Published online 21 June 2006
Published in Soil Sci Soc Am J 70:1272-1280 (2006)
DOI: 10.2136/sssaj2005.0291
© 2006 Soil Science Society of America
677 S. Segoe Rd., Madison, WI 53711 USA
Soil Physics
Hydraulic Conductivity, Immobile Water Content, and Exchange Coefficient in Three Soil Profiles
L. Allettoa,
Y. Coquetb,*,
P. Vachierb and
C. Labatb
a Ecole Supérieure d'Agriculture de Purpan–75, voie du TOEC BP 57 611, 31 076 Toulouse Cedex 3, France
b UMR INRA/INAPG Environment and Arable Crops, Institut National de la Recherche Agronomique/Institut National Agronomique Paris-Grignon, BP 01, 78 850 Thiverval-Grignon, France
* Corresponding author (coquet{at}grignon.inra.fr)
 |
ABSTRACT
|
|---|
We used tension disk infiltrometers to determine the hydraulic conductivity at the matric potential of –0.1 kPa (K–0.1), the immobile water fraction (
im/), and the mass exchange coefficient (
) in three soil profiles of an agricultural field cropped with winter wheat (Triticum aestivum L.). A significant effect of soil horizonation was found for K–0.1, im/, and and could be related to soil structure as determined either by tillage for surface horizons or by pedogenetic factors for subsurface horizons. The seedbed had lower K–0.1 values (31 to 99 mm h–1) than the plowed layer under the seedbed (94–430 mm h–1). The im/ was highly variable in the seedbed (0.213 to 0.882), while had values ranging from 0.0006 to 0.0115 h–1. Values of im/ and in the plow layer under the seedbed were similar but less variable than those found in the seedbed. In the subsoil, clay-enriched illuvial horizons had K–0.1 values similar to those found in the seedbed, but lower than those of the heavy clay horizon (average value of 260 mm h–1) and the limestone horizons (average value of 450 mm h–1). In the illuvial and heavy clay horizons, both im/ and parameters had a small variability. In the limestone, we found a high variability of both im/ values (0.506–0.933) and values (0.0006–0.0424 h–1) associated with high average values. The presence of subsoil horizons with large im/ and low values may have a significant impact on ground water contamination.
Abbreviations: DFBA, 2,6-diflurobenzoate • MIM, Mobile–Immobile Model • PFBA, pentafluorobenzoate • TFBA, trifluoromethylbenzoate
|
INTRODUCTION
|
|---|
THE LARGEST AQUIFER in France is the limestone aquifer of the Beauce region with an estimated groundwater resource of nearly 20 Mm3. The Beauce region is also one of the most intensive agricultural areas in France. This intensive agriculture has led to contamination of the Beauce groundwater by several pesticides (Institut Français de l'Environnement, 2004). Understanding the mechanisms of groundwater recharge is essential to assess pollutant transfer to the water table. Many studies have shown that water and solutes can move through soils according to different kinds of preferential flow processes such as macropore flow, fingering, or nonequilibrium flow (van Genuchten and Wierenga, 1976; Thomas and Philips, 1979; Baker and Hillel, 1990; Brusseau et al., 1992; Jarvis, 1998). Using these preferential flow pathways, solutes can move rapidly below the biologically active root zone down to subsoil layers and the vadose zone, where degradation rates are generally low (Mills et al., 2001; Wood et al., 2001).
To better describe field and laboratory observations, solute transport models including physical nonequilibrium have been developed. The Mobile–Immobile Model (MIM) considers that the water-filled pore space is partitioned into two domains: a mobile domain where water can move and solute transport is due to advection and dispersion, and an immobile domain where water is stagnant and solutes move only by diffusion (Coats and Smith, 1964; van Genuchten and Wierenga, 1976). According to this concept, the one-dimensional transport of a nonsorbing conservative solute under steady-state water flow can be written
 | [1] |
where m and im [L3 L–3] are the mobile and immobile volumetric water contents, Cm and Cim [M L–3] are the solute concentrations in the mobile and immobile domains, t [T] is time, Dm [L2 T–1] is the hydrodynamic dispersion coefficient for the mobile domain, z [L] is depth, and q [L T–1] is the water flux density. Exchange between the two domains is described by the relation
 | [2] |
where [T–1] is the first-order mass exchange coefficient between the mobile and immobile domains.
A field method to estimate im/ and was proposed by Jaynes et al. (1995). The method uses a sequence of nonreactive conservative tracers applied to the soil by using a tension infiltrometer under steady-state water flow. Bromide and fluorobenzoates are commonly used tracers. They are considered to be conservative and inert, are easily extracted and quantified from soil, and have a negligible background in most soils (Bowman, 1984a, 1984b; Benson and Bowman, 1994; Kung et al., 2000a, 2000b; Flury and Wai, 2003).
