Structure and pore space
Although the mineralogical nature of the clay fraction and the textural classification of the soil are of basic importance to soil physical properties, other factors often have even greater importance. Decomposed organic matter, or humus, and certain inorganic compounds, such as iron and aluminum oxides, form coatings on soil particles, and these, along with some clays, bind soil particles together into granules or aggregates. The size and distribution of such aggregates are major factors in determining pore size and distribution in soil and are important in the physical behavior of soils. Additionally, fragmented organic materials, undecomposed or only partially so, if present in soil in significant quantity, also can play a major role in determining physical properties. The manner in which mineral soil particles are assembled and maintained in aggregate form, together with quantity, size, and distribution of fragments of partially decomposed organic materials, is referred to as soil structure. Soil structure may develop as a consequence of natural processes such as wetting and drying, freezing and thawing, transport of minerals in moving water and their deposition, or mechanical forces exerted by plant roots or by other biological agents. Natural structure also may be modified by plowing, cultivation, and mechanical forces associated with cultural activities.
There are two aspects to the evaluation of soil structure which have major practical importance: measurements of pore size and configuration, and assessments of the stability of these characteristics against various natural and cultural forces which act to change the physical arrangement of primary particles and secondary aggregates. Pore size and configuration determine water retention and water and air transport properties of a soil and, in some cases, the ease with which plant roots penetrate and living organisms can move. The equivalent pore size of the largest water-filled pore in a soil sample (that is, the radius of a cylindrical pore which would behave with respect to water retention similarly to the largest water-filled soil pore) may be measured by using the hanging water column apparatus (Fig. 19), and the equation for capillary rise, r = 2s/Dgh, where s is the surface tension of water, D its density, g the acceleration of gravity, and h the height of rise of water in a capillary tube of radius r.
Fig. 19 Measurement of equivalent pore size. (a) Porous plate apparatus. (b) Schematic representations of the air-water interfaces and (c) the particle-water interfaces.
The porous plate apparatus for relating water-filled pore size to the force of removal exerted by the hanging water column (Fig. 19a) is based upon a capillary tube model where soil pores are presumed to be equivalent to small sections of capillary tubes (Fig. 19b). Pores in the porous plate are all considerably smaller than the size of the capillary tube (Fig. 19a), so that they remain water-filled as h is varied. More typically, the shape of air-water interfaces (Fig. 19c) has a radius of curvature, R, given as a combination of a positive radius r1 measured axially and a negative radius r2 at right angles, or 1/R = (1/r1) + (1/r2). A plot of water content of a soil sample in the hanging column apparatus against the height of the hanging water column characterizes the pore size distribution of the sample. A large change in water content with a small change in hanging water column length indicates a large volume of pores of similar size, whereas a small change in water content with a large change in hanging water column length indicates a relatively small volume associated with pores over a wide size range. More direct evidence of pore size and arrangement may be obtained by microscopic examination.
The second factor which is basic to determining the structural state of soil has to do with soil stability against disruptive forces. The existence of binding forces between clay particles themselves and between clay and other mineral surfaces, together with colloidal organic decomposition products, polysaccharides and polyuronides, and inorganic cementing substances, determines how well a particular geometrical organization of soil particles resists change against disruptive forces. Hence, structural evaluation usually involves both geometrical properties and some indication of how stable a particular arrangement of particles will be against the disruptive forces of a plow, falling raindrops, the weight of an animal, or some other force. Although the nature of the pore space, by implication at least, is a part of any characterization of soil structure, often the measurement involves application of a particular disruptive force and noting the degree to which an existing arrangement is destroyed.
An important property which can be used to characterize the structural state of a soil is the bulk density. Where the particle density of soil materials is uniform (often taken as 2.65 g/cm3), it is possible to determine the total pore space fraction of a soil from measurement of the bulk density (mass of dry soil per unit bulk volume) and use of the formula, pore space fraction = 1 − (Db/Dp), where Db and Dp are the bulk and particle densities. The presence of organic materials having particle densities differing appreciably from those of the mineral particles introduces some error into such computations. Agricultural soils in the plow depth ordinarily have bulk densities ranging from 1.0 to 1.5 g/cm3 and a pore-space fraction from 0.4 to 0.6, or roughly half particles and half pore space.
From some perspectives the physical properties of soil may be presented adequately by a description of the properties of a sample of surface soil. However, horizonation, whether produced by natural or cultural processes, may profoundly affect water flow and retention near the surface or within the rooting zone of plants. Such influence is particularly noticeable at boundaries between materials having different pore sizes.
Horizonation may be the result of the method of original deposition, for example, alluvial or loess derived profiles, or it may develop over time as a consequence of weathering processes which differ near the soil surface from those at depths beneath. Root growth and development and organic matter decomposition near the surface may promote both the development of aggregates and their stabilization. Also, fine soil particles, silts and clays, may be displaced downward through a profile by moving water, accumulating in layers below. Solutes carried in water also may dissolve minerals, carrying them in the direction of water flow and depositing them as precipitates or leaving them behind as water evaporates.
Horizonation also may occur in soil as a consequence of plowing and cultivation operations. It is not unusual for a plow pan to develop in cultivated soils (Fig. 20) as a consequence of the smearing action of an implement surface as it slides through the soil. Decreased porosity at an interface between disturbed and undisturbed soil can profoundly affect water flow and bring about both filtering of fine particles, which may be carried in moving water, and deposition of substances by precipitation. Also, soil may become severely compacted in the surface by vehicle and animal traffic, with a consequent decrease in porosity and increase in soil hardness.
