. The heterogeneity is defined using homogeneous sublayers forming repeated unit cells. Results from previous studies indicated that for a large cell length, the effective hydraulic conductivity approaches an arithmetic average of the pressure heads which develop in each sublayer. For a cell of length approaching zero, the effective conductivity becomes the harmonic average of the individual sublayer hydraulic conductivities weighted according to the cell fraction each occupies. The effective hydraulic conductivity functions for the finite cell lengths fall within the envelope formed by the two limiting cases for small and large cell lengths. But the effective conductivity in previous studies was defined as the steady downward flow velocity for an infinitely deep profile. In this study, we examine the effective hydraulic conductivities for a finitely deep profile which is typical for many applications such as water fluxes for both infiltration and evaporation between ground surface and water table. We address the significance of profile depth, upper and bottom conditions in the profile on the effective hydraulic conductivities. The contrasting extent of the hydraulic parameters in the sublayers, which define the heterogeneity degree, and its influence on the effective hydraulic conductivities are also explored.
H13F-0981
Generating hydraulic properties from non-equilibrium water-retention curves
Water retention curves, particularly those obtained from ceramic plate experiments, tend overestimate the equilibrium water content at high tensions and thus imply unrealistic amounts of water present in films etc. We have previously reported a method to predict non-equilibrium water-retention data from known fractal equilibrium water-retention curves. This method relates actual water loss to the product of the equilibrium water loss and the ratio of two different percolation-based calculations of the hydraulic conductivity. Here we reverse the process and generate the equilibrium water-retention curve from data. No particular form of pore- size distribution is assumed, although extrapolation procedures and a posteriori checks are used. The results we obtained make sense and imply water film contents of roughly 0.01 instead of 0.05 or higher in the coarse Hanford site soils investigated. From a reliable equilibrium pore-size distribution, percolation theory yields all flow and transport properties. Work supported by NSF grant EAR-0609884.
H13F-0982
On the Influence of Topography Upon Scaling Characteristics of Soil Hydraulic Parameters
One of the most important issues concerning studies into the hydrologic cycle and climate prediction today is the upscaling of soil hydraulic parameters in the unsaturated zone. Ecological phenomena occur differently, and due to different causes, at a wide range of scales. Efforts to model hydrologic processes and phenomena, with particular emphasis on those occurring in the unsaturated zone, are currently ongoing at various scales. Input data are required for these models at their representative scales. However, measurement of parameter data at all such required scales is impractical as it entails huge outlay of finances, time and effort. Inter-connections often exist between information across these scales. However, the exact mathematical or physical nature of these connections is generally a mystery. Over the past few decades, numerous efforts have been conducted to either understand and solve these mysteries, or to find a way around them to obtain effective parameters at multiple scales. Most upscaling efforts thus far have opted to ignore the effect of topography in their derivation of effective parameter values. This approach is reasonable as long as the coarser support dimensions are smaller than hill slopes. When upscaling fine scale hydraulic parameter data to hillslope scales and beyond, however, topography plays a bigger role and cannot be ignored. We present a study of the influence of topographic variations on the effective, upscaled soil hydraulic parameters under different hillslope configurations. Fine resolution parameters were upscaled using the Power Averaging Operator methodology which incorporates features from both mean-type and mode-type aggregation. Simulations of multiple hypothetical scenarios were conducted using the HYDRUS- 3D hydrologic modeling software to develop empirical relationships between the topography and the soil hydraulic parameters for matching hydrologic responses. These relationships may be assimilated into currently existing schemes to derive a more comprehensive algorithm to upscale fine resolution soil hydraulic parameters to any footprint dimension.
H13F-0983
Upscaling the Unsaturated Hydraulic Conductivities of Isotropic Soils
Accurate simulation and prediction of flow and transport of solutes in a heterogeneous vadose zone requires the appropriate hydraulic properties corresponding to the spatial scale of interest. Upscaling techniques provide effective properties to describe the vadose zone system's behavior with information collected at a much smaller scale. Realizing that a saturated system can be considered as a special state of the unsaturated system, the methodologies for upscaling the saturated hydraulic conductivity of heterogeneous isotropic porous media under steady-state flow conditions can be extended for upscaling the unsaturated hydraulic conductivity. An advantage of this approach is that the extended upscaling methods are independent of the choice of hydraulic function models. The Matheron, small-perturbation, and self- consistent upscaling methods were used to demonstrate the approach. The extended upscaling methods were tested using multi-step numerical experiments of gravity-induced flow into Miller-similar synthetic soils with different levels of heterogeneity. Results show that, under 3-D flow conditions in isotropic soils, the self- consistent method applies to all the soil heterogeneity conditions considered while the Matheron and small- perturbation methods are acceptable for soil of relatively low variability.
