Soil Science Society of America Journal 66:1501-1504 (2002)
© 2002 Soil Science Society of America
DIVISION S-1—NOTES
Determination of hydraulic behavior of hillsides with a hillslope infiltrometer
Guillermo Mendozaa,b and
Tammo S. Steenhuis*,a
a Dep. Bio. and Env. Eng., Riley-Robb Hall, Cornell Univ., Ithaca, NY 14853-5701
b The New York City Dep. of Environ. Protection, Kingston, NY 12401
* Corresponding author (tss1{at}cornell.edu)
 |
ABSTRACT
|
|---|
Watersheds, in many parts of the world, consist of sloping soils with a dense subsoil at shallow depth. Very few measurement techniques exist for realistically determining hydraulic properties in situ on these hillside soils. A hillslope infiltrometer, open at the bottom, top, and downhill sides, was developed that can measure the vertical saturated conductivity of the horizon sequence and lateral hydraulic conductivity of the horizon most conductive to water. The infiltrometer was tested on the steeply sloping soils in Honduras. Increasing rainfall intensities were applied and lateral flow rate was measured in several troughs at the downslope end of the hillslope infiltrometer. Vertical infiltration rates for the horizon sequence and lateral conductivity were calculated at each rainfall rate after steady state conditions were established. The hillslope infiltrometer proved useful in the characterization of subsurface flow on the steep slopes in Honduras.
|
INTRODUCTION
|
|---|
HILLSIDES DOMINATE THE HYDROLOGY of watersheds in many parts of the world. Hillside soils often have a permeable root zone overlying more dense subsoil. Both vertical and lateral saturated hydraulic conductivity affects the amount of runoff, interflow, and deep percolation. In particular, the saturated conductivity through the horizon sequence is an important parameter because runoff is only generated for large storms when the rainfall exceeds percolation. There has been a great interest in measuring hillslope hydraulic characteristics. Dunne and Black (1970) conducted a hillslope experiment by measuring subsurface flow in a trench in Vermont. Similarly, Mosley (1982) measured hillslope hydraulic characteristics by excavating a trench along the contour to the bedrock and applying water with a line source uphill. Also around that time, Harr (1977) installed piezometers and measured flow from a steep forested slope in Oregon. Later, Anderson et al. (1997) and Torres et al. (1998) conducted tracer studies in a steep watershed in Oregon using piezometers and suction lysimeters. The main interest of these hillslope experiments was to determine flow paths. It was difficult to derive saturated vertical and lateral conductivities from these experiments, since the specific discharge was not known accurately because of the transient flow conditions. There are few methods to measure the conductivity in situ for these types of soils and the studies mentioned above used laboratory derived hydraulic parameters. The existing field methods for vertical saturated hydraulic conductivity usually assume that the soil is homogeneous in the vertical direction (Bouwer and Rice, 1964; Green et al., 1986). If the conductivity is changing with depth, lateral flow downslope can occur, invalidating the basic assumption that the specific discharge is constant. Laboratory measurements are not always a good alternative to in situ methods for measuring saturated conductivity since the measured values are either too low when the soil is disturbed or too high when undisturbed cores are used. In undisturbed cores, pores are open that are dead ended in the field.
There is an urgent need for accurately measuring hydraulic properties of hillslopes in situ (Sherlock et al., 2000). In most hillslope soils, the saturated conductivity decreases with depth (Ambroise et al., 1996; Scanlon et al., 2000). In this note, we report on an apparatus (called the hillslope infiltrometer) for determining the hydraulic properties of soils where the conductivity is decreasing with depth.
|
Materials and Methods
|
|---|
The hillslope infiltrometer is a metal box that slides around and isolates a soil column. It has open ends at the bottom, top, and downhill sides (Fig. 1)
. The box is constructed of three plates of 22 U.S. gauge (1 mm thick) stainless steel (two plates are 36 by 41 cm and one is 30 by 41 cm, Fig. 1a). To make the infiltrometer portable, the plates are connected using steel piano hinges so they can be folded into one flat piece. The open downhill side has three connectors to hold the plates of the box tightly around the soil column, and three lateral collectors to catch and measure the water flow. The connectors fit into slits made from metal (Fig. 1b). The lateral flow collector consists of a trough and lip (Fig. 1c). The 5 by 30.5 cm lip is inserted into the soil. The trough, 30.5 cm long and 6 cm deep, is sloped so that water will accumulate at the downslope end, making collection easier.
