Institute of Ecosystem Studies

2008 BES Annual Meeting Presentation and Poster Abstracts

Can we observe urban landscape fragmentation globally?
Ellis, Erle
Co-Authors: Erle C. Ellis, Scott Sener1, David Potere2 and Mutlu Ozdogan3 1Department of Geography & Environmental Systems, University of Maryland, Baltimore County, Baltimore, MD 21250 2 Office of Population Research, Princeton University, 207 Wallace
Abstract: Urban landscapes are generally believed to be the most heterogeneous on Earth due to their fine-scale fragmentation into built and non-built land covers. However, most studies of urban landscape fragmentation are either city-scale case studies of limited geographic scope, or are conducted at relatively coarse resolutions that preclude observations on the fine-scale heterogeneity that is characteristic of urban environments. Here we present preliminary results of a global study of fine- scale landscape fragmentation based on a global stratified random sample of terrestrial landscapes. Landscape samples were selected using a two-stage design based on an equal-area discrete global grid system stratified by population and cultivated area. Sample sites were then mapped to eight land cover classes through interpretation of high-resolution Google Earth imagery. Results demonstrate that urban areas may not be the most fragmented of Earth's terrestrial ecosystems; sample sites with intermediate population densities also show a high degree of landscape fragmentation. Multi-stage stratified global landscape sampling designs based on discrete global grids offer new opportunities to make global observations on the ecology of urban and other highly heterogeneous ecosystems, and to make associated inferences about the global causes and consequences of long-term ecological change in anthropogenic biomes.