Effects And Simulation Of Biological Soil Crusts On Runoff On Slope On The Loess Plateau

In order to prevent soil erosion,protect and improve the ecological environment,the Grain for Green Program was implemented in the Loess Plateau in 1999.Biological soil crusts（biocrusts）were widely developed in the area on the soil surface of the revegetated grassland,and its average coverage is more than 70%after the implementation of the Grain for Green Program,because of the reduction of human disturbance and suitable climatic conditions.Biocrusts significantly affect the physical and chemical properties of surface soil,soil permeability and surface micro-topography,and then affect the infiltration and runoff of the slope of revegetated grassland,which become an important factor in the hydrological process of the basin and even the region.So far,due to the complexity of biocrusts and differences in research conditions,the impact of biocrusts on slope runoff has not formed a unified understanding.In addition,hydrological model as an important tool for hydrological phenomena and hydrological process research has been widely used in runoff simulation of different regions and land use types.However,the existing hydrological models rarely consider the influence of biocrusts on model parameters and model fitting accuracy.The existing models,such as SCS-CN model,have unclear simulation effect on slope runoff with biocrusts.In order to analyze the influence and mechanism of biocrusts on slope runoff,clarify the effect of biocrusts on slope runoff,reveal the influence of biocrusts on hydrological model parameters,and improve the prediction accuracy of hydrological model in the revegetated grassland slope with widely distributed biocrusts on the Loess Plateau,based on the widely used SCS-CN model,this paper took the slope with different coverage and composition of biocrusts as the object.Through indoor simulated rainfall experiment and field simulated rainfall experiment,this paper studied the influence of biocrust coverage and composition on runoff characteristic parameters such as the initial runoff time,runoff velocity,runoff yield,runoff coefficient and runoff process.The influence and quantitative relationship of biocrust coverage on the CN value and initial abstraction ratio（λ）of SCS-CN model parameters were clarified.The main conclusions are as follows:（1）Biocrusts significantly affect the runoff on the slope,and the degree of influence is related to the biocrust coverage.The initial runoff time,runoff velocity,runoff yield,runoff coefficient and runoff process on the slope under different of biocrust coverage are significantly different.The initial runoff time,runoff velocity,runoff yield,runoff coefficient and runoff generation rate of the slope covered by moss crusts and algal crusts decreased with the increase coverage of biocrusts.The initial runoff time of moss crust slope with 80%～100%coverage was 63.6%lower than that of slope with 0%～20%coverage,the runoff velocity was 69.7%lower than that of bare soil slope,the runoff was 84.1%lower than that of bare soil slope,and the runoff coefficient was 75.2%lower than that of bare soil slope.The initial runoff time of biocrusts mixed with algae and mosses slope with 80%～100%biocrust coverage was 33.8%lower than that of slope with 0%～20%biocrust coverage,and the runoff velocity of slope was 87.1%lower than that of bare soil slope,the runoff yield was 71.2%lower than that of bare soil slope,and the runoff coefficient was 75.7%lower than that of bare soil slope.Under the condition of algal crusts covering,the initial runoff time on the slope first increased and then decreased with the increase of algal crust coverage,and the runoff velocity,runoff yield and runoff coefficient first decreased and then increased with the increase of algal crust coverage,and the inflection points were all at 40%coverage.（2）The influence of biocrusts on slope runoff was related to the composition of biocrusts.The initial runoff time,runoff velocity,runoff yield,runoff coefficient and runoff process are significantly different under different biocrust compositions.Under the same biocrust coverage,runoff yield,runoff generation rate and runoff coefficient of moss crust slope were less than those of algal crust slope.The initial runoff time on the algal crust slope with 20%～80%coverage was significantly longer than that on the moss crust slope,which was opposite under other biocrust coverage conditions.The runoff velocity of moss crust slope with 0%～40%biocrust coverage was higher than that of algal crust slope,and the other coverage conditions were opposite.The initial runoff time on the slope covered by algal-moss mixed biocrusts increased with the increase of moss crust ratio,and the runoff velocity,runoff yield,runoff coefficient and runoff generation rate decreased with the increase of moss crust ratio.The initial runoff time of the slope with algal-moss ratio of 1:2 was 2.26 times that of the slope with algal-moss ratio of 2:1,and the runoff yield of the slope with algal-moss ratio of 1:2 was reduced by77.2%compared with the slope with algal-moss ratio of 2:1.In the stable runoff generation stage,the runoff velocity on the slope with algal-moss ratio of 1:2 decreased by 108.2%compared with that on the slope with algal-moss ratio of 2:1,and the runoff yield and runoff coefficient per 3 minutes decreased by 72.8%compared with that on the slope with algal-moss ratio of 2:1.（3）Biocrusts significantly affected SCS-CN model parameters,the CN value andλwere mainly affected by the coverage of biocrusts in the widely developed loess plateau.With the increase of biocrust coverage,the CN value decreased in a linear function,andλdecreased in a logarithmic function.The parameter CN of SCS-CN model was significantly negatively correlated with the biocrust coverage.The CN value decreased linearly with the increase of biocrust coverage,which could be expressed as:CN=-0.2867×C_BSC+79.012,R²=0.4898(C_BSC represents biocrust coverage,unit:%).The SCS-CN model parameterλwas significantly negatively correlated with the biocrust coverage.Theλvalue decreased logarithmically with the increase of biocrust coverage,which could be expressed as:λ=0.08·e^{^}(-0.015·C_BSC),R²=0.6012(C_BSC represents biocrust coverage,unit:%).（4）After the SCS-CN model parameters were modified according to the biocrust coverage,the fitting accuracy of this model for runoff yield on biocrust slope with different coverage was significantly improved.The SCS-CN model with revised parameters was used to simulate the runoff of 39observation plots in simulated rainfall events.Compared with the standard SCS-CN model,the E value（Nash Efficiency Coefficient）increased by 155%,RMSE value（Root Mean Square Error）decreased by 51.1%,and QR value（Qualified Rate）increased by 165%.The modified SCS-CN model was used to fit the runoff of 12 observation plots in 3 natural rainfall events.The results showed that the E value of the modified SCS-CN model was 5.5 times higher than that of the standard SCS-CN model,and the RMSE value was reduced by 41.7%.