Fantasia, R.L.; Bricelj, V.M., and Ren, L., 2017. Phytoplankton community structure based on photopigment markers in a mid-Atlantic U.S. coastal lagoon: Significance for hard-clam production. In: Buchanan, G.A.; Belton, T.J., and Paudel, B. (eds.), A Comprehensive Assessment of Barnegat Bay-Little Egg Harbor, New Jersey.

Phytoplankton community structure at four contrasting sites in the Barnegat Bay-Little Egg Harbor (BB-LEH) Estuary was determined microscopically and by photopigment-CHEMTAX analysis. It was related to temperature, salinity, and weekly growth rates of juvenile hard clams, Mercenaria mercenaria, deployed at these sites during the summer in 2012 and 2013, pre- and post-Hurricane Sandy. Results indicated distinct differences in phytoplankton composition among sites, with a greater contribution of chlorophytes and cyanobacteria at the northernmost site, Island Beach State Park (IBSP). Photopigment analysis was useful in improving upon the taxonomic assessment of pico-coccoid (<3 μm) algae that are difficult to identify microscopically. The presence of the brown tide alga, Aureococcus anophagefferens, was confirmed by immunofluorescence during both years, with peak densities of 4.4 × 105 cells ml−1 in June 2013 at Sedge Island, in a Marine Conservation Zone that supports clam seeding. Concentrations of 19′but-fucoxanthin and A. anophagefferens showed a strong linear relationship, suggesting that this pigment is a good indicator of this pelagophyte in BB-LEH. The occurrence of brown tide was important to consider in CHEMTAX analysis as it affected the estimated contribution of diatoms to chl a, given that A. anophagefferens is also characterized by a relatively high fucoxanthin:chlorophyll a ratio. Generally, hard-clam tissue-growth rates were greatest at Sedge Island and Tuckerton, LEH, and least at IBSP, a lower salinity site, and off Harvey Cedars, a more developed site along a bulkheaded shoreline. Significant linear relationships were found between clam growth rate and diagnostic photopigments, with positive relationships for indicators of diatoms and negative relationships for those of cyanobacteria and chlorophytes.

The Barnegat Bay-Little Egg Harbor (BB-LEH) Estuary is a mid-Atlantic, coastal lagoon ecosystem spanning ∼70 km of the state of New Jersey coastline, U.S.A. The estuary is shallow (∼1–6 m) and vertically well mixed, with long residence times (up to 70 days) in the lowest energy portions of the bay, especially in summer (Defne and Ganju, 2014; Guo et al., 2004). It is characterized by a strong N-S gradient in salinity, with the northern portions of the bay exhibiting the lowest salinity values because of the influence of the Toms River, the major source of freshwater to the estuary. The highest nitrogen loading from anthropogenic sources and chlorophyll a (chl a) concentrations typically occur in July in the northern reaches of the estuary (Kennish and Fertig, 2012), although salinity in this region typically lies below <15 psu, the tolerance range of adult Mercenaria mercenaria (Bricelj, Kraeuter, and Flimlin, 2017). In recent decades, the BB-LEH estuary has shifted from classification as a moderately eutrophic (Seitzinger and Pilling, 1993) to a highly eutrophic estuary (Bricker et al., 2007; Kennish et al., 2007).

The BB-LEH estuary has historically supported high rates of primary production and large stocks of the commercially valuable hard clam, M. mercenaria, although a marked historical decline of stocks has been observed (see Bricelj, Kraeuter, and Flimlin, 2017). A pronounced decline was also documented since the 1980s in Long Island's South Shore Estuaries, New York, largely attributable to overfishing; recovery of clam populations has been slow, despite a marked reduction in commercial fishing pressure (reviewed by Bricelj, 2009). Several possible causes have been suggested, including recruitment limitation and changes in phytoplankton composition associated with increased eutrophication, e.g., a potential shift from diatoms and other large microalgae to picoplankters (Newell et al., 2009) that are poorly retained and digested and/or toxic for hard clams (Bricelj, 2009; Grizzle, Bricelj, Shumway, 2001).

