Size of Sand Grains Controls Pore Structure and Water Dynamics: Implications for Water Retention and Hydraulic Conductivity

Abstract

Sand grain size strongly influences the physical and hydraulic behaviour of sandy soils, particularly water retention, pore distribution, and water movement under unsaturated conditions. This study evaluated the effect of five sand grain-size classes, ranging from very coarse to very fine, on pore distribution, aeration, water retention, and unsaturated hydraulic conductivity. Quartz sand samples with different particle sizes were saturated and subjected to matric tensions ranging from 10 to 15,000 hPa. Very fine sand (0.053–0.106 mm) showed the highest field capacity (0.38 m³ m⁻³) and available water content (0.30 m³ m⁻³), which were associated with a predominance of pores between 0.2 and 3 μm in diameter. In contrast, coarser sand fractions were dominated by macropores (>50 μm) and exhibited lower water retention. Permanent wilting point values remained low and similar among grain-size classes (≈0.02 m³ m⁻³). Under unsaturated conditions (matric tensions > 100 hPa), very fine sand exhibited hydraulic conductivity values up to ten times greater than those of coarser fractions. Overall, decreasing sand particle size increased water retention and plant-available water while reducing macroporosity and aeration capacity. These findings demonstrate that sand grain-size distribution plays a major role in regulating water dynamics in sandy soils and may support the development of more efficient irrigation and soil management strategies to improve water conservation and plant water availability in drought-prone environments.

1. Introduction

Soil texture plays a significant role in regulating ecosystem water limitation, as it controls water retention and availability across a wide range of environmental conditions [1]. Although widely recognised for its low water retention capacity compared to silt and clay, sand forms the matrix of sandy soils found in various regions of the world [1], often associated with sedimentary rocks [2]. This lower water retention is attributed to the predominance of macropores, which promote aeration and drainage at the expense of micropores responsible for capillary retention and adsorption [3,4]. In addition, the smaller specific surface area of sand particles limits their participation in physical–chemical water retention processes [5]. These characteristics have been consistently reported in studies on water dynamics in sandy soils [2,6,7,8]. These processes are complex because pore size and shape vary, as reported in the study by [9].

However, water retention capacity in sandy soils is not uniform, as it varies according to the diameter of the sand particles, which is influenced by the source material and pedogenic processes [10,11,12]. In monodisperse systems, in which particles have similar diameters, larger pores are formed. On the other hand, in polydisperse systems, the presence of smaller particles filling the spaces between larger ones results in smaller diameter pores, significantly altering the distribution of pore space and, consequently, water retention and movement [3,4].

Water retention in the soil is governed by two mechanisms: adsorption, which occurs on surfaces charged with clay minerals, oxides, and organic matter, and capillarity, which is dominant in small-diameter pores [13]. While adsorption is related to the specific surface area, capillarity results from the forces of adhesion, cohesion, and surface tension, which allow water to rise and be retained in micropores [3]. Near saturation, water flow is governed by volumetric capillary forces in water-filled pores, yielding high hydraulic conductivity regardless of texture. As the soil dries and capillary continuity breaks down, transport becomes predominantly dependent on adsorbed water films at particle surfaces, a mechanism that is more efficient in finer fractions due to their higher specific surface area. Thus, the distribution of pore size, determined by the arrangement and diameter of the particles, defines which mechanism will predominate and, consequently, control water dynamics within the soil profile.

The USDA classification system [14] divides the sand fraction into five classes: very coarse (2–1 mm), coarse (1–0.5 mm), medium (0.5–0.25 mm), fine (0.25–0.106 mm) and very fine (0.106–0.053 mm). This subdivision reflects a variation of approximately 19-fold variation in the average particle diameter, which can dramatically influence porosity, pore size distribution and, consequently, water retention and conduction processes. Although some studies have suggested that finer sand fractions increase water retention [11,15,16], there is still a lack of research that systematically evaluates the effect of each grain-size class in monodisperse systems, especially on aeration and unsaturated hydraulic conductivity.