View larger version (38K):
[in this window]
[in a new window]
|
Fig. 1. Schematic diagram of the hillslope infiltrometer: (A) Infiltrometer, (B) Connector, (C) Lateral flow collector.
|
|
Water is applied with a rainfall simulator consisting of a plexiglass container 30 cm wide, 30 cm long, and 30 cm high. The raindrops are formed by 64 capillary tubes (2.4-mm i.d.) 30 mm apart (Ogden et al., 1997). The rate is regulated by setting the pressure head with a Mariotte-type bubbling tube inserted through the top of the container.
The hillslope infiltrometer is installed in the field by carving out a soil block that is slightly smaller than the infiltrometer. An one to two spade wide trench is left around the block for ease of installation and collection of water from the troughs. Once the block has the correct dimensions, the folded out steel plates are laid horizontally on the ground, newspapers are taped to the inside, and an expandable foam (used for insulation of houses) is sprayed evenly over the newspapers. Then, at once, the plates (with the foam) are put around the soil column. To hold them in place, the three connectors (Fig. 1b) are inserted into the slits of the box before the foam fully expands. Lastly, the flow collectors are put into the soil (Fig. 1c). Runoff is collected in the topmost collector 2 to 4 cm from the surface (Fig. 2)
. Interflow originating from the A-B soil horizon (A-B interflow) is measured with a lateral flow collector located at the B-C soil horizon. A third collector near the bottom end of the infiltrometer (Fig. 2) captures interflow from the C soil horizon (C interflow). The vertical percolation rates for the different layers are calculated from these outflow rates and the rainfall intensity. The lateral conductivity in the A and B soil horizon is determined by measuring the difference in hydraulic head between two 2-cm wide mini wells ending at the B-C interface, spaced 18 cm apart and located 11 cm from the outlet face (Fig. 2). At the surface, small mounds are made around the well so that the surface runoff cannot flow into the wells. Note, that above we assumed the typical A B C horizon sequence. Obviously, any horizon sequence can be studied.
Rainfall is applied in an ascending sequence of constant rates. We selected the highest application rate to be close to the maximum observed rainfall rate in the area of 20 cm h-1. The rainfall rate is increased after the outflow is at steady state. The experiment is stopped when the highest rate is reached or the front of the soil column breaks off because of over saturation of the soil block. The latter does not compromise the experiment because all the required data will have been gathered at a lower rainfall rate.
Outflow volumes are measured approximately every 5 min (high rates of flow are measured at shorter time intervals before the trough overflows) by pumping water out of the collectors (Fig. 2) with a hand pump and measuring the volume with a graduated cylinder. When volume measurements indicate that steady state is reached, water levels in the mini-wells are measured.
The hillslope infiltrometer was tested in an area with steep hills in the Lavanderos community in Honduras, on a plot with conventional agriculture. Maize (Zea mays) and beans (Phaseolus vulgaris) were grown in rotation without any conservation practices on a slope of about 26% on a sandy loam soil A-B horizon overlying a clay loam C horizon starting at a depth of 20 cm. Rainfall rates used were 47, 112, and 175 mm h-1.
|
Theory
|
|---|
Three flow rates in the vertical direction were computed during steady state for each rainfall rate: the infiltration into the A and B horizon, iA-B; percolation into the C horizon, iC; and the recharge flowing out of the bottom of the infiltrometer soil through the C horizon, iR:
 | [1] |
 | [2] |
 | [3] |
where the rainfall rate, p, steady state runoff, r, steady state interflow in the A-B soil horizon, fA-B, and steady state C interflow at the lower end of the infiltrometer box, fC, are all measured. Initially, at low flow rates, the soil is unsaturated and an increase in rainfall rate results in an increase in the infiltration rate that is equal to the rainfall rate increase. When the soil becomes saturated, the infiltration rate remains constant or increases slightly because of small increases in hydraulic gradient.
When the soil is just saturated throughout a layer, the matric and pressure potentials in this layer are near zero and, consequently, the hydraulic gradient is close to unity. Then, for steady state conditions when the soil is saturated, assuming an unit gradient, the vertical saturated conductivities are equal to the infiltration rates into the respective horizons (i.e., KvA-B and KvC = iC).
