Research on anti-scourability of slope eroded Soil

Purpose】Soil aggregates are of great significance to soil and water conservation and ecological environment construction in arid area of northwest district.【Methods】Exploring the effects of different vegetation includes types and land use methods on the stability of soil aggregates in the Loess Plateau, and provide reference for the rational use and management of land, also the improvement of soil structure in the region. Select 9 types of samples of 0-30 cm of typical soil plots as the research objects, compare and analyze the particle size index, stability differences and anti-erodibility of soil aggregates under various vegetation cover. 【Results】The results show that P value, MWD value, GMD value, D value, and AI value of the 0-10cm surface soil all show the maximum value. As the depth increases, the size distribution of the above index values of each soil sample in the 10-20cm and 20-30cm layers is different; P value in the 0-30cm depth layer is linearly positively correlated with the AI value and MWD value, and linearly negatively correlated with the D value. The correlation coefficient R between each variable is in the range of 0.78-0.97, and the D value reflects the Loess Plateau area stability and erosion resistance of soil aggregates better. GMD and MWD value show an exponential relationship, the correlation coefficient R value of 10-20cm height layer is 0.46; AI and MWD value in 0-10cm, 20-30cm height layer have a power function relationship, 10-20cm height layer has a polynomial function, the correlation coefficient R value is Preprints (www.preprints.org) | NOT PEER-REVIEWED | Posted: 13 May 2021 doi:10.20944/preprints202105.0291.v1 © 2021 by the author(s). Distributed under a Creative Commons CC BY license. 2 0.97. The scour coefficient of different soil samples has a high degree of dispersion, the maximum CV value is 1.92, and the minimum value is 0.49.【Conclusions】The results of this study can provide a theoretical basis for the ecological and hydrological benefit evaluation of slope erosion control and vegetation restoration on the Loess Plateau.

0.97. The scour coefficient of different soil samples has a high degree of dispersion, the maximum CV value is 1.92, and the minimum value is 0.49.【Conclusions】The results of this study can provide a theoretical basis for the ecological and hydrological benefit evaluation of slope erosion control and vegetation restoration on the Loess Plateau.
Keywords Soil anti-scourability; Soil erodibility; Artificial simulated rainfall；Soil aggreg ates; Erodibility Highlights ➢ The K value of soil erodibility is one of the important factors to calculate the amount of soil erosion.
➢ Soil aggregates play a key role in the differences of soil structure at small watershed scale.
➢ Provide reference for the Loess Plateau and other areas with similar conditions.