While the influence of particle size distribution on soil physical properties is a classic topic, studies that isolate the exclusive effect of sand particle diameter in monodisperse systems remain scarce. Unlike some research focusing on samples with preserved structure or natural mixtures of sand, silt, and clay [2,5,12], this work analyzes water dynamics by quantifying changes in pore size distribution, water retention, and unsaturated hydraulic conductivity across a wide range of matric tensions (from 10 to 15,000 hPa). By controlling the variability inherent to fine and organic fractions, this study provides quantitative parameters that allow the isolation of how the reductions in particle diameter affects water retention and hydraulic conductivity.

Furthermore, the sand particle size distribution directly influences soil bulk density, as particles of different sizes affect not only the total porosity and pore size distribution but also the arrangement of the soil matrix, which ultimately governs the bulk density.

Hydraulic conductivity, a fundamental property for describing the water transmission capacity in porous media, is directly influenced by the grain size of sand [17,18]. This property is crucial in practical applications such as water resource management, infiltration modelling [19], geotechnical stability, and erosion prediction [8]. Studies such as Tsai et al. [18] have shown that reducing particle diameter decreases the infiltration rate in Green-Ampt models, which are widely used in hydrological and erosion simulations [20]. However, more detailed assessments are still needed to determine how each sand fraction affects unsaturated hydraulic conductivity, especially at higher matric tensions.

Given the above, the hypothesis of this study is that reducing the diameter of sand particles increases microporosity and water retention at field capacity, with minimal change in the permanent wilting point, resulting in greater available water and unsaturated hydraulic conductivity, but with a reduction in macroporosity and aeration capacity. The overall objective was to evaluate how five sand particle size classes affect pore size distribution, aeration, water retention, and unsaturated hydraulic conductivity under controlled conditions in a monodisperse system.

2. Materials and Methods

2.1. Study Location and Material

The study was conducted at the Soil Physics and Management Laboratory of the Santa Catarina State University, in Lages, Santa Catarina, Brazil. The soil was classified as Quartzipsamment (USDA), equivalent to Neossolo Quartzarenico (Brazilian Soil Classification), confirmed by preliminary petrographic analysis, with a particle density of 2.60 g cm⁻³.

2.2. Sample Preparation and Particle Size Fractionation

The sand was air-dried, disaggregated, and subjected to sequential sieving using mesh openings of 2.0, 1.0, 0.5, 0.25, 0.106, and 0.053 mm. The fractions retained in each interval were classified according to the USDA system [14] as: very coarse sand (1.0–2.0 mm), coarse (0.5–1.0 mm), medium (0.25–0.5 mm), fine (0.106–0.25 mm) and very fine (0.053–0.106 mm). Material above 2.0 mm and below 0.053 mm was discarded.

Each fraction was washed sequentially with running water, a 5% HCl solution (to remove carbonates and impurities) and distilled water. The samples were then dried in an oven at 105 °C for 48 h and stored in airtight containers prior to use.

2.3. Assembly of Cylinders and Saturation

Volumetric steel cylinders (70.7 cm³; 6 cm diameter × 2.5 cm height) were prepared with a fine fabric fixed to the base with elastic, to allow saturation without loss of material. Each sand fraction was placed in the cylinders and subjected to mechanical vibration for 1 min to ensure uniformity and particle settling. Five replicates were used per grain-size fraction.

The cylinders were placed on trays, saturated by capillarity for 24 h, and then weighed on an analytical balance with an accuracy of 0.001 g.

2.4. Water Retention Curve Determination

The saturated samples were subjected to sequential matric tensions of 10, 60 and 100 hPa on a sand tensions table, and 330, 1000, 3000, 5000 and 15,000 hPa in Richards chambers with porous plates, according to the methodology described by [21]. After equilibrium at each tension (usually 72 h), samples were weighed to determine the retained water content. Finally, the samples were dried in an oven at 105 °C for 48 h to determine the dry mass.

