HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Figures Only Full Text (PDF) Free Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in Web of Science Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Web of Science (5) Citing Articles via Google Scholar Google Scholar Articles by Allred, B. J. Articles by Brown, G. O. Search for Related Content PubMed Articles by Allred, B. J. Articles by Brown, G. O. GeoRef GeoRef Citation Agricola Articles by Allred, B. J. Articles by Brown, G. O. Related Collections Soil Hydrology Soil Mineralogy Vadose Zone Processes and Chemical Transport

ORIGINAL RESEARCH

The Impact of Clay Mineralogy on Nitrate Mobility under Unsaturated Flow Conditions

^a USDA-ARS Soil Drainage Research Unit, 590 Woody Hayes Dr., Rm. 234, Columbus, OH 43210
^b School of Natural Resources, Ohio State Univ., 2021 Coffey Rd., 210 Kottman Hall, Columbus, OH 43210
^c Biosystems and Agricultural Engineering Dep., Oklahoma State Univ., 109 Agricultural Hall, Stillwater, OK 74078
^* Corresponding author (allred.13{at}osu.edu).

All rights reserved. No part of this periodical may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, recording, or any information storage and retrieval system, without permission in writing from the publisher.

Received 27 April 2006.

	ABSTRACT

TOP ABSTRACT INTRODUCTION Materials and Methods Results and Discussion Summary and Conclusions REFERENCES

 
Transient unsaturated horizontal column experiments were conducted to assess clay mineralogy impacts on electrostatic processes affecting nitrate (NO₃^–) mobility. Replicated tests were conducted on quartz sand, mixtures of the sand and kaolinite, illite, and montmorillonite, and two natural soils with organic matter removed. In each test, a 200 mg L^–1 nitrate–nitrogen (NO₃^––N) solution was injected at the inlet of dry soil columns. Comparison of corresponding NO₃^––N concentration and volumetric water content profiles from the column tests provided valuable information regarding soil mineral composition impacts on NO₃^– transport. With the exception of a small peak at the wetting front, NO₃^––N concentrations for the quartz sand were consistently near the 200 mg L^–1 injection level within the wetted portion of the columns, indicating that NO₃^– electrostatic interactions were negligible. Anion adsorption processes in the 25% kaolinite–75 % sand mixture produced a result in which the NO₃^––N concentrations adjacent to the inlet of the columns were approximately 20% greater than that of the injected solution. Anion exclusion was the dominant electrostatic interaction affecting NO₃^– mobility in the 25% illite–75% sand, 25% montmorillonite–75% sand, and 15% kaolinite–7.5% illite–7.5% montmorillonite–70% sand mixtures and in the two natural soils. Evidence of anion exclusion in these artificial and natural soils includes NO₃^––N concentrations near the column inlet that were 11 to 19% less than the injected solution concentration, and NO₃^––N concentrations near the wetting front that were greater than the injected solution concentration by factors of 1.7 to 5.4. These results indicate that anion adsorption is an important process affecting NO₃^– mobility in low pH soils, with limited amounts of organic matter, and having a clay-size fraction dominated by kaolinite, while anion exclusion is a key electrostatic interaction influencing NO₃^– mobility in near-neutral to high pH soils, especially if significant amounts of montmorillonite are present.

Abbreviations: CEC, cation exchange capacity • PZC, point of zero charge • XRD, X-ray diffraction.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION Materials and Methods Results and Discussion Summary and Conclusions REFERENCES

 
Nitrate is the most widespread contaminant found in groundwater (Freeze and Cherry, 1979; USEPA, 1990). In particular, inorganic and organic fertilizer application on farm fields is commonly responsible for NO₃^– contamination in shallow aquifers (Nolan and Stoner, 2000). Investigations conducted by the USGS (2001) have found that 20% of the shallow wells sampled in agricultural areas exceeded the USEPA drinking water standard of 10 mg L^–1 NO₃^––N. In addition, NO₃^– from fertilizer applied to midwestern U.S. farm fields is often intercepted by buried agricultural drainage pipes and then discharged into local waterways at concentrations commonly approaching 50 mg L^–1 NO₃^––N (Zucker and Brown, 1998). For most NO₃^– involved with aquifer contamination or degradation of midwestern U.S. surface waters, initial transport is through the soil profile. Nitrate mobility in the soil environment is in turn governed largely by electrostatic interactions between negatively charged NO₃^– ions and either soil clay minerals or soil organic matter.

Electrostatic attraction, or anion adsorption, occurs when NO₃^– ions become attached to positively charged exchange sites on a soil surface. A significant percentage of exchange sites on soil mineral and soil organic matter are pH dependent. Under low pH conditions, positively charged hydrogen ions become attached to certain mineral exchange site functional groups, thereby causing these exchange sites to become positively charged (Foth, 1984; Bohn et al., 1985). Examples include kaolinite, a layered aluminosilicate mineral, allophane, an amorphous aluminosilicate mineral, and various iron oxide-hydroxide and aluminum oxide-hydroxide minerals, all of which will develop a net positive charge given a sufficiently low pH environment (Sposito, 1984; Gustafsson, 2001).

The pH level of interest is defined by the point of zero charge (PZC), which is the pH value of a soil solution when the total net charge on a mineral particle is zero. If the pH is below the PZC, then the mineral particle has a positive charge, and if the pH is above the PZC, then the mineral particle has a negative charge. The PZC for kaolinite is 4.6 (Sparks, 2003) and proto-imogolite allophane has a PZC of 6 to 7 (Gustafsson, 2001). Reported PZC values for the iron hydroxide goethite ranges from 6.1 ± 0.6 (Sposito, 1984) to 7.3 (Stumm, 1992) to 7.8 (Stumm and Morgan, 1981). Illite and chlorite, two other common layered aluminosilicate minerals found in soil, have a moderate to substantial percentage of ion exchange sites that are pH dependent and therefore probably have PZC values similar to that of kaolinite. Consequently, NO₃^– mobility can potentially be reduced, through anion adsorption, in soils that have lower pH values and substantial amounts of kaolinite, illite, chlorite, allophane, and/or iron or aluminum oxides-hydroxides. Anion adsorption of NO₃^– has been documented for kaolinite-rich Ultisols in the southeastern USA (Toner et al., 1989; Eick et al., 1999), Andisols containing large amounts of allophane (Katou et al., 1996; Ryan et al., 2001), and Oxisols with iron and/or aluminum oxide-hydroxides from the tropics (Wong et al., 1990).

Electrostatic repulsion, or anion exclusion, occurs with NO₃^– when ion exchange sites on soil mineral or organic matter surfaces are negatively charged. If anion exclusion processes dominate, NO₃^– is repelled and does not come into contact with soil surfaces and, as a result, moves freely through the soil environment. Although pH dependent, the net charge on soil organic matter is always negative (Bohn et al., 1985). Dissociation of H⁺ from carboxyl and phenol functional groups increases from low to high pH, causing a corresponding increase in the net negative charge of soil organic matter. Nearly all ion exchange sites on the layered aluminosilicate minerals smectite and vermiculite are permanent (not pH dependent) and negatively charged (Bohn et al., 1985). The PZC value for smectite (montmorillonite) is 2.5 (Sparks, 2003), and vermiculite probably has a similar PZC value. Therefore, ignoring chemical and biological transformations, NO₃^– is expected to have high mobility in soil environments with moderately low to high pH and significant amounts of organic matter, smectite, and/or vermiculite.

