Ecology of phytoplankton communities dominated by Aureococcus anophagefferens: the role of viruses, nutrients, and microzooplankton grazing

We investigated the impact of viruses, nutrient loading, and microzooplankon grazing on phytoplankton communities in two New York estuaries that hosted blooms of the brown tide alga Aureococcus anophagefferens during 2000 and 2002. The absence of a bloom at one location during 2002 allowed for the fortuitous comparison of a bloom and non-bloom year at the same location as well as a comparison of two sites experiencing bloom and non-bloom conditions during the same year. During the study, blooms were found at locations with high levels of dissolved organic nitrogen and lower nitrate concentrations compared to a non-bloom location. Experimental additions of inorganic nitrogen and phosphorus yielded growth rates within the total phytoplankton community which significantly exceeded control treatments in 83% of experiments, while A. anophagefferens experienced significantly increased growth during only 20% of experimental inorganic nutrient additions. Consistent with prior research, these results suggest brown tides are not caused by eutrophication, but instead are more likely to occur when sources of labile DOM are readily available. Microzooplankton grazing rates on the total phytoplankton community during a bloom were lower than grazing rates at a non-bloom site, and grazing rates on A. anophagefferens were lower than grazing rates on the total community on some dates, suggesting that reduced grazing mortality may also promote brown tides. Mean densities of viruses during blooms (3 × 10⁸ ml⁻¹) were elevated compared to most estuarine environments and were twice the levels found at a non-bloom site. Experimental enrichment of the natural viral densities yielded a significant increase in A. anophagefferens growth rates relative to control treatments when background levels of viruses were low (<1.7 × 10⁸ ml⁻¹), suggesting that viruses may promote bloom occurrence by regenerating DOM or altering the composition of microbial communities.     

HPLC pigment records provide evidence of past blooms of Aureococcus anophagefferens in the Coastal Bays of Maryland and Virginia, USA

Concentrations of the accessory phytoplankton pigment 19′-butanoyloxyfucoxanthin (but-fuco), derived from archived high performance liquid chromatography (HPLC) data from the Atlantic coastal bays of Maryland and Virginia (1993–1995 and 1999–2002), were used to determine the presence of Aureococcus anophagefferens at 18 stations. Paired data of direct cell counts of A. anophagefferens and but-fuco concentration data from 2000 to 2002 were linearly regressed (R² = 0.78). This regression was used to estimate historical cell densities from 1994 to 1995 and to improve the spatial resolution from 1999 to 2002. Although the HPLC method used did not permit quantification of but-fuco before 1994, the records indicate that qualitatively A. anophagefferens was present in 1993. Quantitative measurements grouped into bloom index categories showed that annually, peak densities occurred in May to July. Severe Category 3 blooms (>200,000 cells ml⁻¹) were found in 1995, 2001, and 2002. Spatially, concentrations of but-fuco were higher in the northern extent of the study region than in the lower Chincoteague Bay, and along the western shore of Chincoteague Bay than on the eastern side. On an interannual basis, it appeared that A. anophagefferens became more geographically widespread and blooms more intense throughout the study period.     

Potential impacts of brown tide, Aureococcus anophagefferens, on juvenile hard clams, Mercenaria mercenaria, in the Coastal Bays of Maryland, USA

The rate of growth of juvenile hard clams, Mercenaria mercenaria, was studied in the Coastal Bays of Maryland during an outbreak of the brown tide, Aureococcus anophagefferens. Brown tide dominated the plankton community during the month of June 2002, with cell densities at several sites reaching category 3 (>200,000 cells ml⁻¹) levels. Temperatures during the bloom were 18.6–27.5 °C. Nutrient conditions preceding and during the bloom were conducive for the proliferation of A. anophagefferens: while inorganic nitrogen and phosphorus were <1 μg at N or P l⁻¹, urea was elevated during bloom development. Organic nitrogen, phosphorus and carbon were in the range of levels observed in previous brown tide blooms and increased following the collapse of the bloom. Growth rates of juvenile clams were significantly lower during the period of the brown tide bloom than following its collapse. Growth rates of M. mercenaria were found to be negatively impacted at brown tide densities as low as 20,000 cells ml⁻¹, or category 1 levels. The low growth rates of M. mercenaria could not be explained by temperature, as the lowest growth rates were found when water temperatures were at levels previously found to be optimal for growth.     

