论文部分内容阅读
植被的结构参数如植被高度、生物量、水平和垂直分布等,是影响陆地与大气能量交换乃至生物圈多样性的重要因素。多数遥感系统虽然可以提供植被水平结构的图像,但是不能提供植被成分垂直分布的信息。大尺度激光雷达仪器如LVIS产生的激光雷达信号,已成功地用于估计树高和森林生物量,然而大多数激光雷达仪器不具备图像能力,只能提供一个区域内的采样数据。其他的遥感数据如多角度高光谱、多频率多时相辐射计或雷达数据,可根据GLAS(Geoscience Laser Altimeter System)采样的测量用来推断出连续的森林结构区域覆盖参数。MISR(Multi-angle Imaging Spectrometer)对陆表多角度的成像能力,可以通过BRDF的各向异性提供植被的结构信息。结合激光雷达的垂直采样和MISR的图像,区域内乃至全球性的森林空间参数的成像是可能的。ICESat卫星上的GLAS数据、Terra卫星上的MISR数据为区域或全球性森林结构参数提供了可能。本文的研究目的是评估GLAS数据,分析类似于MISR的数据对森林结构参数的估计能力。本文中使用了LVIS、AirMISR和GLAS数据。通过对GLAS树高的测量与GLAS像元内来自LVIS的平均树高对比,发现它们是高度相关的。同时还探讨了多角度频谱成像仪数据预测树高信息的能力,这将在今后区域内森林结构参数映射加以研究。
Vegetation structural parameters such as vegetation height, biomass, horizontal and vertical distribution are the important factors affecting the exchange of land and atmosphere energy and even the diversity of the biosphere. Although most remote sensing systems can provide images of vegetation horizontal structure, they do not provide information about the vertical distribution of vegetation components. Lidar signals produced by large-scale lidar instruments such as LVIS have been successfully used to estimate tree height and forest biomass, yet most lidar instruments do not have the image capability to provide only sample data within a region. Other remote sensing data such as multi-angle hyperspectral, multi-frequency multiphase radiometer or radar data can be used to infer continuous forest structure area coverage parameters from measurements taken with the GLAS (Geoscience Laser Altimeter System). MISR (Multi-angle Imaging Spectrometer) multi-angle imaging spectrometer on the multi-angle imaging capability of the ground, through the BRDF anisotropy to provide vegetation structure information. Combining vertical sampling with LIDAR and MISR images, it is possible to image in-region and globally global forest spatial parameters. GLAS data on ICESat satellites and MISR data on Terra satellites offer the possibility of regional or global forest structure parameters. The purpose of this paper is to evaluate GLAS data and to analyze the ability of MISR-like data to estimate forest structural parameters. LVIS, AirMISR, and GLAS data are used in this article. By comparing the GLAS tree height measurements with the average tree height from the LVIS within the GLAS pixels, they were found to be highly correlated. The ability of multi-angle spectral imager data to predict tree height information is also explored, which will study the mapping of forest structural parameters in the future.