Abstract
An optical algorithm is developed for the retrieval of the downwelling diffuse attenuation coefficient K_{d} (490) in the Red Sea using a comprehensive hydrolight simulated data set ( N = 5000 ). We found a robust relationship between the K_{d} (490) and the ratio of remote sensing reflectance R_{\text {rs}} (443)/ R_{\text {rs}} (555), with an excellent determination coefficient ( R^{2} = 0.999 ) and a low root-mean-square error (RMSE = 0.00033). The performance of the developed algorithm is evaluated with in situ data collected in the Red Sea by comparing obatined model output with existing empirical (NASA, Morel et al. , Zhang and Fell, Tiwari and Shanmugam) and semianalytical (Lee et al. ) algorithms. On the used in situ data from the Red Sea, the new algorithm shows good retrievals of K_{d} (490) with a low bias, and a low RMSE compared to that of the existing algorithms. For satellite application, we applied our algorithm to selected MODIS-Aqua images acquired over the Red Sea, which captured spatial features of phytoplankton blooms and physical processes (e.g., cyclonic and anticyclonic circulations) in the Red Sea. The new algorithm has the potential to improve our understanding of water transparency and photosynthetic processes that rely on the availability of solar radiation.