Studies on Brackish and Freshwater Cyanobacteria

Tidal-influenced stormwater detention ponds sampled by the SCHABP range in salinity from low brackish to marine, and not surprisingly, the type of HAB is associated with salinity properties.  In general, dinoflagellate and raphidophyte blooms are relatively prevalent under mid-brackish to marine conditions (Kempton et al. 2002, Lewitus et al. 2003, 2004), while cyanobacteria are commonly the dominant phytoplankton species in low brackish waters.  For example, a compilation of samples from 2002-2003 demonstrates the occurrence of several different potentially toxic cyanobacterial species spanning all salinities, but 85% of these occurred at salinity < 20 ppt and 66% below 10 ppt (Fig. 1).  It is important to note that these are not freshwater ponds – they experience tidal exchange, the salinity dictated by the distance from tidal creeks and the hydrological properties of surface and groundwater flow.  Assuming that the salinity distribution in Fig.5 reflects habitat preference, another important point is that salinity optima vary between cyanobacterial species.  Among the four most prevalent species, Oscillatoria occurred most often in low salinities (67% occurrence at 1 to 7.5 ppt), Microcystis occurred in salinities ranging from 2 to 10 ppt (73%), and Anabaena ranged from 5 to 20 (69%), and Planktothrix’s distributional pattern was more euryhaline. 

Figure 1.  Occurrence of potentially toxic cyanobacterial species at various salinity ranges in samples collected from South Carolina lagoonal ponds in 2002-2003.

Microcystins are cyanobacterial hepatotoxins that have caused animal poisonings and human health problems, usually through drinking water and recreational activities.  During the summer of 2000, a Microcystis spp. bloom occurred in Lake Edmunds, a freshwater-to-low salinity pond in Charleston, SC.  In the summers of 2001 and 2002, monitoring studies on cyanobacteria and microcystin were conducted in this pond, and in 2002, regulation of growth and toxin production by M. aeruginosa cultures were studied.  In both years, microcystin was consistently detected in water samples. In 2003, we added other freshwater sites, a pond in Charles Towne Landing, a public park, and Goose Creek reservoir, formerly a drinking water source and now a site of recreational activity (Fig. 2).  The primary bloom producing species found in all bodies of water were Microcystis aeruginosa, Anabaena spp., and Aphanizomenon flos-aquae. Goose Creek Reservoir Power Plant was the only site consistently below 1 ppb, the threshold of safe levels indicated by World Health Organization guidelines. Highest whole water toxin content, greater than 10 ppb, occurred during June and July for all other sites. On July 2 and 14, 2003, the toxin content of Lake Edmunds Shaffer Road was 1926 ppb (whole water) and 316 ppb (filtrate), respectively. 

During the Lake Edmunds survey, an inverse relationship was observed between pond phosphate concentration and microcystin levels.  In laboratory experiments, toxin production was higher in exponential growth phase than stationary phase, independent of the type of nutrient limiting growth (N or P).  We conclude that the potential for microcystin production is greatest in actively dividing cells, and that the inverse relationship between microcystin and P in pond samples reflects post-bloom periods of nutrient depletion after cellular toxin accumulation. Predictably, increased nutrient loading would lead to greater ambient concentrations of microcystin not only by supporting higher Microcystis biomass, but by prolonging the period of active cellular division in bloom populations. 

Figure 2.  Microcystin concentrations in filtrate (top) and whole water (bottom) from 5 SC freshwater sites monitored from May to September 2003. 

The low to mid-brackish ponds are almost always phosphate-rich and support high biomass cyanobacterial blooms.  Such blooms can contain mixed cyanobacterial populations and exhibit successions of dominant types, or remain dominated by a single species over time (Fig. 3).

 

High biomass blooms can be sustained for long periods and are common visual and odiferous nuisances to residents and tourists.  The public is less aware of the toxic potential of these blooms.  Nevertheless, our studies indicate that high microcystin concentrations are commonly associated with these blooms.  For example, toxin levels analyzed by ELISA were always > 1 ppb, and ranged to > 10,000 ppb from September through November 2004 in a Kiawah Island brackish pond (Fig. 4).  The Microcystis bloom lasted five months. For comparison, the World Health Organization threshold for safe drinking water is 1 ppb.  

Figure 4.  Temperature and salinity (top) and Microcystis abundance and microcystin concentrations (bottom) in a South Carolina lagoonal pond from September 2004 to mid-January 2005.  Note that high biomass (> 100,000 cell/ml) was sustained throughout this 5-month period.  Abundance estimates are the means from 3 sampling sites within the pond.  Microcystin concentrations for each site are shown (note log scale).  This is the pond used to test the use of constructed wetlands as a supplementary BMP to mitigate pond HABs.