【作者】 范文杰；

【导师】 张洪江；

【作者基本信息】 北京林业大学 ， 水土保持与荒漠化防治， 2014， 博士

【摘要】 土壤及土壤的肥力状况是农林业生产的重要基础以及可持续发展的重要保障,土壤肥力的高低对人类社会的发展有着重要的意义。因此,对土壤肥力状况及其空间分布特征进行系统研究成为当下热点问题之一,研究土壤肥力的空间分布,可以更好地了解人为活动对土壤肥力的影响。本论文以冀东平原地区河北省吴桥县为研究地点,分析了吴桥县耕地土壤属性的基本统计特征,通过ArcGIS10.1软件对土壤肥力要素的空间分布进行插值并绘制了其空间分布图,运用熵权法与模糊评价法相结合的方法,对吴桥县县域的土壤肥力进行了综合分析与评价。吴桥县耕地土壤主要养分含量除速效钾降幅大外,全氮、有机质、碱解氮、速效磷与全国第二次土壤普查相比均明显上升,微量元素缺乏的农耕地面积很小,研究区微量元素整体上基本处于中等偏上水平,其中铁和铜两种微量元素含量较高。通过分析发现土壤各有关指标之间存在着一定的相关性。采用主成分分析的方法分析了吴桥县不同乡镇耕地土壤各属性的综合得分值,结果为宋门乡、铁城镇、杨寺乡及曹洼乡土壤养分综合得分值较高,其次为桑园镇、何庄乡、沟店铺乡、梁集镇、于集镇,安陵镇最低。应用SPSS13.0软件对土壤pH值、有机质与各自相关的变量进行回归拟合分析,结果表明逐步回归分析较之一元回归分析、一元非线性回归分析、多元线性回归分析在拟合度方面有着明显的提高,说明逐步回归模型比其它几种模型的预测能力要强,在预测能力上也较为优化。通过分析吴桥县土壤各属性要素的全局性趋势效应得出,土壤pH值、有机质、全钾、速效钾及有效铜为近似无趋势效应；土壤有效锰、有效锌、全氮、碱解氮、全磷、速效磷及有效铁的趋势效应为一阶。从土壤养分半方差函数相关参数及理论模型拟合的结果来看,土壤pH值、有机质、全氮、碱解氮、全磷、全钾、有效铁、有效猛、有效铜、有效锌符合指数模型,土壤速效磷、速效钾符合线性模型。土壤属性空间变异规律为：碱解氮>有效锌>全磷>有效锰>有机质>全氮>全钾>pH值>有效铜>有效铁>速效钾>速效磷。土壤各肥力要素变程之间有明显差异,最大的为全氮为2592.0m,最小的为碱解氮和全钾为30.0m,仅为全氮的1.16%。按照空间依赖性的大小土壤各属性排列为：全氮>pH>有效铁>有效铜>速效钾=速效磷>全磷=有效锰>有机质=有效锌>全钾=碱解氮。土壤的肥力元素空间分布是土壤自身存在的一种自然特性,结合地统计学方法对研究区土壤肥力因子进行普通克里格法插值,分析和探讨了吴桥县土壤肥力要素有关指标的空间分布规律,并通过交叉检验法来对各土壤肥力指标要素的普通克里格插值进行精度检验,结果表明各项土壤指标插值精度较高,结果可靠。运用熵权法和模糊评价法相结合的方法对吴桥县4965个采样点的土壤样品分析结果,进行了土壤养分指标的综合评价,并绘制了土壤肥力空间分布图。研究结果表明吴桥县土壤肥力水平以三级为主,三级地面积最大,占县域面积的69.98%；其次为二级地,占县域面积的29.16%；四级地面积占县域面积的0.85%；一级地面积最少,仅为吴桥县域面积的0.01%。从全县整体情况来看,三级地主要分布在吴桥县西北部的桑园镇、安陵镇和梁集镇；二级地主要分布在东南部的曹家洼乡、铁城镇和杨家寺乡；四级地主要分布在安陵镇的西北部的徐口村、范庄村及莫场等村庄,以及梁集镇的徐连九村和牟家庵村；一级地分布在宋门乡的西宋门村的西部,仅有较少的面积分布。从土壤养分的高低区域性分布来看是：东北部较高,西南部较低,东部较高,西部较低。更多还原

