The Effect of the Freeze–Thaw Process on Plant Available Water and Water-Stable Aggregates as a Function of Soil Tillage and Soil Chemical Quality

Abstract

The goal of this work was to determine the effect of soil freeze–thaw processes on the formation of water-stable aggregates (WSA) and plant available water (PAW) in soils of different textures, depending on the intensity of tillage: conventional tillage (CT), reduced tillage (RT) and no-tillage (NT). The WSA value (0.4%) and PAW mean (5.5%) in sandy loam were higher than in loam. The average content of WSA and PAW tended to decrease in the following order: air-dry soil > soil with water content at field capacity > soil near full saturation. These results indicate that WSA in soils that are close to full saturation upon freezing will be less stable after thawing and will decrease the PAW. The content of WSA in NT was 9.4% higher than in RT and 14% higher than in CT. The content of PAW in NT was 5.6% higher than in CT and 13.6% higher than in RT. The effects of various physical and chemical properties on PAW as a function of water level during the freeze–thaw process indicate that WSA content acted as a direct factor for PAW. In a temperate climate zone under dry meteorological conditions, NT would have a promising future for soil stability by maintaining higher WSA and PAW.

1. Introduction

Global climate change leads to thinning snow cover and increased soil moisture, changing the frequency of freezing–thawing in winter and spring, especially in regions with temperate climates [1]. Global warming affects the freezing and thawing of soils, which will significantly change the structure of soils in areas with seasonal freezing and thawing [2]. Winter conditions with seasonally frozen soils can have a profound effect on the structure and, therefore, the water-holding capacity of the soil [3].

The climate of Lithuania includes precipitation, frost, and drying of soil [4], which leads to the instability of soil aggregates and the water-holding capacity of the soil. Cambisols in temperate climate zones are among the most productive soils on earth. These soils are intensively used as agricultural lands. Cambisol covers approximately 17% of Lithuania’s territory [5].

Freeze–thaw cycling is a process of energy input and output in soil [6] and can greatly influence the arrangement and binding of soil particles and, hence, soil structure. Soil aggregates are the main unit of structure of soil [7], and perform the functions of soil stability, as well as maintaining soil nutrients and microbial activity [8,9]. Studies by Tang et al. [10] and Xiao et al. [11] showed that physical protection in aggregates is one of the key mechanisms for soil organic carbon conservation. The contents of soil organic matter and clay minerals are important in the formation of the water-stable aggregates [12]. In addition, silt and clay bind soil particles better than sand. It is also maintained that soil management practices, such as reduced tillage with crop residue conservation, can improve soil quality. Macroaggregates (>0.25 mm) provide a wider pore space and better structural stability, which play an important role in reducing soil erosion and increasing water-holding capacity [13]. In general, water-resistant aggregates and soil erosion resistance are used to characterize the stability of aggregates, with higher values indicating greater soil stability [14]. Soil organic matter, soil texture and pH influence the stability of soil aggregates [15,16].

The freezing–thawing process influences the chemical and physical properties of the soil, such as the structure of the soil, water-stable aggregates, organic matter content [17], and soil phosphorus [18]. The changing trends of soil properties are related to the freezing–thawing degree, water content, and soil texture [19]. Starkloff et al. [20] found the negative impact of the freezing–thawing process on the properties of macropores in silt and sandy soils, which, in turn, reduces water absorption by plants.

The influence of tillage on the development of agricultural crops is as diverse as that on the physical properties of the soil [21]. Agricultural crop production depends on soil structure, soil type, and weather conditions [22]. The influence of tillage methods on the development of agricultural crops is significantly dependent on local growing conditions. These conditions include the soil’s response to freezing–thawing and wetting–drying cycles [23]. Factors caused by soil cultivation are related to changes in the physical properties of the soil and affect the distribution of water in the soil and the growth of roots and, thus, the absorption of water by plants [24,25].

Freezing and thawing cycles change the internal structure of the aggregate in the form of increased porosity and more asymmetrical pores [13,26], and lead to a decrease in soil nutrient content and changes in physicochemical properties such as pH and electrical conductivity [27,28]. However, the actual influence of freezing–thawing cycles on the stability of aggregates is still poorly understood.

Soil aggregation is one of the most important indicators of soil quality; high aggregation increases porosity and, therefore, increases the water-holding capacity of the soil [29].

The aim of this study was to determine the effect of soil freeze–thaw processes on plant available water and water-stable aggregates in Cambisol in relation to basic soil chemical properties, soil texture and soil tillage intensity.

2. Materials and Methods

2.1. Soil Description and Site

This research was carried out at the Institute of Agriculture, the Lithuanian Research Centre for Agriculture and Forestry, Central Lithuania, on two long-term field trials with loam and sandy loam textures prevailing (Figure 1). The soil of the local site is classified as Cambisol (loamic, drained), according to WRB [30]. A long-term study examined different soil cultivation methods, since 2004 in loam and since 1999 in sandy loam. The experimental treatments were set up on two textural (sandy loam and loam) soils (Figure 1a). Three soil tillage (NT—no-tillage, RT—reduced tillage and CT—conventional tillage) systems were investigated in both experiments, with plant residues removed and fertilized with mineral NPK fertilizers according to the soil properties. The crop in the investigation field in 2021 was winter wheat (Triticum aestivum L.). More detailed information on soil texture and soil treatment is provided by Kochiieru et al. [31].

2.2. Sampling

Soil cores (two replications) with steel cylinders for determination of water-stable aggregates (WSA) and plant available water (PAW) were collected from both field (loam and sandy loam) experiments in 2021 (Figure 1b). Each soil sample was taken using a plastic cylinder and a steel cylinder (Figure 1c) from the middle of the soil layers at 0–10 and 10–20 cm depths of soil. The plastic cylinder (diameter—46 mm; height—51 mm) was used to freeze–thaw the soil (Figure 1c). The steel cylinder (diameter—53 mm; height—50 mm) was used to transport the plastic cylinder with the soil sample (Figure 1c).