2.5. Porosity and Water Retention Calculations

Based on the retained moisture data at each tension, the main physical–hydrological attributes were calculated: total porosity (θs), microporosity (water content at a tension of 60 hPa), macroporosity (difference between total porosity and microporosity), field capacity (FC—retention at 100 hPa), permanent wilting point (PWP—retention at 15,000 hPa), available water (difference between FC and PWP), aeration capacity (air volume at 100 hPa) and bulk density, according to the methodology established by [21]. To analyse the relationship between average diameter of sand particles and each of these attributes, second-order inverse polynomial nonlinear regression models were fitted using the REG procedure of SAS (version 9.2) [22]. The significance of the fits was evaluated by the F-test at a 5% probability level.

2.6. Calculations of Pore Volume in Each Sand Diameter Class

Pore volume in different diameter classes was calculated based on the capillarity equation, which relates the matric tension (h) to the pore radius (r), as shown in Equation (1):

$$r={\displaystyle \frac{2.\sigma .\cos \alpha }{\rho .g.h}}$$

(1)

where r = pore radius (m, converted to μm); σ = surface tension of water (0.07287 N m⁻¹ at 20 °C); α = contact angle (0°); ρ = density of water (1000 kg m⁻³); g = acceleration due to gravity (9.81 m s⁻²); and h = matric tension (hPa).

Based on this, six pore diameter classes were defined, corresponding to the applied matric tension ranges: 300 μm (0–10 hPa); 300–50 μm (10–60 hPa); 50–30 μm (60–100 hPa); 30–3 μm (100–1000 hPa); 3–0.2 μm (1000–15,000 hPa); and <0.2 μm (>15,000 hPa).

Pore volume in each class was obtained by the difference in retained moisture between the corresponding limit matric tensions. The data were subjected to analysis of variance in a completely randomised design, using factorial arrangement consisting of pore diameter classes and sand grain-size classes, with five replicates. When significant by the F test (p ≤ 0.05), the means were compared by Tukey’s test at 5% probability, using the SAS statistical package [22].

2.7. Adjustment of the Water Retention Curve

The water retention curves were fitted using the model proposed by [23], which incorporates a correction factor for residual moisture, as shown in Equation (2):

$$\theta \left(\psi \right)=\left(1-\left({\displaystyle \frac{\mathrm{l}\mathrm{n}(1+\raisebox{1ex}{$\psi $}\!\left/ \!\raisebox{-1ex}{${\psi }_r$}\right.)}{\mathrm{l}\mathrm{n}[1+(\raisebox{1ex}{${10}^6$}\!\left/ \!\raisebox{-1ex}{${\psi }_r$}\right.\left)\right]}}\right)\right)\times {\displaystyle \frac{{\theta }_s}{{\left\{\mathrm{l}\mathrm{n}[e+{\left(\raisebox{1ex}{$\psi $}\!\left/ \!\raisebox{-1ex}{$\alpha $}\right.\right)}^n]\right\}}^m}}$$

(2)

where θ(ψ) = volumetric moisture (m³ m⁻³); ψ = matric tension (hPa); ψr = tension corresponding to residual moisture (hPa); θs = volumetric moisture content at saturation (m³ m⁻³); α, n, m = model adjustment parameters; ln = natural logarithm; and e = base of the natural logarithm (≈2.718).

The fitted parameters in this model were used to discuss the differences in the characteristic water curves between the sand classes. The comparison of water retention curves for the five sand fractions was performed by multiple contrasts, using SAS generalised linear mixed models (PROC GLIMMIX) (Cary, NC, USA), using the estimated parameters (α, n, m) for the equation [23], and the covariance matrix of the estimated parameters for the sand fractions, generated by SAS’s Nonlinear Regression Procedure (PROC NLIN) [22].

