The Short-Term Efficacy of Straw Incorporation on Soil Detachment in Sloping Farmland

Abstract

Straw incorporation is applied in sloping farmland to coordinate soil water, fertilizer, air, heat, and soil erosion control in soil loss areas. Straw incorporation is considered to significantly affect soil detachment. However, the knowledge about the influence of soil consolidation by rainfall and mechanical effect by straw incorporation in short-term on soil detachment capacity (Dc) by rill flow is still limited. The current study was carried out to quantify the impact of soil consolidation by rainfall and mechanical effect under straw incorporation on Dc. The soil samples were collected from seven different plots (straw incorporation rates of 0 (CK), 0.2, 0.4, 0.6, 0.8, and 1.2 kg m⁻² with rainfall simulation of 30 mm and without straw incorporation and rainfall simulation (CK0)) and subjected to flume scoring experiments. The results indicated that the Dc with different straw incorporation rates significantly differed and decreased by 39.16–60.04%, compared with CK. The Dc exhibited a power function relationship with hydraulic parameters and stream power was the most appropriate hydraulic variable to express Dc for different straw incorporation rates. The hydraulic characteristics, straw incorporation rates, and interaction between them have a significant impact on the Dc, and Dc was more sensitive to hydraulic characteristics. The contribution rates to Dc reduction benefits by soil consolidation exceeded those by mechanical effect of incorporated straw. The impact of incorporated straw with rainfall simulation in short-term on Dc has a threshold of a straw incorporation rate of 0.4 kg m⁻². The Dc for different straw incorporation rates could be satisfactorily simulated using the composite equation of binary power-exponential function of stream power and soil cohesion. This research reveals the impacts of soil consolidation by rainfall and the mechanical effect of incorporated straw on Dc, and offers a framework for predicting and managing soil erosion in areas susceptible to soil loss.

1. Introduction

Severe soil erosion caused topsoil loss and land degradation in sloping farmland, which posed a serious challenge to food security and the high-quality development of regional environments [1,2]. Soil detachment, a crucial sub-process in soil erosion, is defined as the soil particles’ displacement from the soil mass through raindrop splash and flowing water scoring [3,4,5]. The mass of soil eroded per unit area and time was defined as soil detachment rate and the Dc was measured using clear water with no sediment concentrations. Dc serves as the principal metric for assessing soil detachment intensity [6,7,8,9]. The soil detachment holds a crucial role for soil erosion process, which provides non-cohesion sediment for subsequent sediment transport and deposition [10,11]. Consequently, accurately estimating Dc is indispensable for both soil erosion prediction and control.

Soil erosion is particularly serious in typical agricultural areas with long-term intensive cultivation [1,12,13]. One of the most important needs for human beings is food, particularly in developing countries with a large population to feed [14]. On account of precipitation, terrain, soil texture, soil properties, and tillage practice, the sloping farmland serves as the primary source of soil losses in the Loess Plateau [14,15]. The sloping farmland is a common and indispensable land use type for food production [15,16]. To enhance soil quality and reduce soil erosion, local farmers usually break up crop straw and mix it into the soil, after the wheat is harvested in autumn [17,18,19,20]. At present, many scientists have shifted their focus to investigating the soil and water conservation benefits of crop straw incorporation and the mechanisms affecting them [21,22,23]. Ma et al. [23] conducted rainfall simulation, concluding that incorporated straw reduces the runoff by 3.11–18.18% and sediment yields by 30.71–57.71%. The straw returning to the field significantly reduced Dc through its mechanical effect and by improving soil stability through straw decomposing [24,25,26]. As a result of the differences in crop yields and farming practices, a discrepancy in straw incorporation rates also exists, which is of great significance to the variability of Dc.

The tillage disturbance, straw decomposition, and mechanical effect of incorporated straw greatly influence soil properties and soil erodibility, which vary with temperature, precipitation, and tillage practice, and greatly influence Dc [18,25,27,28,29,30]. The incorporated straw was wrapped in splashed soil particles during the rainfall and it formed soil–straw matrix [9,24,25,30]. The soil tensile strength and anti-erosion abilities were enhanced due to the formation of soil–straw matrix, thus reduced Dc [9]. According to Yao et al. [9], compared with bare soil control, the organic matter and mean weight diameter of straw-incorporated soil increased by 7.73% and 35.18%, respectively, and Dc reduced by 25.92–38.76%. Liu et al. [30] conducted a plant litter incorporation test over 524 days and concluded that critical shear stress enhanced 8–14 times and rill erodibility decreased by 77–96%. The soil anti-erosion ability and Dc also differed for different straw incorporation rates and exhibited temporal variation [9].

