Responses of Soil Infiltration and Erodibility to Vegetation Succession Stages at Erosion and Deposition Sites in Karst Trough Valleys

Abstract

The topographies of soil erosion and deposition are critical factors that significantly influence soil quality, subsequently impacting the erodibility of soils in karst regions. However, the investigation into the effects of erosion and deposition topographies on soil erodibility across different stages of vegetation succession in karst trough valleys is still at a preliminary stage. Therefore, three distinct topographic features (dip slopes, anti-dip slopes, and valley depressions) were selected at erosion (dip/anti-dip slope) and deposition sites (valley) to investigate the spatial heterogeneity of soil physicochemical properties, infiltration capacity, aggregate stability, and erodibility in karst trough valleys. Additionally, five different stages of vegetation succession in karst forests were considered: Abandoned land stage (ALS), Herb stage (HS), Herb-Shrub stage (HES), Shrub stage (SHS), and Forest stage (FS). Additionally, the relationships among these factors were analyzed to identify the key driving factors influencing soil erodibility. The results revealed that soil physicochemical properties and soil aggregate stability at the deposition site were significantly superior to those at the erosion site. The FS resulted in the best soil physicochemical properties, whereas the HS resulted in the highest soil aggregate stability within the deposition site. However, the soil infiltration capacity at the erosion site was significantly greater than that at the deposition sites. The ALS had the strongest soil infiltration capacity at both the erosion and deposition sites. The soil erodibility at erosion sites (0.064) was significantly greater than that at deposition sites (0.051), with the highest soil erodibility observed on anti-dip slopes during the HES at erosion sites (0.142). The structural equation model reveals that erosion and deposition topographies, vegetation succession, soil physicochemical properties, soil aggregates, and soil infiltration characteristics collectively account for 88% of the variation in soil erodibility under different conditions. Specifically, both direct and indirect influences on soil erodibility are most significantly exerted by soil aggregate stability and vegetation succession. This study provides scientific evidence to support the management of soil erosion and ecological restoration in karst trough valleys while offering technical assistance for regional ecological improvement and poverty alleviation.

1. Introduction

Soil erosion is a global challenge that results in the loss of topsoil and nutrients, ultimately leading to land degradation [1,2]. This issue is particularly severe in karst regions, where the soil layer is shallow and rock exposure is high, making the soil prone to erosion due to rainfall [3]. Karst regions account for approximately 15.6% of the world’s land area [4]. Additionally, the unique double-layer geological structure of karst regions facilitates the flow and loss of sediment both on the surface and underground [1,5]. This process increases the exposure of surface bedrock and reduces the erodible soil area, ultimately leading to rocky desertification [6,7]. Compared with in non-karst regions, soil formation in karst regions is challenging and slow. Studies by Chen et al. have shown that soil in karst regions consists mainly of acid insoluble matter [8], accounting for only 10% of the total material, which contributes to a slower rate of soil formation than erosion [4]. Consequently, soil erosion in karst regions is more complex than that in non-karst regions [9]. Due to the severity of rocky desertification in karst regions, soil erosion has emerged as a significant environmental issue, impeding social and economic development while accelerating land degradation [10]. Therefore, managing soil erosion and preventing soil and water loss are crucial for combating rocky desertification in karst regions [1,5].

Soil erodibility refers to the susceptibility of soil to erosion and is a critical indicator for assessing soil loss and implementing soil and water conservation measures [11,12]. It is influenced by various factors, including soil physicochemical properties, organic matter content, aggregate structure, soil texture, and infiltration capacity [13,14]. Generally, soil erodibility decreases with increasing soil organic matter and bulk density, and more stable soil aggregate structures result in lower erodibility [15]. However, compared with non-karst regions, karst regions are characterized by severe rocky desertification, fragmented soils, and extensive bare bedrock, resulting in numerous microhabitats, such as stone gullies, stone grooves, stone crevices, and stone clusters [16]. Consequently, the spatial heterogeneity of soil physicochemical properties in karst regions is highly pronounced, thereby resulting in significant variations in soil erodibility between adjacent topographical areas within these regions due to differences in infiltration capacity and soil aggregate stability [17,18]. Therefore, exploring the relationships between soil erodibility and its influencing factors under different topographical conditions in karst regions and identifying the key influencing factors are crucial for managing soil and water loss and preventing rocky desertification in these areas [19,20].

Vegetation restoration is an important measure for controlling soil erosion and rocky desertification in karst regions [21,22]. Different vegetation succession can alter the characteristics of soil fauna, microorganisms, and litter, resulting in variations in soil physicochemical properties, soil infiltration characteristics, and soil aggregate stability [23,24]. These changes subsequently impact soil erodibility. Currently, there are numerous studies on the influence of different vegetation successions on soil erosion. For example, Lei et al. investigated soil erodibility in forestlands, croplands, and grasslands at different slope positions on karst hillslopes [4]. Their study confirmed that organic carbon and soil aggregates have a significant effect on soil erodibility. Guo investigated how different patterns of vegetation restoration can modify land use types and alter the nature of the soil and roots, ultimately affecting soil erosion [25]. Additionally, a previous study examined the weakening effect of various vegetation types on reducing soil erodibility on the Loess Plateau, and reported that grasslands were more effective at reducing soil erodibility [3]. However, some studies demonstrated that forested areas had lower erodibility than other land use types in a typical purple soil small watershed [15]. Furthermore, previous studies have utilized multiple soil indices to quantify erodibility while calculating comprehensive indices for different models of vegetation restoration; these studies have established erodibility models indicating that forested areas are better able to mitigate soil erosion [26,27,28]. These discrepancies may be attributed to variations in factors such as climatic conditions, topographic features, or specific characteristics related to the soil or vegetation successional stages of each study area. Therefore, future research should focus on investigating how different vegetation successions affect soil erosion while exploring their underlying influencing factors. Notably, current research on soil erodibility predominantly focuses on homogeneous soils in regions such as the Loess Plateau and purple soil areas [29,30]. Owing to the complex microhabitats in karst regions, vegetation successional stages under different topographies significantly affect soil erodibility. However, a definitive conclusion has yet to be reached, necessitating further investigation.

China is one of the countries with the largest karst regions globally, with a karst area spanning approximately 3.44 × 106 km², accounting for approximately 13% of its total land area [4,28]. The majority of this karst landscape is located in the southwestern region. Furthermore, the karst trough valley is one of eight major topography types (karst trough valley, karst plateau, karst depression, peak depression, karst canyon, karst fault basin, peak plain, and middle mountain) in the southwestern karst regions of China [5,16]. It not only features a typical double-layer geological structure of ground and underground, but also has a unique topographical characteristic of “one mountain, two ridges, and one trough”. Compared with those of other karst topographies, the cross-sections of the karst trough valley region exhibit a “U” shape [21]. The bedrock strata dip on both sides of the trough valley slope either aligns with or opposes the slope direction, forming typical structures such as dip slopes and anti-dip slopes; this results in more complex microhabitats in the karst trough valley, leading to greater spatial heterogeneity in soil physicochemical properties and soil erodibility [5,7,9]. Previous studies have shown that the dip in bedrock strata affects soil physicochemical properties, influences soil infiltration, and ultimately impacts soil erodibility [16,31]. At the erosion site, the alignment or opposition of the bedrock dip and slope directions significantly impacts the hydrodynamic mechanisms of the slope, as well as the distributions of runoff and sediment [10,16]. Consequently, the differences in soil texture between dip slopes and anti-dip slopes, along with variations in vegetation succession, ultimately influence soil erodibility. At the deposition site, sediments from the erosion site accumulate in the trough valley. The gentler slopes and slower runoff of deposition sites compared with erosion sites increase the water retention time at deposition sites, thereby affecting soil physicochemical properties and infiltration [10]. Additionally, human activities are frequent at deposition sites, with most agricultural activities concentrated there [19]. This concentration further complicates the relationship between soil physicochemical properties, soil aggregates, soil infiltration, and soil erodibility in erosion areas. Therefore, it is imperative to clarify the relationships between soil erodibility and its influencing factors at erosion-deposition sites in karst trough valleys.

According to a review of the existing theoretical knowledge, we found that research on soil erodibility primarily involves developing models or calculating indices to evaluate soil erodibility under various vegetation covers and to analyze its variations across different regions and vegetation succession [27,32,33,34]. Furthermore, research on karst trough valleys has focused primarily on soil physicochemical properties, soil aggregate stability, or soil quality under different land use types. However, owing to the complex microhabitats in karst areas through valleys, vegetation succession under different topographies significantly affects soil erodibility [5,7,21]. Few studies have specifically investigated the factors influencing soil erodibility and their direct and indirect impacts. These elements are rarely integrated to examine their direct and indirect effects on soil erodibility [30,35,36]. We need to integrate these elements to study their direct and indirect effects on soil erodibility.

