Copulas as a link for modeling stand structure and development

良好的森林管理需要良好的管理工具和良好的森林调查数据。森林管理计划通常跨越60年或更长时间。规划涉及到对市场和资源流动的预测有很大的不确定性。传统上，森林管理者依赖于生长和产量模型，这些模型为森林中的每一种独特条件提供了单一的估计。与这些预测相关的误差经常被忽视。这里提出了一种新的统计方法，以利用新的激光技术来增强对当前森林状况的估计，并开发一种新型的单木生长和产量模型，使森林管理者能够直接评估不确定性，并将不确定性纳入其规划模型。该项目的目标是从整个林区的航空激光数据中提取单株树木信息，并将这些数据作为生长模型的输入，以预测森林发展，评估潜在气候变化情景下的物种组成变化，为管理加拿大的森林设计创新的营林方案，并评估森林管理规划方案的风险。这项工作将使森林管理人员能够更好地预测森林的变化，并规划保障措施，以确保加拿大森林今后的可持续利用。

Good forest management requires good management tools and good forest inventory data. Forest management plans often span 60 or more years. Planning involves accounting for much uncertainty in projections of markets and resource flows. Traditionally forest managers rely on growth and yield models that provide a single estimate for each unique condition within the forest. Errors associated with those predictions are often ignored. A novel statistical approach is proposed here to enhance estimation of current forest conditions using new laser technology and to develop a new type of individual tree growth and yield model that will enable forest managers to directly assess uncertainty and incorporate uncertainty into their planning models. The goal of this project is to extract individual tree information from airborne laser data across an entire forest area, and to use this data as inputs into the growth model to predict forest development, assess species composition change under potential climate change scenarios, design innovative silvicultural options for managing Canada’s forests, and to assess risk of forest management planning options. This work will enable forest managers to better anticipate changes in the forest and to plan safeguards to ensure sustainable use of Canada’s forest into the future.