This method has been applied to field conditions mainly on surface soil layers (Casey et al., 1997; Lee et al., 2000; Al-Jabri et al., 2002). References about MIM parameters in subsurface layers are scarce. Understanding the transport properties of solutes in subsoil horizons is essential to describe solute behavior in the whole soil profile and assess the groundwater pollution risks by contaminants. The objective of this study was to evaluate the MIM solute transport parameters and near-saturated hydraulic conductivity in the different horizons from the surface down to the 1-m depth in three soil profiles of an agricultural field. Statistical analysis was used to detect horizonation effects on hydraulic and MIM parameters.
|
MATERIALS AND METHODS
|
|---|
Field Site
The 25-ha field site is situated in the Beauce agricultural region, France. The field had been plowed to a depth of 28 cm in September 2003 and a 7-cm-deep seedbed was prepared before winter wheat (Triticum aestivum L.) was sown in November 2003. Field measurements of hydraulic conductivity and MIM parameters were performed in the last week of April and the first 3 wk of May 2004.
A strong soil heterogeneity was found in the field. According to the World Reference Base for Soil Resources (ISSS-ISRIC-FAO, 1998) the two main soil types are Orthic Luvisol and Eutric Cambisol. The substrate, which appeared at a variable depth, is a limestone having different levels of alteration.
Three 200-m2 plots were selected to represent the main soil types encountered in the field. Plots 1 and 3 corresponded to Orthic Luvisols, and Plot 2 to a Eutric Cambisol. In each plot, excavations were dug to describe and sample the soil profile (Table 1). Samples were collected from the 28-cm-thick plowed organo-mineral layer (LA1, LA2, and LA3) and from subsoil layers (BT1, BT3, B3, C1, C2, and C3) from each of the three plots. In Plots 1 and 3, BT1 and BT3 corresponded to clay-enriched illuvial horizons and had similar soil characteristics. In Plot 3, B3 corresponded to a heavy red clay horizon. Horizons C1, C2, and C3 corresponded to the weathered limestone. Each sample was characterized by its particle size distribution (g kg–1 dry soil) after removal of the organic matter by H2O2 and decarbonation by HCl (Association Française de Normalisation, 1983), its organic C content (g kg–1 dry soil) by sulfochromic oxidation (Association Française de Normalisation, 1999b), its pH in water, and its CaCO3 content (g kg–1 dry soil) by the volumetric method (Association Française de Normalisation, 1999a).
View this table:
[in this window]
[in a new window]
|
Table 1. Characteristics of the surface (LA1, LA2, LA3) and subsurface soil horizons (BT1, C1, C2, BT3, B3, C3) of the three plots (1, 2, and 3).
|
|
Hydraulic Conductivity
We used tension disk infiltrometers to establish constant water infiltration rates at a soil water pressure of –0.1 kPa. For each soil horizon, we did two to four replicates located at 50 cm from each other horizontally. An 8-cm-diameter base was used and contact between the disk and the soil surface was ensured by a fine, well-sorted Fontainebleau sand (Coquet et al., 2005). The cumulative infiltration, I (mm), could be expressed by a two-term equation (Haverkamp et al., 1994; Vandervaere et al., 2000a):
 | [3] |
where S(i,f) (mm h–0.5) is sorptivity, i and f are initial and final volumetric water contents respectively, A (mm h–1) is a parameter describing the effects of gravity and lateral absorption by soil on infiltration, and t is elapsed time (h). Hydraulic conductivity, K–0.1 (mm h–1), could be obtained using the following equation for parameter A (Haverkamp et al., 1994):
 | [4] |
where ß is a constant in the interval (0, 1) taken as 0.6 (Vandervaere et al., 2000b), r is the radius of the disk permeameter, and 
is a constant equal to 0.7 in most soils. Derivatives of I vs. 
t were used as a function of t to calculate K–0.1 as recommended by Vandervaere et al. (2000a).
Tracer Infiltrations
After water flow had reached steady-state infiltration, a sequence of four anionic tracer solutions was applied at the same water pressure (–0.1 kPa). The four tracers applied were Br–, pentafluorobenzoate (PFBA), 2,6-diflurobenzoate (DFBA), and trifluoromethylbenzoate (TFBA) at concentrations of 250 mg L–1 for Br– and PFBA and 200 mg L–1 for DFBA and TFBA. These high concentrations were necessary to limit the disturbances produced by a recent N application on the field. Two infiltrometers were alternately filled with the four solutions. The first solution contained only Br–. The second solution contained Br– and PFBA, the third Br–, PFBA, and DFBA, and the fourth solution contained all the tracers. The time of application for each tracer varied and depended on the infiltration rate. We did two to four replicate measurements for each soil horizon. After the penetration of the last solution, the infiltrometer was removed and the contact material scraped off. Soil samples were taken beneath the center of the disk using a 25-mm-diameter stainless steel cylinder at 10-mm intervals down to 100 mm (in the absence of coarse elements). Only the first centimeter under the infiltrometer was used to determine im and . Other samples were used to established concentration profiles of each tracer. Samples were transported to the laboratory in hermetic jars and stored at 4°C until analysis. The gravimetric water content was determined using a part of each sample and was used to calculate the bulk density, 
b (g cm–3), and the volumetric water content of the whole sample based on its total wet mass.
