Ship-of-opportunity based phycocyanin fluorescence monitoring of the filamentous cyanobacteria bloom dynamics in the Baltic Sea

Distribution of cyanobacteria cannot be evaluated using chlorophyll a (Chla) in vivo fluorescence, as most of their Chla is located in nonfluorescing photosystem I. Phycobilin fluorescence, in turn, is noted as a useful tool in the detection of cyanobacterial blooms. We applied phycocyanin (PC) fluorometer in the monitoring of the filamentous cyanobacterial bloom in the Baltic Sea. For the bloom forming filamentous cyanobacteria Aphanizomenon flos-aquae and Nodularia spumigena, PC fluorescence maximum was identified using the excitationeemission fluorescence matrix. Consequently, the optical setup of our instrument was noted to be appropriate for the detection of PC, and with minor or no interference from Chla and phycoerythrin fluorescence, respectively. During summer 2005, the instrument was installed on a ferryboat commuting between Helsinki (Finland) and Travemu¨nde (Germany), and data were collected during 32 transects providing altogether 200 000 fluorescence records. PC in vivo fluorescence was compared with Chla in vivo fluorescence and turbidity measured simultaneously, and with Chla concentration and biomass of the bloom forming filamentous cyanobacteria determined from discrete water samples. PC fluorescence showed a linear relation to the biomass of the bloom forming filamentous cyanobacteria, and the other sources of PC fluorescence are considered minor in the open Baltic Sea. Estimated by PC fluorescence, cyanobacterial bloom initiated late June at the Northern Baltic Proper, rapidly extended to the central Baltic Proper and the Gulf of Finland, and peaked in the mid-July with values up to 10 mg l 1 (fresh weight). In late July, bloom vanished in most areas. During single transects, or for the whole summer, the variability in Chla concentrations was explained more by PC fluorescence than by Chla fluorescence. Thus, filamentous cyanobacteria dominated the overall variability in phytoplankton biomass. Consequently, we show that during the cyanobacterial blooms, the estimation of Chla concentration using only Chla in vivo fluorescence is not applicable, but PC in vivo fluorescence is required as a predictor as well.