Study on Shear Characteristics of Herbs Plant Root–Soil Composite System in Beiluhe Permafrost Regions under Freeze–Thaw Cycles, Qinghai–Tibet Highway, China

Abstract

In order to study the root–soil composite system shear characteristics under the action of freeze–thaw cycles in the permafrost regions along the Qinghai–Tibet Highway (QTH) from the Beiluhe–Tuotuohe (B-T) section, the slopes in the permafrost regions along the QTH from the B-T section were selected as the object of the study. The direct shear test of root–soil composite systems under different amounts of freeze–thaw (F-T) cycles and gray correlations were used to analyze the correlation between the number of F-T cycles, water content, root content, and the soil shear strength index. The results show that the cohesion of the soil in the area after F-T cycles exhibits a significant stepwise decrease with an increase in F-T cycles, which can be divided into three stages: the instantaneous stage (a decrease of 46.73–56.42%), the gradual stage (a decrease of 14.80–25.55%), and the stabilization stage (a decrease of 0.61–2.99%). The internal friction angle did not exhibit a regular change. The root–soil composite system showed significant enhancement of soil cohesion compared with soil without roots, with a root content of 0.03 g/cm³ having the most significant effect on soil cohesion (increasing amplitude 65.20–16.82%). With an increase in the number of the F-T cycles, while the water content is greater than 15.0%, the greater the water content of the soil, the smaller its cohesion becomes. Through gray correlation analysis, it was found that the correlation between the number of F-T cycles, water content, root content, and soil cohesion after F-T cycles were 0.63, 0.72, and 0.66, respectively, indicating that water content had the most significant impact on soil cohesion after F-T cycles. The results of this study provide theoretical support for further understanding the variation law of the shear strength of root–soil composite systems in permafrost regions under F-T cycles and the influencing factors of plant roots to enhance soil shear strength under F-T cycles, as well as for the scientific and effective prevention and control of retrogressive thaw slump in the study area, the QTH stretches across the region.

1. Introduction

The Qinghai–Tibet Plateau (QTP) is the largest perennial permafrost distribution area at mid-latitudes in the world [1], and it is also known as the ‘Roof of the World’ and the ‘Third Pole’ [2]. Its complex terrain and continuous uplift produce strong dynamic and thermal effects, which have a significant impact on China’s and even the global climate system [1,2,3]. It is the ‘driver’ and ‘amplifier’ of global climate change [3]. It has been shown that the distribution area of perennial permafrost on the QTP is 1.01 × 10⁶ km² [4], and perennial permafrost plays an important role in the impact of ecological and environmental changes in the alpine region as well as in the planning of infrastructure projects [5]. In recent years, due to the influence of global warming, the freeze–thaw (F-T) cycles have been significantly strengthened, leading to changes in soil cohesion in the permafrost regions of the QTP, which affects the shear strength of permafrost, reduces the stability of slopes, and triggers a certain degree of damage and destabilization of infrastructure construction such as highways and railways in the cold region [6]. In particular, in recent years, with the continuous increase in infrastructure construction such as railways and highways in the permafrost region of the QTP, a series of disaster problems such as retrogressive thaw slump, solifluction, and thermokarst have occurred due to permafrost degradation [6,7,8,9,10,11], which has resulted in certain impacts and potential threats on adjacent transportation trunk lines and infrastructure facilities. Based on this, in order to ensure the coordinated development between the construction of infrastructure and the protection of the ecological environment in the region, it is necessary to carry out research on the shear characteristics of soil in permafrost regions under the action of F-T cycles. For the scientific and effective prevention and control of a series of permafrost disasters such as retrogressive thaw slump and solifluction in the permafrost regions of the QTP, research has become an urgent issue for the Qinghai–Tibet Highway (QTH) and its surrounding areas to ensure the smooth construction of infrastructure such as roads.

It has been shown that the shear strength of soil after F-T cycles is an important index to reflect the mechanical properties of soil and describe the characteristics of soil after F-T cycles [12,13,14], and the shear strength parameters of soil mainly includes the cohesion and internal friction angle of soil [15,16,17,18,19,20]. According to relevant research results at home and abroad, through conducting F-T cycle tests on different types of soil, it is shown that there are significant differences in the variation law of soil shear strength, which is characterized by an increase, decrease, and basically unchanged variation law of soil shear strength with an increasing number of F-T cycles [6,12]. Sun et al. (2021) conducted a study on the shear characteristics of silt in Huainan, Anhui province, under F-T cycles. The study showed that the decrease in soil cohesion after 1 and 3 F-T cycles was relatively large, and the decrease in cohesion afterward gradually slowed down. After 10 cycles, the cohesion tended to stabilize, while the internal friction angle of the soil showed a trend of decreasing first, then increasing, and then decreasing again [21]. Liu et al. (2021) studied the changes in shear strength of undisturbed loess samples on the Loess Plateau after F-T cycles and found that with the increase in the number of F-T cycles, the soil cohesion decreased exponentially and the internal friction angle of the soil slightly increased [22]. Lv et al. (2019) conducted a remolding experiment on silty clay from northeastern China and then studied the changes in shear strength of the remolded soil samples under different amounts of F-T cycles. They pointed out that there is a linear positive correlation between the number of F-T cycles and the cohesion of the soil, and there is no significant correlation between the number of F-T cycles and the internal friction angle [23]. Shima et al. (2022) has investigated the effect of freeze–thaw cycles on the pure mode I, pure mode II and mixed-mode I/II fracture resistance of cement emulsified asphalt mortar (CEAM). The results showed that the application of freeze–thaw cycles reduced the fracture energy of CEAM in all three modes [24]. Arash et al. (2022) investigated the effect of adding nanocement to improve the dynamic behavior of clay under freeze–thaw cycles. The result showed the addition of nanocement increased the shear modulus and decreased the damping ratio [25].

