| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
a College of Natural Resources, Univ. of Wisconsin, Stevens Point, WI 54481 USA
b Dep. of Soil and Crop Sciences, Texas A&M Univ., College Station, TX 77843-2474 USA
hlin{at}uwsp.edu
 |
ABSTRACT |
|---|
 TOPNOTES ABSTRACT INTRODUCTIONMaterials and methodsResults and discussionConclusionsREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMaterials and methodsResults and discussionConclusionsREFERENCES |
|---|
While qualitative use of soil morphology has been widely applied, quantification of such data is lacking. Rawls et al. (1993b) noted that a quantitative description of the effects of soil morphological properties on soil water movement is yet to be established. Quisenberry et al. (1993) also commented that while the importance of structure in water flow and solute transport has been recognized, the quantification of structure in a manner that would be useful for modeling transport is lacking.
Texture traditionally has been used to infer soil hydraulic parameters (Rawls and Brakensiek, 1983; Cosby et al., 1984; Wösten and van Genuchten, 1988). However, equations that relate hydraulic conductivity to texture alone cannot correctly predict Ksat and other hydraulic values for soils that contain large cracks, worm holes, or root channels (Campbell, 1985). Increasing evidence has shown that many clayey soils riddled with biopores or cracks, especially those having strong fine blocky or granular pedality, had Ksat values as high as or even higher than coarse-textured soils (O'Neal, 1949, 1952; McKeague et al., 1982; Coen and Wang, 1989; Bouma, 1991). Furthermore, shrinking or swelling upon drying or wetting creates dynamic hydraulic behavior in active clayey soils that are not accounted for by texture alone (Lin et al., 1998). Although pore space in a porous medium is related to particle-size distribution, it cannot be predicted solely from texture except for homogeneous, randomly packed, and nonexpansive coarse-textured materials. Childs (1969) differentiated "structural pore space" (determined by the arrangement of aggregates) from "textural pore space" (defined by the spatial distribution of primary particles). He pointed out that the tendency to correlate texture and hydraulic conductivity is "hazardous". From fractal analysis, Rawls et al. (1993a) suggested the ascending ranking of Ksat of the matrix: clay < silty clay < silty clay loam < sandy clay < sandy clay loam < clay loam < silt loam < sandy loam < loam < loamy sand < sand. Lin et al. (1997) measured near-saturated hydraulic values of more than ninety soils in their natural state and found the textural ranking to differ from the above ranking. Many of the soils Lin et al. (1997) studied contained a significant amount of large interconnected pores that were not particularly related to particle size. Rawls et al. (1993a) acknowledged the significance of large pores by making a separate calculation for macropore Ksat.
Besides texture, initial moisture state, pedality, macroporosity, and root density also influence flow and transport in field soils. The macromorphological aspects of these features are often characterized in soil surveys using qualitative descriptors (i.e., classes) (Soil Survey Division Staff, 1993); however, such descriptive representations are less useful in quantitative modeling. In a systematic investigation of field indicators for estimating soil permeability, O'Neal (1949) found that no field clues, when taken singularly, were reliable indicators of soil permeability. Each indicator must be considered with reference to the others; however, the qualitative nature of field indicators made the cross-reference difficult. Therefore, a means of quantifying soil morphology will be helpful in quantitative evaluation of the interrelationships among different morphological features and their combined effects on soil hydraulic properties.
In this study, we examined basic relationships between in situ measured steady infiltration rates and five major soil morphological features: texture, initial moisture, pedality, macroporosity, and root density, for 96 soil horizons of varying structure. A point scale system was then developed based on these relationships to convert descriptive morphological classes into numerical values with the aid of a computer optimization program. Emphasis was placed on clayey soils. We propose such a system as a means of quantifying soil structure that may be incorporated into pedotransfer functions.