The lateral conductivity in the A-B soil horizon, KhA-B, is determined with the water height in the two mini wells as boundary conditions. The distance of the mini wells are x1 and x2 from the lower end of the infiltrometer. The water heights in the mini wells, h1 and h2, are measured from the horizontal plane at the average height of the B-C interface. Using the Dupuit-Forschheimer assumptions, the height of the water table, h, in the A and B horizon during steady state can then be expressed as (after Fetter, 1994, p. 143),
 | [4] |
where fA-B is the net influx in the A and B horizon. The flux in the A and B horizon per unit width, qx, at any distance x from the origin can be found by substituting H for h2 in Eq. [4] and differentiating. Since qx = Khdh/dx = 0.5KdH/dx, we find that:
 | [5] |
By evaluating Eq. [5] at x = 0, noting that q0 is equal to the product of the net inflow, fA-B, and the length, L, over which the water was applied, the saturated lateral conductivity in the A and B horizon can be expressed as:
 | [6] |
|
Results
|
|---|
The data gathered with the hillslope infiltrometer are given in Fig. 3
. The rainfall rate (thick line) was increased twice. Initially, the rate was 47 mm h-1, at 60 min it was increased to 112 mm h-1 and, finally, at 85 min increased to the rate of 175 mm h-1. Runoff, A-B interflow, and C interflow were measured (Fig. 3). The flows are given in millimeters per hour, obtained by dividing the volumes collected in the troughs by the corresponding cross-sectional areas. There was no interflow until about 16 min after the beginning of the experiment (Fig. 3). After 16 min, interflow began and increased until it became steady at 35 min for the A-B soil horizon (21 mm h-1) and 45 min for the C soil horizon (9 mm h-1). After the rainfall rate was increased to 112 mm h-1, steady state A-B interflow was 74 mm h-1 (reached at 70 min), and C interflow remained unchanged at 9 mm h-1. Substantial surface runoff occurred only after 100 min during the 175 mm h-1 rainfall rate (Fig. 3). Steady state flow rates during the 175 mm h-1 rainfall intensity were 10 mm h-1 of surface runoff, 119 mm h-1 of A-B interflow, and 13 mm h-1 of C interflow.
View larger version (22K):
[in this window]
[in a new window]
|
Fig. 3. Runoff, lateral flow for the A-B horizon and C horizon, infiltration into the A-B horizon, percolation into the C horizon, and recharge for increasing rainfall intensities for one replicate of the agricultural field in the Lavanderos community without soil and water conservation practices.
|
|
The calculated infiltration, percolation, and recharge rates are also shown in Fig. 3. The A-B infiltration (solid triangles), C percolation (solid circles), and recharge curves (solid inverted triangles) were calculated from the rainfall applied and collected lateral flows using Eq. [1] through [3]. The A-B infiltration followed the rainfall intensity until the 175 mm h-1 rate when it became 165 mm h-1. Thus, initially, the A-B soil horizon was unsaturated and only under the highest rainfall intensity, the infiltration rate became less than the rainfall rate, indicating that the interface was near saturation. The C horizon percolation is more complex. From 0 to 20 min, the C percolation rate was equal to the rainfall intensity and then decreased to a steady state rate of 26 mm h-1. To understand this, we note that the C percolation rate (solid circles) was calculated as the difference in the A-B infiltration and B interflow assuming steady state conditions (Eq. [2]). Initially, from 0 to 20 min, the system is not at steady state because part of the water infiltrating in the A-B horizon is used for building up the saturated layer above the B-C interface. Thus, as long as the flow is transient, the C percolation rate in Fig. 3 overestimates the actual infiltration rate. Only at steady state (from 40–60 min) Eq. [2] is valid and the correct percolation rate can be obtained from Fig. 3 as 26 mm h-1. Immediately after the rainfall intensity was increased to 110 mm h-1, a spike in the C percolation rate occurred. The spike was, again, a result of the system not being at steady state because the saturated layer above the B-C interface was being built up to facilitate the higher interflow. Eventually, after some time, the soil moisture storage became constant and the spike disappeared.
