Analysis of Moisture Characteristic Curves for Two Different Soils

The soil water characteristic curve is a graphical representation of the relationship between soil water content and matric potential (suction), reflecting the functional relationship between soil water energy and soil water content. It is also an important indicator for characterizing the fundamental hydraulic properties of soil. The concept of "macropores" is relative; generally speaking, any pore—regardless of its size or shape—that facilitates preferential migration of water and solutes can be classified as a macropore. Numerous experiments and studies have shown that macropores are widely present in natural soils. Although macropores account for only 0.1% to 5% of the total soil volume, they significantly influence the movement of water and solutes within the soil. Therefore, when studying the soil water characteristic curve, it is essential to take into account the effects of macropores.

This paper takes the soil from Dongyang Town, Qixia District, Nanjing City as an example. Using the soil moisture meter method, we measured the soil moisture characteristic curves for both undisturbed soil columns containing large pores and homogeneous soil columns (disturbed soil) lacking large pores. We then fitted the measured data using the Van Genuchten model. Finally, we analyzed and discussed the effects of large pores on the soil moisture characteristic curves, the fitted model parameters, and key water constants such as field capacity and wilting coefficient. The soil samples used in this experiment were collected from Dongyang Town, Qixia District, Nanjing City. The sampling depths were 0–20 cm and 20–40 cm. According to the International Soil Science Society’s classification criteria for soil particle size, we determined the composition of clay (<0.002 mm), silt (0.002–0.02 mm), and sand (0.02–2 mm) in the soil samples using the pipette method, and thereby identified the soil texture. Each soil sample was tested twice, and the average values were taken to obtain the mechanical composition, as shown in Table 1. The centrifuge used in the experiment was a Hitachi CR21 high-speed refrigerated centrifuge with rotor number 60; the inner diameter and height of the ring knife used were both 5 cm.

In the field, undisturbed soil samples were collected using ring samplers. Since the ring samplers have a diameter of only 5 cm, the actual sampling depths ranged from 7.5 to 12.5 cm and from 27.5 to 32.5 cm, respectively. After air-drying, the soil samples were passed through a 2-mm sieve and then uniformly packed into ring samplers according to their bulk densities, yielding homogeneous soil samples. The prepared soil samples were saturated in deaerated water for 48 hours, after which their initial masses were measured. They were then placed into centrifuge cups and subjected to centrifugation at increasing speeds—500, 1000, 1500, 2000, 3000, 4000, 5000, and 6000 rpm—in sequence. At each rotation speed, the samples were centrifuged for 100 minutes, after which their weights were recorded, thus generating a series of data points relating soil moisture content to the corresponding soil water potential. Using the mathematical software MATLAB, these data points were fitted using the nonlinear least-squares fitting function lsqcurvefit to obtain the parameters of the Van Genuchten model, ultimately yielding an expression for the soil moisture characteristic curve.

The formula for centrifugal force is:

h = 1.118 × 10⁻⁵ × n² × h' × (r₀ - h'/2) (Equation 1) where: h is the soil water suction (cm H₂O); n is the centrifuge rotation speed (rpm); h' is the length of the soil sample in the ring knife (cm); and r₀ is the radius of rotation at the reference water surface (cm), which is the distance from the center of the centrifuge rotor to the farthest point of the soil sample in the centrifuge chamber.

The Van-Genuchten model, which describes the soil water characteristic curve, features both continuity and continuous slope, yielding a smooth curve that is widely applicable to most soils over a fairly broad range of water potentials or moisture contents. Moreover, it allows for an analytical solution of the relative hydraulic conductivity. Consequently, this model has found the most extensive application in the field of soil physics. Its mathematical expression is as follows:

θ = θr + θs - θr[1 + ah^n]

In equation (2), mh<0: θ represents the volumetric water content (cm³/cm³); θr denotes the residual water content (cm³/cm³); θs stands for the saturated water content (cm³/cm³); h is the suction (cmH₂O); and a, n, and m are fitting parameters, where a is the scale parameter, inversely proportional to the average pore diameter, given by a = 1/hb, with hb being the air-entry suction; m = 1 - 1/n (where n > 1).

