Unsaturated Hydraulic Conductivity Data Research Articles

Soil Classification: A New Approach for Grouping Soils Using Unsaturated Hydraulic Conductivity Data

AbstractGrouping soils based on similarities in their textural, taxonomic, and/or structural properties has broad applications to pedology, hydrology, and soil science. In this study, we present a new approach for classifying soils using hydraulic conductivity data. We apply concepts from critical path analysis and calculate critical pore sizes at various water saturations from the unsaturated hydraulic conductivity curves. Soils with similar critical pore size at the same effective water saturation are then grouped into the same class. To demonstrate the practical application of the proposed soil classification method, we use 102 samples including nine soil textures from the UNSODA database. Applying a curve clustering method, we find eight different soil classes within the studied data set.

The importance of unsaturated hydraulic conductivity measurements for green roof detention modelling

Characterising the unsaturated hydraulic conductivity of a green roof substrate is essential for accurately modelling runoff detention in response to rainfall events. In this paper, the unsaturated hydraulic conductivities for four representative green roof substrates were determined in an infiltration column using steady state and transient techniques. The conventional Durner-Mualem Hydraulic Conductivity Function (HCF) model, for which parameters were calibrated based on the measured Soil Water Release Curve (SWRC) data, was shown to provide a poor fit to the experimental data. A new three-segment HCF was, therefore, proposed to fit measured unsaturated hydraulic conductivity data. Detention tests were carried out on 100 mm and 200 mm deep substrates using four simulated storm events. The runoff and moisture content data collected during the detention tests was used to validate the HCFs using the Richards Equation. The new three-segment HCF resulted in simulated runoff and moisture content profiles that closely matched the measured data (with mean Rt2 = 0.754 for modelled runoff), in contrast to predictions made using the conventional Durner-Mualem model (with mean Rt2 = 0.409 for modelled runoff). It was also demonstrated that further simplification of the HCF to a function defined by moisture content at just two points – the saturated hydraulic conductivity and at an unsaturated hydraulic conductivity of 0.1 cm/min – provides a model that is fit-for-purpose for green roof runoff estimation (with mean Rt2 = 0.629 for modelled runoff).

Scaling of Hydraulic Functions in Heterogeneous Soil Using Nonlinear Least Squares Minimization Method

Presenting soil heterogeneity precisely in various spatial scales is the main key to simulate water and solute transport through it. The method described by Richards is mostly used to study water flow through vadose zone. It requires spatial representation of hydraulic functions and water retention relationship in the soil. To represent the spatial relationship of soil hydraulic functions, scaling approach is being used since the last few decades. In this study, a simple scaling method using nonlinear least squares minimization technique has been used to scale soil matric potential, hydraulic conductivity as well as simultaneous scaling of soil matric potential and hydraulic conductivity data. Simultaneous scaling is necessary as it reduces the volume of data by producing a single set of scale factors for hydraulic functions in a heterogeneous soil. Van Genuchten’s semi-empirical expressions were used in this study to parameterize soil hydraulic functions. Results showed that correlation coefficient from raw and descaled data was superior when soil matric potential and hydraulic conductivity data were scaled separately than simultaneously. Improvement of correlation coefficient in simultaneous scaling can be obtained by adding more weight to the soil matric potential data than unsaturated hydraulic conductivity data, which enhances the overall correlation coefficient in simultaneously scaling. Statistical analysis of the scale factors showed that they are lognormally distributed. Scale factors calculated by solving simple equations obtained using the method described in this study can be used to simulate water movement through heterogeneous soil conditions using HYDRUS model.

Soil hydraulic functions of international soils measured with the Extended Evaporation Method (EEM) and the HYPROP device

The paper presents a data base of 173 soil hydrological data (raw data) from 71 sites all over the world (Asia, Africa, Australia America and Europe). The samples were mainly collected and measured as part of research projects. The soils cover a wide range of texture classes and dry bulk densities. The data base consists of water retention and unsaturated hydraulic conductivity data. Additional data of the particle size distribution, the dry bulk density, the organic matter content and other variables are included. The hydraulic data were measured with the Extended Evaporation Method (EEM) and the HYPROP device in the range between saturation and close to the wilting point. Information to the geo reference, the soil type, the sampling depth and the horizon are given. Data access via DOI 10.4228/ZALF.2003.273.

