Study on Soil Stabilization and Slope Protection Effects of Different Plants on Fully Weathered Granite Backfill Slopes

Abstract

The slope erosion in the distribution area of completely weathered granite is often relatively severe, causing serious ecological damage and property loss. Ecological restoration is the most effective means of soil erosion control. Taking completely weathered granite backfill soil as the research object, two types of slope protection plants, Vetiver grass and Pennisetum hydridum, were selected. We analyzed these two herbaceous plants’ soil reinforcement and slope protection effects through artificial planting experiments, indoor simulated rainfall experiments, and direct shear tests. The test results showed that the runoff and sediment production rates of the two herbaceous plant slopes were significantly lower than those of the bare slope, with the order of bare slope > Vetiver grass slope > Pennisetum hydridum slope. Compared with the bare slope, the cumulative sediment production of the Vetiver grass slope at 60 min decreased by 56.73–60.09%, and the Pennisetum hydridum slope decreased by 75.97–78.45%. The indoor direct shear test results showed that soil cohesion decreases with increasing water content. As the root content of Vetiver grass roots increases, soil cohesion first increases and then decreases, reaching a maximum value when the root content is 1.44%. As the root content of Pennisetum hydridum increases, soil cohesion increases. The internal friction angle increases slightly with increasing water content, while the root content does not significantly affect the internal friction angle. Therefore, the shear strength of soil decreases when the water content increases. The shear strength of the Vetiver grass root-soil composite reaches a peak at a root content of 1.44%, while the shear strength of the giant king grass root-soil composite increases as the root content increases. At the same root content, the shear strength of the Vetiver grass root-soil composite is slightly higher than that of giant king grass. The reinforcement effect of roots on shallow soil is better than on deep soil. Both herbaceous plants have an excellent soil-fixing and slope-protecting impact on the fully weathered granite backfill slope. Pennisetum hydridum’s soil and water conservation effect is significantly better than that of the Vetiver grass. In contrast, Vetiver grass roots slightly outperform Pennisetum hydridum in enhancing the shear strength of the soil. The research results can provide a theoretical basis for the vegetation slope protection treatment of fully weathered granite backfill slopes.

1. Introduction

In human engineering activities, excavating and backfilling the slope is often necessary. However, the original surface vegetation cover was destroyed after slope excavation and backfilling, forming secondary bare land, causing soil erosion and further leading to environmental issues such as soil erosion and landslides [1]. Currently, the main slope protection measures in engineering activities include engineering measures, vegetation measures, and a combination of engineering and vegetation measures. Vegetation slope protection, also known as plant slope protection, ecological slope protection, or plant slope stabilization, is widely applied in slope erosion control due to its advantages, such as soil stabilization, slope protection, improvement of the ecological environment, and low prevention and control costs [2]. This technology improves the shear strength and permeability of soil through hydrological and mechanical effects [2,3], thereby enhancing the stability of slopes.

From the perspective of the hydrological effect of plants, vegetation slope protection mainly manifests in the interception and buffering effect of plant stems and leaves on rainwater [4]. Many scholars at home and abroad have conducted numerous indoor and outdoor experiments to study the hydrological effects of vegetated slopes. The results show that vegetation can significantly reduce the erosion caused by rainwater on the slope surface. Absorbing and dispersing rainwater energy reduces the slope surface’s erosion effect [5,6,7,8]. Jia Lianlian et al. [9] and Wang Lianrui et al. [10] analyzed the influence of different types of surface cover on erosion and sediment production in loess and red clay slopes through simulated rainfall experiments. The study found that vegetation significantly reduces runoff on the slope surface, and the runoff and sediment production are minimal in forest and grassland types. Liu et al. [11] studied the runoff and sediment yield regulation effects of two Poaceae plants (Elymus dahuricus and Bromus inermis) and two leguminous plants (Alfalfa and White Clover) under simulated rainfall conditions and compared them with those of bare slopes. The results showed that Poaceae plants performed better in reducing sediment and maintaining runoff due to their dense fibrous roots and high biological crust coverage.