Furthermore, given near-neutral to high pH soil environments, even kaolinite, illite, and chlorite particles have a net negative charge, thereby contributing to the anion exclusion process. The respective range of cation exchange capacities (CEC) for kaolinite, illite, chlorite, vermiculite, and smectite are 1 to 15, 10 to 40, 20 to 40, 100 to 150, and 70 to 120 cmolc kg^–1, respectively (Bohn et al., 1985, McBride 1994). These CEC values show that, for a given amount of material, the number of negatively charged ion exchange sites on vermiculite and smectite far outnumber the exchange sites for kaolinite, illite, and chlorite, implying that vermiculite and smectite will have the greatest impact on anion exclusion processes. Quartz, the most abundant mineral in the sand- and silt-size fraction, has a very low CEC and should have minimal electrostatic interaction with NO₃^–.

Steady-state saturated and unsaturated soil column leaching experiments have been used to assess the anion exclusion effect by measuring the effective percentage of pore water unavailable for transport of negative chloride (Cl^– ) ions (Thomas and Swoboda, 1970; Appelt et al., 1975; James and Rubin, 1986). These studies indicate that the percentage of pore water excluded decreases with increased Cl^– concentration (Thomas and Swoboda, 1970) and becomes greater with larger soil CEC (Appelt et al., 1975). James and Rubin (1986) demonstrated that the percentage of pore water excluded increased as the overall water content was reduced, which resulted from the actual exclusion volume remaining constant. Furthermore, steady-state saturated column testing by Shukla and Cepuder (2000) showed that anion exclusion volume for Cl^– decreases as pore-water velocity increases. Smith and Davis (1974) conducted steady-state unsaturated column tests on eight soils and found that the percentage of excluded water ranged from 5 to 39% and was independent of whether the anion was Br^– or NO₃^–. Melamed et al. (1994) demonstrated enhanced anion exclusion of Br^– with reduction in PZC of an Oxisol.

Bond et al. (1982) and Bond and Phillips (1990) quantified the anion exclusion effect on Cl^– transport using constant-flux transient unsaturated horizontal column experiments performed with a syringe pump on a clayey soil having a moderate initial volumetric water content of 0.16 to 0.18. Both studies found that, as a result of anion exclusion, the Cl^– concentration front extended significantly farther into the soil column than the theoretical "piston front" separating the injected solution from the initial soil solution. On the basis of the injected Cl^– concentration and the overall soil solution Cl^– concentration at the column inlet, Bond et al. (1982) developed an equation to calculate the effective column inlet water content for which Cl^– was excluded. Bond et al. (1982) also ascertained that the anion exclusion effect does not require the presence of immobile pore water. Bond and Phillips (1990) determined that the type of cation present in the injection solution, whether it was Na⁺, K⁺, or Ca²⁺, had little effect on Cl^– anion exclusion as indicated by the column inlet soil solution Cl^– concentration or the position of the Cl^– concentration front. Using variable-flux transient unsaturated horizontal soil column experiments with initial water contents of 0.18 and column inlet water contents maintained at approximately 0.55, Smiles and Gardiner (1982) measured the anion exclusion effect on Cl^– transport in a clay soil by noting the positional difference between the Cl^– concentration front and the piston front. Similar variable-flux transient unsaturated horizontal soil column experiments performed by Bond et al. (1984) showed that increasing the Cl^– concentration decreases the anion exclusion impact on Cl^– transport.

Phillips et al. (1988) conducted two transient unsaturated vertical column tests in dry soil, showing that anion exclusion processes produced peak Cl^– concentrations coinciding with the wetting front. Research at a larger scale with 6-m-long vertical soil columns and at field test plots confirmed that anion exclusion impacts the mobility of Cl^–, Br^–, and even sulfate (Gvirtzman et al., 1986; Hills et al., 1991; Porro and Wierenga, 1993; Porro et al., 1993). Allred et al. (2007) determined anion exclusion to be the dominant electrostatic process affecting NO₃^– transport under unsaturated flow conditions in four initially dry soils having widely different physicochemical and mineralogical properties. Additionally, Allred et al. (2007) found that for three of the four initially dry soils evaluated using transient unsaturated column tests, the wetting front NO₃^– concentration was significantly greater than the NO₃^– concentration of the solution injected at the column inlet.

Figure 1 depicts a pore-scale conceptualization of anion adsorption and anion exclusion processes under unsaturated flow conditions. For the purpose of this conceptualization, the water present in an unsaturated pore is divided into two zones. Zone 1 is adjacent and in direct contact with soil mineral and organic particles along the side of the pore. Zone 2 is closer to the center axis of the pore and is bounded on one side by the air–water interface and on the other side by Zone 1. The pore-water velocity has a parabolic distribution based on the Hagen–Poiseuille law for laminar flow through a small diameter circular tube (Daugherty et al., 1985; de Marsily, 1986; Tindall and Kunkel, 1999). From the side toward the center of the pore, the water velocity increases from zero at the soil–water interface and reaches a maximum value at the air–water interface. Therefore, as illustrated in Fig. 1, the average pore-water velocity in Zone 2 will be greater than the average pore-water velocity of Zones 1 and 2 combined, while the average pore-water velocity in Zone 1 will be less than the average pore-water velocity on Zones 1 and 2 combined.

View larger version (25K): [in this window] [in a new window]   FIG. 1. Pore-scale conceptualization of anion adsorption and anion exclusion processes under unsaturated flow conditions.

Solving environmental problems associated with NO₃^– will require, at least in part, computer modeling programs that are capable of accurately predicting NO₃^– movement through the soil profile. Computer modeling programs used to simulate nitrogen movement and transformations within the soil profile (Davidson et al., 1978; Selim and Iskandar, 1981; Ma et al., 1999; Hanson et al., 1999) could potentially be improved greatly by incorporating NO₃^– anion adsorption and exclusion processes, once there is a better understanding of these effects under unsaturated flow conditions. Although there have been some limited field investigations of NO₃^– anion adsorption within very specific soil environments, almost all laboratory research into anion transport processes in soil has focused only on tracers such as Br^–, and particularly Cl^–, but not NO₃^–, which is not only an important agricultural nutrient but also the most widespread environmental contaminant. Investigation of NO₃^– anion adsorption and exclusion effects under unsaturated flow conditions has been especially limited.