H13F-0984 INVITED
Soil Moisture Processes in the Near Surface Unsaturated Zone: Experimental Investigations in Multi-scale Test Systems
Understanding the dynamics of soil moisture distribution near the ground surface is of interest in various
applications involving land-atmospheric interaction, evaporation from soils, CO2 leakage from carbon
sequestration, vapor intrusion into buildings, and land mine detection. Natural soil heterogeneity in
combination with water and energy fluxes at the soil surface creates complex spatial and temporal
distributions of soil moisture. Even though considerable knowledge exists on how soil moisture conditions
change in response to flux and energy boundary conditions, emerging problems involving land atmospheric
interactions require the quantification of soil moisture variability both at high spatial and temporal resolutions.
The issue of up-scaling becomes critical in all applications, as in general, field measurements are taken at
sparsely distributed spatial locations that require assimilation with measurements taken using remote sensing
technologies. It is our contention that the knowledge that will contribute to both improving our understanding
of the fundamental processes and practical problem solution cannot be obtained easily in the field due to a
number of constraints. One of these basic constraints is the inability to make measurements at very fine
spatial scales at high temporal resolutions in naturally heterogeneous field systems. Also, as the natural
boundary conditions at the land/atmospheric interface are not controllable in the field, even in pilot scale
studies, the developed theories and tools cannot be validated for the diversity of conditions that could be
expected in the field. Intermediate scale testing using soil tanks packed to represent different heterogeneous
test configurations provides an attractive and cost effective alternative to investigate a class of problems
involving the shallow unsaturated zone. In this presentation, we will discuss the advantages and limitations of
studies conducted in both two and three dimensional intermediate scale test systems together with
instrumentation and measuring techniques. The features and capabilities of a new coupled porous
media/climate wind tunnel test system that allows for the study of near surface unsaturated soil moisture
conditions under climate boundary conditions will also be presented with the goal of exploring opportunities to
use such a facility to study some of the multi-scale problems in the near surface unsaturated zone.
http://www.cesep.mines.edu
H13F-0985
Fractal Model of Bundle Capillary Tubes for Infiltration
Green-Ampt model is commonly used in situ due to its simplicity and satisfactory performance for a wide variety of infiltration problems. The model assumes a piston flow with a sharp wetting front propagating to deeper depths. However, the assumption of a sharp wetting front will deviate from the reality because of the effect of diffusion. To apply Green-Ampt model to field, there is a need to explore the effect of diffusion. We proposed a fractal model of bundle capillary tubes to describe the velocity deviation of wetting front. The proposed model assumes that the pore space of the soil is represented by a bundle of capillary tubes with a fractal law distribution of pore sizes. The distribution of travel distances can be estimated at any given time, and the effective arrival time can be estimated for a given location. The result shows that the travel distance of the wetting front increases with the decrease of probability density function. The relationship between the effective arrival time and the average travel distance indicates that average travel distance advances with respect to the square root of effective arrival time.
H13F-0986
Mean and Variance Relationships for Soil Moisture at Multiple Scales
Hydrologic studies of the land surface have revealed that observable relationships exist between the mean and variance of soil moisture fields. In these investigations uncertain trends occur where sometimes a decreasing soil moisture mean produces an increase in soil moisture variance; other times a directly proportional relationship exists. Recent studies to model and explain these paradoxical trends have focused on moisture thresholds that depend on physical properties of the soil itself. These examinations have used sparse point location data to model vertical transport processes only. In this work we pursue a continuum approach using digital spatial soil data to examine the trends between the mean and variance of soil moisture fields. We account for the additional effects of lateral dispersion on moisture movement as we investigate the mean and variance relationships of soil moisture at multiple scales.
