论文部分内容阅读
以于田绿洲为例,基于地统计学的反距离加权法(IDW)、径向基函数法(RBF)、普通克里金法(OK)及局部多项式法(LPI)等4种不同插值方法,经过交叉验证对比确定了适用的方法进行插值,揭示了盐渍化土壤含盐量、电导率、PH值和含水量的空间变异特征。研究表明:(1)于田绿洲含盐量均值为9.29 g kg-1属重盐土类型,同时土壤电导率、含盐量、含水量和pH分别表现为强、中等、弱空间变异性;(2)土壤特性的结构性因子C/(C0+C)均大于75%,表现出空间自相关较强,是结构性因素主导作用的影响;(3)在电导率和含盐量的空间变异分析采用IDW法较为适合,对pH值的空间插值采用OK法精度最高,含水量则除了LPI法,其他三种方法都可以任意选用;(4)土壤含盐量和电导率空间分布表现出北高南低的分异规律,绿洲外围高于绿洲内部,pH值呈现从东部至西部递减的趋势,易受随机性因素的影响,含水量则出现南北方向明显的差异,研究区中、东部绿洲内高值存在。通过不同插值方法对于田绿洲盐渍化土壤特性的空间异质性研究,揭示了土壤特性空间变异特征及空间分布规律,可为解决土壤盐渍化问题提供基础参考。
Taking Yutian Oasis as an example, four different interpolation methods based on geo-statistics, such as inverse distance weighted (IDW), radial basis function (RBF), ordinary kriging (OK) and local polynomial (LPI) After cross validation, the suitable methods were interpolated to reveal the spatial variability of salinity, conductivity, pH value and water content of salinized soils. The results showed that: (1) The average salt content of Yutian oasis was 9.29 g kg-1, and the soil conductivity, salt content, water content and pH showed strong, moderate and weak spatial variability respectively; ( 2) The structural characteristics C / (C0 + C) of soil properties were all more than 75%, showing the strong spatial autocorrelation and the dominant effect of structural factors. (3) In the spatial variability of conductivity and salinity The method of IDW is more suitable for analysis, the OK method is the most accurate for the spatial interpolation of pH values, and the other three methods can be selected for the water content in addition to the LPI method. (4) The spatial distribution of salt content and conductivity in soil shows north The differentiation of high and low south is higher than that of the oasis in the periphery of the oasis. The pH shows a decreasing trend from the east to the west and is easily affected by the randomness. The water content shows a significant difference in the north and south directions. In the middle and eastern oases High value exists. Spatial heterogeneity of soil properties of Yutian oasis salinization by different interpolation methods reveals the spatial variability and spatial distribution of soil characteristics and provides a reference for solving the problem of soil salinization.