The types and amounts of clay minerals present will almost certainly impact NO₃^– mobility within the soil profile. However, little research has been done to date on this aspect of NO₃^– transport in unsaturated soil. To fill this research need, an investigation was initiated to rigorously examine the effects of clay mineralogy on NO₃^– transport under unsaturated flow conditions, using transient unsaturated horizontal column experiments to assess clay mineralogical impacts on soil NO₃^– mobility. Characteristics of the water content and NO₃^– concentration profiles obtained from these column tests were used to quantify anion adsorption–exclusion processes affecting NO₃^– transport. The NO₃^– anion adsorption–exclusion effects of kaolinite vs. illite vs. montmorillonite were compared based on test results with four artificial soils: (i) sand (100%), (ii) kaolinite (25%) and sand (75%), (iii) illite (25%) and sand (75%), and (iv) montmorillonite (25%) and sand (75%). In addition, one artificial soil (kaolinite [15%], illite [7.5%], montmorillonite [7.5%], and sand [70%]) and two natural soils (Teller loam and Slaughterville sandy loam) were tested to assess NO₃^– transport under a more typical soil scenario, in which combinations of clay minerals are present. The governing hypothesis for this research project can be stated as follows: Anion adsorption is an important process affecting NO₃^– mobility in low pH soils, with limited organic matter, having a clay-size fraction dominated by kaolinite, while anion exclusion is the most significant electrostatic process influencing NO₃^– mobility under near-neutral to high pH soil conditions, especially if significant quantities of montmorillonite are present.

	Materials and Methods

TOP ABSTRACT INTRODUCTION Materials and Methods Results and Discussion Summary and Conclusions REFERENCES

 
Soil Materials
Clean quartz sand and mixtures of the sand and one or more reference clay minerals (kaolinite, illite, and montmorillonite) were used for part of the study. The course- to medium-grained quartz sand was purchased from ELE International LLC (Loveland, CO). Samples of kaolinite (KGa-1b, Washington County, GA), illite (IMt-2, Silver Hill, MT), and montmorillonite (a member of the smectite group of clay minerals, SAz-1, Apache County, AZ) were obtained from the Source Clays Repository of the Clay Minerals Society (West Lafayette, IN). The kaolinite and montmorillonite were delivered in powdered form, while the illite came as crushed shale that was then ground to a powder. Values for pH and CEC for the sand and clay minerals are provided in Table 1. The pH values of the sand, illite, and montmorillonite were modestly above neutral, while the kaolinite pH was quite low. The sand and kaolinite had very low CEC values, the illite had a moderate CEC, and the montmorillonite, as expected, had a high CEC (Table 1).

The two natural soil materials tested were each collected at a depth between 5 and 20 cm beneath the ground surface, at locations near Perkins, OK, one from the Slaughterville series (course-loamy, mixed, superactive, thermic Udic Haplustolls) and the other from the Teller series (fine-loamy, mixed, active, thermic Udic Argiustolls). Geographically, soils of the Teller and Slaughterville series are found in the Central Rolling Red Prairies within the Great Plains region of the USA, where NO₃^– contamination of aquifers is often a problem. To focus strictly on the effects of soil mineral (especially clay) composition on nitrate mobility, we removed organic matter from both the Slaughterville and Teller soils by oxidation using a 30% hydrogen peroxide solution (Klute, 1986). Following organic matter oxidation, these soils were oven dried at 100°C to remove residual hydrogen peroxide. Afterward, the Slaughterville and Teller soils were washed with a 2 mol L^–1 CaCl₂ solution to saturate cation exchange sites with Ca²⁺ and to displace any adsorbed NO₃^–. This initial wash was followed by six additional washes with distilled water to remove excess salts and any NO₃^– originally in the soil. At the end of the complete washing cycle, the Slaughterville and Teller samples were again dried at 100°C, this time to suppress microbial activity and ensure that column tests were conducted with initially dry soil material. Finally, the soils were ground to pass a 2-mm sieve.

Table 2 gives the pH, CEC, and particle-size distribution of the Slaughterville and Teller soils. Values for pH were determined from the soil solution of a saturated paste (Richards, 1954). Cation exchange capacity values were obtained by first saturating the exchange sites with sodium cations, followed by extraction of the sodium with a solution of ammonium acetate (Black, 1965). Wray (1986) described methods used to determine the soil particle-size distribution. Based on classification of the particle-size distribution, the Slaughterville is a sandy loam and the Teller is a loam. The Slaughterville soil, as shown in Table 2, has greater pH, a lower CEC, and a smaller percentage of clay-size particles than the Teller soil.

View this table: [in this window] [in a new window]   TABLE 3. Slaughterville and Teller soil mineralogy (particle size <2 µm).

Testing of Nitrate Transport
Nitrate mobility under unsaturated flow conditions was assessed using transient unsaturated horizontal column experiments. Transient unsaturated horizontal flow in a soil column can be described by the following relationship:

[1]

where t is time (T) from test initiation, x is distance (L) from the column inlet, is volumetric water content (dimensionless), and D() is the soil water diffusivity (L²/T), itself a function of . Soil water diffusivity can in turn be expressed as

[2]

where is pore-water pressure potential (L) and K() is unsaturated hydraulic conductivity (L/T), which is dependent on . Bruce and Klute (1956) showed that by using the Boltzmann transformation, = x/t,, having dimensions L T^–0.5, Eq. [1] can be reduced to an ordinary differential equation of the form

[3]

Equation [3] is valid for quantifying unsaturated flow in column tests only if the following water content boundary conditions are maintained:

[4]

and

[5]

The boundary condition given by Eq. [4] simply indicates that the volumetric water content beyond the wetting front is the same as the initial volumetric water content, _i, of the soil column. The second boundary condition, Eq. [5], states that the volumetric water content at the inlet of the soil column, ₀, is kept constant throughout the test.

A computer-controlled syringe pump apparatus described by Brown and Allred (1992) and shown in Fig. 2 was used to inject a 200 mg L^–1 NO₃^––N solution (1.44 g KNO₃ L^–1) into the inlet of dry soil columns. A computer-controlled syringe pump, which was first used by Bond (1984), allows flexibility in maintaining the inlet volumetric water content at a constant value (Eq. [5]) throughout the time duration of the test. The Eq. [5] boundary condition was accomplished by regulating the instantaneous injected flow at a rate inversely proportional to the square root of elapsed time. The specific equation for the instantaneous injected flow rate, Q (L³ T^–1), is

[6]

where A is the cross-sectional area of the soil column (L²), S is the sorptivity (L T^–0.5), and t is the time (T) since the experiment began. Sorptivity can be expressed in the following form:

[7]

where V is the total volume (L³) of the of the solution injected at the inlet of the soil column and t_T is the total test duration time (Brown and Allred, 1992). Substituting the last term in Eq. [7] into Eq. [6] yields

[8]

showing that the proportionality constant between Q and t^–1/2 in the injection rate equation is simply a function of the chosen injection volume and test duration that are both input into the computer controlling the syringe pump. Consequently, for consistency between experiments, the same S of 0.0073 cm s^–0.5 and A of 9.35 cm² was used in each column experiment, giving a consistent instantaneous injection rate of

[9]

It is important to note here that the actual value of the inlet volumetric water content maintained during a test depends not only on the programmed injection rate equation but also on soil hydraulic properties.

View larger version (65K): [in this window] [in a new window]   FIG. 2. (a) Schematic of the computer-controlled syringe pump apparatus, (b) the computer-controlled syringe pump apparatus, (c) syringe pump and horizontal soil column with controller in background, and (d) syringes inserted into column inlet.