Stimulation of the brown tide organism, Aureococcus anophagefferens, by selective nutrient additions to in situ mesocosms

The influence of nutrient additions and sediment exchange on Aureococcus anophagefferens growth was studied using 200 l mesocosms deployed in situ at the Southampton College Marine Science Center in Long Island, New York. A. anophagefferens cell density increased in mesocosms with separate additions of the following materials: urea + glucose and desiccation-stressed Enteromorpha tissue. A decrease in A. anophagefferens cell density was observed in mesocosms with either no additions (control) or with added nitrate. A treatment containing a sediment layer exhibited an increase in average cell densities, but the increase was not statistically significant (P = 0.15). In the 9 day experiment, net growth of A. anophagefferens was only observed during the last 3 days, which corresponded to a period of high solar irradiation. Total chlorophyll concentration declined in all treatments during the first 6 days, which corresponded to relatively low daily irradiance, suggesting low-light stress on the phytoplankton assemblage during the initial phase of the experiment. During the ensuing sunny period, a 4–5-fold increase in chlorophyll concentration was observed in the nitrate and urea treatments with lesser increases in the other treatments. A. anophagefferens density increased relative to total phytoplankton biomass (Chl basis) in the urea + glucose and Enteromorpha treatments. Results are consistent with a prevailing hypothesis that organic nitrogen nutrients favor the growth of A. anophagefferens. Specifically, our evidence indicates that A. anophagefferens exhibited net population growth under organic N, but not inorganic N nutrient (specifically NO₃⁻) loading.     

First occurrence of Cochlodinium blooms in Sabah, Malaysia

Harmful algal blooms (HABs) resulting in red discoloration of coastal waters in Sepanggar Bay, off Kota Kinabalu, Sabah, East Malaysia, were first observed in January 2005. The species responsible for the bloom, which was identified as Cochlodinium polykrikoides, coincided with fish mortalities in cage-cultures. Determinations of cell density between January 2005 and June 2006 showed two peaks that occurred in March–June 2005 and June 2006. Cell abundance reached a maximum value of 6 × 10⁶ cells L⁻¹ at the fish cage sampling station where the water quality was characterized by high NO₃–N and PO₄–P concentrations. These blooms persisted into August 2005, were not detected during the north–east monsoon season and occurred again in May 2006. Favorable temperature, salinity and nutrient concentrations, which were similar to those associated with other C. polykrikoides blooms in the Asia Pacific region, likely promoted the growth of this species. Identification of C. polykrikoides as the causative organism was based on light and scanning microscopy, and confirmed by partial 18S ribosomal DNA sequences of two strains isolated during the bloom event (GenBank accession numbers DQ915169 and DQ915170).     

Using clay to control harmful algal blooms: deposition and resuspension of clay/algal flocs

Harmful algal blooms (HABs) may be legitimate targets for direct control or mitigation, due to their impacts on commercial fisheries and public health. One promising control strategy is the rapid sedimentation of HABs through flocculation with clay. The objective of this study was to evaluate flow environments in which such a control strategy might be effective in removing harmful algae from the water column and depositing a layer of clay/algal flocs on the sea floor. We simulated the natural environment in two laboratory flumes: a straight-channel “17 m flume” in which flocs settled in a still-water column and a “racetrack flume” in which flocs settled in flow. The 17 m flume experiments were designed to estimate the critical bed shear stress for resuspension of flocs that had settled for different time periods. The racetrack flume experiments were designed to examine the deposition and repeated resuspension of flocs in a system with tidal increases in flow speed. All flume runs were conducted with the non-toxic dinoflagellate Heterocapsa triquetra and phosphatic clay (IMC-P4). We repeated the experiments with a coagulant, polyaluminum hydroxychloride (PAC), expected to enhance the removal efficiency (RE) of the clay. Our experiments indicated that at low flow speeds (≤10 cm s⁻¹), phosphatic clay was effective at removing algal cells from the water column, even after repeated resuspension. Once a layer of flocs accumulated on the bed, the consolidation, or dewatering, of the layer over time increased the critical shear stress for resuspension (i.e. decreased erodibility). Resuspension of a 2 mm thick layer that settled for 3 h in relatively low flow speeds (≤3 cm s⁻¹) would be expected at bed shear stress of 0.06–0.07 Pa, as compared to up to 0.09 Pa for a layer that was undisturbed for 9 or 24 h. For the same experimental conditions, the addition of PAC decreased the removal efficiency of algal cells in flow and increased the erodibility of flocs from the bottom. By increasing the likelihood that flocs remain in suspension, the addition of PAC in field trials of clay dispersal might have greater impact on sensitive, filter-feeding organisms. Overall, our experiments suggest that the flow environment should be considered before using clay as a control strategy for HABs in coastal waters.     