2.3. Freezing–Thawing Process

Undisturbed monoliths of soil were adjusted for three levels of water content (air-dry soil, soil at field capacity and soil near saturation) in a lab. Soil samples with different water contents were subjected to three freezing–thawing cycles. The soil monoliths were frozen at −5 °C, and they thawed at 5 °C. A single freezing–thawing cycle was 120 h. More details on this procedure can be found in Kochiieru et al.’s study [32].

2.4. Hydro-Physical Analyses of the Soil

Undisturbed soil samples after the freezing–thawing processes were further investigated for determination of soil water retention characteristics from 0–10 and 10–20 cm depths. Water retention properties were studied at −4, −10, −30 and −100 hPa (in a sandbox) and at −300 hPa (in a sand–kaolin box). The water content was determined at −15,500 hPa of suction using sieved soil samples [33]. The water content levels at −100 hPa and–15,500 hPa were considered as the field capacity (in Europe) and the permanent wilting point, respectively. The amount of water between these two suctions was regarded as the plant-available water content (the water-holding capacity of the soil).

2.5. Water-Stable Aggregate (WSA) Analysis

After the freezing–thawing cycles, WSAs were estimated in soil samples with different water contents from different layers of soil. Air-dry soil was sieved using a Retsch sieve shaker (Retsch AS200 basic, Retsch GmbH, Haan, Germany). Water-stable aggregates were determined from 0–10 and 10–20 cm depths by using a wet-sieving apparatus (Eijkelkamp Agrisearch Equipment, Giesbeek, The Netherlands). Four replications of 4 g of air-dried soil aggregates (1–2 mm size) were wet-sieved (0.25 mm mesh size) in distilled water, and then the stable aggregates were destroyed by a 0.2% NaOH solution, oven-dried at 105 °C for 24 h and weighed [15]. More details are provided by Kochiieru et al. [32].

2.6. Agrochemical Analysis

Soil samples (two field replications) were collected from (0–10 and 10–20 cm) soil layers randomly by using a steel auger (Figure 1c) at the same place as the samples for water-stable aggregates and plant available water content. The disturbed soil samples were air-dried; visible roots and plant residues were manually removed. Then, the samples were crushed, sieved through a 2 mm sieve and homogeneously mixed. Three replicate samples were used for each field sample in the laboratory analysis. Soil organic carbon (SOC) was identified by the titrimetric (classical) method of Tyurin [34]. Total nitrogen (N) was determined by the Dumas method using a Vario EL III CNS autoanalyzer [35]. Soil pH_KCl was determined using an AS-3010D potentiometer in 1M KCl (1:2.5, w:v). Available potassium and phosphorus were identified by the Egner–Riehm–Domingo (A–L) method [36].

2.7. Meteorological Conditions

The annual mean air temperature was 7.5 °C, and the mean air temperature during the growing season was 14.7 °C. The total amount of rainfall during the 2021 crop-growing period was 357.4 mm, with a total annual precipitation of 565.9 mm (data from the Dotnuva Meteorological Station).

2.8. Statistical Analysis

The statistical software package SAS 7.1 was used to calculate mean values and standard errors. Mean values were compared by Duncan’s multiple range tests at a probability level of p < 0.05. Correlation–regression analysis was also implemented. The standard error values are presented as error bars. A path analysis for statistical data treatment was performed [37]. The principal diagram for the path coefficient analysis is presented in Figure 2.

Px_iy represents the path coefficient that shows the direct effect of an independent variable x_i on the dependent variable Y. rx_ix_j represents the simple indirect correlation between two independent variables x_i and x_j. The Y demonstrates the correlation between Y and x_i, i.e., the sum of the entire path (sum of effects) connecting the two variables (more details are provided in Kochiieru et al.’s study [32]).

3. Results

3.1. Content of Soil Water-Stable Aggregates

The average content of water-stable aggregates was higher in sandy loam than in loam but was not significantly different at p < 0.05 (

Table 1. Water-stable aggregates (±standard error) in different soil textures, tillage systems, soil depths and water contents at freezing.

Factors	Soil Texture	Tillage System	Soil Layer (cm)	Water Content at Freezing	Water-Stable Aggregates (%)	F	Pr > F
Soil texture	Sandy loam				60.43 a ± 0.75	0.05	0.8201
	Loam				60.20 a ± 0.64
Tillage system		No-tillage			65.43 a ± 0.54	37.42	0.0001
		Reduced tillage			59.26 b ± 0.90
		Conventional tillage			56.26 c ± 0.81
Soil layer (cm)			0–10		62.63 a ± 0.64	23.59	0.0001
			10–20		58.01 b ± 0.70
Water content at freezing				Air-dry	61.03 a ± 0.83	0.58	0.5625
				Field capacity	60.14 a ± 0.93
				Near full saturation	59.77 a ± 0.80

Water-stable aggregate data followed by the same letters are not significantly different at p < 0.05.

). The content of WSA, averaged across soil textures, soil depths and water contents at freezing, in no-tillage (NT) soil was 9.4% higher than in reduced tillage (RT) and 14.0% higher than in conventional tillage (CT) (Table 1).

The mean WSA values were higher at the 0–10 cm soil depth than at the 10–20 cm soil depth (Table 1). The content of WSA, averaged across soil textures, tillage systems, and soil depths, tended to increase in the following order: NS (59.77%) < FC (60.14%) < AD (61.03%). These results indicate that aggregates in soils that are close to full saturation upon freezing will be less stable after thawing. However, the effect of water content on the averaged WSA data during freezing was insignificant (Table 1).

3.2. The Effect of Tillage Systems on Water-Stable Aggregate Content

The effect of the tillage system was significant (p < 0.01) on the WSA content at different water levels under freezing–thawing processes in different soil textures (Figure 3 and

Table 2. The effect of tillage systems on WSA content at different water levels under freezing–thawing processes in different soil textures (data averaged across soil depths).

