东莞市森林生态系统土壤有机碳含量的地统计学分析  

Geostatistical Analysis of Soil Organic Carbon Content in Forest Ecosystem of Dongguan

作　　者：沈德才 刘婷 莫罗坚 周海琪 Shen Decai;Liu Ting;Mo Luojian;Zhou Haiqi(Dongguan Institute of Forestry,Dongguan Guangdong 523106)

机构地区：[1]东莞市林业科学研究所,广东东莞523106

出　　处：《热带林业》2022年第3期45-49,共5页Tropical Forestry

基　　金：广东省科技基础条件建设项目(2013B06050074):东莞市森林生态效益监测网络体系建设研究。

摘　　要：该研究以2 km×2 km的UTM网格为研究单元,结合经典统计学与地统计学方法研究了东莞市森林生态系统的土壤有机碳含量的空间分布规律。结果表明:四层土层有机碳含量的半方差函数值均较小。L1与L2两层的土壤有机碳含量为线性模型,空间结构特征较弱;L3与L4两层为指数模型,空间结构特征为中等强度。各土层空间自相关指数Moran’I数值基本上接近0,表明空间上的正负自相关均不明显。这种结果进一步证明了控制土壤有机碳含量空间异质性的关键影响因素是林分因子,包括林分类型、林龄、凋落物状况等因素。经克里格插值绘制的各土层土壤有机碳含量分布图显示,L1、L2两层土壤有机碳含量变异特征差异较小,空间变异性小,依赖性小,破碎化程度较高;而L3与L4层的土壤有机碳含量变异特征较相似,空间依赖性相对较高,破碎化程度较低。Forest ecosystem soil organic carbon content spatial characteristics were investigated with geostatistical methods based on a 2 km×2 km UTM grid system overlaid on the forested land of Dongguan. The results are as follows:the soil organic carbon content in varoius soil layers had small semivariance. SOC content in soils of L1-L2 was best fitted by the linear model,showing weak spatial structure;whereas SOC content in soils to a depth of L3-L4 was best fitted by the exponential model,exhibiting spatial structures with medium strength.Only weak spatial autocorrelation of SOC content in each soil layer was detected,with the autocorrelation index Moran’s I close to 0. The results from geostatistical analysis validated the findings that the key factors controlling spatial variability of SOC content were those of forest types,stand age,and litter regimes. A kriging map for the SOC content in each of the soil layer showed in details smaller spatial heterogeneity and interdependance and higher fragmentation for the L1 and L2 soil layers,and similar spatial variabilty and interdependance and lower fragmentation for the L3 and L4 soil layers.

关 键 词：土壤有机碳含量 地统计学 森林生态系统 东莞 

分 类 号：S714.5[农业科学—林学]