Tracer Analysis
The soil was extracted using a 1:1 soil/deionized water ratio. Each sample was shaken for 24 h then centrifuged for 15 min at 9000g. The supernatant was filtered through a 0.45-µm nylon filter. Analysis of the fluorobenzoate tracers was done using a Waters HPLC Alliance chain equipped with a photodiode array detector (Waters 996, Waters Corp., Milford, MA) with an anion exchange column (Platinum Sax 100A, 5u 250 mm by 4.6 mm, Alltech France, Templemars, France). The mobile phase was a KH2PO4 solution at 0.005 mol L–1 with 20% of acetonitrile and a flow of 2 mL min–1. The detection wavelength was 205 nm. The pH was adjusted to 2.2 by adding H3PO4. Injection volume was 50 µL. The duration of the analysis was 15 min. Under these conditions, retention times of TFBA, DFBA, PFBA, and Br– were 2.9, 5.6, 9.2, and 12.5 min, respectively. In some samples, NO3–, with a retention time of 13.1 min, has perturbed the Br– analysis. In these cases, Br– ions were quantified using the same HPLC with another anion exchange column (A-2 Anion 7u 100 mm by 4.6 mm, Alltech). The eluting solution was 2.2 mmol L–1 Na2CO3 and 2.8 mmol L–1 NaHCO3 with a flow of 2 mL min–1 and a detection wavelength of 205 nm. Under these conditions, retention times of Br– and NO3– were 5 and 6.5 min, respectively (analysis duration: 10 min). Tracer concentrations in the infiltrating solutions were also measured and checked for stability in time and served as input concentrations C0.
Immobile Water Content and First-Order Mass Exchange Coefficient
For each tracer, the mean resident solution concentration, C [M L–3], is expressed by
 | [5] |
where [L3 L–3] is the total volumetric water content:
 | [6] |
Under the assumptions that the initial tracer concentration in soil is zero and that the tracer concentration in the mobile domain (Cm) is constant and equal to the injection concentration (C0), which means that dispersion in the mobile domain does not affect Cm in the sampling volume, im and can be determined from
 | [7] |
The regression of ln(1 – C/C0) vs. time should result in a straight line with the intercept ln(im/) and the slope –(/im) (Jaynes et al., 1995). Snow (1999) has defined validity domains for the assumptions of this method. She showed that the quality of im/ and estimates was dependent on soil dispersivity, infiltration rate, and the cumulative infiltration of tracers. Dispersivity values in the three profiles have been estimated from independent field Br–-tracing experiments and ranged between 1.5 and 2.3 cm. We used large cumulative infiltrations for the last tracer (TFBA) when facing high infiltration rates to ensure limited errors on estimates as suggested by Snow's analysis. Although fluorobenzoates have a lower diffusion coefficient than Br– (Bowman, 1984b), it appears not to have affected our results, because there was no systematic trend of Br–-normalized concentration ln(1 – C/C0) data points located below the regression lines. Such tracer effect seems to be negligible in proportion to other experimental errors. Macropore flow could also have affected our MIM parameter measurements. Following Nachabe (1995), the volumetric content of macropores (mac) could be calculated from the water flux densities at –0.3 and –0.1 kPa. We did not have multipotential infiltration measurements in the same three profiles but we used some data obtained on similar soils in the same catchment (Coquet et al., 2005). The maximum mac value (0.0085) was found for a limestone C horizon. Correcting for macroporosity effect would result in the replacement of the left-hand side of Eq. [7] by ln(1 – C/C0 – mac); however, such correction was found to be unnecessary because mac was always negligible in proportion to 1 – C/C0, even when im/ was high.
Statistical Analysis
An unbalanced analysis of variance was done to study the effect of soil horizonation on the hydraulic and transport properties. Bonferroni tests were used to distinguish the soil horizons that had significantly different K–0.1, im/, and values. We calculated the linear correlation coefficients between the measured values of , im/, , and q.
|
RESULTS AND DISCUSSION
|
|---|
Hydraulic Conductivity
Surface Layer
Measurements were made at 3-cm depth in the surface layer (LA1, LA2, and LA3). Steady infiltration rates at –0.1 kPa were achieved within 
5 to 10 min and ranged from 15 to 139 mm h–1 (Table 2). Hydraulic conductivity at –0.1 kPa ranged from 31 to 99 mm h–1, with an arithmetic average and a median for all the surface layer measurements of 65 mm h–1. These hydraulic conductivity values were close to previous results obtained for the seedbed of similar loamy soils (Coutadeur et al., 2002). Coefficients of variation of K–0.1 were 36, 16, and 35% for LA1, LA2, and LA3, respectively. Cameira et al. (2003) found similar CV values after disk plowing, but several studies indicate a CV of near-saturated conductivity >100% in plowed soils due to heterogeneity in macroporosity distribution (Reynolds et al., 1995; Coutadeur et al., 2002).
Observation of the tracer concentration profile in the surface LA experiments (Fig. 1) revealed a permanent jump between the 6- and 8-cm depths (3- and 5-cm relative depths). This discontinuity corresponds to the depth of the superficial tillage that had been done after plowing for the preparation of the seedbed and marks the separation between the seedbed (0–7 cm) and the underlying undisturbed plow layer (7–28 cm). We made complementary measurements at 8-cm depth in Plot 2 (called LA2b in Table 2). A higher mean K–0.1 value (263 mm h–1) than in the seedbed was obtained, but also a higher variability between replicates (CV >90%). The low K–0.1 values and low K–0.1 variability in the seedbed compared with the underlying plowed layer suggests that the rotary harrow tillage used for seedbed preparation has disrupted macropore continuity and homogenized the upper part of the plowed horizon.