There are few past studies of the phytoplankton community in the BB-LEH estuary, with only one comprehensive, baywide study conducted within the past two decades (Olsen and Mahoney, 2001) until such studies were resumed in 2011–13 (Ren, 2013, 2015; Ren et al., 2017). Recurring brown tides of the harmful, picoplanktonic alga Aureococcus anophagefferens (Pelagophyceae) occurred in BB-LEH, especially from 1995 to 2004 when concentrations attained up to 2.6 × 106 cells ml−1. These levels exceeded threshold concentrations (5–8 × 104 ml−1) known to adversely affect growth and survival of larval and juvenile hard clams (Bricelj, 2009; Bricelj and MacQuarrie, 2007; Bricelj, MacQuarrie, and Smolowitz, 2004), yet routine monitoring for this alga ceased thereafter (Bricelj, Kraeuter, and Flimlin, 2017; Mahoney, Olsen, and Jeffress, 2006; Pecchioli, Lathrop, and Haag, 2006). Aureococcus anophagefferens produces a dopamine-mimetic toxin that, on contact, inhibits the beat of gill lateral cilia in many bivalves, including juvenile and adult M. mercenaria, thus suppressing feeding (Gainey and Shumway, 1991). Prior studies identified a diverse and seasonally dynamic phytoplankton composition in BB-LEH, with larger diatoms and dinoflagellates dominating during the winter and spring, leading to a greater contribution of small forms (<5 μm) in summer and early fall (Mountford, 1984; Olsen and Mahoney, 2001; Ren, 2013, 2015).

Microscopic analysis of the phytoplankton community is laborious and costly. Alternatives such as the analysis of phytoplankton diagnostic photopigments by high-performance liquid chromatography (HPLC) can provide biomass estimates of various phytoplankton groups and quantify the phytoplankton composition at the class level in a relatively rapid, cost-effective, and reproducible manner (Paerl et al., 2003, 2007). Microscopic analysis can also sometimes be biased, favoring large forms, as nano- (2–20 μm) and picoplankton (0.2–2 μm) are often difficult to identify and quantify using this method. Analysis of photopigments, combined with microscopy data, can prove useful in identifying these small forms. The program CHEMTAX (Mackey et al., 1996) uses factor analysis and a steepest descent algorithm to fit a matrix of marker pigment ratios to actual ratios found in samples and thus provide biomass estimates of multiple taxa as a fraction of chl a. Accurate resolution of phytoplankton taxa using photopigments determined by HPLC and CHEMTAX analysis requires an initial matrix comprising pigment ratios approximating those likely to be present for the classes occurring in samples taken. This method has been used to estimate the biomass of phytoplankton taxa in numerous mid-Atlantic estuaries (e.g., Laza-Martinez et al., 2007; Lewitus et al., 2005; Li et al., 2004; Paerl et al., 2014) but was not previously implemented in the BB-LEH estuary.

Therefore, the main objective of the present study was to provide a first systematic characterization of the summer phytoplankton assemblage in the BB-LEH estuary from photopigment analysis at four representative sites along a N-S latitudinal gradient during 2012 and 2013. Microscopically determined taxonomic analysis was used in parallel to pigment analysis at two of these sites to inform the selection of initial pigment ratios from the literature that were best suited to the samples. A secondary objective was to determine the presence of A. anophagefferens in the estuary and compare results of photopigment analysis with a species-specific immunofluorescence assay for this microalga. Finally, this study made a first attempt to relate phytoplankton group parameters and thus food quality to in situ weekly growth rates of juvenile M. mercenaria at specific sites. Note that none of the sampling sites selected for this study coincided with the NJ Department of Environmental Protection (NJDEP) phytoplankton monitoring stations and thus provided complementary information on the phytoplankton assemblage (in addition to its relationship to hard-clam growth). In particular, no prior phytoplankton characterization is available for the Marine Conservation Zone (MCZ) Sedge Islands (Sedge Is.) area (Figure 1), where clam-stock enhancement (seeding) efforts and recreational harvesting of hard clams are important activities.