H13F-0987
A new Multi-Scale Data Assimilation Algorithm to Downscale Satellite-Based Soil Moisture Data
The study focuses on downscaling of soil moisture from coarse remote sensing footprints to finer scales. Two approaches are proposed for soil moisture downscaling. The first approach provides the probability distribution functions at the finer scales with no information about the spatial organization of soil moisture fields. The second approach implements a multiscale ensemble Kalman filter (EnKF) that assimilates remotely sensed coarse scale soil moisture footprint, attributes of fine scale geophysical parameters/variables (i.e., soil texture: %sand, vegetation: NDVI, topography: slope, and precipitation) and coarse/fine scale simulation into a spatial characterization of soil moisture evolution at the finer scales. To downscale the remotely sensed coarse scale soil moisture to another spatial scale, the multiscale EnKF uses a bridging model. The bridging model infers the pixel-specific scaling coefficient from the compatible geophysical parameters/variables that influence the soil moisture evolution process at that particular spatial scale. Data from diverse hydroclimatic regions from the semiarid Arizona, the agricultural landscape of Iowa, and the grassland/rangeland of Oklahoma are used in the study to implement the multiscale downscaling algorithm. The results demonstrate that the bridging model of multiscale EnKF helps to characterize the evolution of soil moisture within the remotely sensed footprint. Validation conducted at the finest scale also shows reasonable agreement between the measured field data and the simulated downscaled soil moisture evolution.
H13F-0988
New Methodology to Assess Soil Internal Stress-Strain Interaction
The Restrained Ring Method (RRM) (Abou Najm et al., 2008) was presented as a tool to understand and measure the evolution of internal stresses in soils as they dry and shrink (Figure 1). This understanding holds key significance in soil characterization due to the effects that soil's internal stresses have on major soil behaviors including strain (shrinkage curve), cracking, and soil strength. This study builds on the previous research by integrating the RRM and digital imagery processing (Figure 2) to assess the relationships between stress and strain in soils. A linear relationship between the logarithm of the internal stress and the strain resulting from shrinkage was experimentally calculated for the first time in the soil literature. Results of stress-strain relationships from different soil textures will be presented. This methodology is expected to bridge a major gap between the soil physics and mechanics communities by establishing the tool to quantify internal stresses in soils due to internal stresses thus allowing for agricultural and environmental applications to be considered using mechanistic approach. It will also complement the methods and procedures developed in soil mechanics to assess stress-strain relationships as function of external loads and foundations. References: Abou Najm, M., Mohtar, R., Weiss, J., and Braudeau, E. Assessing Internal Stress Development in Unsaturated Soils. Water Resources Research, 2007WR006484R, Special issue on Hydrology and Mechanical Coupling in Earth Sciences and Engineering: Interdisciplinary Perspectives (accepted).
H13F-0989
Efficient uncertainty quantification techniques in inverse problems for Richards' equation using coarse-scale simulation models
In this talk, we discuss efficient uncertainty quantification techniques for Richards' equation which use coarse-scale simulation models. When solving Richards' equation in heterogeneous porous media some type of upscaling or coarsening is needed due to scale disparity. We describe multiscale techniques used for solving the spatial component of the stochastic flow equations. These techniques allow us to simulate the flow and transport on the coarse grid and thus reduce the computational cost. Additionally, we discuss techniques to combine multiscale methods with stochastic solution techniques, specifically, sparse grid collocation methods. These techniques are further used in constructing efficient uncertainty quantification techniques consisting of sampling hydraulic conductivity given dynamic measurements. We propose several efficient sampling algorithms based on Langevin diffusion and the Markov chain Monte Carlo (MCMC) method.
H13F-0990
Scaling Environmental Processes in Heterogeneous Arid Soils (SEPHAS) Weighing Lysimeter Facility: Construction, Installation, and Preliminary Infiltration Experiment Data
The inability to upscale or downscale arid environmental processes influences research areas of hydrology,
biogeosciences, mathematical modeling, and global environmental change. Research facilities that span
small (column) scales to large (basin) scales are either rare or nonexistent. To address this issue,
researchers from Nevada's universities constructed a weighing lysimeter facility in Boulder City, NV, under
the NSF funded program: Scaling Environmental Processes in Heterogeneous Arid Soils (SEPHAS). Four
lysimeters are weighed on separate balances, each with a live mass of approximately 28,000 kg with a
resolution of roughly 100 g or 0.025 mm of water. Each lysimeter is equipped with data loggers that can be
accessed remotely so investigators can monitor individual sensors and weather systems as needed. This
meso-scale facility is devoted to investigating the near-surface interactions of soil, water, biota, and
atmospheric processes that affect desert environments, similar to those found in the southwestern United
States such as the Mojave Desert, and will bridge existing eco-scale, laboratory, and nano-scale research
efforts. Each lysimeter contains 12 m3 of either repacked or intact desert soil (dimensions: 2.258 m
diameter and 3 m deep) and is instrumented with 13 different sensor technologies to measure state variables
including water content, matric potential, and thermal properties at 15 depth planes. Furthermore, a
relatively new technology called distributed temperature systems was installed to obtain temperature profiles.