|
Materials and methods |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMaterials and methodsResults and discussionConclusionsREFERENCES |
|---|
In Situ Measurements of Steady Infiltration Rates
Apparent steady-state infiltration rates (i) were measured in situ for each soil horizon using tension infiltrometers at sequential supply water potentials of -0.24, -0.12, -0.06, -0.03, -0.02, -0.01, and 0 m (i-0.24 to i0). Following the scheme described by Lin et al. (1997), measurements were made using 0.25-m-diameter infiltrometers similar to that of Perroux and White (1988). It was assumed, based on the capillary theory, that supply water potentials of -0.24, -0.12, -0.06, -0.03, -0.02, and -0.01 m exclude pores of radius or fissures of width greater than 0.063, 0.125, 0.25, 0.5, 0.75, and 1.5 mm from participating in flow. Five infiltrometers, placed 1 to 2 m apart, were used simultaneously to obtain replicated data in each horizon. The method of moments was used to estimate the means of hydraulic properties within each horizon (Parkin and Robinson, 1992). For each measurement, a level soil surface was prepared with care to avoid smearing the soil surface and blocking the macropores (Lin et al., 1997). Soil morphology within each infiltration intake area was recorded using the scheme described below prior to each infiltration run. Adjacent soil was sampled for gravimetric determination of initial soil moisture content. Soil samples were also collected for laboratory determination of particle-size distribution, organic C content, and dry bulk density. Because of uncertainties associated with extracting hydraulic conductivity from tension infiltrometer data (Lin and McInnes, 1995), we used the steady infiltration rates as quantifiers of the soils' capacity to transmit water.
Field Descriptions of Soil Morphological Features
Soil morphology of each horizon was described following standard soil survey procedures (Soil Survey Division Staff, 1993) at the time when infiltration measurements were made. A sketch of the measured soil surface was sometimes made, in addition to a photograph, to help correlate morphological features with infiltration data. Morphology descriptions were focused on texture, initial moisture state, pedality, macroporosity, and root density. Texture was determined by hand-texturing in the field, and later in the laboratory by particle-size analysis. Initial soil moisture condition was recorded in terms of saturated, wet, moist, dry, or very dry states. These five classes are a simplified version of the eight classes outlined in the Soil Survey Manual (Soil Survey Division Staff, 1993). Saturated state herein is equivalent to satiated class in the Soil Survey Manual; wet state equivalent to nonsatiated and very moist classes; moist state equivalent to moderately moist and slightly moist classes; dry state equivalent to slightly dry and moderately dry classes; and very dry equivalent to air-dried condition. The abundance and size (diameter) of plant roots were recorded using the outlines in the Soil Survey Manual (Soil Survey Division Staff, 1993). The class placement for root abundance of each root size was done based on the number of roots per unit area (Soil Survey Division Staff, 1993).
Pedality was described in terms of the grade, size, and shape of peds. We made the differentiation in using the terms of soil pedality and soil structure. Traditionally, the term soil structure has been used in U.S. soil surveys to refer to the shape, size, and grade of soil units composed of primary particles. This concept of soil structure indeed means pedality; however, pedality by itself does not provide sufficient information about porosity, such as the size of interpedal pores and the quantity and size of intrapedal and transpedal pores. Since soil porosity is critical in determining flow and transport characteristics in field soils, we chose to use the term soil structure to encompass both pedality and porosity. Consequently, interpedal, intrapedal, and transpedal pores that are not well represented by pedality were described in the porosity item.
Detailed observations of soil macroporosity are generally lacking in soil surveys, but they are recognized as important aspects of soil morphology. Special efforts were made in this study to record the size, type, and quantity of visible pores (macropores) in situ. Size was classified based on radius (for cylindrical pores) or width (for planar pores). Six classes used were: very fine (<0.5 mm), fine (0.5–1 mm), medium (1–2.5 mm), coarse (2.5–5 mm), very coarse (5–10 mm), and extremely coarse (>10 mm). Type referred to the general shape of pores and the associated potential pore continuity and connectivity. Four types considered were: vughs (small spherical or elliptical cavity), channels (cylindrical and elongated), fractures (planar), and packing voids (prolate and irregular). Quantity for each combination of pore size and type was recorded by visually comparing the soil surface to charts of various pore areal percentages. Five classes used were: very few (<0.25%), few (0.25–0.5%), common (0.5–2%), many (2–5%), and very many (>5%).