Figure 1.

Map of the study area in the Barnegat Bay-Little Egg Harbor (BB-LEH) coastal lagoon estuary, New Jersey (NJ), and its watershed (boundary indicated by bold gray outline). Circles indicate present study sampling stations (black) and NJDEP subsurface water quality and phytoplankton sampling stations (white). Inset: Study area in relation to U.S. mid-Atlantic states. LI = Long Island; NY = New York. Station Coordinates: Island Beach State Park (IBSP), 39.906, −74.088; Sedge Is.; 39.795, −74.119; Harvey Cedars, 39.708, −74.138; Tuckerton Cove, 39.563, −74.340; BB12, 39.582, −74.679; and BB05, 39.884, −74.113.

Figure 1.

Map of the study area in the Barnegat Bay-Little Egg Harbor (BB-LEH) coastal lagoon estuary, New Jersey (NJ), and its watershed (boundary indicated by bold gray outline). Circles indicate present study sampling stations (black) and NJDEP subsurface water quality and phytoplankton sampling stations (white). Inset: Study area in relation to U.S. mid-Atlantic states. LI = Long Island; NY = New York. Station Coordinates: Island Beach State Park (IBSP), 39.906, −74.088; Sedge Is.; 39.795, −74.119; Harvey Cedars, 39.708, −74.138; Tuckerton Cove, 39.563, −74.340; BB12, 39.582, −74.679; and BB05, 39.884, −74.113.

Close modal

Study Area

Four contrasting study sites in the BB-LEH coastal lagoon ecosystem were chosen for weekly phytoplankton characterization based on diagnostic photopigments (Figure 1). The four sites were representative of different habitats in the BB-LEH; from north to south they included (1) Island Beach State Park (IBSP) located in northern BB, SE of Toms River, in a lower salinity sector of the bay; (2) Sedge Is. MCZ, directly exposed to oceanic influence by its proximity to Barnegat Inlet and characterized by extensive eelgrass, Zostera marina, cover; (3) Harvey Cedars, a site along Long Beach Island, south of Barnegat Inlet, adjacent to a highly developed and bulkheaded shoreline; and (4) Tuckerton Cove, located within the Jacques Cousteau National Estuarine Research Reserve, on the western shore of LEH, surrounded by extensive marshland and historically supporting highly productive hard-clam populations (Bricelj, Kraeuter, and Flimlin, 2017). All sites were located on the eastern shore of the estuary, except for Tuckerton Cove, located in shallow water (<3 m in depth), and thus were vertically well mixed. No sites north of IBSP were included, as low salinities (typically <15) below the tolerance range of M. mercenaria (Grizzle, Bricelj, and Shumway, 2001) have historically excluded native hard-clam populations in this region. Data from two NJDEP water-quality and phytoplankton monitoring stations (BB12 and BB05; Figure 1), located near the Tuckerton Cove and IBSP sites, respectively, were also used to supplement phytoplankton microscopy data for these sites when sampling periods overlapped.

Field Water-Column Sampling

Weekly field sampling was split into two periods each year: Trial I (5 June–6 July and 5 June–10 July in 2012 and 2013, respectively) and Trial II (24 July–12 September and 12 August–10 September in 2012 and 2013, respectively). Sampling trials encompassed the major period of hard-clam growth at this latitude, although detectable shell growth starts in May (juveniles from hatcheries are not available this early in the year) and continues into October (Grizzle, Bricelj, and Shumway, 2001). Sampling was conducted on consecutive days for the southern and northern sites because of logistic constraints. Seawater (10 L) was collected from a depth ∼0.5 m off-bottom with a battery-powered Masterflex peristaltic pump to prevent bottom disturbance and resuspension and filtered immediately through a 150-μm Nitex mesh screen to remove zooplankton and large detrital particles. Pre-sieved seawater was transported in a cooler on ice to a nearby location, where it was filtered within ∼1 hour of collection onto Whatman GF/F glass-fiber filters (2.5-cm diameter; nominal pore size = 0.7 μm) in duplicate. Filters were stored at −80°C prior to overnight shipping on dry ice for photopigment analysis. Duplicate water samples were taken for microscopic analysis of phytoplankton (at Sedge Is. and IBSP), and immunofluorescent enumeration of A. anophagefferens was conducted only at key sites and dates (Harvey Cedars, Tuckerton [2012–13] and Sedge Is. [2013]), based on the presence of 19′ butanoylfucoxanthin (19′but). This pigment is diagnostic for pelagophytes (DeYoe et al., 1997) and some dinoflagellates (Zapata et al., 2012). The former are generally rare in estuaries other than for the brown tide species A. anophagefferens and Aureoumbra lagunensis in southern U.S. estuaries. This analysis was not conducted for samples at IBSP given that A. anophagefferens typically proliferates at higher salinities than found at this site (Gobler, Lonsdale, and Boyer, 2005; LaRoche et al., 1997). Water samples were fixed in situ at a final seawater concentration of 1.5% glutaraldehyde and stored at 4°C until analysis.