The mechanical effect of vegetation slope protection is mainly reflected in the reinforcement of short roots and the anchoring effect of long roots [12]. Many scholars at home and abroad have researched the relationship between water content, plant roots, and soil shear strength. Studies have shown that there is an optimal water content to make the shear strength of the soil reach the peak value, and the shear strength is the smallest under saturated water content. The influence of moisture content on shear strength is mainly reflected in the change in cohesion [13,14]. The root-soil composite considers roots and soil as an integrated system, and its shear strength is attributed to the interaction between the root system and the soil [15]. Roots’ ability to enhance the soil’s shear strength is related to root diameter, length, growth angle, and root content [16]. Some studies have pointed out that the root-soil composite’s shear strength increases with the root morphology’s complexity [17,18,19]. Li Zhenyu et al. [16,20] used the combination of water and fertilizer to regulate the root architecture of plants and studied the effect of root architecture on soil strength. They pointed out that with the increase in secondary and tertiary roots of Vetiver grass after regulation, the shear strength of the soil significantly improved. To quantitatively study root content’s contribution to soil shear strength, Zhang Xingling et al. [21] used the roots of Achnatherum splendens in the loess region of the Qinghai-Tibet Plateau as reinforcement material for reinforced soil and pointed out the exponential relationship between the shear strength and root content of root-reinforced soil. Yu Yi et al. [22] and Wang Lianrui et al. [10] studied the slope protection effect of herbaceous plants with red clay slopes as the research object and pointed out that the improvement of root weight density (RWD) on the strength of red clay slopes is mainly reflected in strengthening cohesion. At the same time, it has little effect on the internal friction angle. Moreover, the impact of Vetiver grass on enhancing soil shear strength is more pronounced than that of Cynodon dactylon. Some studies further pointed out that with the increase in root content, the shear strength of the complex increased first and then decreased. There is an optimal root content [23,24,25].

In summary, the current research mainly evaluates slope protection with vegetation by combining hydrological and mechanical effects. However, the research focuses on loess and red clay areas, and more research still needs to be performed in entirely weathered granite areas. Completely weathered granite is widely distributed in Longling County, Baoshan City, Yunnan Province. Due to the pipe trench excavation, the surface vegetation and the red soil layer on the slope’s surface are damaged, and the lower completely weathered layer is exposed, which can easily cause shallow surface landslides and slope erosion disasters [26,27]. Therefore, it is urgent to conduct relevant research on vegetation slope protection in fully weathered granite areas. This study, taking fully weathered granite backfill soil as the research object, carries out indoor artificial rainfall simulation experiments and indoor shear tests and explores the impact of two herbaceous plants on slope runoff, sediment production, and soil shear strength. The research results can provide a theoretical basis for the ecological management of fully weathered granite backfill soil slopes.

2. Overview of the Study Area

The study area is in Longling County, Baoshan City (Figure 1a), Yunnan Province, at the intersection of the south subtropical climate zone and the north tropical zone. With abundant rainfall, the annual average precipitation falls between 1400 and 2100 mm (Figure 1b), mainly concentrated in May and October, accounting for approximately 88–90% of the annual rainfall. The average yearly precipitation in other areas is mostly 800–1100 mm. The annual average temperature is 14.9 °C. Due to the complex terrain and significant vertical elevation difference, Longling County possesses climatic characteristics of both low latitude, monsoon, and mountain terrain, forming a subtropical mountainous monsoon climate with minor temperature differences throughout the four seasons, distinct dry and wet seasons, and prominent vertical variation.

Simultaneously, the vegetation coverage in the study area is relatively low, and the plant species are relatively homogenous. Among the completed slope protections, the combination of herbaceous plants and shrubs is primarily used, which solves the shortcomings of shallow root systems and poor slope stabilization effects of herbaceous plants. This approach also reduces ecological safety risks and achieves fast and durable slope protection, favoring the positive succession of the ecosystem. The primary plants used are foxtail grass, pine trees, artificially planted Vetiver grass, and giant king grass.

The granite is widely distributed in the study area. Due to abundant rainfall and low vegetation coverage, the granite is strongly weathered, which constitutes its unique geological characteristics [28,29]. Pipe trench excavation is required during the laying of oil and gas pipelines. Due to the specificity of oil and gas pipelines, the backfilled soil cannot be compacted after excavation, resulting in insufficient compactness of the backfilled soil, reduced shear strength, and susceptibility to deformation and damage under the action of rainwater erosion and dynamic loads generated during pipeline operation [27].