In addition, many scholars have conducted a large number of soil F-T cycles tests in order to study the changing laws of soil shear properties after F-T cycles tests. As shown in

Table 1. The results of the values of freeze–thaw temperature, freeze–thaw duration, number of freeze–thaw cycles, and other related parameters in the freeze–thaw cycles test.

Researcher	Test Object	Temperature/°C		Duration/h		Number of Freeze–Thaw Cycles
		Freeze	Thaw	Freeze	Thaw
Tang et al. (2023) [26]	Q3 loess	−15	Room temperature	12	12	0, 2, 4, 6, 8, 10
Qiu et al. (2022) [27]	Drifted soil	−15	25	12	12	0, 1, 3, 6, 10
Yang et al. (2023) [28]	Solidified soil	−15	20	12	12	0, 1, 3, 5, 7, 10
Li et al. (2022) [29]	Clay	−20	20	12	12	0, 1, 3, 5, 7, 10
Zhang et al. (2022) [30]	Coarse grain soil	−15	5	16	8	7
Quan et al. (2023) [6]	Clay	−15	20	12	12	0, 1, 2, 3, 5, 7, 10
Guo et al. (2022) [31]	Clay	−10	20	6	6	1, 2, 5
Cai et al. (2021) [32]	Silt	−23	20	24	24	0, 10
Lin et al. (2022) [33]	Clay	−20	20	12	12	0, 1, 10

, scholars have selected test objects, F-T temperatures, F-T durations, the number of F-T cycles, and other relevant test parameters when conducting soil F-T cycles test. According to this table, different types of test soil and different test parameters result in differences in the shear properties and variation laws of soil after F-T cycles. Based on the above actual situation, relevant research has shown that unifying the parameters such as F-T temperature, freeze–thaw duration, and the number of F-T cycles can effectively reduce the differences in the variation laws of soil shear properties after F-T cycles [12].

To summarize, the current deficiencies in the research conducted mainly lie in the different test parameters such as F-T temperature, F-T duration, the number of F-T cycles, etc., in the F-T cycles test, and there is no relatively unified standard yet, which results in the differences in the change rules of the shear properties of the soil after the F-T cycles test. Secondly, the existing research on how to enhance the shear strength of the soil after F-T cycles and its benefits to the ecological environment protection is relatively insufficient, mainly manifested in the addition of materials such as curing agents, fibers, and industrial waste residues to the soil, which, although has an improvement effect on the soil, the harmony between the improved material and the ecological environment is not shown.

In view of the existing deficiencies, in this paper, the native herbs in the QTP are explored as an effective strategy to prevent and control the thaw slumping disaster along the Beiluhe–Tuotuohe section of the QTH. The effect of plant roots on the shear strength of soil in permafrost regions was evaluated. The objectives of this study are as follows: (1) the variation in soil shear properties after F-T cycles was discussed; (2) we evaluated the effect of plant roots as a reinforcement material on the shear strength index cohesion c value of soil after F-T cycles; and (3) the correlation between the number of F-T cycles, water content, root content, and soil shear strength index was compared.

2. Materials and Methods

2.1. Study Site

The study area is located in a retrogressive thaw slump area on the G109 National Highway from Beiluhe–Tuotuohe at K3074 + 100 m. The bottom of the slump body is adjacent to the QTH and is about 500 m away from the Qinghai–Tibet Railway. The slump block is equipped with important infrastructure facilities such as communication optical cables and engineering pipelines [34]. The geographical location and traffic conditions of the study area are shown in Figure 1. Its geographical coordinates are 92°53′40″ E and 34°43′38″ N, with an altitude of 4760 m. The study area has a semi-arid cold climate, belonging to the plateau sub-cold semi-arid climate zone [34,35]. The annual average temperature is −5.0~−3.8 °C, the annual average evaporation is 1316.9 mm [34], and the annual average rainfall is 290~300 mm [35]. The perennial permafrost in the study area is extremely developed, with a thickness of active layer generally ranging from 2 to 3 m, and the surface soil layer is dominated by clay [35], belonging to the alluvial and pluvial high plain landform of the Hoh Xil region on the QTP [36]. The main vegetation type in the region is alpine meadow, with sedges dominating the meadow plants. The main plant species are Carex myosuroides Vill., Kobresia tibetica Maxim., and Garex hirta Linn., etc. The soil types are mainly alpine meadow soil, alpine shrub meadow soil, and meadow marsh soil [37].

2.2. Materials

The soil samples used in the laboratory test for this study were taken from the permafrost regions of the Beiluhe on the QTH. The soil type is silty clay, and the soil particle size distribution curve in the study area is shown in Figure 2. From

Table 2. Results of soil particle analysis in the study area.

Specific Gravity of Soil Particles/Gs	Coefficient Uniformity/Cu	Coefficient Curvature/Cc	Particle Composition/%			Soil Type
			>0.075 mm	0.075~0.005 mm	<0.005 mm
2.71	68.00	13.70	38.00	48.30	13.70	Silty clay

, it can be seen that the soil particle specific gravity is 2.71, the unevenness coefficient (C_u) is 68.00, and the curvature coefficient (C_c) is 13.70. The soil particle size mainly ranges from 0.075 to 0.005 mm. In addition, it can be seen from the table that C_u > 5 and C_c < 3 of the slope soil in the area, which reflects the poor grading of the soil in the area. During the indoor root–soil composite system shear test, the root material used was a combination of herbs plant roots grown in the native meadow layer of the region, consisting of the main herbs plant types of Carex myosuroides Vill., Kobresia tibetica Maxim., Garex hirta Linn., etc. The pulling effect of roots in the study area is shown in Figure 3. The sample of the root–soil composite system was prepared in July 2022 in the field study area. The specific sample preparation method is as follows: use a shovel to dig a meadow layer composed of a root–soil composite system at a depth of 30 cm below the ground surface, wrap it with plastic wrap, and then take it back to the laboratory to prepare remolded soil samples.