1.Introduction
Over the past decade, with the implementation of ecological management projects such as returning farmland to forests and grasslands, closing mountains , prohibiting grazing, and protecting natural forests, the ecological environment has improved significantly. However, due to the special topography of the Loess Plateau, soil erosion is still the main ecological problem in the region. (Yoder R E, et al ,1936).Soil erosion first occurs in the surface of the soil. Therefore, the stability of the topsoil is very important (Li Z X, et al, 2005), Soil water-stable aggregates are key indicators for evaluating soil structural stability, erosion resistance, and runoff generation. They also play an important role in the sustainable development and utilization of land, maintenance of productivity, and nutrient cycling. (Amezketa E ,et al,1999;Six J, et al, 2004;Bronic C J, et al, 2005;Zhou H, et al, 2012;). It is generally believed that the degradation of soil structure is mainly manifested by the decrease in the stability of soil aggregates (Pinheiro E F M,et al,2004 ). Soil aggregates are affected by vegetation types, planting years and farming methods, and also vary with land use patterns.  believes that agricultural farming is the main factor that destroys the structure of the surface soil; (Pinheiro et al 2004) have shown that the content of soil aggregates decreases after grassland becomes agricultural land ; have found that the stability of agricultural soil aggregates is less than that of forest land significantly. Soil aggregates are the basic unit structure of soil and are important indexes for studying soli structure and anti-erosion ability (Six J, 2004). soil aggregates are formed by natural processes such as cementation (Shinjo et al., 2000).
Soil aggregate stability is one of the most important factors affecting soil erosion and soil structure. The soil aggregate stability was controlled by several soil primary characteristics, such as soil texture, clay mineralogy, contents of organic matter, which influence many soil properties, including soil porosity, hydraulic conductivity, water-holding capacity, and soil erosion process (Zhu et al., 2018). The stability of soil aggregates is affected by scour under different slopes and different flow rates, which leads to the difference in soil erosion rates.
Soil water-stable aggregates are the best indicator for soil anti-erosion ability. They are usually evaluated by three indicators: average weight diameter (MWD), geometric average diameter (GMD), and fractal dimension (D) of soil water-stable aggregates the stability of soil aggregates, and the particle size> 0.25 mm the water-stable aggregates are regarded as the most structural aggregates, and the higher the content, the stronger the erosion resistance of soil. Many soil erosion studies focus on the effects of runoff power, whereas much less attention has been paid to the effects of soil anti-scourability on water erosion rates. Soil erosion is a process, in which soil and other ground substances were destroyed, denudated, transported and induced deposition by water, wind, freezethaw and gravity acts in the land surface. In semiarid areas, soil erosion is a serious threat to land productivity and sustainability for natural and human-managed ecosystems (Su Z A, et al., 2010;Fu B J et al., 2011). Water erosion is one of the most important forms of erosion in soil erosion, and soil anti-scourability is the index that best reflects the process and law of regional soil erosion.In addition to absorbing and storing water and nutrients, plant roots can also provide stability to the plant itself, which means that the soil containing roots has resistance to wind and water attraction (Reubens et al., 2007). It is generally believed that the root system improves soil properties (such as soil cohesion, soil stability and shear strength) through its tensile strength, friction and adhesion characteristics, therebyimproving soil resistance to water erosion Ola et al., 2015;.
With the acceleration of ecological civilization construction, under the premise of revealing the law of slope erosion and soil erosion mechanism in the Loess Plateau, this article takes the soil from different typical forest and grass vegetation types as the research object to clarify the law of erosion resistance. Analyze the influence of the MWD, GMD, D, and P values of soil aggregates on the distribution and stability of soil water-stable aggregates, study the differences of soil waterstable aggregates under different vegetation types. However, the relationship between Water stability of soil aggregates and soil scour resistance is not clear. We attempt to clarify the following scientific issues: (1) How does the stability of soil aggregates change during rainfall process? (2) What is the contribution of waterstable aggregate content to the Soil and Water Conservation?

Study area and sampling design
The landform type of the study site is a typical loess hilly and gully area. The soil type is mainly loessial soil with sparse vegetation. In the summer rainstorm season, soil erosion is serious and the surface is washed away by thousands of gullies. The average annual rainfall is mainly concentrated on July and August each year. The precipitation is 600mm, and the average annual potential evaporation is 1050mm.  Table 1. Chose 5 points for each plot, and collected the original soil samples of 0-10cm, 10-20cm, and 20-30cm soil layers with a 100 cm 3 ring knife at each point. After the soil was taken, backfilled layer by layer, and took each layer with 3 repetitions. Put it in an aluminum box to avoid squeezing, tried to keep the original shape of the soil sample, brought it back to the laboratory, and air-dry for testing.
In the laboratory, we broke the soil samples into 5 to 8 mmdiameter blocks according to the natural structure. Stones, plant roots and plant residues were removed from the soil samples, which were airdried at room temperature to preserve. Using the dry sieving method separted soil aggregates to obtain different aggregate size fractions, including 0.106-0.25, 0.25-0.5, 0.5-1,1-2, mm. The soil bulk density was measured by the ring knife method (Hossain et al., 2015).

Table 1. Coverage of forest and grass vegetation in different sampling areas and sampling locations 2.2. Calculation of soil anti-scouring
Soil erodibility defined as the inherent susceptibility of soil to detachment and transport by rainfall and runoff. As an indicator of soil erodibility, the stability of soil aggregates is often used since aggregate breakdown is closely related to crusting that reduces the infiltration capacity drastically and increases runoff, thus leading to water erosion (Jacobs, M, et al,2018). The soil antiscour coefficients and soil erosion energy dissipation have been viewed as the most appropriate index for describing the stability of soils and can be calculated as follows: Where, c K is soil anti-scour coefficients, L.min/g; Q is erosion flow, L; t is time, min; w is soil quality after washing, g.
Where, e K is erosion energy dissipation, J/g; m is erosion water quality, kg; v is velocit y, m/s; w is soil quality after washing, g.  values are associated with greater particle size agglomeration and greater stability; conversely, a smaller D leads to better structure and higher soil stability). MWD, GMD and D were calculated using the following formulas:

