基于多变量经验态式分解的喀斯特洼地土壤碳氮多尺度特征  

Multi-scale characteristics of soil carbon and nitrogen in karst peak-cluster depression based on multivariate empirical mode decomposition

作　　者：吴敏[1,2,3] 叶莹莹[1,2,3] 肖霜霜[1,2,3] 张伟[1,2] 王克林[1,2] 陈洪松[1,2] 

机构地区：[1]中国科学院亚热带农业生态研究所、亚热带农业生态过程重点实验室,长沙410125 [2]中国科学院环江喀斯特生态系统观测研究站,广西环江547100 [3]中国科学院大学,北京100049

出　　处：《生态学杂志》2016年第6期1453-1459,共7页Chinese Journal of Ecology

基　　金：国家自然科学基金项目(41471445);中国科学院科技服务网络计划(STS计划);广西喀斯特区生态服务提升与民生改善研究示范项目(KFJEW-STS-092)资助

摘　　要：利用样线法采集喀斯特典型峰丛洼地单元内土壤样品(50个),采用多变量经验态式分解方法,分析喀斯特典型峰丛洼地单元内土壤有机碳(SOC)和全氮(TN)的多尺度特征,探讨了SOC和TN多尺度特征的大小、差异及其主要影响因素。结果表明,SOC和TN分别与环境因子存在4和5个特征尺度。其中SOC主导尺度为68.6与140.9 m,TN主导尺度为72.3与141.1 m,两者主导尺度范围相似,在主导尺度内预测土壤碳、氮含量的决定系数也最高。裸岩率在各特征尺度下始终是土壤C、N空间分布的重要影响因素,而其他环境因子对土壤C、N空间分布的影响在各特征尺度间存在差异,表现为在较小的尺度上(141.0 m)主要受地形湿度指数和土壤水分的影响,在中等尺度内(229.2 m)主要受p H的影响,在较大的尺度内(362.9 m)主要受海拔与坡度的影响。以上研究表明,通过对土壤C、N进行尺度分解,有助于深入认识其空间分布主导尺度的来源以及各尺度内的主要影响因素,研究结果对喀斯特地区适应性景观格局设计和土地利用结构调整具有重要的理论和实践意义。The objective of this study was to apply multivariate empirical mode decomposition（ MEMD） in revealing scale-specific characteristics of soil organic carbon（ SOC） and total nitrogen（ TN） and their main affecting factors. Using the transect method（ 20 m interval）,50 soil samples were collected from the peak-cluster depression. SOC and TN and environmental factors were separated into different intrinsic mode abductions（ IMFs） and residues represented different scales. The multi-characteristic scales were IMF1,IMF2,IMF3 and IMF4 for SOC,and IMF1,IMF2,IMF3,IMF4 and IMF5 for TN. The dominant scales for SOC were similar with for TN.The dominant scales were IMF1 and IMF2（ scale： 68.6 and 140.9 m） for SOC,and IMF1 and IMF2（ scale： 73. 2 and 141. 1 m） for TN. IMF1 and IMF2 presented the main contribution to overall prediction of both SOC and TN. At each scale,SOC and TN were controlled by one or multiple factors. Significant correlations between rock ratio and SOC and TN were found both at measured scales and all IMFs. However,the relations between SOC and TN and the other envi-ronmental factors were complex. The significant correlations between SOC,TN and topographic wetness index and soil water at smaller scales（ 141.0 m） were strongest. The significant correlations between SOC and TN and p H at moderate scales（ 229.2 m） were strongest. The significant correlations between SOC and TN and elevation and slope at larger scales（ 362.9 m） were strongest. Our results showed that SOC and TN were controlled by different environmental factors at different specific scales. Our findings provide a better understanding of the relationships between SOC and TN and their main affecting factors at specific scales,and provide scientific basis for adaptive landscape pattern design and land-use structure adjustment.

关 键 词：多变量经验态式分解 喀斯特 土壤碳氮 特征尺度 

分 类 号：S153.6[农业科学—土壤学]