Incorporating fuzzy sets in spatially explicit ecological modelling

将动物行为和种群动态与地理信息系统（GIS）中表示的空间数据联系起来的模拟模型正在成为保护生物学和景观生态学等领域的一个突出特征。这些模型在保护工作中很有价值，因为它们在对比管理方案中很有用。然而，由于景观数据、栖息地建模以及代表我们对动物分散行为理解的参数的不确定性，它们的有效性受到限制。使用基于代理的、面向对象的数据集成和综合建模框架，我们将探索空间景观数据的模糊表示及其在空间显式人口模型（sepm）中的后续使用之间的相互作用。将制定基于经验的方法的战略，以便根据档案现场数据得出生境的空间明确模糊表示。基于gis的栖息地表示的进展将被纳入一个建模框架，允许探索分散行为的理论模型，这些模型将纳入感知范围的各向异性性质以及景观矩阵对个体行为的影响。这扩展了最近的工作，将GIS数据库模块中的模糊表示与使用可能对象概念的动物运动行为模糊控制集成在一起。一般的模型构建目标是能够明确地表示行为权衡，或者不同的决策策略，分散者可以使用这些策略来导航环境。它将促进我们对动物在不同环境下如何分散行为的理解，从而有助于设计空间上明确的模拟，从而合理地代表景观作为物种相关的空间信息和基于个体的动物运动模型。这种系统的改进应能提高为保护工作提供的生态情报的质量。

Simulation models that connect animal behavior and population dynamics with spatial data represented in a geographic information system (GIS) are becoming a prominent feature in the fields such as conservation biology and landscape ecology. These models are valued in conservation efforts because of their utility in contrasting management scenarios. However, their validity is limited due to uncertainties concerning landscape data, habitat modeling, as well as parameters representing our understanding of the behavior of dispersing animals. Using an agent-based, object oriented modeling framework for data integration and synthesis, we will explore the interplay between the fuzzy representation of spatial landscape data and their subsequent use in spatially explicit population models (SEPMs). Strategies for an empirically based methodologies for deriving a spatially explicit fuzzy representation of habitat based on archival field data will be developed. The advance in GIS-based habitat representation will be incorporated into a modeling framework allowing exploration of theoretical models of dispersal behavior that will incorporate the anisotropic nature of perceptual range as well as landscape matrix effects on an individual's behavior. This extends recent work that integrates fuzzy representation in the GIS database module with the fuzzy control of animal movement behavior using the concept of Probeable objects. A general model building objective is to be able to explicitly represent the behavioral tradeoffs, or differing decision strategies, that dispersers may use to navigate the landscape. It will advance our understanding of how dispersing animals may behave under different circumstances, thus helping design spatially explicit simulations that plausibly represent both the landscape as species-relevant spatial information and the individual based model of animal movement. The improvements in such a system should advance the quality of ecological intelligence provided to conservation efforts.