Uncovering the Spatial Pattern of Feedback Effects at Alpine Treeline

该研究将评估积极反馈在空间模式上的结合方式，以改变冰川国家公园高山林木线现有树木和krummholz植物附近的环境。该项目将开发一个细胞自动机和一个随机版本，使用遗传编程来揭示反馈过程的空间模式。将使用遗传编程，以便根据模拟产生的模式与多光谱遥感图像中观察到的模式相似的程度改变规则。模拟和遥感模式的分辨率都将是一米。不同的非生物（预反馈）场地质量模式、植被的静态模式、非生物场地质量的变化率和不同的尺度关系将通过创建的规则进行检查。这项研究在更精细的尺度上检查空间细节，将允许探索响应的非线性和作为复杂适应系统的过渡带地区组织的潜力。将过渡带的生物地理现象与复杂性理论联系起来，可以提高我们对生物群落边界组织的认识，并对复杂性理论作出地理贡献。特别是，这项研究将增加我们对地理现象如何与其他空间系统共享特征的理解。这项研究还将通过在遗传算法中结合适合度的空间测量来推进地理建模领域。将空间模式作为遗传规划的关键参数，分析遗传规划生成规则的空间依赖关系。最后，这项研究将促进对一个重要的生物地理学概念——过渡带的机制的理解。由于正反馈引起的潜在非线性，这将与过渡带作为气候变化影响指标的想法密切相关。

The research will assess the way in which positive feedbacks combine in spatial patterns to alter the environment in the neighborhood of existing tree and krummholz plants at alpine treeline in Glacier National Park. The project will develop a cellular automaton, and a stochastic version, that uses genetic programming to uncover the spatial patterns of feedback processes. Genetic programming will be used so that the rules change in response to how well the simulations produce patterns similar to patterns observed in multispectral remote sensing imagery. Both the simulations and the remotely-sensed patterns will be at one-meter resolution. Different patterns of abiotic (pre-feedback) site quality, static patterns of vegetation, changing rates of abiotic site quality, and different scale relations will be examined through the rules created.This research to examine spatial detail at a finer scale will allow exploration of nonlinearities in responses and the potential for the organization of ecotone areas as complex adaptive systems. Linking the biogeographical phenomena of ecotones to the body of work on complexity theory can improve our understanding of the organization of biome boundaries and make a geographical contribution to the theory. In particular, the research will increase our understanding of how a geographic phenomenon may share traits with other spatial systems. This research will also advance the area of geographical modeling by incorporating spatial measures of fitness in genetic algorithms. Spatial pattern will be used as the key parameter in genetic programming, and the spatial dependencies of the rules produced by this programming will be analyzed. Lastly, this research will advance understanding of the mechanisms underlying an important biogeographical concept, the ecotone. It will be germane to the idea of ecotones as indicators of the impacts of climatic change because of the potential nonlinearities induced by positive feedbacks.