3. Materials and Methods

3.1. Test Materials

3.1.1. Basic Properties of Fully Weathered Granite Backfill Soil

The completely weathered granite backfill soil used in this experiment was taken from Longling District, Baoshan City, Yunnan Province. The soil sample appears grayish-white and mainly consists of medium-to-coarse-grained quartz with a small amount of potassium feldspar and kaolinite. The soil has a high gravel content, poor inter-particle cohesion, loose structure, and poor physical and mechanical properties, making it susceptible to erosion by running water and extremely difficult for ecological restoration. Before the experiment, measuring the fundamental physical properties of the soil was strictly according to the “Standard for Soil Test Method” GB/T 50123-2019 [30]. The results are shown in

Table 1. Physical property indicators of completely weathered granite backfill soil.

Soil Name	Natural Density/g·cm−3	Dry Density /g·cm−3	Natural Moisture Content/%	Volumetric Weight /KN·m−3	Plastic Limit /%	Liquid Limit /%	Natural Void Ratio
Completely weathered granite backfill soil	1.78	1.52	14.9	17.8	21.3	30.9	0.733

. The particle size gradation is shown in Figure 2, which indicates that the soil has a good gradation and uniform particle size distribution.

3.1.2. Herbaceous Plants

Select the well-growing Pennisetum hydridum from the study area and purchase Vetiver grass with good soil fixation and slope protection functions. Transplant and cultivate these plants at the School of Land Resource Engineering testing site, Kunming University of Science and Technology, from February to September 2023 (Figure 3). The specifications of the steel troughs for planting the two herbaceous plants are 3 m (length) × 0.5 m (single width) × 0.5 m (height). Watering was performed daily during the initial planting stage to ensure plant survival. After the plants survive, watering is stopped to allow them to grow in a natural environment. After planting, conduct indoor simulated rainfall tests and root-soil composite shear strength tests.

3.2. Test Method

3.2.1. Artificial Simulated Rainfall Test

The indoor artificial simulated rainfall test was conducted from September to November 2023, using a self-designed ZYKX-DZ02 artificial simulated rainfall system. The experimental device is illustrated in Figure 4. Based on rainfall data from Longling County from 2012 to 2022 and preliminary field investigations in the study area, the rainfall intensity was designed to be 110 mm/h (heavy rain), and the slope angles were set at 15° and 30°. Before the experiment, the soil was naturally air-dried and sieved through a 10 mm mesh. A layered soil filling method was employed, with each layer having a thickness of 10 cm until the total thickness reached 50 cm. Compacting while filling and controlling the soil moisture content at approximately 10%. After the completion of soil filling, the soil bulk density of the slope was measured to verify whether it meets the experimental design requirements. Vetiver grass were planted in Tank A, Pennisetum hydridum was planted in Tank B, and a bare slope was established as a control group. Due to limited space in the steel tank area, the experiment was conducted in two separate sessions. After completing the preparatory work, carry out the indoor simulated rainfall test. Starting from the runoff generation on the slope surface, collect runoff every 2 min using a 1 L container, each lasting 30 s. Each rainfall event lasted for 1 h. Calculate the slope surface’s runoff generation rate and sediment yield rate based on the experimental results. The runoff generation rate is calculated as the ratio of the runoff volume (mL) collected in a single sample to the duration of the sample collection (s). The sediment yield rate is calculated as the ratio of the sediment yield (g) collected in a single sample to the duration of the sample collection (s). The runoff generation on the slope mainly depends on soil infiltration conditions, while sediment generation mostly comes from gully wall collapse and runoff erosion. Both are essential indicators reflecting slope erosion.

3.2.2. Shear Strength Test

Determination of gradients of water content and root content: Since slope erosion in the study area mainly occurs during the rainy season, three levels of mass water content, namely 15%, 19%, and 23%, are set between natural water content and saturated water content to determine the optimal water content, from now on referred to as water content. Referring to the studies by Chen Weijie [31] and Zeng Qingjun et al. [32] on the content of Vetiver grass and Pennisetum hydridum in soil and based on on-site longitudinal profiling of the planted vegetation, we have determined that the maximum mass root content rate for fresh Vetiver grass is approximately 1.92%, with a variation gradient of 0.48%. The maximum mass root content rate for fresh Pennisetum hydridum roots is approximately 1.41%. Two similar root content rates were set for Pennisetum hydridum to facilitate comparison with Vetiver grass, using a variation gradient of 0.23%. Five levels of root content gradients were established for both plants, from now on referred to as the root content rate. Details are provided in

Table 2. Root rate gradient table.

