Published in Vadose Zone Journal 4:139-144 (2005)
© 2005 Soil Science Society of America
677 S. Segoe Rd., Madison, WI 53711 USA
ORIGINAL RESEARCH
The Effect of Entrapped Air on the Quasi-Saturated Soil Hydraulic Conductivity and Comparison with the Unsaturated Hydraulic Conductivity
A. Sakaguchi,
T. Nishimura* and
M. Kato
Department of International Environmental and Agricultural Sciences, Graduate School of Agriculture, Tokyo University of Agriculture and Technology, Saiwaicho, Fuchu 183-8509, Tokyo, Japan
* Corresponding author (takun{at}cc.tuat.ac.jp)
Received 1 March 2004.
 |
ABSTRACT
|
|---|
Entrapped air can greatly affect the hydraulic conductivity at or near saturation. In this study, we measured the hydraulic conductivity and volume of entrapped air in a quasi-saturated soil. Two soils, a Masa sandy loam soil from weathered granite rock and a TUAT light clay andisol from volcanic ash, were used. The soils, with three different dry bulk densities, were packed into a steel cylinder. To attain complete saturation, the packed soil samples were immersed in a 0.02 mol L–1 gypsum solution under vacuum conditions. The soil samples were then left on a sintered porous plate with suction of –17.0 kPa for different periods of time to allow drainage and air intrusion. After this drainage process, the samples were again immersed in water to permit air entrapment. The hydraulic conductivity was measured using the falling head method, and the amount of entrapped air was determined gravimetrically. The quasi-saturated hydraulic conductivity was found to decrease with increasing entrapped air content until the soil had the maximum fraction of entrapped air, approximately 10% of the bulk soil volume. A comparison of the quasi-saturated and unsaturated hydraulic conductivities of the soil samples at or near saturation, when the suction of soil water was greater than the air-entry value, showed that the quasi-saturated hydraulic conductivity was smaller than the unsaturated hydraulic conductivity.
Abbreviations: TUAT, Tokyo University of Agriculture and Technology
|
INTRODUCTION
|
|---|
R EPEATED WETTING–DRYING cycles often leads to entrapped air in the pores of a water-saturated soil. This entrapped air may then form an isolated (discontinuous) air phase that is no longer connected to the atmosphere (Youngs and Peck, 1964; Luckner et al., 1989). The term quasi-saturated soil defines such a soil with entrapped air, while the term quasi-saturated hydraulic conductivity is used here for the relationship between hydraulic conductivity and the entrapped air content (Faybishenko, 1995). "Unsaturated" here refers to the state in which soil air is connected to the atmosphere, and soil water is kept under negative pressure, while quasi-saturated conditions occur even when the pressure is positive, such as under surface ponding.
The entrapped air content typically ranges from 0 to 20% of the bulk soil volume (Fayer and Hillel, 1986). In previous studies, the entrapped air content was determined to be 3 to 15% for a loam soil (Seymour, 2000), up to 10% for another loam soil (Faybishenko, 1995), and 6.5% for a fine sandy loam (Fayer and Hillel, 1986). Other studies found entrapped air to be 2 to 10% (Hanson 1977) and 15% (Bouwer and Rice 1983) of the porosity of a soil. Most previous studies used unstructured loam or sandy loam soils. It is not clear how much air can be entrapped in aggregated soils.
Various papers have reported the effect of entrapped air on flow phenomena in soils. Most field soils contain some amount of entrapped air; hence, field soils are often quasi-saturated, even under ponding conditions. Thus, samples may not always be wholly saturated also for laboratory experiments designed to determine the saturated hydraulic conductivity. Entrapped air that is released from or dissolved into soil water may change the permeability and affect groundwater recharge, remediation of contaminated soils, and irrigation (Koga, 1987).
Entrapped air affects hydraulic conductivity (Gupta and Swartzendruber, 1964) and infiltration (Bond and Collis-George, 1981). Koga (1987) reported that the saturated hydraulic conductivity of a highly compacted andisol decreased by a factor of 100 from its initial value with a 5% increase in the entrapped air fraction. In addition, flow instability and initiation of preferential flow are affected by entrapped air (Wang et al., 1998). Entrapped air is also a dominant factor when surge irrigation is implemented to save irrigation water (Seymour, 2000), and entrapped air also contributes to the instability of soil structure. However, it is still not clear how and to what extent entrapped air blocks pores and reduces the permeability of a soil.