Tillage System	Water-Stable Aggregates After Freezing–Thawing at Different Water Content (%)
	WSAAD		WSAFC		WSANS
	Sandy Loam	Loam	Sandy Loam	Loam	Sandy Loam	Loam
No-tillage	67.23 a ± 1.51	67.35 a ± 0.80	68.07 a ± 1.30	63.91 a ± 1.45	63.41 a ± 1.15	62.59 a ± 1.04
Reduced tillage	57.69 b ± 1.68	58.14 b ± 2.21	62.01 b ± 2.05	55.00 b ± 2.84	62.76 a ± 2.24	59.98 a ± 1.71
Conventional tillage	54.05 b ± 2.08	61.75 b ± 1.19	59.94 b ± 1.85	51.94 b ± 1.44	48.71 b ± 1.69	61.16 a ± 0.87
F	14.71	9.28	5.73	9.46	22.53	1.07
Pr > F	0.0001	0.0004	0.0061	0.0004	0.0001	0.3514

Water-stable aggregate data followed by the same letters are not significantly different at p < 0.05. WSA—water-stable aggregate. WSA_AD—air-dry WSA, WSA_FC—WSA at field capacity, and WSA_NS—WSA near full saturation.

). The decrease in WSAs was related to the soil texture and tillage system. The content of WSA in sandy loam was significantly lower than in loam at two soil water (AD and NS) levels. But the content of WSA at FC in sandy loam was higher than in loam (Table 2). In the soil of sandy loam, at AD, FC and NS water contents, conventional tillage decreased the WSA content by 1.24-, 1.14- and 1.30-fold, and in the soil of loam by 1.09-, 1.23- and 1.02-fold, respectively, compared with water-stable aggregates in no-tillage (Table 2). The water-stable aggregate content tended to decrease in the following order: NT (66.24%) > RT (60.82%) > CT (54.23%) in sandy loam, and NT (64.62%) > CT (58.28%) > RT (57.71%) in loam (Table 2).

3.3. Soil Chemical Properties

The soil organic carbon (SOC) content within the 0–20 cm soil layer averaged from 7.8 to 8.6 g kg⁻¹ in sandy loam and from 11.1 to 12.5 g kg⁻¹ in loam. The content of SOC in the soil of sandy loam was 29.9% lower than in the soil of loam. The content of SOC consistently decreased in the downward direction of soil layers in sandy loam (NT, RT, and CT) and in loam (NT and RT), while in CT, the loam content of SOC was higher at the 10–20 cm soil depth (

Table 3. Soil chemical properties (±standard error) in relation to different soil textures, tillage systems, and soil layers.

Soil Texture	Tillage	Soil Layer	pHKCl	Soil Organic Carbon (SOC)	Total Nitrogen (N)	Available Phosphorus (P)	Available Potassium (K)	C/N
		cm		g kg−1
Sandy loam	NT	0–10	5.4 ± 0.4	9.2 ± 0.8	1.13 ± 0.17	0.284 ± 0.05	0.249 ± 0.03	8.4 ± 2.0
		10–20	6.0 ± 0.5	8.1 ± 1.1	0.87 ± 0.19	0.148 ± 0.03	0.114 ± 0.02	9.5 ± 0.8
		Mean: 0–20	5.7 ± 0.3	8.6 ± 0.6	1.00 ± 0.13	0.216 ± 0.05	0.181 ± 0.04	9.0 ± 0.9
	RT	0–10	5.8 ± 0.1	8.5 ± 0.5	0.97 ± 0.03	0.366 ± 0.04	0.326 ± 0.06	8.7 ± 0.3
		10–20	6.2 ± 0.1	7.1 ± 0.6	0.85 ± 0.04	0.181 ± 0.01	0.131 ± 0.03	8.4 ± 0.3
		Mean: 0–20	6.0 ± 0.1	7.8 ± 0.5	0.91 ± 0.04	0.273 ± 0.06	0.228 ± 0.06	8.6 ± 0.2
	CT	0–10	5.4 ± 0.1	8.3 ± 0.7	1.00 ± 0.07	0.189 ± 0.01	0.168 ± 0.01	8.2 ± 0.1
		10–20	5.6 ± 0.4	8.1 ± 0.3	0.89 ± 0.07	0.154 ± 0.01	0.166 ± 0.01	9.2 ± 1.0
		Mean: 0–20	5.5 ± 0.2	8.2 ± 0.3	0.94 ± 0.05	0.171 ± 0.01	0.167 ± 0.01	8.7 ± 0.5
Sandy loam (average)			5.7 ± 0.1	8.2 ± 0.3	0.95 ± 0.04	0.220 ± 0.03	0.192 ± 0.02	8.8 ± 0.3
Loam	NT	0–10	6.8 ± 0.0	13.0 ± 0.5	1.54 ± 0.13	0.379 ± 0.13	0.356 ± 0.05	8.5 ± 0.4
		10–20	7.0 ± 0.0	12.1 ± 0.9	1.39 ± 0.21	0.361 ± 0.16	0.169 ± 0.05	8.8 ± 0.7
		Mean: 0–20	6.9 ± 0.1	12.5 ± 0.5	1.46 ± 0.11	0.370 ± 0.08	0.263 ± 0.06	8.6 ± 0.3
	RT	0–10	6.9 ± 0.1	12.2 ± 1.1	1.45 ± 0.09	0.337 ± 0.02	0.324 ± 0.01	8.4 ± 0.2
		10–20	7.1 ± 0.0	10.7 ± 0.7	1.27 ± 0.17	0.328 ± 0.01	0.157 ± 0.03	8.5 ± 0.6
		Mean: 0–20	7.0 ± 0.1	11.4 ± 0.7	1.36 ± 0.09	0.332 ± 0.01	0.241 ± 0.05	8.4 ± 0.3
	CT	0–10	6.7 ± 0.1	10.4 ± 0.1	1.13 ± 0.08	0.266 ± 0.01	0.201 ± 0.03	9.3 ± 0.7
		10–20	6.8 ± 0.1	11.7 ± 1.2	1.29 ± 0.11	0.265 ± 0.01	0.159 ± 0.01	9.2 ± 1.7
		Mean: 0–20	6.7 ± 0.1	11.1 ± 0.6	1.21 ± 0.07	0.265 ± 0.01	0.180 ± 0.02	9.2 ± 0.8
Loam (average)			6.9 ± 0.0	11.7 ± 0.4	1.34 ± 0.06	0.332 ± 0.03	0.228 ± 0.03	8.8 ± 0.3

NT—no-tillage, RT—reduced tillage and CT—conventional tillage.