2.8. Unsaturated Hydraulic Conductivity Calculations

To estimate unsaturated hydraulic conductivity under different matric tensions (K(ψ)), the water retention curves were initially fitted to the van Genuchten model [24], according to Equation (3), using the Mualem criterion (Equation (4)):

$$\theta \left(\psi \right)={\theta }_r+{\displaystyle \frac{{\theta }_s-{\theta }_r}{{[1+{\left[\alpha \left|\psi \right|\right]}^n]}^{(1-(\frac{1}{n}\left)\right)}}}$$

(3)

$$m=1-{\displaystyle \frac{1}{n}}$$

(4)

where θ(ψ) = volumetric moisture at tension ψ; θs = volumetric moisture at saturation; θr = residual volumetric moisture (tension of 15,000 hPa); α and n = dimensionless adjustment parameters. ψ = matric tensions (hPa); units of θ in m³ and ψ in hPa.

From the parameters α and n obtained, the unsaturated hydraulic conductivity K(ψ) was estimated using Equation (5), which incorporates the saturated hydraulic conductivity (Ks) as a reference:

$$K\left(\psi \right)=k_s{\displaystyle \frac{{[1-{\left(\alpha \psi \right)}^{\left(n-1\right)}+{(1+{\left(\alpha \psi \right)}^n)}^{-m}]}^2}{{[1+{\left(\alpha \psi \right)}^n]}^{m/2}}}$$

(5)

The parameters of the van Genuchten model were fitted using the Marquardt method [25] through the PROC NLIN procedure of SAS [22], using the experimental data of moisture versus matric tension as input. Saturated hydraulic conductivity (Ks) was estimated using Mualem–van Genuchten approach based on the fitted parameters (α and n).

3. Results and Discussion

3.1. Bulk Density

Bulk density varied nonlinearly with increasing mean sand particle diameter and was described by a quadratic polynomial equation (Figure 1). The lowest density (1.45 g cm⁻³) occurred in the coarse sand fraction (average diameter of 0.87 mm), associated with lower particle packing and the formation of larger diameter pores, as also observed by [15]. This behaviour is related to the characteristic structural arrangement of each sand class, since spheroidal particles of the same size, arranged identically (e.g., closed cubic), result in similar porosity and density [26]. However, the variation observed indicates that each grain-size class developed distinct arrangements, influenced by the shape and heterogeneity of the particles, which directly affected packing and porosity.

In the finer fractions (very fine and fine sand), the density increased significantly due to the greater degree of accommodation between smaller particles. In the coarser fraction (very coarse sand), although larger diameter pores are formed, the greater unit mass of the particles increases the soil density. These results reinforce that, as highlighted by Hillel and Parahyba [3,12], particle size heterogeneity promotes a reduction in pore diameter, since the spaces formed by larger particles are filled by smaller particles.

3.2. Water Retention Curve by Fredlund & Xing

Analysis of water retention curves adjusted by the Fredlund & Xing model [23] revealed three distinct patterns (Figure 2). The first, represented by very fine sand, exhibited higher total porosity (θs), a gentler slope between 0 and 10 hPa, and a higher permanent wilting point. The second pattern, common to the fine, medium, and coarse fractions, was characterised by a small variation between 0 and 10 hPa and a greater decay between 10 and 100 hPa. The third, represented by very coarse sand, exhibited a steep slope between 0.1 and 10 hPa and a lower slope in the drier branch of the curve. Piecewise regression analysis of pore size distribution identified a breakthrough point at 10 hPa (Figure 2), marking the critical transition from drainage-dominated macropores to retention-dominated micropores (60 hPa), which explains the observed shifts in soil water dynamics across granulometric fractions. Similar results were reported by Costa et al. [2] for sandy soils in southern Brazil.

The parameter θs, which corresponds to total porosity, was similar among the medium, coarse, and very coarse classes, intermediate in fine sand, and higher in very fine sand (

Table 1. Values of the adjusted parameters of the Fredlund & Xing [23] model for the water retention curve, determined for the five granulometric classes of the sand fraction.

