基于TM的北亚热带高山区日本落叶松林生物量模型  

Model of biomass for Larix kaempferi forest in northern sub-tropical Alpine area based on TM

作　　者：王宏全[1] 孙晓梅[1] 张灿明[2] 李子敬[1] 李忠国[1] 

机构地区：[1]中国林业科学研究院林业研究所国家林业局林木培育重点实验室,北京100091 [2]湖南省林业科学研究院,湖南长沙410004

出　　处：《中南林业科技大学学报》2010年第10期10-17,共8页Journal of Central South University of Forestry & Technology

基　　金：国家"十一五"科技支撑课题(2006BAD24B06)和(2006BAD01A14)

摘　　要：在北亚热带高山区日本落叶松林内均匀布设51块标准地并测定实际生物量,从同期的TM数据中提取多种植被指数、波段均值,和现地调查的海拔、坡度一起作为自变量进行建模。结果表明:单波段TM4与生物量的相关性最强(R=0.814),TM3与生物量的相关系数为最小的0.074;植被指数INDVI、IDVI、ISAVI、IMSAVI和IRVI均与生物量极显著相关,其相关系数分别为0.912、0.906、0.901、0.891和0.896;生物量与坡度负相关而与海坡正相关;INDVI对该地区生物量的指示效果最佳,且二次多项式模型的拟合精度最高,其R值和平均残差分别为0.912和-1.5。The actual biomass of 51 plots that were set for Larix kaempferi in the northern sub-tropical alpine area were measured.Vegetation index and average band value were extracted from TM data at the same time.All the index including the average elevation and slope were set as independent variables.The results show that single band TM4 had strongly correlation with biomass（R=0.814） while the correlation between TM3 and biomass was weak（R=0.074）;Vegetation index such as INDVI,IDVI,ISAVI,IMSAVI and IRVI were significantly associated with biomass and the correlation coefficient of them were 0.912,0.906,0.901,0.891 and 0.896 respectively;The slope of the plots were negatively related to the biomass while the correlation coefficient between average elevation and biomass were positive;The INDVI was appropriate for the biomass estimation in this areas,and polynomial model was with highest accuracy,the R value and average residual were 0.912 and-1.5 respectively.

关 键 词：日本落叶松 生物量模型 TM数据 植被指数 

分 类 号：S791.223[农业科学—林木遗传育种] S771.8[农业科学—林学]