Each test was programmed for a set time duration and 200 mg L^–1 NO₃^––N solution (1.44 g KNO₃ L^–1) injection volume. To determine if results could be replicated, three tests were conducted with 100% sand, and two tests were performed for each of the sand–soil mixtures and the two natural soils. The replicate tests for a particular soil material had different time durations and solution injection volumes but, as already noted, the same sorptivity, injection rate function, and hence, inlet water content, ₀ (Eq. [5] boundary condition). Table 4 provides information on column length, time duration, injection volume, dry bulk density, and initial water content, _i, for all 15 tests performed in this investigation. Table 4 indicates that all replicated tests for a given soil material had essentially the same column packing density and _i (Eq. [4] boundary condition). Packed dry bulk densities ranged from approximately 1.6 to 1.9 g cm^–3. All _i values in Table 4 are less than 0.011 (average = 0.005), indicating that all soil material was very dry before column test initiation.

Soil material samples for determining soil solution NO₃^––N concentration were placed in 50 mL centrifuge tubes, and 45 mL of 1 M KCl solution then added to disperse soil particles and desorb NO₃^–. The centrifuge tubes containing soil and solution were next placed on a shaker at 300 rpm for 1 h to obtain thorough mixing. After being shaken, each soil–solution mixture was centrifuged at 800 g (2500 rpm) for 1 h. Following the centrifuge step, soil and solution were refrigerated. The final step involved filtering and analyzing the centrifuged solutions for NO₃^––N with a Lachat Instruments QuikChem 8000 Flow Injection Analysis System (Milwaukee, WI), using a modified version of the cadmium reduction column method (USEPA, 1979). For all 15 experiments, an average of 100.1% (SD = 3.3%) of the injected NO₃^––N was accounted for by laboratory analysis.

Data Analysis
Values of volumetric water content and NO₃^––N concentration (reported with respect to total soil solution) from each experiment were plotted against the Boltzmann transformation (distance from the column inlet divided by the square root of the test duration time). Assessment of NO₃^– adsorption or exclusion processes was based on data from these column profiles for volumetric water content and soil solution NO₃^––N concentration. Hypothetical volumetric water content and soil solution NO₃^––N profiles plotted versus the Boltzmann transformation are shown in Fig. 3 . These transient unsaturated horizontal column profiles are representative of soil conditions in a column that is initially dry, _i 0, and has no nitrate (NO₃^–) present, C_i (initial concentration) = 0. There are three hypothetical NO₃^––N concentration profiles, one based on no electrostatic interactions, one the result of anion adsorption, and one the result of anion exclusion (Fig. 3). Given no electrostatic interactions between NO₃^– and soil surfaces, and assuming hydrodynamic dispersion to be negligible, the NO₃^––N soil solution concentration profile will have a constant value from the column inlet to the wetting front equaling that of the original injected solution.

View larger version (30K): [in this window] [in a new window]   FIG. 3. Ideal transient unsaturated horizontal column water content and soil solution NO3––N concentration profiles for an initially dry soil with no NO3––N initially present.

When anion exclusion is the dominant electrostatic interaction, NO₃^– is repelled from soil surfaces and is concentrated more toward the center of the pore, where it travels at a rate faster than the overall average pore-water velocity. As a consequence of anion exclusion and initial test conditions that include dry soil with no NO₃^– present, the NO₃^––N concentration profile will exhibit lower values near the column inlet and higher values at the wetting front (Fig. 3). The magnitude of the wetting front NO₃^––N concentration peak produced by anion exclusion will be significantly greater than the NO₃^––N concentration of the original injected solution, while at the column inlet, soil solution NO₃^––N values will be less than the NO₃^––N concentration of the original injected solution. (The inlet NO₃^––N concentration represents an average of the NO₃^––N concentrations in both the "more-accessible" and "less-accessible" pore waters.)

The inlet water content and NO₃^––N concentration conditions will depend only on the solution injection rate function, soil physiochemical and clay mineralogical properties, and injection solution characteristics, and not the initial soil wetness or dryness. Consequently, the same anion exclusion processes are in operation during a column test, regardless of whether the soil is initially dry or moist. For this research, peak NO₃^––N concentrations at the wetting front greater than the original injection solution concentration (200 mg L^–1 NO₃^––N) are a product of anion exclusion processes coupled with wetting front penetration into a relatively dry soil. For a soil that is initially much wetter, a NO₃^––N concentration peak might not even develop, but instead, the NO₃^––N concentration front would penetrate significantly farther into the column than the theoretical "piston front" separating injected water from displaced initial water (Smiles and Gardiner, 1982; Bond et al., 1982; Bond and Phillips, 1990). Again, the same anion exclusion processes are in operation whether a soil is initially dry or moist. However, since the piston and actual wetting fronts essentially coincide for tests performed with initially dry soils, the amount of NO₃^––N that would normally be transported beyond the piston front in tests conducted with initially moist soil will instead simply build up at the wetting front for tests performed with a dry soil.

A simple method for quantifying whether anion adsorption or anion exclusion processes dominate NO₃^– transport in unsaturated soil columns involves comparing X_C (Boltzmann transformation value at the centroid of a column test NO₃^––N concentration profile) to X_C-NEI, (Boltzmann transformation value at the centroid for a theoretical NO₃^––N concentration profile based on no electrostatic interactions). The centroid is simply the location of the center "balance" point of a geometrical shape such as a water content or NO₃^––N concentration profile, and its calculation is described in numerous calculus and engineering mechanics textbooks (Leithold, 1976; Meriam and Kraige, 1986). The calculation for X_C-NEI assumes a 200 mg L^–1 NO₃^––N concentration from the inlet to the wetting front position obtained with the actual experiment. If X_C < X_C-NEI, anion adsorption is indicated, and if X_C > X_C-NEI, anion exclusion is implied.

If anion adsorption processes dominate, the NO₃^––N concentration near the column inlet, C₀, should equal the maximum NO₃^––N profile concentration, C_MAX, and will be greater than the NO₃^––N concentration of the original injected solution, C_Orig. Consequently, given anion adsorption, C₀/C_Orig = C_MAX/C_Orig > 1, and the value of C₀/C_Orig or C_MAX/C_Orig can then be used to gauge the magnitude of the anion adsorption effect. The greater C₀/C_Orig or C_MAX/C_Orig, the greater the anion adsorption effect.

The magnitude of the anion exclusion effect in a transient unsaturated horizontal column experiment can be quantified several ways. Anion exclusion, as already noted, involves NO₃^– being repelled from soil surfaces and moving at a rate faster than the average pore-water velocity, in turn causing the NO₃^––N concentration near the column inlet, C₀, to be less than the NO₃^––N concentration of the original injected solution, C_Orig. Therefore, one measure of the magnitude of the anion exclusion effect is the ratio C₀/C_Orig, which is less than 1 when anion exclusion dominates. The ratio C₀/C_Orig can then be used to calculate the effective excluded pore-water volume at the inlet, _EX, that is unavailable to NO₃^–. The value of _EX is calculated in the following manner (Bond et al., 1982; Fetter, 1993):

[10]

where ₀ is the volumetric water content at the column inlet. The proportion of inlet moisture content that is excluded becomes simply _EX/₀. Equation [10] is based on an idealized model in which the overall column inlet pore water is divided into two zones. The first zone, in which NO₃^– is excluded, is adjacent and in direct contact with soil mineral or organic matter surfaces, and has a NO₃^––N concentration equal to zero. The second zone, with a NO₃^––N concentration equal to C_Orig, is separated from soil mineral and organic matter surfaces by the NO₃^––excluded first zone. Although the pore-scale model reflected by Eq. [10] is not completely realistic (see Fig. 1), the Eq. [10] relationship does allow certain inlet parameters to be determined in a simple manner, and these inlet parameters are useful for quantifying the magnitude of the NO₃^– anion exclusion effect.