Water temperature was recorded in situ every 15 minutes using an Onset HOBO® temperature logger deployed at the site to determine short-term variability, as well as taken with a handheld thermometer at discrete time points when water sampling occurred. Salinity was also measured at discrete weekly intervals using a refractometer. Tidal height (centimeters above mean lower low water) at the time of sampling was calculated by fitting the tide predictions from the nearest available National Oceanic and Atmospheric Administration (NOAA) site (NOAA, 2017) to a sine function and solving for tidal height. The nearest sites to IBSP, Sedge Is., Harvey Cedars, and Tuckerton were Seaside Park (+39.9217, −74.0833), Sedge Is. (+39.7883, −74.0983), Loveladies Harbor (+39.7250, −74.1367), and the Tuckerton Creek Entrance (+39.6017, −74.3417), respectively. All tide predictions used are subordinate to Sandy Hook, New Jersey.

Clam Deployment and Sampling

Juvenile hard clams (∼8–9 mm initial shell length [SL], 13.1 mm in Trial I, 2012) were obtained from New Jersey commercial hatcheries that did not experience brown tide. A new cohort of juvenile hard clams, also obtained from a local hatchery source, was deployed for Trial II in both years. Clams were deployed ∼20 cm off-bottom in 4-mm-square mesh bags, within cages (46 × 46 × 46 cm high, divided into three levels) to exclude predators at each site (three to four cages per site, one mesh bag per cage). Placement of clams above bottom prevented confounding effects of substrate type on growth. Cages were deployed in relatively shallow water (≤2 m), marked by surface buoys, and weighed with concrete blocks inserted in the bottom shelf. They were cleaned of fouling organisms on a weekly basis at the time of sampling. Each mesh bag contained 300 to 500 clams, depending on initial size, which is a low density that precludes density-dependent growth inhibition. At each sampling, ∼30 to 50 clams were removed from each cage to determine their SL and tissue dry weight (DW). The DW of dissected tissues was determined by oven-drying to a constant weight (24 to 48 hours depending on size, at 60°C) and weighing with a Cahn electrobalance (±0.1 μg).

Growth rates were determined as kDW, the instantaneous growth coefficient [percentage change in DW of soft tissues day−1]. This parameter was calculated as kDW = [(ln Xf – ln Xo)/time interval in days] × 100, where Xf and Xo are the mean final and initial DW of clams in each cage. Soft-tissue growth is more markedly and immediately affected by changes in environmental conditions than shell growth and thus provides a more sensitive index of short-term environmental change.

Phytoplankton Analysis

For microscopic analysis, fixed phytoplankton samples were size-fractionated by filtration onto 0.2-μm, 3-μm, and 8-μm pore-size filters, with 0.03% proflavine hemisulfate used to stain the latter two size fractions. The 0.2 to 3 μm fractions were counted immediately after filtration. Phytoplankton identification and enumeration were conducted under an epifluorescence microscope (Leica DM L) with blue and green excitation and transmitted light. All species were identified to the lowest taxonomic level possible (see Ren [2013, 2015] for detailed methodology). The biovolume of each taxon was calculated based on microscopic measurements of cell dimensions and geometric models of phytoplankton (e.g., Hillebrand et al., 1999). Aureococcus anophagefferens cell concentrations were determined by immunofluorescence using a monoclonal antibody conjugated to fluorescein isothiocyanate and enumerated by flow cytometry following methods of Stauffer et al. (2008).