2.3. Specimen Preparation

According to the “National Standard of China for Geotechnical Test Methods” (GB/T 50123-2019) [38], the soil sample was measured using the ring method to determine the density of the soil in the area. The water content of the soil was measured using the drying method. In this study, the root–soil composite ring knife samples prepared in the area were cleaned with clean water to retain the roots, and the roots were dried at 25 °C for 8 h, and Formula (1) was used to obtain the root content of the sample. The measured water content of the soil in the area was 30.0%, and the density was 1.38 g/cm³, with a root content of 0.03 g/cm³. The experimental samples were remolded soil. To maintain consistency in water content and density between the remolded soil and the in situ field soil in the area, four different gradients of root content were set in the ring sample of the root–soil composite system, including 0 g/cm³, 0.01 g/cm³, 0.03 g/cm³, and 0.05 g/cm³. At the same time, in order to maintain consistency in root content and density between the remolded soil and the in situ field soil in the area, the root content of soil in the ring sample was controlled to be 0.03 g/cm³, and the density was set to 1.38 g/cm³ during the indoor sampling process. Additionally, according to the water content of in situ field soil in the study area, which was 30.0% in the study area, four different gradients of water content were set for each ring sample during the test process, including 10.0%, 15.0%, 20.0%, and 25.0%.

$$m_r=\frac{m_s}{v}$$

(1)

In Equation (1), $m_r$ is the root content of the root–soil composite sample (g/cm³); $m_s$ is the dry root mass (g) contained in the root–soil composite sample; $v$ is the volume of the root–soil composite sample; and the total volume of each group of 4 samples is calculated to be 240 cm³.

• Steps for preparing remodeled soil samples

① Drying: place the soil sample in an oven and dry it at a constant temperature of 105 °C for 8 h to remove water. ② Sieving soil: after crushing the dried soil sample, use a sieve with a 2 mm pore size for screening. ③ Preparing soil sample: after mixing the dried soil with water according to a certain ratio, place it in a sealed bag and leave it to stand and wet for 24 h for use. ④ Check the uniformity of soil sample moisture by sampling at different locations, using a three-parameter instrument to measure the moisture content of the remolded soil sample. The measured result should not be more than ±1% of the preparation result before use.

2.

Straight shear specimen preparation steps

① Calculate the mass (m_s) of soil samples required for each specimen based on the predetermined density. For plain soil, weigh the soil sample with a mass of m_s at a particular time and add it to the compaction cylinder (diameter 61.8 mm, height 125 mm). For the root–soil composite system, first mix the plant roots with a mass of m with the soil sample evenly, and then add it to the sample preparation device in 4 increments. ② Weigh the mixed roots and soil samples with a mass of m_s + m or plain soil with a mass of m_s at a particular time, add them to the compaction cylinder of the sample preparation device, and compact them layer by layer. The contact surface of each layer is scraped with a knife. ③ Take out the prepared soil samples, cut them into 4 straight shear samples with a diameter of 61.8 mm and a height of 20 mm with a soil-cutting knife, and use them as a group. ④ Repeat the above steps to prepare 3 groups of samples, and then carry out the test.

2.4. Freeze–Thaw Cycles Direct Shear Test

According to the measured temperature data in the study area in recent years, the annual maximum temperature is 19.2 °C, and the minimum temperature is −27.9 °C [34]. Referring to the values of three test parameters used in F-T cycles test conducted by scholars, such as F-T temperature, F-T duration, and the number of F-T cycles, the freezing temperature can be selected between −15 °C and −20 °C, the thawing temperature can be selected between 15 °C and 20 °C, the F-T duration is selected as 24 h, and the number of F-T cycles is preferably 10–15 [12]. In Bakary et al.’s study (2023), the axial strain and temperature of samples were measured during 12 F-T cycles conducted between 10 °C and −20 °C [39]. Elisabeth et al. (2022) assessed the F-T behavior of bonded fasteners in concrete according to the European Assessment Document 330499-01-0601 with F-T condition tests, which include 50 temperature cycles with a duration of 24 h between −20 °C and +20 °C on constantly loaded anchors [40]. Therefore, during the indoor F-T cycles test, this study used the TMS9018-500 F-T cycles system, with temperature changes using a sinusoidal pattern. During the test, the minimum freezing temperature was set to −20 °C, the freezing duration was 12 h, the maximum thawing temperature was set to 20 °C, the thawing duration was 12 h, and the amounts of cycles were 0, 1, 3, 5, 7, and 10. In addition, after F-T cycles are completed and the sample returns to room temperature, a rapid direct shear test is immediately conducted. The rapid direct shear test equipment uses the ZJ strain-controlled direct shear apparatus. The rapid direct shear test uses 50 kPa, 100 kPa, 200 kPa, and 300 kPa vertical pressure, with a shear rate of 2.4 mm/min [41]. The data are collected and processed through the geotechnical test data acquisition and processing system to obtain the shear strength indicators cohesion c value and internal friction angle φ value of the sample. The principle is that the shear strength of soil is the ultimate strength of shear resistance when one part of the soil slides against another part of the soil under the action of external force. In the direct shear test, four samples of the same kind of soil were subjected to four different vertical pressures, and the horizontal shear force was directly applied along the fixed shear surface to obtain the shear stress at the time of failure. Then, according to the Coulomb’s law (2), the shear strength index cohesion c value and internal friction angle φ value of the soil were determined. The laboratory reconstruction soil F-T cycles direct shear test process conducted in this study is shown in Figure 4, and the required parameters are shown in Table 3.

$$\tau =\sigma tan\phi +c$$

(2)

In Equation (2), $\tau$ is the shear strength (kPa), and $\sigma$ is the normal stress (kPa).

Figure 4. Indoor freeze–thaw cycles direct shear test process of remolded soil samples in the study area. (A) Root–soil composite system in situ sample. (B) Dry soil weighing. (C) Water weighing. (D) Remolded root–soil composite system specimens. (E) Preparation of root–soil composite ring knife sample. (F) Root–soil composite system ring knife sample. (G) The sample was subjected to freeze–thaw cycles test. (H) Freeze–thaw cycles box setting temperature.

