Introduction

Anisotropy, which is the opposite of “isotropy,” is a term used to denote preferential flow direction in soils and other geologic materials. If soil consisted of perfectly spherical grains, flow rates would be isotropic – the same in all directions, other factors being equal. Soil doesn’t consist of perfectly spherical grains, however.

It’s commonly understood that flow of air and water through soils is greatly influenced by grain size, but grain shape is also key. Clay minerals have a platy crystaline habit and tend to lie flat in alluvial deposits. Sand and silt grains made of micas or metamorphic rock fragments with slatey cleavage tend to land flat side down when deposited.

In soils, there is usually a substantial decrease in vertical hydraulic conductivity going from the A (topsoil) to the B (subsoil) horizon. Soil-forming processes create weathering products such as clay, sesquioxides (iron and aluminum oxyhydroxides), carbonates, and silicates that accumulate in the B horizon. Roots and burrowing soil fauna increase conductivity in the A horizon.

Depending on the physical properties of the soil, preferential flow could occur in any direction, but the tendency is for higher horizontal rates. For many layered soils, the hydraulic conductivity perpendicular to the soil layers is slower than the conductivity parallel to soil layers. The lateral conductivity is determined by the layer with highest conductivity, while the vertical conductivity is controlled by the layer with lowest hydraulic conductivity (Zaslavsky and Rogowski, 1969; Todd, 1980).

Zaslavsky and Rogowski (1969) found the combined effects of slope and soil anisotropy caused subsurface lateral flow in both saturated and unsaturated conditions. Subsurface lateral flow has important water quality ramifications by limiting the vertical absorptive capacity of the soil.

Steenhuis and Muck (1988) studied the downslope movement of chlorine and nitrate during runoff events and found “interflow,” a type of subsurface lateral flow that occurs when the soil moisture is between field capacity and saturation, to be an important transport process and a necessary precursor to surface runoff. The interflow was set up by a reduction in hydraulic conductivity in an underlying layer. Instead of allowing steady infiltration to the deeper soil, the anisotropic properties of the subhorizon promoted saturated flow in the upper, more permeable soil. Rapid subsurface lateral flow and its associated runoff may be capable of transporting a variety of pollutants, for example: road salt, fertilizers, pesticides, animal wastes, engine fluids from salvage yards, and leachate from on-site waste disposal systems.

Mechanisms of Anisotropy

Todd (1980) noted that, especially for unconsolidated undisturbed alluvial material, anisotropic behavior is the norm. Two reasons for anisotropy in unconsolidated sediments include: i)individual particles are rarely spherical. When they settle out of the water column, they land with their flat sides down.

Sheet silicate minerals (phyllosilicates) such as micas and most clay minerals have a platy crystalline habit, and would tend to land on their flat sides); and ii) alluvium tends to exist in layers of different grain sizes that reflect changes in the flow regime or sediment source during deposition. Each layer has its own unique hydraulic conductivity.

For a set of horizontal layers, if one layer has a relatively low hydraulic conductivity, vertical flow is restricted, but horizontal flow is rapid through overlying layers with relatively high hydraulic conductivity. Hydraulic conductivities in layered sediments tend to be higher in the horizontal direction (Todd, 1980).

Soil-forming process, including leaching of materials downward by infiltrating water, upward migration of materials by capillary draw, and faunal turbation by earthworms and other animals produce “horizons” that may have different hydraulic behavior than the horizons above or below. Generally, layered soils will have higher hydraulic conductivities in the lateral direction than in the vertical direction (Zaslavsky and Rogowski, 1969), although this is not always true (Hammermeister et al., 1982a; Dabney and Selim, 1987).

Soil “B” horizons are zones of accumulation where translocated materials are deposited onto mineral grain surfaces and into pores, effectively reducing the hydraulic conductivity of that soil layer (Whipkey and Kirkby, 1978). Argillic horizons are B horizons where the accumulated material is clay; spodic horizons are B horizons where the accumulated material is humus and amorphous compounds of iron and aluminum; calcic horizons form as the result of precipitation of calcite mobilized from calcite-rich materials above or below the calcic horizon. All of these B horizons can potentially restrict the downward infiltration of water and cause saturated conditions near the soil surface.