Fig. 20 Interface between disturbed soil and compact soil produced by the smearing action of a tillage implement as shown by a scanning electron micrograph. Water flow across such an interface is greatly reduced.
Forces acting on soil water
Flow of water in soil depends upon forces existing within the soil matrix and upon forces associated with gravity acting upon the water itself. In saturated soil the force of gravity acting upon water within the soil and by standing water upon the soil surface constitutes the moving force, and it can cause flow in any direction depending upon the geometry of the system. In unsaturated soil, gravity still acts upon the water but, except in very wet soil, this force may be small compared to forces due to the attraction of solid surfaces for water. These are the same forces that cause water to rise in a capillary tube, and they are sometimes referred to as capillary forces. In soil, such forces are called matric forces. In addition to matric forces, three other forces may be present in a soil system: osmotic forces, relating to the attraction between solutes and water; local gas pressures, which may exert a force upon air-water interfaces within soil pores; and overburden forces, which may arise where substances dissolved or suspended in water add to its density and increase its weight. Overburden forces may exist also in saturated soil, but usually such forces are neglected under both saturated and unsaturated conditions.
The nature of the various forces acting on water in soil differ appreciably, and the direction in which each acts may be highly complicated. Hence, it is difficult to apply them quantitatively to soil water. To circumvent this difficulty, the contribution of the various forces to the energy state of soil water is considered, thus permitting the addition of the contributions of each type of force. This involves measuring the work required to remove a unit quantity of water, and is known as the potential. Such work depends upon the degree of wetness, so that the energy state is not a linear function of water content. The units of measurement are energy content per unit of liquid volume of water. The commonly used units for potential are the bar, the millibar, and centimeter of water. Above the water table, work is required to remove water, and by convention, potential is negative. The potential at a flat or free air–water interface is taken to be zero.
Practical problems of water flow in unsaturated soil involve mostly the matric and gravity potentials. For flow into roots, because they are semipermeable membranes, additional work must be done by a plant to remove water from those ions which do not readily pass through root membranes. Thus, in dealing with flow of water in soil-plant systems, osmotic potentials also must be considered. However, gravity potential often is small in unsaturated soil compared to matric and osmotic potentials, and sometimes may be neglected. Matric potential is a function of water content, pore size, and pore size distribution in the soil. As with capillary tubes, small pores hold water tightly, and considerable work must be done to remove it. The capillary rise equation may be used to relate the size of the largest water-filled pores (the size of a cylindrical pore with similar water-retentive properties) to the force required to remove water from wet soil, h = 2s/dgr as described earlier. Such air-water interfaces in small pores are essentially hemispherical so that this radius also approximates the spherical equivalent of the irregularly shaped pore.
The capillary rise equation can be applied to a porous soil system in the wet range (Fig. 19). Pores in the porous plate are all small enough to remain water-filled at the elevation of the hanging water column (smaller than the radius of a capillary tube which would raise water to this elevation). The radii of the air-water interfaces in the soil are given by the capillary rise equation. If the elevation h is decreased, larger pores become water-filled; if h is increased, only smaller pores remain water-filled. The elevation is an index of the matric potential, and the potential on a mass basis may be obtained by multiplying by the acceleration of gravity: hg (or if potential is on a volume basis, the value is hDg, where D is the density of water). The pressure in the water just below the interface as given by the equation p = 2s/r, and if the radius is taken as negative (the radius of a raindrop is positive and the pressure inside also is positive), the pressure may be seen to be negative or less than the pressure of the atmosphere. The use of the hanging water column apparatus (Fig. 19) is confined to the very wet range because of the limitation in the length of a hanging water column due to possible cavitation. However, positive pressures applied in a pressure chamber over a porous plate upon which samples are placed achieve the same result without any serious limitation to the range of measurements which can be made.
Matric potential in wet soil equals the work done per unit mass or volume to remove water from the air-water interface or against surface tension forces. As the soil becomes drier, larger and larger proportions of the water are associated with particle surfaces, and more work per unit quantity must be done to remove this water, which is tightly held by adsorptive forces in the particle surface. Hence, matric potential may be seen to involve both surface tension and the attraction of particle surfaces for water. Curves showing this relationship over a wide range of water content are obtained experimentally and used to characterize different soils (Fig. 21). However, since it is the porosity (a characteristic dependent not only upon textural classification but also upon the degree and kind of aggregation) that defines the relationship, different curves can exist for the same basic soil material treated differently.
Fig. 21 Relationship between water potential (index of retentive forces) and water content for different soil materials. Wide variations often exist, even for materials in the same textural class. 1 bar = 102 kPa; 1 cm3 = 0.6 in.3
The water potential–water content curves (Fig. 21) are desorption curves obtained in the process of removing water from saturated soil. A slightly different curve would exist for the wetting cycle because filling and emptying of large pores is controlled by the size of entryways rather than the size of the pore itself. Air is easily trapped in large pores during filling, and water is retained in large pores during drying, until entryways are emptied. This hysteresis phenomenon sometimes complicates use of water potential–water content curves, since it is impossible to determine one from the other without knowledge of the wetting history.