Figure 4. Indoor freeze–thaw cycles direct shear test process of remolded soil samples in the study area. (A) Root–soil composite system in situ sample. (B) Dry soil weighing. (C) Water weighing. (D) Remolded root–soil composite system specimens. (E) Preparation of root–soil composite ring knife sample. (F) Root–soil composite system ring knife sample. (G) The sample was subjected to freeze–thaw cycles test. (H) Freeze–thaw cycles box setting temperature.

Table 3. Parameter setting results of indoor freeze–thaw cycles direct shear test on soil remodeling specimens in the study area.

Water Content/%	Root Content/g/cm3	Number of Freeze–Thaw Cycles	Temperature/°C		Duration/h
			Freeze	Thaw	Freeze	Thaw
30.0	0	0, 1, 3, 5, 7, 10	−20	20	12	12
30.0	0.01	0, 1, 3, 5, 7, 10	−20	20	12	12
30.0	0.03	0, 1, 3, 5, 7, 10	−20	20	12	12
30.0	0.05	0, 1, 3, 5, 7, 10	−20	20	12	12
25.0	0.03	0, 1, 3, 5, 7, 10	−20	20	12	12
20.0	0.03	0, 1, 3, 5, 7, 10	−20	20	12	12
15.0	0.03	0, 1, 3, 5, 7, 10	−20	20	12	12
10.0	0.03	0, 1, 3, 5, 7, 10	−20	20	12	12

Table 3. Parameter setting results of indoor freeze–thaw cycles direct shear test on soil remodeling specimens in the study area.

Water Content/%	Root Content/g/cm3	Number of Freeze–Thaw Cycles	Temperature/°C		Duration/h
			Freeze	Thaw	Freeze	Thaw
30.0	0	0, 1, 3, 5, 7, 10	−20	20	12	12
30.0	0.01	0, 1, 3, 5, 7, 10	−20	20	12	12
30.0	0.03	0, 1, 3, 5, 7, 10	−20	20	12	12
30.0	0.05	0, 1, 3, 5, 7, 10	−20	20	12	12
25.0	0.03	0, 1, 3, 5, 7, 10	−20	20	12	12
20.0	0.03	0, 1, 3, 5, 7, 10	−20	20	12	12
15.0	0.03	0, 1, 3, 5, 7, 10	−20	20	12	12
10.0	0.03	0, 1, 3, 5, 7, 10	−20	20	12	12

2.5. Gray Relational Analysis

Sun (2010) pointed out that gray correlation analysis is a quantitative description of the characteristics of mutual changes between factors. Although the Pearson correlation, linear regression, and other related analysis methods can also help obtain answers, they often require a large amount of data and obvious data distribution characteristics. And, when the distribution characteristics of multiple factors is not obvious, the use of the Pearson correlation, linear regression, and other related analysis methods is difficult. Relatively speaking, the gray correlation analysis requires less data and lower requirements for data, and the principle is simple [42]. The principle of gray relational analysis is an analytical method that examines the geometric proximity between factors to analyze and determine the degree of influence between factors or the contribution of several sub-sequences to the parent sequence [43].

Let there be a reference sequence X₀ = {x₀(k), k = 1, 2, …, n} and a comparison sequence X_i = {x_i(k), k = 1, 2, …, n}. The mean value $\overline{X}$ of the parent sequence X₀ = {x₀(k), k = 1, 2, …, n} and the subsequent X_i = {x_i(k), k = 1, 2, …, n} is obtained by using Formula (3). Then, by dividing each element x₀(k) and x_i(k) in the sequence by the mean value of the sequence, the new mother sequence X₀ = {x₀(k), k = 1, 2, …, n} and the subsequent X_i = {x_i(k), k = 1, 2, …, n} are obtained. The correlation coefficient ζ_i(k) between X₀ and X_i at the k point can be expressed by Equation (4) [44]:

$$\overline{X}=\frac{1}{n}{\sum }_{k=1}^{n}X\left(k\right)$$

(3)

$${\zeta }_i\left(k\right)=\frac{\left[{min}_i{min}_k\left|X_i\left(k\right)-X_0\left(k\right)\right|+\rho {max}_i{max}_k\left|X_i\left(k\right)-X_0\left(k\right)\right|\right]}{\left[\left|X_i\left(k\right)-X_0\left(k\right)\right|+\rho {max}_i{max}_k\left|X_i\left(k\right)-X_0\left(k\right)\right|\right]}$$

(4)

In Equation (4), ${\zeta }_i\left(k\right)$ represents the correlation coefficient between each corresponding data of the reference sequence and the comparison sequence; $\left|X_i\left(k\right)-X_0\left(k\right)\right|$ represents the absolute difference between the two sequences; ${min}_i{min}_k\left|X_i\left(k\right)-X_0\left(k\right)\right|$ represents the minimum value of the absolute difference between the two sequences; ${max}_i{max}_k\left|X_i\left(k\right)-X_0\left(k\right)\right|$ represents the maximum value of the absolute difference between the two sequences; and $\rho$ represents the resolution coefficient, which is taken as 0.5.

The correlation of the two sequences can be expressed by Equation (5) [45]:

$${\gamma }_i=\frac{1}{n}{\sum }_{k=1}^n{\zeta }_i\left(k\right)$$

(5)

In Equation (5), ${\gamma }_i$ represents the correlation between the reference sequence and the comparison sequence.

3. Results

3.1. Direct Shear Test Results and Analysis

3.1.1. The Variation in Shear Resistance of Soil with Different Root Content under F-T Cycles

As shown in Figure 5, when the plain soil and root–soil composite samples in the area are subjected to shear stress, the shear stress and shear displacement curves are approximately linear in the initial stage. When the shear displacement exceeds 0.1 mm, the linear relationship is transformed into a nonlinear relationship. At the same time, under the same shear stress, compared with the initial stage, the nonlinear stage forms a relatively significant shear displacement, and the shear stress of the sample shows a significant increase with the increase in vertical pressure.