During packing, four conservative tracers were applied uniformly at four depths from 0.15 to 0.55 m.
Solution samplers installed at seven depths from 0.50 to 2.9 m will sample soil solution during irrigation
experiments. Native desert shrubs will be installed in two replicate lysimeters in the fall of 2008 and horizontal
rhizotron tubes installed at three depths from 0.60 to 1.50 m and one vertical rhizotron tube will be used to
examine rooting behavior and water balance in recently disturbed soil. This presentation will describe the
construction and installation of this unique facility, and present preliminary data collected during an infiltration
experiment.
http://www.sephas.dri.edu
H13F-0991
Relating Temperature Propagation and Wetting Front in Arid Soils Using Distributed Temperature Sensing (DTS) Coupled With Multiple Technologies
Increased demand for water resources in the arid SW US requires innovative ways to measure and describe the movement of water through arid soils. Several types of sensors, utilizing both destructive and non- destructive measurement techniques, are available. A research facility located approximately 40 km southeast of Las Vegas, NV houses three weighing lysimeters, each equipped with multiple sensors to observe temperature changes and soil moisture content. A fourth lysimeter is planned for Fall 2008 installation. Fiber-optic Distributed Temperature Sensing (DTS) is used in the lysimeters along with a variety of soil moisture sensors. DTS systems monitor temperature and the rate of temperature change with a high temporal and spatial resolution by measuring the temperature dependent and independent energy reflectance (Raman backscattering). Two DTS cables were placed in each lysimeter. A vertical cable was positioned from 5-300 cm by wrapping cable around a 5-cm PVC pipe with a vertical measurement resolution of 1.17 cm/m of fiber. A lateral cable with 1.5-m diameter coils at depths of 5, 25, 50, 75, 95 and 200 cm and 2-m diameter coils at depths of 25 and 95 cm was installed. In addition to the DTS, we have installed dual- probe and triple-probe heat-pulse sensors and soil thermistors. Thermal response curves from the heat- pulse sensors were converted to water content and thermal properties. A precision rainfall simulator was constructed to apply approximately 1-1.5 cm/day/lysimeter. Using each technology, we compared the temperature responses to the wetting front position through the upper 50-75 cm of soil, and to the overall water mass changes in the lysimeters. The nested measurement technologies allow us to examine the DTS technology for monitoring wetting front position, water content, soil heat flux, and thermal regimes. This presentation will provide the preliminary data analysis of the irrigation system and the comparison of these different technologies in arid soils.
H13F-0992
PUMPING TEST DETERMINATION OF UNSATURATED AQUIFER PROPERTIES
Tartakovsky and Neuman [2007] presented a new analytical solution for flow to a partially penetrating well pumping at a constant rate from a compressible unconfined aquifer considering the unsaturated zone. In their solution three-dimensional, axially symmetric unsaturated flow is described by a linearized version of Richards' equation in which both hydraulic conductivity and water content vary exponentially with incremental capillary pressure head relative to its air entry value, the latter defining the interface between the saturated and unsaturated zones. Both exponential functions are characterized by a common exponent k having the dimension of inverse length, or equivalently a dimensionless exponent kd=kb where b is initial saturated thickness. The authors used their solution to analyze drawdown data from a pumping test conducted by Moench et al. [2001] in a Glacial Outwash Deposit at Cape Cod, Massachusetts. Their analysis yielded estimates of horizontal and vertical saturated hydraulic conductivities, specific storage, specific yield and k . Recognizing that hydraulic conductivity and water content seldom vary identically with incremental capillary pressure head, as assumed by Tartakovsky and Neuman [2007], we note that k is at best an effective rather than a directly measurable soil parameter. We therefore ask to what extent does interpretation of a pumping test based on the Tartakovsky-Neuman solution allow estimating aquifer unsaturated parameters as described by more common constitutive water retention and relative hydraulic conductivity models such as those of Brooks and Corey [1964] or van Genuchten [1980] and Mualem [1976a]? We address this question by showing how may be used to estimate the capillary air entry pressure head k and the parameters of such constitutive models directly, without a need for inverse unsaturated numerical simulations of the kind described by Moench [2003]. To assess the validity of such direct estimates we use maximum likelihood- based model selection criteria to compare the abilities of numerical models based on the STOMP code to reproduce observed drawdowns during the test when saturated and unsaturated aquifer parameters are estimated either in the above manner or by means of the inverse code PEST.
