Bringing Uncertain Ecosystem Services into Forest Optimization

森林生态系统为人类福利提供了许多服务，但目前尚不清楚如何在森林优化中考虑生态系统服务（ES），特别是在ES不确定的情况下。支持多目标下森林决策的各种方法可能为改进森林优化提供宝贵的机会。可以用适当的指标很好地描述ES，这些指标已经部分可用。在考虑多准则的连续优化方法中，有一些很有前途的方法可以最小化不良结果（折衷规划）。这些方法可以很好地与鲁棒优化相结合。鲁棒方法通常基于预定义不确定性空间的非随机优化。承认生态系统在森林管理中的重要性的一种替代多准则决策方法是考虑森林优化中的经济价值系数。这些可以纳入现代组合方法，以支持关于未来最佳森林组成和管理的决策。基于经济价值对所有ES进行一致的考虑具有吸引力，但也有风险。如果某一种服务价值很高，它可能会取代其他在未来可能仍然重要的基本服务。本研究计划希望开发和分析一种优化方法，以测试将ES及其基于各种信息源的不确定性整合到森林优化中的结果。这种方法必须能够考虑多个标准来表征ES或用于优化过程的聚合经济价值系数。这种办法可以根据可比较的生态系统的不同信息来源，分析生态系统一体化对所产生的最佳森林组成和管理的影响。

Forest ecosystems provide numerous services for human welfare, but it is less clear how ecosystem services (ES) can be considered in forest optimization, particularly, if the ES are uncertain. Various methods of supporting forest decision-making under multiple objectives may provide valuable opportunities to improve forest optimization. ES may be described quite well with appropriate indicators, which are already partly available. Among the continuous 8optimization methods for considering multiple criteria, are promising approaches which minimize undesirable outcomes (compromise programming). These approaches can be combined well with robust optimization. Robust methods are often based on non-stochastic optimization using pre-defined uncertainty spaces. An alternative to multi-criteria decision methods for acknowledging the importance of ES in forest management would be to consider economic value coefficients in forest optimization. These could be integrated into modern portfolio approaches to support decisions about the optimal future forest composition and management. A consistent consideration of all ES based on their economic value has appeal, but also bears risks. If a certain ES is valued very highly, it could displace other essential ES which may still be important in the future. This research proposal would like to develop and analyse an optimization approach to test the consequences of integrating ES, and their uncertainties based on various information sources, into forest optimization. This approach must be capable of considering either multiple criteria to characterize ES or aggregated economic value coefficients for the optimization process. Such an approach would allow the influence of the integration of ES to be analysed, based on different information sources for comparable ES, on the resulting optimal forest composition and management.