Given the aforementioned scientific gaps, we hypothesize that topographical positions, vegetation successional stages, soil physicochemical properties, soil aggregates, and soil infiltration have direct and indirect impacts on soil erodibility. Therefore, we have selected five vegetation successional stages of karst forests for both erosion and deposition sites in karst trough valleys: Abandoned land stage (ALS), Herb stage (HS), Herb-Shrub stage (HES), Shrub stage (SHS), and Forest stage (FS). We have utilized a structural equation model (SEM) to investigate the relationships between soil erodibility and its influencing factors. The specific objectives are as follows: (1) to examine how different erosion and deposition topographies and changes in vegetation successions affect the physicochemical properties of soils at erosion and deposition sites in karst trough valleys; (2) to explore the relationships between soil erodibility and its influencing factors such as physicochemical properties of soils, soil aggregates, and infiltration characteristics; and (3) to identify key influencing factors for soil erosion in the erosion and deposition sites of karst trough valleys. This study aims to provide a scientific basis for controlling soil erosion and water loss while facilitating ecological restoration efforts in areas affected by deposition caused by erosion in karst trough valleys [1,21]. Additionally, it seeks to establish a stable and sustainable ecosystem in karst trough valleys, offering technological support for regional ecological improvement and the resolution of poverty issues [10].

2. Materials and Methods

2.1. Study Area

This study was conducted in the Qingmuguan karst trough valley area of Chongqing City, located in Southwest China (coordinates: 29°40′–29°47′ N and 106°16′–106°20′ E) (Figure 1). The region experiences a subtropical monsoon humid climate, with most rainfall occurring during the summer and autumn seasons [16]. The annual average precipitation is approximately 1100 mm, and the average annual temperature is about 18 °C. The climate is characterized by warm winters, hot summers, and variable spring and autumn seasons [16]. The Qingmuguan karst trough valley is situated in the southern section of the Jinyun Mountain–Wentang Gorge anticline in the parallel ridge–valley region of Eastern Sichuan, with the overall terrain sloping from northeast to southwest. The northern part of the trough valley features dip slopes where the bedrock dip aligns with the slope direction, while the southern part features anti-dip slopes where the bedrock dip opposes the slope direction. Based on slope direction, position, and bedrock strata dip, the Qingmuguan karst trough valley can be divided into three distinct parts: (1) dip slopes on the northern side of the valley where the slope and bedrock strata dip align; (2) anti-dip slopes on the southern side of the valley where the slope and bedrock strata dip oppose; and (3) the valley depression, referred to as the trough [9,10,37]. Due to erosion, soil from the dip and anti-dip slopes is deposited in the trough, defining the dip and anti-dip slopes as erosion sites and the valley depression as a deposition sites. The average elevation of the erosion and deposition site is 509 m, with an average slope of 30.9 degrees, and the topography of the deposition site exhibits a flatter terrain [21]. The specific topographic characteristics of the erosional and deposition sites are presented in Table 1. Due to the unique biological, climatic, and geomorphic conditions of Chongqing, as well as the abundant precipitation resulting from its humid climate, the soil is prone to aluminization and iron oxide hydration. These factors make it suitable for the development of yellow soil in Qingmuguan karst trough valley [16]. Consequently, the predominant soil types in this area are zonal yellow soil and non-zonal lime soil. According to the World Reference Base for Soil Resources (WRB), the soil in this area is classified as cambisols (primary unit). Additionally, due to a thin soil layer in karst areas, rocky desertification is more severe here. Consequently, throughout most stages of vegetation succession in this area, the presence of a humus layer on the surface is almost negligible. However, forested land exhibits a thin humus layer with an average thickness of 1.1 cm. Furthermore, the average particle size composition of the soil consists of clay (15.7%), silt (79.1%), and sand (5.2%) [5]. The main land use types in the area are grassland, forest land, and cropland, accounting for 19.62%, 61.52%, and 9.42% of the total trough valley area, respectively.

2.2. Soil Sampling and Preparation

According to the characteristics of vegetation succession in the karst area and previous research findings, the Qingmuguan karst trough valley is classified into five stages. Each stage of vegetation succession in the Qingmuguan trough valley presents distinct habitats [5,16]. Abandoned land represents the initial phase of vegetation succession and arises from the gradual abandonment of cultivated areas, exacerbated by intense hydraulic erosion in karst regions, resulting in soil loss, reduced soil depth, and increased exposure of underlying rock formations. Throughout this process, the depletion of essential nutrient elements such as nitrogen, phosphorus, potassium, and organic matter within the soil leads to a decline in soil fertility, ultimately culminating in abandoned land. In the Abandoned land stage (ALS), there is a lack of ground cover with barren soil and relatively thin soil layer, rendering it unsuitable for biological growth. During the Herb stage (HS), certain herbaceous plants such as Setaria viridis (L.) P. Beauv begin to thrive, gradually improving soil conditions and laying a foundation for further species growth [38]. The Herb-Shrub stage (HES) witnesses an improvement in both soil quality and habitat due to the presence of Imperata cylindric (L.) P. Beauv. As shrubs like Imperata cylindrica (L.) P. Beauv. dominate during the Shrub stage (SHS), habitats continue to improve. Finally, during the Forest stage (FS), tree species like Masson pine emerge Pinus massoniana Lamb are present, leading to resource redistribution and forming a stable community structure [16]. The essential information on vegetation succession are provided in Table 1.

From September to December 2023, a comprehensive field survey of the main vegetation succession in the Qingmuguan karst trough valley of Chongqing City was conducted. The sampling points were selected based on the severity of erosion observed in the remote sensing satellite image, followed by field investigations to confirm their suitability. Then, we carried out a field investigation in Aoki Pass, to assess features including, but not limited to, slope direction, slope, rock formation tendency, soil layer thickness, vegetation succession stage, etc. After the field investigation, we selected five vegetation succession stages in the Qingmuguan karst trough valley area [23]. Additionally, the sampling was conducted in September and the sampling was one-off. This approach is more representative as it aligns with the growth cycle of plants, where sprouting occurs from April to May, followed by the growth season from June to August. The plants mature in September and gradually enter the yellow period from October to November [39]. Therefore, five different vegetation succession stages—Abandoned land stage (ALS), Herb stage (HS), Herb-Shrub stage (HES), Shrub stage (SHS), and Forest stage (FS)—were selected as the research subjects under the same site conditions in erosion and deposition sites. Three replicate plots, each measuring 6 × 6 m, were established for each vegetation succession stage, and within each plot, five sampling points were arranged in an S-shape [10,40]. Before sampling, the surface litter and crust were carefully removed, and the aboveground parts of plants were trimmed. Soil samples were collected using 100 cm³ ring cutters and aluminum boxes to measure soil infiltration, bulk density, capillary porosity, and total porosity. Additionally, 2 kg of loose soil was collected from each of the sampling points, air-dried in plastic bags, and analyzed for soil particle composition, aggregate characteristics, and organic matter content. The soil infiltration rate was determined using the constant head method. Soil bulk density, porosity, and water-holding capacity were measured using the ring cutter method and oven-drying method. Soil aggregates were measured using the wet sieving method, and soil organic matter (SOM) was determined using the potassium dichromate external heating method [20].

Since the sample points selected in this study are located within the same topographic region, namely, the karst trough valley area, their soil types exhibit consistency. Moreover, karst areas suffer from severe rocky desertification and face challenges in soil formation. Furthermore, due to the unique dual-layer geological structure above and below ground level, surface, and underground flow/leakage easily occur in these areas, resulting in a thin layer of soil [21]. Additionally, our research primarily focused on studying the erodibility of different vegetation successions across erosion and deposition sites. The most significant interaction between plants and soil lies within the eluvial horizon; hence, our work predominantly investigated this particular layer [41].