Water movement and retention
Water moves in soil in response to the net force acting. Above the water table in the region of a soil profile primarily involved with growing plants, matric forces are the major forces involved. Only near the water table where soil is very wet do gravitational forces play an important role. Where matric forces dominate, the water flow (flux) is the product of a conductivity factor and the moving force, which is the gradient of the matric potential. The conductivity factor is expressed as the product of a dimensionless number between 0 and 1 which expresses the degree to which the channel is filled with water and the value of the conductivity for saturated flow. This relationship describes saturated flow when the channels are full and where the matric potential is replaced by an equivalent potential derived from gravity and positive pressure, which is associated with saturation.
The channel-filling factor falls off rapidly as large pores empty, so that the cross section available for liquid flow is greatly reduced. As the soil desaturates, the pressure potential disappears and is replaced by a negative matric potential term. Matric forces are “pulling” rather than “pushing” forces, and water is pulled from regions of high potential into regions of low potential. Liquid water flows through water-filled interstitial space and along surfaces. The air-filled pores contribute nothing to liquid flow. Under such circumstances it may be seen that coarse materials, such as gravels, which would have high conductivity when saturated, would have extremely low conductivity at low water content. Thus water movement in unsaturated soil may be retarded in regions of large pores as well as in regions with fine pores. Aside from the influence of air-filled pores in reducing conductivity, conductivity also is reduced, beyond the amount expected from reduction of flow cross section, as pore size is reduced. The conductivity of capillary tubes varies with the fourth power of the capillary radius (Hagen-Poiseuille equation). On a unit area basis, the variation is with the square of the radius. Hence, with the force term held constant, reducing pore radius by a factor of 2 would reduce flow by a factor of 22, or 4.
Movement of water into soil, infiltration, and redistribution of water in soil following water addition slow down with time as a consequence of both changing gradients and reduction of unsaturated hydraulic conductivity as water moves from wet zones into drier zones. Slow redistribution is of particular importance to consideration of water retention in soil. Water is nearly always moving, downward into dry soil or toward a water table, and upward in response to evaporative forces at the soil surface. Hence, soil has no unique retentive capacity. Water-retentive capacity is a dynamic property and must be defined in the context of change with time. A practical field capacity may be specified as the quantity of water in a defined depth of soil when rate of downward flow, or loss from a designated part of the profile, reduces to a value beyond which any further loss may be regarded as negligible. For agricultural purposes, and depending upon the nature of the profile, this rate would be reached in a period of time ranging from a fraction of a day to 10 or more days. The presence of soil horizons, involving either coarse or fine layers, reduces liquid water flow rates and increases water retention, generally reducing the time required to reach a negligible loss value.
An important application of unsaturated flow of water in soil concerns contamination of soil by radioactive materials or by other chemical or biological substances in spills. The seriousness of contamination depends upon the amount of water that is applied to soil as precipitation or through artificial means. From an agricultural point of view, slow movement of liquid water deep in the profile below the normal depth of roots may be of little consequence. However, when soil contains contaminants, particularly long-half-life isotopes, such flow could become important. Under some circumstances, the presence of layers of sand or gravel might constitute an effective barrier to the spread of some contaminants, but not when long periods might allow diffusion of substances in extremely thin films, which might under some conditions be present on particle surfaces or in vapor form.
Water vapor is present in soil air with relative humidity ranging from 98.9 to 100%, existing in soil wet enough to support plant growth. At uniform temperature the sum of matric and osmotic potentials, which affect evaporation, may be equated to relative humidity by the equation: matric + osmotic potential = (RT/M) ln (p/p0), where R is the universal gas constant, T the Kelvin temperature, M the molecular weight of water, and p/p0 the vapor pressure divided by the vapor pressure of saturated air at the same temperature, or the relative humidity. Where gradients of matric plus osmotic potential exist in soil, vapor pressure gradients also exist, and vapor flow can take place, even when liquid flow is negligibly small. If temperature gradients also exist, appreciable vapor flow can occur because of large vapor density gradients. However, in the absence of temperature gradients, vapor flow over large distances is small in dry soil. Because of low hydraulic conductivity values in dry soil, evaporative water loss from below depths of 25–50 cm (10–20 in.), in the absence of plants, generally is negligibly small, and moist soil, even in deserts, may be found below such depths. Flow of water in dry soil is primarily in vapor form, and the insulation properties of dry soil are high, so that high temperature gradients required for rapid diffusion of vapor do not exist. Where deep-rooted plants, particularly trees, are present without a water source, soil can be dried to great depths through the process of plant water uptake and evapotranspiration. Plant uptake has been used to remove unwanted chemical substances from soil and, in some coastal reclamation projects, to dry soil.
Air composition and flow
Gaseous composition of soil air, apart from water vapor discussed above, tends toward the composition of the atmosphere, with somewhat higher concentrations of carbon dioxide and lower concentrations of oxygen due to metabolic processes which utilize oxygen and give off carbon dioxide. Exchange with the atmosphere is most rapid in dry soil having high porosity and least rapid in dense or wet soil. The composition of the soil air affects plant growth and the growth of microorganisms and insects which inhabit the soil. Hence, limited air exchange with the atmosphere often has a marked effect on plant disease and plant growth generally. Oxygen requirements of plant roots vary with different plants, so that plant composition often is dependent upon aeration characteristics of soils.