Calculation of waterstable aggregate content
where i d is the mean diameter (mm) of the soil aggregate size fractions, i w is the proportion of the total soil in the ith size fraction (%), is the mass of aggregates smaller than xi (mm), T m is the total mass of aggregates, and dmax is the maximum diameter of the soil aggregate size fractions.
In this study, before the analysis of variance, all data was tested for normality and homogeneity of variance. When the test of homogeneity of variance was passed, the Duncan method (confidence interval is within 95%) was used to analyze the differences in multiple comparisons (Steel and Torrie, 1980). SPSS 24.0 software was used for all statistical analyses. The graphs and tables in this study were produced using Excel 2016, Origin 8.5.

Distribution characteristics of soil aggregates and soil stability
As shown in Fig.1 (a) show that the soil formation process is very slow, which further proves that it is a collapsed loess parent material.
As shown in Fig.1 (b). The rule is more obvious for soil sample No. 1, that is, as the depth of As shown in Fig.1 (c). The general rule of GMD agglomerate particle size is basically similar to the distribution rule of MWD, and at the same time, there is a certain correlation between them.
Except for No. 2 and No. 4 soil samples, the GMD values of the 10-20cm layer are higher than the other two layers (0-10 and 20-30cm).
As shown in Fig.1 (d) As shown in Fig.1 (e). The minimum D value of the aggregate particle size distribution is about 2.61, and the maximum value is about 2.98. Different from the above indicators, the smaller the D value, the larger the aggregate particle size, the larger the D value, and the aggregate the smaller the proportion of the particle size, the opposite relationship. The D value of No. 6 soil sample is the

Correlation of each stability index of aggregate
As shown in Fig.2 (a-e), Preliminary analysis of the correlation between the 5 indicators of 0-30 cm soil layer samples. It can be seen from the figure that the correlation between the soil samples of each layer is relatively significant, the difference is that the correlation coefficient is different between different soil samples and different layers. In which, P has a positive linear relationship with MWD and AI values, and the correlation coefficients are 0.9517, 0.8901, 0.8854, 0.9849, 0.9608, 0.9843, as show in Fig.2 (e, d), the formula is as follows: Where, the a and b are the coefficient.
However, as shown in Fig.2 (c), P and D values show a negative linear correlation, that is, the higher the P, MWD, and GMD values are, the more stable the soil structure is. The smaller the D value, the smaller the dispersion of soil aggregates and the stronger the soil agglomeration. The D value has an opposite relationship with the P value, indicating that in the large aggregates, the large aggregates may contain more cementing substances, have strong polymerization and more important contribution to maintaining soil stability.
As shown in Fig.2 (a), the fitting of GMD and MWD found to show an exponential function relationship, the correlation coefficients are 0.98, 0.98 and 0.46, the formula is as follows: bx y ae = Where, the a and b are the coefficient. The correlation coefficients of the above indicators are relatively high, most of which are greater than 0.95, and a very few are about 0.8, which shows that the indicators can be substituted for each other. In the subsequent analysis, a certain index can be used to indicate the particle size distribution characteristics of the agglomerates.

Correlation of each stability index of aggregate
As shown in Figure 3, the soil erosion resistance coefficients of the 0-30cm soil layer samples are calculated. The difference from Section 3.1 is that this issue will be further explained. Six typical soil samples are selected.
For the No. 1 soil sample, the 0-10cm depth has the largest erosion resistance coefficient and the largest erosion energy consumption, respectively 51.12 and 5.74; with the increase of depth, the erosion resistance coefficient and erosion energy consumption appear at 10-20cm and 20-30cm, the decreasing trend is 2.46 and 0.29. The impact coefficients of samples No. 3 and No. 4 at 0-10cm are relatively small, which is mainly due to the surface soil is loosen, which leads to large runoff sand content and low scour resistance, the erosion energy consumption is correspondingly small, only for 0.33 and 0.56. During the impact test of different samples, as the water flow soaked and scoured the soil sample continuously, the binding force between soil particles gradually decreased, and the holding capacity of some root systems to the soil gradually weaken. When the water flow energy accumulated to a certain extent, the holding system of some soil blocks instantly collapsed, resulting in a sharp decline in soil erosion resistance; while the remaining soil samples became relatively stable due to the interaction of soil and root system, and the soil erosion resistance ability suddenly increased subsequently.
It can be seen from the order of appearance of the anti-scouring peaks in different samples and different depth layers that most of them are in the 20-30 cm soil layer, and the root density has a high correlativity. Therefore, after the relatively loose sediment on the surface of some parts, the root system hold the remaining soil firmly, even if the water continues to wash away the soil particles, there will be less lose. This is also the main reason for the increase in the impact resistance value at the end of the test.
Tab 2. CV value of soil anti-scouring coefficient