2.3. Determination of Indices of Soil Erodibility

2.3.1. Determination of Soil Infiltration Capacity

In this study, the soil infiltration capacity was measured using a double-ring infiltrometer [16,42]. Four different vegetation successional stages were selected in karst trough valley erosion and deposition sites, with abandoned land serving as the control. Five replicates were conducted for each vegetation type, totaling 75 measurement points. At each site, a flat plot of land was selected for the infiltration experiment. All vegetation was cut at ground level, and the litter layer was carefully removed. PVC pipes with an inner diameter of 16 cm, an outer diameter of 32 cm, and a height of 20 cm were placed concentrically on the soil surface. Each PVC pipe was inserted into the soil to a depth of approximately 10 cm, ensuring that the ring remained level during insertion. The soil surrounding the PVC pipe was compacted to prevent water from leaking out. Water was then simultaneously added to the inner and outer rings until it reached a depth of 5 cm. The time required for the water level in the inner ring to drop by 0.5 cm was recorded using a stopwatch, continuing until the infiltration rate of three consecutive measurements remained constant, indicating a steady-state infiltration rate. Throughout the experiment, the water levels in both the inner and outer rings remained consistent. In this study, the total amount of infiltration within 90 min was the cumulative infiltration volume. The initial infiltration rate (IIR) was calculated based on the first three minutes of the infiltration process. The stable infiltration rate (SIR) was determined by averaging the last three relatively constant infiltration measurements [16].

2.3.2. Determination of Soil Aggregates and Soil Erodibility

The study area experiences severe soil erosion, primarily due to hydraulic erosion. In the dip slope, the bedrock strata dip aligns with the direction of the slope, resulting in enhanced water flow, heightened erosion capacity, and a higher degree of soil erosion. Conversely, the bedrock strata dip on the anti-dip slope exhibits an opposing orientation that impedes water flow to some extent and possesses a slightly lower erosion capacity compared to that of the dip slope. The deposition sites represent a sedimentary depression characterized by a gentle gradient and relatively weaker erosion compared to other areas. Previous research has indicated that intensive human activities and unique natural conditions contribute to significant soil and water loss in the Qingmuguan karst trough valley. Specifically, the Longdong Trough and Laodongcun Depression have accumulated a total sediment deposition of 3754.09 t and 2424.79 t, respectively, while the Changjiawa Basin, situated within the karst depression of the Three Gorges Reservoir area, has experienced sediment deposition amounting to 557.0 t. These figures underscore the alarming extent of soil erosion in this particular area. The soil aggregate stability was measured using the wet sieving method with soil sieves of 0.25 mm, 0.5 mm, 1 mm, 2 mm, and 5 mm [43]. Initially, the weight of the aggregates in each size class was determined using the dry sieving method. Subsequently, a 50 g soil sample was prepared proportionally for wet sieving [39]. The soil sample was placed on the corresponding sieve and wetted using a spray head. It was then immersed in distilled water and subjected to repeated vertical shaking with a 4 cm amplitude for 10 min. After shaking, the aggregates retained on each sieve were washed into aluminum boxes, dried at 105 °C, and weighed. The proportion of material for each sieve size was then calculated [44].

WSAwet is the proportion of >0.25 mm soil aggregates in the total mass of soil aggregates, which have a significantly positive correlation with soil productivity [21,45].

WSAwet = CW/CT × 100%(1)

where CW is the mass weight (g) of wet screened soil and CT is the total aggregate weight (g) of soil.

The value of the soil erodibility factor K reflects the soil erosion ability and damage resistance to external soil erodibility [2].

(2)$\mathrm{K}=7.954\times \left\{0.0017+0.0494\times \exp \left[-0.5\times {\left({\displaystyle \frac{\mathrm{lgGMD}+1.67}{0.6986}}\right)}^2\right]\right\}$

(3)$\mathrm{GMD}=\exp \left[\left({\sum }_{\mathrm{i}}^{\mathrm{n}}{\mathrm{y}}_{\mathrm{i}}\ln {\mathrm{x}}_{\mathrm{i}}\right)/{\sum }_{\mathrm{i}}^{\mathrm{n}}{\mathrm{y}}_{\mathrm{i}}\right]$

MWD (mm) is an important index for evaluating soil aggregate stability. The larger the MWD value, the better the soil water aggregate stability.

(4)$\mathrm{MWD}={\sum }_{\mathrm{i}=1}^{\mathrm{n}}{\mathrm{x}}_{\mathrm{i}}{\mathrm{y}}_{\mathrm{i}}$

where ${\mathrm{x}}_{\mathrm{i}}$ is the mean diameter of soil aggregates with different particle sizes (mm) and ${\mathrm{y}}_{\mathrm{i}}$ is the mass percentage of soil aggregates with different particle sizes (%).

2.4. Data Analysis

A one-way analysis of variance (ANOVA) was used to analyze the soil physicochemical properties under vegetation successional stages. The ANOVA was performed using SPSS software (version 21.0). The potential impacts of interactions among soil physicochemical properties, soil infiltration, and soil aggregate stability on soil erodibility were determined using Structural Equation Modeling (SEM). SEM, also known as covariance structure analysis, is used to analyze complex interrelationships among multiple independent variables or between one or more dependent variables [46]. Based on existing theoretical knowledge, hypothetical conceptual models were developed to elucidate the potential pathways of influence of these key factors on soil erodibility. The goodness of fit of the conceptual models was evaluated using indicators such as the chi-square to degrees of freedom ratio (χ²/df), model p-value (p), Goodness of Fit Index (GFI), Adjusted Goodness of Fit Index (AGFI), and Root Mean Square Error of Approximation (RMSEA). Generally, a model is considered to have good predictive performance if χ²/df is less than 2, p > 0.05, GFI and AGFI are greater than 0.90, and RMSEA is less than 0.05 [46]. Structural Equation Modeling (SEM) was conducted using Amos (version 24).

3. Results

3.1. Soil Physicochemical Properties in Karst Trough Valley Erosion and Deposition Sites

As a reference index for evaluating soil quality, soil physicochemical properties have an important influence on soil infiltration and soil erodibility. In this study, the soil physicochemical properties in the erosion and deposition sites were significantly different. Figure 2 shows that soil total porosity in deposition sites exceeded that of dip slopes and anti-dip slopes in erosion sites by factors of 1.04 and 1.07, respectively. Moreover, the capillary and field water capacities in sedimentary areas surpassed those in erosion areas. Notably, the field water capacity reached 270.69 g·kg⁻¹ in deposition sites, whereas it was only 226.34 g·kg⁻¹ and 251.76 g·kg⁻¹ in dip/anti-dip slopes of erosion sites, indicating superior soil water conditions in deposition site. Generally, soil physicochemical properties were more favorable in deposition sites compared to erosion sites, particularly regarding soil porosity.

In the dip slope of the erosion site, the soil physicochemical properties of the SHS were superior. For example, the capillary porosity and organic matter content, which reached 62.13 g·kg⁻¹, were significantly higher (p < 0.05) compared to those of the other stages. In contrast, the organic matter content of the HES was notably lower at 26.88 g·kg⁻¹, differing by 2.31 times. Noteworthy was the relatively low non-capillary pore space of SHS, measuring only 8.51, followed by the FS, HES, HS, and ALS. Conversely, on the anti-dip slope of the erosion site, the FS exhibited superior soil physicochemical properties, with significantly higher saturation moisture content, capillary moisture content, water retention capacity, non-capillary porosity, and organic matter compared to other plots. The organic matter content of the FS was notably high at 60.09 g·kg⁻¹, surpassing the HS, ALS, HES, and SHS by factors of 1.28, 1.50, 2.86, and 1.48, respectively. In the deposition site, a similar trend to the anti-dip slope of erosion site was observed, with the FS demonstrating superior soil physicochemical properties, notably in saturated water capacity, non-capillary porosity, and total porosity following a descending order of FS, HS, SHS, ALS, and HES. However, a peculiar phenomenon emerged: the ALS exhibited the highest organic matter content, reaching 40.79 g·kg⁻¹, contrary to the erosion site.

3.2. Soil Infiltration Characteristics in Karst Trough Valley Erosion and Deposition Sites

3.2.1. Soil Infiltration Process

The changes in soil infiltration rate over time in erosion and deposition sites, as depicted in Figure 3, can be categorized into three distinct stages: the initial rapid decline stage (0–10 min), followed by a subsequent slow decline stage (10–60 min), and finally reaching a stable stage (60–90 min). The decrease in soil infiltration rate was more pronounced in the erosion site compared to the deposition site, with average reductions of 37.3% and 41.0%, respectively. Additionally, the erosion and deposition sites experienced the greatest decline in soil infiltration rate in the ALS, with an average decrease of 6.60 mm·min⁻¹.

