【作者】 曹斌；

【导师】 谭炳香；

【作者基本信息】 中国林业科学研究院 ， 森林经理学， 2010， 硕士

【摘要】 森林是重要的再生资源,作为陆地生态系统的主体,森林是陆地上面积最大、分布最广、组成结构最复杂、物质资源最丰富的生态系统,也是自然界功能最完善的资源库,生物基因库及能源储存调节库。对改善生态环境,维护生态平衡具有不可替代的作用。叶面积指数(leaf area index, LAI)又称为叶面积系数,是一块地上植被叶片的单面总面积与占地面积的比值。叶面积指数是植物生态研究中一个十分重要的植被结构参数,是表达植被冠层结构的最基本参数之一,其已成为一个重要的森林定量评价指标。本文利用多角度高光谱CHIRS数据对森林叶面积指数进行反演研究；通过对CHRIS传感器高光谱、多角度特点的分析,估算叶面积指数。主要研究内容与结论如下：(1)对CHRIS数据进行深入研究,找到适合CHRIS数据的预处理方法。通过正射纠正,对CHRIS数据5个角度图像进行配准。(2)通过归一化植被指数(normalized difference vegetation index, NDVI)、比值植被指数(ratio vegetation index, RVI)、修正简单植被指数(modified simple ratio index, MSR)对针叶林、针阔混交林进行LAI反演研究。在相同条件下,针叶林、针阔混交林的RVI与LAI的相关性最好。(3)对CHRIS数据不同波段组合的NDVI、RVI、MSR与LAI的相关性进行分析,针叶林与针阔混交林对不同的波段的敏感性不同。针叶林三种植被指数与LAI相关性最好的为红光10波段(中心波长：709nm),最大R2为0.7098；针阔混交林三种植被指数与LAI相关性最好的为红光8波段(中心波长：697nm),最大R2为0.6385。(4)对CHRIS数据不同角度图像间NDVI、RVI、MSR与LAI相关性进行分析,针叶林、针阔混交林0°图像植被指数与LAI的相关性最好。0°图像针叶林样地植被指数与LAI相关性最好,最大R2为0.7098,-36°图像次之,+36°图像植被指数与LAI没有明显的相关性。针阔混交林0°图像与LAI的最大R2为0.6385,-36°、+36°图像植被指数与LAI的相关性分别次之。更多还原

【Abstract】 Forests are important regeneration resources. As the principal part of terrestrial ecosystems, forests are the largest, the most widely distributed, the most complex structure and the most abundant resources of ecosystem; also it is the most comprehensive resource library in nature, biological gene bank and energy storage regulation library. It has an irreplaceable role on improving environment and maintaining ecological balance.Leaf area index (LAI) is the ratio between a total area of standing vegetation and single leaf area. Leaf area index is a significant vegetation structural parameter in the study of plant ecological and one of the most basic parameter of vegetation canopy structure, it also has become an important forest quantitative evaluation standard. In this paper, the multi-angle hyperspectral CHIRS data is used to inverse leaf area index. Through the analysis of hyperspectral and multi-angle sensor CHRIS, estimate leaf area index. Main content and conclusion include the followings:(1) The preprocessing method for CHRIS data can be found through analysing the CHRIS data. By ortho correction, five CHRIS images have been matched.(2) Use normalizing difference vegetation index (NDVI), simple vegetation index (RVI), modified simple ratio index (MSR) to inverse the conifer forest and the mixed coniferous broad leaved forest’s LAI. It showed that RVI correlated best with LAI under the same conditions.(3) Analyzing the relevance of LAI with NDVI, RVI, MSR that were in different bands of the CHRIS data, and found that the coniferous forest and the mixed coniferous broad leaved forest had different sensitivity to different band. The maximum correlation of LAI with NDVI, RVI, MSR of Coniferous forest happened in the red band 10 (middle wavelength:709nm), the R2 was 0.7098; meanwhile, that of mixed coniferous broad leaved forest happened in the red band 8 (middle wavelength:697nm), the maximum R2 was 0.6385.(4) The correlation of LAI with NDVI, RVI, MSR from CHRIS image data with different angles has been studied, it revealed that 0°image of NDVI and LAI had the best correlation.0° image of NDVI of coniferous forest had the best correlation with LAI, the maximum R2 can reach 0.7098;-36°image took the second place, but the correlation in +36°image was not good. At the same time, the maximum R2 of 0°image of NDVI with LAI of mixed coniferous broad leaved forest was 0.6385, that of -36°,+36°image took the second place, respectively.更多还原