Multitemporal principal component analysis of spectral and spatial features of the Venice Lagoon

This study focuses on a multitemporal spectral analysis of morphologic features of the Venice Lagoon. Principal Component Analysis, applied in a multidate merged rotation to standardized visible and infrared Landsat Thematic Mapper spectral bands, has allowed discrimination of differences in spatial distribution and spectral characteristics of the observed features. A comparative analysis has been carried out on the derived PC based on visual and statistical interpretation. Near- and medium-infrared bands are revealed to be suitable to return significant information in a very shallow water environment and under particular weather conditions. A map of part of the Northern Lagoon has been produced out of the first Principal Component of multitemporal TM4s.