Quantification of Soil Morphology
A point scale system was developed as a means of quantifying soil morphology. A hypothetical structureless and nearly impermeable clay soil was assumed as the reference. The reference clay, assigned one point, was assumed to be massive, contain no macropores and no roots, and at fully swollen saturated state. For each morphological feature to be quantified, descriptive classes were scaled (rated) relative to the reference clay in terms of the assumed increasing capacity to transmit water vertically. Based on functional and/or empirical relationships between relevant morphology and hydraulic properties found in this study and in the literature, points were assigned to each morphological class for the amount of possible permeability increase relative to the reference clay (see details below). A computer optimization program, written in BASIC, was developed to facilitate the testing of various point assignments to different morphological classes. Points that yielded the best correlation with the measured infiltration rates were selected using a sectioning or "one-at-a-time" search method (Shoup, 1982; McCuen and Snyder, 1986). To reduce the number of variables, the optimal points of macropore quantity and ped size were assigned point scales relative to the macropore areal porosity and the maximum length of interpedal voids, respectively, without using the computer optimization program. The descriptors of pedality, macroporosity, and root density contain more than one attribute, thus the optimal points for each attribute were evaluated separately and then combined by multiplication. Combination of the attributes within a morphological feature was found to be better expressed by multiplication (reflecting interactions) than addition. Total optimal points of a morphological feature were then divided by its maximum value to obtain an index ranging from 0 to 1. This provided equal weight for each morphological feature in final comprehensive evaluations. These 0 to 1 range values were called morphometric indices.
|
Results and discussion |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMaterials and methodsResults and discussionConclusionsREFERENCES |
|---|
:
. Similar exponential relationships were also reported for Ksat and clay or sand contents by Cosby et al. (1984) and Campbell (1985). Such exponential relationships were used as the basis for rating the twelve textural classes in the USDA textural triangle. We adopted the approach of Cosby et al. (1984) to use the midpoint values of mass fractions in each textural class for calculating the ratio of the steady infiltration rate of a textual class over clay class. The ratios thus calculated from the empirical equations were used as the initial points assigned to each textural class. Further, through numerical tests with the computer optimization program, optimal points that yielded the best correlation with the measured infiltration rates at supply water potentials less than -0.03 m were selected (Table 1)
. Particle size >2 mm is not considered in textural classes. If a soil's texture had gravelly as a modifier, 20 extra points were needed.
|
|
Pedality
Through aggregation, pedality alters a soil's pore-size distribution and pore continuity and connectivity. Notably, the formation of peds creates interpedal pores that differ in size and shape from intrapedal pores. Dye-tracing studies have demonstrated that preferential flow can occur along interpedal pores in structured soils (Ritchie et al., 1972; van Stiphout et al., 1987; Lin et al., 1996). Conceptually, we would expect the following trend of increasing capacity to transmit water vertically among different shapes of peds: (massive) < platy < blocky, prismatic < granular. (Columnar ped shape was not encountered in this study, thus not considered herein). Higher Ksat has been reported for prismatic soil than blocky silty clay loam (Bouma and Anderson, 1977a, 1977b). Our study showed that, depending on texture and initial moisture, soils with prismatic peds had either higher or lower i's than similar soils with blocky peds (Table 2)
. For fine-textured soils (clay and clay loam), prismatic pedality had higher flow rates than blocky pedality because of common occurrence of interprism macropores; whereas in medium-textured soils (sandy clay loam and silt loam), interprism pores were less significant in soils with prismatic pedality, leading to lower water flow rates than that in blocky soils. Because of such varying situations, similar points were assigned to prismatic and blocky ped shapes in our study. The hydraulic distinctions between these two ped shapes were left to the associated difference in macroporosity and texture. Numerical tests using the computer program showed that the points assigned to ped shapes shown in Table 1 yielded optimal correlation with the measured steady infiltration rates.
|
The stronger the ped grade (i.e., the distinctness or strength of ped units), the more significant the interpedal pores. Because a quantitative relationship between ped grade and soil hydraulic properties is not available, we used the computer program to numerically select the optimal points for each ped grade following the order: weak < moderate < strong (Table 1). Single grain is not considered as a ped grade from a pedological point of view; we listed it as a very strong class under ped grade in order to compare sands with other textural soils.