The rill flow was the primary driving factor for Dc and the variability of Dc was closely linked with hydraulic characteristics of rill flow [31,32,33,34]. The influencing mechanism of rill flow on Dc was partially investigated in a previous study, but the relationship between them is still inconsistent [7,35,36,37]. The previous study reported that Dc can be described as the linear, power, or exponential function of hydraulic parameters [11,38,39]. In addition, the optimal hydraulic parameter to characterize the Dc is still controversial. Wang et al. [39] found mean flow velocity (V) to be a reliable predictor of Dc. In contrast, the flow shear stress (τ) was used to predict Dc in the WEPP model. Geng et al. [40] and Ma et al. [41] determined that stream power (w) is the most suitable hydraulic variable for expressing Dc. Additionally, Yang et al. [42] proposed that unit stream power (U) can effectively estimate Dc. Overall, the optimal parameter characterizing Dc and the relationships between them have been explored, but there is still no unified conclusion in recent research. Therefore, further studies are still being conducted to elucidate the dynamic mechanism for Dc.

Due to poor soil structure, intensive cultivation, and rainfall characteristic, the Loess Plateau is globally recognized for its significant soil and water loss, with soil loss rates of 2000–2500 t km⁻² a⁻¹ [33,43,44]. The soil erosion rate maintained a high level in the initial stage after sowing seeds in sloping farmland [9]. In more recently research, the results concluded that straw incorporation could reduce sediment yields by more than 30% and decrease Dc by more than 25% by reducing direct raindrop splash and enhancing hydraulic roughness [9,23]. The Dc and soil loss decrease rapidly with crop growth, litter decomposition, and soil consolidation [14,25,45]. During the early phases of straw incorporation, Dc significantly increases due to the tillage disturbance effect of the straw incorporation process. However, Dc also decreases with soil consolidation by rainfall and the mechanical effect of incorporated straw. The tillage disturbance and mechanical effect of incorporated straw, along with soil consolidation, are associated with both straw incorporation rates and rainfall. While the impact of straw incorporation on Dc has been investigated, limited attention has been given to the straw incorporation rates. Especially in the early stage after straw incorporation, it was believed that the soil structure damage was caused by tillage disturbance, leading to higher Dc. Additionally, the interaction effect of straw incorporation and rainfall on the variability of Dc remains poorly understood. Investigating and distinguishing the short-term impact of soil consolidation, tillage disturbance, and mechanical effect of incorporated straw on Dc, and their variability with different straw incorporation rates, holds crucial importance in elucidating the influence of straw incorporation on Dc.

Thus, the current research aims to (1) identify the variability of Dc with straw incorporation rates in the short-term and distinguish the impact of soil consolidation by rainfall and straw incorporation on Dc, (2) explore the relationship between Dc and hydraulic parameters, and clarify optimal hydraulic parameters that characterize Dc, and (3) establish a model for Dc prediction that could be used for straw incorporation.

2. Materials and Methods

2.1. Study Area and Test Soil

This research was conducted in the Laboratory of Hydraulic Engineering, Northwest A&F University (34°15′ N, 108°04′ E). The Loess Plateau, known for its fertile soil and favorable climatic conditions, holds significant potential for agricultural, forestry, and animal husbandry development. Meanwhile, the Loess Plateau is renowned for its rich historical and cultural significance. However, it is also prone to soil erosion, a phenomenon that is the focus of this study. The test soil used in the current study was collected from local farmland, and specific soil properties are listed in

Table 1. The specific soil properties for test soil.

Soil Texture			Mean Weight Diameter	Soil Organic Matter	Median Particle Size
Sand Content	Silt Content	Clay Content
10.20%	61.60%	28.20%	0.96 mm	21.26 g kg−1	50.70 μm

.

