Satellite LiDAR enhancement of Forest Inventory and Production Forecast Capabilities

卫星激光雷达增强森林清查和产量预测能力

• 批准号: NE/F021437/1
• 负责人: Peter North
• 金额: $ 30.51万
• 依托单位: Swansea University
• 依托单位国家: 英国
• 项目类别: Research Grant
• 财政年份: 2009
• 资助国家: 英国
• 起止时间: 2009 至 无数据
• 项目状态: 已结题
• 来源: https://gtr.ukri.org/projects?ref=NE%2FF021437%2F1
• 关键词: Satellite; LiDAR; enhancement; Forest; Inventory

Knowledge of carbon distribution stored within vegetation is an important factor necessary for reporting changes in carbon stock and recognised in international agreements such as the Kyoto Protocol to the United Nations Framework Convention on Climate Change 1997. Quantifying tree volume is also of great importance for forest management to monitor stand performance and assess commercial potential. Additionally, the vertical structure of vegetation can provide useful information regarding the quality of woodlands as species habitats. Satellite-derived optical data can provide a two-dimensional perspective of land class distribution and therefore permits the delineation of forests and stands according to the reflective properties for large areas. However, estimates to quantify vegetation from optical data rely on indirect assumptions based on its reflectivity at different wavelengths. Light Detection and Ranging (LiDAR) provides a direct means of estimating vegetation height, vertical profile, volume and canopy cover using the structural properties of the vegetation itself. Full waveform LiDAR uses the ability for laser pulses, emitted from terrestrial, airborne or satellite platforms, to penetrate gaps between vegetation foliage. Energy is reflected and returned to the sensor from all intercepted surfaces within the illuminated area (footprint), meaning that the returned waveform represents both the canopy structure and surface topography. The time taken for the returned energy to be detected at the sensor can be converted into distance using the speed of light which allows elevation differences between the intercepted surfaces to be calculated, therefore providing a vertical canopy profile. The Geoscience Laser Altimeter System (GLAS) aboard the Ice, Cloud and land Elevation Satellite (ICESat) provides near global coverage three times annually, sampling the Earth's surface using approximately 64 metre diameter LiDAR footprints. This innovative satellite therefore offers an unprecedented opportunity for seasonal biophysical parameter retrieval at regional to global scales. As part of the the National Forest Inventory, land cover maps identifying forested areas will be produced by the Forestry Commission during the course of this research. These will be used to define areas of Interpretive Forest Types (i.e. broad vegetation classes including conifers, broadleaves, newly planted stands, etc.) which will be further segmented into species classes using optical remotely sensed data. Returned LiDAR waveforms will then be used for sample plots of these classes to establish relationships with important forest parameters used in vegetation analysis. These will include direct relationships with vegetation height profile, canopy cover and stemwood volume as well as the indirect estimation of parameters used in forestry applications such as mean diameter distribution and basal area. These models will then be used to extend the estimates across classified areas leading to the production of a nation-wide cartographic product. This is be of particular interest for forest inventory purposes as, whilst comprehensive information is available relating to public woodland, this is not the case for privately owned land. As private land accounts for 60% of Britain's woodlands and 80% of timber production, methods of validating estimates of vegetation parameters for these areas would serve to reduce uncertainty in carbon accounting. Furthermore, the three/dimensional perspective of this product would allow habitat structure to be related to land cover type and species distribution for a more comprehensive analysis of habitat properties and fragmentation. Vertical vegetation structure can also be used to better understand ecosystem fluxes and the effects of humans and the environment on these.

了解植被中碳的分布情况是报告碳储量变化的一个重要因素，并在1997年《联合国气候变化框架公约京都议定书》等国际协定中得到认可。量化树木蓄积量对于森林管理监测林分绩效和评估商业潜力也非常重要。此外，植被的垂直结构可以提供有关作为物种生境的林地质量的有用信息。从卫星获得的光学数据可以提供土地类别分布的二维透视图，因此可以根据大面积的反射特性对森林和林分进行划界。然而，从光学数据中量化植被的估计依赖于基于其在不同波长下的反射率的间接假设。光探测和测距（LiDAR）提供了一种直接的手段，利用植被本身的结构特性来估计植被高度、垂直剖面、体积和树冠覆盖。全波形激光雷达利用从地面、空中或卫星平台发射的激光脉冲穿透植被之间的间隙的能力。能量被反射并从被照射区域（覆盖区）内的所有被拦截表面返回到传感器，这意味着返回的波形代表了冠层结构和表面地形。在传感器处检测到返回能量所花费的时间可以使用光速转换成距离，这允许计算被拦截表面之间的高度差，因此提供垂直的树冠轮廓。冰、云和陆地高程卫星（ICESat）上的地球科学激光高度计系统（GLAS）每年提供三次接近全球的覆盖，使用直径约64米的激光雷达足迹对地球表面进行采样。因此，这颗创新卫星为区域乃至全球范围的季节性生物物理参数反演提供了前所未有的机会。作为国家森林资源清查的一部分，林业委员会将在研究过程中制作确定森林地区的土地覆盖图。这些将用于定义解释性森林类型的区域（即广泛的植被类别，包括针叶树、阔叶树、新植林等）。利用光学遥感数据将其进一步划分为物种类别。返回的激光雷达波形将用于这些类别的样地，以建立与植被分析中使用的重要森林参数的关系。这将包括与植被高度剖面、树冠覆盖和树干体积的直接关系以及对林业应用中使用的参数的间接估计，例如平均直径分布和断面积。然后将利用这些模型将估计数扩大到所有分类地区，从而制作全国范围的制图产品。这对森林清查特别有意义，因为虽然可以获得有关公共林地的全面信息，但私人拥有的土地却不是这种情况。由于私人土地占英国林地的60%和木材产量的80%，验证这些地区植被参数估计值的方法将有助于减少碳核算的不确定性。此外，该产品的三维视角将使生境结构与土地覆盖类型和物种分布相关，以便更全面地分析生境特性和破碎化。垂直植被结构也可用于更好地了解生态系统通量以及人类和环境对这些通量的影响。

项目成果

Evaluating Prospects for Improved Forest Parameter Retrieval From Satellite LiDAR Using a Physically-Based Radiative Transfer Model

使用基于物理的辐射传输模型评估改进卫星激光雷达森林参数检索的前景

• DOI: 10.1109/jstars.2013.2244199
• 发表时间: 2013
• 期刊: IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
• 影响因子: 5.5
• 作者: Rosette J

Slope Estimation from ICESat/GLAS

• DOI: 10.3390/rs61010051
• 发表时间: 2014
• 期刊: Remote. Sens.
• 作者: C. Mahoney;N. Kljun;S. Los;L. Chasmer;J. Hacker;C. Hopkinson;P. North;J. Rosette;E. Gorsel

Forestry Applications for Satellite Lidar Remote Sensing

• DOI: 10.14358/pers.77.3.271
• 发表时间: 2011
• 期刊: Photogrammetric Engineering and Remote Sensing
• 影响因子: 1.3
• 作者: J. Rosette;J. Suárez;P. North;S. Los