Variability Of Soil Nutrients Of Dryland In Microtopography Of Mid-Sichuan Hilly Region

Terrain is a condition impact on material and energy exchange between soil and environment.By other soil forming factors,surface material and energy redistribute.In the past,spatial variability characteristics of soil physico-chemical properties were analyzed more through spatial interpolation methods of GIS.Because of the complexity of the terrain and the improvement of spatial interpolation methods of GIS,there are some errors. They were analyzed mostly based on large and medium-scale.However,the study of spatial variation of soil nutrients through digital terrain model access to relevant quantitative environmental variables under the micro-scale topographic in hilly area is still uncommon reported.Spatial variability of soil nutrients in microtopography of Mid-Sichuan Hilly Region is discussed.The research results are that:(1) OM,Total N,Total P,Total K,Avail.N,Avail.P and Avail.K content are 10.05 g·kg^-1, 0.35 g·kg^-1,0.60 g·kg^-1,9.22 g·kg^-1,29.38 mg·kg^-1,9.24 mg·kg^-1 and 86.06 mg·kg^-1 in the study area.According to nutrient classification standards,in the study area soil organic matter,Total N,Avail.P and Avail.N are deficient,but Avail.K is rich.From the coefficient of variation,it is all medium variant,which is mainly due to soil erosion in the region.The coefficient of variation of SOM is similar to that of Total N,which confirms that there is a positive correlation between them.The coefficient of variation of Total K is minimum, 18.0%,which is perhaps connected with organic carbon content.According to the K-S-sample tests,all are normal distribution.(2) There are specific differences for the soil nutrient content in different landscape position,the diversification of OM and Total N is:hilltop＜middle slope＜upper slope＜lower slope＜foot slope.The content of Avail.N,Avail.P and Total N manifest:hilltop＜upper slope＜middle slope＜lower slope＜foot slope.The content of Total K reveals: lower slope＜middle slope＜upper slope＜hilltop＜foot slope;the content of Avail.K manifests:hilltop＜upper slope＜middle slope＜lower slope＜foot slope.(3)The variation order of soil nutrient content among slope shape is:concave slope＞linear slope＞convex slope.(4)The characteristics of soil nutrient content among different slope are:because of the soil erosion and the impact of terrain on solar radiation,there is a tendency:with the increase of slope,the soil nutrient content gradually decreases.(5) The difference of soil nutrient content among different slope aspect is not obvious, which might be the result of the relative lowness of hill in the study area,and little difference about the aspect of solar radiation received.However,it shows a certain trend: Avail.K:Shady slope＞Sunny slope＞semi-Shady slope.Other soil nutrient content in the shade and the sunless is higher than the sunlit.(6) By analyzing topography indices extracted from spatial analysis on GIS platform, and soil nutrient content,it indicates that there is positive correlation between OM,Total N, Total P,Avail.N,Avail.P,Avail.K content and elevation(h),profiles Curvature(Kv)(a= 0.01),there is an obvious negative correlation between OM,Total N,Avail.N,Avail.P, Avail.K content and compound topographic index(CTI)(a=0.01),there is significant positive correlation between OM,Total N content and Convergence momentum index(SPI) (a=0.01).there is a significant positive correlation between Avail.N content and the convergence of momentum index(SPI)(a=0.05),there is a significant positive correlation between Total P content and the convergence of momentum index(SPI) and compound topographic index(CTI),there is a significant negative correlation between Total K content and slope(β)(a=0.01),there is a negative correlation between Total K content and slope shape(SS)(a=0.05),there is an obvious negative correlation Total K between and elevation(h)(a=0.01).Thus,h and kv is the main impact factors of soil nutrients in the hilly region.(7) The Multi-dimensional stepwise regression analysis is carried on with various nutrients as the dependent variable and the corresponding terrain indicator as the independent variable under the SPSS 12.0 software platform.The analysis result indicates that the final model decision coefficients of R² of the models of OM,Total N,Avail.N, Total P,Avail Avail.P,Total K,and Avail.K are:0.579,0.470,0.419,0.394,0.410,0.386, and 0.471,respectively.After the model examination,it was found that the MPE of the OM, Total N,Total P,Total K,Avail.N,Avail.P and Avail.K content of the soil are-1.940,0.072, -0.368,-0.782,-6.341,-1.340,-8.041,the RMSPE are 2.570,0.096,0.388,3.141,8.828, 2.003,24.422 respectively.