2.2. Experimental Treatments and Soil Sampling

The collected test soil was screened through a 10 mm screen to eliminate any potential interference from crop roots and gravel. The soil boxes with a length, width, and depth of 2.0, 2.0, and 0.25 m, respectively, were used for straw incorporation and soil sampling. The screened soil was packed into the box for the straw incorporation experiment. Initially, the sand with a 2 cm thick, was placed at the bottom of the soil box to allow for permeation and facilitate drainage. According to the determination of the soil bulk density (BD) in the field, the soil was uniformly filled to a BD of 1.25 g cm⁻³. During soil filling into the soil boxes, a board was used to compact and scrape after each layer was filled to remove the effect of soil stratification. To ensure that the soil and straw were more fully mixed, the prepared soil boxes were divided into several soil plots (length × width: 0.2 m × 0.2 m). Wheat straw was gathered from local farmland after the wheat harvest. Following thorough consultations with experienced local farmers and on-site investigations regarding both the straw-to-grain yield ratio of 1:1, incorporation rates of 0, 0.2, 0.4, 0.6, 0.8, and 1.2 kg m⁻² were designated for the topsoil of 10 cm. The straw incorporation rates of 0, 0.2, 0.4, 0.6, 0.8, 1.2 kg m⁻² were recorded as CK, T1, T2, T3, T4, and T5, respectively. Considering the rainfall conditions and the real situation in the early stages of straw incorporation and soil consolidation (forming stable soil crust), a side-spray rainfall simulator was utilized to conduct rainfall simulations, maintaining a rainfall intensity of 90 mm h⁻¹ for a duration of 30 min across all treatments to standardize the influence of soil crust on the Dc [46,47]. After rainfall, the soil box was left in natural conditions for one day. To calculate the impact of soil consolidation on Dc, the soil samples without straw incorporation and rainfall were also prepared and recorded as CK0. The circle cutting ring (diameter × depth, 9.8 cm × 5.0 cm) was used for soil sampling. During the soil sampling, be as careful as possible to minimize damage to the soil surface crusts. An Eijkelkamp pocket vane tester and pocket penetrometer were used to take soil cohesion (Coh) and soil penetration resistance (PR) measurements, with 20 replicates.

2.3. The Hydraulic Parameters Measurement

The rectangular steel fume was used to simulate rill flow (Figure 1, length × width × depth: 4.0 m × 0.2 m × 0.1 m). The collected soil sample was positioned in a soil sample container of hydraulic flume. After adjusting the slope gradient (s, %) and flow discharges (q, m² s⁻¹) to the designed values, and achieving stable rill flow, flow velocity was measured fifteen times for each s and q using KMnO₄. The mean flow velocity (V, m s⁻¹) was calculated by multiplying the observed mean values by a correction factor in accordance with the flow regime. The flow depth (h, m), flow shear stress (τ, pa), stream power (w, W m⁻²), and unit stream power (U, m s⁻¹) were calculated as

$$h={\displaystyle \frac{Q}{BV}}$$

(1)

$$\tau =\rho gsh$$

(2)

$$w=\tau V$$

(3)

$$U=Vs$$

(4)

where Q, B, ρ, and g are the flow discharge (m³ s⁻¹), fume width (m), density of water (kg m⁻³), and gravity constant (m s⁻²), respectively. The 20 combinations of slope gradient (8.75%, 17.63%, 26.79%, and 36.40%) and flow discharge (8.33 × 10⁻⁴, 1.25 × 10⁻³, 1.67 × 10⁻³, 2.1 × 10⁻³, and 2.5 × 10⁻³ m² s⁻¹) were designed to achieve τ of 2.31–17.22 Pa.