View larger version (18K):
[in this window]
[in a new window]
|
Fig. 1. Resident concentration of the four anionic tracers (Br–, pentafluorobenzoate [PFBA], 2,6-diflurobenzoate [DFBA], and trifluoromethylbenzoate [TFBA]) at the end of experiment in surface layers (a) LA1, (b) LA2, and (c) LA3. Depth is relative to the infiltration surface, which was located 3 cm below the actual soil surface.
|
|
Subsurface Horizons
Among the subsurface horizons, the clay-enriched illuvial horizons subjacent to the plowed horizon of Plots 1 (BT1) and 3 (BT3) had similar soil characteristics (Table 1). Values of K–0.1 ranged from 28 to 85 mm h–1 (Table 2) with CVs of 43 and 18% for BT1 and BT3, respectively. These horizons were not affected by tillage and that could explain their low K–0.1 variability. Morphological description of these horizons indicated that clay illuviation formed coatings around peds that had a partially compact and subangular structure. Intraaggregate sorption and interaggregate flow may have been limited because of clay illuviation. The K–0.1 values found in BT1 and BT3 horizons may be the result of both clay accumulation and the absence of disturbance by tillage. Bejat et al. (2000) observed a similar trend for a surface horizon with a decrease in saturated hydraulic conductivity when clay content increased and intensity of tillage practices was reduced. Heuvelman and McInnes (1997) found the highest water flow rate variability in a BT horizon, but their BT horizon presented well-structured prismatic aggregates.
In the heavy clay horizon, B3, the mean K–0.1 value was 258 mm h–1 with a CV of 13%. We observed that this horizon had no sign of clay illuviation (no coatings) and had large prismatic aggregates (>20 cm) between which water could easily run. In the particular case of prismatic aggregates, it has been shown that, depending on initial water content, the hydraulic conductivity could be higher than for other types of structure (Bouma and Anderson, 1977; Lin et al., 1998). An important macroporosity due to biological activity was also clearly visible. Despite a texture that would have indicated a low hydraulic conductivity, this horizon had one of the highest conductivities measured in the three soil profiles (Table 2). Several studies have shown that clayey soils with biopores or cracks, such as the one we observed, could have higher hydraulic conductivity values than coarse-textured soils (Bouma, 1991; Lin et al., 1997, 1998, 1999).
At the base of the soil profiles, the limestone substrates had the highest K–0.1 values that ranged between 114 and 852 mm h–1, with a global arithmetic mean for the three plots of 446 mm h–1. Coefficients of variation were 37, 54, and 39% for C1, C2, and C3, respectively. These substrates correspond to different facies of the Beauce limestone. These limestones have a high capacity to conduct water because they are densely fractured or have a high water-conducting porosity that has been created by chemical and mechanical weathering. Their facies also presented a high spatial variability within our experimental field.
Solute Transport Properties
Surface Layer
Tracer concentration data vs. time were correctly described by Eq. [7] with an average R2 of 0.82 (Fig. 2). Values of im/ ranged from 0.213 in LA1 to 0.882 in LA3 (Table 2), with a global arithmetic mean value of 0.533, a median of 0.482, and a CV of 52% (Fig. 3). Values of ranged from 0.0006 h–1 in LA1 to 0.0115 h–1 in LA2 (Table 2), with a global arithmetic mean value of 0.0031 h–1, a median of 0.0017 h–1, and a CV of 130%. Casey et al. (1998) found im/ to range from 0.28 to 0.95 and to range from 0.015 to 0.3 h–1, Al-Jabri et al. (2002) found im/ to range from 0.36 to 0.88 and to range from 0.002 to 0.12 h–1, and Ilsemann et al. (2002) found im/ to range from 0.11 to 0.27 and to range from 0.015 to 0.034 h–1, all for the same soil type (Nicollet loam) but with different techniques. Ilsemann et al. (2002) also found im/ values between 0.04 and 0.07 and values between 0.001 and 0.008 h–1 for a fine-loamy to medium sand. Kookana et al. (1993) measured values between 0.008 and 0.6 h–1 for various flow rates in sand columns. Our results for im/ were similar to these studies, but values were rather small (except one measure in LA2). The values that we measured in the surface layer correspond to characteristic diffusion times, tc = ln2/, between the mobile and immobile domains of 2.5 to 49 d. Such slow exchange rates between the mobile and the immobile water domains suggests that single-tracer techniques (Clothier et al., 1992, 1995) could be used to simply assess im/ in these soils. Another important consequence is that earlier breakthrough of solutes may be expected when im/ is large and is small, as in the case of the LA3 horizon, than if preferential flow didn't exist and all the soil water contributed to solute transport.
