Close Window

Urban environments are characterized by high spectral andspatial heterogeneity and, as a consequence, most urban pixels in moderateresolutionimagery contain multiple land-cover materials. Despite these complexities,virtually all urban land cover can be generalized as a combination ofvegetation, impervious surfaces, and soil (V-I-S components), in addition towater. Previous work has demonstrated the potential of multiple endmemberspectral mixture analysis (MESMA) to model the subpixel abundance of V-I-Scomponents. Here, the authors test whether the technique is sufficiently robustto map V-I-S components for a diverse set of cities, selecting 10 urban centersin the state of Rondônia, Brazil, to represent a range of populations, developmenthistories, and economic activities. For each urban sample, a 20 km × 20km region centered over the built-up area was subset from Landsat EnhancedThematic Mapper Plus (ETM+) imagery. MESMA was applied to all subscenesusing the same spectral library, model constraints, and selection rules. Accuracy of the modeled V-I-S fractions was assessed using high-resolution imagesmosaicked from digital aerial videography. Modeled fractions and referencefractions were highly correlated, with R2 values exceeding 0.75 for all materialsin multiple cities across a region. Model complexity, or the number of endmembersrequired to accurately model each pixel, was correlated with thedegree of human impact on the landscape. Built-up areas, as delineated bymodel complexity, exhibited a strong fit to the well-established relationshipbetween the built-up area of a settlement and its population. Finally, this workdemonstrates that the V-I-S components as modeled by MESMA can captureboth inter- and intraurban variability, suggesting that these data products couldcontribute to comparative studies of urbanizing areas through time and acrossregions.

Close Window