The Threshold Values Of Soil Erosion Factors On Typical Slopes Of Loess Plateau

Soil erosion caused by rainfall is a worldwide environmental problem.Rainfall,vegetation cover,slope intensity and slope length are important factors affecting soil erosion,and quantifying the thresholds of these erosion-influencing factors is of great significance for land management in erosion areas.In this paper,we analyze the effects of rainfall intensity,vegetation cover,slope gradient and slope length on soil erosion using a combination of simulated rainfall tests and field runoff plot location observations,and quantify the rainfall threshold（the minimum amount of rainfall that generates runoff on a slope）,the vegetation cover threshold that weighs the regulation of runoff and soil erosion,the slope and slope length thresholds for soil erosion（the slope gradient and slope length at the point or extent of maximum soil erosion）.slope and slope length),a slope erosion prediction model considering soil erosion thresholds was constructed to provide basic support for regional land use management.The main findings are as follows:（1）The rainfall threshold for soil erosion decreases as an exponential function with increasing rainfall intensity,with different vegetation types having different cover coefficients,as follows:woodland slopes y=58.943e^-1.668x（R²=0.84）,pasture slopes y=64.596e^-2.034x（R²=0.83）,barren slopes y=74.809e^-2.308x（R²=0.80）,and farmland slopes y=59.986e^-1.888x（R²=65）.The rainfall threshold of soil erosion gradually decreases with finer soil texture,and the rate of decrease tends to slow down with increasing rainfall intensity.The differences in soil texture and regional rainfall erosion forces are important reasons for the differences in rainfall thresholds.（2）The increase in vegetation cover increased soil infiltration,reduced runoff volume,slowed runoff velocity,and increased soil resistance to erosion.However,the increments of the above effects vary with increasing the same cover.Slope runoff decreased as a power function or exponential function with vegetation cover,and soil erosion decreased as an exponential function,power function or logarithmic function with vegetation cover.Overall,the lower threshold of vegetation cover for controlling soil erosion in different areas ranged from 20%to 30%,and the upper threshold ranged from50%to 70%.When the vegetation cover in Sui-Te area is 40%and below,the forest land can prevent soil erosion better.When the vegetation cover is greater than 40%,barren slope weeds and pasture can prevent soil erosion better.The barren slope land in Xifeng area can prevent soil erosion better,followed by forest land cover.The choice of vegetation to prevent soil erosion and soil conservation in Tianshui area can be both woodland and pasture.（3）The slope effect of soil erosion on slopes is closely related to soil texture and slope erosion mode（development of fine furrows）.In the theoretical derivation,the slope threshold varies with the erosion mode,and the slope factor threshold is between 22°～26°when the erosion mode is dominated by surface erosion and around 36°when the erosion mode is dominated by furrow erosion.In field observation,the slope threshold for soil erosion in forest land is in the range of 36°～40°,and the slope threshold for soil erosion in pasture land is in the range of 26°～30°.The slope thresholds for soil erosion on agricultural land in Suide area are in the range of 26°～30°,in Xifeng area are in the range of 16°～20°,and in Tianshui area are in the range of 11°～15°.According to the difference of slope thresholds of different cover types,you can choose to configure agricultural land and pasture on slopes less than 26°,woodland on slopes from 26°～36°,and pasture or wasteland on slopes greater than 36°.（4）Slope length increased infiltration volume per unit area,runoff velocity,runoff shear and runoff power,and decreased runoff depth per unit area;the effect of slope length on erosion volume per unit area was controlled by the effect of slope length on runoff erosion capacity,fine channel density and sediment deposition.In the field survey,the erosion modulus of woodland,pasture and barren land cover slopes decreased with increasing slope length from 8°～15°;from 15°～25°,the erosion modulus fluctuated with increasing slope length,and the slope length thresholds all appeared at 16 m-20 m,with woodland significantly increasing at the slope length threshold;from 25°～35°,the slope length thresholds of erosion modulus appeared at 16 m-20 m,11 m-15 m,21 m-25 m,respectively.The slope length thresholds of erosion modulus under 25°～35°are at 16 m-20 m,11 m-15 m,21 m-25 m slope lengths,respectively,and all of them increase insignificantly at the slope length threshold;>35°slope,the slope length thresholds of erosion modulus are at 16 m-20 m,46 m-50 m slope lengths,respectively,among which the forest land increases significantly at the slope length threshold.On agricultural slopes,erosion modulus increases with slope length in Xifeng and Tianshui areas from 8°～15°,but the increase in erosion modulus is not significant in the investigated slope length range;erosion modulus fluctuates and increases with slope length in Suide and Xifeng areas from 15°～25°,and the slope length thresholds occur at 21 m-25 m and 16 m-20 m slope lengths,with a significant increase in erosion modulus at the slope length threshold in Xifeng.The erosion model increases significantly at the slope length of 26 m-30 m.The runoff depth and erosion modulus of Suide area increase with the slope length of 25°～35°slope.（5）Soil erosion on different cover slopes was simulated and evaluated based on the improved RUSLE model,and the Nash-Suttcliffe coefficients of the model ranged from 0.504 to 0.707.Overall,the model has good simulation results for smaller erosion events and unsatisfactory simulation results for larger soil erosion events.The prediction model of soil erosion was constructed using modified rainfall erosion force,runoff power and soil texture,considering the influence of the threshold value of soil erosion influence factors and the interaction between factors on the slope,and quantified the relationship between soil texture characteristics and the coupling effect of runoff power and rainfall erosion force,and the Nash-Suttcliffe coefficients of the model were all greater than 0.8,which significantly improved the prediction effect.