The aim of our study was to evaluate the relationship between the volume of entrapped air and the hydraulic conductivity of a structured quasi-saturated soil to discuss the mechanism that causes entrapped air to reduce soil hydraulic conductivity and to compare the quasi-saturated and unsaturated hydraulic conductivities of two soils
|
MATERIALS AND METHODS
|
|---|
Soils
Two soils, a Masa sandy loam soil (sand: 0.8, silt: 0.05, clay: 0.15 kg kg–1) and a Tokyo University of Agriculture and Technology (TUAT) light clay andisol (sand: 0.32, silt: 0.32, clay: 0.36 kg kg–1), were used in this study. The Masa soil was collected at Miharu in the Fukushima prefecture of Japan. The parent material of the soil was weathered granite rock. The andisol, derived from volcanic ash, was collected at the Field Science Center of the Tokyo University of Agriculture and Technology. The particle densities of the soils were 2.74 and 2.67 g cm–3 for Masa soil and TUAT andisol, respectively. The soils were sieved through a 3-mm mesh screen and then stored in plastic bags until packing to maintain gravimetrical water contents of 0.138 and 0.674 g g–1 for the Masa soil and TUAT andisol, respectively.
Sample Preparation and Determination of the Quasi-Saturated Hydraulic Conductivity
The soils were packed into 5.0-cm-diameter, 5.1-cm-long steel cylinders. Dry bulk densities for the Masa soil samples were 1.52, 1.58, and 1.71 g cm–3, with porosities of 0.446, 0.422, and 0.376, respectively. Packing dry bulk densities for the andisol samples were 0.68, 0.82, and 0.89 g cm–3, with porosities of 0.446, 0.422, and 0.376, respectively. At the bottom of each steel cylinder, a 20-µm mesh membrane filter was installed to prevent the flashing of fine particles. Because of its much higher hydraulic conductivity than that of the soils, the membrane did not affect flow through the soil samples.
Complete water saturation was attained using the following procedure. The packed soil column was kept on a deep tray in a vacuum chamber. To prevent additional dispersion of soil colloids, the samples were saturated using a 0.02 mol L–1 gypsum solution provided through a siphon installed inside the vacuum chamber (Fig. 1)
. Before saturation, air was evacuated for more than 30 min using an air pressure inside the chamber of approximately –96.8 kPa until no bubbles were released from the gypsum solution. The vessel containing the gypsum solution was tilted to allow solution to be added into the tray through the siphon. As a result, the packed soil samples were entirely immersed in the solution. Although some entrapped air may still have remained in the soil, it likely shrank or dissolved into the gypsum solution because of the pressure increase when the chamber was opened to the atmosphere.
The saturated sample was weighed and the saturated hydraulic conductivity measured using the falling head method (Klute and Dirksen 1986). The sample was subsequently placed on a porous plate and subjected to a negative pore water pressure of –17 kPa to cause drainage. The pore water of the soil sample was forced to drain through the porous plate, while at the same time air intruded into voids of the soil during a period of a few minutes to several hours.
At the cessation of drainage, the samples were again immersed in the gypsum solution for resaturation while allowing for the presence of entrapped air. The amount of entrapped air was determined by the difference in the weight of the completely saturated soil and that after resaturation. After measuring the hydraulic conductivities and weights of the resaturated samples, the samples were again placed onto the porous plate. Measurements of the hydraulic conductivity and the weight following drainage and resaturation were repeated until the amount of entrapped air, or the weight of the quasi-saturated soil sample, reached a constant value.
