Potential Of Landsat-Oli For Seagrass And Algae Species Detection And Discrimination In Bahrain National Water Using Spectral Reflectance

Seagrass (Halodule uninervis and Halophila stipulacea) and algae (green and brown) species are widely distributed along the coastal zones of the Bahrain national water. In this study the potential of Landsat-OLI VNIR spectral bands was investigated for distinction and discrimination among these species using spectral reflectances. The measured spectra's of each species considering different coverage rate (0, 10, 30, 75 and 100%) were transformed using continuum-removed (CR) approach, resampled and convolved in the solar-reflective spectral bands of OLI using a radiative transfer code, then converted to water vegetation indices (WVI). Regression analysis were performed between the transformed WVI and the coverage rates of each species individually (seagrass and algae) and mixed; as well between WVI and NIR reflectances. Spectral and CR analyses showed that the blue and the green bands perform better than the coastal and the red bands for seagrass and algae classes' discrimination. This result was further corroborated by the WVI. Regression results between the coverage rates and WVI calculated with green and NIR bands showed that the TDAVI and WAVI discriminate significantly among the mixed species (R-2 of 0.70), and between individual species (R-2 of 0.80 for algae and for seagrass). Accomplished between WVI and NIR reflectances, regression correlations were more significant when all mixed samples (R-2 of 0.95) have been considered, likewise when we consider individually the two seagrass (R-2 of 0.95) and the two algae species (R-2 of 0.82).