In the erosion site, the average decrease in soil infiltration rate on the dip slope, from high to low, was observed as follows: HS, FS, HES, SHS, and ALS. During the rapid decline stage, the ALS exhibited the smallest decline range of only 4.3%, while the HS had the largest decline range reaching up to 37.7%. Notably, the decline range of the HS was 8.77 times that of the ALS. A significant finding is that the inverse slope of the ALS layer and the layer with different slopes exhibited varying rates of decrease in soil infiltration. Notably, the inverse slope of the ALS layer experienced the most rapid decline, reaching up to 54.3%. Furthermore, during the stage of rapid infiltration, both the ALS and HS demonstrated substantial reductions of 35.2% and 38.6%, respectively, while the SHS showed a comparatively smaller decrease of 21.1%. In the deposition site, the average infiltration rate in the FS experienced the most significant decrease, reaching 67.9%, despite its initial infiltration rate being only 3.1 mm·min−1. In contrast, the ALS exhibited a relatively smaller reduction of 37.6%, but had an initially high infiltration rate of 22.7 mm·min⁻¹. Furthermore, the HS demonstrated the least decline in average soil infiltration rate at 13.5%. During the rapid decline stage, the FS displayed the largest decrease of 37.1%. It is noteworthy that the HS did not experience a decrease in soil infiltration rate and even showed a slight increase of 1.6%. The differences observed during the remaining stages were not statistically significant.

3.2.2. Soil Infiltration Index

The soil infiltration characteristics in the karst trough erosion deposition area are influenced by vegetation successional stages (Figure 4). The soil infiltration characteristics of the HES, SHS, and HS in the erosion site exhibited significantly (p < 0.05) higher performance compared to those in the deposition site. Overall, the soil infiltration characteristics in the erosion site surpassed those observed in the deposition site.

In the erosion site, the soil permeability of the dip slope area, in order from best to worst, was ALS, HS, HES, SHS, and FS. The initial infiltration rate, stable infiltration rate, total infiltration amount, and average infiltration rate of the ALS were significantly higher compared to the HS, HES, SHS, and FS. Specifically, the initial infiltration rate, stable infiltration rate, total infiltration amount, and average infiltration rate of the ALS were 5.53, 8.36, 7.78 and 7.77 times, respectively. The soil permeability of the anti-dip slope is also the best in the ALS and the worst in the HES and HS. The results depicted in Figure 2 demonstrate that the initial infiltration rate, stable infiltration rate, total infiltration amount, and average infiltration rate of the ALS were significantly higher than those of the HS. Specifically, they were 3.13, 3.42, 3.21 and 3.2 times greater than the HS’s rates, respectively. Conversely, there was minimal disparity observed in soil permeability among other vegetation successional stages. In the deposition site, the order of soil permeability in different vegetation successional stages was as follows: ALS > SHS > HS > FS > HES. As depicted in Figure 3, the soil permeability of the ALS was significantly higher than that of other plots, particularly the HES. The average infiltration rate of the ALS was 13.32, whereas the average infiltration rate of the HES was merely 1.69. Furthermore, the initial infiltration rate, stable infiltration rate, and total infiltration rate of the ALS were significantly higher than those of the HES; they were, respectively, 6.57 times, 7.75 times, and 7.897 times greater than those observed in the HES. There were no significant differences in soil permeability among the FS, SHS, and HS.

3.3. The Content of Water-Stable Aggregates in Karst Trough Valley Erosion and Deposition Sites

Soil water-stable aggregates significantly affect soil anti-erodibility, with higher contents of >0.25 mm soil water-stable aggregates indicating stronger soil anti-erodibility. Figure 5 shows the distribution of soil water-stable aggregate content in the karst trough erosion and deposition sites. The deposition site exhibited a significantly higher content of >5 mm water-stable aggregates compared to the erosion site, showing a 1.58-fold increase. Conversely, the dip slope of the erosion site had the lowest content (only 12.6%) of >5 mm soil water-stable aggregates. Additionally, the highest percentage (40.01%) of <0.25 mm water-stable aggregates was observed in the dip slope within the erosion site, while it reached its lowest value (32.71%) in the deposition area. No significant differences were found among other particle sizes for water-stable aggregates.

In the erosion site, the content of >5 mm water-stable aggregates in the HS of the dip slope was as high as 21.18%, which was significantly higher than that of other samples and 6.62 times that of the HES. The order of the contents of >0.25 mm water-stable aggregates from high to low was HS, FS, ALS, HES, and SHS. Additionally, the content of water-stable aggregates with a size smaller than 0.25 mm in the SHS was the highest at 67.84%, which was 1.30, 1.65, 3.30 and 3.67 times higher that of the HES, ALS, FS, and HS. The content of >5 mm water-stable aggregates in the dip slope with the ALS was significantly higher compared to other plots, reaching up to 27.3%, which was 13.13 times greater than that in the HES. At the same time, in the anti-dip slope, the highest > 0.25 mm water-stable aggregate content was observed in the HS (85.5%), followed by the ALS (83.62%) and HES (29.44%). In the deposition site, the >5 mm water-stable aggregates content was observed as follows: HS > FS > ALS > HES, SHS. In contrast to the erosion site, the highest percentage (80.82%) of >0.25 mm water-stable aggregates was found in the FS, while the lowest percentage (50.24%) was observed for SHS. Furthermore, the SHS and HES contents of <0.25 mm water-stable aggregates were found to be significantly higher in comparison to other plots, but still lower than that in the erosion site. No significant differences were observed among other sizes of water-stable aggregates.

3.4. Soil Erodibility Factor K in Karst Trough Valley Erosion and Deposition Sites

The value of the soil erodibility factor K reflects the soil erosion ability and damage resistance to external erodibility. Figure 6 below illustrates the K values of soil corrosion in the distribution of the erosion and deposition sites. The average value of the soil erodibility factor K in the dip slope of the erosion site was as high as 0.064, while it was only 0.051 in the deposition site. Furthermore, the maximum K value of soil erodibility was observed in the HES of the anti-dip slope, reaching 0.142, which was 4.73 times higher than that of the FS in the deposition site. Overall, there was a significant disparity between the K values of soil erodibility in the erosion and deposition sites.

In the erosion site, the K value of the dip slope was higher than that of the anti-dip slope. The highest K value of soil erodibility was 0.087 in the HES of the dip slope, which was significantly higher than the other plots except the SHS. The K values of soil erodibility from high to low were HES, SHS, ALS, FS, and HS. The distribution of soil erodibility K value in the anti-dip slope was similar to that in the dip slope, and the HES was significantly higher than that in the other plots. The K value of the HES was 0.142, which was 1.97, 3.33, 4.38, and 4.65 times higher than that of the SHS, ALS, FS, and HS.

In the deposition site, the K value of soil erodibility exhibited a similar distribution pattern as the erosion site, with the HES showing significantly higher values compared to other plots. The specific distribution order was observed as HES > SHS > ALS > HS > FS. Notably, the soil erodibility of the HES (0.087) was 2.89 times greater than that of the FS, while no significant differences were found among the remaining plots.

3.5. The Effects of Soil Properties and Soil Infiltration on Soil Erodibility

Soil erodibility is significantly influenced by soil physicochemical properties, as well as soil aggregates. A Spearman correlation analysis revealed a negative correlation between soil erodibility and capillary water capacity, field water capacity, geometric mean diameter (GMD), and organic matter content (p < 0.05). These findings suggest that greater capillary water capacity, field water capacity, GMD, and organic matter content are associated with reduced soil erodibility (Figure 7).

In order to investigate the relationship between soil infiltration, soil erodibility, and their influencing factors, the entropy method was employed to standardize soil physicochemical properties as well as soil infiltration characteristics, and then a structural equation model was established (Figure 8). As presented in Table 2, the goodness of fit indices of this model were within an acceptable range and all these fit indices suggest a satisfactory and robust model fit. The results from the structural equation analysis revealed that erosion and deposition topographies, vegetation successional stages, soil physicochemical properties, soil aggregates, and soil infiltration characteristics collectively accounted for 88% of the variation in soil erodibility under different conditions.

The path coefficients of the structural equation model are shown in Figure 9. The effects of vegetation successional stages on soil aggregates and aggregates on soil erodibility reached a significant level. The results showed that erosion and deposition topographies, vegetation successional stages, soil physicochemical properties, soil aggregate, and soil infiltration characteristics all had direct effects on soil erodibility, especially when the path coefficient of soil aggregate reached −0.90 (p < 0.05). In addition, we found that erosion and deposition topographies and vegetation successional stages had direct effects on soil physicochemical properties, soil aggregates, and soil infiltration characteristics. Vegetation successional stage had a significant effect on soil aggregates (p < 0.05), with a path coefficient of −0.69. It is worth noting that the erosion and deposition topographies and vegetation successional stage had indirect effects on soil erodibility with path coefficients of 0.14 and 0.62, respectively.