A comparative experimental study was conducted to examine the shapes of soil moisture characteristic curves for both macroporous soils and homogeneous soils. The relationships between soil water suction and volumetric water content at depths of 0–20 cm and 20–40 cm are presented in Table 2, while the corresponding soil moisture characteristic curves are shown in Figures 1 and 2. As can be seen from the figures: the moisture characteristic curve of undisturbed soil is steeper than that of homogeneous soil. That is, as suction increases, the decrease in volumetric water content in undisturbed soil is smaller than in homogeneous soil. This indicates that within the effective suction range determined by the centrifuge method, undisturbed soil exhibits better water retention and stronger water-holding capacity. The reason for this phenomenon lies in the presence of macropores of varying sizes in undisturbed soil. These large soil pores have relatively low matrix potentials, allowing soil water to preferentially move through them. When the suction exceeds 100 cmH2O, the macropores are already drained, leaving only water remaining in the smaller pores. Since smaller pores exert greater suction on soil water, the same increase in suction results in less water being released from the soil matrix compared to homogeneous soil—thus manifesting as a steeper slope on the soil moisture characteristic curve. In contrast, homogeneous soil disrupts both the macropore and micropore structures of undisturbed soil, promoting the development of medium-sized pores that are more uniformly distributed. Consequently, as suction increases, the volumetric water content of homogeneous soil decreases gradually, resulting in a flatter slope on its soil moisture characteristic curve. From Table 2 and Figures 1 and 2, it can also be observed that when suction is below 408.44 cmH2O, the volumetric water content of undisturbed soil is consistently lower than that of homogeneous soil; however, when suction exceeds 919.98 cmH2O, the volumetric water content of undisturbed soil becomes higher than that of homogeneous soil. The lower volumetric water content of undisturbed soil at suction levels below 408.44 cmH2O is attributable to the fact that the water in the macropores is primarily gravitational water. Under conditions where the macropores remain interconnected, this portion of water will immediately drain due to gravity as soon as the soil sample is removed from the air-water interface. Even if the macropores are not interconnected, the water in these pores will have already been completely lost at suction levels ranging from 408.44 to 919.98 cmH2O. Hence, in the low-suction region of the soil moisture characteristic curve, the volumetric water content of undisturbed soil is lower than that of homogeneous soil. On the other hand, when suction exceeds 919.98 cmH2O, the volumetric water content of undisturbed soil becomes higher than that of homogeneous soil. This is because, at suction levels above 919.98 cmH2O, the water in undisturbed soil is predominantly retained in the fine-pore structure. Due to capillary forces, fine pores exhibit better water retention, making it difficult for this water to be lost even as suction increases. In contrast, homogeneous soil, while disrupting the macropore structure of undisturbed soil, also destroys its fine-pore structure. As a result, homogeneous soil shows poorer water retention at suction levels above 919.98 cmH2O, leading to a higher volumetric water content in undisturbed soil in the high-suction region of the soil moisture characteristic curve.

The suction range for measuring the soil moisture characteristic curve using the centrifuge method is 100–15,849 cmH₂O; however, this method fails to capture the process of water loss from large pores. To describe in detail the evolution of the soil moisture characteristic curve during water loss from large pores, the soil column method can be employed. The suction range for the soil column method is 0–100 cmH₂O. For more details, refer to the book "Methods for Determining Soil Physical Properties," compiled by the Japanese Committee for the Determination of Soil Physical Properties. To compare the parameters of the soil moisture characteristic curves between coarse-pore soils and homogeneous soils, we used MATLAB’s nonlinear function lsqcurvefit to fit the measured data points. The results are shown in Table 3. The sum of squared residuals refers to the sum of the squares of the differences between the measured values and the corresponding values on the fitted curve; a smaller value indicates a better fit. Other parameter meanings are consistent with Equation (2). The magnitudes of the sum of squared residuals obtained after fitting the suction and moisture content data for the four soil samples were all on the order of 10⁻⁵, indicating an excellent fit. The fitted data also reveal the following: The residual moisture contents of undisturbed soils are consistently higher than those of homogeneous soils; the saturated moisture contents of undisturbed soils are consistently lower than those of homogeneous soils; and the scale parameter a of undisturbed soils is consistently smaller than that of homogeneous soils. The higher residual moisture content of undisturbed soils compared to homogeneous soils stems from the fact that undisturbed soils contain fine pores with excellent water-retention capacity, making them less prone to water loss even under high suction. In contrast, homogeneous soils have had their fine pores disrupted, leading to the development of medium-sized pores that are less capable of retaining water under high suction. Consequently, the residual moisture content of undisturbed soils is greater than that of homogeneous soils. The scale parameter a of undisturbed soils is about one-sixth that of homogeneous soils—meaning the average pore diameter of undisturbed soils is larger than that of homogeneous soils. This is because undisturbed soils, subjected to alternating wetting and drying cycles, freeze-thaw processes, dissolution of soluble substances in the soil, plant growth, and animal activities, exhibit a greater abundance of coarse-pore structures, resulting in a larger average pore diameter. On the other hand, homogeneous soils have had their coarse-pore structures disrupted, promoting the development of medium-sized pores and thus reducing the average pore diameter and increasing the value of a.
   