Soil structure, i.e., the pattern and size of soil aggregates, soil texture, and depth are key variables in determining anisotropic flow in soil profiles. The weight of overlying materials, or “geostatic pressure,” further tends to reduce porosity and hydraulic conductivity with depth (Whipkey and Kirkby, 1987).

Layered soils do not always have higher hydraulic conductivity in the horizontal direction as opposed to the vertical direction (Bouma et al., 1982). Hammermeister et al. (1982a, b) found soil “A” horizons on Oregon forested hillslopes having higher hydraulic conductivity in the vertical direction than in the horizontal direction, presumably due to faunal turbation and vertically-oriented root channels. In their study, surface lateral flow was caused by saturation of soil horizons immediately overlying a dense till parent material. The till acted as an aquitard. Changes in lithology are often associated with anisotropic flow of roundwater.

Some soils, such as Vertisols, and Mollisols develop large, continuous cracks due to shrinking and swelling of expansible clays, such as smectite. Such soils may have very high vertical hydraulic conductivity, compared to horizontal conductivity (Dabney and Selim, 1987). This situation can be potentially hazardous by threatening groundwater quality if contaminants are introduced to the soil.

Hydrologic Implications

Anisotropy and Hillslope Hydrology
Chorley (1978) took issue with the Hortonian model (Horton, 1941) of overland flow. The Hortonian runoff model essentially says that when the rate of precipitation exceeds the infiltration capacity of the soil, overland flow will occur (precipitation-infiltration capacity = overland runoff). Instead, Chorley formulated a more elaborate model of hillslope hydrology that accounted for layered profiles with flow restricting layers, subsurface lateral flow or throughflow,” and the focusing of infiltrating water to the bottom of the hill (Fig. 1).

Because of downhill drainage, footslopes tend to have higher antecedent moisture than higher backslope and summit positions. He also noted that hydraulic conductivities in coarse soils were faster when the soils were at least 80% saturated and capillary forces were negligible. This is a significant departure from the Hortonian model, since Chorley’s scenario can lead to rapid, saturated flow in select areas of a watershed, even though precipitation rates never exceeded the infiltration rates of the soil. These limited areas prone to surface runoff include: i) footslope positions near surface water bodies, ii) concavities where streamlines converge near the land surface (Fig. 2), and iii) areas of thin soil cover, where infiltration is limited by an underlying impermeable formation.

The “concavities” mentioned above include both profile concavities and contour (lateral) concavities. On a topographic map, contour concavities appear bay-like or horseshoe-shaped. Both profile and lateral concavities force flowlines to converge at an area that becomes more saturated than the rest of the landscape. Anisotropic behavior, with horizontal hydraulic conductivity (Kx) greater than vertical hydraulic conductivity (Ky), enhances the process of subsurface lateral flow and partial area saturation (Fig. 3).

Zaslavsky and Rogowsky (1969) and Burt and Trudgill (1985) presented a model for the combined effects of hill slope angle and soil anisotropy (Figure 3). The angle of the flowline through the soil (beta) can be calculated from the land surface slope angle (alpha), the lateral hydraulic conductivity (Kx), and the vertical conductivity (Ky) using the equation:

[tan β = 1/U tan α],
where U = Kx/Ky, Kx/Ky > 1 (Figure 3). This relationship indicates that as the slope becomes steeper and U becomes greater, i.e., Kx/Ky gets greater, the flowlines run more shallow and parallel to the land surface.

Contaminant Transport

Anisotropic behavior is important to consider when selecting sites for waste disposal, or when applying to land any chemical intended to stay where it was put. Steenhuis and Muck (1988) studied the migration of chloride and nitrate due to shallow interflow and subsequent overland flow. They noted the influence of preferential lateral flow and the contribution of upslope soils to partial area saturation and overland flow. Onsite waste disposal systems (OSWDS), or septic systems, are designed to allow wastewater to slowly infiltrate soil where microbial activity is sufficient to decompose waste products. If these systems are located in a soil prone to anisotropic behavior, and in a landscape position which promotes lateral flow, rapid throughflow and even surface runoff could pose a serious health risk (Rahe et al., 1978).