A New Approach to Estimate Soil Hydraulic Parameters Using Only Soil Water Retention Data

Traditional approaches for estimating soil hydraulic parameters (such as the RETC code) perform well when experimental data for both the retention curve and hydraulic conductivity function are available; however, unsaturated hydraulic conductivity data are often unavailable. The objective of this work was to develop an approach to estimate robust soil hydraulic parameters from water retention curve data alone. The proposed approach, called the Multiobjective Retention Curve Estimator (MORE), is based on the Multiobjective Shuffled Complex Evolution Metropolis (MOSCEM‐UA) algorithm and estimates an optimal parameter set by simultaneously minimizing two objective functions each representing water content and relative unsaturated hydraulic conductivity residuals. To address the lack of observed unsaturated hydraulic conductivity data, MORE transforms both predicted and observed water contents into the hydraulic conductivity space using a pore‐size distribution model (e.g., the Mualem model) and also optimizes the soil hydraulic parameters in this fictitious space. We applied MORE to two cases. In the first case, MORE was used to estimate soil parameters using only retention curve data for 12 random soils selected from the UNSODA database. While the soil hydraulic parameters estimated using RETC and MORE fit the retention curve similarly, the MORE approach consistently decreased the error in fitting unsaturated hydraulic conductivities by as much as 5% compared with RETC. The second case involved using the parameters fitted using the MORE and RETC approaches to model a field‐scale experiment. Compared with RETC, the error in predicted water contents decreased by 25% using parameters predicted by MORE. The MORE approach was shown to fit robust soil hydraulic parameters; however, the approach is relatively slower and more time consuming than RETC.

Estimation of effective unsaturated hydraulic conductivity tensor using spatial moments of observed moisture plume

Knowledge of unsaturated zone hydraulic properties is critical for many environmental and engineering applications. Various stochastic methods have been developed during the past 2 decades to estimate the effective unsaturated hydraulic properties. Independent of these stochastic methods, we develop in this paper a practical approach to estimate the three‐dimensional (3‐D) effective unsaturated hydraulic conductivity tensor using spatial moments of 3‐D snapshots of a moisture plume under transient flow conditions. An application of the new approach to data collected at the Hanford Site in southeastern Washington State yields an effective unsaturated hydraulic conductivity tensor that exhibits moisture‐dependent anisotropy. The effective hydraulic conductivities compare well with laboratory‐measured unsaturated hydraulic conductivity data from small core samples; they also reproduce the general behavior of the observed moisture plume at the site. We also define a moisture diffusivity length concept which we use in conjunction with estimated correlation scales of the geologic media at the field site to explain deviations between the observed and simulated plumes based on the derived effective hydraulic properties.

Neural Networks Prediction of Soil Hydraulic Functions for Alluvial Soils Using Multistep Outflow Data

Indirect methods for prediction of soil hydraulic properties play an important role in understanding site‐specific unsaturated water flow and transport processes, usually via numerical simulation models. Specifically, pedotransfer functions (PTFs) to predict soil‐water retention have been widely developed. However, few datasets that include unsaturated hydraulic conductivity data are available for prediction purposes. Moreover, those available employ a variety of measurement techniques. We show that prediction of soil‐water retention and unsaturated hydraulic conductivity curves from basic soil properties can be improved if hydraulic data are determined by a single measurement method that is consistently applied to all soil samples. Here, we present a unique dataset that consists of 310 soil‐water retention and unsaturated hydraulic conductivity functions, all of which were obtained from the multistep outflow method. With this dataset, neural networks coupled with bootstrap aggregation were used to predict the soil‐water retention and hydraulic conductivity characteristics from basic soil properties, that is, sand, silt, and clay content, bulk density (ρb), saturated water content, and saturated hydraulic conductivity. The prediction errors of water content were about 3 to 4% by volume. Unsaturated hydraulic conductivity predictions improved significantly when a so‐called performance‐based algorithm was utilized to minimize residuals of soil hydraulic data rather than hydraulic parameters. The root mean squared of residuals for predicted values of water content and unsaturated hydraulic conductivity were reduced by about 50% when compared with predicted hydraulic functions using a published neural networks program Rosetta Results from a sensitivity analysis suggest that the hydraulic parameters are mostly sensitive to sand content and saturated water content.

