Multiscale Context for Change in Alpine Tundra

点击翻译按钮获取中文摘要

Alpine tundra is threatened by climate change because the low-temperature climates where it exists may be reduced globally. This study will provide a context for the response of alpine tundra in western North America to climate change by investigating the community structure of alpine tundra at multiple spatial scales. The focus is on the spatial extent of processes relevant to alpine tundra diversity. This study examines the conjecture that the mix of plant species growing together (or secondarily species diversity) differs along fine spatial scale gradients in soil characteristics. Soil gradients are nested within broader geographic-scale gradients determined by topography such as slope shape and position and these environments are further nested within regional gradients in climate variables (radiation, temperature, and precipitation). For example, the amount of water from precipitation at a site is modified by how the terrain alters the redistribution of snow, and the snowmelt available to plants is affected by soil texture. New data on the abundance of species on alpine tundra sites will be collected and used to derive biodiversity (dependent) variables. Site conditions (independent variables) of soil and surficial conditions will be examined at fine scale; topographic measures, which are more general conditions given that they are at larger extents than the plants, will be examined at a middle scale, and climate variables from established weather stations will be examined at coarse scale based on extrapolations developed for mountain topography. In order to explore possible relationships, three analytical methods will be applied for species composition: Mantel regression with pairwise dissimilarities of vegetation and environmental difference; multiscale ordination using canonical correspondence analysis; and multilevel regression with community composition dependent variables statistically derived by additional ordinations. For community diversity, multilevel regression will be used with species richness as the dependent variable. This research will advance our understanding of biodiversity and community biogeography by testing hierarchical relations across spatial scales. Multilevel models will be developed to address the kind of hierarchical controls that affect plant diversity and community structure across greater range of scales than done in prior ecological research. This research will provide new insight into potential ecological responses to climate change, which is a large-scale phenomenon, because it will take into account the spatial heterogeneity that exists at the finest scale. The results will inform National Park Service resource managers about possible future change in alpine tundra biodiversity, an important visual and aesthetic resource. More broadly, this study will improve interpretation and contribute to the findings of a global alpine monitoring network.