Unsaturated Soil Hydraulic Properties
The same soil column was used to measure the unsaturated hydraulic properties. The hanging water column method was used to determine the water retention characteristic of the soils. The steady-state pressure control method (Klute and Dirksen, 1986) was used to determine the unsaturated hydraulic conductivity of the TUAT andisol for pressure heads ranging from –1.3 to –6.6 kPa. The pressure control method was used for pressure heads from –3.6 to –7.5 kPa for the Masa soil. Instead of the pressure control method, the steady-state flux method (Klute and Dirksen, 1986) was employed for the Masa soil for pressure heads greater than –3.6 kPa when the unsaturated hydraulic conductivity of the soil was approximately 4.0 x 10–6 m s–1 or greater. Measurements of the unsaturated hydraulic conductivity were conducted for the drainage process.
|
RESULTS AND DISCUSSION
|
|---|
Hydraulic Conductivity and Entrapped Air
Figures 2 and 3
show relationships between the entrapped air volume and the quasi-saturated hydraulic conductivity. The entrapped air content is the ratio of volume of entrapped air to the volume of the entire soil column. Figure 2 presents results for the Masa soil, having a dry bulk density of 1.52 g cm–3, while Fig. 3 shows results of the TUAT andisol, having a dry bulk density of 0.82 g cm–3. Results for the three replicates, shown in both figures, agree fairly well.
View larger version (13K):
[in this window]
[in a new window]
|
Fig. 2. Quasi-saturated hydraulic conductivity of the Masa sandy loam as a function of entrapped air content ( b = 1.52 g cm–3).
|
|
View larger version (13K):
[in this window]
[in a new window]
|
Fig. 3. Quasi-saturated hydraulic conductivity of the TUAT andisol as a function of entrapped air content (b = 0.82 g cm–3).
|
|
For the Masa soil, the fully saturated hydraulic conductivity was 5.98 x 10–6 m s–1, while the quasi-saturated hydraulic conductivity decreased to 3.80 x 10–7 m s–1. For the TUAT andisol, the quasi-saturated hydraulic conductivity decreased from 3.31 x 10–6 to 2.45 x 10–7 m s–1. The maximum entrapped air volume was 10.5% of the entire volume (23% of the void volume) for the Masa soil and 9.0% of whole volume (13% of the void volume) for the TUAT andisol, respectively. These results agree with those by Fayer and Hillel (1986), Faybishenko (1995), and Seymour (2000), who found the maximum entrapped air to be 6.3 to 15% of the entire soil volume, while the hydraulic conductivity with entrapped air decreased to approximately one-tenth of the initial value.
Faybishenko (1995) modified Averjanov's power function (Mualem, 1986) to obtain the following expression for the quasi-saturated hydraulic conductivity of soils:
 | [1] |
where 
, s, and qs are the volumetric water content, saturated volumetric water content, and volumetric water content with maximum entrapped air, respectively (m3 m–3); 
max = s – qs (m3 m–3) is the maximum entrapped air; and = s – (m3 m–3) is the content of entrapped air; K(), Ks, and K0 are the quasi-saturated hydraulic conductivity, saturated hydraulic conductivity, and quasi-saturated hydraulic conductivity with maximum entrapped air content, respectively (m s–1); and n is a fitting parameter.
The solid lines in Fig. 2 and 3 were obtained by fitting Eq. [1] to the measured data. The fit of the data was quite good, yielding regression coefficients for the Masa and TUAT soils of 0.98 and 0.96, respectively. The Masa soil had a relatively poor soil structure, while the TUAT andisol was much more aggregated. Figures 2 and 3 suggest that Eq. [1] correctly reflects a trend in hydraulic conductivity changes for both aggregated and poorly structured soils, as a function of entrapped air content.
Effect of Dry Bulk Density on Quasi-Saturated Hydraulic Conductivity
Figures 4 and 5
show how the hydraulic conductivity decreased as a function of the entrapped air content for different values of dry bulk density. Both soils showed the same trend in that the soils having a lower dry bulk density (b) or a higher saturated hydraulic conductivity exhibited a drastic decrease in hydraulic conductivity with a small increase in entrapped air content when the entrapped air content was very small. Further increases in the volume of entrapped air caused much smaller drops in the hydraulic conductivity. Soils having higher dry bulk densities or lower saturated hydraulic conductivities behaved differently in that the hydraulic conductivity decreased more linearly (on a log scale) as a function of entrapped air content.
View larger version (16K):
[in this window]
[in a new window]
|
Fig. 4. Effect of dry bulk density on the quasi-saturated hydraulic conductivity of the Masa sandy loam.
|
|
The empirical parameter, n, in Eq. [1] reflects the shape of the curvature of the conductivity function. The relationship between n and the fully saturated hydraulic conductivity, Ks, is shown in Fig. 6
. This figure also contains results obtained by Faybishenko (1995). The relationship between n and the saturated hydraulic conductivity is uniquely expressed by the empirical power function proposed by Faybishenko (1995).
