OPTICAL DETECTION OF CYANOBACTERIAL BLOOMS Shipboard observation and remote sensing

Planktonic cyanobacteria exhibit buoyancy regulation and morphologic adaptations by which population losses due to sinking and grazing are minimised. This may lead to rapid bloom formation, given favourable light and nutrient availability. Blooms will initiate at sites best offering these conditions – ‘hot spots’ or ‘fronts’ of productivity – and spread to other areas depending on the basin’s morphometry and hydrology (Oliver and Ganf, 2000). By their buoyancy control several species are able to dwell in particular depth intervals. Colonies of Microcystis species exhibit vertical migration, and are known to form surface scums, which can be wind-swept to other lake parts and accumulate on the shore. Due to these factors, temporal and spatial distribution of cyanobacterial biomass is highly complex. Because monitoring of the spatial distribution of plankton is costly and therefore sparse, spatio-temporal patterns of bloom formation in large inland waters are poorly known. Availability of rapid and cost-effective methods to acquire the distribution of biomass and ecophysiological characteristics is important from scientific and water management points of view. Worldwide, there are more than 2000 lakes with a surface area exceeding 10 km (Wetzel, 1990) for which assessments of the spatial distribution of plankton biomass could be greatly improved by remote sensing. Advances in hydrologic optics (Kirk, 1994), and the advent of satellite-based digital imaging spectrometers with a less than 1-km spatial resolution and suitable spectral band width now enable mapping of biomass distribution in many of these inland waters. The gap between measurements in situ and satellite imagery can be bridged by shipboard and airborne spectrometry (Kirk, 1994; Bukata et al., 1995). Moreover, the data from the different platforms are complementary (Vos et al., 2003). The methods used by these platforms are here collectively referred to as optical teledetection, which can be broadly defined as macroscopic observation of a