Measuring Anisotropy

One laboratory tool used to measure anisotropy is the permeameter. Permeameters work reasonably well in estimating hydraulic conductivity in “undisturbed” soil cores (Freeze and Cherry, 1979). The “constant head” permeameter determines K in coarser-textured soils from the volume of water (Q) that flows through the soil core over a period of time. A constant hydraulic gradient (H) is maintained over a soil core of known length (L) and cross-sectional area (A), so that K = QL/AH.

A “falling head” permeameter uses a variable hydraulic gradient to determine “K,” and is recommended over the constant-head permeameter for fine-textured soils. I have personally used the constant head permeameter to observe hydraulic conductivities Kx, Ky, and Kz in oriented soil cores taken from a single hillsope and found that K varied over almost two orders of magnitude, with considerable variance within replicates.

Conclusion

Anisotropy needs to be accounted for in a variety of land-use decisions. It is a factor in watershed response to snowmelt and precipitation events, contaminant migration, and aquifer performance. Commonly used hyrologic models, such as the Theis method for predicting aquifer drawdowns assume homogenous and isotropic conditions. Such models should be used cautiously, with anisotropy evaluated as much as practicable.

Anisotropy is not a static quality; rather, it may change over time. Burt and Trudgill (1985) suggested soil formation and anisotropy were linked in a positive feedback relationship: infiltration causes soil horizonation, horizonation causes anisotropy, anisotropy alters the direction of soil water flow, and water flow redistributes solutes.

Reneau et al (1989) described the development of clogging mats in septic systems, which altered the hydraulic behavior of the filter fields.

Land use professionals need to be aware of such processes which naturally, or artificially alter the hydrologic character of earthen materials. Regulators need to understand things can change over time, and hydrologic evaluation need to be periodically updated.

References

Bouma, J., C. F. M. Belmas, and L.W. Dekker. 1982. Water infiltration and redistribution in a silt loam subsoil with vertical worm channels. Soil Sci. Soc. Am. J. 46:917-921.

Burt, T.P. and S.T. Trudgill. 1985. Soil properties, slope hydrology and spatial patterns of chemical denudation. Pp. 13-16 In Richards, K.S., R.R. Arnett and S. Ellis, Geomorphology and Soils. George Allen & Unwin, London.

Chorley, R.J. 1978. The hillslope hydrologic cycle. Pp. 1-42. In M.J. Kirkby (ed.) Hillslope Hydrology. John Wiley and Sons, New York.

Dabney, S.M. and H.M. Selim. 1987. Anisotropy of a fragipan soil: vertical vs. horizontal hydraulic conductivity. Soil Sci. Soc. Am. J. 51:3-6.

Hammermeister, D.P., G.F. Kling, and J.A. Vomocil. 1982a. Perched water tables on hillsides in the western Oregon: I. Some factors affecting their development and longevity. Soil Sci. Soc. Am. J. 46:819-826.

Hammermeister, D.P., G.F. Kling, and J.A. Vomocil. 1982b. Perched water tables on hillsides in western Oregon: II. Preferential downslope movement of water and ions. Soil Sci. Soc. Am. J. 46:819-826.

Rahe, T.M., C. Hagedorn, E.L. McCoy and G.F. kling. 1978. Transport of antibiotic-resistant Escherichia coli through western Oregon hillslope soils under conditions of saturated flow. J. Environ. Qual. 7:487-494.

Reneau, R.B. Jr., C. Hagedorn, and M.J. Degan. 1989. Fate and transport of biological and inorganic contaminants from onsite disposal of domestic wastewater. J. Environ. Qual. 18:135-144.

Steenhuis, T.S. and R.E. Muck. 1988. Preferred movement of nonadsorbed chemicals on wet, shallow, sloping soils. J. Environmental Quality 17:376-384.

Todd, D.K. 1980. Groundwater Hydrology. John Wiley & Sons, New York.

Whipkey, R.Z. and M.J. Kirkby. 1978. Flow within soil. Pp. 1121-1144. In M.J. Kirkby (ed.) Hillslope Hydrology. John Wiley & sons, New York.

Zaslavsky, D. and A.S. Rogowski. 1969. hydrologic and morphologic implications of anisotropy and infiltration in soil profile development. Soil Sci. Soc. Proc. 33:594-599.