Published in Vadose Zone Journal 3:964-970 (2004)
© 2004 Soil Science Society of America
677 S. Segoe Rd., Madison, WI 53711 USA
ORIGINAL RESEARCH
Determination of Hydraulic Properties in Sloping Landscapes from Tension and Double-Ring Infiltrometers
Waduwawatte Bodhinayakea,
Bing Cheng Sia,* and
Kosuke Noboriob
a Dep. of Soil Science, Univ. of Saskatchewan, Saskatoon, SK Canada
b Faculty of Agriculture, Iwate Univ., Morioka, Iwate 0208550, Japan
* Corresponding author (Bing.Si{at}usask.ca)
Received 10 November 2003.
 |
ABSTRACT
|
|---|
The majority of landscapes, natural or cultivated, are nonlevel. However, specifically designed instruments are not available for estimation of soil hydraulic properties in sloping landscapes. The objective of this study is to examine if tension and double-ring infiltrometers are suitable for determination of soil hydraulic properties on sloping soil surfaces. A field experiment was conducted in a silt loam soil (Typic Haplustolls) in Saskatchewan, Canada to explore the usefulness of tension and double-ring infiltrometers for the determination of soil hydraulic properties in sloping landscapes. Soil surfaces were created to represent four treatments, 0 (level), 7, 15, and 20% slopes. For each treatment, water infiltration rates were measured using a double-ring infiltrometer and a tension infiltrometer at –3, –6, –10, –13, –17, and –22 cm water pressure heads. In addition, three-dimensional computer simulation studies were performed for a tension infiltrometer with various disc diameters and water pressure heads for different surface slopes. Steady-state infiltration rate, field-saturated hydraulic conductivity, unsaturated hydraulic conductivity as a function of water pressure head, macroscopic capillary length parameter, and water-conducting macro- and mesoporosity were compared for different surface slopes. These parameters were not significantly different (p < 0.05) between level and sloping lands. Experimental and numerical results of this study suggest that both tension and double-ring infiltrometers are suitable for characterization of surface soil hydraulic properties in landscapes with slopes up to 20%.
Abbreviations: TDR, time domain reflectometry
|
INTRODUCTION
|
|---|
TENSION INFILTROMETERS (Perroux and White, 1988) and double-ring infiltrometers (Bower, 1986) are useful instruments that offer a simple, fast, and convenient means of determining soil hydraulic properties based on in situ infiltration measurements at the soil surface. Tension infiltrometers have proven useful for characterizing soil hydraulic conductivity near saturation (Ankeny et al., 1991; Messing and Jarvis, 1993), sorptivity (Zhang, 1997), mobile–immobile water content (Angulo-Jaramillo et al., 2000), and water-conducting porosity (Watson and Luxmoore, 1986; Dunn and Phillips, 1991; Cameira et al., 2003). Double-ring infiltrometers have also been widely used for estimation of field-saturated hydraulic conductivity under ponded conditions without much disturbance to the soil surface at the measurement site (Bower, 1986).
In many parts of the world, most of the landscapes under crop cultivation and watersheds are nonlevel (slopes > 1%). Conversely, very few measurement techniques are available for determining hydraulic characteristics in situ on hillslopes. Dunne and Black (1970) and Mosley (1982) estimated hillslope hydraulic characteristics by measuring subsurface flow into excavated trenches. Tracer studies with tensiometers, piezometers, and suction lysimeters have been used to identify subsurface flow pathways in a steep watershed by Harr (1977), Anderson et al. (1997) and Torres et al. (1998). These methods, however, are time-consuming, tedious to perform under field conditions, and sometimes require laboratory derived hydraulic parameters for determination of hydraulic conductivity. The hillslope infiltrometer, which is open at the bottom, top, and downhill sides, was introduced by Mendoza and Steenhuis (2002) for determination of vertical and horizontal saturated hydraulic conductivity of soil horizons in steep lands. The installation of this device requires the carving of a soil block that is slightly smaller than the infiltrometer (36 cm in length, 30.5 cm in width and 41 cm in height) and excavation of a trench around the block for ease of installation of the device and water collectors. This particular method is destructive, time-consuming, and somewhat cumbersome for routine field use.