Characterization of phytoplankton groups at the four sites was conducted from the analysis of photopigments by HPLC coupled with in-line photodiode array spectrophotometry. Pigments were extracted in 100% acetone at 20°C, filtered on a 0.45-μm filter, and injected into an HPLC system equipped with a series of C18 reverse-phase columns. Pigments were detected by absorbance in the range 380–700 nm and identified by comparing retention times and peak areas with pigment standards, following methods of Paerl et al. (2014) and references therein. The relative contribution of phytoplankton classes to chl a was estimated using the software CHEMTAX (Mackey et al., 1996).

Statistical Analysis

To determine the photopigments suitable for use as markers of phytoplankton taxonomic groups, covariance between all photopigments measured by HPLC was determined using the corr( ) function in R. Linear regression analysis was conducted for those pigments with significant correlations (P < 0.05) to determine the typical ratios between these pigments in the samples and to provide a basis for selecting CHEMTAX pigment ratios.

The initial accessory pigment:chl a matrix for CHEMTAX (Supplementary Appendix 1A) was constructed based on published values, with preference given to ratios from studies on phytoplankton community structure from temperate U.S. Atlantic estuaries (e.g., Lewitus et al., 2005; Li et al., 2004), cultures of known dominant species in BB-LEH (e.g., Nannochloris atomus; Olsen, 1989), and those that approximated the ratios derived from linear regressions between photopigments calculated in the present study. Pigments used as chemotaxonomic markers of microalgae classes in CHEMTAX were chl a, chlorophyll c (c1+c2, chl c), chlorophyll b (chl b), fucoxanthin (fuco), alloxanthin (allo), peridinin (peri), zeaxanthin (zea), neoxanthin (neo), violaxanthin (viol), and lutein (lut).

From the initial CHEMTAX ratio matrix (Supplementary Appendix 1A), 60 pigment ratio tables were generated by multiplying each cell of the initial table by a scaled random number to adjust each ratio by ±50% (Mackey et al., 1997; Wright et al., 2009) and used as the starting point for a CHEMTAX run. From the 10% of CHEMTAX results with the smallest residual (n = 6) for each site/year, the one providing the best correlations with microscopically determined biovolume and resolution of chl b-containing classes was selected for all further analyses. For Harvey Cedars and Tuckerton Cove, where microscopy was not available, the initial ratio matrix from Sedge Is. was used. Separate CHEMTAX analyses were conducted for each site and year to minimize errors arising from different phytoplankton assemblages and environmental parameters.

Linear regression analyses of CHEMTAX estimates vs. microscopically derived cell biovolume and clam growth coefficients vs. phytoplankton-related parameters were conducted using the lm package in R. Only correlations between weekly clam growth rates and phytoplankton pigment metrics are reported in the present paper. A follow-up paper will present data on seasonal survival and condition index and also compare weekly growth rates in terms of both soft tissue DW and SL to determine the allocation between shell and tissue growth in relation to seston parameters, including particulate organic carbon and nitrogen (Bricelj et al., unpublished data).

Differences in mean class contribution to chl a estimated by CHEMTAX were determined between sites and study years by two-factor analysis of variance (ANOVA), followed by Tukey's a posteriori multiple comparison tests. For this analysis, sampling dates were grouped by site and trial, and any data without a corresponding sampling date (falling within 1 week) in the first or second year were omitted. Weekly percentage instantaneous growth coefficients (kDW) were compared within each trial/site using ANOVAs followed by Tukey's a posteriori multiple comparisons tests. Percentages were arcsine transformed to meet the assumptions of normality prior to conducting ANOVAs.