Neural Networks Prediction of Soil Hydraulic Functions for Alluvial Soils Using Multistep Outflow Data

Indirect methods for prediction of soil hydraulic properties play an important role in understanding site-specific unsaturated water flow and transport processes, usually via numerical simulation models. Specifically, pedotransfer functions (PTFs) to predict soil-water retention have been widely developed. However, few datasets that include unsaturated hydraulic conductivity data are available for prediction purposes. Moreover, those available employ a variety of measurement techniques. We show that prediction of soil-water retention and unsaturated hydraulic conductivity curves from basic soil properties can be improved if hydraulic data are determined by a single measurement method that is consistently applied to all soil samples. Here, we present a unique dataset that consists of 310 soil-water retention and unsaturated hydraulic conductivity functions, all of which were obtained from the multistep outflow method. With this dataset, neural networks coupled with bootstrap aggregation were used to predict the soil-water retention and hydraulic conductivity characteristics from basic soil properties, that is, sand, silt, and clay content, bulk density (ρb), saturated water content, and saturated hydraulic conductivity. The prediction errors of water content were about 3 to 4% by volume. Unsaturated hydraulic conductivity predictions improved significantly when a so-called performance-based algorithm was utilized to minimize residuals of soil hydraulic data rather than hydraulic parameters. The root mean squared of residuals for predicted values of water content and unsaturated hydraulic conductivity were reduced by about 50% when compared with predicted hydraulic functions using a published neural networks program Rosetta Results from a sensitivity analysis suggest that the hydraulic parameters are mostly sensitive to sand content and saturated water content.

Improved Prediction of Unsaturated Hydraulic Conductivity with the Mualem‐van Genuchten Model

In many vadose zone hydrological studies, it is imperative that the soil's unsaturated hydraulic conductivity is known. Frequently, the Mualem–van Genuchten model (MVG) is used for this purpose because it allows prediction of unsaturated hydraulic conductivity from water retention parameters. For this and similar equations, it is often assumed that a measured saturated hydraulic conductivity (Ks) can be used as a matching point (Ko) while a factor SLe is used to account for pore connectivity and tortuosity (where Se is the relative saturation and ). We used a data set of 235 soil samples with retention and unsaturated hydraulic conductivity data to test and improve predictions with the MVG equation. The standard practice of using and resulted in a root mean square error for log(K) (RMSEK) of 1.31. Optimization of the matching point (Ko) and L to the hydraulic conductivity data yielded a RMSEK of 0.41. The fitted Ko were, on average, about one order of magnitude smaller than measured Ks Furthermore, L was predominantly negative, casting doubt that the MVG can be interpreted in a physical way. Spearman rank correlations showed that both Ko and L were related to van Genuchten water retention parameters and neural network analyses confirmed that Ko and L could indeed be predicted in this way. The corresponding RMSEK was 0.84, which was half an order of magnitude better than the traditional MVG model. Bulk density and textural parameters were poor predictors while addition of Ks improved the RMSEK only marginally. Bootstrap analysis showed that the uncertainty in predicted unsaturated hydraulic conductivity was about one order of magnitude near saturation and larger at lower water contents.

NEW METHOD FOR DETERMINING HYDRAULIC PROPERTIES OF UNSATURATED SOIL USING EVAPORATION DATA

Solving soil unsaturated flow problems requires knowledge of the water retention θ(ψ) and unsaturated hydraulic conductivity k (θ) relationships.This study introduces a new evaporation method for simultaneous estimation of both retention and unsaturated hydraulic conductivity data. Newton-Raphson iterative method was used to solve non-linear equations in time. Simultaneous transport of liquid and vapor flow in the unsaturated soil was modeled. Evaporation from a vertical column of soil material (tile) was simulated. Hydraulic parameters were estimated by minimizing the sum of squared differences between measured and simulated evaporation rate through the whole period of the experiment. Two laboratory experiments were carried out on the same sample under different conditions to evaluate the ability of this new method to provide good estimation of the retention and unsaturated conductivity curves. The results show that this method is stable, accurate, and can be applied for other types of soil.