【Abstract】 Soil and soil fertility are the important foundation for agroforestry production and guarantee of sustainable development, soil fertility level has an important significance for development of human society.Wherefore, the systematic study has become one of the current hot issues on the soil fertility status and spatial distribution characteristics.Studing the spatial distribution of soil fertility can better understand the impact of human activities on soil fertility.This paper analyzed the properties of soil basic statistical characteristics taking Wuqiao County as the investigative object.Through ArcGIS10.1, the spatial distribution of soil fertility elements were interpolated and spatial distribution were plotted.The soil fertility was conducted a comprehensive evaluation taking advantage of the method of entropy and fuzzy evaluation.The content of total nitrogen, organic matter,nitrogen,available phosphorus were significantly increased than the second national soil survey in Wuqiao county in addition to the large decline of available K.The lack of trace elements of land area was small,and the content of trace elements was basically in the middle level. Among them, two kinds of trace elements iron and copper content was higher.There were some correlations among indicators of soil by the correlation analysis.The composite score of farmland were analyzed by principal component analysis in different towns of Wuqiao County.The result showed:Songmen township, Tiecheng town, Yangsi township and Caowa township were relatively higher in composite score value, followed by Sangyuan town, Hezhuang township, Goudianpu township, Liangji town, and Yuji town.Anling town was the lowest value in composite score.The regression analysis were fitted on soil pH, organic matter associated with each variable. The result showed:the accuracy of stepwise regression analysis was improved compared with one of Linear regression analysis with one factor, Non-linear regression analysis with one factor, multivariate linear regression analysis. The predictive ability of stepwise regression model was stronger than other models, then stepwise regression model predictive ability than several other models stronger.The soil pH, organic matter, total potassium, available potassium and available copper were approximate no trend effect by analyzing the effects of the global trends of Wuqiao soil derived elements.The trend effect available Mn, available Zn, total nitrogen, available nitrogen, total phosphorus, available phosphorus and available iron were a valid order. The results showed:the soil pH, organic matter, total N, total K, total P, available Fe,available Mn, available Cu and available Zn belonged to the exponential model and available P and available K belonged to the linear model from soil nutrients semivariogram parameters and theoretical models.The Variation was available N> available Zn>Total P>available Mn>organic matter>Total N>total K>pH>available Cu> available Fe>available K> available P by accordance with the nugget and sill ratio(C0+C1).There were significant differences among the change process of elements of soil fertility,the maximum for the total nitrogen was2592.0m,and the smallest for nitrogen and potassium were30.0m,only1.16%of the maximum value.Where fertility elements of spatial dependence of the size can be arranged as follows:Total N>pH>available Fe>available Cu>available K=available P>Total P=available Mn>organic matter=available Zn> total K=available N.The spatial distribution of soil fertility elements is a natural characteristics.The ordinary Kriging interpolation combined statistical methods was implemented for soil fertility elements in experimental zone, and analyzes and discussed the spatial distribution on soil fertility indicators in Wuqiao County. The interpolation accuracy was relatively high and reliable, through cross-validation method to test the accuracy of the various elements of soil fertility4965soil sample analysis results were evaluated comprehensively on soil nutrient indicators by entropy and fuzzy evaluation method, and the spatial distribution of soil fertility was drawn. The results showed that:Wuqiao County fertility levels was mainly third major place, followed by the sencond major place.The largest was the third major place, accounting for the county area of69.98%, followed by sencond major place, accounting for the county area of29.16%, the fourth major place area accounting for the county area of0.85%, the first major place is only accounting for the county area of0.01%.Judging from the county as a whole, the third major place of Wuqiao County were in the town of Sangyuan, Anling and Liangji in the northwest; and the second major place were in Caojia, Tiecheng and Yangjiasi in the southeastern, and the fourth major place were in Xukou, Fanzhuang and Mochang and other villages in the northwestern Anling town, and Xulianjiu villages and Moujiaan in LiangJi town. The first major place were in the west Xisongmen Village, only less area distribution.The level of nutrients from the soil of regional distribution, that is:the higher the northeast, southwest lower, the higher the eastern and western lower.更多还原