Cell counts of A. anophagefferens obtained using the immunofluorescence method were related to concentrations of 19′but by linear regression analysis. Based on immunofluorescence counts of A. anophagefferens and high concentrations of 19′but detected at Sedge Is. in 2013, a second CHEMTAX analysis was conducted for this site, which included a category for Aureococcus in the ratio matrix (Supplementary Appendix 1J,K).

Physical Variables: Temperature and Salinity

The two northern study sites contrasted greatly in their temperature and salinity regimes, reflecting the influence of the Toms River plume at IBSP and that of oceanic exchange via Barnegat Inlet at Sedge Is. Water temperature observed at Sedge Is. was consistently lower than at the other three study sites, especially during Trial I (2.2 and 4.1°C lower mean than at other sites in 2012 and 2013, respectively; Figure 2). During Trial II in both years, the temperature at Sedge Is. approached that of the other three sites but generally remained lower (by 1–2°C on average; Figure 2). High daily temperature fluctuations were also a consistent feature at Sedge Is. because of the proximity to Barnegat Inlet. The maximum 2-hour temperature variation at Sedge Is. was 10°C day−1 in 2012 and 16°C day−1 in 2013 (data not shown). Maximum summer temperatures remained below 29.7°C at all study sites, and temperature fell sharply (∼0.5°C day−1) during the first week of September.

Figure 2.

Water temperatures (daily means) as measured by HOBO® Temperature Loggers during the 2012 and 2013 study periods. Note that sampling in Trial II 2013 started ∼2 weeks later than in 2012.

Figure 2.

Water temperatures (daily means) as measured by HOBO® Temperature Loggers during the 2012 and 2013 study periods. Note that sampling in Trial II 2013 started ∼2 weeks later than in 2012.

Close modal

Mean salinities were highly consistent between the two years at all four study sites (Table 1), differing by ≤0.7 between 2012 and 2013. Overall, IBSP showed the lowest salinities and Sedge Is. the highest, with Tuckerton and Harvey Cedars showing intermediate salinities. In both years, salinity minima recorded at IBSP (19 and 16 in 2012 and 2013, respectively; Table 1) occurred following large precipitation events (more pronounced in 2013 when rainfall, as measured at Toms River, exceeded 7 cm over a period of up to 1 week [NJ Weather & Climate, 2017]). In 2013, relatively low salinities at IBSP were maintained for 2 weeks (12 June to 16 June) (Table 1).

Table 1.

Salinities measured at the four study sites during the 2012–13 study periods (northern and southern sites sampled on consecutive dates). Dashed lines indicate gaps in sampling period. ND = not determined.

Salinities measured at the four study sites during the 2012–13 study periods (northern and southern sites sampled on consecutive dates). Dashed lines indicate gaps in sampling period. ND = not determined.
Salinities measured at the four study sites during the 2012–13 study periods (northern and southern sites sampled on consecutive dates). Dashed lines indicate gaps in sampling period. ND = not determined.

Characterization of Phytoplankton Based on CHEMTAX Analysis

Mean chl a concentrations were generally highest at Tuckerton and IBSP (11.89 and 10.31 μg l−1, respectively; Figure 3), although these two sites were characterized by a very different phytoplankton composition (see the next section). They were lowest at Harvey Cedars and Sedge Is. (8.03 and 4.84 μg l−1, respectively; Figure 3). The highest chl a level of all study sites was measured at Tuckerton (29.31 μg l−1 in 2012 and 21.71 μg l−1 in 2013; Figure 3g,h), and the lowest was measured at Sedge Is. (0.53 μg l−1 in 2012 and 2.31 μg l−1 in 2013; Figure 3b,c). Weekly chl a at Sedge Is. was significantly lower than at Tuckerton in both years and IBSP in 2013 (P < 0.01). Overall, mean chl a concentration was lowest in Trial I 2012, with no significant differences among sites. Concentrations of chl a were generally higher in 2013 (Figure 3), although no significant differences between yearly means were observed at any of the four sites.

Figure 3.