2.4. Dc Measurements

After adjusting the s of the rectangular steel flume and the q, the surface of the soil sample was parallel to the bottom. Prior to commencing the experiment, thin iron sheets were laid over the soil sample’s surface. Once the flowing water became stable, the thin iron sheet was removed. Sediment samples were collected, and the duration of scouring was recorded (t, s). To standardize the impact of the cutting ring on Dc, the scouring experiment concluded after the scouring depth of 2 cm [9,23]. The collected sediment samples were dried, and the dry mass (M_a, kg) was recorded. The Dc was calculated as

$$Dc={\displaystyle \frac{M_a}{At}}$$

(5)

where Dc and A are the soil detachment capacity (kg m⁻² s⁻¹) and section area of soil samples (m²), respectively. The Dc with a different straw incorporation rate was measured for five repetitions at each s and q, and mean values were further analyzed. The soil sample was collected after rainfall simulation and the variation in Dc was influenced by rainfall simulation. Thus, relative soil detachment capacity (RDc, 0–1) was calculated by the Dc for straw incorporation (Dcs) divided by Dc for CK (Dck).

$$\mathit{RDc}={\displaystyle \frac{\mathit{Dcs}}{\mathit{Dck}}}$$

(6)

2.5. Data Calculation and Statistical Analysis

The Dc for bare soil without rainfall and straw incorporation, bare soil with rainfall, and straw incorporation with rainfall were recorded as Dck0, Dck, and Dcs, respectively. The Dc reduction benefits for soil consolidation by rainfall (DRS), mechanical effect of incorporated straw (DRM), and straw incorporation (DRR) were calculated as follows:

$$DRS={\displaystyle \frac{Dck0-Dck}{Dck0}}\times 100\%$$

(7)

$$DRM={\displaystyle \frac{Dck-Dcs}{Dck}}\times 100\%$$

(8)

$$DRR={\displaystyle \frac{Dck0-Dcs}{Dck0}}\times 100\%$$

(9)

The contribution rates for soil consolidation by rainfall (CRS) and mechanical effect of incorporated straw (CRM) were calculated as follows:

$$CRS={\displaystyle \frac{DRS}{\sum DRR}}\times DRS$$

(10)

$$CRM={\displaystyle \frac{DRM}{\sum DRR}}\times DRS$$

(11)

ANOVA was used to test the differences in Dc between different treatments. The simple regression method was used to analyze the relationship between Dc, hydraulic parameters, and soil properties. A response surface analysis experiment was conducted, with hydraulic variable and straw incorporation rates serving as independent variables and Dc as the dependent variable. The determination of the prediction model for Dc was tested by the multiple regression method. The R² and NSE were used to evaluate the performance of the regression equation for Dc.

3. Results

3.1. The Variability of Dc

The variability characteristics of Dc for CK, T1, T2, T3, T4, and T5 presented a clear trend in the current study (Figure 2). The Dc for CK had the maximal value, with a range of 6.12 × 10⁻³ to 1.84 kg m⁻² s⁻¹ and an average of 0.33 kg m⁻² s⁻¹. With the increase in straw incorporation rates, the Dc decreased first and then increased, suggesting that straw incorporation had an inconsistent impact on Dc and there was a threshold effect on Dc. The Dc for T1, T2, T3, T4, and T5 ranged from 1.18 × 10⁻² to 0.49, 1.09 × 10⁻² to 0.37, 5.41 × 10⁻³ to 0.45, 1.05 × 10⁻² to 0.62, and 5.44 × 10⁻³ to 0.77 kg m⁻² s⁻¹, respectively, with averages of 0.17, 0.13, 0.16, 0.17, and 0.20 kg m⁻² s⁻¹, respectively. The mean values of Dc for T1, T2, T3, T4, and T5 decreased by 50.09%, 60.04%, 51.41%, 49.51%, and 39.16%, respectively, compared to CK. The Dc distribution features from Figure 2 indicate that the data point for CK was relatively dispersed, and that, for straw incorporation, was relatively concentrated. The corresponding variation coefficients of Dc for these treatments further proved this (

Table 2. The statistical parameters of soil detachment capacity for different straw incorporation rates.

Straw Incorporation Rates (kg m−2)	Min	Max	Mean Values	n	Coefficient of Variation
0	6.12 × 10−3	1.84	0.33	20	1.41
0.2	1.18 × 10−2	0.49	0.17	20	0.97
0.4	1.09 × 10−2	0.37	0.13	20	0.75
0.6	5.41 × 10−3	0.45	0.16	20	0.82
0.8	1.05 × 10−2	0.62	0.17	20	1.10
1.2	5.44 × 10−3	0.77	0.20	20	1.03

). The ANOVA analysis showed that significant discrepancies existed among the different straw incorporation rates, indicating that straw incorporation had a remarkable effect on the variability of Dc.