Focusing efforts to quantify NO₃^– anion adsorption–exclusion effects based on column inlet conditions, C₀/C_Orig for anion adsorption and C₀/C_Orig, _EX, and _EX/₀ for anion exclusion, was appropriate in this study for two reasons. First, it is the inlet portion of the column that has been the most thoroughly flushed by the injection solution. Second, given physical and chemical equilibrium conditions and regardless of initial soil moisture conditions, the water content and NO₃^––N concentration at the column inlet will have specific values that remain constant throughout the time duration of a test.

Because anion exclusion results in C₀ < C_Orig, and given initial test conditions with dry soil and no NO₃^– present, mass balance considerations alone suggest that there will be some soil solution NO₃^––N concentrations greater than C_Orig. Therefore, the ratio of the maximum soil solution NO₃^––N concentration to C_Orig, or C_MAX/C_Orig, is greater than 1 and is an additional valid measure of the magnitude of the anion exclusion effect. As defined by these quantities, the larger the magnitude of the NO₃^– anion exclusion effect, the smaller the value of C₀/C_Orig, the greater the value of _EX, the greater the value of _EX/₀, and the greater the value of C_MAX/C_Orig.

	Results and Discussion

TOP ABSTRACT INTRODUCTION Materials and Methods Results and Discussion Summary and Conclusions REFERENCES

 
Profiles of the transient unsaturated horizontal column experimental data are provided in Fig. 4 and 5 . Water content and NO₃^––N concentration values are plotted with respect the Boltzmann transformation. The vertical scales on the NO₃^––N concentration profiles are varied to better display data (Fig. 4 and 5). For each soil material, different symbols are used to separate the data from the two or three individual tests conducted with that particular soil material. The two or three column experiments for each soil material had the same hydraulic boundary conditions (Eq. [4] and [5]) but different time durations and NO₃^––N solution injection volumes.

View larger version (26K): [in this window] [in a new window]   FIG. 4. Experimental data profiles for (a) 100% sand, water content, (b) 100% sand, NO3–-N concentration, (c) 25% kaolinite, 75% sand, water content, (d) 25% kaolinite, 75% sand, NO3–-N concentration, (e) 25% illite, 75% sand, water content, (f) 25% illite, 75% sand, NO3–-N concentration, (g) 25% montmorillonite, 75% sand, water content, and (h) 25% montmorillonite, 75% sand, NO3–-N concentration.

View larger version (29K): [in this window] [in a new window]   FIG. 5. Experimental data profiles for (a) 15% kaolinite, 7.5% illite, 7.5% montmorillonite, 70% sand, water content, (b) 15% kaolinite, 7.5% illite, 7.5% montmorillonite, 70% sand, NO3––N concentration, (c) Slaughterville, water content, (d) Slaughterville, NO3––N concentration, (e) Teller, water content, and (f) Teller, NO3––N concentration. Organic matter was removed from the Slaughterville and Teller soils before testing.

[11]

The boundary condition given by Eq. [11] states that the NO₃^––N inlet concentration is kept constant throughout the test.

Data from the column tests for each soil were used to construct a composite volumetric water content profile and a composite soil solution NO₃^––N concentration profile for each soil material. The characteristics of the composited water content and NO₃^––N concentration profiles for each soil material were quantified and are presented in Table 5. Wetting penetration was quantified by calculating, X_W, the Boltzmann transformation value at the composited water content profile centroid. The greatest wetting penetration, based on an X_W value of 0.0324 cm s^–0.5, occurred with the 100% sand, which was expected because clean sands typically have high hydraulic conductivity. The kaolinite–sand mixture, illite–sand mixture, and Slaughterville soil all had intermediate wetting penetrations as determined by X_W values of 0.0177, 0.0230, and 0.0196 cm s^–0.5, respectively. The montmorillonite–sand mixture, kaolinite–illite–montmorillonite–sand mixture, and Teller soil all had fairly low wetting penetrations as verified by X_W values of 0.0106, 0.0140, and 0.0129 cm s^–0.5, respectively.

The soil solution NO₃^––N concentration profiles for the 100% quartz sand (Fig. 4b) are similar to the hypothetical NO₃^––N concentration profile displayed in Fig. 3, based on no electrostatic interactions. The value of X_C was 0.0389 cm s^–0.5, only slightly larger than the 0.0380 cm s^–0.5 value of X_C-NEI, a further indication that electrostatic interactions did not substantially impact NO₃^– mobility in the 100% sand (Table 5). From the column inlet and almost to the wetting front (Fig. 4a), values of NO₃^––N for the 100% quartz sand are approximately equal to C_Orig.

The column tests with 100% sand did show a modest NO₃^––N concentration peak located at the wetting front (Fig. 4a and 4b) that on average was greater than C_Orig by a factor of 1.4 (Table 5). This is most likely the result of a very weak anion exclusion effect. The sand had a CEC of 0.8 cmolc kg^–1 (Table 1), and after preparation, this material had a fairly neutral pH of 6.58, indicating, given PZC considerations, that a small amount of net negative surface charge was present that would contribute to a weak anion exclusion effect, in turn producing a modest NO₃^––N concentration peak at the wetting front. One cannot rule out evaporation and vapor phase transport at the wetting front as another possible explanation to account for the modest NO₃^––N concentration peak found at the wetting front. Regardless of the actual process affecting the wetting front NO₃^––N concentration, however, it is clear overall that there was very little impact as a result of electrostatic interactions on NO₃^– mobility in the 100% quartz sand material. Consequently, any anion adsorption or anion exclusion effects exhibited by the sand–clay mixtures or the natural soils tested were due to the clay minerals present and not sand- or silt-size quartz particles.

The configuration of the NO₃^––N concentration profiles for the kaolinite–sand mixture (Fig. 4d) is similar to the hypothetical anion adsorption NO₃^––N concentration profile displayed in Fig. 3. Specifically, the highest NO₃^––N concentrations are adjacent to the column inlet and are greater than the 200 mg L^–1 NO₃^––N concentration of the injected solution (C₀/C_Orig = C_MAX/C_Orig = 1.2, Table 5). This result indicates significant NO₃^– adsorption occurring near the inlet. Beyond a Boltzmann transformation value of 0.3 cm s^–0.5, the NO₃^––N values begin to decrease and reach zero at the wetting front (Fig. 4c and 4d). Additional evidence of anion adsorption affecting NO₃^– mobility in the kaolinite–sand mixture is an X_C value of 0.0179 cm s^–0.5 that is less than the corresponding X_C-NEI value of 0.0210 cm s^–0.5 (Table 5). The pH for the kaolinite–sand mixture is 4.08, well below the PZC for kaolinite. As a result, the kaolinite particles in this soil material mixture had a net positive surface charge, in turn promoting anion adsorption processes that reduced NO₃^– mobility.