Fractal processes in soil water retention

Numerous empirical models exist for soil water retention and unsaturated hydraulic conductivity data. It has generally been recognized that the empirical fitting coefficients in these models are somehow related to soil texture. However, the fact that they are empirical means that elaborate laboratory experiments must be performed for each soil to obtain values for the parameters. Moreover, empirical models do not shed insight into the fundamental physical principles that govern the processes of unsaturated flow and drainage. We propose a physical conceptual model for soil texture and pore structure that is based on the concept of fractal geometry. The motivation for a fractal model of soil texture is that some particle size distributions in granular soils have already been shown to display self‐similar scaling that is typical of fractal objects. Hence it is reasonable to expect that pore size distributions may also display fractal scaling properties. The paradigm that we use for the soil pore size distribution is the Sierpinski carpet, which is a fractal that contains self similar “holes” (or pores) over a wide range of scales. We evaluate the water retention properties of regular and random Sierpinski carpets and relate these properties directly to the Brooks and Corey (or Campbell) empirical water retention model. We relate the water retention curves directly to the fractal dimension of the Sierpinski carpet and show that the fractal dimension strongly controls the water retention properties of the Sierpinski carpet “soil”. Higher fractal dimensions are shown to mimic clay‐type soils, with very slow dewatering characteristics and relatively low fractal dimensions are shown to mimic a sandy soil with relatively rapid dewatering characteristics. Our fractal model of soil water retention removes the empirical fitting parameters from the soil water retention models and provides parameters (fractal dimension) which are intrinsic to the nature of the fractal porous structure. The relative permeability functions of Burdine and Mualem are also shown to be fractal directly from fractal water retention results.

Hydraulic properties and ‘field capacity’ of Himatangi sand

Abstract Bulk density, particle size, soil water retentivity, and saturated and unsaturated hydraulic conductivity data from Himatangi sand are reported. The soil under a 4 m2 area was wet by ponding and the subsequent decline in soil water contentmonitored using a neutron probe. The drainage rate from the top 550 mm of the soil profile was greater than 1 mm/day for 6 daysafter ponding ceased, indicating that the commonly accepted concept of field capacity is inappropriate for this soil. Simple theory is developed which allows prediction of drainage losses under different soil water and evapotranspiration conditions. If the upper limit of readily available water is taken as the storage when the drainage rate from a 550 mm deep root zone is 1 mm/day, and the lower limit is taken as the storage at a matric potential of-50 kPa, then the readily available water holding capacity is only 8 mm. This suggests that carefully controlled high frequency irrigation is necessary if both drainage losses and plant water ...

Comparison of measured and estimated unsaturated hydraulic conductivity

Most studies of empirical estimates of unsaturated hydraulic conductivity functions do not account for water which may be relatively immobile under the conditions in which field measurements of conductivity are made. To investigate this, unsaturated hydraulic conductivity data were obtained for three monolith lysimeters, 80 cm in diameter by 135 cm deep, using the instantaneous profile technique. The lysimeters contained well‐structured, freely draining loam soil and moisture measurements were made using a neutron probe. Conductivity estimates were also obtained from laboratory measurements of soil moisture characteristics using the modified Millington and Quirk computational method. Ratios of the calculated to measured conductivities at a matching point near saturation were so large as to suggest that only a minor proportion of the soil pore space was contributing to flow through the whole profile.

Scaling field‐measured soil hydraulic properties using a similar media concept

With a view toward developing techniques for studying water flow in spatially varying soils, field data for soil water characteristic relationships and unsaturated hydraulic conductivity are scaled by using the concept of similar media. Data observed by different investigators at three geographic areas are used. The soil water characteristic data consisted of 840, 900, and 512 observations, while those for the unsaturated hydraulic conductivity (available from only one of the three sources) consisted of 2640 observations. In the process of scaling the data a best fit for the scaled data is defined in terms of a sum of squares about an ‘average’ curve using one value for the scaling parameter for each sampling location. Comparisons made between curves fitting the data and those fitting the scaled data show that scaling reduces the sums of squares by amounts varying from 34 to over 90%. For similar media the scaling parameter determined at a given sampling location for the soil water characteristic relationship (α from h(S)) should be identical to that for the unsaturated hydraulic conductivity (αr from K(S)). Although these parameters were highly correlated (r = 0.91 ) for the only set of data available, α r from h(S) values are shown to be more effective in scaling the unsaturated hydraulic conductivity data than are αr from K(S) values in scaling soil water characteristic data.