4. Discussion

4.1. Changes in Soil Inflation Characteristics and Soil Erodibility Under Different Erosion and Deposition Topographies

The bedrock strata dip plays a crucial role in regulating soil particle transfer and water loss at the karst trough valley erosion and deposition sites. It not only alters the contact angle between soils, but also redirects surface runoff and underground leakage [7,21]. In this study, the infiltration characteristics and erodibility of soil varied significantly across different environments. The infiltration characteristics of the soil at the deposition site were superior to those at the erosion site, whereas the erodibility coefficient K demonstrated contrasting attributes. Regional precipitation may contribute to accelerated soil erosion, resulting in the displacement of soil particles on slope layers towards trough valley depressions [9,10]. This phenomenon leads to variations in the soil properties of different depressions, which subsequently affect infiltration characteristics and soil erodibility [47]. There is a tendency of erosion sites, bedrock strata, and slopes to accelerate or hinder surface runoff at the deposition site, that is, to cause the surface runoff velocity to be slow [5]. The faster the surface runoff is, the worse the soil infiltration characteristics and the greater the degree of soil erodibility [48]. In addition, bedrock strata can also affect soil infiltration and erodibility by affecting key factors such as soil bulk density, porosity, and organic matter. Compared with the anti-dip slopes, dip slopes have greater bulk density and lower porosity and organic matter content. This may be due to the opposite rock tendency and slope direction of the anti-dip slope, which has a blocking effect on surface runoff, causing the runoff to penetrate into the ground through the cracks in the rock layer. Therefore, in most cases of sediment deposition at underground erosion sites, the soil aggregate content tends to be low. A higher soil aggregate content reduces the clay content and consequently increases soil erosion [7,18,41]. Moreover, variations in soil infiltration result in changes in physicochemical properties that subsequently affect soil erodibility [44,49]. Studies have shown that increased soil infiltration capacity leads to increased soil water content and prolonged soil water retention time, which is more conducive to the leaching of soil materials, leading to an increase in the soil organic matter content [50]. The soil water content and soil organic matter content are significantly positively related. The cementation of soil organic matter affects soil erodibility, with a relatively high soil organic matter content decreasing soil erodibility [19,25,51].

4.2. Responses of Soil Infiltration and Soil Erodibility to Vegetation Successional Stages

The soil infiltration characteristics and soil erodibility can vary significantly under different vegetation successional stages. A previous study has indicated that vegetation successional stages could alter the functioning of ecosystem services, thereby influencing soil physicochemical properties [23,26,30]. Consequently, this ultimately affects the soil infiltration capacity either directly or indirectly [26]. The findings of this study indicate that the ALS exhibited the highest infiltration capacity at both the erosion and deposition sites, while the FS demonstrated the lowest infiltration capacity specifically at the erosion site, and the HES showed reduced infiltration capacity particularly at the deposition site. These results are consistent with Gan’s previous research [16]. However, the research findings of Yan indicate that vegetation in karst regions can enhance soil structure and improve soil infiltration capacity, which may be due to differences in soil thickness and geological structure [1]. In this study, the study area featured a substantial soil layer with a high gravel content, increasing the porosity of the soil [47]. This increase in porosity subsequently enhanced the soil’s capacity to store water and improved its infiltration capabilities. Notably, the gravel content of the ALS surpassed that of the other samples, further substantiating its significant contribution to the enhancement of soil water infiltration capacity [48]. Therefore, the superior infiltration capacity of the ALS soil may be attributed to its high porosity, significant gravel content, and wide range of coarse particles [16]. However, under such circumstances, fine sand particles are prone to being lost through pores, thereby accelerating soil erosion and subsequently leading to the degradation in soil quality [42]. The process of soil degradation results in a reduction in the soil organic matter content, the collapse of soil aggregates, decreased soil stability, and increased susceptibility to erosion, thereby leading to increased soil erodibility [22]. At the karst trough valley erosion and deposition sites, the infiltration capacity of the four vegetation successional stages was lower than that of the ALS. The relatively poor infiltration capacity of the vegetation successional stages, such as the FS and SHS, may be because there are fewer large particles in the soils of the FS and SHS, which leads to a decrease in soil porosity and a reduction in soil infiltration efficiency [47,49]. Additionally, the FS, SHS, and HS exhibited minimal susceptibility to anthropogenic activities because of their abundant vegetation coverage and well-developed subterranean root systems [47]. The biochemical impacts of root exudates can increase soil adhesion, stabilize organic matter and aggregates within the soil matrix, and diminish soil porosity [18,43]. Consequently, this leads to a reduction in infiltration capacity and erodibility [20]. Moreover, the roots of the FS, HS, and SHS accumulate and intercept soil particles, resulting in soil particle swelling and a decrease in soil permeability [23]. Notably, the soil erodibility in the HES appeared to be significantly greater. This could be attributed to regular grazing activities leading to interference and negative impacts on soil properties such as nutrient content, aggregate stability, organic matter content, and porosity [12,39,49]. The grazing activities in the HES region have led to an increase in soil permeability and porosity, accelerated mineralization of organic matter, and enhanced crushing of soil aggregates. As a result, these factors contribute to the loosening of surface soil, ultimately affecting its erodibility [22,28].

4.3. Relationship Between Soil Corrosion and Its Influencing Factors

The results of structural equation modelling revealed that soil erodibility under different vegetation successional stages was influenced primarily by aggregate content, which exhibited a significant direct negative impact on soil erodibility. The standardized path coefficient of the soil aggregate was −0.9. This indicates a significant correlation between soil aggregates and soil erosion, which is consistent with the findings of Gan et al. [21]. In general, there exists a significant correlation between aggregate content and soil properties in terms of their influence on erosion susceptibility [18,20]. Higher levels of soil aggregates are associated with reduced vulnerability to erosion due to improved water stability and increased clay and silt particle contents, resulting in enhanced water adsorption capacities and more compact soil structures [10,37]. The results presented in Figure 6 demonstrate the significant negative impact of vegetation successional stage on soil aggregates, as indicated by a standardized path coefficient of −0.69. Notably, vegetation successional stage exerts a significant indirect in-fluence on soil erodibility, with a path coefficient of 0.62. This indicates that soil aggregates and soil erodibility can have a significant relationship, which is consistent with the results of Peng et al. [44]. This may be due to changes in soil physicochemical properties, soil aggregates, and soil infiltration that can be caused by vegetation successional stages [6,26,45]. The diversification of plant species resulted in an increase in litter accumulation in the FS, SHS and HS. This subsequently increased the input of soil organic matter into the soil matrix, promoted organic cementation between soil particles, increased the content of soil macroaggregates, and ultimately reduced soil erodibility [19,27]. In addition, the intermingling of the FS, SHS, and HS roots was advantageous for increasing soil porosity, thereby increasing soil permeability [50]. Furthermore, the exudates released by plant roots facilitates cementation between soil aggregate particles [13,21]. However, the HES and ALS exhibited sparse vegetation cover, shallow root systems, and fragmented soil aggregates with low contents of large aggregates due to artificial tillage practices [10,50]. These factors contributed to the observed increase in soil erodibility in this study. The direct effect of soil infiltration on soil erosion was found to be nonsignificant, with a standardized path coefficient of 0.05. Previous research has demonstrated the correlation between soil infiltration and soil erosion, as both are influenced primarily by surface soil dispersion [51,52]. Consequently, as the erosion degree of infiltration and its velocity decrease, the HES clearly had the highest soil erodibility and a lower soil infiltration capacity, which aligns with previous findings [16]. This could be attributed to tillage-induced destruction of the soil structure, leading to crust formation in the surface layer, which diminishes its ability to infiltrate water, and subsequently increases runoff and exacerbates soil erosion [4]. Soil erodibility is influenced by a multitude of factors, including the physicochemical properties of the soil, its infiltration capacity, the formation of soil aggregates, and the presence and activity of soil microorganisms [4,12,53,54]. Hence, there are several significant challenges that need to be addressed. Firstly, our exploration of soil erodibility was limited to the Qingmuguan karst trough valley; therefore, future studies should expand this investigation to encompass the entire karst region and even a global scale in order to draw more robust conclusions. Secondly, we did not consider the influence of plant roots on soil erodibility in our study of this region. It is important to note that plant roots can impact soil erodibility by altering soil physicochemical properties and aggregate stability. Hence, future research should further investigate the influence of plant roots on soil erodibility. In addition, the unique topographic structure of the karst trough valley area, characterized by a dual aboveground and belowground configuration, results in soil erosion not only on the surface, but also underground. Monitoring underground loss poses significant challenges [8]. Therefore, for further investigations, we will conduct indoor tests and simulate outdoor conditions to comprehensively assess soil erosion in this region. Hence, in any subsequent study conducted on this subject matter, it is imperative that all these contributing factors are comprehensively considered to establish a solid foundation for effective soil erosion control and ecological restoration at karst trough valley erosion and deposition sites [6,8,9].