In soils with compound pedality, in which large units were composed of smaller units separated by planes of weakness, points were assigned to both secondary and primary ped units. To obtain an optimal weighing of secondary and primary pedality for their combined impact on the steady infiltration rates, we tested different weighting approaches and found that using only the points of secondary pedality in a soil with both secondary and primary pedality yielded optimal correlation with the measured infiltration rates. The addition of primary pedality points to secondary pedality significantly decreased the correlation with macropore flow rate, and only slightly enhanced the correlation with micropore flow rate. This suggests that primary pedality might have limited impact on infiltration at a water potential of greater than -0.03 m in a soil with compound pedality. Quisenberry et al. (1993) and Nortcliff et al. (1994) also reported that in a soil with both primary and secondary peds, preferential flow occurred mostly along secondary ped faces.
Macroporosity
Macroporosity showed a decreasing trend of correlation with the steady infiltration rates as supply potential decreased (Fig. 1), reflecting the physical exclusion of macropores from flow. In comparison, the correlation between total soil porosity determined from clods in laboratory and the steady infiltration rates was rather weak. Quantity of macropores was directly assigned a point scale relative to the macropore areal porosity, with a base point of one for "very few" class (Table 1).
The ratio of hydraulic conductivities of two soils that have similar pore spaces but with pores that differ in size is Nb, where N is the ratio in linear dimension of a larger body over a smaller body and b = 2 (Childs, 1969). Stakman (1969) reported a b = 1.93 of sand separates, with mean pore radius ranging from 0.005 to 0.5 mm (i.e., very fine size class). In a study done by Wang et al. (1985), two soil horizons with similar texture and pedality but different sizes of very fine interpedal planar voids — one with 0.2 to 0.5 mm in width and another <0.2 mm — had the difference in measured vertical Ksat of the order N2.8 to N3.68, where N was 3.5. The somewhat higher b in this case might be related to different quantities of planar voids. On the other hand, a numerical study done by Edwards et al. (1979) showed that a worm hole with a radius of 5 mm (very coarse size class) had only N0.8 higher cumulative infiltration than a similar worm hole with a radius of 2.5 mm (coarse size class), where N = 2. Numerical tests with our computer program indicated that using N2 to N3 for scaling the points of pore size classes less than or equal to "medium" and using N0.8 to N1 for pore size classes greater than or equal to "coarse" gave better correlation with the measured steady infiltration rates than using N2 across all macropore size classes. This is reflected in the optimal points obtained for macropore size classes shown in Table 1. The reduced impact on water flow with macropore size class greater than or equal to "coarse" in an initially unsaturated soil can be explained by film flow along the walls of large macropores rather than the flow filling up entire macropore space (Bouma and Dekker, 1978; Philips et al., 1989).
By definition, the general trend of increasing capacity to transmit water among the four types of macropores is: vugh < channel < fracture < packing void. Such a trend is supported by dye-tracing studies of Bouma et al. (1977) and Lin et al. (1996). Both found a similar order, as evidenced by dye-stained areas in soils with various types of macropores. Once the order of macropore types in their capacity to transmit water was identified, optimal points for each macropore type were selected using the computer optimization program (Table 1).
Root Density
Several studies showed that preferential water flow occurred along living roots of trees (Aubertin, 1971), corns (Zea mays L.) (Warner and Young, 1991), and grasses (Lin, 1995). Through thin section, FitzPatrick (1993) showed that active pore space around plant roots was generally <1 mm. We took into account the effect of living roots on water flow by estimating the root density in the soils. The pore areal fraction at the root–soil interface (
root) was estimated based on the number and size of roots in a unit area using the formula:
 | (1) |
root thus estimated for the 96 soil horizons showed a moderate degree of correlation with i-0.06 to i-0.24, but not with i0 to i-0.03 (Fig. 1). Consequently, the average ratio of root between root size classes was used as the basis for assigning initial points to each root size class, and the average number of roots per unit area was used to assign initial points to each root abundance class. Through the tests with the optimization program, final optimal points were selected (Table 1).
