| The Southwest Karst region centered in Guizhou is the largest and most continuous karst area in the world,and also the most complex,typical,and diverse in terms of landscape types.Regionally,soil and water loss occurs frequently due to the effects of the special geological background（extensive soluble carbonate rocks）,humid climate,and dense population,with serious hydraulic erosion.The karst slope has high heterogeneity,with a large amount of exposed carbonate rocks on the surface,making soil and water loss in the area extremely complex.Soil erosion experience models such as RUSLE,which are established based on homogeneous soil regions（non-karst regions）,face significant challenges when applied in the karst region.Currently,the quantitative relationship between exposed rock and soil loss has not been systematically established,and the establishment of this relationship provides an opportunity for optimizing the application of soil erosion experience models such as RUSLE in the karst region.In this study,the Yaji River Basin in Guizhou Province was selected as the study area,and natural rainfall experiments were conducted in the basin from 2021to 2022.Five runoff plots with different levels of rock exposure（0%,10%,30%,50%,and 70%）were set up in the experiments,each with a size of 3 m x 12 m and a slope of 15°.Based on two years of meteorological data and observation of runoff and sediment yield,this study explored the characteristics of runoff and sediment yield on karst slopes under erosive rainfall conditions and the impact of exposed rocks on soil and water loss.An attempt was made to establish a quantitative relationship between exposed rocks and soil loss,and a formula for estimating the karst desertification factor（D）was established based on the relationship between the two.Subsequently,a D-RUSLE model was constructed,and its accuracy was evaluated.The D-RUSLE model was applied to estimate the soil erosion status in the Yeji River Basin and to explore its spatial differentiation characteristics and potential influencing factors.The results showed that:（1）During the field positioning observation,a total of 32 erosive rainfall events were collected,which can be classified into three types:A（short duration,moderate rainfall intensity,moderate rainfall amount）,B（medium duration,heavy rainfall intensity,heavy rainfall amount）and C（long duration,moderate rainfall intensity,light rainfall amount）.The A type is the dominant type.There were significant differences（P<0.05）in the runoff and sediment yield of the catchment under B type rainfall compared to A type and C type rainfall,but there was no significant difference in the runoff and sediment yield between A and C types.Overall,the runoff and sediment in the catchment area is consistent with the variation of rainfall,with B>A>C.The runoff is mainly affected by the amount of rainfall,while sediment is mainly affected by the rainfall erosivity.（2）As the rock exposure rate increased,the runoff and sediment yield of karst slopes showed a decreasing trend.Among them,the difference in runoff yield between the 0%,10%,and 30%rock exposure runoff plots was not significant,while the difference in runoff yield between the 50%and 70%rock exposure runoff plots was not significant,but significantly lower than that of the 0%rock exposure plots（P<0.05）.The difference in sediment yield between the 0%and10%rock exposure runoff plots was not significant,but both were significantly higher than the 30%rock exposure plots（P<0.05）.The difference in sediment yield between the 50%and 70%rock exposure plots was not significant,but both were significantly lower than the 10%rock exposure runoff plots.（3）Under erosive rainfall conditions,soil loss on karst slopes showed an exponential function relationship with rock exposure rate,and the fitted model was Ai=ae-b RER（R2=0.834）.Where,a was 0.967 and b was 0.036,reflecting the degree of impact of rock exposure on soil erosion.（4）Compared with RUSLE,the MAE,RMSE and Reof the D-RUSLE model constructed by introducing the karst desertification factor（D）were reduced.That is,the D-RUSLE model had higher accuracy and was more suitable for application on large scales in karst areas.（5）The Yeji River Basin is mainly dominated by slight erosion,accounting for 74.20%of the total basin area,and erosion types of light erosion and above account for 25.80%,that is,25.80%of the areas in the basin exceed the allowable soil loss amount(5 t·ha-2·a-1)in karst areas,which are key areas for soil and water conservation.（6）Soil erosion in the Yeji River Basin was mainly concentrated in the cold regions,which accounted for 67.41%of the total area.The cold regions mainly included urban areas in the central part of the basin and the northern regions with more wooded areas.The erosion hotspots were distributed in some areas of towns such as Renhe Township,Jinpo Township,Gantang Town,Yongxing Township,Guli Town,Yuduo Town,Xinren Township,Jinbi Town,and Jinlan Town.（7）Among the selected factors,elevation,surface roughness,lithology,slope curvature,population density,and per capita GDP all had a certain explanatory power for the spatial differentiation of soil erosion in the basin.Elevation had the highest explanatory power,while population density had the lowest explanatory power.The explanatory power of any two-factor interaction was higher than that of a single factor,with the interaction between elevation and slope curvature having the strongest explanatory power,and the interaction between population density and per capita GDP having the weakest explanatory power.The research results have enriched the soil erosion theory in karst areas,and have helped to deepen the understanding of the soil erosion mechanism on slopes in karst regions,providing a reference for the comprehensive management of soil erosion and desertification in the Southwest Karst Region.The optimization experiments and practices of the RUSLE model can provide valuable reference for further optimizing related soil erosion models like RUSLE in the future. |
PDF Full Text Request