3.2. The Relationship Between Hydraulic Parameters and Dc

The relationships between hydraulic parameters and Dc were analyzed and illustrated in Figure 3. The regression analysis revealed that Dc could be simulated as a power function of hydraulic parameters. Additionally, around the regression curves of V and U, Dc values are relatively dispersed under the same hydrodynamic conditions, but relatively concentrated around the regression curves of τ and w. The empirical equations between Dc and hydraulic parameters were established and the R² was derived. The R² of established empirical equations for V, τ, w, and U ranged from 0.42 to 0.67, 0.89 to 0.98, 0.92 to 0.97, and 0.74 to 0.89, respectively. However, the performance of the prediction equation using the mean flow velocity for the straw incorporation of 0.2, 0.4, 0.6, and 0.8 kg m⁻² was relatively poor, with R² values of 0.42, 0.50, 0.54, and 0.55, respectively. Additionally, the empirical equation of τ and w offered a better prediction for Dc, with average R² values of 0.938 and 0.942. But, the prediction equations of V were relatively poor, with an average R² of 0.54. According to the prediction equations of hydraulic parameters, the V, τ, w, and U could explain 54.20%, 93.81%, 94.20%, and 84.43% of the variation in Dc, respectively, suggesting that w was optimizing the hydraulic index to characterize the Dc change, and revealing that the influence of hydraulic characteristics on soil detachment might be better reflected from the perspective of energy consumption.

3.3. The Effect of Straw Incorporation on Dc and Dc Reduction Benefits

The comparison of Dc between straw incorporation treatment and CK revealed that the Dc of soil after straw incorporation was significantly lower than that of CK. The soil consolidation induced by rainfall, mechanical effects of incorporated straw, and straw incorporation had significant impacts on Dc and contributed to Dc reduction. In order to calculate the contribution rate of each influencing factor to Dc reduction, the Dc for the soil sample without rainfall and without straw incorporation, with rainfall and without straw incorporation, and with straw incorporation and rainfall were collected and measured.

Table 3. The soil consolidation, mechanical effect of incorporated straw, and straw incorporation for reduction in soil detachment capacity (%).

Straw Incorporation Rate (kg m−2)	Soil Consolidation	Mechanical Effect	Straw Incorporation
0.2	82.42	50.09	91.22
0.4	82.42	60.04	92.97
0.6	82.42	51.41	91.46
0.8	82.42	49.51	91.12
1.2	82.42	39.16	89.30

presents the calculated Dc reduction benefits achieved through soil consolidation by rainfall, mechanical effects of incorporated straw, and the combined effect of straw incorporation and rainfall. The results revealed that the reduction benefits in Dc for the soil consolidation effect by rainfall were 82.42%. The reduction benefits in Dc by mechanical effect of incorporated straw ranged from 39.16% to 60.04%. The reduction benefits in Dc for straw incorporation with rainfall were 91.22%, 92.97%, 91.46%, 91.12%, and 89.30%, respectively. The initial increase and subsequent decrease in Dc reduction benefits with increasing straw incorporation rates, suggesting that there exists a threshold effect for Dc at a straw incorporation of 0.4 kg m⁻². The contribution rates of soil consolidation to the reduction in Dc with 0.2, 0.4, 0.6, 0.8, and 1.2 kg m⁻² of straw incorporation rates were 56.74%, 53.79%, 56.32%, 56.93%, and 60.54%, respectively, while the mechanical effect of incorporated straw were 34.48%, 39.18%, 35.13%, 34.20%, and 28.77%, respectively (Figure 4). The results indicate the contribution of soil consolidation by rainfall for reduction in Dc were higher than that of the mechanical effect of incorporated straw and suggest that the effect of incorporated straw on Dc is one of a disturbance and protection effect.