Figure 6 shows the change trend of soil cohesion under F-T cycles for soil samples with the same water content and density but different root content in the study area. From the figure, it can be seen that with the increase in the number of F-T cycles, the cohesion of soil with different root content decreases significantly in a stepwise trend, which can be divided into three stages: transient stage (after one F-T cycle), gradual stage (after three to seven F-T cycles), and stable stage (after seven or more F-T cycles). The variation process is as follows: after one F-T cycle, compared with the unfrozen soil samples, the soil samples with four different gradients of root content, 0 g/cm³, 0.01 g/cm³, 0.03 g/cm³, and 0.05 g/cm³, have a decrease in cohesion by 56.42%, 50.91%, 46.73%, and 52.22%, respectively. After three F-T cycles, compared with the soil samples after one F-T cycle, the soil samples with four different gradients of root content, 0 g/cm³, 0.01 g/cm³, 0.03 g/cm³, and 0.05 g/cm³, have a decrease in cohesion by 23.51%, 21.44%, 25.55%, and 14.80%, respectively. Similarly, after 10 F-T cycles, compared with the soil samples after 7 F-T cycles, the soil samples with four different gradients of root content, 0 g/cm³, 0.01 g/cm³, 0.03 g/cm³, and 0.05 g/cm³, have a decrease in cohesion by 0.61%, 2.99%, 1.29%, and 1.63%, respectively. From the above analysis, it can be seen that the decrease in the cohesion of soil samples with different root contents under the same number of F-T cycles is relatively similar. This result indicates that the effect of F-T cycles on the cohesion of soil samples with different root contents is relatively consistent.

As shown in

Table 4. The increase in cohesion of different root contents soils after freeze–thaw cycles compared with that of plain soil (unit: %).

Root Content/g/cm3	Number of Freeze–Thaw Cycles
	0	1	3	5	7	10
0.01	11.42	25.50	28.90	30.41	12.20	9.51
0.03	16.82	42.81	38.99	65.20	42.07	41.10
0.05	6.68	16.96	30.28	40.94	12.50	11.35

, it can be seen that the cohesion of the soil containing roots is significantly higher than that of plain soil. In addition, when the number of F-T cycles is less than three, the cohesion of soil with different root contents is increased by 0.03 g/cm³ > 0.01 g/cm³ > 0.05 g/cm³ compared to plain soil. Correspondingly, when the number of F-T cycles is greater than three, the cohesion of soil with different root contents is increased by 0.03 g/cm³ > 0.05 g/cm³ > 0.01 g/cm³ compared to plain soil, which further reflects that the root–soil composite system significantly enhances the cohesion of soil compared to plain soil, with the most significant increase in cohesion occurring in soil with a root content of 0.03 g/cm³. The main reason for the above changes is that according to Li et al. (2015), there is an optimal root content in the root–soil composite shear strength index, which is reflected in the root–soil composite sample. With the increase in root content, the shear strength index c value gradually increases. When the root content in the composite sample exceeds a certain amount, the shear strength index c value shows a decreasing trend, reflecting the existence of the optimal root content in the root–soil composite sample [46]. Based on this, the root content corresponding to the shear strength index c of the root–soil composite obtained in this study is the optimal root content.

Figure 7 shows the change characteristics of the internal friction angle of soil samples with the same water content and density in the study area but different root content under the action of F-T cycles. From the figure, it can be seen that with the increased number of F-T cycles, the change characteristics of the internal friction angle of four different soil samples display no obvious change pattern. Wang (2005) pointed out that the internal friction angle of clay did not show obvious regularity with the increase in freeze–thaw cycles [47]. Lv et al. (2019) conducted a remolding experiment on silty clay from northeastern China, and then studied the changes in shear strength of the remolded soil samples under different amounts of F-T cycles. They pointed out that there is no significant correlation between the number of F-T cycles and the internal friction angle [23]. This is consistent with the results obtained in this study.

3.1.2. The Variation in Shear Resistance of Soil with Different Water Content under F-T Cycles

In order to further explore the change pattern of soil shear strength in the region under different water content conditions after freeze–thaw cycles, this study selected four different gradient conditions with a root content of 0.03 g/cm³ and a water content of 10.0%, 15.0%, 20.0%, and 25.0%, respectively. The change pattern of soil shear strength after F-T cycles of the root–soil composite system was tested. The test results are shown in Figure 8. From the figure, it can be seen that the change characteristics of soil cohesion under the action of F-T cycles for soil samples with the same root content and density but different water content in the region manifested as a gradual decrease in cohesion with the increase in F-T cycles for soil with different water content, which can be divided into three different stages: transient stage, gradual stage, and stable stage. This is consistent with the decreasing trend of cohesion for four different soil samples with different root contents under the action of F-T cycles. As shown in

Table 5. The cohesion of different water content soils after freeze–thaw cycles (unit: kPa).

Water Content/%	Number of Freeze–Thaw Cycles
	0	1	3	5	7	10
10.0	11.97	6.35	4.90	3.81	3.20	2.96
15.0	12.24	6.99	5.44	4.66	3.45	3.23
20.0	9.21	6.27	4.85	4.07	2.77	2.75
25.0	8.25	5.44	4.76	3.71	2.54	2.50

, it can be concluded that as the number of F-T cycles increases with a water content of 15.0%, the greater the water content of the soil, the smaller its cohesion, and among them, the soil with a water content of 15.0% has the largest cohesion after F-T cycles.

In addition, Figure 9 shows the variation characteristics of the internal friction angle of soil samples with the same root content and density but different water content under the action of F-T cycles. From the figure, it can be seen that with the increased number of F-T cycles, the internal friction angle of the four different water content soil samples did not show a significant change pattern. Further analysis showed that after 7 F-T cycles, compared with the samples that had undergone 10 F-T cycles, the internal friction angle of the soil samples with four different water content gradients of 10.0%, 15.0%, 20.0%, and 25.0% changed by 1.35%, 1.39%, 0%, and 0.94%, respectively, reflecting a stable trend of change in the internal friction angle of the soil. This is consistent with the results obtained by many scholars [23,47].