Assessment of brown tide blooms, caused by Aureococcus anophagefferens, and contributing factors in New Jersey coastal bays: 2000–2002

A 3 year study (2000–2002) in Barnegat Bay-Little Egg Harbor (BB/LEH), New Jersey (USA), was conducted by the New Jersey Department of Environmental Protection, Division of Science Research and Technology (DSRT) in cooperation with several partners to assess brown tide blooms in coastal waters in NJ. Water samples were collected by boat and helicopter at coastal stations from 2000 to 2002 along with field measurements. Aureococcus anophagefferens were enumerated and associated environmental factors were analyzed. A. anophagefferens abundances were classified using the Brown Tide Bloom Index and mapped, along with salinity and temperature parameters, to their geo-referenced location using the ArcView GIS. The highest A. anophagefferens abundances (>10⁶ cells ml⁻¹), including category 3 blooms (≥200,000 cells ml⁻¹) and category 2 blooms (≥35,000 to ≤200,000 cells ml⁻¹), recurred during each of the 3 years of sampling and covered significant geographic areas of the estuary, especially in Little Egg Harbor. While category 3 blooms were generally associated with warmer water temperatures (>16 °C) and higher salinity (>25–26 ppt), these factors were not sufficient alone to explain the timing or distribution of A. anophagefferens blooms. There was no significant relationship between brown tide abundances and dissolved organic nitrogen measured in 2002 but this was consistent with other studies. Extended drought conditions, with corresponding low freshwater inputs and elevated bay water salinities, occurring during this time were conducive to blooms. A. anophagefferens abundances were well above the reported levels that have been reported to cause negative impacts on shellfish. It was shown that over 50% of the submerged aquatic vegetation (SAV) habitat located in Barnegat Bay/Little Egg Harbor was categorized as having a high frequency of category 2 or 3 blooms for all 3 years.     

Control of the harmful alga Cochlodinium polykrikoides by the naked ciliate Strombidinopsis jeokjo in mesocosm enclosures

Red tides dominated by the harmful dinoflagellate Cochlodinium polykrikoides have caused annual losses of USD $5–60 million to the Korean aquaculture industry annually since 1995 and a loss of USD $3 million during a 1999 net-pen fish mortality event in Canada. In order to evaluate the potential to control C. polykrikoides red tides dominated by using mass-cultured heterotrophic protistan grazers, we monitored the abundance of Strombidinopsis jeokjo (a naked ciliate) and C. polykrikoides after mass-cultured S. jeokjo was introduced into mesocosms (ca. 60 l) deployed in situ and containing natural red tide waters dominated by C. polykrikoides. Water temperature, salinity, and pH, as well as the abundance of co-occurring other protists and metazooplankton were measured concurrently. To compare the growth and ingestion rates of S. jeokjo feeding on cultured versus natural populations of C. polykrikoides, we also monitored the abundance of cultured C. polykrikoides and S. jeokjo in bottles during laboratory grazing experiments. S. jeokjo introduced into the mesocosms grew well, effectively reducing natural populations of C. polykrikoides from approximately 1000 cells ml⁻¹ to below 10 cells ml⁻¹ within 2 days. The growth and ingestion rates of cultured S. jeokjo on natural populations of C. polykrikoides in the mesocosms for the first 30 h (0.72 day⁻¹ and 51 ng C grazer⁻¹ day⁻¹) were 84% and 44%, respectively, of those measured in the laboratory during bottle incubations with similar initial prey concentrations. The calculated grazing impact of S. jeokjo on natural populations of C. polykrikoides suggests that large-scale cultures of this ciliate could be used for controlling red tides by C. polykrikoides in small areas.     