The configuration of the soil solution NO₃^––N concentration profiles for the illite–sand and montmorillonite–sand mixtures (Fig. 4f and 4h) are quite similar to that of the hypothetical anion exclusion affected NO₃^––N concentration profile displayed in Fig. 3. Specifically, NO₃^––N concentrations at the column inlet are less than the 200 mg L^–1 injected concentration, and the highest NO₃^––N concentrations were found in a peak positioned at the wetting front (Fig. 4e–h). Additionally, X_C > X_C-NEI for the illite–sand and montmorillonite–sand mixtures (Table 5). Consequently, there is little doubt that anion exclusion is the dominant electrostatic interaction impacting NO₃^– mobility in these two soil material mixtures. This result is not surprising because the illite–sand mixture had a pH of 7.85, and the montmorillonite–sand mixture had a pH of 7.44. These pH values are clearly above the PZC for illite or montmorillonite, meaning that the surface charge on the illite or montmorillonite particles present in these two sand–clay mixtures had a net negative charge, thereby promoting NO₃^– anion exclusion.

The quantities used to gauge the magnitude of the anion exclusion effect are described in the Materials and Methods section and in Table 5. Compared with the illite–sand mixture, the montmorillonite–sand mixture had a lower C₀/C_Orig and greater _EX, _EX/₀, and C_MAX/C_Orig (Table 5), clearly indicating that the anion exclusion effect on NO₃^– mobility was greater with the montmorillonite–sand than the illite–sand. This is because the amount of surface charge on the montmorillonite particles is substantially larger than the amount of surface charge on the illite particles (Table 1). It is interesting to note that the peak NO₃^––N concentrations that occurred at the wetting front for the illite–sand and montmorillonite–sand mixtures were greater than C_Orig by factors of 3.95 and 5.4, respectively (Table 5). Approximately 10% of the NO₃^––N injected at the column inlet ended up at the wetting front concentration peak in tests conducted with the illite–sand mixture, while an impressive 50% of the NO₃^––N injected at the column inlet ended up at the wetting front concentration peak in tests conducted with the montmorillonite–sand mixture. These results demonstrate that in dry soils, with little NO₃^– initially present, and having near-neutral or higher pH, anion exclusion not only makes the NO₃^– that is added more mobile but can potentially produce high concentration NO₃^– "pulses" that move through the soil profile.

Testing conducted on the kaolinite–illite–montmorillonite–sand mixture and the two natural soils, Slaughterville and Teller (organic matter removed), indicated that all three exhibited NO₃^––N concentration profiles (Fig. 5b, 5d, and 5f) similar to that of the hypothetical anion exclusion affected NO₃^––N concentration profile displayed in Fig. 3. The lowest NO₃^––N concentrations for these soil materials were adjacent to the column inlet. The highest NO₃^––N concentrations were found in a peak located at the wetting front (Fig. 5). Accordingly, X_C > X_C-NEI for the kaolinite–illite–montmorillonite–sand mixture and the Slaughterville and Teller soils (Table 5). Therefore, the evidence is strong that anion exclusion is the dominant electrostatic interaction impacting NO₃^– mobility in these three soil materials. Most important, the results show anion exclusion to be an important process affecting NO₃^– mobility not only in some artificial sand–clay mixtures but also in natural soils that commonly contain a variety of clay minerals. (Comparison of some of the test results obtained by Allred et al. [2007] with the test results acquired in this study with the Slaughterville and Teller soils indicates that for these two natural soils, neither organic matter removal by hydrogen peroxide oxidation nor oven drying the soil at 100°C alters clay mineralogy or soil physicochemical properties to the extent that it affects water or NO₃^– movement.)

Based on C₀/C_Orig, _EX, _EX/₀, and C_MAX/C_Orig (Table 5), no clear trend indicates whether the anion exclusion effect on NO₃^– mobility was greatest in the kaolinite–illite–montmorillonite–sand mixture, the Slaughterville soil, or the Teller soil. Consequently, these quantities may not be appropriate for comparing the magnitude of the anion exclusion effect between different soils having a variety of clay minerals. Rather, C₀/C_Orig, _EX, _EX/₀, and C_MAX/C_Orig may be better used on an individual soil to gauge changes in the anion exclusion effect due to varying conditions within that particular soil. The kaolinite–illite–montmorillonite–sand mixture, the Slaughterville soil, and the Teller soil had pH values of 7.39, 7.35, and 6.53, respectively (Table 2). These near-neutral pH values and considerations with regard to PZC indicate that essentially all the clay mineral particle surfaces within these three soil materials had a net negative charge, which accounted for the strong anion exclusion effect influencing NO₃^– mobility. From considerations of clay mineral CEC, the montmorillonite present in these three soil materials undoubtedly had the greatest impact on NO₃^– anion exclusion effect. The peak NO₃^––N concentrations that occurred at the wetting front for the kaolinite–illite–montmorillonite–sand mixture, the Slaughterville soil, and the Teller soil were greater than C_Orig by factors of 4.40, 1.95, and 1.7, respectively (Table 5). With respect to the total amount NO₃^––N injected at the column inlet, 17 to 26% ended up in wetting front concentration peaks for tests performed with the kaolinite–illite–montmorillonite–sand mixture and the two natural soils. These results further confirm that in dry soils, with near-neutral to high pH, and containing a variety of clay minerals, anion exclusion not only makes NO₃^– more mobile but can potentially produce high concentration NO₃^– "pulses" that move through the soil profile.

	Summary and Conclusions

TOP ABSTRACT INTRODUCTION Materials and Methods Results and Discussion Summary and Conclusions REFERENCES

 
Nitrate mobility within the soil environment is governed to a significant extent by electrostatic interactions between negatively charged NO₃^– ions and charged soil particle surfaces (either mineral or organic). Anion adsorption occurs when NO₃^– ions become attached to positively charged exchange sites on a soil surface. When anion exclusion processes dominate, NO₃^– ions are repelled from soil particle surfaces and move at a rate faster than the overall average pore-water velocity. Transient unsaturated horizontal column experiments were conducted to assess clay mineralogy impacts on the electrostatic processes affecting NO₃^– transport in soil under initially dry conditions. Replicated tests were conducted on quartz sand, mixtures of the sand and one or more clay minerals (kaolinite, illite, and montmorillonite), and two natural soils (with organic matter removed). In each test, a 200 mg L^–1 NO₃^––N solution was injected at the inlet of dry soil columns. Comparison of corresponding NO₃^––N concentration (reported with respect to soil solution) and volumetric water content profiles from the column tests provides valuable information regarding the impact of soil mineral composition on NO₃^– transport.

