Backscattering of light by Karenia brevis and implications for optical detection and monitoring

Kevin Leo Mahoney

Abstract

A need exists for a rapid and efficient method of detecting and monitoring natural populations of Karenia brevis (K. brevis ). Optical instrumentation could meet this need; however, the optical properties of K. brevis must first be examined. Consequently, measurements of optical properties of K. brevis , grown under laboratory conditions, were made. Also, Mie Theory was evaluated in estimating these IOPs. With characteristics of K. brevis as input parameters and an inferred index of refraction (n = 1.053), Mie calculations provided a close estimate of absorption, scattering, and backscatter by K. brevis . The impact of natural Karenia brevis populations on bulk in situ IOPs was unknown. This experiment evaluated K. brevis ' influence on the IOPs of natural waters, and evaluated Mie Theory in approximating the IOPs of natural K. brevis assemblages. The experiment was conducted during two bloom events off the western coast of Florida. The Mie Model performed well in estimating the IOPs of the K. brevis blooms. Karenia brevis contributed substantially to bulk in situ IOPs. The presence of Karenia brevis coincided with low backscatter: total scatter and backscatter: total absorption ratios. Waters dominated by K. brevis differed optically from of waters with little or no K. brevis present. The extent of Karenia brevis ' contribution to measured remote sensing reflectance (Rrs ) was unknown. Based on K. brevis characteristics, Mie calculations approximated IOPs for two bloom events. The modeled IOPs served as input parameters to Hydrolight 4.1 resulting in modeled Rrs of K. brevis . The Mie-Hydrolight 4.1 simulations yielded comparable results to observed Rrs . The model's input parameters were then manipulated to reflect the dynamics of K. brevis blooms. When K. brevis was aggregated in surface waters, Rrs was sensitive to K. brevis cell concentration and CDOM (gelbstoff ) concentration. For a subsurface population of K. brevis , Rrs was sensitive to the depth of the subsurface layer but not cell concentration. Also, Rrs for K. brevis differed from the Rrs of waters without K. brevis .