Physical conditions of soils affect solute movement. Water is necessary in the vicinity of roots to solubilize nutrients contained in clays and other minerals and to form liquid paths through which the nutrients may diffuse and become available for absorption by roots. Also, solutes may be carried in water taken in through roots to form the transpirational stream. See also: Plant mineral nutrition
Solutes may be carried with moving water downward out of the rooting zone of plants and upward in moving water due to evaporation at the soil surface. In the latter case, transport to the surface and deposition through evaporation of pure water constitute a concentration process which may lead to high salt concentration in and on the soil surface. See also: Root (botany)
Single-grain soil materials like dry sands have only a limited capability to withstand stress or compressive forces when unconfined. With small amounts of water added, this capability is increased because of surface tension forces in air–water interfaces which tend to hold particles together. Elimination of air–water interfaces by further additions of water, together with lubrication of particle surfaces, reduces the ability to withstand stress. Aggregated soil behaves somewhat in the same fashion except that stabilizing materials such as humus tend to bind particles into aggregates. Soil, particularly when moist, has some tensile strength. However, tensile strength is limited, as may be seen with swelling soils which may shrink and form large cracks upon drying. Compressive strength varies with degree of compaction brought about by animal or vehicle traffic, particularly at critical water contents somewhere between wet and dry. The water content for maximum compaction depends upon textural characteristics and types of clay minerals present. In a like manner an optimum water content generally exists for maximum effectiveness of tillage operations designed to fragment the soil and reduce hardness.
Mechanical strength is an important consideration in agriculture as well as construction engineering. The agriculturalist needs soil which is strong enough to support small plants and trees, yet soft enough to permit easy movement of roots. Additionally, the soil should be sufficiently resistant to the mechanical forces of rain and running water to hold its position. The engineer needs soil with minimum shrink and swell properties and maximum strength over a wide range of water contents. See also: Soil mechanics
Soil temperature depends upon absorption of solar radiation, reradiation from the surface, conductive exchange with the air, heat flow within, and the heat capacity of the soil. Soil color and surface texture influence both absorption and reradiation. Smooth, light-colored materials generally reflect light energy and are poor radiators, while rough, dark materials absorb or reradiate energy best. Thus, rough, dark soils tend to warm faster than smooth, light-colored materials. Organic residues on the soil surface play a major role in determining soil temperature through interception of both incoming and outgoing radiation and reduction in the velocity of air movement at the soil surface. Water content is the major factor involved in both heat transfer and heat retention, increasing both thermal conductivity and thermal capacity. Change of state of soil water—freezing, thawing, and evaporating—involves significant quantities of energy as latent heat. Soil temperature at a given depth and time depends on both heat conductance and storage, and complicated mathematical models are required for its prediction.
Plant growth and biological activity
Growth of plants, microorganisms, and insects in soil all involve establishment of an optimum physical environment. In turn, the presence in soil of organic materials and their decomposition products profoundly affects these physical properties. Broken or partially decomposed organic materials behave somewhat the same as mineral particles, and the decomposed material or humus acts as a binding substance imparting stability to mineral particle arrangements of soil aggregates. From a soil physical point of view, the major factors affecting biological activity are water, temperature, and gas composition. With some exceptions, soil supporting growing plants must have a matric-plus-osmotic potential ranging from about −15 bars (−1.5 megapascals) up to a small fraction of −1 bar (−0.1 MPa). If the soil is too wet, aeration becomes limiting. But, because aeration depends upon the quantity of air–filled pores, which will be different in different soils, it is difficult to specify a definite water potential value which will be limiting. Also, numerous agricultural plants grow best when the water potential is above −1 bar, even though plants might survive at potentials even below −15 bars.
Soils having mostly fine pores tend to remain wet in the spring and to warm slowly, so that many plants are delayed in growing because of low temperature. Optimum temperature for plants varies widely with species, and temperature thresholds exist for many plants; for example, corn and tomatoes grow poorly or not at all until day-time temperatures are well above 59°F (15°C), and growth is reduced if soil temperature greatly exceeds 86°F (30°C). However, many plants do well in cold soil.
Mechanical properties of soil can restrict root growth. Hard pans or plow pans formed by tillage operations often limit the growth of all but the hardiest of plants. Such pans may reduce the depth of the rooting zone. The hardness of the soil also is a factor in plowing and tillage operations. Since soil water content affects hardness, certain tillage operations are timed so as to take advantage of a water content that produces softer soil. If too wet, however, a soil may puddle badly when tilled. Thus an optimum water content for such operations exists. Appreciable attention is being given to no-till or minimum tillage to avoid creating hard soil by excessive traffic compaction and to maintain the protection of plant cover against erosion. Weed control, one of the important functions of tillage, is done chemically when these practices are followed. See also: Herbicide; Plant growth
Soil properties vary vertically at a given site, horizontally from place to place, and both vertically and horizontally with time. Determination of the spatial variability of soil can be accomplished by kriging, a geostatistical technique that was developed to aid mining engineers make better estimates of ore deposits in unsampled locations. Geostatistics is the statistics of spatially correlated data; since soil properties are generally correlated spatially, geostatistics and kriging have become popular methods for describing and dealing with this complex aspect of soils. Kriging is a technique of making optimal, unbiased estimates of soil properties at unsampled locations.
Describing spatial variability of soil is important only insofar as it enables users of land resources to make better predictions of soil behavior and performance. The precision with which soil behavior and performance can be predicted depends on the uniformity of the land in question. While it may be possible to use an average value to estimate soil behavior or performance, it is often not very useful and sometimes dangerous to depend on averages, especially if different parts of a land parcel behave very differently. For example, a home owner who plans to install a septic tank needs to know the hydraulic characteristics of the plot and not the average value for the neighborhood. Since the performance of each plot as a septic tank field depends on the hydraulic behavior of the field, there is economic value in any method that enables a user to estimate the local value with greater precision than the regional mean. Kriging achieves this aim by using observed values in the neighborhood to estimate values at unsampled locations.