Ingestion of the dinoflagellate, Pfiesteria piscicida, by the calanoid copepod, Acartia tonsa

The dinoflagellate, Pfiesteria piscicida, can form harmful algal blooms in estuarine environments. The dominant copepod species usually found in these waters is Acartia tonsa. We tested the ability of A. tonsa to graze the non-toxic zoospore stage of P. piscicida and thus serve as a potential biological control of blooms of this algal species. A. tonsa grazed the non-toxic zoospore stages of both a non-inducible P. piscicida strain (FDEPMDR23) and a potentially toxic strain (Tox-B101156) at approximately equal rates. Ingestion of P. piscicida increased with cell concentration and exhibited a saturated feeding response. Both the maximum number of cells ingested (I_max) and the slope of the ingestion curve (α) of A. tonsa feeding on P. piscicida were comparable to these ingestion parameters for A. tonsa fed similar-sized phytoplankton and protozoan species. When these laboratory ingestion rates were combined with abundance estimates of A. tonsa from the Pocomoke Estuary and Chesapeake Bay, we found that significant grazing control of the non-toxic zoospore stage of P. piscicida by A. tonsa would only occur at high copepod abundances (>10 copepods L⁻¹). We conclude that under most in situ conditions the potential biological control of blooms of P. piscicida is exerted by microzooplankton grazers. However, in the less saline portions of estuaries where maximum concentrations of copepods often occur with low abundances of microzooplankton, copepod grazing coefficients can be similar to the growth rates of P. piscicida.     

Intraspecific variability in Karlodinium veneficum: Growth rates, mixotrophy, and lipid composition

We isolated eleven strains of the harmful algal bloom (HAB)-forming dinoflagellate Karlodinium veneficum during a bloom event in the NW Mediterranean coastal waters and we studied the inter-strain variability in several of their physiological and biochemical traits. These included autotrophic growth parameters, feeding capabilities (mixotrophy), lipid composition, and, in some cases, their responses to biotic and abiotic factors. The strains were found to differ in their growth rates (0.27–0.53 d⁻¹) and in the maximum cell concentrations achieved during stationary phase (6.1 × 10⁴–8.6 × 10⁴ cells mL⁻¹). Their ingestion performance, when offered Rhodomonas salina as prey, was also diverse (0.22–1.3 cells per K. veneficum per day; 8–52% of their daily ration). At least two strains survived for several months under strict heterotrophic conditions (no light, low inorganic nutrients availability, and R. salina as food source). These strains also showed very distinct fatty acid compositions, with very low contents of monounsaturated and polyunsaturated fatty acids. According to a Bray Curtis similarity analysis, three or four strain groups able to perform different roles in bloom development were identified. We further analyzed one strain from each of the two most distinct groups with respect to prey concentration, light intensity, nutrient availability, and we determined the functional responses (growth and feeding rates) to food concentration. Taken together, the results served to highlight the role of mixotrophy and clone variability in the formation of HABs.     

A comparison of N and C uptake during brown tide (Aureococcus anophagefferens) blooms from two coastal bays on the east coast of the USA

Blooms of the brown tide pelagophyte, Aureococcus anophagefferens, have been reported in coastal bays along the east coast of the USA for nearly two decades. Blooms appear to be constrained to shallow bays that have low flushing rates, little riverine input and high salinities (e.g., >28). Nutrient enrichment and coastal eutrophication has been most frequently implicated as the cause of A. anophagefferens and other blooms in coastal bays. We compare N and C dynamics during two brown tide blooms, one in Quantuck Bay, on Long Island, NY in 2000, and the other in Chincoteague Bay, at Public Landing, MD in 2002, with a physically similar site in Chincoteague Bay that did not experience a bloom. We found that the primary forms of nitrogen (N) taken up during the bloom in Quantuck Bay were ammonium and dissolved free amino acids (DFAA) while the primary form of N fueling production at both sites in Chincoteague Bay was urea. At both Chincoteague sites, amino acid carbon (C) was taken up while urea C was not. Even though A. anophagefferens has the ability to take up organic C, during the bloom at Chincoteague Bay, photosynthetic uptake of bicarbonate was the dominant pathway of C acquisition by the >1.2 μm size fraction during the day. C uptake by cells <5.0 μm was insufficient to meet cellular C demand based on the measured N uptake rates and the C:N ratio of particulate material. While cells >1.2 μm did not take up much organic C during the day, smaller cells (>0.2 μm) did. Peptide hydrolysis appeared to play an important role in mobilizing organic matter in Quantuck Bay, where amino acids contributed substantially to N and C uptake, but not in Chincoteague Bay. Dissolved organic N (DON), dissolved organic C (DOC) concentrations and the DOC/DON ratio were higher and total dissolved inorganic N (DIN) concentrations were lower at the bloom site in Chincoteague Bay than at the nonbloom site in the same bay. We conclude that A. anophagefferens is capable of using a wide variety of N and C compounds, and that nutrient inputs, biotic interactions and the dominant recycling pathways determine which compounds are available and which metabolic pathways are active at a particular site.     