View larger version (37K):
[in this window]
[in a new window]
|
Fig. 2. Normalized resident concentrations (C is the mean resident solution concentration and C0 is the input concentration) of each tracer vs. application time for all experiments. In each graph, points with the same symbol correspond to the same experiment (four tracers). Linear regression lines are shown for each experiment.
|
|
Despite a relatively low variability of the hydraulic conductivity in the seedbed, the amplitude of variation of im/ was high, especially in LA3 (Fig. 3). This variability might be due to a small-scale variability of soil structure, which greatly influences the characteristics of the transport processes (Casey et al., 1997; Okom et al., 2000).
We looked at the potential impact of the rotary harrow tillage (done before sowing) on MIM parameters and made measurements at the 8-cm depth in the LA2 horizon (LA2b samples). Although K–0.1 values were different from the upper part of the horizon, we obtained similar but less variable values of im/ and (Table 2). These differences in MIM parameters are probably related to differences in soil structure. High variability of im/ and values in the seedbed may be due to a larger number of pore discontinuities created by superficial tillage than in the older, partially settled, plowed layer.
In surface horizons (Table 3), we found positive correlations between and the steady-state infiltration rate q at the 0.01 level as observed by Kookana et al. (1993) and Casey et al. (1997).
Subsurface Horizons
As for surface horizons, tracer concentrations could be described adequately by Eq. [7] (Fig. 2) with an average R2 of 0.78. In the clay-enriched illuvial horizons, BT1 and BT3, im/ values ranged from 0.350 to 0.521 and ranged from 0.0007 to 0.0027 h–1, corresponding to characteristic diffusion times between 11 and 41 d. The low variability between replicates for im/ (CV = 43% for BT1 and CV = 18% for BT3) might be due to the homogeneity of the media. These horizons were massive, with compact and subangular aggregates joined by coatings.
For the heavy clay horizon, B3, the arithmetic mean im/ value was 0.304 with a 0.263 to 0.356 range and values ranged from 0.0009 to 0.0016 h–1 (characteristic diffusion times of 18–32 d) with an arithmetic mean value of 0.0012 h–1. Significant positive correlations were found between and im as well as and , and a negative correlation was found between and im/ (Table 3). The heavy clay B3 horizon had im/ values close to those of the BT3 horizon, and values similar to those of the BT1 horizon (Fig. 3). According to the literature, the immobile water fraction increases with clay content (Lee et al., 2000; Okom et al., 2000; Kjaergaard et al., 2004) and aggregate size (Rao et al., 1980; Nkedi-Kizza et al., 1983, 1984). In the heavy clay horizon with a coarse prismatic structure, we thus expected higher values of the immobile water fraction than in the other horizons but instead we found values to be among the lowest (Fig. 2). In this clay horizon, morphological description of the prismatic aggregates showed that the number of biological macro- and mesopores (>100 µm) was important. The high hydraulic conductivity at –0.1 kPa in this horizon suggests that these biopores were well interconnected, as observed in several clay soils by Lin et al. (1996, 1997), thus limiting dead-end pores and stagnant intraaggregate water.
In the limestone materials, im/ values were very high and ranged from 0.506 to 0.933 (CV = 20%) with values between 0.0006 and 0.0424 h–1 (CV = 87%), corresponding to characteristic diffusion times ranging from 0.7 to 48 d (Table 2). These values of were higher and more variable than those of the other horizons (Fig. 3). The limestone horizons had a granular structure with small peds, resulting in large exchange surfaces. Diffusion may be enhanced and could explain high values. The high variability in im/ values prevented us from finding any relation with the composition or structure variability of these horizons. We found significant correlations between and im and between and , as for the heavy clay B3 horizon. We also found in these horizons a significant correlation between and q (Table 3).
The correlation analysis based on all the transport parameter values of all horizons showed a positive correlation at the 0.01 probability level between im and (Fig. 4a), im and , im/ and , im/ and q, and and q (Table 3). Other studies reported positive correlation between and im (Skopp et al., 1981; Casey et al., 1997; Lee et al., 2000, Al-Jabri et al., 2002), which could be explained by the mechanical diffusion model. When the regions of contact between the mobile and the immobile domains expand, as the immobile domain increases, exchanges between the two domains become faster and so increases. We found a significant relation between log() and log(v) where v is the average pore water velocity. Unlike Casey et al. (1997), our data did not fit Kookana et al. (1993)'s regression line (Fig. 4b). This is because the values that we measured in our soils were more than one order of magnitude lower than those reported by Kookana et al. (1993) in the same range of pore water velocities. Such difference is probably related to the type of soils considered in this study. Most of the soils considered by Kookana et al. (1993) were sandy or well-aggregated soils. In our study, the soils have textures dominated by the silt and clay fractions (Table 1) and tend to have coarse structures (BT and B horizons) and vertically oriented tubular pores. We also found a negative correlation at the 0.01 probability level between im/ and and between and q.
