Assessment of ASTER land cover and MODIS NDVI data at multiple scales for ecological characterization of an arid urban center

Study of the detailed structure and ecological functioning of urban and peri-urban systems is intensifying due to increasing concentration of the human population into urban centers. Much of this increase is expected to occur in semiarid to arid cites. Data from new high spatial, temporal, and spectral resolution satellite-based sensors promise to increase our understanding of global urban ecological and climatic processes and improve city and land planning capabilities. Two of these sensors, the Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER) and the Moderate Resolution Imaging Spectroradiometer (MODIS), provide an opportunity to compare urban spatial structure (distribution and configuration of discrete land cover/land use classes on the landscape) with contemporaneous measurements of surface biophysical composition at a variety of spatial and temporal scales. Such combined measurements are useful for modeling changes to urban climate, hydrology, and biogeochemical cycles caused by modification of the landscape. We compare gridded landscape metrics derived from expert system land cover classification of ASTER to corresponding MODIS NDVI data at scales of 250 m/pixel, 500 m/pixel, and 1 km/pixel in order to determine which of these scales is optimal for monitoring of urban biophysical processes and landscape structure change. Weak positive and negative correlations between NDVI and landscape structure were observed at all three spatial scales for the metrics Class Area, Mean Patch Size, Edge Density, and Interspersion/Juxtaposition Index.