5. Conclusions

The present study has shown that the soil physicochemical properties and soil aggregate stability at the deposition site were significantly superior to those at the erosion site. The FS resulted in the best soil physicochemical properties, whereas the HS resulted in the highest soil aggregate stability within the deposition site. However, the soil infiltration capacity at the erosion site was significantly greater than that at the deposition sites. The ALS had the strongest soil infiltration capacity at both the erosion and deposition sites. The soil erodibility at the erosion sites (0.064) was significantly greater than that at the deposition sites (0.051), with the highest soil erodibility observed on the anti-dip slopes during the HES at the erosion sites (0.142). The structural equation model reveals that erosion and deposition topographies, vegetation succession, soil physicochemical properties, soil aggregates, and soil infiltration characteristics collectively account for 88% of the variation in soil erodibility under different conditions. Specifically, both direct and indirect influences on soil erodibility are most significantly exerted by soil aggregate stability and vegetation succession.

Author Contributions

Conceptualization, X.T., Y.F. and J.P.; Methodology, H.S., F.G., L.J., D.L. and Y.Y.; Software, H.S. and F.G.; Validation, H.S.; Formal analysis, H.S. and F.G.; Investigation, H.S.; Data curation, H.S.; Writing—original draft, H.S.; Project administration, J.P.; Funding acquisition, F.G. All authors have read and agreed to the published version of the manuscript.

Data Availability Statement

The data that have been used are confidential.

Conflicts of Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Footnotes

Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Figures and Tables

Figure 1. Location of the study area and photos of the experimental sites.

Figure 2. Soil physicochemical properties in erosion and deposition sites. Note: Different lowercase letters indicate significant differences between groups in the same column (p [less than] 0.05). Abandoned land stage (ALS), Herb stage (HS), Herb Shrub stage (HES), Shrub stage (SHS), and Forest stage (FS). The lines in the figure represent the average values of each index at different vegetation succession stages, illustrating their changing trends across vegetation succession stages. Lines of varying colors correspond to the dip slope (Erosion site), anti-dip slope (Erosion site), and deposition site.

Figure 3. Soil infiltration process in erosion and deposition sites.

Figure 3. Soil infiltration process in erosion and deposition sites.

Figure 4. Soil infiltration characteristics in erosion and deposition sites.

Figure 5. The content of water-stable aggregates in erosion and deposition sites.

Figure 5. The content of water-stable aggregates in erosion and deposition sites.

Figure 6. The value of the soil erodibility factor K in erosion and deposition sites. Note: Different lowercase letters indicate significant differences between groups in the same column (p [less than] 0.05).

Figure 7. Spearman correlation between soil erodibility and their influencing factors. Note: CMC is capillary moisture capacity, SMC is saturation moisture capacity, TP is total porosity, FC is field capacity, BD is bulk density, SIR is stable infiltration rate, CP is capillary porosity, NCP is non-capillary porosity, GMD is geometric mean diameter, OM is organic matter, IIR is initial infiltration rate, TI is total infiltration amount, AIR is average infiltration rate, and K is soil erodibility. * represents significant correlation and ** represents extremely significant correlation.

Figure 8. Effect of the interactions between erosion and deposition positions, vegetation successional stages, soil physicochemical properties, soil aggregate, and soil infiltration characteristics on soil erodibility. Note: CMC is capillary moisture capacity, SMC is saturation moisture capacity, TP is total porosity, FC is field capacity, BD is bulk density, WSA is the proportion of >0.25mm soil aggregates in the total mass of soil aggregates, MWD is mean weight diameter, GMD is geometric mean diameter, SIR is stable infiltration rate, CP is capillary porosity, NCP is non-capillary porosity, OM is organic matter, IIR is initial infiltration rate, TI is total infiltration amount, AIR is average infiltration rate. ** represents extremely significant correlation.

Figure 9. The standardized path coefficients for direct, indirect, and total effects of erosion and deposition topographies, vegetation successional stages, soil physicochemical properties, soil aggregate, and soil infiltration characteristics on soil erodibility.

Table 1

Basic information on erosional and deposition sites.

EADT	BSD	VSS	BSDA(°)	Latitude(N)	Longitude(E)	Slope(°)	Altitude(m)	Soil Depth (cm)
		ALS	55	29.629	106.267	35	587.6	20–50
	Dip	HS	45	29.628	106.268	24	533.1	50–60
	erosion slope	HES	30	29.632	106.271	33	511.3	30–40
		SHS	20	29.629	106.268	27	511.7	40–50
		FS	18	29.632	106.27	31.5	522.1	55–65
Erosion
site		ALS	40	29.714	106.301	39	512.2	30–60
	Anti-dip	HS	34	29.714	106.298	36	504.9	40–60
	erosion slope	HES	25	29.714	106.3	32	522.1	50–60
		SHS	22	29.714	106.299	35.5	535.1	40–50
		FS	20	29.714	106.302	31	525.7	50–60
		ALS	3	29.725	106.305	2	473.8	40–60
Deposition		HS	5	29.724	106.304	3.5	470	50–60
sites	Valley	HES	3	29.725	106.306	2.5	465.4	45–55
		SHS	2	29.724	106.305	4	468.7	40–60
		FS	4	29.725	106.306	3	470.5	55–65

Note: EADT is erosion and deposition topography, BSD is bedrock strata dip, VSS is vegetation succession stage, and BSDA is bedrock strata dip angle.

Table 2

Fitting coefficients of the soil erodibility and its impact factors model.

Indices Name	χ2/df	GFI	TLI	RMSEA	IFI	CFI
Evaluation criterion	<3	Close to 1	Close to 1	<0.08	Close to 1	Close to 1
Results	1.22	0.97	0.98	0.05	0.99	0.99

References

1. Yan, Y.; Hu, Z.; Wang, L.; Jiang, J.; Dai, Q.; Gan, F.; Almaliki, A.H.; Hashim, M.A.; Hussein, E.E.; Ghoneim, S.S.M. Impact of extreme rainfall events on soil erosion on karst Slopes: A study of hydrodynamic mechanisms. J. Hydrol.; 2024; 638, 131532. [DOI: https://dx.doi.org/10.1016/j.jhydrol.2024.131532]

2. Yang, M.; Yang, Q.; Zhang, K.; Pang, G.; Huang, C. Global soil erodibility factor (K) mapping and algorithm applicability analysis. Catena; 2024; 239, 107943. [DOI: https://dx.doi.org/10.1016/j.catena.2024.107943]

3. Li, Z.-w.; Zhang, G.-h.; Geng, R.; Wang, H. Rill erodibility as influenced by soil and land use in a small watershed of the Loess Plateau, China. Biosyst. Eng.; 2015; 129, pp. 248-257. [DOI: https://dx.doi.org/10.1016/j.biosystemseng.2014.11.002]

4. Chen, L.; Li, Y.; Zhang, Z. Impact of land use type and slope position on the erodibility of karst hillslopes in Southwest China. Catena; 2023; 233, 107498. [DOI: https://dx.doi.org/10.1016/j.catena.2023.107498]

5. Gan, F.; He, B.; Qin, Z. Hydrological response and soil detachment rate from dip/anti-dip slopes as a function of rock strata dip in karst valley revealed by rainfall simulations. J. Hydrol.; 2020; 581, 124416. [DOI: https://dx.doi.org/10.1016/j.jhydrol.2019.124416]

6. Bi, M.; Zhang, S.; Xu, Q.; Hou, S.; Han, M.; Yu, X. Coupling and synergistic relationships between soil aggregate stability and nutrient stoichiometric characteristics under different microtopographies on karst rocky desertification slopes. Catena; 2024; 243, 108142. [DOI: https://dx.doi.org/10.1016/j.catena.2024.108142]