H13F-0993 INVITED
Linking Pore- and Darcy-scale Models of Reactive and Non-Reactive Transport
Pore-scale simulations of flow, transport, and reactions in porous media (in which the geometry of solid grains and pore spaces is explicitly quantified) are being used to demonstrate links between microscopic and macroscopic phenomena. We have developed pore-scale models of saturated and unsaturated flow, and of reactive transport in saturated media, based on the mesh-free method Smoothed Particle Hydrodynamics (SPH) and mesh-based computational fluid dynamics (CFD) methods. We are utilizing two approaches for representing pore-scale behavior in larger-scale models: 1) upscaling using volume-averaging and other methods in which macroscopic properties are estimated from pore-scale numerical simulations, and 2) hybrid multi-scale methods in which pore- and continuum-scale models in different sub-portions of a model domain are directly coupled. We illustrate these approaches based on applications in mixing-controlled mineral precipitation, intragranular dissolution and diffusion-controlled mass transfer, and effective pore-scale dispersion and dilution of conservative tracers.
H13F-0994
UPSCALING REACTIVE TRANSPORT USING A PORE-NETWORK MODEL
The main objective of this research is a better understanding of the relations between the adsorption rate constants and Darcy scale velocity under saturated and unsaturated flow conditions using a 3D pore-scale network model. First, we carry out numerical "experiments" in single tubes with circular and triangular cross sections imposing a concentration front moving into the tube and assuming equilibrium adsorption at its wall. Then we use volume-averaged concentration breakthrough curves to find upscaled rate constants as functions of tube geometry, subscale adsorption parameters, and average flow velocity. Next, we have constructed a 3D pore-network model which is composed of a large number of interconnected tubes. Transport equations for adsorbing solutes are solved within each tube. Single tube results are used in these equations. The pore-network model is used to simulate core-scale solute transport. Numerical data are used to calculate flux-averaged concentration and to determine upscaled kinetic rate coefficients. In this way, we have been able to develop a relationship between core-scale adsorption rate constants and local-scale equilibrium coefficient and average flow velocity.
H13F-0995 INVITED
Effect of Hierarchical, Multi-Scale Heterogeneity on Long-Term Nitrate Transport in a Deep Vadose Zone
Worldwide, unconsolidated sedimentary basins provide a significant groundwater resource. Deep
unsaturated zones are common within these sedimentary basins, particularly within arid and semi-arid climate
zones with strong groundwater development. For the vertical flow and transport through these unsaturated
sediments, two to three hierarchical heterogeneity scales can readily be identified: sedimentary sequences,
sedimentary strata within a sequence, and within-stratum heterogeneity. Based on a detailed field
reconnaissance, we developed three different representations of two-dimensional and three-dimensional
water flow and nitrate transport in a 16 m thick alluvial unsaturated zone below an irrigated orchard in semi-
arid Fresno County, California: The first model representation is a homogeneous lithofacies representation
which explicitly identifies major textural facies within the unsaturated zone. The second is a heterogeneous
lithofacies model which includes heterogeneity within each facies, where heterogeneity is represented by the
scaling factor technique (single-parameter heterogeneity). The third is also a heterogeneous model where
the heterogeneity in each lithofacies is characterized by random fields of each parameter in the van
Genuchten soil hydraulic function (multi-parameter heterogeneity). Scaling factors and van Genuchten
parameters are generated using a geostatistical model that is based on extensive field site borehole data
and undisturbed core analyses. We compared the effect of the different modeling representations on water
flow and nitrate transport during seven years of two different fertilization treatments. Estimated nitrate levels
in the vadose zone and nitrate leaching to groundwater differ significantly between fertilizer treatments, yet
are nearly identical between the three model representations and model dimensionality. In all cases, the
deep vadose zone nitrate mass is at least four times larger than the measured nitrate mass at the site. The
high discrepancies between the measured and model estimated nitrate at the site underscore the need for a
rigorous examination of alternative modeling strategies to simulating deep vadose zone flow and transport
processes, such as dual porosity flow and finger flow.
http://groundwater.ucdavis.edu/gw_205.htm