Tension and single-ring or double-ring infiltrometers are primarily designed and tested on horizontal surfaces. However, the equipment has been extensively used in the past to obtain saturated and near-saturated soil hydraulic properties on sloping lands. Watson and Luxmoore (1986) and Wilson and Luxmoore (1988) used tension infiltrometers in conjunction with double-ring infiltrometers for measuring infiltration rates (hydraulic conductivity) and water-conducting macro- and mesoporosity in forest watersheds with slopes up to 20%. Using single-ring infiltrometers, Elliott and Efetha (1999) measured infiltration rates in conventionally tilled and zero-tilled fields with slopes ranging from 6 to 30% in a rolling landscape in Canada. Using tension infiltrometers, Joel and Messing (2000) compared two methods for estimating the hydraulic conductivity near saturation on sloping lands. In the first method (the split-location method), the tension infiltrometer was moved to an adjacent location after infiltration measurement at each applied water pressure head. In this method, the estimate of unsaturated hydraulic conductivity, K(h), was based on measured sorptivity, steady-state infiltration rate, initial moisture content, and the volumetric water content at the applied water pressure head. In the second (one-location method), the tension infiltrometer was not moved during the measurements of infiltration at each sequence of applied water pressure head, and the estimate of K(h) was based only on steady-state infiltration rate. Based on the simplicity of operation and the amount of soil disturbance, Joel and Messing (2000) recommended the one-location method over the split-location method. Casanova et al. (2000) studied the influence of aspect and slope on hydraulic conductivity measured by tension infiltrometer. They found smaller K(h) values for the north-facing locations than for the south-facing locations. The differences were attributed to differences in texture and organic matter contents observed for the two soils. Both Joel and Messing (2000) and Casanova et al. (2000), however, did not evaluate the appropriateness of tension infiltrometers for the determination of hydraulic properties in lands with different slopes and recommended further studies on the influence of slope on tension infiltrometer measurements. Despite the extensive use of tension and double-ring infiltrometers in determining surface soil hydraulic properties in sloping lands, to the best of our knowledge no systematic studies have been conducted on the suitability of this equipment for estimation of these properties in sloping lands. Therefore, the objective of this study was to evaluate the suitability of tension and double-ring infiltrometers for the estimation of surface soil hydraulic properties at saturated and near-saturated conditions.
|
MATERIALS AND METHODS
|
|---|
Field experiments using a tension infiltrometer and a double-ring infiltrometer were performed at Laura (approximately 50 km west of Saskatoon), Saskatchewan, Canada (51°52'N lat., 107°18'W long.). The soil in that area is described as an Elstow association: Dark Brown Chernozems (Typic Ustolls) developed on loamy glaciolacustrine parent material with a silt loam texture in both Ap (0–14 cm) and Bm (14–25 cm) soil horizons. The lacustrine sediments are underlain by glacial till that is drained by the Tessier aquifer. The water table occurs at approximately 15 m below the surface within a sand layer (Dyck et al., 2003). Since 1996, the site has been under a crop–fallow rotation dominated by wheat (Triticum aestivum L.), with some barley (Hordeum vulgare L.). During the study period (August 2002), the site had been summer fallowed. The long-term (23-yr) average precipitation at the site is 321 mm yr–1.
Four slopes—0 (level), 7, 15, and 20%—were chosen as treatments. Generally, the maximum slope of the land used for agricultural purposes in Saskatchewan, Canada is 15%. The treatments were selected to represent level as well as one below and one above the maximum slope of the agricultural lands. A field with 0 to 5% slope and about 10 m wide and 100 m long was selected as the experimental site. An area of 9 m2 (3 by 3 m) was demarcated in the field as an experimental block. Each experimental block was divided into eight experimental units (approximately 1 m2 per experimental unit). Each slope had two experimental units, one for tension infiltrometer measurements and the other for double-ring infiltrometer measurements. Surface slopes were then artificially created on the experimental units to represent 0, 7, 15, and 20% slopes. In each experimental unit, an undisturbed soil core sample (0.075 m in diameter and 0.05 m in height) was taken from an area near the measurement location. These cores were used for the determination of bulk density by oven drying at 105°C. Total porosity was estimated from soil bulk density and particle density (Flint and Flint, 2002). To measure soil moisture content, a time domain reflectometry (TDR) probe was installed before the tension infiltration measurements. This TDR probe (consisting of two parallel, 15-cm-long, 0.2-cm-diam. stainless-steel rods with a separation of 1 cm) was inserted horizontally, at 2- to 2.5-cm depth, in line with the center of the infiltrometer disk and in a direction perpendicular to the slope. A TDR reading was recorded just before the commencement of infiltration measurement to get antecedent soil moisture content.
Infiltration measurements were performed using a tension infiltrometer with a 20-cm-diameter disk that was attached to the reservoir tower and tension control tube via a flexible tube (Soil Measurement Systems, Tuscon, AZ). A thin layer (<0.5 cm) of moist fine testing sand was applied over the prepared surface at each measurement location in a circular area with a diameter equal to that of the infiltrometer disk. This smoothed out any irregularities of the soil surface and ensured good contact between the soil surface and the infiltrometer membrane. The testing sand is reported to have a saturated hydraulic conductivity of 5.3 x 10–5 m s–1 and an air-entry value slightly higher than –3.0 cm water pressure head. The testing sand was of sufficient porosity not to be a hydrologically limiting layer, yet had fine enough pores to remain saturated at the pressure heads used in this study. The nylon mesh attached to the infiltrometer disc had an air-entry value of about –3.0 cm water pressure head. Although the testing sand layer may have an impact on early-time hydraulic conductivity (Vandervaere et al., 2000), the thin layer of sand we used would not affect the steady-state infiltration rates, and the concomitant estimated hydraulic properties of the surface soil (Vandervaere et al., 2000). The infiltration rates were measured at –3, –6, –10, –13, –17, and –22 cm water pressure heads (corresponding to 1 x 10–3, 5 x 10–4, 3 x 10–4, 2.31 x 10–4, 1.76 x 10–4, and 1.36 x 10–4 m equivalent pore diameters). Measurements were performed from low to high pressure heads, that is, beginning with the –22 cm water pressure head. For the beginning of the infiltration measurements, the pressure head of the infiltrometer was adjusted to –22 cm water. The unit, consisting of the reservoir tower and the tension control tube, was placed on a wooden bench kept beside the measurement location. Simultaneously the infiltrometer disc, inclined at an angle (to the horizontal) corresponding to that of the soil surface, was placed on the contact sand, so the pressure head in the middle of the disc was similar to the pressure head of the bubbling outlet at the bottom of the water supply tube.