7. Gan, F.; He, B.; Qin, Z.; Li, W. Role of rock dip angle in runoff and soil erosion processes on dip/anti-dip slopes in a karst trough valley. J. Hydrol.; 2020; 588, 125093. [DOI: https://dx.doi.org/10.1016/j.jhydrol.2020.125093]

8. Ran, C.; Bai, X.; Tan, Q.; Luo, G.; Cao, Y.; Wu, L.; Chen, F.; Li, C.; Luo, X.; Liu, M. et al. Threat of soil formation rate to health of karst ecosystem. Sci. Total Environ.; 2023; 887, 163911. [DOI: https://dx.doi.org/10.1016/j.scitotenv.2023.163911]

9. Gan, F.; Shi, H.; Yan, Y.; Pu, J.; Dai, Q.; Gou, J.; Fan, Y. Soil quality assessment of karst trough valley under different bedrock strata dip and land-use types, based on a minimum data set. Catena; 2024; 241, 108048. [DOI: https://dx.doi.org/10.1016/j.catena.2024.108048]

10. Liu, S.; Wang, R.; Yang, Y.; Shi, W.; Jiang, K.; Jia, L.; Zhang, F.; Liu, X.; Ma, L.; Li, C. et al. Changes in soil aggregate stability and aggregate-associated carbon under different slope positions in a karst region of Southwest China. Sci. Total Environ.; 2024; 928, 172534. [DOI: https://dx.doi.org/10.1016/j.scitotenv.2024.172534]

11. Gayen, A.; Haque, S.M. Soil erodibility assessment of laterite dominant sub-basin watersheds in the humid tropical region of India. Catena; 2022; 213, 106161. [DOI: https://dx.doi.org/10.1016/j.catena.2022.106161]

12. Yamaguchi, A.; Kanashiki, N.; Ishizaki, H.; Kobayashi, M.; Osawa, K. Relationship between soil erodibility by concentrated flow and shear strength of a Haplic Acrisol with a cationic polyelectrolyte. Catena; 2022; 217, 106506. [DOI: https://dx.doi.org/10.1016/j.catena.2022.106506]

13. Dai, W.; Feng, G.; Huang, Y.; Adeli, A.; Jenkins, J.N. Influence of cover crops on soil aggregate stability, size distribution and related factors in a no-till field. Soil Tillage Res.; 2024; 244, 106197. [DOI: https://dx.doi.org/10.1016/j.still.2024.106197]

14. Wang, D.; Niu, J.; Yang, T.; Miao, Y.; Zhang, L.; Chen, X.; Fan, Z.; Dai, Z.; Wu, H.; Yang, S. et al. Soil water infiltration characteristics of reforested areas in the paleo-periglacial eastern Liaoning mountainous regions, China. Catena; 2024; 234, 107613. [DOI: https://dx.doi.org/10.1016/j.catena.2023.107613]

15. Chen, X.; Zhou, J. Volume-based soil particle fractal relation with soil erodibility in a small watershed of purple soil. Environ. Earth Sci.; 2013; 70, pp. 1735-1746. [DOI: https://dx.doi.org/10.1007/s12665-013-2261-y]

16. Gan, F.; Shi, H.; Gou, J.; Zhang, L.; Liu, C. Effects of bedrock strata dip on soil infiltration capacity under different land use types in a karst trough valley of Southwest China. Catena; 2023; 230, 107253. [DOI: https://dx.doi.org/10.1016/j.catena.2023.107253]

17. Wang, B.; Liu, J.; Li, Z.; Morreale, S.J.; Schneider, R.L.; Xu, D.; Lin, X. The contributions of root morphological characteristics and soil property to soil infiltration in a reseeded desert steppe. Catena; 2023; 225, 107020. [DOI: https://dx.doi.org/10.1016/j.catena.2023.107020]

18. Li, H.; Zhang, Q.; Wu, J.; Zou, H.; Xia, X.; Dan, C.; Liu, C.; Guo, Z.; Zhang, Y.; Liu, G. Response of soil aggregate disintegration to antecedent moisture during splash erosion. Catena; 2024; 234, 107633. [DOI: https://dx.doi.org/10.1016/j.catena.2023.107633]

19. Guo, G.; Kong, Y.; Xu, Y.; Peng, X.; Niu, M.; Zeng, G.; Ouyang, Z.; Liu, J.; Zhang, C.; Lin, J. Soil organic matter components and sesquioxides integrally regulate aggregate stability and size distribution under erosion and deposition conditions in southern China. J. Hydrol.; 2024; 639, 131588. [DOI: https://dx.doi.org/10.1016/j.jhydrol.2024.131588]

20. Chen, L.; Zhang, K.; Li, Y. Spatial variations in soil erodibility induced by rock outcropping on sloping cropland in the karst region of Southwest China. Geoderma; 2023; 440, 116705. [DOI: https://dx.doi.org/10.1016/j.geoderma.2023.116705]

21. Gan, F.; Shi, H.; Gou, J.; Zhang, L.; Dai, Q.; Yan, Y. Responses of soil aggregate stability and soil erosion resistance to different bedrock strata dip and land use types in the karst trough valley of Southwest China. Int. Soil Water Conserv. Res.; 2024; 12, pp. 684-696. [DOI: https://dx.doi.org/10.1016/j.iswcr.2023.09.002]

22. Chen, S.; Zhang, G.; Zhu, P.; Wang, C.; Wan, Y. Impact of land use type on soil erodibility in a small watershed of rolling hill northeast China. Soil Tillage Res.; 2023; 227, 105597. [DOI: https://dx.doi.org/10.1016/j.still.2022.105597]

23. Li, Z.-W.; Zhang, G.-H.; Geng, R.; Wang, H.; Zhang, X.C. Land use impacts on soil detachment capacity by overland flow in the Loess Plateau, China. Catena; 2015; 124, pp. 9-17. [DOI: https://dx.doi.org/10.1016/j.catena.2014.08.019]

24. Ma, Y.; Liu, Y.; Comino, J.R.; Vicente, M.L.; Shi, Z.; Wu, G.L. Artificially cultivated grasslands decrease the activation of soil detachment and soil erodibility on the alpine degraded hillslopes. Soil Tillage Res.; 2024; 243, 106176. [DOI: https://dx.doi.org/10.1016/j.still.2024.106176]

25. Guo, M.; Chen, Z.; Wang, W.; Wang, T.; Wang, W.; Cui, Z. Revegetation induced change in soil erodibility as influenced by slope situation on the Loess Plateau. Sci. Total Environ.; 2021; 772, 145540. [DOI: https://dx.doi.org/10.1016/j.scitotenv.2021.145540]

26. Zhang, M.; Dong, L.; Wang, Y.; Bai, X.; Ma, Z.; Yu, X.; Zhao, Z. The response of soil microbial communities to soil erodibility depends on the plant and soil properties in semiarid regions. Land Degrad. Dev.; 2021; 32, pp. 3180-3193. [DOI: https://dx.doi.org/10.1002/ldr.3887]

27. Arunrat, N.; Sereenonchai, S.; Kongsurakan, P.; Hatano, R. Soil organic carbon and soil erodibility response to various land-use changes in northern Thailand. Catena; 2022; 219, 106595. [DOI: https://dx.doi.org/10.1016/j.catena.2022.106595]

28. Chen, S.; Zhang, G.; Zhu, P.; Wang, C.; Wan, Y. Impact of slope position on soil erodibility indicators in rolling hill regions of northeast China. Catena; 2022; 217, 106475. [DOI: https://dx.doi.org/10.1016/j.catena.2022.106475]

29. Zhou, P.; Guo, M.; Zhang, X.; Zhang, S.; Qi, J.; Chen, Z.; Wang, L.; Xu, J. Quantifying the effect of freeze–thaw on the soil erodibility of gully heads of typical gullies in the Mollisols region of Northeast China. Catena; 2023; 228, 107180. [DOI: https://dx.doi.org/10.1016/j.catena.2023.107180]

30. Zhang, B.-j.; Zhang, G.-h.; Yang, H.-y.; Wang, H. Soil resistance to flowing water erosion of seven typical plant communities on steep gully slopes on the Loess Plateau of China. Catena; 2019; 173, pp. 375-383. [DOI: https://dx.doi.org/10.1016/j.catena.2018.10.036]