Pelagic-benthic transition of the harmful alga, Heterosigma akashiwo: Changes in swimming and implications for benthic cell distributions

Many harmful algal blooming (HAB) species transition between a vegetative, motile phase in the water column and a dormant, non-motile resting phase in the sediments. These life history transitions potentially regulate the timing, location and persistence of bloom events. Motility promotes aggregation and influences vertical distributions in the water column. However, the contribution of this behavior to benthic distributions of resting cells is currently unknown. We used video-tracking techniques to test the hypothesis that algal cells use active down-swimming during pelagic-benthic transition to favorably influence benthic distributions. In an experimental water column, we monitored cell swimming trajectories of Heterosigma akashiwo for 14 days after cells were signaled to enter the benthic resting stage. Using the statistical characteristics of individual cell trajectories, we developed a video-based motion assay to assign each tracked Heterosigma cell to one of three cell states known to occur during pelagic-benthic transition: induced motile, transitional and resting. The primary swimming characteristic influencing benthic distribution, net vertical velocity, was essentially the same for all three cell states. Hence, we found no evidence that active down-swimming influences benthic distributions. Our data suggest that benthic distributions of Heterosigma resting cells are similar to distributions of slowly sedimenting passive particles. These observations suggest that Heterosigma benthic resting cell distributions can be predicted by modeling the effects of cell sedimentation rates combined with geophysical flow patterns.     

The potential threat of algal blooms to the abalone (Haliotis midae) mariculture industry situated around the South African coast

Toxic algal blooms are common world-wide and pose a serious problem to the aquaculture and fishing industries. Dinoflagellate species such as Karenia brevis, Karenia mikimotoi, Heterosigma akashiwo and Chatonella cf. antiqua are recognised toxic species implicated in various faunal mortalities. Toxic blooms of Karenia cristata were observed on the south coast of South Africa for the first time in 1988 and were responsible for mortalities of wild and farmed abalone. K. cristata and various other dinoflagellate species common along the South African coast, as well as K. mikimotoi (Isolation site: Norway, Univ. of Copenhagen) and K. brevis (Isolation site: Florida, BIGELOW), were tested for toxicity by means of a bioassay involving Artemia larvae as well as abalone larvae and spat. K. cristata, like K. brevis, contains an aerosol toxin; however, the toxin present in K. cristata has not yet been isolated and remains unknown. K. brevis was, therefore, used to determine which developmental phase of the bloom would affect abalone farms most, and whether ozone could be used as an effective mitigating agent. Of the 17 dinoflagellate species tested, K. cristata, Akashiwo sanguinea, K. mikimotoi and K. brevis pose the greatest threat to the abalone mariculture industry. K. brevis was most toxic during its exponential and stationary phases. Results suggest that ozone is an effective mitigation agent but its economic viability for use on abalone farms must still be investigated.     

Cochlodinium polykrikoides blooms and eco-physical conditions in the South Sea of Korea

Eco-physical conditions for the initiation and termination of Cochlodinium polykrikoides blooms in the South Sea of Korea are examined in this paper. The C. polykrikoides blooms generally occur in the sea near Naro-Do in late August every year. The submarine canyon near Naro-Do plays an important role in surface water intrusion from the open ocean driven by northeasterly winds. In late August, the monsoonal wind system in Korea changes from southwesterly to northeasterly winds, causing Ekman transport of warm, fresh Changjiang Diluted Water (CDW) into the sea near Naro-Do and creating a front between inland sea water and CDW. Along the front, aggregation of single C. polykrikoides cells in the CDW and downwelling yield favorable eco-physical conditions for development of C. polykrikoides blooms. When typhoons and strong northeasterly winds bring vertically well-mixed East China Sea water into the sea near Naro-Do again in September, the eco-physical conditions favor diatom growth and lead to the termination of C. polykrikoides blooms.     