View larger version (19K):
[in this window]
[in a new window]
|
Fig. 4. Relationships between (a) the immobile volumetric water content and the total volumetric water content and (b) the logarithm of the mass exchange coefficient (, h–1) and the logarithm of the average pore water velocity (v, cm h–1).
|
|
Statistical Analysis
We analyzed the entire K–0.1, im/, and data using ANOVA to study the effect of soil horizonation on these hydraulic and transport properties. We found a significant effect of soil horizonation on K–0.1 at a probability level of 0.001, but also on im/ at a probability level of 0.001 and at a probability level of 0.02. We used a Bonferroni test to group the different horizons with similar K–0.1, im/, or values. For the K–0.1 values, we found a significant difference between the C1 horizon and the seedbed (LA1, LA2, and LA3) and illuvial (BT1 and BT3) horizons. The C1 horizon had a mean K–0.1 one order of magnitude larger than the seedbed and BT horizons. These results are consistent with those of Coquet et al. (2005), who found the same differentiation in the same type of soils between the tilled layer, having a low near-saturated hydraulic conductivity, and the limestone substrate (C horizon), having a high near-saturated hydraulic conductivity, with B horizons having intermediate K values. Our results support the observation that pedological organization and tillage practices can result in hydraulic conductivity differentiation within agricultural soil profiles.
The C1 and C3 limestone horizons had im/ mean values >0.75—significantly different from the LA1, BT3, and B3 horizons, which had im/ values <0.40. No horizon could be differentiated from the other from the point of view of their parameter values by the Bonferroni test at the 0.05 level. Our results show that different soil horizons can have different MIM parameters. In our study, limestone C horizons had large MIM parameter values compared with the rest of the soil. This highlights the importance of the vadose zone, where significant MIM preferential flow may occur. Preferential flow can seriously modify initial estimates of groundwater contamination based on the application of the classical convection–dispersion equation to vadose zone transport.
|
CONCLUSIONS
|
|---|
We used the Jaynes et al. (1995) method to study hydraulic and transport properties of several materials from the surface down to the 1-m depth in a heterogeneous field of the Beauce region in France. Results show that the mobile–immobile type of preferential transport was indeed a characteristic of Beauce soils. By using only a part of the total porosity, qualified as "hydraulically active," solutes can move within the soil profile at a higher rate than if all the porosity was active.
Tillage had a large influence on the hydraulic conductivity at –0.1 kPa of surface horizons. A high variability in solute transport properties was recorded and seemed to be dependent on the structure created by tillage. In particular, the seedbed, with highly erratic immobile water fractions and low values, was different from the underlying plowed layer that had less variable im/ and values. In subsurface horizons, an immobile water fraction was also detected. The occurrence of an immobile water fraction was not correlated with clay content but seemed to be more dependent on soil structure. In limestone materials, large immobile water fractions were found, and values were larger than for the upper horizons. We did not find any clear general relationship between MIM parameters and soil properties, but it appeared, as suggested by Heuvelman and McInnes (1999), that the interplay of several soil characteristics, such as texture, structure, and pore geometry, greatly influence the occurrence of preferential transport. To better describe groundwater contamination risks, MIM parameters not only of the surface but also of the subsurface soil layers should be used in models.
|
ACKNOWLEDGMENTS
|
|---|
We wish to thank Nathalie Bernet and Véronique Etiévant for their help in laboratory analysis. This work was financially supported by the Ministry of Ecology and Sustainable Development of France.
Received for publication September 5, 2005.
|
REFERENCES
|
|---|
- Al-Jabri, S., R. Horton, D. Jaynes, and A. Gaur. 2002. Field determination of soil hydraulic and chemical transport properties. Soil Sci. 167:353–368.[CrossRef]
- Association Française de Normalisation. 1983. Soil quality. Particle size determination by sedimentation. Pipette method. AFNOR standard X31–107. (In French.) AFNOR, Paris.
- Association Française de Normalisation. 1999a. Soil quality. Determination of carbonate content. Volumetric method. AFNOR standard NF ISO 10693:1995. (In French.) AFNOR, Paris.
- Association Française de Normalisation. 1999b. Soil quality. Determination of organic and total carbon after dry combustion (elementary analysis). AFNOR standard NF ISO 14235:1998. (In French.) AFNOR, Paris.
- Baker, R.S., and D. Hillel. 1990. Laboratory tests of a theory of fingering during infiltration into layered soils. Soil Sci. Soc. Am. J. 54:20–30.[Abstract/Free Full Text]
- Bejat, L., E. Perfect, V.L. Quisenberry, M.S. Coyne, and G.R. Haszler. 2000. Solute transport as related to soil structure in unsaturated intact soil blocks. Soil Sci. Soc. Am. J. 64:818–826.[Abstract/Free Full Text]
- Benson, C., and R. Bowman. 1994. Tri- and tetrafluorobenzoates as nonreactive tracers in soil and groundwater. Soil Sci. Soc. Am. J. 58:1123–1129.[Abstract/Free Full Text]
- Bouma, J. 1991. Influence of soil macroporosity on environmental quality. Adv. Agron. 46:1–37.
