Impact of reduced spectral resolution on cloud detection and altitude estimation

Atmospheric effects, especially the impact of clouds, have always played a major role in remote sensing applications. Advances in optic, sensor, micro-electronic, communication and processing technologies enable the development and deployment of versatile remote sensors. Nevertheless, deployed systems remain costly so tailoring them to specific missions continues to be a key element of design. Autonomous interpretation of the complex terrestrial environment becomes easier as spectral, spatial and temporal information is enhanced. By nature, cloud identification improves in value as larger areas are observed so spectral and spatial resolution trades with coverage area become important. An algorithm to autonomously identify cloud type from high spatial and moderate spectral resolution data was presented at the 1998 IEEE Aerospace Conference (IAC). The purpose of this paper is to evaluate the performance of the CLOUD Detection and Identification (CLOUDDI) algorithm in the presence of reduced spectral and spatial resolution. Results against both flight data and model predictions are considered.