Comparison of Unsaturated Hydraulic Conductivity and Quasi-Saturated Hydraulic Conductivity
Figures 7 and 8
show unsaturated and quasi-saturated hydraulic conductivities of the Masa and TUAT soils, respectively. The air phase fraction of the unsaturated soil was estimated from the measured pressure head and the soil water retention curve (Fig. 9)
. It is interesting that between near complete saturation and the maximum entrapped air content (at approximately 10% entrapped air content), the quasi-saturated hydraulic conductivity was always smaller than the unsaturated hydraulic conductivity.
The saturated hydraulic conductivity is an important parameter needed for evaluating the unsaturated hydraulic conductivity function because most unsaturated hydraulic conductivity models include two terms, such as the saturated hydraulic conductivity and the relative hydraulic conductivity as a function of matric potential, volumetric water content, or degree of saturation (Mualem 1986). Moreover, several unsaturated hydraulic conductivity models assume that the unsaturated hydraulic conductivity is equivalent to the saturated hydraulic conductivity when the pressure head is greater than the air-entry value (Brooks and Corey 1964). However, Fig. 7 and 8 suggest that the quasi-saturated hydraulic conductivity, which is commonly measured in situ and often assumed to be equal to the saturated hydraulic conductivity, is quite sensitive to entrapped air. The air-entry values of the Masa and TUAT soils were –3.5 and –2.5 kPa, respectively. These values correspond to air phase fractions of approximately 3.2 and 2.5%, respectively, and to unsaturated soil hydraulic conductivities of 4.22 x 10–6 and 2.37 x 10–6 m s–1, respectively (Fig. 7 and 8).
The quasi-saturated hydraulic conductivity was distinctly lower than the unsaturated hydraulic conductivity at the air-entry value. Although the tensiometer pressure of soil water of the sample during quasi-saturated conditions was equal to or greater than zero (the samples were submerged), the quasi-saturated hydraulic conductivity was still less than the hydraulic conductivity of the unsaturated soils having the same air phase ratio.
The following may explain why the unsaturated hydraulic conductivity was greater than the quasi-saturated hydraulic conductivity at the same air fraction. When a pressure head less than the air-entry value is applied to the soil, this will lead to drainage. This drainage starts with the larger pores. One may assume that soil pores having diameters greater than that corresponding to the applied pressure head through the capillary rise equation will now become filled with air. At the same time, water may flow through the water-filled pores having smaller diameters than that corresponding to the applied suction. At quasi-saturation, entrapped air spreads to the soil pore system, and may not only clog or block bigger pores, but also may stay in and/or block finer pores and pore throats. This may cause the hydraulic conductivity of quasi-saturated soil to become smaller than the unsaturated hydraulic conductivity at the same air phase fraction.
The maximum entrapped air content corresponded to pressure heads of –5.0 and –4.0 kPa for the Masa soil having a dry bulk density of 1.52 g cm–3 and the TUAT andisol having a dry bulk density of 0.82 g cm–3, respectively (Fig. 9). These values are just below the air-entry value of the soil samples. Values of K0 and qs in Eq. [1] can be estimated from the unsaturated hydraulic conductivity and the water retention curve at pressure heads slightly smaller than the air-entry value. Values of Ks, and hence n, in Eq. [1] may now be determined from direct measurements of the completely saturated hydraulic conductivity (Fig. 6), while s is now equivalent to the porosity of the soil. Thus, if basic soil hydraulic properties are measured using common laboratory methods, the quasi-saturated field soil hydraulic conductivity can be predicted quickly and easily from field volumetric water content measurements as obtained with time domain reflectometry. Our method may provide better predictions of highly transient infiltration and drainage events at or near saturation such as during surge irrigation, ponding and drainage of paddy fields, groundwater recharge, and remediation of contaminated soils.
|
CONCLUSIONS
|
|---|
Our investigations of two structured soils, a Masa sandy loam and a TUAT andisol, showed the same volume of entrapped air as that of less structured soils studied previously. The power function proposed by Faybishenko (1995) correctly expressed the trend of decreasing hydraulic conductivities of aggregated soils with increasing entrapped air. Since the saturated hydraulic conductivity is a key parameter in many unsaturated hydraulic conductivity models, the presence of entrapped air may affect not only Ks itself, but also the entire unsaturated hydraulic conductivity function. Our study shows that parameters of the power function for the quasi-saturated hydraulic conductivity can be determined using common laboratory measurements.