31. Luo, Z.; Lian, J.; Nie, Y.; Zhang, W.; Wang, F.; Huang, L.; Chen, H. Improving soil thickness estimations and its spatial pattern on hillslopes in karst forests along latitudinal gradients. Geoderma; 2024; 441, 116749. [DOI: https://dx.doi.org/10.1016/j.geoderma.2023.116749]

32. Moraes, J.F.S.; Cantalice, J.R.B.; Singh, V.P.; Piscoya, V.C.; Cunha Filho, M.; Guerra, S.M.; Almeida, P.L.R. Lateral sediment connectivity by curve number and a proposed approach to soil erodibility at the watershed scale. Catena; 2024; 234, 107611. [DOI: https://dx.doi.org/10.1016/j.catena.2023.107611]

33. Sirjani, E.; Sameni, A.; Mahmoodabadi, M.; Moosavi, A.A.; Laurent, B. In-situ wind tunnel experiments to investigate soil erodibility, soil fractionation and wind-blown sediment of semi-arid and arid calcareous soils. Catena; 2024; 241, 108011. [DOI: https://dx.doi.org/10.1016/j.catena.2024.108011]

34. Sun, L.; Liu, F.; Zhu, X.; Zhang, G. High-resolution digital mapping of soil erodibility in China. Geoderma; 2024; 444, 116853. [DOI: https://dx.doi.org/10.1016/j.geoderma.2024.116853]

35. Lu, S.; Liu, M.; Yi, J.; Li, S.; Xu, Y.; Zhang, H.; Ding, F. Lateral partition patterns and controlling factors of soil infiltration at a steep, near-stream, and humid hillslope scale. Catena; 2024; 239, 107917. [DOI: https://dx.doi.org/10.1016/j.catena.2024.107917]

36. MakMensah, E.; Zhao, W.; Zhou, X.; Zhang, D.; Zhao, X.; Wang, Q.; Obour, P.B. Impact of maize straw biochar and tied-ridge-furrow rainwater harvesting on soil erosion and soil quality in a semiarid region. Soil Use Manag.; 2023; 39, pp. 1304-1320. [DOI: https://dx.doi.org/10.1111/sum.12880]

37. Chen, Z.; Guo, M.; Wang, W.; Wang, W.; Feng, L. Response of soil erodibility of permanent gully heads to revegetation along a vegetation zone gradient in the loess-table and gully region of the Chinese Loess Plateau. Sci. Total Environ.; 2023; 892, 164833. [DOI: https://dx.doi.org/10.1016/j.scitotenv.2023.164833]

38. Zeng, C.; Li, T.; He, B.; Feng, M.; Liang, K.; Xu, Q.; Zhao, X.J.C. Vegetation succession increases soil organic carbon density and decreases soil erodibility: Evidence from a karst trough valley experiencing farmland abandonment. Catena; 2024; 246, 108359. [DOI: https://dx.doi.org/10.1016/j.catena.2024.108359]

39. van den Bergh, S.G.; Chardon, I.; Leite, M.F.A.; Korthals, G.W.; Mayer, J.; Cougnon, M.; Reheul, D.; de Boer, W.; Bodelier, P.L.E. Soil aggregate stability governs field greenhouse gas fluxes in agricultural soils. Soil Biol. Biochem.; 2024; 191, 109354. [DOI: https://dx.doi.org/10.1016/j.soilbio.2024.109354]

40. Wang, L.; Guo, M.; Chen, Z.; Zhang, X.; Zhou, P.; Liu, X.; Qi, J.; Wan, Z.; Xu, J.; Zhang, S. Quantifying the contributions of factors influencing the spatial heterogeneity of soil aggregate stability and erodibility in a Mollisol watershed. Catena; 2024; 239, 107941. [DOI: https://dx.doi.org/10.1016/j.catena.2024.107941]

41. Yang, W.; Peng, X.; Dai, Q.; Li, C.; Xu, S.; Liu, T. Storage infiltration of rock-soil interface soil on rock surface flow in the rocky desertification area. Geoderma; 2023; 435, 116512. [DOI: https://dx.doi.org/10.1016/j.geoderma.2023.116512]

42. Zhu, P.; Chen, S.; Wang, C.; Xu, Z. Short-term conservation tillage degrades soil infiltration properties in the black soil region of Northeast China. Eur. J. Soil Sci.; 2024; 75, e13479. [DOI: https://dx.doi.org/10.1111/ejss.13479]

43. Liu, J.; Yang, Y.; Zheng, Q.; Su, X.; Zhou, Z. Response of soil aggregate stability and rill erodibility to soil electric field. Catena; 2022; 215, 106338. [DOI: https://dx.doi.org/10.1016/j.catena.2022.106338]

44. Peng, J.; Wang, J.; Yang, Q.; Long, L.; Li, H.; Guo, Z.; Cai, C. Spatial variation in soil aggregate stability and erodibility at different slope positions in four hilly regions of northeast China. Catena; 2024; 235, 107660. [DOI: https://dx.doi.org/10.1016/j.catena.2023.107660]

45. Wang, X.; Huang, P.; Ma, M.; Shan, K.; Wu, S. Effects of riparian pioneer plants on soil aggregate stability: Roles of root traits and rhizosphere microorganisms. Sci. Total Environ.; 2024; 940, 173584. [DOI: https://dx.doi.org/10.1016/j.scitotenv.2024.173584]

46. Zhu, P.; Zhang, G.; Wang, C.; Chen, S.; Wan, Y. Variation in soil infiltration properties under different land use/cover in the black soil region of Northeast China. Int. Soil Water Conserv. Res.; 2024; 12, pp. 379-387. [DOI: https://dx.doi.org/10.1016/j.iswcr.2023.07.007]

47. Zhang, W.; Zhu, X.; Xiong, X.; Wu, T.; Zhou, S.; Lie, Z.; Jiang, X.; Liu, J. Changes in soil infiltration and water flow paths: Insights from subtropical forest succession sequence. Catena; 2023; 221, Pt A, 106748. [DOI: https://dx.doi.org/10.1016/j.catena.2022.106748]

48. Qiu, D.; Xu, R.; Gao, P.; Mu, X. Effect of vegetation restoration type and topography on soil water storage and infiltration capacity in the Loess Plateau, China. Catena; 2024; 241, 108079. [DOI: https://dx.doi.org/10.1016/j.catena.2024.108079]

49. Liu, Y.-F.; Zhang, Z.; Liu, Y.; Cui, Z.; Leite, P.A.; Shi, J.; Wang, Y.; Wu, G.-L. Shrub encroachment enhances the infiltration capacity of alpine meadows by changing the community composition and soil conditions. Catena; 2022; 213, 106222. [DOI: https://dx.doi.org/10.1016/j.catena.2022.106222]

50. Feng, Z.J.; Nie, W.B.; Ma, Y.P.; Li, Y.C.; Ma, X.Y.; Zhu, H.Y. Effects of urea solution concentration on soil hydraulic properties and water infiltration capacity. Sci. Total Environ.; 2023; 898, 165471. [DOI: https://dx.doi.org/10.1016/j.scitotenv.2023.165471]

51. Liao, Y.; Dong, L.; Li, A.; Lv, W.; Wu, J.; Zhang, H.; Bai, R.; Liu, Y.; Li, J.; Shangguan, Z. et al. Soil physicochemical properties and crusts regulate the soil infiltration capacity after land-use conversions from farmlands in semiarid areas. J. Hydrol.; 2023; 626, Pt A, 130283. [DOI: https://dx.doi.org/10.1016/j.jhydrol.2023.130283]

52. Zhu, P.; Feng, T.; Yang, L.; Wang, D.; Chen, X.; Zhang, F.; Li, C. Biological soil crusts decrease soil erodibility of economic fruit forests land through its consolidation effect in the Three Gorges Reservoir area. Catena; 2024; 243, 108200. [DOI: https://dx.doi.org/10.1016/j.catena.2024.108200]

53. Wang, G.; Fang, Q.; Wu, B.; Yang, H.; Xu, Z. Relationship between soil erodibility and modeled infiltration rate in different soils. J. Hydrol.; 2015; 528, pp. 408-418. [DOI: https://dx.doi.org/10.1016/j.jhydrol.2015.06.044]

54. Liang, J.; Huo, P.; Mo, X.; Zhang, L.; Fan, X.; Sun, S. Fostering sustainable banana cultivation: Maximizing red soil performance with lignin-based humic acid liquid fertilizer. J. Agric. Commun.; 2023; 1, 100018. [DOI: https://dx.doi.org/10.1016/j.agrcom.2023.100018]