Sand Diameter (mm)	θs (m3 m−3) *	Ψr (hPa)	α (hPa)	n	m
1.0 to 2.0	0.51	0.1	0.07	0.59	2.4
0.5 to 1.0	0.52	0.1	1.0	0.32	96.5
0.25 to 0.50	0.50	30	1.0	0.53	25.7
0.106 to 0.25	0.55	20	1.1	0.57	7.0
0.053 to 0.106	0.67	187	53.7	2.30	0.4

* θs = total porosity; ψr, α, n and m = fitted parameters of the Fredlund & Xing [23] model for the water retention curve.

).

Considering a hypothetical tetrahedral arrangement between spheroidal particles of the same diameter, as described by [27], the diameter of the pore formed inside the tetrahedron varies according to the sand class. Tetrahedral packing is the most predominant arrangement in granular media subjected to sedimentation and mechanical settling, as it minimizes void space and maximizes particle contact points, thereby achieving a state of greater mechanical stability. For very fine sand (lower limit of 53 μm), the smallest pore formed would have a diameter of 24 μm, being drained from 123 hPa. For very coarse sand (upper limit of 2000 μm), the largest pore would have a diameter of 899 μm, being drained at only 3 hPa—a 41-fold difference in drainage tension (

Table 2. Average, minimum and maximum diameter of particles and pores formed in a theoretical tetrahedral arrangement, and corresponding drainage tension, for the five classes of the sand fraction.

	Particle Diameter			Pore Diameter *			Tension **
Sand	Average	Min	Max	Average	Min	Max	Average	Min	Max
	------------------------- μm ----------------------------						---------- hPa ----------
Very coarse	1500	1000	2000	674	449	899	4	7	3
Coarse	750	500	1000	337	225	449	9	13	7
Medium	375	250	500	169	112	225	17	26	13
Fine	178	106	250	80	48	112	37	62	26
Very fine	79.5	53	106	36	24	48	82	123	62

* Pore diameter calculated for a tetrahedral arrangement system, formed by spheroidal particles with the same diameter, based on Klein and Hurlbut [27]. ** Matric tensions to which water is subjected at that pore diameter, calculated by the capillarity equation.

). This mechanism helps explain the differences observed in the retention curves.

Furthermore, when analysing soil composed of particles with more uniform diameters (monodisperse system), the smallest particle in the coarse sand class has a diameter of 1000 μm, which is greater than the diameter of the largest pore (899 μm) formed inside the tetrahedron of this class (1000 to 2000 μm). Therefore, the pore remains free of solid particles inside (Table 2). This pattern occurs for the other sand classes. The exception is fine sand (106 to 250 μm), in which the tetrahedrons formed by the largest particles (250 μm) have pores with a diameter of 112 μm. These pores may contain fine sand particles with a smaller diameter (106 to 112 μm), which reduce the porosity formed but create pores with a smaller diameter inside these tetrahedrons. However, under natural conditions, the systems are polydisperse, in which smaller particles progressively fill the larger pores, increasing the number of pores but reducing their diameter, as highlighted by [3,12,16].

The matric tension limit at which the decrease in soil moisture content becomes small in relation to the increase in matric tension is called Ψr (tension at the residual moisture point). This limit varied between sand classes, increasing from 0.1 hPa in very coarse sand to 187 hPa in very fine sand (Table 1). These differences are due to the low volume of small-diameter pores in very coarse sand, resulting in low water retention. The α parameter is related to the matric tension at which air enters the pores. It can be observed that very coarse sand had the lowest α (Table 1), indicating greater water loss under lower tensions (up to 10 hPa) when compared to other fractions, such as very fine sand, in which higher tension is required to remove water from the pores (Figure 2). This behaviour in coarse sandy soils was reported by [12,28].

The other sand fractions showed intermediate α values, with fitted models presenting similar parameter estimates, ranging from 1.0 to 1.1 hPa (Table 1). As illustrated in Table 2, in a purely tetrahedral theoretical arrangement, water drainage would begin at 3 hPa for very coarse sand and 62 hPa for very fine sand. However, in practice, the arrangement of particles is not always tetrahedral. It is assumed that arrangements such as open cubic can also occur [3], forming pores with even larger diameters within the structure. In such cases, water drainage would begin under even lower tensions, explaining the greater water loss observed in the coarser fractions at low applied tensions.

