基于GIS与地统计学原理，使用ArcGIS地统计分析模块研究了湖南省青石冈林场土壤(0～60 cm) pH值、有机质、水解性氮、全氮、全磷和全钾空间变异情况。结果表明：实验区土壤为酸性，pH值为4.36，有机质、水解性氮、全氮、全磷和全钾平均含量依次为20.47 g/kg、206.41 mg/kg、1.02 g/kg、0.32 g/kg和11.80 g/kg，该地区pH值变异系数最小，属于小变异，其他土壤养分变异系数在35%以上，为高度变异，变异强度为全钾(72.26%)＞有机质(68.96%)全氮(66.96%)＞全磷(47.06%)＞水解性氮(37.16%)。全钾(49.58%)＞全氮(46.08%)＞水解性氮(40.09%)＞全磷(35.38/%)＞有机质(33.85%)；半方差最优模型拟合分析表明，全氮、全钾的分布符合指数模型，pH值和水解性氮符合高斯模型，有机质符合球状模型，全磷符合有理二次函数模型；从空间结构特征看， pH值和全钾表现出强烈的空间相关性，其变异主要由地形、土壤母质、气候等自然因素(结构性因素)引起，有机质、全氮和全磷表现出中等空间相关性，主要是结构因素及随机因素共同影响造成，而水解性氮空间相关性很弱，表明随机因素对其空间分布的贡献较大；土壤养分的空间分布特征：有机质、全氮、水解性氮和全磷整体分布趋势比较相似，水解性氮的分布情况相对简单，等值线与等高线基本平行，但5种养分均在低海拔富集，有机质和全氮的分布图零碎，受人为及外界干扰明显。采用克里格插值方法进行最优内插，绘制了养分含量分布图，可更直观地了解县域的土壤养分丰缺状况，为土壤可持续利用、次生林的恢复与重建提供科学依据。%Combined the methods and geo-statistics with GIS, the spatial variability of soil nutrients in Hunan province, such as pH, organic matter(OM), hydrolyzable nitrogen(HN), total nitrogen(TN), total phosphorus (TP), Total potassium (TK), were analyzed with ArcGIS geo-statistical analyst. In this study area, soil is acidic. The average content of pH, organic matter(OM), hydrolyzable nitrogen(HN), total nitrogen(TN), total phosphorus (TP) and Total potassium (TK) are 4.36, 20.47 g/kg, 206.41 mg/kg, 1.02 g/kg, 0.32 g/kg and 11.80 g/kg. The results show that the pH had a weak variation, and the variation coefficients of five kinds of soil nutrients over 35%, belong to highly variation. The variation intensity of soil nutrients and soil pH ranked from high to low as follows:total potassium (72.26%), organic matter (68.96%) total nitrogen (66.96%), total phosphorus (47.06%), hydrolysable nitrogen (37.16%).. The parameters of theoretical variogram models for soil nutrients show that the distributions of soil pH and total potassium fitted the exponential model,the pH and hydrolyzable nitrogen fitted Gaussian model, the organic matter fitted Pentaspherical model, the total phosphorous fitted Ratonal Quadratic model. In terms of spatial structure, the pH and total potassium showed high spatial autocorrelation, which was resulted from topography, soil parent material and climate (structural factors) factors. The organic matter, total nitrogen and total phsphorous indicated medium spatial correlation each other, which mainly due to structural factors and random factors, but thr hydrolyzable nitrogen’s spatial autocorrelation was very weak, it was intensive affected by human and outside factors. The random factors made a bigger contribution to the spatial distribution. The spatial distributions of soil nutrients had different characteristics. Organic matter, total nitrogen, hydrolyzable nitrogen and total phosphorous had similar overall distribution trend, but the distribution of hydrolysis nitrogen was relatively simple. The lines of constant altitude and contour line on a map were basically parallel. However, the five kinds of nutrients had obvious enrichment phenomenon at low altitude. The distribution of organic matter and total nitrogen was a little complex, was significantly influenced by humanity. It was more directly to understand the soil nutrients status of Schima superba secondary forests with distribution maps of the nutrients content which were drew by the Bestinterpolation of Kriging method. And it also may give a scientific foundation for sustainable use of soil and the recovery and reconstruction of secondary forests.
展开▼