). The average across depth contents of SOC tended to decrease in the following orders: NT (8.6 g kg⁻¹) > CT (8.2 g kg⁻¹) > RT (7.8 g kg⁻¹) in sandy loam, and NT (12.5 g kg⁻¹) > RT (11.4 g kg⁻¹) > CT (11.1 g kg⁻¹) in loam (Table 3).

The content of total nitrogen (N) within the 0–20 cm soil layer amounted to 0.91–1.00 g kg⁻¹ in sandy loam and 1.21–1.46 g kg⁻¹ in loam. The soil total N content in the soil of sandy loam was 29.1% lower than in the soil of loam (Table 3). The content of total N consistently decreased in the downward direction of soil layers in sandy loam (NT, RT, and CT) and in loam (NT and RT), while in CT, the loam content of total N was higher at the 10–20 cm soil depth (Table 3). The average across depths of total N tended to decrease in the following orders: NT (1.00 g kg⁻¹) > CT (0.94 g kg⁻¹) > RT (0.91 g kg⁻¹) in sandy loam, and NT (1.46 g kg⁻¹) > RT (1.36 g kg⁻¹) > CT (1.21 g kg⁻¹) in loam (Table 3).

The content of available phosphorus (P) within the 0–20 cm soil layer amounted to 0.171–0.273 g kg⁻¹ in sandy loam and 0.265–0.370 g kg⁻¹ in loam. The content of available P in the soil of sandy loam was 33.7% lower than in the soil of loam (Table 3). The content of available P consistently decreased in the downward direction of soil layers in different tillage systems of sandy loam and loam. The average across depth of available P tended to decrease in the following orders: RT (0.273 g kg⁻¹) > NT (0.216 g kg⁻¹) > CT (0.171 g kg⁻¹) in sandy loam, and NT (0.370 g kg⁻¹) > RT (0.332 g kg⁻¹) > CT (0.265 g kg⁻¹) in loam (Table 3).

The soil available potassium (K) content in sandy loam was 15.8% lower than in loam. The content of K within the 0–20 cm soil layer amounted to 0.167–0.228 g kg⁻¹ in sandy loam and 0.180–0.263 g kg⁻¹ in loam (Table 3). The content of available K consistently decreased in the downward direction of soil layers in different tillage systems of sandy loam and loam. The average across depth of available K tended to decrease in the following orders: RT (0.228 g kg⁻¹) > NT (0.181 g kg⁻¹) > CT (0.167 g kg⁻¹) in sandy loam, and NT (0.263 g kg⁻¹) > RT (0.241 g kg⁻¹) > CT (0.180 g kg⁻¹) in loam (Table 3).

The pH values observed at the 0–20 cm soil depth indicated a moderately acidic pH in sandy loam (5.4–6.2 pH) and a neutral pH in loam (6.7–7.1 pH). The soil pH in loam was 17.4% higher than in sandy loam (Table 3).

3.4. Content of Plant Available Water

The plant available water content was higher in sandy loam than in loam, but was not significantly different at p < 0.05 (

Table 4. Plant available water (±standard error) in different soil textures, tillage systems, soil layers and water contents at freezing.

Factor	Soil Texture	Tillage System	Soil Layer (cm)	Water Content at Freezing	Plant Available Water (m3 m−3)	F	Pr > F
Soil texture	Sandy loam				0.156 a ± 0.003	2.73	0.1031
	Loam				0.148 a ± 0.004
Tillage system		No-tillage			0.162 a ± 0.003	7.43	0.0012
		Reduced tillage			0.140 b ± 0.004
		Conventional tillage			0.153 a ± 0.005
Soil layer (cm)			0–10		0.151 a ± 0.004	0.02	0.8756
			10–20		0.152 a ± 0.003
Water content at freezing				Air-dry	0.159 a ± 0.005	5.22	0.0078
				Field capacity	0.155 a ± 0.004
				Near full saturation	0.141 b ± 0.004

Plant available water data followed by the same letters are not significantly different at p < 0.05.

). The content of PAW, averaged across soil textures, soil layers and water contents at freezing, in NT was 5.6% higher than in CT and 13.6% higher than in RT (Table 4).

The mean PAW was the same in the 0–10 cm soil layer and in the 10–20 cm soil layer (Table 4). The PAW content, averaged across soil textures, tillage systems, and soil layers, tended to decrease in the following order: AD (0.159 m³ m⁻³) > FC (0.155) > NS (0.141 m³ m⁻³) (Table 4). The effect of the tillage system and water content at freezing was significant (p < 0.01) on the PAW content (Table 4).

3.5. The Influence of Tillage Systems on Plant Available Water (PAW) Content

The effect of the tillage system was significant (p < 0.01) on the PAW content at different water levels under freezing–thawing processes in different soil textures (Figure 4 and

Table 5. The influence of tillage systems on plant available water contents at different water levels after freezing–thawing processes in different soil textures (data averaged across soil layers).

