Phosphorus Sequestration for Control of Cyanobacterial Growth in Drinking Water Reservoirs

Abstract

Engineered drinking water reservoirs are designed to facilitate particle settling for reduction of turbidity prior to conveyance to a drinking water treatment plant (DWTP). Fine cohesive sediment particles can carry significant loads of adsorbed phosphorus (P) that can desorb into the water column and intensify the growth of cyanobacteria (CB), causing problematic and potentially toxic CB blooms. In light of these reservoir sediment dynamics, strategies for mitigating rapid CB proliferation through sequestration of P were investigated. A series of bench scale experiments were conducted to examine the impact of managing dissolved and sediment-associated P for controlling CB growth. The first phase of testing involved batch experiments with fine reservoir sediments to determine their P release characteristics and the amount of dissolved P potentially available for CB uptake. The utility of sequestering this soluble reactive P (SRP) with a common metal salt coagulant, ferric chloride (FeCl3), was also investigated. These adsorption / desorption experiments showed that a dose as low as 25 mg/L was effective in precluding SRP desorption from the sediment over a relatively wide range of solution SRP concentrations. These results were critical to provide an understanding of the SRP-sediment dynamics after treatment with FeCl3. The second phase of testing involved confirmation of the importance of sediment-associated SRP on the growth of a commonly found CB, Microcystis aeruginosa and evaluation of the utility of FeCl3 coagulation for limiting M. aeruginosa growth through sequestration of SRP. Standard methods for culturing / growing M. aeruginosa were adapted for a series of experiments, at near bloom cell counts, in the presence and absence of sediment to demonstrate the potential utility of SRP sequestration with a common coagulant used during drinking water treatment to inhibit CB growth. While the lab-scale experiments could not, and were not expected to exactly mimic reservoir behavior, they were conducted to demonstrate proof-of-concept. They were successful in doing so because M. aeruginosa growth was inhibited with adequate FeCl3 application. Significantly lower FeCl3 doses were effective when the high levels of sediment (analogous to previously deposited sediment) were removed from the system. The results of this study have several implications for controlling the proliferation of CB through nutrient sequestration. SRP can be sequestered very effectively at doses of FeCl3 typical of DWTP operations. Growth of M. aeruginosa can even be inhibited by sequestering P when CB cell counts are elevated to levels consistent with those that may be expected at bloom conditions; as would be expected, relatively higher FeCl3 doses are then required. Further experimental work to determine the optimal dose of FeCl3 at different sediment loads and lower M. aeruginosa starting cell populations should be considered.