摘要
采用半方差函数、克里格空间插值等地统计学方法分析了吉林省蛟河市阔叶红松林土壤养分的空间异质性。结果表明：各养分指标的变异系数均在6．29％～34．43％，除土壤pH值外，其他养分指标均表现为中等程度变异；全氮、速效磷、速效钾、pH值和有机质的最佳拟合模型为指数模型，全磷和全钾的最佳拟合模型为球状模型：全钾、速效磷和pH值的块金值和基台值的比值较小，表现出强烈的空间自相关性，全磷、有机质、全氮和速效钾表现为中等程度的空间自相关性：土壤养分各指标的自相关距离在3．5～92．2m，全氮、速效磷、速效钾、pH值和有机质的分维数比较大。说明其空间分布格局复杂程度比较大；全氮和全磷的分维数比较小，说明其结构性比较好，空间分布格局简单。
We used geostatistic methods of semivariogram and Kriging interpolation to analyze the spatial variability of soil nutri- ents in broad-leaved Korean pine forest of Jiaohe, Jilin Province. The coefficient of variation for soil nutrients ranges from 6.29% to 34.43%, and it can be considered as relatively moderate except soil pH. The variogram of total nitrogen （TN）, available phosphorus （AP）, available potassium （AK）, pH and organic matter （OM） can be well fitted by expo- nential model, and total phosphorus （TP） and total potassium （TK） can be well fitted by spherical model. The ratio of nugget and sill of TP, AP and pH is lower with strong spatial autocorrelation, while that of TP, OM, TN and AK is with moderate spatial autocorrelation. The regressed range of soil nutrients is 3.9-92.2 m. The fractal dimension of TN, AP, AK, pH and OM is higher with more complex spatial distribution pattern, while the fractal dimension of TN and TP is low er with simple spatial distribution pattern.
出处
《东北林业大学学报》
CAS
CSCD
北大核心
2013年第10期98-102,共5页
Journal of Northeast Forestry University
基金
"十二五"国家科技支撑计划项目(2012BAC01B03-5)
关键词
阔叶红松林
土壤养分
地统计学
空间异质性
克里格插值
Broad-leaved Korean pine forest
Soil nutrients
Geostatistics
Spatial heterogeneity
Kriging interpolation