Sample Shape	Moisture Content /%	Root Ratio /%	Root Diameter /mm
Plain soil	15%	-	-
	19%	-	-
	23%	-	-
Vetiver grass	15%	0.48	0∼2.5
		0.96	0∼2.5
		1.44	0∼2.5
		1.92	0∼2.5
Pennisetum hydridum	15%	0.72	0∼3.0
		0.95	0∼3.0
		1.18	0∼3.0
		1.41	0∼3.0

.

The two types of roots used in the experiment were intact, fresh, and undamaged. Two plants of Vetiver grass and Pennisetum hydridum with consistent growth were selected. Through longitudinal profiling, it was found that the root systems of both plants are mainly distributed at a depth of 0–30 cm below the soil surface, and both have many lateral and fibrous roots. The root diameter was measured using a vernier caliper.

In this experiment, remolded soil samples were used. Given that the pore distribution of the pressed sample is more uniform and closer to the ultimate strength of the undisturbed soil [33], the pressing method is used to prepare disturbed samples of plain soil and root-soil composites in this study. The sample was prepared using a self-designed soil sampling ring knife with an inner diameter of 61.8 mm and a height of 20 mm, along with a matching loading device. Due to the limitations of the experimental setup, the roots were cut to a length of approximately 2 cm (Figure 5). Following the principle of dry density control, ring knife samples were prepared according to preset moisture content and root content gradients, specifically referring to the “Standard for Soil Test Method” GB/T 50123-2019.

The experiment was conducted at the Soil Engineering Laboratory of Kunming University of Science and Technology’s School of Land Resources Engineering from November to December 2023. Use the four-linkage strain-controlled direct shear tester to perform direct shear tests on 44 samples under four levels of vertical stress: 50, 100, 150, and 200 kPa. The shear rate was set at 0.8 mm/min. Setting up three parallel groups for each sample to ensure the accuracy of the experiment, and the results were averaged, as shown in

Table 3. Statistics of test results.

Sample Shape	Moisture Content /%	Root Ratio /%	Shear Strength/kPa
			50 kPa	100 kPa	150 kPa	200 kPa
Plain soil without root	15%	-	43.09	80.17	114.99	147.35
	19%	-	38.91	77.27	108.41	138.08
	23%	-	38.91	69.90	104.87	126.35
Vetiver grass	15%	0.48	43.27	80.72	114.90	146.35
		0.96	45.91	83.72	125.44	148.02
		1.44	52.09	90.99	129.99	152.44
		1.92	51.18	88.63	126.62	150.71
Pennisetum hydridum	15%	0.72	43.53	82.54	115.33	149.08
		0.95	44.44	83.17	127.94	148.99
		1.18	49.25	87.45	130.50	149.71
		1.41	51.46	86.36	130.60	150.92

. Based on the shear strength of the soil under different vertical stresses, the corresponding shear strength indicators were calculated using Coulomb’s strength theory.

4. Results and Analysis

4.1. Analysis of Rainfall Erosion Test Results

4.1.1. Runoff Characteristics of Different Plant Slopes

Figure 6 shows the variation trend of the runoff generation rates of the three kinds of slopes with the same rainfall duration. At the beginning of rainfall, the runoff generation rate on the slope is the lowest. As the rainfall progresses, the soil moisture content gradually increases, soil infiltration decreases, and slope runoff increases rapidly. When the soil moisture content reaches saturation, the runoff generation rate on the slope stabilizes. Therefore, the runoff generation rate rises quickly and then shows large fluctuations. The runoff generation rate on a bare slope has the most extensive fluctuation range. This is because the bare slope forms gullies that can block runoff channels after being eroded by raindrops and surface erosion. When the slope runoff channels are flushed open, there are relatively unstable oscillatory changes. The presence of plants can weaken the kinetic energy of falling raindrops to a certain extent, effectively reducing erosion damage to the soil surface and slowing down the runoff speed on the slope. The average runoff generation rate for the three types of slopes is bare slope > Vetiver grass slope > Pennisetum hydridum slope. This indicates that both herbaceous plants can effectively reduce slope runoff, and the effect of lowering runoff is more significant on the Pennisetum hydridum slope.