Tillage System	Plant Available Water (PAW) After Freezing–Thawing at Different Water Contents (m3 m−3)
	PAWAD		PAWFC		PAWNS
	Sandy Loam	Loam	Sandy Loam	Loam	Sandy Loam	Loam
NT	0.162 ab ± 0.006	0.172 a ± 0.009	0.165 a ± 0.004	0.171 a ± 0.008	0.145 a ± 0.003	0.158 a ± 0.006
RT	0.142 b ± 0.003	0.143 a ± 0.007	0.166 a ± 0.008	0.129 b ± 0.008	0.140 a ± 0.009	0.122 b ± 0.012
CT	0.176 a ± 0.012	0.161 a ± 0.016	0.157 a ± 0.004	0.142 b ± 0.010	0.151 a ± 0.007	0.131 ab ± 0.010
			Contrasts
NT vs. CT	−0.014 *	0.011 ns	0.008 ns	0.029 *	−0.006 ns	0.027 *
NT vs. RT	0.020 **	0.029 *	−0.001 ns	0.042 **	0.004 ns	0.036 **
RT vs. CT	−0.034 **	−0.018 ns	0.009 ns	−0.013 ns	−0.011 ns	−0.009 ns

Plant available water (±standard error) data followed by the same letters are not significantly different at p < 0.05. PAW_AD—air-dry PAW, PAW_FC—PAW at field capacity, PAW_NS—PAW near full saturation, NT—no-tillage, RT—reduced tillage, and CT—conventional tillage. * p < 0.05; ** p < 0.01; and ^ns not significant.

). The decrease in the PAW content was related to the soil texture and tillage system. The content of PAW in sandy loam was significantly higher than in loam in the two tillage (RT and CT) systems, while the PAW under NT in sandy loam was lower than in loam (Table 5).

The average across water contents of the PAW content tended to decrease in the following order: CT (0.161 m³ m⁻³) > NT (0.157 m³ m⁻³) > RT (0.149 m³ m⁻³) in sandy loam, and NT (0.167 m³ m⁻³) > CT (0.145 m³ m⁻³) > RT (0.131 m³ m⁻³) in loam. In sandy loam, under NT, RT and CT, NS decreased the PAW content by 1.12-, 1.01- and 1.17-fold, and in the soil of loam by 1.09-, 1.17- and 1.23-fold, respectively, compared with the PAW content at AD (Table 5).

3.6. The Effect of Water Content During Freezing–Thawing Processes on Plant Available Water (PAW) Content

The decrease in plant available water (PAW) was related to the water content during freezing–thawing processes in different tillage systems and soil textures. The content of PAW in sandy loam was higher than in loam at the three soil water levels. In the soil of sandy loam, under NT, RT and CT, NS decreased the PAW content by 1.12-, 1.01- and 1.17-fold, and in the soil of loam by 1.09-, 1.17- and 1.23-fold, respectively, compared with the PAW content at AD (Table 5). The average across tillage systems of the PAW content tended to decrease in the following order: FC (0.163 m³ m⁻³) > AD (0.160 m³ m⁻³) > NS (0.145 m³ m⁻³) in sandy loam, and AD (0.159 m³ m⁻³) > FC (0.147 m³ m⁻³) > NS (0.137 m³ m⁻³) in loam (Table 5). These results indicate that soils that are close to full saturation upon freezing will have a lower content of PAW after thawing (Table 5).

3.7. The Effect of Soil Chemical Properties on Water-Stable Aggregates

Water-stable aggregates (WSA) positively correlated with the chemical properties of soil in loam (

Table 6. Correlation matrix of soil quality parameters at different water contents during the freeze–thaw processes in loam.

Parameters	NT	RT	CT	WSA	PAW	SOC	N	P	K
Air-dry soil
WSA—water-stable aggregates (%)	67.35	58.14	61.75
PAW—plant available water (m3 m−3)	0.172	0.143	0.161	0.48 *
SOC—soil organic carbon (g kg−1)	12.53	11.40	11.05	0.56 *	0.78 **
N—total nitrogen (g kg−1)	1.46	1.36	1.21	0.40 ns	0.55 *	0.94 **
P—available phosphorus (g kg−1)	0.370	0.332	0.265	0.46 *	0.18 ns	0.69 *	0.85 **
K—available potassium (g kg−1)	0.263	0.241	0.180	0.38 ns	0.28 ns	0.68 *	0.73 **	0.54 *
C/N—carbon/nitrogen ratio	8.64	8.42	9.25	0.09 ns	0.16 ns	−0.44 ns	−0.71 **	−0.81 **	−0.56 *
Soil at field capacity
WSA—water-stable aggregates (%)	63.91	55.00	51.94
PAW—plant available water (m3 m−3)	0.171	0.129	0.142	0.72 **
SOC—soil organic carbon (g kg−1)	12.53	11.40	11.05	0.73 **	0.49 *
N—total nitrogen (g kg−1)	1.46	1.36	1.21	0.76 **	0.38 ns	0.94 **
P—available phosphorus (g kg−1)	0.370	0.332	0.265	0.83 **	0.51 *	0.69 *	0.85 **
K—available potassium (g kg−1)	0.263	0.241	0.180	0.75 **	0.17 ns	0.68 *	0.73 **	0.54 *
C/N—carbon/nitrogen ratio	8.64	8.42	9.25	−0.51 *	0.03 ns	−0.44 ns	−0.71 **	−0.81 **	−0.56 *
Soil near full saturation
WSA—water-stable aggregates (%)	62.59	59.98	61.16
PAW—plant available water (m3 m−3)	0.157	0.121	0.131	−0.10 ns
SOC—soil organic carbon (g kg−1)	12.53	11.40	11.05	0.65 *	0.11 ns
N—total nitrogen (g kg−1)	1.46	1.36	1.21	0.52 *	0.10 ns	0.94 **
P—available phosphorus (g kg−1)	0.370	0.332	0.265	0.23 ns	0.42 *	0.69 *	0.85 **
K—available potassium (g kg−1)	0.263	0.241	0.180	0.81 **	0.03 ns	0.68 *	0.73 **	0.54 *
C/N—carbon/nitrogen ratio	8.64	8.42	9.25	−0.08 ns	−0.01 ns	−0.44 ns	−0.71 **	−0.81 **	−0.56 *

NT—no-tillage, RT—reduced tillage, and CT—conventional tillage. * p < 0.05; ** p < 0.01; and ^ns not significant.

