耕作层土壤养分含量的空间自相关分析——以秦安县郭嘉镇为例

Spatial autocorrelation analysis on soil nutrient contents of plough layer——A case from Guojia Town,Qinan County

为了探究土壤养分含量的空间分布特征,运用空间自相关理论,在Geo Da095i,Arcgis等软件的基础上,以甘肃省秦安县郭嘉镇为例,采用Moran,I统计量研究了土壤主要养分元素含量在三种不同权重矩阵下的空间自相关程度以及空间分布规律。结果表明:秦安县郭嘉镇耕作层土壤中各主要养分空间分布并非随机状态,而是存在一定的空间自相关特性;全域型空间自相关的研究表明,rook权重下秦安县郭嘉镇耕作层土壤中有机质、全氮、全磷、全钾、有效磷和速效钾含量的Moran's I指数分别为0.3088、0.0024、-0.0522、-0.2220、0.2678和0.2264,表明有机质、全氮、有效磷和速效钾的含量在空间分布上呈现空间正相关,而全磷和全钾的含量在空间分布上呈现空间负相关;局域空间自相关结合Moran散点图和LISA聚集图的分析结果揭示出郭嘉镇各行政村养分元素含量分布的空间分布规律,可以直观地得到各养分元素含量分布的"高—高"、"低—低"聚集区和"低—高"、"高—低"孤立区的具体位置,其中有机质、全氮、有效磷和速效钾都在空间分布上出现"高-高"聚集区和"低-低"聚集区,而全磷的含量只有"低-低"聚集区,全钾的含量在空间分布上较为分散,没有明显的聚集区。

In order to explore the spatial distribution of soil nutrient content at Qinan County of Gansu Province, we employed the method of spatial autocorrelation using software like GeoDa095i and ArcGIS. The results showed that： （1） the spatial distribution of the soil nutrient elements in Guojia, Qinan County had a fixed spatial distribution rather than random; （2） Under the rook weight matrix, global- type spatial autoeorrelation analysis indicated positive spatial correlations among soil organic matter, total nitrogen content, available phosphorus and available potassium content showed, with the Moran＇ s I indexes being 0. 3088, 0. 0024, 0. 2678, 0. 2264, respectively. By contrast, negative correlation existed between total phosphorus and total potassium content, the Moran＇ s I values - 0. 0522 and - 0.2220, respectively. （3） Local spatial autocorrelation analysis revealed a ＂high- high＂ and ＂low- low＂ concentrating pattern and a ＂low- high＂ and ＂high - low＂ dispersing pattern for nutrients content, a ＂high - high＂ and ＂low - low＂ pattern for organic matter, total nitrogen, available phosphorus and available potassium, and a ＂low- low＂ pattern for total phosphorus.