4.1.2. Sediment Yield Characteristics of Different Plant Slopes

Different slopes have similar trends in sediment production rate (Figure 7). Some loose soil particles are on the slope surface at the beginning of rainfall. As the runoff increases rapidly, soil particles are quickly transported. Therefore, the sediment production rate rapidly increases at the initial rainfall stage of three types of slopes under two different gradients [10]. After the first rainfall on the bare slope (15° slope), the loose soil and soil particles on the slope surface have been washed away. Due to insufficient backfilling, the sediment production rate at the beginning of rainfall on the 30° bare slope shows a downward trend. As the rainfall progresses, the soil moisture content reaches saturation, and the slope runoff stabilizes. The loose soil on the surface has been depleted, and the slope runoff’s ability to transport soil particles decreases. The sediment production rate on the slope reaches a dynamic equilibrium, with the bare slope showing the most extensive fluctuation range. Under both slopes, the sediment production rate is highest to lowest at the bare, Vetiver grass, and Pennisetum hydridum slopes. This means that giant king grass has a more significant effect on reducing sediment production on the slope.

The reasons for the above phenomena are: (1) during rainfall, the kinetic energy of raindrops serves as the initial driving force for soil erosion on slopes. For bare slopes without vegetation cover, the rainfall directly hits the slope, when raindrops fall onto the surface layer of the slope, they cause splash erosion of the surface soil, and runoff will form rapidly on the slope, leading to the development of rills on the slope. This process results in edge erosion and accumulation within the rills, thereby increasing sediment production on the slope, and the sediment production rate fluctuates significantly. However, on slopes with planted vegetation, part of the rainfall is retained on the stems and leaves, and the interception effect of the stems and leaves can reduce its kinetic energy and weaken the erosion effect of raindrops on the shallow soil of the slope. The other part seeps into the ground through the root system [34,35]. (2) The stems and leaves, while reducing and dispersing runoff, increase the roughness of the slope surface and the resistance of slope flow, thereby slowing down the flow velocity and reducing the sediment yield rate. (3) Planting vegetation on slopes can improve the physical and chemical properties of the soil through its root system, delaying the formation of gullies on the surface layer of the slope. This effectively reduces the erosion of the slope’s shallow surface layer and enhances the surface soil’s overall erosion resistance on the slope [36,37].

4.1.3. Cumulative Runoff and Sediment Production

Table 4. Cumulative runoff and sediment yield in 60 min for different slopes.

Slope	Slope Type	Cumulative Runoff Volume/L	Cumulative Sediment Yield/g	Reduction Range of Sediment Yield/%
15°	bare slope	40.85	1084.10	-
	Vetiver grass slope	39.80	469.13	56.73
	Pennisetum hydridum slope	28.45	233.59	78.45
30°	Bare slope	30.95	2258.20	-
	Vetiver grass slope	28.70	901.34	60.09
	Pennisetum hydridum slope	20.60	542.71	75.97

shows the cumulative runoff and sediment yield after 60 min of rainfall at a rainfall intensity of 110 mm/h for different slopes. The cumulative runoff for the two slopes is 15° > 30°, while the cumulative sediment production is 30° > 15°. This is because the amount of water retained on the slope surface decreases as the slope increases. Still, it accelerates the conversion of potential energy into kinetic energy of the slope runoff, increasing the runoff velocity and allowing more sediment to be transported. The cumulative runoff and sediment production for both slopes are most significant for the bare slope, followed by the Vetiver grass slope and the Pennisetum hydridum slope. The cumulative sediment production for the Vetiver grass slope is reduced by 56.73–60.09% compared to the bare slope. In comparison, the cumulative sediment production for the Pennisetum hydridum slope is reduced by 75.97–78.45%. Based on the comprehensive analysis of Figure 6 and Figure 7, both herbaceous plants have significant soil and water conservation effects, and the impact of Pennisetum sinese is significantly better than that of Vetiver grass.

4.1.4. Slope Erosion Characteristic

An analysis was conducted on various slopes after rainfall to more intuitively compare the slope protection effects of plants. As shown in Figure 8, there are significant differences in slope erosion patterns under different slope gradient conditions.

For a bare slope with a gradient of 15°, the primary erosion patterns are surface erosion and raindrop splash erosion. The slope surface has sporadically distributed small pits, while the rest of the area exhibits a relatively flat eroded surface without distinct erosion gullies. On the slope surfaces of both Vetiver grass and Pennisetum hydridum, there is a slight formation of erosion gullies in the middle and lower parts, with no other significant changes observed on the overall slope surface.

