Role of Shellfish Aquaculture in the Reduction of Eutrophication in an Urban Estuary.

Land-based management has reduced nutrient discharges; however, many coastal waterbodies remain impaired. Oyster "bioextraction" of nutrients and how oyster aquaculture might complement existing management measures in urban estuaries was examined in Long Island Sound, Connecticut. Eutrophication status, nutrient removal, and ecosystem service values were estimated using eutrophication, circulation, local- and ecosystem-scale models, and an avoided-costs valuation. System-scale modeling estimated that 1.31% and 2.68% of incoming nutrients could be removed by current and expanded production, respectively. Up-scaled local-scale results were similar to system-scale results, suggesting that this up-scaling method could be useful in bodies of water without circulation models. The value of removed nitrogen was estimated using alternative management costs (e.g., wastewater treatment) as representative, showing ecosystem service values of $8.5 and $470 million per year for current and maximum expanded production, respectively. These estimates are conservative; removal by clams in Connecticut, oysters and clams in New York, and denitrification are not included. Optimistically, the calculation of oyster-associated removal from all leases in both states (5% of bottom area) plus denitrification losses showed increases to 10%-30% of annual inputs, which would be higher if clams were included. Results are specific to Long Island Sound, but the approach is transferable to other urban estuaries.

Profiles of paralytic shellfish toxins in bivalves of low and elevated toxicities following exposure to Gymnodinium catenatum blooms in Portuguese estuarine and coastal waters.

Profiles of paralytic shellfish toxins (PSTs) were examined in 405 composite samples of Mytilus spp., Cerastoderma edule, Donax trunculus and Spisula solida collected between 2007 and 2012 from natural production areas in two estuaries (Aveiro and Mondego), two coastal lagoons (Óbidos and Formosa), and three open coastal areas (Aguda, Comporta and Culatra). Toxin concentrations were obtained from the biotoxin monitoring programme database. Episodes of PST toxicity in Portugal have been associated with Gymnodinium catenatum blooms. Toxin profiles for each species showed no trend over the surveyed years. In general, profiles differ only slightly among areas, except for Óbidos. However, toxin profiles in bivalves varied between low and elevated toxicities, corresponding to below and above the PST regulatory limit, respectively. The ratio R1=(C1+2):B1, which were the main toxins produced by G. catenatum cells, decreased considerably between elevated and low toxicity cockles, indicating the elimination of C1+2 or conversion of compounds into B1. R2=[(dcSTX)+(dcGTX2+3)]:[(C1+2)+(B1)], which represents the ratio of minor to major toxins in G. catenatum cells, increased substantially in wedge clams (D. trunculus) of low toxicity and less markedly in cockles (C. edule) and mussels (Mytilus spp.). These differences are interpreted as the predominance of a biotransformation phase after exposure to the algal bloom. The toxin profile of surf clams (S. solida) was dominated by decarbamoyl compounds, reflecting intense biotransformation during exposure to blooms. The higher ratio R2 in low toxicity samples suggests that elimination of the produced decarbamoyl toxins was slower than biotransformation.

Partitioning of paralytic shellfish toxins in sub-cellular fractions of the digestive gland of the cockle Cerastoderma edule: Changes under post-bloom natural conditions.

Concentrations of paralytic shellfish toxins (C1+2, B1, dcGTX2+3, dcSTX, GTX2+3 and STX) were determined by LC-FLD in composite samples of digestive glands of the cockle Cerastoderma edule and in each sub-cellular particulate fractions obtained after differential centrifugation (nuclei+debris, mitochondria, lysosomes and microsomes). The specimens were sampled during the exposure to a bloom of Gymnodinium catenatum (day 0) and in the subsequent 8, 12, 14, 19, 21 and 25 days under natural depuration conditions. Toxin profiles of digestive glands were dominated by C1+2 followed by B1 and dcGTX2+3, although the proportion between C1+2 and B1 contents decreased with the time, indicating a slower elimination of B1. All toxins, except GTX2+3 and STX, were quantified in the four sub-cellular fractions. The content of the quantified toxins decreased most markedly in nuclei+debris and microsomal fractions, during the first eight and 12 days, respectively. Conversely, different patterns were observed among toxins in mitochondrial and lysosomal fractions. The less accentuated decreases of dcGTX2+3 and dcSTX contents in the mitochondrial fraction may have resulted from the conversion of other toxins, like C1+2 and B1, associated with enzymatic activities in that fraction. The largest discrepancy was registered in lysosomal fraction for B1, since its content increased after eight days of post-bloom conditions. Input of B1 may come from the conversion of other toxins, like the abundant B2 and C1+2. These transformations are associated to the major role of lysosomes in the intra-cellular digestive process of materials acquired through vesicular transport.

Uptake and release of paralytic shellfish toxins by the clam Ruditapes decussatus exposed to Gymnodinium catenatum and subsequent depuration.

A laboratory experiment was performed with the clam Ruditapes decussatus, fed with the toxic dinoflagellate Gymnodinium catenatum and the non-toxic algae Isochrysis galbana (14 days) and subsequently only with I. galbana (15 days). Individual paralytic shellfish toxins were determined by LC-FLD in G. catenatum cells, whole clam tissues, and particulate organic matter (POM) produced by clams. The toxins dcSTX and dcGTX2 + 3 in the algae were less abundant than C1 + 2 and B1, but were predominant in clams during both the exposure and depuration phases. The toxin dcNEO was only detected in clams during a short period, indicating conversion from other compounds. The toxin composition of the POM indicated the export of dcSTX as faeces or pseudo-faeces along the entire experiment (2.5-14 nmol mg(-1)), B1 was present in a short period of the exposure and C1 + 2 and dcGTX2 + 3 absent. A mass balance calculation indicated that approximately 95% of C1 + 2 and 85% of B1 supplied to the clams were converted into other toxins or lost in solution. Conversely, the net gain of 512, 61 and 31 nmol for dcSTX, dcGTX2 + 3 and dcNEO, respectively, suggests the conversion from other assimilated compounds by clams during exposure and depuration phases.

Application and sensitivity testing of a eutrophication assessment method on coastal systems in the United States and European Union.

The Assessment of Estuarine Trophic Status (ASSETS) screening model has been extended to allow its application to both estuarine and coastal systems. The model, which combines elements of pressure, state and response, was tested on four systems: Maryland Coastal Bays and Long Island Sound in the United States and The Firth of Clyde (Scotland) and Tagus Estuary (Portugal) in the European Union. The overall scores were: Maryland Coastal Bays: Bad; Firth of Clyde: Poor; Tagus Estuary: Good. Long Island Sound was modelled along a timeline, using 1991 data (score: Bad) and 2002 data (score: Moderate). The improvement registered for Long Island Sound is a consequence of the reduction in nutrient loading, and the ASSETS score changed accordingly. The two main areas where developments are needed are (a) In the definition of type-specific ranges for eutrophication parameters, due to the recognition that natural or pristine conditions may vary widely, and the use of a uniform set of thresholds artificially penalizes some systems and potentially leads to misclassification; (b) In the definition and quantification of measures which will result in an improved state through a change in pressures, as well as in the definition of appropriate metrics through which response may be assessed. One possibility is the use of detailed research models where different response scenarios potentially produce changes in pressure and state. These outputs may be used to drive screening models and analyze the suitability of candidate metrics for evaluating management options.