3.4. The Response of Dc to Straw Incorporation and Hydraulic Characteristics

Rill flow provided the energy for soil detachment, which determined the soil detachment intensity. Suitable straw incorporation rates may enhance soil fertility and soil structural stability by releasing fragmented organic substances. The response model of Dc to hydraulic characteristics (stream power) and straw incorporation was established, including independent variables (w, stream power, and S, straw incorporation rates) and dependent variables (Dc). As depicted in Figure 5, the curvature of the response surface illustrates the variation and sensitivity of Dc to both stream power and straw incorporation. Initially, Dc decreased with straw incorporation rates, but subsequently increases. The variability of Dc with w tended to be slight at first but later became more significant. When there was no straw incorporation, the effect of hydraulic characteristics on Dc became more significant. The measured soil detachment capacity for CK and straw incorporation plots in the current study was used to simulate Dc through the response surface model using coded factors (A-stream power and B-incorporation rates), as follows:

$$D_c=0.17+0.21A-0.065B-0.14AB+0.19B^2+0.39{AB}^2 (n=120)$$

(12)

Equation (12) with an R² of 0.82 and F value of 105.74 (

Table 4. Variance analysis and regression equations of soil detachment capacity to interactions between stream power and straw incorporation rates.

Source	Sum of Squares	df	Mean Square	F Value	p-Value Prob > F
Model	5.91	5	1.18	105.74	<0.0001
A-stream power	0.56	1	0.56	49.96	<0.0001
B-incorporation rates	0.19	1	0.19	16.84	<0.0001
AB	0.24	1	0.24	21.43	<0.0001
B2	0.68	1	0.68	61.17	<0.0001
AB2	0.78	1	0.78	69.37	<0.0001
Residual	1.27	114	0.011
Cor Total	7.18	119

) indicates that the equation is reliable. The p-value of stream power, straw incorporation rates, and interaction terms were less than 0.01, and the F value of hydraulic characteristics and straw incorporation rates were 49.96 and 16.84, respectively, suggesting that hydraulic characteristics, straw incorporation rates, and the interaction between them had a significant impact on the Dc, and the Dc was more sensitive to hydraulic characteristics.

4. Discussion

4.1. The Variability of Dc with Straw Incorporation

As depicted in Figure 2, straw incorporation notably lowered Dc. Compared with CK, the mean value of Dc for straw-incorporated soil decreased from 39.16% to 60.04% in the current study. This finding aligns with the results of Yao et al. [9], who emphasized the significant impact of straw incorporation on decreasing Dc. The reduction efficacy in Dc for straw incorporation may be attributed to soil consolidation and the mechanical effect of straw. After straw incorporation and rainfall simulation, the incorporated wheat straw was wrapped in splashed soil particles, forming a soil–straw matrix, and the soil physical crust was also formed by rainfall impact [48,49]. The soil–straw matrix and soil physical crust that formed increased soil tensile strength and soil erosion resistance, thus reducing Dc [30]. In addition, the exposed straw on the soil surface enhanced surface roughness and increased energy consumption of rill flow [23]. Thus, we can conclude that straw incorporation influences soil erosion resistance and hydraulic characteristics of rill flow, which in turn affect Dc.

It is also worth noting that Dc significantly differed for different straw incorporation rates. The difference may be induced by discrepancies in mechanical effect for different straw incorporation rates. Except for soil consolidation, the variability of Dc was also influenced by the mechanical effect of incorporated straw [47,50,51]. The Dc for CK was higher than that for straw incorporation, indicating that the influence of the mechanical effect of incorporated straw had a notable effect on Dc [47,51,52]. Initially, Dc decreased with straw incorporation rates but subsequently increased, and the Dc with straw incorporation rates of 0.4 kg m⁻² was lower than any other treatment in the current study. The variability of Dc indicated that Dc was influenced by the balance and trade-off between the mechanical effect of incorporated straw, including the disturbance magnitude and protection effect of the incorporated straw. This finding aligns with the results of Yao et al. [9], who observed that both the mechanical effect of incorporated straw and the magnitude of disturbance escalated with straw incorporation rates.