As the slope gradient increases, the primary erosion patterns on the bare slope surface are erosion gullies and discontinuous scour marks, mainly distributed in the middle and lower parts of the slope. A tension crack develops at the top of the slope. A distinct erosion gully forms on the slope surface of Vetiver grass, and a few erosion holes appear at the top of the slope. On the slope surface of the Pennisetum hydridum, a slight erosion gully forms at the lower part, with several erosion holes at the top. The rest of the eroded surface is uniform, with no significant changes on the slope.

4.2. Analysis of Shear Strength Test Results of Root-Soil Composite

4.2.1. Analysis of Shear Stress and Shear Displacement Curves

This study conducted direct shear tests on two types of herbaceous plant root-soil composite samples and soil samples without roots under four levels of vertical stress: 50, 100, 150, and 200 kPa and obtained the shear stress and shear displacement curves of the soil. Here, we illustrate with examples of shear stress and shear displacement curves for plain soil at three moisture content levels of 15%, 19%, and 23%, as well as for root-soil composites with Vetiver grass contents of 0.96% and 1.92%, and giant king grass contents of 0.95% and 1.41% at 15% moisture content (Figure 9). The figure shows that: (1) Under the same moisture content and root content conditions, shear stress is positively correlated with shear displacement. At the beginning of the shear test, the relationship between shear stress and shear displacement is roughly linear. However, as the shear test progresses, the relationship between shear stress and displacement becomes nonlinear. (2) Under any of the four levels of vertical stress, when the shear stress reaches a specific value, the shear displacement gradually increases with increasing moisture content. For example, at a vertical stress of P₄ = 200 kPa and a shear stress of 120 kPa, the shear displacements of plain soil with moisture contents of 15%, 19%, and 23% are 2.06 mm, 2.74 mm, and 3.94 mm, respectively. Compared to plain soil with 15% moisture content, the shear displacements of plain soil with 19% and 23% moisture content increase by 33.00% and 91.26%, respectively. This indicates that the soil’s resistance to shear deformation gradually decreases when the water content is relatively high. Similarly, the relationship between shear stress and shear displacement for root-soil composites at different root contents follows the same trend as that for plain soil.

Under the same shear stress, the shear displacement of the root-soil composite is smaller than that of plain soil. When the shear stress reaches 120 kPa and under vertical stress of P4 = 200 kPa, the shear displacements for the Vetiver grass slope with root contents of 0%, 0.96%, and 1.92% are 2.06 mm, 1.87 mm, and 1.62 mm, respectively. For Pennisetum hydridum, the shear displacements with root contents of 0%, 0.95%, and 1.41% are 2.06 mm, 1.97 mm, and 1.85 mm, respectively. Compared with the shear displacement of plain soil, the shear displacement of root-soil composites with Vetiver grass root content of 0.96%, 1.92%, and Pennisetum hydridum root content of 0.95%, 1.41% decreased by 9.22%, 21.36%, 4.37%, and 10.19%, respectively. This indicates that roots effectively enhance the soil’s resistance to shear deformation.

4.2.2. Analysis of Shear Strength of Plain Soil under Different Moisture Contents

The variation of shear strength of plain soil with moisture content was obtained through direct shear tests; the results are shown in Figure 10. Under the same vertical stress, the shear strength of plain soil is generally negatively correlated with moisture content. The shear strength of the soil at different moisture content levels is ranked as 15% > 19% > 23%. Fully weathered granite is rich in hydrophilic clay minerals. When the moisture content is high, a thicker bound water film forms of soil particles, causing the volume of clay minerals to expand and accelerating the expansion of adjacent cracks. The difference in porosity leads to an uneven overall structure of the sample, thereby reducing the shear strength of the soil [38].