The amount of water infiltrating into the soil was measured by recording the water level drop in the graduated reservoir tower as a function of time. When the amount of water entering the soil did not change with time for three consecutive measurements taken at 5-min intervals, steady-state flow was assumed, and the steady-state infiltration rate was calculated based on the last three measurements. The pressure heads were then set sequentially to –17, –13, –10, –6, and –3 cm water, and the corresponding steady-state infiltration rates were obtained. Generally, steady state was achieved within 20 to 30 min. Once the steady-state infiltration rate was attained at –3.0 cm water pressure head, a TDR reading was taken to determine soil moisture content. Macroporosity (fraction of total soil volume occupied by pores >1 x 10–3 m; Luxmoore, 1981) was determined as the difference between total porosity and volumetric moisture content held at the –3.0 cm water pressure head.
A double-ring infiltrometer with inner and outer rings of 20 and 30 cm in diameter, respectively, was used to determine steady infiltration rate at a constant head (Reynolds et al., 2002) on the selected experimental units. Steel rings were pushed into the soil concentrically and parallel to the measurement surface to a depth of 5 cm with minimum soil disturbance. After the insertion of the rings the contact between the inside surface of the ring and the soil was tamped lightly using the blunt edge of a pencil to minimize short-circuit flow along the inside wall of the ring. A steel pointer was positioned vertically at the center of the inner cylinder with a height of 3.0 cm above the soil surface. The inner cylinder was then filled with water equivalent to a water head of 4.0 cm initially. The time taken to drop the water level by 1.0 cm in the inner cylinder (to the pointer) was recorded. Thereafter, a measured volume of water, equivalent to 1.0 cm in depth, was successively added to the inner ring until the infiltration time did not change for three consecutive measurements taken at 5-min intervals. At this point steady-state flow was assumed, and the steady-state infiltration rate was calculated based on the last three measurements. Generally, steady-state flow was achieved within 30 to 60 min. Water level in the outer ring was maintained at exactly the same height as that in the inner ring. There were five replications in this experiment.
Estimation of Soil Hydraulic Properties from Tension Infiltrometer Measurements
The three-dimensional steady infiltration rates obtained at different pressure heads were used to obtain unsaturated hydraulic properties. For Gardner's (1958) exponential hydraulic conductivity function,
 | [1] |
where K(h) is the unsaturated hydraulic conductivity (L T–1) for a given pressure head, h (L) and 
is the inverse soil macroscopic capillary length parameter (L–1), Wooding (1968) derived the following approximate solution for steady-state infiltration rate under a shallow circular disc:
 | [2] |
where q
is steady-state infiltration rate (L T–1) corresponding to the applied water pressure head h, rd is radius of the disc (L), and Kfs is the field-saturated hydraulic conductivity (L T–1). Equation [2] has two unknown parameters, Kfs and . Following Logsdon and Jaynes (1993), these parameters were estimated through nonlinear regression of q as a function of h using MathCad 2000 (MathSoft, Cambridge, MA). Values for K(h) were then estimated by substituting the resulting values of Kfs and into Eq. [1].
Estimation of Field-Saturated Hydraulic Conductivity from Double-Ring Infiltrometer Measurements
Field-saturated hydraulic conductivity, Kfs (L T–1), was estimated for each experimental unit from the one-dimensional steady-state infiltration rates obtained from the double-ring infiltrometer following the procedure outlined by Reynolds et al. (2002). The Kfs is given by
 | [3] |
where qs (L T–1) is the quasi-steady infiltration rate, H (L) is the steady depth of ponded water in the ring, c1 = 0.316
and c2 = 0.184 are dimensionless quasiempirical constants, d (L) is depth of ring insertion into the soil, ar (L) is radius of the inner ring, and (L–1) is the inverse soil macroscopic capillary length parameter (Gardner's ). The respective a values obtained from Eq. [2] for different slopes were used for the calculation of Kfs. The water pressure head at the center of the inner ring varied from 4.0 to 3.0 cm during infiltration measurements. Therefore, the midway elevation between 3.0 and 4.0 cm (3.5 cm) was taken as the steady depth of ponded water in the inner ring (H).