The parameter n, which determines the slope of the retention curve, varied between fractions. Lower values were observed for coarse and medium sand, reflecting a sharp decrease in water content from a tension of 10 hPa. In contrast, the higher value of n in very fine sand indicates a more balanced distribution of pore diameters in the range of 0.2 to 300 μm. For very coarse sand, the high water loss under tensions below 10 hPa (expressed by the low value of α) promoted rapid stabilization of the curve, resulting in a reduced n value. This behaviour is accentuated by the lower total porosity (θs) of this fraction, from which the retention curve originates (Table 1).

The parameter m, which regulates the degree of inflection of the retention curve under high tension, showed variation (between 25.7 and 96.5) in the medium and fine sand classes, indicating a pronounced inflection in this segment of the curve. Very fine sand stood out with a significantly higher m value, corresponding to a low inflection near the residual moisture point. This demonstrates a more gradual variation in the volume of smaller diameter pores as tension increases. Together, these results show a more homogeneous distribution of pore diameter in the finer fractions, a pattern that corroborates the findings of [10,16].

3.3. Pore Diameter Distribution and Available Water

The distinct shapes of the water retention curve (Figure 2) were reflected in significant differences in pore volume distribution among the sand fractions (Figure 3). In very coarse sand, pores larger than 300 μm predominated, whereas in the coarse, medium, and fine fractions, pores between 50 and 300 μm represented the largest share of the pore space. In contrast, very fine sand showed a marked predominance of pores within the smallest diameter range (0.2–3 μm), indicating a stronger contribution of microporosity to water retention. The other fractions exhibited reduced pore volumes in the 30–50 μm range and, especially, in the interval associated with plant-available water (0.2–30 μm), which helps explain their limited capacity to store water for plant uptake. Similar trends have been reported in sandy soils, where higher contents of fine and very fine sand are associated with increased water retention and availability [11,12,15,29].

Among the five classes evaluated, very fine sand presented the most favorable pore distribution, combining a relevant proportion of aeration pores (>30 μm) with a substantial volume of pores in the plant-available range (0.2–30 μm). This balanced pore system indicates improved pore connectivity and a more gradual drainage pattern compared with the coarser fractions. These results agree with previous studies reporting that sandy soils with a greater proportion of finer sand particles tend to exhibit a more homogeneous pore network and improved water storage capacity [10,16]. Additionally, the association between fine particles (very fine sand and silt) and greater water availability has also been highlighted in sandy soil profiles from different Brazilian regions [6,10].

The physical attributes associated with water retention, total porosity (TP), microporosity, field capacity (FC), and permanent wilting point (PWP), decreased as particle diameter increased, following a second-order inverse polynomial model (Figure 4a,d–f). The highest values were observed in very fine sand (0.053–0.106 mm), followed by fine sand (0.106–0.25 mm) and the lowest values in the medium, coarse, and very coarse fractions. This greater water retention in the finer fractions, at all tensions evaluated, is a result of the more compact arrangement of particles, which reduces interparticle spacing, and the greater specific surface area available for adsorption and capillarity [3,30,31]. Consequently, a greater number of capillary micropores (diameter < 53 μm) are formed, which, together with the large contact surface, promote the retention of a significantly higher volume of water than that found in coarser fractions. The positive relation between fine sand content and greater water retention and availability in sandy soils has also been documented in field studies, such as those by [11,12] in sandstone-derived soils, highlighting the agronomic relevance of this attribute for mitigating risks to crops during periods of water stress.

In the medium, coarse, and very coarse sand classes, large-diameter pores predominate (>169 μm, >337 μm, and >674 μm, respectively—Table 2), which are classified as macropores. As a direct consequence of this distribution of pore space, the attributes related to water retention in micropores showed reduced values: microporosity and field capacity remained close to 0.05 m³ m⁻³, while the permanent wilting point was approximately 0.02 m³ m⁻³. This pattern of low water retention in coarser sand fractions is consistent with the results reported by [19,28].