3.2. Calculation Results and Analysis of Gray Correlation Degree

After the soil in the study area undergoes F-T cycles, its cohesion shows significant regular patterns of change with the number of F-T cycles, water content, and root content. However, the internal friction angle does not show significant regular patterns of change. In this study, the gray correlation method is used to analyze the correlation between the cohesion of the F-T cycles soil and three factors: the number of F-T cycles, water content, and root content.

Table 6. Calculated correlations between parent sequence cohesion and subsequent number of freeze–thaw cycles, water content, and root content.

X0 (Cohesion/kPa)		X1 (Number of Freeze–Thaw Cycles)		X2 (Water Content/%)		X3 (Root Content/g/cm3)
Original Value	Processing Value	Original Value	Processing Value	Original Value	Processing Value	Original Value	Processing Value
6.54	1.54	0	0.00	30.0	1.26	0.00	0.00
7.29	1.72	0	0.00	30.0	1.26	0.01	0.38
7.64	1.80	0	0.00	30.0	1.26	0.03	1.14
6.98	1.64	0	0.00	30.0	1.26	0.05	1.90
11.97	2.82	0	0.00	10.0	0.42	0.03	1.14
12.24	2.88	0	0.00	15.0	0.63	0.03	1.14
9.21	2.17	0	0.00	20.0	0.84	0.03	1.14
8.25	1.94	0	0.00	25.0	1.05	0.03	1.14
2.85	0.67	1	0.23	30.0	1.26	0.00	0.00
3.58	0.84	1	0.23	30.0	1.26	0.01	0.38
4.07	0.96	1	0.23	30.0	1.26	0.03	1.14
3.33	0.78	1	0.23	30.0	1.26	0.05	1.90
6.35	1.50	1	0.23	10.0	0.42	0.03	1.14
6.99	1.65	1	0.23	15.0	0.63	0.03	1.14
6.27	1.48	1	0.23	20.0	0.84	0.03	1.14
5.44	1.28	1	0.23	25.0	1.05	0.03	1.14
2.18	0.51	3	0.69	30.0	1.26	0.00	0.00
2.81	0.66	3	0.69	30.0	1.26	0.01	0.38
3.03	0.71	3	0.69	30.0	1.26	0.03	1.14
2.84	0.67	3	0.69	30.0	1.26	0.05	1.90
4.90	1.15	3	0.69	10.0	0.42	0.03	1.14
5.44	1.28	3	0.69	15.0	0.63	0.03	1.14
4.85	1.14	3	0.69	20.0	0.84	0.03	1.14
4.70	1.11	3	0.69	25.0	1.05	0.03	1.14
1.71	0.40	5	1.15	30.0	1.26	0.00	0.00
2.23	0.53	5	1.15	30.0	1.26	0.01	0.38
2.83	0.67	5	1.15	30.0	1.26	0.03	1.14
2.41	0.57	5	1.15	30.0	1.26	0.05	1.90
3.81	0.90	5	1.15	10.0	0.42	0.03	1.14
4.66	1.10	5	1.15	15.0	0.63	0.03	1.14
4.07	0.96	5	1.15	20.0	0.84	0.03	1.14
3.71	0.87	5	1.15	25.0	1.05	0.03	1.14
1.64	0.39	7	1.62	30.0	1.26	0.00	0.00
1.84	0.43	7	1.62	30.0	1.26	0.01	0.38
2.33	0.55	7	1.62	30.0	1.26	0.03	1.14
1.85	0.44	7	1.62	30.0	1.26	0.05	1.90
3.20	0.75	7	1.62	10.0	0.42	0.03	1.14
3.45	0.81	7	1.62	15.0	0.63	0.03	1.14
2.77	0.65	7	1.62	20.0	0.84	0.03	1.14
2.54	0.60	7	1.62	25.0	1.05	0.03	1.14
1.63	0.38	10	2.31	30.0	1.26	0.00	0.00
1.79	0.42	10	2.31	30.0	1.26	0.01	0.38
2.30	0.54	10	2.31	30.0	1.26	0.03	1.14
1.82	0.43	10	2.31	30.0	1.26	0.05	1.90
2.96	0.70	10	2.31	10.0	0.42	0.03	1.14
3.23	0.76	10	2.31	15.0	0.63	0.03	1.14
2.75	0.65	10	2.31	20.0	0.84	0.03	1.14
2.50	0.59	10	2.31	25.0	1.05	0.03	1.14

shows the cohesion of the soil as the mother sequence, denoted as X₀, and the number of F-T cycles, water content, and root content as the sub-sequences, denoted as X₁, X₂, and X₃. According to Formulas (1) and (2), the correlation between the cohesion of the soil in the study area and the three factors the number of F-T cycles, water content, and root content is calculated. As shown in

Table 7. Calculated correlation between soil cohesion and three factors such as number of freeze–thaw cycles, water content, and root content in the study area.

Parameter	Number of Freeze–Thaw Cycles	Water Content/%	Root Content/g/cm3
Cohesion/kPa	0.63	0.72	0.66

, the correlation between these three factors and the cohesion of the soil in the study area is greater than 0.5, indicating that these three factors have a certain impact on soil cohesion. Additionally, according to the correlation between soil cohesion and these three factors, the order from high to low is water content, root content, and the number of F-T cycles. Furthermore, as shown in Table 5, among these three factors, water content has the highest correlation with the soil cohesion after F-T cycles, which is 0.72. Therefore, water content has the most significant impact on soil cohesion after F-T cycles. Dong et al. (2017) pointed out that the influence on the freeze–thaw characteristics of soil can be explained from the perspective of establishing a new equilibrium after the soil skeleton is unstable [12]. Before freezing and thawing, the soil skeleton composed of soil particles, pores, and water in soil is in a relatively stable static state. After many freeze–thaw cycles, a new stable soil skeleton is formed in the soil, and a new dynamic equilibrium is established, so its physical and mechanical properties gradually tend to be stable. This follow-up study needs to be further explored.