) and sandy loam (

Table 7. Correlation matrix of soil quality parameters at different water contents during the freeze–thaw processes in sandy loam.

Parameters	NT	RT	CT	WSA	PAW	SOC	N	P	K
Air-dry soil
WSA—water-stable aggregates (%)	67.23	57.69	54.05
PAW—plant available water (m3 m−3)	0.162	0.142	0.176	−0.16 ns
SOC—soil organic carbon (g kg−1)	8.63	7.78	8.18	0.49 *	0.18 ns
N—total nitrogen (g kg−1)	1.00	0.91	0.94	0.30 ns	0.19 ns	0.88 **
P—available phosphorus (g kg−1)	0.216	0.273	0.171	0.24 ns	−0.61 *	0.54 *	0.58 *
K—available potassium (g kg−1)	0.181	0.228	0.167	0.14 ns	−0.50 *	0.63 *	0.60 *	0.97 **
C/N—carbon/nitrogen ratio	8.96	8.57	8.74	0.35 ns	−0.03 ns	−0.06 ns	−0.51 *	−0.34 ns	−0.27 ns
Soil at field capacity
WSA—water-stable aggregates (%)	68.07	62.01	59.94
PAW—plant available water (m3 m−3)	0.165	0.166	0.157	0.18 ns
SOC—soil organic carbon (g kg−1)	8.63	7.78	8.18	0.55 *	−0.39 ns
N—total nitrogen (g kg−1)	1.00	0.91	0.94	0.71 **	−0.34 ns	0.88 **
P—available phosphorus (g kg−1)	0.216	0.273	0.171	0.31 ns	−0.11 ns	0.54 *	0.58 *
K—available potassium (g kg−1)	0.181	0.228	0.167	0.18 ns	−0.25 ns	0.63 *	0.60 *	0.97 **
C/N—carbon/nitrogen ratio	8.96	8.57	8.74	−0.38 ns	0.16 ns	−0.06 ns	−0.51 *	−0.34 ns	−0.27 ns
Soil near full saturation
WSA—water-stable aggregates (%)	63.41	62.77	48.71
PAW—plant available water (m3 m−3)	0.146	0.139	0.151	−0.60 *
SOC—soil organic carbon (g kg−1)	8.63	7.78	8.18	0.49 *	0.09 ns
N—total nitrogen (g kg−1)	1.00	0.91	0.94	0.46 *	0.30 ns	0.88 **
P—available phosphorus (g kg−1)	0.216	0.273	0.171	0.83 **	−0.52 *	0.54 *	0.58 *
K—available potassium (g kg−1)	0.181	0.228	0.167	0.71 **	−0.48 *	0.63 *	0.60 *	0.97 **
C/N—carbon/nitrogen ratio	8.96	8.57	8.74	−0.08 ns	−0.40 ns	−0.06 ns	−0.51 *	−0.34 ns	−0.27 ns

* p < 0.05; ** p < 0.01; and ^ns not significant. NT—no-tillage, RT—reduced tillage, and CT—conventional tillage.

) at different water contents during the freeze–thaw processes.

Soil organic carbon (SOC) content positively correlated with WSAs (r = 0.56 * (AD), r = 0.73 ** (FC) and r = 0.65 * (NS); Table 6) in loam. In sandy loam, SOC correlated with WSA (r = 0.49 * (AD), r = 0.55 * (FC) and r = 0.49 * (NS); Table 7). Total nitrogen (N) content positively correlated with WSA (r = 0.40 ^ns (AD), r = 0.76 ** (FC) and r = 0.52 * (NS); Table 6) in loam. In sandy loam, total N positively correlated with WSA (r = 0.30 ^ns (AD), r = 0.71 ** (FC) and r = 0.46 * (NS); Table 7) Available phosphorus (P) content positively correlated with WSA (r = 0.46 * (AD), r = 0.83 ** (FC) and r = 0.23 ^ns (NS); Table 6) in loam. The content of P correlated with WSA (r = 0.24 ^ns (AD), r = 0.31 ^ns (FC) and r = 0.83 ** (NS); Table 7) in sandy loam. Available potassium (K) content positively correlated with WSA (r = 0.38 ^ns (AD), r = 0.75 ** (FC) and r = 0.81 ** (NS); Table 6) in loam. The content of K positively correlated with WSA (r = 0.14 ^ns (AD), r = 0.18 ^ns (FC) and r = 0.71 ** (NS); Table 7) in sandy loam.

3.8. Relationships of Dependent Variable (PAW) with Physical and Chemical Properties of Soil at Different Water Contents During Freezing–Thawing

Plant available water (PAW) content was influenced positively or negatively by the properties of the soil at different water contents during the freeze–thaw processes. In loam, the content of PAW positively correlated with WSA (r = 0.48 * (AD) and r = 0.72 ** (FC); Table 6), while the result in NS was inverse (r = −0.10 ^ns; Table 6). The content of WSA acted most strongly as a direct factor for PAW (Px₁y = −2.219 (AD), Px₁y = 4.893 (FC), and Px₁y = −1.404 (NS);

Table 8. Pathways of plant available water (PAW) response to WSA, SOC, N, P, K and C/N in loam.