Determination of Water-Conducting Porosity
The equivalent radius r (L) of the largest water-filled pore in the soil at a given pressure head can be calculated from the capillary rise equation as given by (Bear, 1972):
 | [4] |
where 
is the surface tension of water (M T–2), ß is the contact angle between water and the pore wall (assumed to be zero), 
is the density of water (M L–3), g is the acceleration due to gravity (L T–2), and h is the applied water pressure head (L).
This study assumes that the equivalent pores smaller than the r value estimated by Eq. [4] are full of water and are responsible for 100% of the water flux for a given water pressure head. Also, it is assumed that the equivalent pores larger than the value of r calculated from Eq. [4] are air-filled and do not contribute to any of the water flux. On the basis of Poiseuille's Law the flow rate through a single vertical macropore due to gravity can be given as
 | [5] |
where Q(r) is the flow rate (L3 T–1) as a function of pore radius r and µ is the dynamic viscosity of water (M L–1 T–1).
Bodhinayake et al. (2004) derived a new equation for determining macroporosity from tension infiltrometer measurements. Bodhinayake and Si (2004) applied this equation to compare the effect of land use on macroporosity. A detailed description of the derivation of the equation is given in Bodhinayake et al. (2004). Briefly, they considered the number of pores per unit area (L2) as a function of pore radius r. The cumulative pore number distribution is then given by
 | [6] |
where n(r) is the total number of pores in a given pore size range and P(r) is the number of pores per unit soil surface area per unit pore radius. The hydraulic conductivity, K (L T–1) at a given pore size r, K(r) can be expressed as
 | [7] |
where r is the upper limit of the integrals determined by the pressure head.
The water-conducting porosity in a given pore size range can be expressed as
 | [8] |
where a and b are pore radii (L). An expression for P(r) can be obtained by taking the derivatives of both sides of Eq. [7]. Substitution of P(r) into Eq. [8] then yields
 | [9] |
Normally, unsaturated hydraulic conductivity of soil is expressed in relation to soil water content or pressure head. In Eq. [9], the hydraulic conductivity is expressed as a function of pore radius. To express the hydraulic conductivity as a function of water pressure head, that is, K = K(h), we rewrite Eq. [4] (assume ß = 0) as
 | [10] |
where h(r) is the water pressure head corresponding to radius r. Substitution of Eq. [5] for Q(r) and Eq. [10] for r into Eq. [9] leads to
 | [11] |
Integration of Eq. [11] by parts and substituting Eq. [2] for K(h) leads to
 | [12] |
Substitution of Eq. [10] into Eq. [12] and subsequent integration gives (Gradshteyn and Ryzhik, 2000; Eq. 2.322, p. 104)
 | [13] |
From Eq. [10], it can be seen that the value of 0 pressure head cannot be related to a pore size. Therefore, the upper limit of the integral of Eq. [12] cannot be defined. However, a small pressure head can be related to a pore size. We assumed that the maximum pore diameter at the sites is 0.50 cm. This is a reasonable assumption because no pores >0.50 cm in diameter existed at any of the infiltration measurement locations. Equation [13] (with Kfs and from Eq. [2]) was then used for the estimation of water-conducting porosity with pore diameter ranges from 1 x 10–3 to 5 x 10–3, 5 x 10–4 to 1 x 10–3, 3 x 10–4 to 5 x 10–4, 2.31 x 10–4 to 3 x 10–4, 1.76 x 10–4 to 2.31 x 10–4, 1.36 x 10–4 to 1.76 x 10–4 and 1.36 x 10–4 to 5 x 10–3 m.
Computer Simulation
Water infiltration from a tension infiltrometer placed at a sloping landscape was simulated with various disk diameters, water pressures applied at the soil surface, and sloping degrees. The numerical solution of Richards' equation and moisture retention relationship given by Russo et al. (1991) were employed for the simulation. The Richards' equation for three-dimensional water flow in a homogenous and isotropic soil at a sloping landscape may be expressed as
 | [14] |
where C(h) is the soil water capacity (L–1), h is the water pressure head (L), K(h) is the unsaturated hydraulic conductivity (L T–1) as defined in Eq. [2], 
is the angle of the slope with respect to the horizontal y axis (rad), t is the time (T), and x, y, z are the axes of the Cartesian coordinate system (L) with z positive upwards. The soil water characteristic curve is described (Russo et al., 1991) as
 | [15] |
where 
, s, and r are volumetric water content (L3 L–3), saturated volumetric water content (L3 L–3), and residual volumetric water content (L3 L–3), respectively. is the inverse soil macroscopic capillary length parameter (L–1), and m is an empirical parameter (0.5). The soil water capacity, C(h), is defined as the slope of the soil water characteristic curve as d/dh.