Disscussion
Soil fractal dimension (D), mean weight diameter (MWD) and geometric mean diameter First, due to the drastic decrease in root biomass, soil erosion resistance is significantly reduced.
For example, the impact resistance of No. 5 soil sample is the highest in the surface layer, but is lower in the 10-30 cm layer, which is about half of the surface layer, and the decline is larger. It is inferred from that the root system has little effect on the soil layer below 10cm, the content of soil cement is low, the aggregate structure is weak, and it cannot resist runoff erosion. The soil moisture content in the sample collection area is relatively low, resulting in poor root development. Although there are dense roots on the surface of the soil, which results in a relatively high erosion resistance of the soil, the erosion resistance below the dense root layer is still rapidly decreasing. Secondly, the soil erosion resistance of different restoration methods is also characteristic under the surface layer, and the soil erosion resistance below the surface layer may be higher than the surface layer, because the properties of the soil are changed after man-made disturbance. As mentioned in Section 3.1, the MWD values of soil aggregates below the surface of No. 2 and No. 4 soil samples are higher than those of the surface, which also indicates that the soil erosion resistance below the surface of the area where such soil samples are located is relatively stronger than that of the surface. For a wide range of slopes dominated by herbaceous plants, the dense layer of roots on the surface is the only barrier against runoff erosion, and the erosion resistance of the soil below the surface will be drastically reduced. Therefore, to prevent slope erosion is to strengthen the surface. The protection and planting of herb roots should also insist on not disturbing the topsoil and focus on natural restoration and succession.
In terms of the coefficient of variation of soil, scour resistance is a soil property with strong variation. In other words, when the sampling point is small, the data cannot accurately represent the overall situation of the sample site. In addition, the obtained results have a high degree of dispersion, it is impossible to determine what interval the average value of a certain soil sample is located in.
Nevertheless, the approximate range of the soil anti-scourability coefficient in this study has obtained ideal results. The maximum values in the results of scour resistance come from samples with special mulching conditions on the ground. In fact, they represent the highest soil scour resistance of the soil in the study area. However, the minimum values of soil scour resistance are those with little surface coverage such as agricultural land. It is reflected in the 20-30cm soil layer.
In future research, we will continue to increase the selection of samples and conduct comprehensive tests of soil erosion resistance for more land use types.

Conclusion
The influence of vegetation on soil water-stable aggregates is generally in the depth range of 0-30 cm. No. 2 soil sample has the largest water-stable large aggregates, MWD and GMD values, the best soil structure, and the strongest anti-erosion ability. The No. 9 soil sample has the smallest value of various indicators of soil water-stable aggregate stability, the soil structure is the worst and erosion resistance are the most serious. There is a good linear relationship between soil fractal dimension D of different land use in small watershed and aggregate index. D not only reflects the quantitative characteristics of large aggregates but also indicates the strength of soil erosion resistance. The variation coefficient of the water-stable aggregates in each soil sample shows a trend of gradually becoming smaller as the depth increases.
In view of the management of different land use methods, the afforestation area and natural restoration grassland of the Loess Plateau should be mainly protected. The cultivated land in the form of stepped stone ridges can increase the number of large aggregates and improve the stability of soil aggregates, thereby slowing down the water and soil loss, preventing and controlling of slope erosion.

Declaration of Competing 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.
carbon and nitrogen in natural restoration grassland and Chinese red pine plantation on the Loess Plateau. Catena 149, 253-260.