The objectives of this project include (1) to understand the presence of toxic cyanobacteria and cyanotoxins in the source water and finished water in Taiwan, (2) to evaluate the management strategy of international organization and other countries for the control of cyanotoxins in drinking water, and (3) to help the establishment of national management strategies for cyanotoxins in drinking water of Taiwan. The outcomes of this three-ear project include establishment of analytical techniques, monitoring of toxins in drinking water systems and analysis of the data, training of reservoir and water utilities personnel, and proposed national framework for the management of cyanobacteria and cyanotoxins in drinking water systems. In the project, 612 water samples were collected and analyzed in 6 major drinking water reservoirs and associated treatment plants. A solid-phase extraction (SPE) concentration technique followed by liquid chromatograph-mass spectrometer (LC-MS) was used for the quantification of six microcystin congeners (LR, RR, YR, LA, LW, and LF), anatoxin-a, nodularin, cylindrospermopsin, and BMAA. The concentrations of microcystins were between N.D. -1.2 g/L in surface water samples, were between N.D. -0.47 g/L in raw water samples, and were all < 0.1g/L for finished water samples. A correlation between major cyanobacteria and their metabolites was conducted for the samples collected from A, B, and T Reservoir. The data from Reservoirs A and B indicated that the concentration of microcystins and that of 2-MIB changed with weather, with higher concentration at warmer seasons. The data from Reservoir T suggested that chlorophyll-a concentration is proportional to microcystins concentration, Microcystis cell concentration, and β-cyclocitral concentration, indicating that chlorophyll-a may be a good indicator for the estimation of the algal metabolites in the reservoir. In addition, when the cell number > 105 cells/ml, the microcystins concentration would be very like to exceed 1 g/L. For the treatment efficiency in the waterworks, 32-65% in average of microcystins removal efficiency were observed for the waterworks with conventional treatment processes, 48-100% were for the conventional waterworks with floatation and slow sand filtration units, and around 53 to 96% were for the advanced waterworks. For all the 42 samples collected from distribution systems, all the concentrations of microcystins were smaller that 0.1 g/L. However, both cylindrospermopsin and 2-MIB need to be further investigated. Real time PCR and other molecular techniques suggested that new strains of Cylindrospermopsis were detected in T and J reservoirs. In addition, Microcystis cells were present in the distribution systems at low level. For the management of cyanobacteria and cyanotoxins in the drinking water systems, four cyanobacteria bloom episodes, including one domestic reservoir, two Australian reservoirs and one New Zealand watershed, were reported and analyzed. In addition, three workshops for the identification of toxic cyanobacteria in drinking water was held, with more than 240 people trained in the workshops. Finally, a national framework was suggested for the management of cyanobacteria and cyanotoxins in drinking water systems for different responsible agencies.

Cyanobacteria,Microcystin,Nodularin,Anatoxin-a,Cylindrospermopsin