As the sand diameter increased, the observed reduction in total porosity and permanent wilting point (Figure 4a,f) was less pronounced than the sharp decrease in microporosity and field capacity (Figure 4d,e). This disparity resulted in a concomitant increase in macroporosity and aeration capacity (Figure 4b,c). To quantify this inverse relation between particle diameter and macroporosity and aeration porosity, a second-order inverse polynomial model was also applied. It should be noted, however, that the coefficients of the terms b/x and c/x² in the equations for macroporosity and aeration capacity had opposite signs to those obtained in the adjustments for microporosity, field capacity, and permanent wilting point.

High macroporosity and aeration capacity were observed in all sand classes. In fractions ranging from fine to very coarse sand, the values of macroporosity and aeration capacity were very similar, differing by only 0.01 m³ m⁻³, with averages of 0.45 and 0.46 m³ m⁻³, respectively. The very fine sand class, however, stood out with a significant reduction in these two attributes, presenting values lower than 0.15 m³ m⁻³ when compared with the other fractions of 0.45 and 0.46 m³ m⁻³.

The relation between available water content and average sand particle diameter was also described by a second-order inverse polynomial model (Figure 5). The available water content showed a strong dependence on particle size, being maximum in very fine sand (0.30 m³ m⁻³), intermediate in fine sand (0.06 m³ m⁻³), and minimum in the medium, coarse, and very coarse fractions (≈0.02 m³ m⁻³).

The higher water retention observed in very fine sand across the entire range of matric tensions evaluated indicates that sandy soils with a predominance of this fraction retain more water than those dominated by medium or coarse sand, especially under matric tensions between 60 and 1500 hPa. This behaviour is attributed mainly to capillary retention, with a secondary contribution from adsorption forces [3]. The pores formed in this class (diameter between 0.2 and 30 μm) act efficiently as capillaries, being directly responsible for supplying water available to plants. The importance of smaller diameter pores for water retention in sandy soils was also evidenced in the study by [5], conducted with soils from the state of Santa Catarina, Brazil.

3.4. Unsaturated Hydraulic Conductivity (Kr) by Van Genuchten

The distinct water retention curves (Figure 2) and corresponding pore diameter distributions (Figure 3) directly influenced the unsaturated hydraulic conductivity among the sand classes (Figure 6). Since water flow in soil is proportional to the fourth power of the pore radius [3], differences in pore space distribution resulted in markedly different conductivity profiles. The higher hydraulic conductivity observed in finer fractions under high matric tensions is sustained by enhanced capillary and adsorptive forces, which facilitate the maintenance of continuous water films at solid–liquid interfaces; this preserves hydraulic pathways in tension regimes where the dominance of macroscopic capillary forces in coarser fractions would otherwise lead to hydraulic disconnection.

Contrary to classical expectations where finer textures exhibit lower K due to reduced pore sizes, the higher unsaturated hydraulic conductivity in very fine sand at high matric tensions (>1000 hPa) reflects a phenomenon driven by enhanced pore connectivity and sustained capillary–adsorptive water films, which maintain continuous flow pathways despite smaller individual pore radii, consistent with theoretical models emphasizing network tortuosity over isolated pore geometry.

In addition to unsaturated conductivity, the literature also establishes a strong relation between sand grain size and saturated hydraulic conductivity. Studies such as those by [7,18,32,33] converge in demonstrating that higher saturated conductivity values are associated with soils with a predominance of coarse and very coarse sand.

Joint analysis of the unsaturated hydraulic conductivity curves (Kr—Figure 6) and the Fredlund & Xing model [23] parameters (α, m, n—Table 1) reveals that Kr was similar between coarse and very fine sand fractions under low matric tensions (up to 10–30 hPa). This behavioral similarity is consistent with the values of the α parameter, which were also close for these classes. However, the very coarse sand fraction had the lowest Kr at low matric tensions, associated with its α value. This relation reflects the lower total porosity available for water flow in the unsaturated condition in this fraction.

