Identifying and mapping night lights using imaging spectrometry

Remote mapping of night lights using the Defense Meteorological Satellite Program (DMSP) has been used for decades to inventory the global distribution of human activity. ^©± The coarse spatial and spectral resolution of DMSP, however, has precluded discrimination of lighting types or spectral characteristics. Recent demonstrations using photography from the International Space Station and airborne multispectral simulations demonstrate significant potential, but high-spectral-resolution field and laboratory measurements indicate that these methods do not take full advantage of the spectral information available. This research demonstrates the use of imaging spectrometer data to identify, characterize, and map urban lighting based on comparison to a lights spectral library. The library provides information about spectral emission lines unique to specific lighting types. ProSpecTIR-VS imaging spectrometer data of Las Vegas, Nevada were analyzed to extract spectral features and these were compared to the spectral library measurements on a pixel-by-pixel basis, resulting in a detailed spatial map showing different lighting types. The nature and distribution of lights can be used as a surrogate for characterization of urban settings, and measurement of urban development.