Efficient Detection of Cloud Scenes by a Space-Orbiting Argus 1000 Micro-Spectrometer

The description, interpretation and imagery of clouds using remote sensing datasets collected by Earth-orbiting satellites have become a great debate spanning several decades. Presently, many models for cloud detection and classification have been reported in the modern literature. However, none of the existing models can efficiently detect clouds within the shortwave upwelling radiative wavelength flux (SWupRF) band. Therefore, in order to detect clouds more efficiently, a method known as radiance enhancement (RE) can be implemented. A satellite remote sensing database is one of the most essential parts of research for monitoring different atmospheric changes. This study proposes an innovative approach using RE and SWupRF to distinguish cloud and non-cloud scenes by using a space-orbiting Argus 1000 spectrometer utilizing the GENSPECT line-by-line radiative transfer simulation tool for space data retrieval and analysis. We apply this approach within the selected wavelength band of the Argus 1000 spectrometer in the range from 1100 nm to 1700 nm to calculate the integrated SWupRF synthetic spectral datasets. We used the collected Argus observations starting from 2009 to investigate radiative flux and its correlation with cloud and non-cloud scenes. Our results show that the RE and SWupRF model can identify most of the cloudy scenes except for some thin clouds that cannot be identified reasonably with high confidence due to complexity of the atmospheric system. Based on our analysis, we suggest that the relative correlation between SWupRF and RE within a small wavelength band can be a promising technique for the efficient detection of cloudy and non-cloudy scenes.