- Bouma, J., and J.L. Anderson. 1977. Water movement through pedal soils. I. Saturated flow. Soil Sci. Soc. Am. J. 41:413–418.[Abstract/Free Full Text]
- Bowman, R. 1984a. Analysis of soil extracts for inorganic and organic tracer anions via high-performance liquid chromatography. J. Chromatogr. 285:467–477.[CrossRef]
- Bowman, R. 1984b. Evaluation of some new tracers for soil water studies. Soil Sci. Soc. Am. J. 48:987–993.[Abstract/Free Full Text]
- Brusseau, M.L., R.E. Jessup, and P.S.C. Rao. 1992. Modeling solute transport influenced by multiprocess nonequilibrium and transformation reactions. Water Resour. Res. 28:175–182.[CrossRef]
- Cameira, M.R., R.M. Fernando, and L.S. Pereira. 2003. Soil macropore dynamics affected by tillage and irrigation for a silty loam alluvial soil in southern Portugal. Soil Tillage Res. 70:131–140.[CrossRef]
- Casey, F.X.M., S.D. Logsdon, R. Horton, and D.B. Jaynes. 1997. Immobile water content and mass exchange coefficient of a field soil. Soil Sci. Soc. Am. J. 61:1030–1036.[Abstract/Free Full Text]
- Casey, F.X.M., S.D. Logsdon, R. Horton, and D.B. Jaynes. 1998. Measurement of field soil hydraulic and solute transport parameters. Soil Sci. Soc. Am. J. 62:1172–1178.[Abstract/Free Full Text]
- Clothier, B.E., L.K. Heng, G.N. Magesan, and I. Vogeler. 1995. The measured mobile-water content of an unsaturated soil as a function of hydraulic regime. Aust. J. Soil. Res. 33:397–414.[CrossRef]
- Clothier, B.E., M.B. Kirkham, and J.E. McLean. 1992. In situ measurements of the effective transport volume for solute moving through soil. Soil Sci. Soc. Am. J. 56:733–736.[Abstract/Free Full Text]
- Coats, K., and B. Smith. 1964. Dead end pore volume and dispersion in porous media. SPE J. 4:73–84.
- Coquet, Y., P. Vachier, and C. Labat. 2005. Vertical variation of near-saturated hydraulic conductivity in three soil profiles. Geoderma 126:181–191.
- Coutadeur, C., Y. Coquet, and J. Roger-Estrade. 2002. Variation of hydraulic conductivity in a tilled soil. Eur. J. Soil Sci. 53:619–628.[CrossRef]
- Flury, M., and N.N. Wai. 2003. Dyes as tracers for vadose zone hydrology. Rev. Geophys. 41:1002 doi:10.1029/2001RG000109.[CrossRef]
- Haverkamp, R., P.J. Ross, K.R.J. Smettem, and J.-Y. Parlange. 1994. Three-dimensional analysis of infiltration from the disc infiltrometer. 2. Physically based infiltration equation. Water Resour. Res. 30:2931–2935.[CrossRef]
- Heuvelman, W.J., and K.J. McInnes. 1997. Spatial variability of water fluxes in soil: A field study. Soil Sci. Soc. Am. J. 61:1037–1041.[Abstract/Free Full Text]
- Heuvelman, W.J., and K.J. McInnes. 1999. Solute travel time distributions in soil: A field study. Soil Sci. 164:2–9.[CrossRef]
- Ilsemann, J., R.R. van der Ploeg, R. Horton, and J. Bachmann. 2002. Laboratory method for determining immobile soil water content and mass exchange coefficient. J. Plant Nutr. Soil Sci. 165:332–338.[CrossRef]
- Institut Français de l'Environnement. 2004. Pesticides in water. Annu. Rep. 6th. 2002 data. (In French.) Etude et Travaux no. 42. IFEN, Paris.
- ISSS-ISRIC-FAO. 1998. World reference base for soil resources. FAO, Rome.
- Jarvis, N. 1998. Modeling the impact of preferential flow on nonpoint source pollution. p. 195–221. In H.M. Selim and L. Ma (ed.) Physical nonequilibrium in soils. Modeling and application. Ann Arbor Press, Chelsea, MI.
- Jaynes, D., S. Logsdon, and R. Horton. 1995. Field method for measuring mobile/immobile water content and solute transfer rate coefficient. Soil Sci. Soc. Am. J. 59:352–356.[Abstract/Free Full Text]
- Kjaergaard, C., T.G. Poulsen, P. Moldrup, and L.W. de Jonge. 2004. Colloid mobilization and transport in undisturbed soil columns. I. Pore structure characterization and tritium transport. Vadose Zone J. 3:413–423.[Abstract/Free Full Text]
- Kookana, R.S., R.D. Schuller, and L.A.G. Aylmore. 1993. Simulation of simazine transport through soil columns using time-dependent sorption data measured under flow conditions. J. Contam. Hydrol. 14:93–115.
- Kung, K.-J.S., E.J. Kladivko, T.J. Gish, T.S. Steenhuis, G. Bubenzer, and C.S. Helling. 2000a. Quantifying preferential flow by breakthrough of sequentially applied tracers: Silt loam soil. Soil Sci. Soc. Am. J. 64:1296–1304.[Abstract/Free Full Text]
- Kung, K.-J.S., T.S. Steenhuis, E.J. Kladivko, T.J. Gish, G. Bubenzer, and C.S. Helling. 2000b. Impact of preferential flow on the transport of adsorbing and non-adsorbing tracers. Soil Sci. Soc. Am. J. 64:1290–1296.[Abstract/Free Full Text]
- Lee, J., R. Horton, and D.B. Jaynes. 2000. A time domain reflectometry method to measure immobile water content and mass exchange coefficient. Soil Sci. Soc. Am. J. 64:1911–1917.[Abstract/Free Full Text]
- Lin, H.S., K.J. McInnes, P.J. Wilding, and C.T. Hallmark. 1996. Effective porosity and flow rate with infiltration at low tensions into a well-structured subsoil. Trans. ASAE 39:131–135.