If all observed values are identical, the average of the observed values would be an excellent estimator of the same property at any point in the region. The problem arises when the variance in the observed data is high, and the user is unwilling to accept the mean as an estimator of values at unsampled locations. In such cases, kriging can provide better results, provided certain conditions are met.
Kriging depends on the fact that closely spaced samples tend to be more alike than samples separated by large distances. The difference in clay content between soil samples taken 1 m apart, for example, more likely will be smaller than that of the samples taken 1 km apart. In kriging, the clay content is expected to vary from point to point, but the difference in clay content between two points is expected to vary only with separation distance. Very often, however, the difference varies not only with separation distance but with orientation as well.
Considering the simpler case, in which the difference varies only with separation distance, that information can be captured and condensed in a statistical representation known as a semivariogram: each point on such a curve (Fig. 22) represents the sum of squared differences for a particular separation distance divided by twice the number of pairs. The semivariogram is calculated from Eq. (1),
where γ(h) is the semivariogram [2γ(h) is the variogram], n is the number of pairs, z is the soil variable at location x, and h is the separation distance. In the “ideal” semivariogram, γ(h) = 0 when h = 0; that is, samples separated by zero distance are identical. In practice, the curve frequently intersects the vertical axis at a value above zero. The point of intersection is called the nugget variance, a term that links geostatistics to its early beginning in the gold mining industry. Another characteristic of the semivariogram is that it levels off to a value called the sill. It turns out that in the ideal case the sill corresponds to the ordinary sample variance. The sill, therefore, corresponds to variances associated with samples that are spatially independent. The maximum distance at which sample pairs are correlated is called the range. For sample distances shorter than the range, the sample property is said to be spatially dependent. Note that if by chance the sampling distance exceeds the range, one obtains a semivariogram in which the nugget variance and sill are the same. For purposes of kriging, separation distances within the range of spatial dependence are required. Choosing the minimum sampling distance so that it falls well within the range requires knowledge of the study area. Trends in the data and anisotropy are frequently encountered. Trends are regular changes in a value, such as soil temperature decreasing with elevation. Special techniques are required to eliminate the trend from the data. Anisotropy is encountered when the difference between sample values depends not only on separation distance but on orientation as well.
Fig. 22 Semivariogram of sand content showing a range of spatial depedence of about 9 mi (15 km). Data points represent the experimental values, and the curve is an exponential model.
In order to use the structure in the variance for kriging, the experimental semivariogram is fitted to a model semivariogram. The three most commonly used models are the spherical, exponential, and linear models: the exponential model (Fig. 23) is fitted to the experimental semivariogram.
Fig. 23 Locations of 268 points for kriging sand content. The points are 1 km (0.6 mi) apart.
In kriging, local values are estimated from the weighted average of observed values within the range of the point to be estimated. The estimated value of the variable z at location x0 is given by Eq. (2),
where n is the number of observed values z(xi), and λi are weights applied to each z(xi). The weights are calculated so that the estimate is unbiased and the estimation variance is minimized. The estimation involves solving a set of n + 1 linear equations with n + 1 unknowns that describe the expected autocorrelation between observed values and the value to be estimated.
For purposes of mapping, the points to be estimated can be located in a square grid (Fig. 23). The estimated values and their corresponding estimation variance can then be used to produce isarithm maps (Fig. 24). Areas with low sampling densities will show high estimation variances, so that the estimation variance map can be used to guide subsequent sampling to improve map quality. If the variable in question is difficult or costly to measure, it is easier and cheaper to measure another variable that correlates highly with the variable of interest, and then compute linear estimates through the technique of cokriging. Local estimates of undersampled variables can also be improved through cokriging. Cokriging is, therefore, an interpolation technique that increases the accuracy of an estimated property by using additional information provided by the structure in the variance of a covarying property. Cokriging can be especially useful in soil science because most soil properties are spatially correlated and intercorrelated.
Fig. 24 Isarithm maps. (a) Map of actual sand content; values at the isarithms are in percent. (b) Map of the estimation variance of percent sand content; values at the isarithms are in percent squared.
It is often more practical to estimate the local value of a block of land than a point in the landscape. This procedure, known as block kriging, provides a way to estimate an average value for an area of land. The area may be the size of a experimental plot, septic tank field, or a corn field. The difference between point kriging and block kriging is in the calculation of the weighing coefficients. In block kriging, the semivariances between observed points and the estimated values of point kriging are replaced by the average semivariances between the observed points and all points in the block. As in point kriging, the weights are calculated so that the estimation variance is minimized.
The ability to determine the property of a soil at any point or block in the landscape provides users of land resources with information that permits sounder economic and safer environmental land-use decisions to be made. Spatial variability can be costly to farmers and other users of land resources. Farmers who apply agricultural chemicals uniformly in a field with spatially varying soils will add too much to places where the requirement for the chemical is low, and too little where the requirement is high. Farmers who cultivate large tracts of land are finding it profitable to invest in farm machinery that can match application rates to spatially varying soil and crop requirements with a high degree of precision. While economic considerations serve as the main incentive, stricter environmental regulations may compel farmers to adopt precision farming, and kriging offers a way to predict optimum application rates of agricultural chemicals.