4. Discussion

4.1. The Influence of Number of F-T Cycles on the Shear Characteristics of Soil under F-T Cycles

In terms of the effect of F-T cycles on the shear characteristics of soil after F-T cycles, Bing et al. (2001) found that saline soil, after experiencing F-T cycles, gradually developed into brittle materials, but after 10 F-T cycles, the shear strength of the soil samples would reach a new equilibrium [48]. Xie et al. (2015) studied the F-T characteristics of silty sands and pointed out that the strength of the silty sand samples decreased with the increased number of F-T cycles and showed a constant strength after 15 F-T cycles [49]. Su et al. (2008) studied the change rule of shear properties of Tibetan clay under F-T cycles, and the results showed that the soil cohesion and internal friction angle of soil samples tended to stabilize after 10 F-T cycles [50]. Ye et al. (2012) conducted loess F-T cycles test and pointed out that the shear mechanical indicators of loess basically reached a stable state after 10 F-T cycles [51]. Javad et al. (2022) pointed out that comprehensive laboratory experiments were first conducted to investigate the deterioration and failure process of Tasmanian sandstones subjected to various numbers of F-T cycles (FTCs). A theoretical damage model was then derived to quantify the degradation of the physical–mechanical properties of the sandstones [52]. Subhashbhai et al. (2022) pointed out that the settlement values determined from the centrifuge run were high during the initial F-T cycles, and it decreased rapidly with an increase in cycles [53]. Dong et al. (2017) pointed out that when soil experienced the first F-T cycle, an effective channel had not yet been formed in the soil, and the water in the soil migrated and changed into an unstable state [12]. In the process of water migration and freezing expansion in soil, the soil particles will be crushed, and the soil particle connection and pore characteristics will be changed. Zhang et al. (2016) pointed out that the soil skeleton was unstable, and the soil structure changed greatly, resulting in a large attenuation of the physical and mechanical properties of the soil after the first freeze–thaw cycle [54]. After experiencing an F-T cycle, an unstable migration channel formed inside the soil. F-T cycles continued to circulate, and the physical and mechanical properties of the soil continued to change, but the change was smaller than the first change. After several F-T cycles, the broken soil particles gradually moved closer under the action of an electric double layer and matrix force. Zhang et al. (2016) pointed out that a relatively stable channel will form in the soil, the soil particle skeleton will gradually stabilize, and the soil structure will gradually reach a new stable dynamic equilibrium, so its physical and mechanical properties will gradually stabilize [54]. In this study, direct shear tests were conducted on the soil mass in the permafrost regions along the Beiluhe–Tuotuohe section of the QTP, and the variation law of the soil shear strength parameters obtained was basically consistent with that obtained by the above scholars; that is, with the increase in the number of F-T cycles, the soil cohesion showed a trend of attenuation, and after one F-T cycle, the soil cohesion decreased the most, and after experiencing seven F-T cycles, it reached a new stable state.

4.2. The Influence of Water Content on the Shear Characteristics of Soil under F-T Cycles

Previous research results have shown that the migration of and phase change in water in soil are important aspects of the attenuation of various physical and mechanical properties of soil after F-T cycles. Therefore, water content is an important factor in the F-T damage of soil [12]. Tian et al. (2014) conducted direct shear tests on loess samples with different water content prepared from the No. 1 tunnel in Yangquan, Shanxi province, and found that the cohesion of the soil decreased after F-T cycles, which was manifested in the higher variation amplitude of the samples with a higher water content than the samples with a lower water content [55]. Zhou et al. (2013) conducted an F-T cycles test on loess samples from Sanyuan County, Xianyang, Shaanxi province, and found that the greater the water content, the greater the reduction in soil cohesion and internal friction angle after F-T cycles [56]. Quan et al. (2023) selected from Qinghai–Tibet a clay and set up three different water content gradient samples of 12%, 15%, and 18% in turn. Through direct shear tests of F-T cycles, they found that the degradation trend of cohesion for the three different water content samples was the same and showed that the greater the water content, the greater the degradation of soil cohesion [6]. Şahin et al.’s (2022) study on the effects of this salt presence on the freeze–thaw strength of the grout is rare in the literature. As a result of the experiments, pertaining to all prepared grout samples, freeze–thaw strength decreased when the water content increased [57]. Dong et al. (2017) pointed out that the mechanism of the influence of water content on the characteristics of soil after F-T cycles is mainly reflected in the process of F-T cycles [12]. When the temperature is reduced to 0 °C, the pore water on the surface of the sample first freezes, the internal pore water migrates to the frozen part, and the freezing gradually extends to the interior. At the same time, the soil particles shrink at a low temperature [6]. In this process, on the one hand, when the water content is relatively high, the frost heave of pore water is greater than the shrinkage of solid particles, and the soil generally presents a frost heave state; when the water content is low, the frost heave of pore water is small, and the overall frost heave of soil is not obvious or even slightly contracted [6,12,54]. On the other hand, the freezing expansion volume of water in soil becomes larger, especially when the water content is larger and the expansion deformation of pore water is higher, which may crush soil particles and change the connection form of soil particles, resulting in the change in pore size and structure. When the temperature rises, the solid water melts and migrates, and the soil damage is difficult to recover, resulting in the uneven settlement of the soil. If the water in the soil cannot be discharged in time, this will cause problems such as reduced soil strength and an insufficient bearing capacity. The soil undergoes multiple F-T cycles, the water in the soil is repeatedly frozen and melted, and the soil structure is constantly changing. The higher the water content, the greater the change [12]. This process occurs until, after several F-T cycles, a stable channel is formed in the soil, a new dynamic equilibrium is established, and the soil reaches a stable state again. This study conducted laboratory direct shear tests on remolded soil samples subjected to F-T cycles and set four different water content gradients of 10%, 15%, 20%, and 25%. The results showed that as the number of F-T cycles increased and with a water content of 15%, the soil cohesion decreased as the water content increased. This result is consistent with the findings of previous studies.