- Lin, H.S., K.J. McInnes, P.J. Wilding, and C.T. Hallmark. 1997. Low tension water flow in structured soils. Can. J. Soil Sci. 25:649–654.
- Lin, H.S., K.J. McInnes, L.P. Wilding, and C.T. Hallmark. 1998. Macroporosity and initial moisture effects on infiltration rates in Vertisols and vertic intergrades. Soil Sci. 163:2–8.[CrossRef]
- Lin, H.S., K.J. McInnes, L.P. Wilding, and C.T. Hallmark. 1999. Effects of soil morphology on hydraulic properties: I. Quantification of soil morphology. Soil Sci. Soc. Am. J. 63:948–954.[Abstract/Free Full Text]
- Mills, M., I. Hill, A. Newcombe, N. Simmons, P. Vaughan, and A. Verity. 2001. Quantification of acetochlor degradation in the unsaturated zone using two novel in situ field techniques: Comparisons with laboratory-generated data and implications for groundwater risk assessments. Pest Manage. Sci. 57:351–359.[CrossRef]
- Nachabe, M.H. 1995. Estimating hydraulic conductivity for models of soils with macropores. J. Irrig. Drain. Eng. 121:95–102.[CrossRef]
- Nkedi-Kizza, P., J.W. Biggar, H.M. Selim, M.Th. van Genuchten, P.J. Wierenga, J.M. Davidson, and D.R. Nielsen. 1984. On the equivalence of two conceptual models for describing ion exchange during transport through an aggregated Oxisol. Water Resour. Res. 20:1123–1130.
- Nkedi-Kizza, P., J.W. Biggar, M.Th. van Genuchten, P.J. Wierenga, H.M. Selim, J.M. Davidson, and D.R. Nielsen. 1983. Modeling tritium and chloride-36 transport through an aggregated Oxisol. Water Resour. Res. 19:691–700.
- Okom, A., R. White, and L. Heng. 2000. Field measured mobile water fraction for soils of contrasting texture. Aust. J. Soil Res. 38:1131–1142.[CrossRef]
- Rao, P.S.C., R.E. Jessup, D.E. Rolston, J.M. Davidson, and D.P. Kilcrease. 1980. Experimental and mathematical description of nonadsorbed solute transfer by diffusion in spherical aggregates. Soil Sci. Soc. Am. J. 44:684–688.[Abstract/Free Full Text]
- Reynolds, W.D., E.G. Gregorich, and W.E. Curnoe. 1995. Characterisation of water transmission properties in tilled and untilled soils using tension infiltrometers. Soil Tillage Res. 33:117–131.[CrossRef]
- Skopp, J., W.R. Gardner, and E.J. Tyler. 1981. Solute movement in structured soils: Two-region model with small interaction. Soil Sci. Soc. Am. J. 45:837–842.[Abstract/Free Full Text]
- Snow, V. 1999. In situ measurement of solute transport coefficients: Assumptions and errors. Soil Sci. Soc. Am. J. 63:255–263.[Abstract/Free Full Text]
- Thomas, G.W., and R.E. Philips. 1979. Consequences of water movement in macropores. J. Environ. Qual. 14:87–91.[Medline]
- Vandervaere, J.-P., M. Vauclin, and D.E. Elrick. 2000a. Transient flow from tension infiltrometers: I. The two-parameter equation. Soil Sci. Soc. Am. J. 64:1263–1272.[Abstract/Free Full Text]
- Vandervaere, J.-P., M. Vauclin, and D.E. Elrick. 2000b. Transient flow from tension infiltrometers: II. Four methods to determine sorptivity and conductivity. Soil Sci. Soc. Am. J. 64:1272–1284.[Abstract/Free Full Text]
- van Genuchten, M.Th., and P. Wierenga. 1976. Mass transfer studies in sorbing porous media: I. Analytical solutions. Soil Sci. Soc. Am. J. 40:473–480.[Abstract/Free Full Text]
- Wood, M., S. Issa, M. Albuquerque, and A. Johnson. 2001. Spatial variability in herbicide degradation in the subsurface environment of a groundwater protection zone. Pest Manage. Sci. 58:3–9.
This article has been cited by other articles:
|
|
|
|
 
J.L. Heitman, A. Gaur, R. Horton, D. B. Jaynes, and T. C. Kaspar
Field Measurement of Soil Surface Chemical Transport Properties for Comparison of Management Zones
Soil Sci. Soc. Am. J.,
March 12, 2007;
71(2):
529 - 536.
[Abstract]
[Full Text]
[PDF]
|
|
|
TOP
ABSTRACT
INTRODUCTION