4.2. The Effect of Hydraulic Characteristics on Dc

The Dc was strongly influenced by hydraulic characteristics represented by V, τ, w, and U [8,35,53,54]. The measured Dc increased with hydraulic parameters in a power function relationship in the current study. This finding is corroborated by Zhang et al. [55]. and Ma et al. [41]. Nevertheless, contrasting reports from Zi et al. [11] and Wang et al. [39] indicated that Dc increased exponentially or linearly with hydraulic parameters. This disagreement may be induced by the presence of soil crust and a newly formed soil–straw matrix. The disturbed soil, which could be uniformly detached, was used by Wang et al. [39] to establish the Dc prediction equation. Due to the existence of soil crust and the soil–straw matrix, Dc increases nonlinearly with hydraulic parameters [47,52,56]. There were two possible explanations for this. Firstly, with the increase in hydraulic parameters and the increase in energy use for soil detachment, Dc thus increased. In addition, because of the existence of soil consolidation and the mechanical effect of incorporated straw, Dc was kept the lower level when the hydraulic parameters were smaller, but the Dc increased quickly when the hydraulic driving force exceeded the soil erosion resistance of the soil mass. Taking stream power, for example: when the stream power was less than 1.0 W m⁻², Dc with incorporated straw at 0 and 1.2 kg m⁻² ranged from 6.12 × 10⁻³ to 1.37 × 10⁻² and 5.44 × 10⁻³ to 1.174 × 10⁻² kg m⁻² s⁻¹, respectively, with mean values of 1.18 × 10⁻² and 5.76 × 10⁻² kg m⁻² s⁻¹, respectively. However, when the stream power was higher than 4.0 W m⁻², the mean values of Dc were 0.89 and 0.51 kg m⁻² s⁻¹, respectively. So, the Dc increased rapidly and nonlinearly with hydraulic parameters and the power function of w exhibited a superior capacity to characterize the variability of Dc, whereas V and U demonstrated less satisfactory results. The power function of w could better characterize the variation trend in Dc, while mean flow velocity and unit stream power were less satisfactory. This finding aligns with the conclusions of Shen et al. [8] and Ma et al. [41], who reported that w was the most suitable hydraulic parameter to express Dc.

4.3. Factors Influencing the Variation in Dc

4.3.1. Soil Consolidation

The result from the current study indicates that soil consolidation significantly reduces Dc. The soil erosion resistance for soil mass with straw incorporation under rainfall simulation changed greatly and the soil physical crust and soil–straw matrix were formed, which significantly influenced the variation in Dc [9,46,52]. The soil consolidation after rainfall could be effectively captured by BD, Coh, and PR [25,57]. Soil cohesion, which represents soil that can withstand external forces, served as a key indicator for characterizing soil erosion resistance [25,30]. The Coh for soil mass with incorporated straw ranged from 6.63 to 8.53 kPa in this study, which was higher than that for CK with the rainfall simulation. Soil erosion resistance was usually negatively correlated with Coh [58,59]. A significant negative relationship was observed between Dc and soil Coh. That is, with the increase in soil cohesion, Dc decreased with the exponential function in the current study (Figure 6a). The results are consistent with the findings of Yang et al. [22] and Ma et al. [23], indicating that straw incorporation significantly increases soil anti-erosion ability and reduces soil erosion. Therefore, the effect of soil cohesion on Dc with straw incorporation was partly due to the enhancement of soil consolidation and soil erosion resistance.

The PR is the anti-erosion index of soil to vertical penetration force, which can effectively capture the extent of adhesion among soil particles and porosity conditions [58,60]. As depicted in Figure 6b, the PR for soil mass with straw incorporation was measured and ranged from 180 to 274 kPa, and Dc exhibited an exponential decline with increasing soil penetration resistance. This finding aligns with prior research conducted by Knapen et al. [57] and Yao et al. [9], who observed a strong correlation between Dc and PR. In this study, the obvious discrepancy was found in PR with different straw incorporation rates. As shown in Figure 2, the Dc initially decreased and subsequently increased with straw incorporation rates, with the same variation trend in soil penetration resistance. Thus, a positive relationship between Dc and soil penetration resistance was found.

4.3.2. Mechanical Effect of Incorporated Straw

The straw-incorporated soil had a lower Dc than bare soil in current study. In previous studies, the effect of the decomposition of incorporated straw and plant litter on seasonal variation in Dc have been investigated, which led to the conclusion that incorporated straw could enhance soil structural stability by releasing fragmented organic substances or alter the shape of flowing water and improve soil erosion resistance by a mechanical effect [18,23,43,61]. After the rainfall and straw incorporation in the short-term, both the effect of its mechanical efficiency and soil consolidation on Dc were concomitant, making it difficult to distinguish them separately [30,47]. To clarify the effect of soil consolidation and mechanical impact on Dc, the RDc was calculated and plotted (Figure 7). As shown in Figure 7, the RDc with different straw incorporation rates was less than 1, and RDc decreased first and then increased with increasing straw incorporation rates. This result also illustrated that the incorporated straw could effectively reduce Dc, and the impact of incorporated straw on Dc reduction reached a threshold with straw incorporation rate of 0.4 kg m⁻².