To numerically solve Eq. [14] a general partial differential equation solver, FlexPDE (PDE Solutions Inc., 2000), which uses the finite element method as a solver, was deployed. Its performance to solve problems on one- and two-dimensional water or solute transport in soil was found elsewhere (Noborio, 2001). The size of triangular elements and a time step increment were automatically changed to meet prescribed criteria. Equation [14] was solved with conditions h = –100 cm for the initial soil water pressure head and h = –3, –7, –10, or –22 cm at the center of the disk at the soil surface for a boundary condition, the natural boundary for other boundaries. The disk diameters of the tension infiltrometer simulated were 4, 20, 60, and 100 cm. Surface slopes of 0, 7, 15, and 20%, with and Kfs values obtained from tension infiltrometer measurements in the field, were also used for the simulation. The calculating domain varied with the disk diameter of the tension infiltrometer to have enough domain space for water flow.
Statistical Analysis
Natural logtransformed Kfs, K(h), and values were used for statistical analysis because their distributions are reported to be lognormal (Nielsen et al., 1973; Russo and Bouton, 1992). Analysis of variance was performed using SAS (SAS Institute, 1990) for a randomized complete block design with five replications. Treatment means for 7, 15, and 20% slopes were compared separately with the mean value for the control (level land = 0% slope) by Dunnett's test. Two treatment means were considered as significantly different whenever the absolute difference between the corresponding estimated means exceeded the calculated Dunnett's critical t value at a 0.05 significance level.
|
RESULTS AND DISCUSSION
|
|---|
The mean bulk density, macroporosity, and soil texture of the surface 0 to 5 cm of soil did not show a significant difference among slopes, suggesting that the experimental blocks were nearly uniform in soil texture and soil structure (Table 1). The steady infiltration rate is a function of the boundary conditions (level or sloping surfaces) and the morphology of the pore system that is controlled by the texture and the structure of the soil, its continuity to the soil surface, and potential energy forces applied to the water (Hillel, 1998). Uniform pore morphology allows us to examine the effect of slope gradient (boundary conditions) on measurement of soil hydraulic properties.
No significant differences were found among slope gradients for the steady infiltration rates obtained from double-ring infiltrometers. Although the infiltration measurements were performed with a constant head at the center of the inner ring, the pressure head varied across the sloping surface with the highest value at the downslope side and lowest value at the upslope side. Therefore, the infiltration rate was lower at the upslope side than that on the downslope side. The lower infiltration rate at the upslope side was most likely offset by the higher infiltration rate at the downslope side, resulting in no substantial differences among slopes. Furthermore, Philip (1991) reported that downslope flow occurs as a result of the downslope component of gravity in sloping lands. The magnitude of slope effects on downslope flow increases with an increase in slope. However, for homogeneous and isotropic soils under constant flux boundary condition, Philip showed theoretically that the infiltration normal to the slope differed relatively little from infiltration on a horizontal surface for slope gradients <58%. Harr (1977) also studied the magnitude and direction of downslope flow under different rainfall intensities in steep forested watersheds (average slope 75%) using tensiometers. He observed that even a slight change in soil water pressure head due to slope could change both magnitude and direction of water flux. Harr related the significant downslope flow during and between storms to soil layering with significant differences in pore-size distribution and saturated hydraulic conductivity between surface (0–30 cm) and subsurface soils. The surface soil of our experimental site was more or less homogeneous due to mixing of the soil during cultivation. Soil layers with abrupt changes in soil texture did not exist within the top 1.2 m of the soil profile (Dyck et al., 2003). As a consequence, the downslope (lateral) flow under a constant head boundary condition would also be small, showing no significant difference in infiltration rates between horizontal surface and lands with relatively small gradients (20%).
Statistical tests showed that there are no significant differences among slope treatments for the infiltration rates measured at different pressure heads using tension infiltrometer with a 20-cm-diameter disc (Fig. 1)
. These experimental results were confirmed by the simulation studies. The 20-min cumulative infiltration under a 20-cm-diameter disc, at –3 cm water pressure head, was not apparently different between level and sloping land surfaces (Fig. 2a)
. The pressure head contours under the disc did not show marked differences among sloping surfaces (data not shown). Simulated cumulative infiltration for a 20-min period with 4-, 20-, 60-, and 100-cm diameter discs at –7, –10, and –22 cm water pressure heads also did not show any visible differences among slope gradients (data not shown). Only the cumulative infiltration simulated with a 100-cm-diameter disc at –3 cm water pressure head showed small differences among slope gradients, and the discrepancies increased with time elapsed (Fig. 2b). In addition to the reasons we stated for the double-ring infiltrometer, gravity is less important under unsaturated conditions than at saturation. As with infiltration rate at different pressure heads for the level soil surface, decreases in pressure head resulted in decreases in mean steady infiltration rates (Fig. 1) and the rates of decrease were equivalent for all slope gradients. A decrease in pressure head results in a decrease in the steady infiltration rate because a decrease in pressure head reduces the size and number of pores that participate in conducting water.