Soil management can be defined simply as the manner in which people use soils to produce food, fiber, and forages. Soil management includes determination and use of many factors and practices, such as land survey maps, cropping systems, organic matter and tilth, soil fertility, salinity, and irrigation. Most often, it is not a single soil that is managed; management is carried out on a field or a portion of a landscape composed of a number of physical and biological features, like climate, vegetation, topography, and drainage. In a true sense, applying soil management strategies involves an integration of a number of factors. Such integration involves a difficult synthesis of many individual characteristics, both measured and observed, to arrive at a meaningful interpretation of how a soil responds to management. In fact, the value of any agricultural soil is determined by how well that soil responds to proper management.
Meaningful assessment of how soils are responding to a particular management system can be judged only if standards are established. There are two primary standards for judging quality of management: prevention of soil deterioration or degradation, and improving the soil system in ways that result in increased plant production. Examples of deterioration that must be prevented are excessive erosion, fertility depletion, and accumulations of salt within the rooting zone of plants. Examples relating to improving the soil system are conservation of precipitation, irrigation development, reclaiming salt-damaged land, and increased water use efficiency by proper fertilization. See also: Agricultural soil and crop practices
The two goals are very broad and therefore require the integration of many aspects of science. Incorporation of the basic areas of chemistry, physics, and microbiology into management planning is essential. The sciences of ecology, genetics, and various types of engineering are also essential to modern soil management. Computer modeling will play an increasingly important role in soil management by making simulation models directly available to farm and ranch operators. See also: Simulation
Land capability classification
The first step in planning a management system is assessment of the soil and environmental resources. This is accomplished through land classification—more specifically, through land capability classification, which groups land according to properties essential for identifying the opportunities or constraints the land offers for various uses.
Interpretations of land capability classifications are most often based on physical, chemical, and economic considerations. Classification provides assessment of land suitability for agricultural and other uses, identifies adapted crops, estimates yields of crops under defined systems of management, indicates presence of specific soil management problems, and delineates opportunities and limitations of various management practices. Through land capability classification, there usually are identified several alternative management strategies for a given type of land. From this information, analyses can be made to determine which land use or management strategy will be most desirable for both economic and wise land utilization.
The basic requirements for developing a good classification system are soil-survey and soil chemical and physical data; climatic, topographic, and hydrologic data; field and laboratory research data; long-time land-use records; and experience and observations of land response to various uses.
Soil surveys are basic in that they identify soil properties important for determining land-use capability. Surveys provide maps showing the location and boundaries of soils. Soil depth, texture, kinds of minerals, salinity and kinds of salts, and acidity are some properties utilized in determining land capability. The particular set of properties needed for determining land capability is dependent on the particular use.
Land classifications cannot be static because they depend upon interpretation. As technology changes and economic and social conditions change, interpretations change. With basic soil, climatic, and topographic data and maps, however, these interpretations can be revised easily. Combinations of alternative land-use and management strategies, as well as social and economic conditions, do change, but the physical and chemical factors of well-managed land do not change very much.
Land classifications have been developed for specific purposes, for example, tax assessment, sales and credit, soil and water conservation, irrigation potential and management, wildlife suitability, watershed management, recreational uses, industrial uses, or highway construction.
Effective extrapolation of management or use strategies to the land depends on an adequate inventory of basic data and how well an existing classification is chosen or a classification developed that fits the specific need at a given time and place. With appropriate facts and maps, through land classification, predictions can be made about the results of using a specific type of land in a particular way. Then, planning of land use includes the practical combinations of management practices required, and the effects of management on the quality of land resources. Without land classification and appropriate maps, it is difficult to extrapolate experience and research results to the land.
Land capability classification systems have been utilized for many years by the Soil Conservation Service, USDA, and other land resource planning and management agencies. Their utilization is being enhanced greatly through the use of computer data storage and management systems. See also: Land-use classes
Modern soil management practices have a cumulative effect on the soil's future productivity. The foremost management question for the soil manager is whether a given practice or system is causing soil degradation. Many current agricultural problems have developed because managers did not realize the long-term impact of their management techniques. Some practices may appear to be sustainable for periods of time equal to a human lifetime, and yet they could be causing slow soil degradation that will ruin the soil for long-term use. Therefore, long-term plans that identify causes of soil deterioration and avoid it successfully are essential to a nation's future agricultural productivity. For example, proper use of fertilizers and manures can enhance long-term productivity and maintain environmental quality. Poor management choices including fertilizers and manures can actually ruin a soil and damage the environment, almost irreparably. Civilizations have often missed the “clues” that a particular system was not sustainable. This has led to loss of their food supply, and with their being conquered eventually or ceasing to exist.
Water conservation and efficient use of stored water are important issues within soil management. Since plant growth is limited by water supply in almost every climate situation at some point in the crop cycle, having adequate stored soil water is critical. Irrigation becomes important in climates where the natural precipitation will not meet plant demands or in areas where rainfall patterns are erratic.
Water conservation principles are the same in either rain-fed or irrigated situations, and can be divided into two phases: water capture in soil, and water retention in the soil. Water capture is maximized by maintaining open pores at the soil surface and protecting these surface pores from raindrop impact which would otherwise destroy surface soil structure and ultimately plug the pores. Crop-residue mulches are effective protectors of structure, and these are best maintained by reducing tillage. Residues also slow runoff water and increase the opportunity for infiltration into the soil. The second phase of water conservation, retention of captured water, requires prevention of weed growth that would waste the captured water, and reduction of evaporation. Protecting the soil surface from exposure to sunlight and wind by maintaining crop residues at the surface is an important management tool for evaporation control. Residues serve as reflectors of light energy, as insulators to keep the soil surface cool, and as wind deflectors. Reducing and even eliminating tillage maximizes the crop residue on the soil surface and consequently maximizes water retention.
