| The Loess Plateau is one of the most serious areas of soil erosion in China.Due to its large pore space,vertical joints and wet sink nature,loess is highly susceptible to soil erosion under rainfall conditions.In this paper,the erodibility of loess is studied.Using simulated rainfall tests and indoor physical and chemical property tests,the erosion characteristics of loess slopes under different rainfall intensities and soil capacity conditions and the correlation between soil capacity and disintegration rate and soil erodibility are analysed,and an estimation model for the soil erodibility factor K is established.The specific findings of the study are as follows:(1)With the increase of soil capacitance and rainfall intensity,the slope surface flow rate gradually increases.(1)The initial flow production time shortens as the rainfall intensity and soil capacity increase.At a certain rainfall intensity,the rate of flow production on the slope surface increases and the rate of sand production on the slope surface decreases as the soil capacity increases;at a certain soil capacity,both the rate of flow production and the rate of sand production on the slope surface increase as the rainfall intensity increases.When the soil capacity is the same,the cumulative runoff volume and cumulative sediment volume increase as the rainfall intensity increases.(2)During rainfall erosion on loess slopes,the soil capacitance of the slope increases and the accumulated sediment volume from rainfall erosion decreases,while the soil capacitance of the slope decreases and the accumulated sediment volume from rainfall erosion increases.There is a significant negative correlation between soil capacitance and soil erodibility factor,and the fitted equation R2=0.985,>0.95.The change of soil capacitance can reflect the change of soil resistance to erosion,and soil capacitance can be used as an indicator to evaluate the erodibility of loess.(3)The most significant effect on the rate of soil disintegration is the content of clay particles in the soil,followed by the content of sand particles,and not significantly related to the content of organic matter and powder particles.The disintegration rate and soil erodibility were significantly correlated with each other,and the fitted equation R2=0.913,>0.9.The disintegration rate can be used as an evaluation index for the erodibility of loess.(4)Based on the simulated rainfall test,a mathematical model was established to estimate the K value of soil erodibility factor for the soil samples in the study area:Y=0.157-0.074X1+0.00009109X2,where X1 is the soil capacity and X2 is the soil disintegration rate,R2=0.991,with a good correlation·hm2·h/(hm2·MJ·mm).(5)Comparing the K values calculated by Nommo’s formula and EPIC formula for error analysis,the MAE,MRE and RMSE are 0.0047,0.0852,0.0048 and 0.0030,0.0569,0.0032 respectively,with values close to 0.The accuracy factors are 1.0936 and1.0605 respectively,close to 1.It indicates that the established estimation model is Loess Plateau Region is well applicable.(6)The highest correlation of K deviation was for sand content,followed by clay and powder content,while the correlation between K deviation and soil organic matter content was not significant.The higher the sand content and the smaller the clay and meal content,the greater the difference between the K values calculated by the estimation model and those calculated by the formula method,and the need to reestablish the estimation model by combining the local soil samples. |