View larger version (18K):
[in this window]
[in a new window]
|
Fig. 2. Simulated cumulative infiltration as a function of elapsed time for different slopes at –3 cm water pressure head; a) 20 cm diameter disk, and b) 100 cm diameter disc.
|
|
Because the differences in infiltration rates measured using double-ring and tension infiltrometers are not statistically significant, in the following we use the method for the level surface to calculate hydraulic properties in sloping soil surfaces. The steady-state infiltration rates obtained from tension infiltrometer fitted very closely (R2 > 0.9) to Eq. [3] for all the slopes tested. The fitting parameters, the saturated hydraulic conductivity (Kfs) and inverse capillary length scale (), were calculated. The Kfs estimated from tension infiltrometer measurements as well as the Kfs obtained from double-ring infiltrometer measurements were not significantly different between horizontal and sloping soil surfaces. The Kfs values were estimated from steady infiltration rates, which are mainly influenced by soil texture and structure. Uniform soil texture and bulk density in the experimental site contribute to nearly similar steady infiltration rates among slopes at different pressure heads. As a consequence, Kfs is more or less uniform among slope gradients evaluated.
The unconfined measurements of Kfs estimated from tension infiltrometers were slightly greater than those from confined infiltration in double-ring infiltrometers (Table 1) for all slope gradients. This may be partly due to the effect of air entrapment, which acts to reduce infiltration rate when water enters a soil containing pockets of air. This is of a less problem with the unconfined measurements because of the radial shape and unconfined nature of the wetting front. The Kfs values are comparable with the values given by Rawls et al. (1993) for the USDA soil textural triangle for silt loam soils (1.88 x 10–6 m s–1).
The values did not significantly vary among slope gradients (Table 1). The K(h) functions estimated from infiltration measurements did not change significantly between level and sloping surfaces (Fig. 3)
. Again, more or less uniform pore-size distribution among slope treatments likely explain such similar K(h) relationships and thereby similar values. Our values are comparable with the values reported by Reynolds et al. (2002) for cultivated soils as cited by Reynolds et al. (2002).
When considering macropores, the total volume occupied by them (macroporosity) and the amount of water they are capable of conducting are important. Macroporosity and water-conducting macro- and total porosity between –0.6 and –22 cm water pressure head range were not significantly different between level and sloping lands, indicating that there will not be substantial differences in water transmission properties. Macropores consisted of 13 to 16% of the total soil volume (Table 1). However, the real water-conducting macroporosity of the four slopes varied from 0.002 to 0.003%. Macroporosity values in our study compared reasonably well with the values reported by Dunn and Phillips (1991) and Cameira et al. (2003) in silt loam soil for minimum tillage and conventional tillage plots.
The experimental and numerical results of this study clearly showed that, for the surface slopes ranging from zero to 20% in a cultivated field, there were no significant differences in hydraulic properties and water-conducting porosity estimated (Table 2) from both tension and double-ring infiltrometer measurements at the soil surface. As such, both the tension infiltrometer and double ring infiltrometer are suitable for the measurement of soil hydraulic properties in level lands as well as nonlevel lands up to slopes of 20%.
View this table:
[in this window]
[in a new window]
|
Table 2. Estimated water-conducting porosity (% of total soil volume) in each pore diameter interval for the different slopes.
|
|
For steep soil surfaces, however, water supply pressure head varies across the sloping surface (from the upslope to the downslope side). Since the infiltrometer operates in the macropore range, where water transmission properties can change drastically with even small changes in pressure head (Reynolds and Zebchuck, 1996), this may have a substantial impact on the validity of the tension infiltrometer results. Our results did not indicate such effects. Furthermore, several researchers (Zaslavsky and Sinai, 1981; McCord et al., 1991) have shown that downslope flow can occur even under unsaturated conditions in the presence of soil layering close to the surface and/or anisotropy favoring downslope flow. However, the existence of a small but constant anisotropy causes little lateral downslope flow relative to vertical flow (Jackson, 1992). Lack of soil layering and probably minor anisotropy, with more or less homogeneous conditions prevailing in soil close to the surface, would have created little downslope flow. As a consequence, the infiltration rate, and the estimated hydraulic properties from infiltration rates would not vary much between level and sloping lands in our experimental site.