Parameters	WSA	SOC	N	P	K	C/N	PAW (rY)
Air-dry soil
WSA—water-stable aggregates (%)	−2.219	1.090	−0.639	1.568	0.371	0.315	0.48 *
SOC—soil organic carbon (g kg−1)	−1.235	1.958	−1.495	2.381	0.671	−1.504	0.78 **
N—total nitrogen (g kg−1)	−0.895	1.846	−1.586	2.914	0.716	−2.444	0.55 *
P—available phosphorus (g kg−1)	−1.011	1.354	−1.342	3.442	0.531	−2.792	0.18 ns
K—available potassium (g kg−1)	−0.839	1.339	−1.158	1.862	0.981	−1.910	0.28 ns
C/N—carbon/nitrogen ratio	−0.204	−0.858	1.129	−2.799	−0.546	3.432	0.16 ns
Soil at field capacity
WSA—water-stable aggregates (%)	4.893	−1.117	2.879	−4.252	−2.385	0.699	0.72 **
SOC—soil organic carbon (g kg−1)	3.559	−1.536	3.581	−3.530	−2.188	0.599	0.49 *
N—total nitrogen (g kg−1)	3.709	−1.448	3.798	−4.319	−2.335	0.975	0.38 ns
P—available phosphorus (g kg−1)	4.077	−1.062	3.215	−5.102	−1.730	1.113	0.51 *
K—available potassium (g kg−1)	3.648	−1.051	2.773	−2.761	−3.198	0.762	0.17 ns
C/N—carbon/nitrogen ratio	−2.499	0.673	−2.704	4.150	1.780	−1.369	0.03 ns
Soil near full saturation
WSA—water-stable aggregates (%)	−1.404	0.409	−0.824	0.499	1.352	−0.130	−0.10 ns
SOC—soil organic carbon (g kg−1)	−0.913	0.628	−1.482	1.476	1.149	−0.752	0.11 ns
N—total nitrogen (g kg−1)	−0.736	0.592	−1.572	1.806	1.226	−1.222	0.10 ns
P—available phosphorus (g kg−1)	−0.329	0.435	−1.330	2.134	0.908	−1.396	0.42 *
K—available potassium (g kg−1)	−1.131	0.430	−1.148	1.154	1.679	−0.955	0.03 ns
C/N—carbon/nitrogen ratio	0.106	−0.275	1.119	−1.735	−0.934	1.716	−0.01 ns

* p < 0.05; ** p < 0.01; and ^ns not significant. Number in bold—direct effect; underlined number—dominant effect.

) in loam.

In sandy loam, WSA acted as a direct factor for PAW (r = 0.18 ^ns and Px₁y = 7.266 (FC); Table 7 and

Table 9. Pathways of plant available water (PAW) response to WSA, SOC, N, P, K and C/N in sandy loam.

Parameters	WSA	SOC	N	P	K	C/N	PAW (rY)
Air-dry soil
WSA—water-stable aggregates (%)	0.330	2.237	−1.355	−0.510	0.089	−0.947	−0.16 ns
SOC—soil organic carbon (g kg−1)	0.162	4.554	−3.930	−1.174	0.408	0.156	0.18 ns
N—total nitrogen (g kg−1)	0.100	4.010	−4.463	−1.245	0.391	1.393	0.19 ns
P—available phosphorus (g kg−1)	0.078	2.473	−2.570	−2.163	0.632	0.941	−0.61 *
K—available potassium (g kg−1)	0.045	2.851	−2.677	−2.095	0.652	0.726	−0.50 *
C/N—carbon/nitrogen ratio	0.114	−0.260	2.275	0.745	−0.173	−2.733	−0.03 ns
Soil at field capacity
WSA—water-stable aggregates (%)	7.266	−2.417	−2.684	−5.772	3.818	−0.027	0.18 ns
SOC—soil organic carbon (g kg−1)	3.965	−4.430	−3.328	−10.190	13.592	−0.004	−0.39 ns
N—total nitrogen (g kg−1)	5.160	−3.900	−3.780	−10.806	13.023	−0.036	−0.34 ns
P—available phosphorus (g kg−1)	2.234	−2.405	−2.176	−18.769	21.030	−0.024	−0.11 ns
K—available potassium (g kg−1)	1.278	−2.773	−2.267	−18.181	21.710	−0.019	−0.25 ns
C/N—carbon/nitrogen ratio	−2.786	0.252	1.927	6.465	−5.765	0.070	0.16 ns
Soil near full saturation
WSA—water-stable aggregates (%)	−2.814	−0.175	0.949	5.724	−4.181	−0.103	−0.60 *
SOC—soil organic carbon (g kg−1)	−1.376	−0.358	1.828	3.766	−3.696	−0.070	0.09 ns
N—total nitrogen (g kg−1)	−1.286	−0.315	2.076	3.994	−3.541	−0.624	0.30 ns
P—available phosphorus (g kg−1)	−2.322	−0.195	1.195	6.937	−5.718	−0.422	−0.52 *
K—available potassium (g kg−1)	−1.993	−0.224	1.245	6.719	−5.903	−0.325	−0.48 *
C/N—carbon/nitrogen ratio	0.238	0.020	−1.058	−2.389	1.568	1.224	−0.40 ns

* p < 0.05; and ^ns not significant. Number in bold—direct effect; underlined number—dominant effect.

), while AD and NS were indirect factors for PAW (Table 9). The content of available P in loam positively correlated with PAW (r = 0.18 ^ns (AD), r = 0.51 * (FC), and r = 0.42 * (NS); Table 6), while the result in sandy loam was inverse (r = −0.61 * (AD), r = −0.11 ^ns (FC), and r = −0.52 * (NS); Table 7). The content of available P acted as a direct factor for PAW (Px₄y = 3.442 (AD), Px₄y = −5.102 (FC), and Px₄y = 2.134 (NS); Table 8) in loam. In sandy loam, WSA acted as a direct factor for PAW (r = −0.52 * and Px₄y = 6.937 (NS); Table 7 and Table 9), while AD (r = −0.61 *) and FC (r = −0.11 ^ns) were indirect factors for PAW (Table 9).

The contents of soil organic carbon, total nitrogen and available potassium acted as indirect factors for PAW in loam and sandy loam (Table 6, Table 7, Table 8 and Table 9).