By substituting the parameters from Table 3 into Equation (2), we can obtain the soil moisture characteristic curves for the four soil samples.

2.3 The Impact of Macropores on Field Capacity and Wilting Coefficient Field capacity refers to the soil moisture content at which capillary-held water in the soil reaches its maximum level. When the soil moisture content attains field capacity, the maximum suction force exerted by soil particles on water molecules is approximately 0.3 atmospheres, or 309.9 cm H₂O. By substituting h = 309.9 into Equations (3), (4), (5), and (6), we obtain the following field capacities for undisturbed soil and homogeneous soil: - For undisturbed soil from 0 to 20 cm: 0.3509 - For homogeneous soil from 0 to 20 cm: 0.3834 - For undisturbed soil from 20 to 40 cm: 0.3588 - For homogeneous soil from 20 to 40 cm: 0.3936 From these results, it is evident that the field capacity of undisturbed soil is lower than that of homogeneous soil. This difference arises because undisturbed soil contains macropores, which have poor water-holding capacity. Under relatively low suction forces, water in macropores tends to be easily lost. In contrast, homogeneous soil features well-developed mesopores that are evenly distributed throughout the soil matrix; thus, it exhibits better water retention at lower suction levels. These findings suggest that agricultural tillage can disrupt the structure of macropores, which are less capable of retaining water, thereby promoting the development of mesopores. As a result, the soil becomes more effective at holding water, providing crops with the essential moisture they need for growth.

The wilting coefficient refers to the soil moisture content at which the soil particles exert an attractive force of 15 atmospheres—equivalent to 15,495 cmH₂O—on water molecules. At this moisture level, the water in the soil can no longer be absorbed by plant roots, leading to permanent wilting of the plants. By substituting these values into Equations (3), (4), (5), and (6), we obtain the following field capacities for undisturbed soil and homogeneous soil: 0–20 cm—0.2319 and 0.2157, respectively; and 20–40 cm—0.2477 and 0.2035, respectively. As can be seen, the wilting coefficient of undisturbed soil is higher than that of homogeneous soil. This difference arises because undisturbed soil contains more fine pores, whereas homogeneous soil has disrupted these fine pores.

This conclusion indicates that when using soil moisture characteristic curves derived from homogenized soils to design and guide agricultural irrigation, it is necessary to adjust these curves so that the irrigation regime can both meet the water requirements of crops and maximize water conservation. In this study, we determined the soil moisture characteristic curves for undisturbed soils and homogenized soils at depths of 0–20 cm and 20–40 cm using the centrifuge method. We then fitted the measured data using the Van Genuchten model. By comparing and analyzing the effects of large pores in undisturbed soils on the soil moisture characteristic curves and their parameters, as well as on field capacity and wilting coefficient, we found that although large pores account for only a small fraction of the total soil volume, they exert significant influences on both the low-suction and high-suction segments of the soil moisture characteristic curve. Specifically, large pores cause undisturbed soils to have lower water contents than homogenized soils in the low-suction segment, but higher water contents than homogenized soils in the high-suction segment. Moreover, the scale parameter of the Van Genuchten model is smaller for undisturbed soils than for homogenized soils. We also noted that the field capacity of undisturbed soils containing large pores is lower than that of homogenized soils, while the wilting coefficient of undisturbed soils is higher than that of homogenized soils.