Contributions of multiangular remote sensing imagery to mapping tree biomass in savannas

世界森林在气候调节方面的作用现已得到广泛承认，利用森林影响碳循环的机制也已建立。这种现象的规模使森林参数的估计和绘图成为全球/国家一级的优先事项。遥感是提供有关尺度所需资料的唯一有效手段。过去的研究已经显示了现有遥感技术在一系列生态系统类型上的潜力和局限性，从植被密集和复杂的潮湿热带森林到稀树草原的稀疏树木分布。在后者中，基于光谱特性的传统技术在区分树木和草的信号方面遇到了问题，因为两者在一年中的大部分时间都同时具有光合活性。本研究的目的是探索图像定向分辨率在提高热带稀树草原树木生物量制图精度方面的潜力。该方法利用稀树草原树木的空间排列所产生的突出的阴影模式来估计树木的密度和大小，并最终将这些信息与树木生物量联系起来。该项目为加拿大地理信息产业巩固其在国际上的领先地位提供了一个机会，并为我们在一个新的应用领域提供了出口专业知识的机会。

The role of the World's forests in climate regulation is now widely recognized and mechanisms are in place to use forests in influencing the carbon cycle. The scale of the phenomenon has made the estimation and mapping of forest parameters a global/national level priority. Remote sensing is the only effective means available to provide the needed information at the relevant scales. Past research has shown potentials and limitations of existing remote sensing techniques for a range of ecosystem types, from the densely vegetated and complex humid tropical forest to the sparse tree distribution of savannas. In the latter, traditional techniques based on spectral properties encounter problems differentiating the tree and grass signals, both being simultaneously photosynthetically active throughout most of the year. The objective of this research is to explore the potential of imagery directional resolution in improving tree biomass mapping accuracy in savannas. The proposed approach takes advantage of the prominent shadow patterns created by the spatial arrangement of trees in the savanna to estimate tree density and size, and ultimately link this information to tree biomass. This project represents an opportunity for the Canadian geomatics industry to consolidate its lead position at the international level and provides openings to export our expertise in a new field of application.