Microbial herbivory on the brown tide alga, Aureococcus anophagefferens: results from natural ecosystems, mesocosms and laboratory experiments

Experiments were conducted with natural plankton assemblages from two areas in Great South Bay (GSB) and the Peconic Bays Estuary System, NY, to compare the rates of growth and pelagic grazing mortality of Aureococcus anophagefferens with co-occurring phytoplankton. We hypothesized that A. anophagefferens would experience low mortality rates by microbial herbivores (relative to feeding pressure on other algae) thus providing it with a competitive advantage within the phytoplankton community. In fact, substantial rates of mortality were observed in nearly every experiment in our study. However, mortality rates of A. anophagefferens were less than intrinsic growth rates of the alga during late spring and early summer in Great South Bay, resulting in positive net growth rates for the alga during that period. This timing coincided with the development of a brown tide in this estuary. Similarly, growth rates of the alga also exceeded mortality rates during bloom development in natural plankton assemblages from the Peconic Bays Estuary System held in mesocosms. In contrast to the situation for A. anophagefferens, growth rates of the total phytoplankton assemblage, and another common picoplanktonic phytoplankter (Synechococcus spp.), were frequently less than their respective mortality rates. Mortality rates of A. anophagefferens in both systems were similar to growth rates of the alga during later stages of the bloom. Laboratory studies confirmed that species of phagotrophic protists that consume A. anophagefferens (at least in culture) are present during brown tides but preference for or against the alga appears to be species-specific among phagotrophic protists. We conclude that two scenarios may explain our results: (1) protistan species capable of consuming the brown tide alga were present at low abundances during bloom initiation and thus not able to respond rapidly to increases in the intrinsic growth rate of the alga, or (2) the brown tide alga produced substance(s) that inhibited or retarded protistan grazing activities during the period of bloom initiation. The latter scenario seems less likely given that significant mortality of A. anophagefferens was measured during our field study and mesocosm experiment. However, even a minor reduction in mortality rate due to feeding selectivity among herbivores might result in a mismatch between growth and grazing of A. anophagefferens that could give rise to significant net population growth of this HAB species. Either scenario infers an important role for trophic interactions within the plankton as a factor explaining the development of brown tides in natural ecosystems.     

Pathology and immune response of the blue mussel (Mytilus edulis L.) after an exposure to the harmful dinoflagellate Prorocentrum minimum

The harmful dinoflagellate Prorocentrum minimum has different effects upon various species of grazing bivalves, and these effects also vary with life-history stage. Possible effects of this dinoflagellate upon mussels have not been reported; therefore, experiments exposing adult blue mussels, Mytilus edulis, to P. minimum were conducted. Mussels were exposed to cultures of toxic P. minimum or benign Rhodomonas sp. in glass aquaria. After a short period of acclimation, samples were collected on day 0 (before the exposure) and after 3, 6, and 9 days of continuous-exposure experiment. Hemolymph was extracted for flow-cytometric analyses of hemocyte, immune-response functions, and soft tissues were excised for histopathology. Mussels responded to P. minimum exposure with diapedesis of hemocytes into the intestine, presumably to isolate P. minimum cells within the gut, thereby minimizing damage to other tissues. This immune response appeared to have been sustained throughout the 9-day exposure period, as circulating hemocytes retained hematological and functional properties. Bacteria proliferated in the intestines of the P. minimum-exposed mussels. Hemocytes within the intestine appeared to be either overwhelmed by the large number of bacteria or fully occupied in the encapsulating response to P. minimum cells; when hemocytes reached the intestine lumina, they underwent apoptosis and bacterial degradation. This experiment demonstrated that M. edulis is affected by ingestion of toxic P. minimum; however, the specific responses observed in the blue mussel differed from those reported for other bivalve species. This finding highlights the need to study effects of HABs on different bivalve species, rather than inferring that results from one species reflect the exposure responses of all bivalves.     

Detection of Karenia brevis blooms on the west Florida shelf using in situ backscattering and fluorescence data

Using shipboard data collected from the central west Florida shelf (WFS) between 2000 and 2001, an optical classification algorithm was developed to differentiate toxic Karenia brevis blooms (>10⁴ cells l⁻¹) from other waters (including non-blooms and blooms of other phytoplankton species). The identification of K. brevis blooms is based on two criteria: (1) chlorophyll a concentration ≥1.5 mg m⁻³ and (2) chlorophyll-specific particulate backscattering at 550 nm ≤ 0.0045 m² mg⁻¹. The classification criteria yielded an overall accuracy of 99% in identifying both K. brevis blooms and other waters from 194 cruise stations. The algorithm was validated using an independent dataset collected from both the central and south WFS between 2005 and 2006. After excluding data from estuarine and post-hurricane turbid waters, an overall accuracy of 94% was achieved with 86% of all K. brevis bloom data points identified successfully. Satisfactory algorithm performance (88% overall accuracy) was also achieved when using underway chlorophyll fluorescence and backscattering data collected during a repeated alongshore transect between Tampa Bay and Florida Bay in 2005 and 2006. These results suggest that it may be possible to use presently available, commercial optical backscattering instrumentation on autonomous platforms (e.g. moorings, gliders, and AUVs) for rapid and timely detection and monitoring of K. brevis blooms on the WFS.     