The results of this experiment may be different for heterogeneous soils that have soil layers close to the surface with contrasting hydraulic properties and/or a broad range of pore sizes. The presence of soil layers may increase the downslope flow component. Hence, further investigation is recommended on the slope effect on tension infiltration in soils with layers and broad pore-size ranges.
|
CONCLUSIONS
|
|---|
The suitability of tension and double-ring infiltrometers for the estimation of hydraulic properties in sloping lands in situ was evaluated using infiltration measurements at 0, 7, 15, and 20% slopes in a cereal crop cultivated field. Water infiltration from the tension infiltrometer was also simulated with various disc diameters and water pressure heads for different surface slopes. The measured steady-state infiltration rates, estimated field-saturated hydraulic conductivity, unsaturated hydraulic conductivity as a function of water pressure head, macroscopic capillary length parameter, and water-conducting macro- and mesoporosity were not significantly different among the slopes evaluated. Experimental and numerical results of this study strongly suggest that both the tension infiltrometer (0.2-m-diameter disc) and double-ring infiltrometer are suitable for characterization of soil hydraulic properties in lands with slopes up to 20%.
|
ACKNOWLEDGMENTS
|
|---|
Funding for this project was provided by the National Science and Engineering Research Council of Canada (NSERC) and the University of Saskatchewan through a graduate student scholarship.
|
REFERENCES
|
|---|
- Anderson, S.P., W.E. Dietrich, D.R. Montgomery, R. Torres, M.E. Conrad, and K. Loague. 1997. Subsurface flow paths in a steep, unchanneled catchment. Water Resour. Res. 33:2637–2653.
- Angulo-Jaramillo, R., J.P. Vandervaere, S. Roulier, J.L. Thony, J.P. Gaudet, and M. Vauclin. 2000. Field measurement of soil surface hydraulic properties by disc and ring infiltrometers: A review and recent developments. Soil Tillage Res. 55:1–29.
- Ankeny, M.D., M. Ahmed, T.C. Kaspar, and R. Horton. 1991. Simple field method for determining unsaturated hydraulic conductivity. Soil Sci. Soc. Am. J. 55:467–470.[Abstract/Free Full Text]
- Bear, J. 1972. Dynamics of fluids in porous media. Elsevier Pub. Co. Inc, New York, NY.
- Bodhinayake, W.L., B.C. Si, and C. Xiao. 2004. New method for determining water-conducting macro- and mesoporosity from tension infiltrometer. Soil Sci. Soc. Am. J. 68:760–769.[Abstract/Free Full Text]
- Bodhinayake, W.L., and B.C. Si. 2004. Near-saturated surface soil hydraulic properties under different land uses in the St. Denis National Wildlife Area, Saskatchewan. Canada. J. Hydrol. Process. (in press).
- Bower, H. 1986. Intake rate. Cylinder infiltrometer. p. 825–843. In A. Klute (ed.) Methods of soil analysis. Part 1. 2nd ed. SSSA Book Ser. 5. ASA and SSSA, Madison, WI.
- Cameira, M.R., R.M. Fernando, and L.S. Pereira. 2003. Soil macropore dynamics affected by tillage and irrigation for a silt loam alluvial soil in southern Portugal. Soil Tillage Res. 70:131–140.
- Casanova, M., I. Messing, and A. Joel. 2000. Influence of aspect and slope gradient on hydraulic conductivity measured by tension infiltrometer. Hydrol. Process. 14:155–164.
- Dunn, G.H., and R.E. Phillips. 1991. Macroporosity of a well-drained soil under no-till and conventional tillage. Soil Sci. Soc. Am. J. 55:817–823.[Abstract/Free Full Text]
- Dunne, T., and R.D. Black. 1970. An experimental investigation of runoff production in permeable soils. Water Resour. Res. 6:478–490.
- Dyck, M.F., R.G. Kachanoski, and E. de Jong. 2003. Long-term movement of a chloride tracer under transient, semi-arid conditions. Soil Sci. Soc. Am. J. 67:471–477.[Abstract/Free Full Text]
- Elliott, J.A., and A.A. Efetha. 1999. Influence of tillage and cropping system on soil organic matter, structure and infiltration in a rolling landscape. Can. J. Soil Sci. 79:457–463.
- Flint, L.E., and A.L. Flint. 2002. The soil solution phase. Porosity. p. 241–254. In J.H. Dane and G.C. Topp (ed.) Methods of soil analysis. Part 4. SSSA Book Ser. 5. SSSA, Madison, WI.
- Gardner, W.R. 1958. Some steady-state solutions of unsaturated moisture flow equations with application to evaporation from a water table. Soil Sci. 85:228–232.
- Gradshteyn, I.S., and I.M. Ryzhik. 2000. Table of integrals, series, and products. 6th ed. Academic Press, San Diego, CA.
- Harr, R.D. 1977. Water flux in soil and subsoil on a steep forested slope. J. Hydrol. (Amsterdam) 33:37–58.
- Hillel, D. 1998. Environmental soil physics. Academic Press, San Diego, CA.
- Jackson, C.R. 1992. Hillslope infiltration and lateral downslope unsaturated flow. Water Resour. Res. 28:2533–2539.
- Joel, A., and I. Messing. 2000. Application of two methods to determine hydraulic conductivity with disc permeameters on sloping land. Eur. J. Soil Sci. 51:93–98.