4.3. The Influence of Improved Materials on the F-T Shear Characteristics of Soil under F-T Cycles

In order to study the influence of improved materials on the shear characteristics of soil after F-T cycles, many scholars have improved the soil by adding solidifiers, fibers, industrial waste residue, and other materials and have conducted F-T cycles tests on the improved soil. The results show that the improved soil has good resistance to F-T [12]. Ghazavi et al. (2010) studied the influence of F-T cycles on the compressive strength of fiber-reinforced clay and pointed out that when 3% polypropylene fiber was added to the soil, the increase in unconfined compressive strength of the soil after F-T cycles was 60% to 160%, and the frost heaving reduction was 70% [58]. Wang et al. (2013) selected a new type of high molecular weight solidification material SH to chemically improve loess and conducted F-T cycle tests on SH-improved loess and cemented loess for comparison. The results showed that SH-solidified-improved-loess had higher strength and better resistance to F-T, while cemented loess had lower strength and poorer resistance to F-T [59]. Kenan et al. (2010) conducted F-T cycles on fine grained soils with added geosynthetic fibers and performed California bearing ratio performance tests on the post-cycle samples. The results showed that fine grained soils with added geosynthetic fibers had good resistance to F-T [60]. Mahya et al. (2023) claimed that the objective of their study was to investigate the effect of nano-clay as a stabilizer on the mechanical properties of clay. The results indicate that the long-term durability of specimens against freeze–thaw cycles increases further with the addition of nano-clay content ranging from 2% to 3% [61]. Gençdal et al. (2023) reported that their study investigated the improvement of a high-plasticity clayey silt soil (MH) with microencapsulated phase change material (mPCM) to prevent changes in mechanical properties when subjected to freeze–thaw cycles. As a result of the tests, it was determined that the most suitable additive mPCM ratio is 10% for the compression and strength behaviors [62]. Adding modified materials to F-T cycles soil samples can prevent the development of tensile cracks in soil during loading. The presence of modified materials can enhance the friction between soil particles and enhance the physical and mechanical properties of soil [58,59,60,61,62]. When the improved material was plant roots, Liu et al. (2018) observed the microstructure characteristics of the soil surface contacted by plant roots by scanning electron microscopy and obtained that the surface morphology of the soil contacted with the roots showed relatively flat and dense characteristics compared with the rootless soil [63]. The soil particles in contact with the root were subjected to traction and extrusion in turn so that the original orientation of the soil particles changed and became compacted and the compactness of the soil increased, and its shear strength would also be enhanced to a certain extent [63]. In this study, by adding herbs roots to the soil as a material to improve the performance of the soil, it was found that the cohesion of the soil with added roots was significantly improved compared with the soil without roots after F-T cycles, and the effect of improving the cohesion of the soil was relatively most significant when the root content was 0.03 g/cm³.

4.4. Microscopic Analysis of the Variation in Shear Characteristics

In F-T cycles, the shear characteristics of soil will go through three stages of ‘stable-unstable-new stable’ [6]. In order to explore the reasons for this change pattern of soil in the area under the action of F-T cycles, this paper discussed the changes in soil microstructure. Zhang et al. (2022) carried out CT scanning on the soil after F-T cycles. The results showed that the water migration in the sample led to changes in the pore structure and particle structure of the soil after multiple freeze–thaw cycles [30]. Dong et al. (2010), Ye et al. (2018), and Ni et al. (2014) pointed out that during the initial freezing, the water mainly freezes in situ, and the ice crystals are evenly distributed among the soil particles, which expands the original pores. However, in the melting stage, the ice–water phase transition causes the ice crystals originally used to cement the soil skeleton to disappear, and the expanded pores are difficult to recover. The number of contact points between soil particles is significantly reduced, the cementation strength between soil particles is reduced, and the soil strength is deteriorated. After repeated F-T, the size, shape, and inter-particle pores of soil particles gradually reach a new balance; that is, the strength degradation trend of soil tends to be gentle [64,65,66]. At the same time, Ning et al. (2018) pointed out that the F-T process will cause cracks in the soil and continue to develop, expanding the original pores and water migration channels and affecting the bonding strength between soil particles, resulting in the deterioration of soil strength [67].

5. Conclusions

Based on the experimental works presented in this study on the soils in the study area, after different amounts of freeze–thaw cycles, the following main conclusions can be summarized:

• After the F-T cycles test on the soils in the study area, its cohesion showed a significant downward trend with the increase in the number of F-T cycles. The change trend of the cohesion of the soils after the F-T cycle test can be divided into three stages: transient stage (after one F-T cycles), gradual stage (after three to seven F-T cycles), and stable stage (after more than seven F-T cycles). The internal friction angle of the internal soils after the F-T cycles showed no significant change with the increase in the number of F-T cycles.
• By comparing the changes in cohesion after F-T cycles in different soils, it can be concluded that the root–soil composite system has a significantly higher effect on improving soil cohesion than plain soil, with the effect being most pronounced when the root content is 0.03 g/cm³. And, it can be concluded that as the number of F-T cycles increases, the greater the water content of the soils, and the smaller the change in soil cohesion after F-T.
• The gray correlation analysis shows that the gray correlation between the number of F-T cycles, water content, and root content on the soil cohesion after F-T is more than 0.5, which reflects that these three factors will have a certain degree of influence on the soil cohesion. The relational degree between the water content and the soil cohesion after F-T is the largest, which is 0.72.

The results of this study show that the dominant herbaceous plant roots that constitute the meadow layer in the area have the ability to improve the shear strength of the soil and have the ability to resist freeze–thaw collapse damage. Therefore, the use of plant measures can effectively prevent and control the thermal melt slump disaster caused by temperature changes in the alpine environment.