Measured quasi-saturated hydraulic conductivities of our soils were smaller than the unsaturated hydraulic conductivities at the same air content. This finding could be explained by the fact that entrapped air clogs the largest water conducting pores. Our results are important for making improved predictions of highly transient infiltration and drainage events at or near saturation.
|
ACKNOWLEDGMENTS
|
|---|
This study was supported by a Grant in Aid for Scientific Research (B-13556036) from the Japan Society for the Promotion of Science (JSPS). We would thank to the Field Science Center of TUAT for sampling soil materials.
|
REFERENCES
|
|---|
- Bond, W.J., and N. Collis-George. 1981. Ponded infiltration into simple soil systems. 1. The saturation and transition zones in the moisture content profiles. Soil Sci. 131:202–209.
- Bouwer, H., and R.C. Rice. 1983. Effect of stones on hydraulic properties of vadose zones. p. 77–93 In Proc. on the Characterization and Monitoring of the Vadose (Unsaturated) Zone, Las Vegas, NV. 8–10 Dec. 1983. National Water Association, Worthington, OH.
- Brooks, R.H., and A.T. Corey. 1964. Hydraulic properties of porous media. Hydrol. Paper 3. Colorado State Univ., Fort Collins
- Faybishenko, B.A. 1995. Hydraulic behavior of quasi-saturated soils in the presence of entrapped air. Water Resour. Res. 31:2421–2435.
- Fayer, M.J., and D. Hillel. 1986. Air encapsulation: I. Measurement in a field soil. Soil Sci. Soc. Am. J. 50:568–572.[Abstract/Free Full Text]
- Gupta, R.P., and D. Swartzendruber. 1964. Entrapped air content and hydraulic conductivity of quartz sand during prolonged liquid flow. Soil Sci. Soc. Am. J. 28:9–12.[Abstract/Free Full Text]
- Hanson, B.R. 1977. Entrapped gas in an unconfined aquifer. Ph.D. diss. Colorado State Univ., Fort Collins. (Diss. Abstr. 77-28309).
- Klute, A., and C. Dirksen. 1986. Hydraulic conductivity and diffusivity: Laboratory methods. p. 687–734. In A. Klute (ed.) Methods of soil analysis. Part 1. Agron. Monogr. 9. ASA and SSSA, Madison, WI.
- Koga, K. 1987. Reduction in entrapped air by percolation and the change in permeability—Study on the permeability of compacted soil (II). (In Japanese with English abstract.) Trans. Jpn. Soc. Irrig. Drainage Reclam. Eng. 131:69–77.
- Luckner, L., M.Th. van Genuchten, and D.R. Nielsen. 1989. A consistent set of parametric models for the two-phase flow if immiscible fluids in the subsurface. Water Resour. Res. 25:2187–2193.
- Mualem, Y. 1986. Hydraulic conductivity of unsaturated soils: Prediction and formulas. p. 799–823. In A. Klute (ed.) Methods of soil analysis. Part 1. Agron. Monogr. 9. ASA and SSSA, Madison, WI.
- Seymour, R.M. 2000. Air entrapment and consolidation occurring with saturated hydraulic conductivity changes with intermittent wetting. Irrig. Sci. 20:9–14.
- Wang, Z., J. Feyen, M.Th van Genuchten, and D.R. Nielsen. 1998. Air entrapped effects on infiltration rate and flow instability. Water Resour. Res. 34:212–222.
- Youngs, E.G., and A.J. Peck. 1964. Moisture profile development and air compression during water uptake by bounded porous bodies. Soil Sci. 98:290–294.
This article has been cited by other articles:
|
|
|
|
 
L. Luo, H. Lin, and P. Halleck
Quantifying Soil Structure and Preferential Flow in Intact Soil Using X-ray Computed Tomography
Soil Sci. Soc. Am. J.,
June 18, 2008;
72(4):
1058 - 1069.
[Abstract]
[Full Text]
[PDF]
|
|
|
TOP
ABSTRACT
INTRODUCTION
