黄土高原南北样带不同土层土壤容重变异分析与模拟

发布时间：2018-04-18 01:02

本文选题：容重 + 状态空间　；参考：《农业机械学报》2017年04期

【摘要】：为探明黄土高原南北样带土壤容重空间分布特征,为土壤水文过程模拟与预测提供水力参数,采用经典统计学方法,分析了样带不同土层深度(0~10 cm、10~20 cm、20~40 cm)土壤容重的空间变异特征,并用多元逐步回归、传递函数方程和一阶自回归状态空间模型方法分别对土壤容重的空间分布进行了模拟。结果表明:样带0~20 cm深度土壤容重的变异为中等程度变异,20~40 cm为弱变异。状态空间方程转换系数表明,不同土层深度土壤容重的影响因素不同,0~10 cm主要为有机碳含量、黏粒和砂粒体积分数,10~20 cm为有机碳含量、黏粒和砂粒体积分数和降水量,20~40 cm为黏粒和砂粒体积分数、降水量和土地利用。状态空间模型的模拟效果均优于经典统计的多元逐步回归方程和传递函数方程,基于黏粒和砂粒体积分数、降水量和土地利用的状态空间模型可以解释样带20~40 cm容重92.3%的变异。一阶自回归状态空间模型可用于田间条件下土壤容重分布特征的预测。
[Abstract]:In order to find out the spatial distribution characteristics of soil bulk density in the north and south transect of Loess Plateau and to provide hydraulic parameters for the simulation and prediction of soil hydrological process, the classical statistical method is used.The spatial variation characteristics of soil bulk density in different soil depth (0 ~ 10 cm ~ 10 ~ 20 cm ~ 20 cm ~ 20 ~ 40 cm) were analyzed. The spatial distribution of soil bulk density was simulated by multivariate stepwise regression, transfer function equation and first order autoregressive state space model.The results showed that the variation of soil bulk density at the depth of 0 ~ 20 cm in the transect was moderate to 20 ~ 40 cm, which was a weak variation.The conversion coefficient of state space equation shows that the main factors affecting soil bulk density of different soil depth are organic carbon content, and the volume fraction of clay and sand is 1020 cm, which is mainly organic carbon content.The volume fraction of clay and sand and precipitation of 20 ~ 40 cm are clay and sand volume fraction, precipitation and land use.The simulation results of the state space model are superior to those of the multivariate stepwise regression equation and the transfer function equation of classical statistics. Based on the volume fraction of clay and sand, the state space model of precipitation and land use can explain the variation of the bulk weight of 2040 cm in the sample belt by 92.3%.The first order autoregressive state space model can be used to predict the distribution characteristics of soil bulk density under field conditions.
【作者单位】：西北农林科技大学资源环境学院;西北农林科技大学黄土高原土壤侵蚀与旱地农业国家重点实验室;
【基金】：国家自然科学基金项目(41530854、41501233)
【分类号】：S152

【参考文献】