Interrelationships among Morphometric Indices
Assigning numerical values to morphological classes permits the determination of interrelationships among different morphological features. As shown in Table 3
, the morphometric index of macroporosity (MIp) had significant positive correlation with both the morphometric indices of initial moisture state (MIm) and pedality (MIs), but not with texture (MIt), indicating that the drier the soil or the stronger the pedality, the higher the macroporosity for the soils studied. Although not statistically significant, the negative correlation coefficients between the MIp and MIt and between the MIs and MIt appear to suggest that the fine-textured soils tended to have higher macroporosity and stronger pedality than the coarse-textured soils in the 96 soil horizons investigated. The MIs also showed positive correlation with the morphometric index of root density (MIr), implying that the more roots in the soil, the stronger the pedality. The MIr also had a significant positive correlation with the MIt, suggesting that plant roots tended to grow more easily in sandy soils (higher MIt) than in clayey soils.
|
|
Conclusions |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMaterials and methodsResults and discussionConclusionsREFERENCES |
|---|
|
ACKNOWLEDGMENTS |
|---|
|
NOTES |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMaterials and methodsResults and discussionConclusionsREFERENCES |
|---|
Received for publication April 17, 1998.
|
REFERENCES |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMaterials and methodsResults and discussionConclusionsREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  A. Lilly, A. Nemes, W. J. Rawls, and Ya. A. Pachepsky Probabilistic Approach to the Identification of Input Variables to Estimate Hydraulic Conductivity Soil Sci. Soc. Am. J., January 11, 2008; 72(1): 16 - 24. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. Alletto, Y. Coquet, P. Vachier, and C. Labat Hydraulic Conductivity, Immobile Water Content, and Exchange Coefficient in Three Soil Profiles Soil Sci. Soc. Am. J., June 21, 2006; 70(4): 1272 - 1280. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. A. Winfield, J. R. Nimmo, J. A. Izbicki, and P. M. Martin Resolving Structural Influences on Water-Retention Properties of Alluvial Deposits Vadose Zone J., May 26, 2006; 5(2): 706 - 719. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Eynard, T. E. Schumacher, M. J. Lindstrom, and D. D. Malo Porosity and Pore-Size Distribution in Cultivated Ustolls and Usterts Soil Sci. Soc. Am. J., November 1, 2004; 68(6): 1927 - 1934. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. H. Young, E. V. McDonald, T. G. Caldwell, S. G. Benner, and D. G. Meadows Hydraulic Properties of a Desert Soil Chronosequence in the Mojave Desert, USA Vadose Zone J., August 1, 2004; 3(3): 956 - 963. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Eynard, T. E. Schumacher, M. J. Lindstrom, and D. D. Malo Aggregate Sizes and Stability in Cultivated South Dakota Prairie Ustolls and Usterts Soil Sci. Soc. Am. J., July 1, 2004; 68(4): 1360 - 1365. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. Lin and H. Lin Hydropedology: Bridging Disciplines, Scales, and Data Vadose Zone J., February 1, 2003; 2(1): 1 - 11. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
W. J. Rawls and Ya. A. Pachepsky Soil Consistence and Structure as Predictors of Water Retention Soil Sci. Soc. Am. J., July 1, 2002; 66(4): 1115 - 1126. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
F.G. Calhoun, N.E. Smeck, B.L. Slater, J.M. Bigham, and G.F. Hall Predicting bulk density of Ohio Soils from Morphology, Genetic Principles, and Laboratory Characterization Data Soil Sci. Soc. Am. J., May 1, 2001; 65(3): 811 - 819. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H.S. Lin, K.J. McInnes, L.P. Wilding, and C.T. Hallmark Effects of Soil Morphology on Hydraulic Properties: II. Hydraulic Pedotransfer Functions Soil Sci. Soc. Am. J., July 1, 1999; 63(4): 955 - 961. [Abstract] [Full Text]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| The SCI Journals | Agronomy Journal | Crop Science | |||
| Journal of Natural Resources and Life Sciences Education |
Vadose Zone Journal | ||||
| Journal of Plant Registrations | Journal of Environmental Quality |
The Plant Genome | |||