4.2.3. Analysis of Shear Strength of Root-Soil Composite under Different Root Content Rates

Figure 11a shows the variation of shear strength with root content for the Vetiver grass root-soil composite under four normal stress levels. As can be seen from the figure, except for a few abnormal samples, the shear strength of Vetiver grass root-soil composites is greater than that of plain soil. The increase in shear strength of the root-soil composite is most significant when the root content is 1.44%. In contrast, the enhancement effect of Vetiver grass roots on shear strength is insignificant when the root content is 0.48%. Furthermore, under the same root content, the shear strength of the root-soil composite tends to increase as the vertical stress rises. When the vertical stress reaches 200 kPa, the increase in shear strength of the root-soil composite compared to plain soil is smaller than the increases under the other three levels of vertical stress. Taking Vetiver grass with a root content of 1.92% as an example, under the four levels of vertical stress, the increases in shear strength of the root-soil composite compared to plain soil are 8.09, 8.42, 13.64, and 3.36 kPa, respectively. This is due to the limitation of rearrangement of soil particles and soil volume changes under high-stress conditions. The roots’ enhancement effect on soil shear strength will be relatively weakened. It also indicates that the reinforcement effect of roots on shallow soil is better than that on deep soil, which is consistent with the research results of Hu Ning et al. [39] and Liao Bo et al. [40].

With the increase in root content, the shear strength of Vetiver grass root-soil composites tends to increase first and then decrease. Roots can restrict the deformation of soil on one hand. When the root content is lower than the optimal root content, as the root content increases, the contact area between roots and soil particles increases, thereby increasing the friction between roots and soil and enhancing the shear strength of the composite. On the other hand, as the root content increases, the number of thick roots in the sample also increases. The significant difference in shear resistance between soil and roots causes uncoordinated deformation between roots and soil, forming cracks around the roots and accelerating the expansion of potential cracks. If the confining pressure is too small, the soil becomes more prone to loosening, resulting in a more significant reduction in strength [40]. Therefore, an optimal root content range of 1.44–1.92% exists for Vetiver grass roots, which maximizes the shear strength of the root-soil composite.

Figure 11b shows the shear strength variation with root content for Pennisetum hydridum root-soil composites under four normal stress levels. As can be seen from the figure, the shear strength of Pennisetum hydridum root-soil composites is greater than that of plain soil, indicating Pennisetum hydridum roots. The shear strength of the root-soil composite increases with increasing root content and reaches a maximum at a root content of 1.41%.

The mechanical effect of vegetation slope protection is mainly reflected in the reinforcement of short roots and the anchoring effect of long roots. On the one hand, the roots agglomerate the soil particles around them, which increases the soil cohesion. At the same time, the roots are surrounded by the surrounding soil particles. As many fine steel bars are added to the soil, the roots have a significant reinforcement effect on the soil, thereby improving its shear strength. In this study, due to the limitation of the test device, the roots were cut to a length of approximately 2 cm, resulting in a small sample size, which could not fully reflect the mechanical properties of the undisturbed soil. This may weaken the unique anchoring effect of long roots in undisturbed soil, so that there is a certain deviation between the test results and the real soil mechanical properties, but the overall change trend of shear strength with root content is consistent.

Selecting two groups of root-soil composite samples with approximately equal root content for comparison, When the Vetiver grass root content is 1.44%, the corresponding shear strengths under the four levels of vertical stress are 52.09, 90.99, 130.00, and 152.44 kPa, respectively. However, when the Pennisetum hydridum root content is 1.41%, the corresponding shear strengths under the same vertical stresses are slightly lower, which are 51.46, 86.36, 130.59, and 150.92 kPa, respectively. Except for 150 kPa, the shear strength values of the Vetiver grass root-soil composites are greater than those of Pennisetum hydridum under the other three levels of vertical stress. Therefore, Vetiver grass roots have a slightly better enhancement effect on soil shear strength than Pennisetum hydridum roots. Although the rhizomes of Vetiver grass are small, they are numerous, and the root morphology is more complex, which can tightly entangle the soil, restricting soil deformation and crack propagation. Therefore, in actual slope engineering projects, the growth characteristics of the Vetiver grass should be fully utilized to promote the proliferation of fine roots within the soil, thereby enhancing the cementation ability between the roots and the soil.

Disassembly and observation of the sheared samples (Figure 12) revealed that most of the roots were not cut off but pulled out when the root-soil composite samples underwent shear failure, and there were noticeable sliding friction marks of roots on the shear plane. This indicates that when the composite sample undergoes shear failure, the sliding friction resistance encountered by the roots is less than the maximum tensile resistance of the roots. Therefore, it can be inferred that the shear strength of the composite is related to the sliding friction resistance encountered by the roots during shear failure [22,40].