The association of algicidal bacteria and raphidophyte blooms in South Carolina brackish detention ponds

Over the past 5 years, raphidophyte blooms have been frequently observed along the South Carolina coastal zone. During the 2002, 2003, and 2004 sampling seasons, we investigated temporal fluctuations of algicidal bacteria abundance against raphidophycean flagellates (Heterosigma akashiwo, Chattonella subsalsa, and Fibrocapsa japonica) using the microplate most probable number (MPN) method in three Kiawah Island brackish stormwater detention ponds (K1, K2, and K75). Local axenic isolates of H. akashiwo, C. subsalsa, and F. japonica were obtained and their susceptibility to algicidal bacteria tested. A total of 195 algicidal bacterial strains were isolated from raphidophyte blooms in the study ponds, and 6 of them were identified at the genus level, and the taxonomic specificity of their algicidal activity was tested against local (pond) and nonlocal isolates of raphidophytes (3 species, 10 total strains). In the ponds, a consistent association was found between raphidophyte bloom development and an increase in bacteria algicidal to the bloom species. In 12 of 15 cases, bloom decline followed the increase in algicidal bacteria to maximum abundances. Although variability was found in the taxonomic specificity of the algicidal bacteria effect (i.e. the number of raphidophyte species affected by a particular bacteria strain) and raphidophyte susceptibility (i.e. the number bacteria strains affecting a particular raphidophyte species), a toxic effect was always found when strains of a raphidophyte species were exposed to algicidal bacteria isolated from a bloom caused by that same species. The results suggest that algicidal bacteria may be an important limiting factor in raphidophyte bloom sustenance and can promote bloom decline in brackish lagoonal eutrophic estuaries.     

Microbial community interactions and population dynamics of an algicidal bacterium active against Karenia brevis (Dinophyceae)

The population dynamics of Cytophaga strain 41-DBG2, a bacterium algicidal to the harmful algal bloom (HAB) dinoflagellate Karenia brevis, were investigated in laboratory experiments using fluorescent in-situ hybridization (FISH) and denaturing gradient gel electrophoresis (DGGE). Following its introduction into non-axenic K. brevis cultures at concentrations of 10³ or 10⁵ bacterial cells per milliliter, 41-DBG2 increased to 10⁶ cells per milliliter before initiation of its algicidal activity. Such threshold concentrations were not achieved when starting algal cell numbers were relatively low (10³ cells per milliliter), suggesting that the growth of this bacterium may require high levels of dissolved organic matter (DOM) excreted by the algae. It remains to be determined whether this threshold concentration is required to trigger an algicidal response by 41-DBG2 or, alternatively, is the point at which the bacterium accumulates to an effective killing concentration. The ambient microbial community associated with these algal cultures, as determined by DGGE profiles, did not change until after K. brevis cells were in the process of lysing, indicating a response to the rapid input of algal-derived organic matter. Resistance to algicidal attack exhibited by several K. brevis clones was found to result from the inhibition of 41-DBG2 growth in the presence of currently unculturable bacteria associated with those clones. These bacteria apparently prevented 41-DBG2 from reaching the threshold concentration required for initiation of algicidal activity. Remarkably, resistance and susceptibility to the algicidal activity of 41-DBG2 could be transferred between K. brevis clones with the exchange of their respective unattached bacterial communities, which included several dominant phylotypes belonging to the α-proteobacteria, γ-proteobacteria, and Cytophaga–Flavobacterium–Bacteroides (CFB) groups. We hypothesize that CFB bacteria may be successfully competing with 41-DBG2 (also a member of the CFB) for nutrients, thereby inhibiting growth of the latter and indirectly providing resistance against algicidal attack. We conclude that if algicidal bacteria play a significant role in regulating HAB dynamics, as some authors have inferred, bacterial community interactions are crucial factors that must be taken into consideration in future studies.