Contribution of phytoplankton classes to total chl a, as estimated by CHEMTAX. The vertical line separate sampling periods (Trials I and II). Note the difference in vertical scales among sites, with lowest chl a concentrations at Sedge Is. and Harvey Cedars.

Figure 3.

Contribution of phytoplankton classes to total chl a, as estimated by CHEMTAX. The vertical line separate sampling periods (Trials I and II). Note the difference in vertical scales among sites, with lowest chl a concentrations at Sedge Is. and Harvey Cedars.

Close modal

Estimates of the individual phytoplankton group contribution to chl a (μg l−1) derived by CHEMTAX are shown in Figure 3. Residuals (root mean square error) for CHEMTAX chl a estimates ranged from 0.03 to 0.07 μg l−1. Final ratio matrices for CHEMTAX analysis at each site are shown in Supplementary Appendix 1(B–I).

Characteristics of the phytoplankton community at each site were generally consistent between the two study years. The summer assemblage was typically dominated by diatoms at Sedge Is., Harvey Cedars, and Tuckerton in both years (Figure 3c–h), with mean contributions of diatom chl a to total chl a ranging from 48.7 to 57.7%. At IBSP, mean contributions of diatom chl a were generally lower (30.3%) than at the other sites (Figure 3a,b). The IBSP site was characterized by a greater mean contribution of cyanobacteria (27.9%) and chlorophytes (20.9%) relative to other sites (range = 1.2–11.0 and 2.1–11.6%, respectively). Throughout the 2-year study period, mean cyanobacterial and chlorophyte contributions to chl a were significantly higher (P < 0.001) at IBSP than at the other sites, both in terms of absolute concentrations and percentage of chl a, except for percentage of chlorophytes at Sedge Is. during 2013 (Figure 3). Mean estimates of cryptophyte chl a were higher at the southern sites, Tuckerton, and Harvey Cedars (20.6 and 24.1%, respectively), than at the northern sites, IBSP and Sedge Is. (5.9 and 10.6%, respectively), in both years (Figure 3). Euglenoids, prasinophytes, and dinoflagellates made a relatively small contribution to chl a at all study sites, ranging from 1.6 to 7.2%, 2.8 to 11.7%, and 0.9 to 6.4%, respectively (Figure 3).

At IBSP, diatoms dominated total chl a in early June (68.4% in 2013 and 80.2% in 2013), when total chl a was relatively low at this site, transitioning to a chlorophyte- and cyanobacteria-dominated community later in the summer (Figure 3a,b). While the percentage of contribution of cyanobacteria was similar in both years at IBSP, mean cyanobacterial chl a (μg l−1) estimates were 70% greater in 2013 than in 2012 (P < 0.05), largely accounting for the increase in mean chl a from 2012 to 2013, especially in June and July (Figure 3a,b). In 2012, the maximum chl a concentration (in μg l−1) contributed by diatoms coincided with the chl a maximum observed on 5 September (Figure 3a). In 2013, the chl a maximum was also observed in early September but was less pronounced, with cyanobacteria and diatoms contributing about equally (Figure 3b).

At Sedge Is., mean chl a concentration was the lowest of all sites and comparable in 2012 and 2013 (4.2 and 5.8 μg l−1, respectively), although differences in the timing of chl a maxima were observed. In 2012, the mean chl a concentration during Trial I was also significantly lower (1.2 μg l−1) than in 2013 (7.1 μg l−1) (Figure 3c,d). In both years, the chl a maxima co-occurred with maximum estimates of diatom chl a (μg l−1), although in 2013 this peak was largely attributed to A. anophagefferens (see below).

At Harvey Cedars, the chl a maximum was observed during July in both years and co-occurred with the greatest relative and absolute diatom concentrations (Figure 3e,f). In 2013, mean chl a concentration was 60% greater than in 2012, largely attributable to higher concentrations in July–September and a spike in cryptophytes during mid-June. During the 2012 chl a peak, diatoms and cryptophytes contributed the bulk of chl a (84.8%; Figure 3e). The chl a maximum in 2013 occurred nearly a month earlier than in 2012, although chl a concentrations remained relatively high (>8 μg l−1) throughout most of the summer (Figure 3f). Cryptophyte contribution to chl a was markedly higher in 2013 than in 2012, particularly during Trial I.

