Introduction / Introduction

Abstract

The inherently complex and dynamic nature of urban land surfaces has recently become the focus of remote sensing research worldwide owing to the growing awareness of the implications of global urbanization. Rapid outward growth of cities not only results in land conversion from natural to man-made surfaces and alterations in the spatial structure of land cover and land use but also influences biophysical processes, local climate, resources consumption, and the environment inside and surrounding these urbanized areas. Improved understanding of these influences of urbanization and the subsequent development of sustainable growth strategies require accurate and in-depth information, which can be derived from remote sensing observation on the current state of urbanization, its historic trends, and the links between land use aspects of cities and the human dimension. We are pleased to present this special issue as a showcase that exhibits some recent developments in urban remote sensing research, both in methodology and applications. Four papers in this issue are focused on promising new methodologies for improving retrieval of land surface information from recent high resolution optical and synthetic aperture radar (SAR) data. The performance of object-based classification of urban land cover based solely on spectral information has not been highly satisfactory because of the spectral similarity among land cover classes and high spectral variability within individual classes. Xu and Li demonstrate that urban land cover classification errors can be significantly reduced by using shape features extracted from segmented objects using invariant moments. In an exploration of the urban surface response in SAR data, Li et al. illustrate that polarimetric attributes can be used within a maximum likelihood classification framework to identify vertical structure components of urban environments. Molch et al. assessed the performance of urban built-up delineation with RADARSAT-1 and ENVISAT images based on a texture-based, anisotropic, rotation-invariant built-up presence index, which provides improved built-up area classification accuracy. Finally, L. Zhang et al. assessed the potential for high resolution airborne E-SAR polarimetric SAR interferometric (PolSAR) data for use in urban land cover extraction and classification. The remaining six papers in this issue are focused on the application of remotely sensed measurements for a broad range of urban issues. In a case study of Boulder, Colorado, Golubiewski and Wessman quantified the fractions of major vegetated and anthropogenic components in urbanized landscapes using an unmixing strategy applied to airborne hyperspectral AVIRIS data. The identification of subclass details Numéro spécial