4.2.4. Analysis of Shear Strength Index

Figure 13 shows the relationship curve between shear strength and vertical stress for plain soil and two herbaceous plant root-soil composites. As can be seen from the figure, when the moisture content and root content are constant, the shear strength increases with increasing vertical stress, and there is a linear relationship between the two. With R² > 0.97, the fitting relationship is good, consistent with Coulomb’s law $\tau =c+\sigma \tan \varphi$. The shear strength indicators, cohesion c and internal friction angle φ are obtained from the fitting curve, and the statistical results are shown in

Table 5. Statistical results of shear strength index.

Sample Shape	Moisture Content/%	Root Ratio/%	Cohesion/kPa	Increment of Cohesion/kPa	Cohesion Growth Rate/%	Internal Friction Angle/°
Plain soil	15%	-	9.50	1.54	18.1	30.7
	19%	-	8.51	0.55	6.9	33.3
	23%	-	7.96	0.00	0.0	34.8
Vetiver grass	15%	0.48	10.45	0.95	10.0	32.4
		0.96	13.75	4.25	44.8	32.8
		1.44	21.36	11.86	124.9	33.5
		1.92	20.14	10.64	112.0	32.9
Pennisetum hydridum	15%	0.72	10.26	0.76	8.0	32.4
		0.95	11.53	2.03	21.4	33.2
		1.18	18.11	8.61	90.7	33.9
		1.41	19.17	9.67	101.8	34.1

.

The variation pattern of cohesion is that under the same root content, the cohesion tends to decrease when the moisture content increases. The increase in cohesion at a moisture content of 15% compared to 23% is significantly greater than that at 19% compared to 23%. When the soil moisture content is low, the suction force between water molecules is weak, and the traction ability of soil particles is also weak. As the soil moisture content gradually increases, the water film begins to form and gradually thickens, and the cohesive effect produced by the water film begins to increase gradually. The soil matric suction and apparent cohesion increase. However, when the moisture content increases to a certain level, further thickening of the water film can weaken the water molecules’ suction force on soil particles, thereby reducing cohesion [41]. This study’s designed moisture content is relatively high, so the cohesion shows a decreasing trend. In addition, the cohesion of the root-soil composite is greater than that of plain soil under the same conditions. The cohesion of the Vetiver grass root-soil composite increases first. Then, it decreases with increasing root content, reaching a maximum at a root content of 1.44%, which is consistent with the root content when the shear strength peaks. The cohesion of Pennisetum hydridum root-soil composite increases with increasing root content.

When the root content of both is the same, the increase in cohesion of the Vetiver grass root-soil composite is significantly more significant than that of the Pennisetum hybrid root-soil composite. This pattern is consistent with their performance in terms of shear strength. The above results indicate that both herbaceous plants enhance the cohesion of the slope soil, but there are slight differences in the enhancement effect. Vetiver grass roots have a more pronounced enhancement effect than Pennisetum hydridum roots.

The variation pattern of the internal friction angle is that, under the same root content, the internal friction angle of the root-soil composite shows a slight increase as the moisture content increases. This is because the increase in moisture causes the water film between soil particles to thicken, which reduces cohesion to some extent. However, due to the lubricating effect of the water film, the internal friction during the mutual movement of soil particles is reduced, increasing the internal friction angle. In addition, under the same moisture content, there is no significant change in the internal friction angle of the root-soil composite with varying root content.

5. Conclusions

(1)

The runoff generation rate, sediment production rate, and cumulative runoff and sediment production of the three types of slopes under different gradients all show the trend of bare slope > Vetiver grass slope > Pennisetum hydridum slope. This indicates that both herbaceous plants have good soil and water conservation effects, and Pennisetum hydridum’s soil and water conservation effect is significantly better than that of Vetiver grass.

(2)

The root systems of both herbaceous plants can effectively improve the shear strength of the root-soil composite. With increasing root content, the shear strength of the Vetiver grass root-soil composite first increases. Then, it decreases, and an optimal root content of 1.44% maximizes the soil’s shear strength. The shear strength of the Pennisetum hydridum root-soil composite increases as the root content increases, reaching a maximum when the root content is at 1.41%. The root content primarily affects the cohesion of the soil, and its variation pattern is consistent with the variation pattern of shear strength; there is no significant effect on the internal friction angle. In general, Vetiver grass has a slightly better enhancement effect on soil shear strength than Pennisetum hydridum, and the reinforcement effect of roots on shallow soil is better than that of deep soil.

(3)

Both herbaceous plants have good soil stabilization and slope protection effects. The combined influence of hydrological and mechanical effects should be comprehensively considered in practical applications.