The impact of incorporated straw on Dc was realized by the protection effect and disturbance effect. The Dc reduction benefits of incorporated straw increased first and then decreased with increasing straw incorporation rates, also indicating that incorporated straw triggers a disturbance effect and protection effect on the variability of Dc. The disturbance effect on Dc was mainly achieved through the destruction of soil structure during straw incorporation [9]. The protection effect was mainly realized by the actions of the incorporated straw. The existence of crop residues on the soil surface could increase soil surface random roughness and soil erosion resistance, while increasing the consumption of the kinetic energy of flowing water [22,23]. After the rainfall simulation, the soil tensile strength was significantly enhanced and soil erodibility reduced through the formation of the soil–straw matrix [25]. As a result of the RDc with straw incorporation being less than 1 in current study, its mechanical effect reduces Dc. The disturbance effect of incorporated straw mainly refers to tillage practice. During the process of straw incorporation, the soil structure underwent significant disruption, resulting in the formation of a vulnerable and easily erodible soil surface layer, which significantly increased Dc [9,57]. As depicted in Figure 7, the RDc increased with straw incorporation rates when the straw incorporation rate was higher than 0.4 kg m⁻², indicating that the disturbance effect of incorporated straw was higher than the protection effect.

4.4. The Model for Dc Prediction

As analyzed and discussed above, significant differences were found in Dc for varied straw incorporation rates. The incorporated straw significantly influenced the variability of Dc by soil consolidation and its mechanical effect. The relationship between Dc and hydraulic parameters was also investigated, revealing that Dc could be fitted as a power function of w. Thus, stepwise regression analysis between Dc, hydraulic parameter, and soil properties was conducted to estimate Dc, and the results indicated that Dc with different straw incorporation rates could be predicted by Equation (13):

$$\begin{gathered} Dc=0.046w^{1.96}exp\left(-0.21Coh\right) \\ {R}^2=0.91, \mathit{NSE}=0.91, n=120 \end{gathered}$$

(13)

The comparison between measured Dc and predicted Dc, using Equation (13), is shown in Figure 8, which clearly exhibits the measured and predicted Dc evenly distributed near the 1:1 line. The R² and NSE, with 0.91 and 0.91, also show that Dc with different straw incorporation rates can be well estimated by Equation (13).

5. Conclusions

This research was carried out to explore the variability of Dc across different straw incorporation rates, and to quantify the reduction in Dc. The results clearly demonstrate significant differences in Dc across various straw incorporation rates: the Dc initially decreased and subsequently increased with the straw incorporation rates. The mean values of Dc with straw incorporation rates of 0.2, 0.4, 0.6, 0.8, 1.2 kg m⁻² decreased by 50.09%, 60.04%, 51.41%, 49.51%, and 39.16%, respectively, compared with CK. The Dc exhibited a power function relationship with hydraulic parameters, and stream power exhibited a superior capacity in characterizing the variability of Dc with hydraulic parameters across different treatments in the current study. The hydraulic characteristics, straw incorporation rates, and the interaction between them have a significant impact on Dc variation, and Dc was more sensitive to hydraulic characteristics. The Dc decreased exponentially with soil cohesion and soil penetration resistance. The Dc reduction benefits and contribution rates by soil consolidation were higher than that of the mechanical effect of incorporated straw. The effect of straw incorporation on Dc had a threshold value under rainfall simulation in the short-term, and the disturbance effect of incorporated straw on Dc was higher than the protection effect when the straw incorporation rate was higher than the threshold value of 0.4 kg m⁻². The Dc with different straw incorporation rates could be satisfactorily simulated using the composite equation of binary power-exponential function of stream power and soil cohesion (R² = 0.91, NSE = 0.91). The findings from the current study are highly significant for comprehending the soil erosion mechanisms and formulating plans for soil and water conservation in soil erosion regions.