Irrigation is not always an option. Many locations in the world do not have adequate water supplies to allow irrigation, even though the crops could benefit from additional water. In some cases, underground water supplies are present but are so deep that the additional crop growth with irrigation would not offset the cost of pumping the water. Soil factors such as texture, depth, and salt content also dictate the feasibility of irrigation. Some very fine textured soils will not absorb water readily and are not good candidates for irrigation; others are so shallow that the storage capacity is too little. If soils have too many soluble salts, they are not suitable for irrigation. There are many combinations of climate, soils, water supply, and economic parameters that must be considered when determining if irrigation will be feasible or profitable.
All irrigation waters, whether they are diverted from rivers or pumped from the underground supply, contain some dissolved salts. Thus, when soils are irrigated they receive a dose of these salts. If the irrigation manager does not plan for these salts, the soil can become saline and lose productivity. See also: Water conservation
Soil management involves integration (Fig. 25). The resource block includes climate, soil, and plants; these are the natural-resource inputs. Modifications can and will be made on them, but for the most part they are the constants of the system. The agriculturist enters the scheme at this point to impose a management system, which has as its goal the production of food, fiber, or forages. Figure 25 also shows that the cropping system imposed is directly related to the soil, climate, and plant resources. The manager implements a cropping system through choice of crop rotation and tillage system. Choice of tillage system is often conditioned by the rotation chosen.
Fig. 25 Diagram showing an integrated system of soil management.
Erosion control, organic matter conservation, and water conservation are the classical concerns of soil management; they are all highly linked. A change in one or more of these three factors creates feedbacks to the others. For example, control of erosion leads to direct savings in soil organic matter and soil water and thus soil conservation. Furthermore, the control of erosion, which relates to the first goal of preventing soil deterioration, is linked directly to the second goal, namely, improvement of the soil system. Erosion prevention techniques also improve soil water storage by reducing water runoff. Conservation of organic matter may increase water infiltration, which decreases runoff and thus decreases erosion. The other blocks in Fig. 25 represent pest management and irrigation and drainage. They are equally important, but have not been studied to as great an extent as the first three factors. In fact, pest and pesticide management are relatively new concepts in soil management. Soil tillage methods and rotations are intimately involved in pest and pesticide management.
Finally, basic economic principles are involved in all sound soil management systems, as indicated in Fig. 25 by the enclosure of all other factors by economics. This is true in even the most primitive civilizations. If the people cannot live on the food or income produced, the system cannot survive. In more developed nations, quality of management has even influenced the potential value of the land. The ability to cope with the problems of a particular soil eventually determines its economic value to society.
Forces external to the system, such as changes in government policy, can greatly impact the overall management system. For example, the government has funded land set-aside programs for multiple reasons, ranging from attempts to decrease production of various crops to efforts to slow soil erosion. These interventions cause multiple, and often unexpected, interactions within the system. The programs encourage producers to take land from production of annual crops, such as corn, cotton, sorghum, and wheat, and to replace them with perennial grasses. During the contract period, the producer cannot harvest or graze the grass, and it is left as a cover crop on the soil.
Quite obviously these land set-aside programs have an economic impact on both the producer and local community. The producer experiences less risk because inputs to the system are low and the income is stable, no matter what the weather is. At the same time, the community merchants and the local labor market are often negatively impacted because the merchant sells fewer inputs and there is less need for seasonal labor. In contrast to the mixed economic outcome, the set-aside programs have nothing but positive influences on the soil itself. Soil erosion is slowed, relative to the conditions prior to the set-aside program, and soil organic matter contents stabilize or rise. However, these potentially positive soil changes may not be long-lasting because government policies change on a short-term basis (5–10 years) while soil changes require multiple decades to be of long-term value to the nation. Depending on the management scheme adopted by the producer, the benefits of 10 years in perennial grass can be lost in 1–2 years.
High-quality, efficient management of the natural resources of climate, water, and soil is needed to provide the food and fiber that will sustain life on Earth. Soil management decisions will have an important effect on the future food supply.
Soluble salt and exchangeable cation concentrations play major roles in determining the pH, physical characteristics, and chemical composition of soils. When a salt dissolves in water, it dissociates or separates into cations and anions. The predominant cations in salt-affected soils are calcium (Ca2+), magnesium (Mg2+), sodium (Na+), and potassium (K+); the predominant anions are chloride (Cl−), sulfate (SO42−), carbonate (CO32−), and bicarbonate (HCO3−).
Clays and organic matter contain negative electrically charged sites. In salt-affected soils, this charge is satisfied by calcium, magnesium, sodium, and potassium ions. The cations, bound to the exchange sites by electrical charge, are known as exchangeable cations because they can be removed from the charged surface only by replacement with another cation from the soil solution. See also: Humus; Soil chemistry
Each soil can be classified as normal, saline, sodic, or saline-sodic, based on its salt content and exchangeable cation ratios.
