I/UCRC FRP: Collaborative Research: Understanding and Modeling Competition Effects on Tree Growth and Stand Development Across Varying Forest Types and Management Intensities

植物群落的竞争是一个基本的生态过程，已经通过实验和模拟进行了研究。由于树木种群相互作用的时间很长，并且随着树木生长和死亡率的长期测量，广泛的数据库的可用性越来越高，该项目旨在使用先进的建模方法来开发增强的竞争定量表达。 需要改进森林种群竞争模型，以进一步提高我们在一系列森林类型中的管理选择的预测能力。了解竞争动态对于评估管理活动至关重要，例如部署遗传物质，规定整地处理和种植密度，预测对植被控制和间伐造成的不同程度的种间和种内竞争的反应，以及通过施肥改善养分不足。增加对竞争过程的了解和制定量化竞争影响的改进模型，将加强对各种森林管理备选办法的评价。这些结果将产生深远的影响，不仅对森林管理决策，而且对森林养护和恢复，以及对了解和模拟环境影响，包括气候变化对森林的影响。其他更广泛的影响包括丰富的教育经验，研究生和博士后研究员，由于行业科学家以及大学教师参与这项研究的进行。本研究项目的总体目标是利用现代建模和统计技术以及计算技术来利用林业实地研究中的信息，努力开发改进的树木和林分水平竞争的定量措施。 建模是一种强大的工具，可以整合和综合现有的理论和经验证据，识别知识差距，并提出有关潜在植物竞争的相关假设，这些假设可以通过实验进行测试。将资源可得性和竞争强度与生长和生存联系起来，是理解和预测林分动态的核心。一个关键的挑战是确定物种的影响，遗传变异，微网站的异质性，并通过时间对树的大小变化的本地邻里的影响。这些影响是混杂的，高度相互作用，特别是难以分开的树木，在很长的时间内相互作用。此外，树对树的竞争不能直接测量，而是基于总体林分密度、相对树大小和/或点密度测量（包括相邻树的大小和距离）的测量来推断。先进的统计分析和建模技术将应用于数据库，以处理长期森林生态系统动态的复杂性。将查明知识方面的差距，并将推进测试竞争过程相关假设的实验方法。

Competition in plant communities is a fundamental ecological process that has been studied through experimentation and by modeling. Given the long time periods over which tree populations interact and the increasing availability of extensive data bases with long-term measurements of tree growth and mortality, this project aims to use advanced modeling methodology to develop enhanced quantitative expressions of competition. Improved models of competition in forest populations are required to further advance our predictive ability for management options in a range of forest types. Understanding competition dynamics is paramount for evaluating management activities such as deploying genetic material, prescribing site preparation treatments and planting density, predicting response to varying levels of inter-and intra-specific competition resulting from vegetation control and thinning, and ameliorating nutrient deficiencies through fertilizer applications. Increased understanding of competition processes and formulation of improved models for quantifying competition effects will enhance evaluation of a wide array of forest management options. These results will have far-reaching implications, not only for forest management decision-making but also for forest conservation and restoration, and for understanding and modeling environmental influences, including climate change, on forests. Additional broader impacts include an enriched educational experience for graduate students and postdoctoral fellows due to the involvement of industry scientists as well as university faculty in the conduct of this research. The overall goal of this research project is to use contemporary modeling and statistical techniques and computing technology to exploit information in forestry field studies in an effort to develop improved quantitative measures of tree- and stand-level competition. Modeling is a powerful tool for integrating and synthesizing existing theory and empirical evidence, identifying knowledge gaps, and suggesting relevant hypotheses regarding underlying plant competition that might be tested through experimentation. Relating resource availability and competition intensity to growth and survival is central to understanding and projecting forest stand dynamics. A key challenge is determining effects of species, genetic variation, microsite heterogeneity, and local neighborhood effects on tree-size variation through time. These effects are confounded, highly interactive, and especially difficult to separate for stands of trees, which interact over long time periods. Furthermore, tree-to-tree competition cannot be measured directly, but rather is inferred based on measurements of overall stand density, relative tree size, and/or point density measures that include sizes of and distances to neighbors. Advanced statistical analyses and modeling techniques will be applied to data bases to deal with the complexity of forest ecosystem dynamics over extended time periods. Gaps in knowledge will be identified and experimental approaches for testing relevant hypotheses regarding competition process will be advanced.