The hillslope infiltrometer was, subsequently, tested in several soils in the Lavanderos community and other parts of Honduras and, recently, in the USA in Idaho and New York. In the Lavanderos community, a farmers' survey was performed on the effectiveness of soil and water conservation practices. Infiltrometer results agreed very well with the farmers' perception when runoff was likely to occur. It appeared that when the saturated hydraulic conductivity of one of the soil layers (C horizon in all cases) was lower than the prevailing rainfall rates, soil and water conservation practices were installed voluntarily.
|
Conclusions
|
|---|
Is the hillslope infiltrometer better than the traditional measurements for determining hydraulic properties of the soil? Traditional conductivity measurement (with ponded water) makes the assumption that the soil is homogeneous and that the cross-sectional area for flow does not increase with depth, which is invalid for sloping lands with a dense layer close to the surface. From the Results section above we can calculate that the saturated hydraulic conductivity in the A-B horizon was 16.5 cm h-1 vertically and 10 cm h-1 horizontally. The vertical conductivity of the C horizon was 4 cm h-1. Although the conductivities derived from the flows of the hillslope infiltrometer depend on how well the assumptions are met, they give a wealth of information on the flow rates of the horizon sequence. The additional effort invested in determining the lateral and vertical flow rates with the hillslope infiltrometer over the traditional methods, therefore, is justified.
Received for publication December 19, 2000.
|
REFERENCES
|
|---|
- Ambroise, B., K. Beven, and J. Freer. 1996. Toward a generalization of the TOPMODEL concepts: Topographic indices of hydrological similarity. Water Resour. Res. 32:2135–2145.
- 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.
- Bouwer, H., and H.C. Rice. 1964. Simplified procedure for calculation of hydraulic conductivity with the double tube method. Soil Sci. Soc. Am. Proc. 28:133–134.
- Dunne, T., and R.D. Black. 1970. An experimental investigation of runoff production in permeable soils. Water Resour. Res. 6: 478–490.
- Fetter, C.W. 1994. Applied Hydrogeology. 4th ed. Prentice Hall, NJ.
- Green, R.E., L.R. Ahuja, and K. Chong. 1986. Hydraulic conductivity, diffusivity and sorptivity of unsaturated soils: Field methods. p. 771–798. In A. Klute (ed.) Methods of Soil Analysis. Part 1. 2nd ed., Agron. Monogr. 9. ASA and SSSA, Madison, WI.
- Harr, R.D. 1977. Water flux in soil and subsoil on a steep forested slope. J. Hydrol. 33:37–58.
- Mosley, M.P. 1982. Subsurface flow velocities through selected forest soils, South Island, New Zealand. J. Hydrol. 55:65–92.
- Ogden, C.B., H.M. van Es, and R.R. Schindelbeck. 1997. Miniature rain simulator for field measurement of soil infiltration. Soil Sci. Soc. Am. J. 61:1041–1043.[Abstract/Free Full Text]
- Scanlon, T.M., J.P. Raffensperger, G.M. Hornberger, and R.B. Clapp. 2000. Shallow subsurface storm flow in a forested headwater catchment: Observations and modeling using a modified TOPMODEL. Water Resour. Res. 36:2575–2586.
- Sherlock, M.D., N.A. Chappell, and J.J. McDonald. 2000. Effects of experimental uncertainty on the calculation of hillslope flow paths. Hydrol. Proc. 14:2457–2471.
- 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.
This article has been cited by other articles:
|
|
|
|
 
V. Bagarello and A. Sgroi
Testing Soil Encasing Materials for Measuring Hydraulic Conductivity of a Sandy-Loam Soil by the Cube Methods
Soil Sci. Soc. Am. J.,
June 18, 2008;
72(4):
1048 - 1057.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
J. L. Starr, A. M. Sadeghi, and Y. A. Pachepsky
Monitoring and Modeling Lateral Transport through a Large In Situ Chamber
Soil Sci. Soc. Am. J.,
October 27, 2005;
69(6):
1871 - 1880.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
W. Bodhinayake, B. C. Si, and K. Noborio
Determination of Hydraulic Properties in Sloping Landscapes from Tension and Double-Ring Infiltrometers
Vadose Zone J.,
August 1, 2004;
3(3):
964 - 970.
[Abstract]
[Full Text]
[PDF]
|
|
|
TOP
ABSTRACT
INTRODUCTION