With the exception of a small peak at the wetting front, concentration profile values for a 100% quartz sand were consistently around 200 mg L^–1 NO₃^––N elsewhere within the wetted portion of the columns, indicating extremely weak electrostatic interactions involving NO₃^–. Anion adsorption processes in a kaolinite–sand mixture produced a result in which the NO₃^––N concentrations adjacent to the inlet of the columns were approximately 20% greater than that of the injected solution. The pH for the kaolinite–sand mixture was 4.08, well below the PZC for kaolinite. As a result, the kaolinite particles in this soil material mixture had a net positive surface charge, in turn promoting anion adsorption processes that reduced NO₃^– mobility.

Anion exclusion proved to be the dominant electrostatic interaction affecting NO₃^– mobility in the illite–sand, montmorillonite–sand, and kaolinite–illite–montmorillonite–sand mixtures and in the Slaughterville and Teller series soils from Oklahoma. One significant implication regarding these findings is that anion exclusion is an important process affecting NO₃^– mobility not only in artificial sand–clay mixtures but also in natural soils containing a variety of clay minerals. Evidence of anion exclusion in these sand–clay mixtures and natural soils includes soil solution NO₃^––N concentrations near the column inlet that were 11 to 19% less than the solution concentration originally injected at the inlet. Further strong anion exclusion evidence includes peak soil solution NO₃^––N concentrations that were greater than 200 mg L^–1, by factors of 1.7 to 5.4, found at column wetting fronts for the sand–clay mixtures and natural soils containing significant amounts of illite and montmorillonite. Consequently, anion exclusion processes can produce high-concentration NO₃^– "pulses" that move through dry soils, such as those present in arid environments. The strongest anion exclusion effect was found with the montmorillonite–sand mixture, indicating that montmorillonite, compared with most other clay minerals, will have the greatest NO₃^– anion exclusion impact. Overall results show that anion adsorption is an important process affecting NO₃^– mobility in low pH soils, with limited organic matter, having a clay-size fraction dominated by kaolinite, while anion exclusion is a key electrostatic interaction influencing NO₃^– mobility in near-neutral to high pH soils, especially when significant amounts of montmorillonite is present.

	ACKNOWLEDGMENTS

 
The authors wish to express their appreciation to Dedra Woner for conducting the soil nitrate analysis, and Sandy Jones for the soil mineralogy lab work.

	REFERENCES

TOP ABSTRACT INTRODUCTION Materials and Methods Results and Discussion Summary and Conclusions REFERENCES

 

