福建武夷山自然保护区森林碳储量遥感估测方法与空间分析  

Estimation and spatial analysis of forest carbon stocks based on remote sensing technology in Wuyi Mountain National Reserve of Fujian province

作　　者：李明阳[1] 吴军[2] 余超[1] 时宇[1] 

机构地区：[1]南京林业大学林学院,江苏南京210037 [2]环境保护部南京环境科学研究所,江苏南京210042

出　　处：《南京林业大学学报（自然科学版）》2014年第6期6-10,共5页Journal of Nanjing Forestry University：Natural Sciences Edition

基　　金：国家环保公益性行业科研专项项目(201509043)

摘　　要：以福建省武夷山国家级自然保护区为研究对象,以2003年24块固定样地数据、同年Landsat TM遥感数据为主要信息源,分别采用多元线性回归、K最邻近分类算法、人工神经网络、土地覆盖分类4种方法,对研究区域2003年的碳储量进行遥感估测。在此基础上,对森林碳密度进行地理加权回归及空间格局分析。研究表明:在4个模型中,人工神经网络的相关系数最高,标准误差、平均相对误差最低,预测精度最高;研究区不同功能分区碳密度差别不大,森林平均碳密度为52.40 t/hm2,碳储量为278.542 3万t;森林碳密度与所处位置的海拔、坡向负相关,与坡度、林地土壤水文状况、植物生长状况正相关;随着海拔的降低和人为干扰活动的增强,核心区、缓冲区、实验区森林碳密度的空间聚集性减弱,破碎化趋势增强。Forest carbon sink is important for mitigating the climate change. Monitoring and quantifying the aboveground forest carbon sink with remote sensing has become a hotspot in the research field of forest remote sensing. Wuyi Mountain National Reserve in Fujian province was chosen as the case study area,while 24 permanent sampling plot data in 2003 and Landsat TM remote sensing images in the same year were collected as the main sources of information to estimate the forest carbon stocks in the study area in 2003 by means of four methods of multiple linear regression,K-nearest neighbor classification,artificial neural networks,and land cover classification,followed by geographically weighed regression and spatial pattern analysis of forest carbon density. Study results show that： 1 Among the four models,artificial neural network outperformed others with the highest correlation coefficient,the lowest standard error and the minimum average relative error; 2 The difference of carbon density was not very obvious among three functional areas of the reserve,average carbon density was 52. 40 t / hm2,totaly 2 785 423 t; 3 Carbon density was negatively correlated with the elevation and aspect,positively correlated with slope,hydrological conditions of forest soil and plant growth conditions; 4 With the decrease of altitude and enhancement of human disturbance,from the core zone,buffer zone to the experimental area,the spatial aggregation of forest carbon density tended to become more weakened and more fragmented.

关 键 词：森林碳储量 遥感估测 空间分析 武夷山自然保护区 

分 类 号：S757.2[农业科学—森林经理学]