遥感在森林地上生物量估算中的应用  

Application of remote sensing in forest aboveground biomass estimation

作　　者：何红艳[1] 郭志华[1] 肖文发[1] 

机构地区：[1]中国林业科学研究院森林生态环境与保护研究所,北京100091

出　　处：《生态学杂志》2007年第8期1317-1322,共6页Chinese Journal of Ecology

基　　金：国家自然科学基金重大研究计划项目(90211006);国家重点基础研究发展规划项目(2002CB412508);国家自然科学基金重大项目(30590383);国家科技十一五攻关计划项目(2006037002027);国家林业局重点试验室开放基金资助项目

摘　　要：生物量是地表C循环研究的重要组成部分,生物量研究有助于深入认识区域乃至全球的C平衡。森林作为地球最重要的陆地生态系统,区域乃至全球尺度的森林地上生物量估算一直是生态学研究的难点之一。在大的空间尺度上,遥感技术是估算森林地上生物量的有效手段。TM、AVHRR、SAR等数据以及多源数据的融合在森林生物量估算方面广泛应用,并取得了显著效果。运用遥感技术进行森林生物量估算时,所采用的数据源不同,分析方法也不相同,主要分析方法有:相关分析、多元回归分析、神经网络和数学模型模拟等。随着测定不同空间、时间和波谱分辨率的各种传感器的广泛使用,以及生物量遥感估算模型的进一步发展和完善,大尺度森林生物量的遥感估算研究必将向前迈进一大步。Biomass is an important variable in many ecological and environmental models, and its large scale estimation would be helpful to understand the global C balance. Forest plays an important role in the global carbon budget because it determines the dynamics of terrestrial carbon cycle, but the measurement of its aboveground biomass involves extensive field surveys. On large regional scale, using satellite data is an effective way to estimate forest biomass. Up to now, TM, AVHRR, radar data and the fusion of multiple satellite data have been widely used in forest biomass estimation. During estimating with satellite data, different mathematic methods may be required if the data sources are different. The main methods for the estimation are correlation analysis, regression analysis, neural network, and other mathematic models. New instruments with different resolution in spatial, temporal and spectrum are devised for the relatively reliable deter mination of forest aboveground biomass. With the development and improvement of the theories and models for biomass estimation by using remote sensing data, great progress will be achieved in the research of forest biomass on large scales.

关 键 词：生物量 遥感 神经网络 数学模型 

分 类 号：S718.5[农业科学—林学]