Study On Soil Erodibility K Values In Small Watersheds Of Purple Soil In The Three Gorges Reservoir Area

As a special soil type of China, the purple soil is characterized by soft texture, loose structure and high erodibility. Despite high productivity, the purple soil is very vulnerable to weathering and soil loss and it is the main sediment source of the Yangtze River and its main tributaries. So as to specifically measure the difficult degree of separation, erosion and transport soil erodibility factor K affected on purple soil and the sensitivity of purple soil erosion, soil erodibility factor K has become one of the key indicators to reflect the property.This paper takes Daijia Watersheds, one typical small watershed of purple soil, in Wanzhou ecological environment monitoring station as the research object. To better understand mechanism characteristics of purple soil erosion, corresponding erosion space and the relationship with the land use types, the paper, by surveying the watershed, collecting and determining the soil samples, and analyzing the data, conducts a systematic and comprehensive calculation and analyses on the soil in the watershed respectively from two study angles which are spatial location of the watershed and land use status based on two kinds of calculation model (EPIC model and NOMO model) of soil erosion.Contrastively analyzing and evaluating the corresponding characteristics of soil erodibility factor K values calculated by EPIC model and NOMO model respectively. The main conclusions of the study are as follows:(1) Based on the EPIC model and NOMO model, the main affecting factors of soil erodibility K values at a scale of small watershed are the characteristics of composition of different size soil particles and organic matter content features, and the two parameters are mainly affected by land use type and basin location.(2) Analyzes on parameters of the EPIC model and the NOMO model (factors of soil particles and soil organic matter) indicate that there are differences of characteristics of soil particles content in the same basin position under different land use types. Among them, it is inferred by taking the sand contents as example that Zea mays L is with the largest variation change of sand content (36.70～20.7%), Lpomoea batatas takes second place with it ranged from 39.70% to 24.21%. And sand content wildgrassground ranges from 61.20% to 46.21% meanwhile that of Pinus massoniana Lamb ranges from 50.20% to 35.81%. Both change range of them are relatively lower. Citrus reticulata Banco has the relatively lowest variation amplitude of sand content (55.25～41.33%). Soil particles such as sand, silt and clay have arisen various degrees of increase or decrease with the change of space position of the watershed under the same land use type but in different basin locations. Soil particles characteristics of six land use types from the upper to lower reaches of the river all express a trend that sand is gradually decreasing and silt and clay are gradually increasing. However wildgrassground, soil organic matter of Pinus massoniana Lamb, Citrus reticulata Banco and Zea mays L changes in the order the upper>the middle reaches> downstream while Robinia pseudoacacia L varies like the upper>downstream>the middle reaches.(3) The correlation analyses between index factors of different soil erodibility matrixes and soil erodibility factor K have been done. Soil organic matter content is negative correlated with KEPIC calculated under model EPIC(R=-0.683**). And content changes of sand, silt and clay are correlated with KEPIC (R=0.941**, R=-0.935**and R=-0.838**) respectively. Then, it is founded that the content changes of sand, silt and clay affect the values of KEPIC more contrasted to soil organic matter content. Thence, the relationship of value of KEPIC with characteristics of soil particles is more intimate. Yet, what more closely relate to soil erodibility K values (KNOMO) are mainly contents of silt, clay and organic matter on the condition of using model NOMO. EPIC model and NOMO model have different emphases and standards on evaluating K values. As a result of it, the same factor has difference in influence degree on K value and correlation with K value in different K-calculated matrixes. However soil organic matter content can be an indicator with significant correlation measuring the relationship between it and soil erodibility K using EPCI and NOMO indicating that organic matter has a better promotion effect on inhibiting soil erosion.(4) It is founded that, values KEPIC of soil surface in the whole watershed are in the order wildgrassground>Citrus reticulata Banco>Pinus massoniana Lamb>Robinia pseudoacacia L>Lpomoea batatas>Zea mays L accroding to analyzing each spatial response characteristics of soil erodibility K values of six land sue types calculated by models of EPIC and NOMO respectively and comparatively analyzing them. Meanwhile the writer finds that spatial response characteristics of the K values under the same land use type differed significantly that the values of KEPIC of the soil on the upstream, midstream and downstream of wildgrassground, Pinus massoniana Lamb, Robinia pseudoacacia L and Citrus reticulata Banco both shown a significant difference (P<0.05). While the values of KNOMO in the whole watershed were in the order Zea mays L>Lpomoea batatas>Pinus massoniana Lamb>Robinia pseudoacacia L> Citrus reticulata Banco>wildgrassground. And what similar to KEPIC was that KNOMO on different positions of these four land use types also show a significance of difference, except Zea mays L and Lpomoea batatas of which values of KNOMOo had no significant differences (P<0.05). The two models show differences and similarities in evaluating spatial response characteristics of soil erodibility K values of six land use types. This is may be due to the different emphasizes of the each models. For example, EPIC focuses on characteristics of soil particles and nutrient as well as NOMO does the same and also involves macroeconomic performances of soil structure constituted by soil particles such as penetration level and structure index, etc. On the another hand, it also prove EPIC and NOMO have similar evaluation result and analyse effect when judge the significant differences of spatial variability of soil erodibility K values under the same type of land use.(5) The writer analyzes response characteristics of different-locations soil erodibility K values to land use types and contrasts them. And it can be infer from analyses on soil erodibility K values under six land use types in the entire basin that wildgrassground and Citrus reticulata Banco have relatively highest KEPIC values on its upstream and midstream. But the values of Pinus massoniana Lamb and Robinia pseudoacacia L in the entire watershed are smaller and the values of Zea mays L and Lpomoea batatas are smallest. And it can be seen from the analysis of significant difference that KEPIC size between each two soil erodibility KEPIC values in the upstream and downstream locations of wildgrassground, Citrus reticulata Banco, Robinia pseudoacacia L and Pinus massoniana Lamb all reflected significance of difference. That were, in the midstream of the watershed, KEPIC size between Pinus massoniana Lamb and Robinia pseudoacacia L, wildgrassground and Citrus reticulata Banco have significant difference while it is not different between Pinus massoniana Lamb and Robinia pseudoacacia L. Yet values of KNOMO of Lpomoea batatas and Zea mays L in the entire watershed relatively are the biggest which ranges from0.0754 to 0.0864 and 0.0746 to 0.0845 respectively. However Pinus massoniana Lamb's KNOMO is relatively lower ranged from 0.0553 to 0.0770 and for wildgrassground and Robinia pseudoacacia L they are the smallest respectively range from 0.0498 to 0.0784and 0.0589to 0.0766.It can be concluded from analysis of difference significance that values of KNOMO of Zea mays L and Lpomoea batatas of the upstream and midstream have no obvious difference the same as values of KNOMO of Pinus massoniana Lamb and Robinia pseudoacacia L. The reasons why differences of response characteristics of values of land use type on KEPIC and KNOMO which are soil erosion indexes emerged are similar to what make spatial response characteristics of KEPIC and KNOMO differed. This may be due to the differences of land use types of the watershed, effects of human disturbance and different degree of transformation vegetation root do to the soil except the different emphasizes the two models hold.Therefore studying on distribution of response characteristics of soil erodibility K values of typical watershed of purple soil is meaningful for discussing influence mechanism of soil erodibility at a scale of small watershed. At the same time it provides some degree of data base and theoretical guidance for building and testing soil erosion prediction model based on land use types and spatial variability in the Three Gorges Reservoir Area. This provides certain foundation and practical significance for comprehensively understanding soil erosion and forecasting soil erosion intensity further.