4. Discussion

Soil aggregates are the basic unit of soil structure [7] and are a key factor for the quality of the soil [38]. The response of water-stable aggregates in the soil to freeze–thaw processes depends on some factors, including the water content during freezing–thawing cycles, the contents of soil organic carbon, total nitrogen, available potassium, and available phosphorus, soil texture, and tillage [18]. The available water-holding capacity of soil depends on water-stable aggregates in the soil. Because soil water-holding capacity is influenced by many factors, path data analysis is used to reveal the actual contribution of the several factors examined within the soil texture under different tillage treatments. Debnath et al. [39] and Deb et al. [40] found that organic carbon, clay, sand, and porosity contents directly influence soil water-holding capacity in sloppy land.

The soil structure has been affected by long-term anthropogenic activities. Intensive tillage reduces aggregate stability, nitrogen content, and the content of soil organic carbon, which, in turn, affects soil quality [18,41]. In turn, changes in the relationships between soil properties showed that tillage had an impact on aggregation and, accordingly, on the water-holding capacity of the soil. These changes depended on the type of soil texture and the direction of climate change. Our research results showed that loam is more resistant to drought conditions in dry spells. It demonstrates a higher readiness to perform tillage activity, providing a better water stability of aggregates in the dry soil state compared with sandy loam. Therefore, if dry climatic conditions were to prevail in the temperate climate zone, reducing the tillage intensity would be a promising way to ensure soil sustainability by maintaining water-stable dry soil aggregates. The soil response to freeze–thaw experiments showed that aggregate stability under excess soil moisture conditions depends on the type of soil texture and tillage type. Conventional tillage produced a higher water-stable aggregate (soil near full saturation) content in loam (61.2%) than in sandy loam (58.3%). However, no-tillage demonstrated a higher number of water-stable aggregates (soil near full saturation) than conventional tillage in loam and sandy loam. The high content of water-stable aggregates under full soil saturation may explain how the soil deals with waterlogging in situations with an excess of water. In such situations, sandy loam has the potential to perform better in supporting water-stable aggregates (soil at field capacity), which have a higher content than in loamy textured soil. It should be noted that the long-term no-tillage management resulted in a higher water-stable aggregate content than under conventional tillage application [42].

Skvortsova et al. [43] reported that the content of organic matter and differences in the mechanical strength of the soil affected the stability of aggregates during freezing–thawing, which confirms our results. The stability of soil water-stable aggregates is associated with the composition of soil organic matter as one of the major factors in aggregate formation and stability [44]. The positive correlation between water-stable aggregates and soil organic carbon (r = 0.49 *–0.73 **; Table 6 and Table 7) demonstrates the dominant effect of the contribution of soil organic carbon to the formation of stability of aggregates. Kochiieru et al. [32] found positive relationships (r = 0.87 **–0.95 **) between soil organic carbon and water-stable aggregates with different water contents during freezing–thawing processes at a 0–10 cm soil depth in Cambisol and Retisol. Haydu-Houdeshell et al. [29] reported that soil organic carbon and aggregate stability were moderately correlated (r² = 0.67), and Erktan et al. [45] found a positive linear correlation between these parameters, which confirms our results. Kochiieru et al. [32,42] have documented that intensive soil tillage reduces the stability of aggregates, soil organic carbon, and nitrogen content, and that, in turn, influences the quality of soil. Soil chemical properties such as total nitrogen, available phosphorus, and available potassium also influence soil aggregate formation and stability, which confirms our results (Table 6 and Table 7).

Tillage affects the structure of the soil and, consequently, the development of plants and the efficiency of the soil water regime and water use [21,46]. The results from the long-term field experiments showed better plant available water contents under no-tillage as well (Table 6). The decrease in plant available water was related to water content during freezing–thawing processes under different tillage applications in contrasting soil textures. The average across tillage systems of plant available water content tended to decrease from air-dry soil and soil with water content at field capacity to soil near full saturation. These results indicate that the plant available water content in soils that are close to full saturation upon freezing will be lower after thawing (Table 8).

Debnath et al. [39] and Deb et al. [40] reported that water-holding capacity and soil organic carbon were positively correlated (r = 0.81 ** and r = 0.80 **). Bhavya et al. [47] found that water-holding capacity increased with an increasing content of organic carbon in the soil. The water-holding capacity depends on the physical and chemical properties of the soil, just as the plant available water depends on these parameters. Soil chemical parameters had a positive effect on plant available water in loam (Table 6). However, a negative relationship was observed between these parameters in sandy loam (Table 7). According to Deb et al. [40] and Debnath et al. [39], the amount of sand negatively affects the water-holding capacity. This confirms our results.

The results also showed that some additional factors that had not been included in this study could influence the plant available water content of the studied soil. One such factor is soil macroporosity, which, in turn, influences water cycle processes such as nutrient transport and infiltration. This suggests that the relationship between the macropore network, nutrients and organic carbon in the soil may lead to differences in soil behavior under different land uses [42]. Wang and Xu [48] wrote that changes in pore structure induced by freeze–thaw processes could contribute to soil organic carbon protection of aggregates. Microbial communities regulate soil organic carbon stability, substantially contributing to soil organic carbon content [49]. These missing factors will be explored in future studies.

If dry meteorological conditions were to prevail in the temperate climate zone, no-tillage would have a promising future for soil stability by maintaining higher water-stable aggregate and plant available water contents.

5. Conclusions

The water-stable aggregate value and plant available water mean in sandy loam were higher than in loam, but were not significantly different. The average contents of water-stable aggregates and plant available water tended to decrease in the following order: air-dry soil > soil with water content at field capacity > soil near full saturation. These results indicate that soil aggregates in soil that is close to full saturation upon freezing will be less stable after thawing and will decrease the plant available water content.

The content of water-stable aggregates in no-tillage was 9.4% higher than in reduced tillage and 14% higher than in conventional tillage. The plant available water content in no-tillage was 5.6% higher than in conventional tillage and 13.6% higher than in reduced tillage.

The contents of soil organic carbon and total nitrogen were positive factors for water-stable aggregates at all water levels of the soil. The effects of various physical and chemical properties on water-holding capacity as a function of water level during the freeze–thaw process indicate that water-stable aggregate contents acted as a direct factor for soil water-holding capacity.