These soils do not contain sufficient soluble salts or exchangeable sodium to adversely affect plant growth or soil physical properties. Normal soils have saturation paste pH values less than 8.3 and saturation paste extract electrical conductivities of less than 4 decisiemens per meter and an exchangeable sodium percentage less than 15. A saturation paste is made by mixing just enough distilled water with a soil sample to fill the voids without having excess water standing on the surface of a well-mixed sample after 4–16 h. These electrical conductivities are defined upper limits, but if salt-sensitive plants are grown on soils with an electrical conductivity of 3.5 dS/m, a significant yield reduction will result. Likewise, using a high-volume sprinkler system to irrigate a soil with an exchangeable sodium percentage of 12 could produce serious runoff rates because of low infiltration rates. See also: Hydrolysis; pH
These soils contain sufficient soluble salts (electrical conductivity greater than 4 dS/m) in the upper root zone to reduce yields of most cultivated crops and ornamental plants. The exchangeable sodium percentage is less than 15, the sodium absorption ratio is less than 13, and the pH is less than 8.3. The predominant cations are calcium, magnesium, and, in a few cases, potassium. The predominant anions are chloride and sulfate. Water entry and movement through these soils is not inhibited by high exchangeable sodium concentrations. Osmotic effects and chloride toxicity are the predominant causes of plant growth reduction. See also: Osmosis
These soils are lower in soluble salts than saline soils (electrical conductivity less than 4 dS/m). The exchangeable sodium percentage is greater than 15 and the sodium absorption ratio of the saturation paste extract is greater than 13. The pH of the saturation paste is greater than 8.5. Bicarbonate, carbonate, and hydroxide (OH−) ions are the predominant anions in these soils; these anions cause calcium to precipitate from solution as calcium carbonate (CaCO3; lime). The combination of high exchangeable sodium percentage and pH, and low electrical conductivity and exchangeable calcium causes the clay and organic matter to disperse, which in turn destroys the soil structure or tilth, causing so-called slick spots. These spots have extremely low rates of water and air exchange. They often have a black, greasy, or oily-looking surface due to the dispersed organic matter. Vegetation may be absent because of low water infiltration and insufficient plant-available water.
These soils are similar to saline soils in that the electrical conductivity is greater than 4 dS/m and the pH is below 8.3. Saline-sodic soils differ from saline soils in that more than 15% of the exchangeable cations are sodium and the saturation-paste sodium absorption ratio is greater than 13. The anions are a mix of bicarbonate, chloride, and sulfate. As long as the electrical conductivity remains above 4 dS/m, infiltration rates and hydraulic conductivities are similar to those of normal or saline soils. Irrigating saline-sodic soils with water having low salt concentrations will convert them into sodic soils if they do not contain gypsum (a calcium sulfate mineral). This happens as the electrical conductivity decreases without a decrease in the exchangeable sodium percentage, causing the undesirable properties of sodic soils to be expressed. It is not uncommon to have a mix of two or more classes of salt-affected soils within a field. Salt-affected soils tend to be highly variable from one part of a field to another.
Most soluble salts and exchangeable cations are derived from rock and mineral weathering of the soil parent materials. In high-rainfall, humid, and tropical areas, rain and melting snow leach the salts from the soil as they form. In arid and semiarid areas, the annual evapotranspiration potential is greater than the total annual precipitation, and the salts are not always leached from the soil as they are released. With time, they may accumulate in the root zone at concentration levels that affect plant growth.
Salts can also accumulate above shallow water tables as water moves to the soil surface by capillary rise (wicking) and evaporates, leaving the salts on or near the surface. Shallow water tables may occur naturally, induced by irrigating poorly drained areas, by irrigating upslope from lowlands, or by construction activity that blocks natural subsurface lateral drainage.
All natural waters contain dissolved salt. In many arid and semiarid areas, good-quality irrigation water (low in salts and low in sodium) is not available; consequently, irrigation water is used that contains more salts or sodium than is desirable. If sufficient water does not move through the soil and leach the salts below the root zone, salts or sodium will accumulate in the soil. It is often stated that under irrigation “hard water makes soft soils and soft water makes hard soils.” This implies that irrigation waters containing predominantly calcium and magnesium salts (sodium absorption ratio less than 3 or 4) tend to promote a more friable soil condition than do waters with high sodium concentrations.
Four conditions must be satisfied in order to remove soluble salts and excess sodium from soils: (1) less salt must be added to the soil than is removed; (2) salts must be leached downward through the soil; (3) water moving upward from shallow water tables must be removed or intercepted to avoid additional salts moving back to the soil surface; and (4) in sodic and saline-sodic soils the exchangeable sodium must be replaced with another cation, preferably calcium, and the sodium leached out. Applications of soil amendments (gypsum, iron sulfate, sulfur, or sulfuric acid) are beneficial only on sodic soils when leaching also occurs and on leaching of saline-sodic soils that do not contain natural gypsum.
Saline and sodic soils are found primarily in arid and semiarid areas of the world. Exceptions are recently drained coastal areas, salt marshes, and soils formed in depressions from marine deposits where the weathering products are not leached from the soil. Sodic and saline-sodic soils can also form where mists are carried from ocean waters by the wind and deposited on soil surfaces in arid and semiarid coastal areas. Aridisols and Entisols include most salt-affected soils. Low rainfall and unweathered soil materials result in insufficient salts leaching from the root zone of these soils. Mollisols, Alfisols, and Vertisols also contain considerable saline and sodic soil areas.
Human activities such as spills or intentional dumping of salts or solutions from oil well drilling-mud ponds, mines, food-processing plants, municipal sewage water, power-plant cooling-tower water, or heavy applications of wood ash can induce saline and sodic conditions in any soil when soluble salts are applied faster than they are leached from the soil.