With increasing matric tensions, unsaturated hydraulic conductivity (Kr) decreased in all sand classes, but at different rates of decline. It was observed that the very fine sand and very coarse sand fractions showed the lowest rates of Kr reduction as a function of increasing matric tensions, i.e., their curves exhibited gentler slopes.

For matric tension above 3000 hPa, the unsaturated hydraulic conductivity (Kr) of very coarse sand became similar to that of the coarse and medium fractions. This behaviour is explained by the low water content retained in these three classes above this potential, where reduced moisture and a lower proportion of pores filled with water drastically limit flow. Under the same conditions, fine sand maintained an intermediate Kr, while very fine sand had the highest Kr values among all classes. The higher conductivity in the finer fractions results from a more favourable pore size distribution, formed by the arrangement of smaller particles, and the greater influence of adsorption and capillarity mechanisms, which sustain water movement in the soil under high matric tensions.

4. Conclusions

• The grain size of sand directly influences pore diameter distribution, water retention, and soil hydraulic conductivity (Figure 7). In very fine sand (0.053–0.106 mm), pores between 0.2 and 3.0 μm predominate, which provide high water retention capacity. In fine, medium, and coarse fractions, pores with diameters between 50 and 300 μm predominate, while in very coarse sand (1.0–2.0 mm), macropores with diameters greater than 300 μm predominate, which favour drainage and aeration at the expense of water retention.

2.

The distribution of pore diameter, defined by the sand’s granulometric class, was reflected distinctly in the soil’s water behaviour. Field capacity was highest in very fine sand (0.38 m³ m⁻³) and reduced in the other fractions (≈0.05 m³ m⁻³). The permanent wilting point, in turn, remained similar across all classes (≈0.02 m³ m⁻³). As a result, available water was strongly dependent on particle size, being highest in very fine sand (0.30 m³ m⁻³), intermediate in fine sand (0.06 m³ m⁻³), and lowest in medium, coarse, and very coarse fractions (≈0.02 m³ m⁻³). These differences result from retention dynamics: in very fine sand, water is gradually removed as tension increases, while in very coarse sand, there is a high water loss already at 10 hPa.

3.

Under low matric tensions (up to 10–30 hPa), unsaturated hydraulic conductivity (Kr) was lower in very coarse sand, while the other fractions showed similar values. With increasing matric tensions, Kr declined sharply in the very coarse, coarse, medium, and fine sand classes. In contrast, very fine sand maintained the highest Kr values under matric tensions above 100 hPa, thus sustaining the highest water flow in the soil under conditions of lower water availability.

4.

The reduction in the diameter of sand particles, particularly in the very fine fraction, promotes an increase in water retention, available water, and unsaturated hydraulic conductivity, to the detriment of macroporosity and aeration capacity. It can be concluded that the water behaviour of sandy soils is significantly modulated by the granulometric composition of the sand fraction, highlighting the importance of detailed characterization of sand fractions for water management in agricultural and environmental systems.

5.

While our monodisperse system isolated the fundamental physical effects of sand grain size on water retention and hydraulic conductivity, it is important to acknowledge that soils are polydisperse (clay, silt, and sand) and structurally complex. Our findings provide the physical foundation for the sand matrix, but the inclusion of finer particles (silt and clay) and organic matter in natural soils modify these results by altering pore geometry and adding water retention mechanisms (capillarity and adsorption). The presence of these finer components would increase total water retention and change the hydraulic conductivity compared to our monodisperse quartz sand, as they contribute to pore clogging and enhance surface charge interactions. Therefore, while our results clarify the contribution of sand fractions, they should be applied to natural soil simulations as a baseline, considering the interactive effects of soil structure and clay and silt fractions.

6.

These findings regarding the effects of sand grains size have direct implications for land management in drought-prone sandy soils. Understanding these differences enables technicians and farmers to more effectively plan appropriate management systems for specific soil types, while improving irrigation efficiency and water conservation to optimize water delivery during critical dry periods.