Decision support models for forest management planning

本提案描述了一个研究计划，其长期目标是开发森林管理规划中必要的决策支持模型，利用它们来解决管理问题，并为研究生和从业人员提供专门的教育和培训。该研究计划的短期目标是探索两个问题领域。第一个问题领域需要制定一个随机优化模型，我建议使用模拟和优化的集成，越来越多地被称为“模拟优化”。我有兴趣在两个项目领域使用这个模型来提升我们的知识。首先，使用这种方法的算法含义值得探究。其次，我也有兴趣使用模型来深入了解问题本身的本质；例如，在不确定的环境中规划木材采伐。我提出的研究的新颖之处在于，尽管已经开发了随机规划模型，但没有一个是空间明确的；越来越需要空间明确的规划模型来规划生物多样性目标。在第二个问题领域，与对未来气候变化的估计有关的高度不确定性要求有计划的适应采用强有力的战略；也就是说，在各种可能的未来中，与备选方案相比表现良好的策略。在变化不确定的环境中使用遗传物质，类似于在不确定的市场中投资资产。出于这个原因，我的方法将是使用基于马科维茨投资组合优化模型的种子投资组合来选择种子来源的最佳投资组合，以部署在不确定的气候变化环境中。种子组合模型将构成我描述遗传改良股票部署边界的方法的基础。加拿大各地的规划者迫切需要解决这些问题和那些继续发展的问题。

This proposal describes a research program with long-term objectives of developing the necessary decision support models in forest management planning, using them to solve management problems, and providing specialized education and training to graduate students and practitioners.      The research program's short-term objectives are to explore two problem areas.  The first problem area requires the formulation of a stochastic optimization model, and I propose using the integration of simulation and optimization, increasingly referred to as "simulation optimization".   I am interested in using this model to advance our knowledge in two project areas.  First, the algorithmic implications of using this method deserve inquiry.  Second, I am also interested in using the model to gain insights into the nature of the problem per se; i.e., of planning timber harvests in an uncertain environment.  The novelty of my proposed research is that, although stochastic planning models have been developed, none have been spatially explicit; and spatially explicit planning models are increasingly required to plan for biodiversity objectives.      In the second problem area, the deep uncertainty associated with estimates of future climate change requires that planned adaptation employ robust strategies; i.e., strategies that perform well when compared with the alternatives across a wide range of plausible futures.  The deployment of genetic material in an environment of uncertain change is analogous to the investment in assets in an uncertain market. For this reason, my approach will be to use a seed portfolio based on Markowitz' portfolio optimization model in selecting an optimal portfolio of seed sources to deploy in an environment of uncertain climate change.  The seed portfolio model will form the basis of my approach to delineating the deployment boundaries of genetically improved stock. Solutions to these problems and those that continue to evolve are desperately needed by planners across Canada.