• Allred, B.J., G.O. Brown, and J.M. Bigham. 2007. Nitrate mobility under unsaturated flow conditions in four initially dry soils. Soil Sci. 172:27–41.
• Appelt, H., K. Holtzclaw, and P.F. Pratt. 1975. Effect of anion exclusion on the movement of chloride through soils. Soil Sci. Soc. Am. Proc. 39:264–267.
• Black, C.A. (ed.) 1965. Methods of soil analysis: Part 2. Chemical and microbiological properties. Agron. Monogr. 6. ASA, Madison, WI.
• Bohn, H.L., B.L. McNeal, and G.A. O'Connor. 1985. Soil chemistry. 2nd ed. John Wiley & Sons, New York.
• Bond, W.J. 1984. Flux-controlled constant potential absorption using a programmable pump. Soil Sci. Soc. Am. J. 48:943–946.[Abstract/Free Full Text]
• Bond, W.J., B.N. Gardiner, and D.E. Smiles. 1982. Constant-flux absorption of a tritiated calcium chloride solution by a clay soil with anion exclusion. Soil Sci. Soc. Am. J. 46:1133–1137.[Web of Science]
• Bond, W.J., B.N. Gardiner, and D.E. Smiles. 1984. Movement of CaCl₂ solutions in an unsaturated soil: The effect of solution concentration. Aust. J. Soil Res. 22:3–58.
• Bond, W.J., and I.R. Phillips. 1990. Ion transport during unsteady water flow in an unsaturated clay soil. Soil Sci. Soc. Am. J. 54:636–645.[Web of Science]
• Borden, D., and R.F. Giese. 2001. Baseline studies of the Clay Minerals Society source clays: Cation exchange capacity measurements by the ammonia–electrode method. Clays Clay Miner. 49:444–445.[Free Full Text]
• Brown, G.O., and B. Allred. 1992. The performance of syringe pumps in unsaturated horizontal column experiments. Soil Sci. 154:243–249.
• Bruce, R.R., and A. Klute. 1956. The measurement of soil moisture diffusivity. Soil Sci. Soc. Am. Proc. 20:458–462.
• Busenberg, E., and C.V. Clemency. 1973. Determination of the cation exchange capacity of clays and soils using an ammonia electrode. Clays Clay Miner. 21:213–217.[CrossRef][Web of Science]
• Daugherty, R.L., J.B. Franzini, and E.J. Finnemore. 1985. Fluid mechanics with engineering applications. 8th ed. McGraw-Hill, New York.
• Davidson, J.M., D.A. Graetz, P.S.C. Rao, and H.M. Selim. 1978. Simulation of nitrogen movement, transformation, and uptake in plant root zone. Technical Report EPA-600/3-78-029. USEPA, Athens, GA.
• de Marsily, G. 1986. Quantitative hydrogeology. Academic Press, San Diego, CA.
• Eick, M.J., W.D. Brady, and C.K. Lynch. 1999. Charge properties and nitrate adsorption of some acid southeastern soils. J. Environ. Qual. 28:138–144.[Web of Science]
• Fetter, C.W. 1993. Contaminant hydrogeology. Macmillan, New York.
• Foth, H.D. 1984. Fundamentals of soil science. 7th ed. John Wiley & Sons, New York.
• Freeze, R.A., and J.A. Cherry. 1979. Groundwater. Prentice Hall, Englewood Cliffs, NJ.
• Gvirtzman, H., D. Ronen, and M. Magaritz. 1986. Anion exclusion during transport through the unsaturated zone. J. Hydrol. 87:267–283.
• Gustafsson, J.P. 2001. Modelling competitive anion adsorption on oxide minerals and an allophane-containing soil. Eur. J. Soil Sci. 52:639–653.[CrossRef]
• Hanson, J.D., K.W. Rojas, and M.J. Shaffer. 1999. Calibrating the root zone water quality model. Agron. J. 91:171–177.[Abstract/Free Full Text]
• Hillel, D. 1980. Fundamentals of soil physics. Academic Press, San Diego, CA.
• Hills, R.G., P.J. Wierenga, D.B. Hudson, and M.R. Kirkland. 1991. The second Las Cruces experiment: Experimental results and two-dimensional flow prediction. Water Resour. Res. 27:2707–2718.[CrossRef]
• James, R.V., and J. Rubin. 1986. Transport of chloride ion in a water-saturated soil exhibiting anion exclusion. Soil Sci. Soc. Am. J. 50:1142–1149.[Web of Science]
• Katou, H., B.E. Clothier, and S.R. Green. 1996. Anion transport involving competitive adsorption during transient water flow in an Andisol. Soil Sci. Soc. Am. J. 60:1368–1375.[Web of Science]
• Klute, A. (ed.). 1986. Methods of soil analysis: Part 1. Physical and mineralogical methods. 2nd ed. Agron. Monogr. 9. ASA and SSSA, Madison, WI.
• Leithold, L. 1976. The calculus with analytic geometry. Harper & Row, New York.
• Ma, L., C.W. Lindau, C. Hongprayoon, W. Burhan, B.C. Jang, W.H. Patrick, Jr., and H.M. Selim. 1999. Modeling urea, ammonium, and nitrate transport and transformations in flooded soil columns. Soil Sci. 164:123–132.[CrossRef]
• McBride, M.B. 1994. Environmental chemistry of soils. Oxford Univ. Press, New York.
• Melamed, R., J.J. Jurinak, and L.M. Dudley. 1994. Anion exclusion–pore water velocity interaction affecting transport of bromine through an Oxisol. Soil Sci. Soc. Am. J. 58:1405–1410.[Web of Science]
• Meriam, J.L., and L.G. Kraige. 1986. Engineering mechanics. Vol. 1. Statics. John Wiley & Sons, New York.
• Moore, D.M., and R.C. Reynolds, Jr. 1997. Sample preparation techniques for clay minerals. p. 211–213. In X-ray diffraction and the identification and analysis of clay minerals. 2nd ed. Oxford Univ. Press., New York.
• Nolan, B.T., and J.D. Stoner. 2000. Nutrients in ground waters of the conterminous United States, 1992–1995. Environ. Sci. Technol. 34:1156–1165.
• Page, A.L. (ed.) 1982. Methods of soil analysis: Part 2. Chemical and microbiological properties. 2nd ed. Agron. Monogr. 6. ASA and SSSA, Madison, WI.
• Phillips, F.M., J.L. Mattick, T.A. Duval, D. Elmore, and P.W. Kubik. 1988. Chlorine 36 and tritium from nuclear weapons fallout as tracers for long-term liquid and vapor movement in desert soils. Water Resour. Res. 24:1877–1891.
• Porro, I., and P.J. Wierenga. 1993. Transient and steady-state solute transport through a large unsaturated soil column. Ground Water 31:193–200.[CrossRef][Web of Science]
• Porro, I., P.J. Wierenga, and R.G. Hills. 1993. Solute transport through large uniform and layered soil columns. Water Resour. Res. 29:1321–1330.[CrossRef]
• Richards, L.A. (ed.) 1954. Diagnosis and improvement of saline and alkaline soils. USDA Agricultural Handbook 60. U.S. Gov. Print. Office, Washington, DC.
• Rutledge, E.M., L.P. Wilding, and M. Elfield. 1967. Automated particle size separation by sedimentation. Soil Sci. Soc. Am. Proc. 31:287–288.
• Ryan, M.C., G.R. Graham, and D.L. Rudolph. 2001. Contrasting nitrate adsorption in Andisols of two coffee plantations in Costa Rica. J. Environ. Qual. 30:1848–1852.[Abstract/Free Full Text]
• Selim, H.M., and I.K. Iskandar. 1981. Modeling nitrogen transport and transformations in soils: Theoretical considerations. Soil Sci. 131:233–241.
• Shukla, M.K., and P. Cepuder. 2000. Anion exclusion during transport of chloride through soil columns. Trans. ASAE 43:1425–1430.[Web of Science]
• Smith, S.J., and R.J. Davis. 1974. Relative movement of bromide and nitrate through soils. J. Environ. Qual. 3:152–155.[Abstract/Free Full Text]
• Smiles, D.E., and B.N. Gardiner. 1982. Hydrodynamic dispersion during unsteady, unsaturated water flow in a clay soil. Soil Sci. Soc. Am. J. 46:9–14.[Web of Science]
• Smiles, D.E., and J.R. Philip. 1978. Solute transport during absorption of water by soil: Laboratory studies and their practical implications. Soil Sci. Soc. Am. J. 42:537–544.[Abstract/Free Full Text]
• Sparks, D.L. 2003. Environmental soil chemistry. 2nd ed. Academic Press, San Diego, CA.
• Sposito, G. 1984. The surface chemistry of soils. Oxford Univ. Press, New York.
• Stumm, W. 1992. Chemistry of the solid–water interface: Processes at the mineral–water and particle–water interface in natural systems. John Wiley & Sons, New York.
• Stumm, W., and J.J. Morgan. 1981. Aquatic chemistry: An introduction emphasizing chemical equilibria in natural waters. John Wiley & Sons, New York.
• Thomas, G.W., and A.R. Swoboda. 1970. Anion exclusion effects on chloride movement in soils. Soil Sci. 110:163–166.
• Tindall, J.A., and J.R. Kunkel. 1999. Unsaturated zone hydrology for scientists and engineers. Prentice Hall, Upper Saddle River, NJ.
• Toner, C.V., IV, D.L. Sparks, and T.H. Carski. 1989. Anion exchange chemistry in Middle Atlantic soils: Charge properties and nitrate retention kinetics. Soil Sci. Soc. Am. J. 53:1061–1067.[Web of Science]
• USEPA. 1979. Methods for chemical analysis of water and wastes. U.S. Gov. Print. Office, Washington, DC.
• USEPA. 1990. National water quality inventory: 1988 report to Congress. EPA 440-4-90-003. U.S. Gov. Print. Office, Washington, DC.
• USGS. 2001. Selected findings and current perspectives on urban and agricultural water quality by the National Water-Quality Assessment Program. U.S. Geological Survey Fact Sheet 047-01. USGS, Reston, VA.
• Wong, M.T.F., R. Huges, and D.L. Rowell. 1990. Retarded leaching of nitrate in acid soils from the tropics: Measurement of effective anion exchange capacity. J. Soil Sci. 41:655–663.[CrossRef]
• Wray, W.K. 1986. Measuring engineering properties of soil. Prentice Hall, Englewood Cliffs, NJ.
• Zucker, L.A., and L.C. Brown. 1998. Agricultural drainage: Water quality impacts and subsurface drainage studies in the Midwest. Ohio State Univ. Ext. Bull. 871, Ohio State Univ., Columbus.

This article has been cited by other articles:

				F. S. J. Martinez, Y. A. Pachepsky, and W. J. Rawls Advective-Dispersive Equation with Spatial Fractional Derivatives Evaluated with Tracer Transport Data Vadose Zone J., March 5, 2009; 8(1): 242 - 249. [Abstract] [Full Text] [PDF]

				T. J. Reilly, N. S. Fishman, and A. L. Baehr Effect of Grain-Coating Mineralogy on Nitrate and Sulfate Storage in the Unsaturated Zone Vadose Zone J., February 10, 2009; 8(1): 75 - 85. [Abstract] [Full Text] [PDF]

This Article Abstract Figures Only Full Text (PDF) Free Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in Web of Science Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Web of Science (5) Citing Articles via Google Scholar Google Scholar Articles by Allred, B. J. Articles by Brown, G. O. Search for Related Content PubMed Articles by Allred, B. J. Articles by Brown, G. O. GeoRef GeoRef Citation Agricola Articles by Allred, B. J. Articles by Brown, G. O. Related Collections Soil Hydrology Soil Mineralogy Vadose Zone Processes and Chemical Transport