- Logsdon, S.D., and D.B. Jaynes. 1993. Methodology for determining hydraulic conductivity with tension infiltrometers. Soil Sci. Soc. Am. J. 57:1426–1431.[Abstract/Free Full Text]
- Luxmoore, R.J. 1981. Micro-, meso- and macroporosity of soil. Soil Sci. Soc. Am. J. 45:671–672.[Free Full Text]
- McCord, J.T., D.B. Stephens, and J.L. Wilson. 1991. Hysteresis and state-dependent anisotropy in modelling unsaturated hillslope hydrologic processes. Water Resour. Res. 27:1501–1518.
- Mendoza, G., and S.T. Steenhuis. 2002. Determination of hydraulic behaviour of hillsides with a hill slope infiltrometer. Soil Sci. Soc. Am. J. 66:1501–1504.[Abstract/Free Full Text]
- Messing, I., and N.J. Jarvis. 1993. Temporal variation in the hydraulic conductivity of a tilled clay soil as measured by tension infiltrometers. J. Soil Sci. 44:11–24.
- Mosley, M.P. 1982. Subsurface flow velocities through selected forest soils, South Island, New Zealand. J. Hydrol. (Amsterdam) 55:65–92.
- Nielsen, D.R., J.W. Biggar, and K.T. Ere. 1973. Spatial variability of field measured soil-water properties. Hilgardia. 42:215–260.[ISI]
- Noborio, K. 2001. Application of a general PDE solver for analyzing water and solute transport in soil. J. Jpn. Soc. Soil Phys. 88:19–25 (in Japanese with English abstract.
- PDE Solutions Inc. 2000. FlexPDE user guide. Version 2.19. PDE Solutions Inc., Antioch, CA.
- Perroux, K.M., and I. White. 1988. Design for disc permeameters. Soil Sci. Soc. Am. J. 52:1205–1215.[Abstract/Free Full Text]
- Philip, J.R. 1991. Hill slope infiltration: Planar slopes. Water Resour. Res. 27:109–117.
- Rawls, W.J., L.R. Abuja, D.L. Brakensiek, and A. Shirmohammadi. 1993. Infiltration and soil water movement. p. 5.1–5.51. In D.R. Maidment (ed.) Handbook of hydrology. McGraw-Hill, New York.
- Reynolds, W.D., D.E. Elrick, and E.G. Youngs. 2002. The soil solution phase. Single-ring and double- or concentric-ring infiltrometers. p. 821–826. In J.H. Dane and G.C. Topp (ed.) Methods of soil analysis. SSSA Book Ser. 5. SSSA, Madison, WI.
- Reynolds, W.D., and W.D. Zebchuck. 1996. Use of contact material in tension infiltrometer measurements. Soil Technol. 9:141–159.
- Russo, D., and M. Bouton. 1992. Statistical analysis of spatial variability in unsaturated flow parameters. Water Resour. Res. 28:1911–1925.
- Russo, D., E. Bresler, U. Shani, and J.C. Parker. 1991. Analyses of infiltration events in relation to determining soil hydraulic properties by inverse problem methodology. Water Resour. Res. 27:1361–1373.
- SAS Institute. 1990. SAS/STAT user's guide. Version 6. SAS Inst. Inc., Cary, NC.
- Torres, R., W.E. Dietrich, D.R. Montgomery, S.P. Anderson, and K. Loague. 1998. Unsaturated zone processes and the hydrologic response of a steep, unchanneled catchment. Water Resour. Res. 34:1865–1879.
- Vandervaere, J.P., M. Vauclin, and D.E. Elrick. 2000. Transient flow from tension infiltrometers. I. The two-parameter equation. Soil Sci. Soc. Am. J. 64:1263–1272.[Abstract/Free Full Text]
- Watson, K.W., and R.J. Luxmoore. 1986. Estimating macroporosity in a forest watershed by use of a tension infiltrometer. Soil Sci. Soc. Am. J. 50:578–582.[Abstract/Free Full Text]
- Wilson, G.V., and R.J. Luxmoore. 1988. Infiltration, macroporosity and mesoporosity distributions on two forested watersheds. Soil Sci. Soc. Am. J. 52:329–335.[Abstract/Free Full Text]
- Wooding, R.A. 1968. Steady infiltration from a shallow circular pond. Water Resour. Res. 4:1259–1273.
- Zaslavsky, D., and G. Sinai. 1981. Surface hydrology. 1. Explanation of phenomena. J. Hydrol. Div. ASCE 107:1–16.
- Zhang, R. 1997. Determination of soil sorptivity and hydraulic conductivity from disk infiltrometer. Soil Sci. Soc. Am. J. 61:1024–1030.[Abstract/Free Full Text]
This article has been cited by other articles:
|
|
|
|
 
N. Lazarovitch, A. Ben-Gal, J. Simunek, and U. Shani
Uniqueness of Soil Hydraulic Parameters Determined by a Combined Wooding Inverse Approach
Soil Sci. Soc. Am. J.,
May 16, 2007;
71(3):
860 - 865.
[Abstract]
[Full Text]
[PDF]
|
|
|
TOP
ABSTRACT
INTRODUCTION