At Tuckerton, mean chl a concentrations were comparable in both years (10.9 and 13.1 μg l−1 in 2012 and 2013, respectively), although concentrations were consistently higher in 2013, especially in early summer. In 2012, the chl a maximum was more pronounced and comprised a greater contribution of diatoms (56.4%) than in 2013 (37.0%), as cryptophytes made a larger contribution to chl a in this year (Figure 3g,h).

Phytoplankton Species Composition

Despite their proximity (∼15 km; Figure 1), the IBSP and Sedge Is. sites differed markedly in their phytoplankton assemblage, as shown by both photopigment (Figure 3) and microscopic (Figure 4) analyses. The cell concentrations of the most common phytoplankton species of each class at these two sites are shown in Table 2. Only species present >50% (diatoms) or >30% (all other classes) of sampling dates are included, although other less frequently occurring species occasionally dominated their respective class. Diatom species richness was higher at Sedge Is. than at IBSP (Table 2), and a larger mean number of diatom species was reported at Sedge Is. (11.9 and 14.5 in 2012 and 2013, respectively) than at IBSP (7 and 8 in 2012 and 2013, respectively). At IBSP, species richness was greatest in early to mid-June and decreased throughout the season as picoplankton became more abundant, but richness remained relatively stable throughout the study period at Sedge Is. (Table 2; Figure 3a–d; Figure 4a,b).

Figure 4.

Cell biovolume concentrations of phytoplankton classes, as determined microscopically. Vertical line separates Trials I and II.

Figure 4.

Cell biovolume concentrations of phytoplankton classes, as determined microscopically. Vertical line separates Trials I and II.

Close modal
Table 2.

Cell concentrations (cells l−1 on a log scale) of frequent (≥50% and ≥30% of sampling dates per season for diatoms and all other classes, respectively) phytoplankton species at Sedge Island (S) and IBSP (I) determined microscopically for the 2012–13 study periods. Cell volume is reported as the mean of the calculated cell volume for all dates when the species was recorded (see “Methods”). Gray area indicates missing sampling dates.

Cell concentrations (cells l−1 on a log scale) of frequent (≥50% and ≥30% of sampling dates per season for diatoms and all other classes, respectively) phytoplankton species at Sedge Island (S) and IBSP (I) determined microscopically for the 2012–13 study periods. Cell volume is reported as the mean of the calculated cell volume for all dates when the species was recorded (see “Methods”). Gray area indicates missing sampling dates.
Cell concentrations (cells l−1 on a log scale) of frequent (≥50% and ≥30% of sampling dates per season for diatoms and all other classes, respectively) phytoplankton species at Sedge Island (S) and IBSP (I) determined microscopically for the 2012–13 study periods. Cell volume is reported as the mean of the calculated cell volume for all dates when the species was recorded (see “Methods”). Gray area indicates missing sampling dates.

At both sites, the most common diatom species were small, cylindrical forms (cell volume <500 μm3; Table 2). Large diatom species, however, including Dactyliosolen fragilissimus, Coscinodiscus concinnus, Pleurosigma salinarium, Guinardia spp., and Eucampia zodiacus were more frequently observed at moderate cell concentrations (∼104 cells l−1) at Sedge Is. than IBSP (Table 2). Skeletonema costatum, a cosmopolitan neritic species, also occurred more frequently and at higher cell concentrations at Sedge Is. (Table 2). The highly euryhaline Cyclotella choctawatchea (Prasad, Nienow, and Livingston, 1990) was the most common diatom species in both years (Table 2), occurring at all but one sampling date, and contributing on average 29.9 and 16.5% of total diatom biovolume at IBSP and Sedge Is., respectively.

At Sedge Is., diatom species in the 100–1000 μm3 cell volume range were most abundant, generally accounting for ≥50% of numerical diatom abundance in 2012 (Table 2). Large (>10,000 μm3), intermittently occurring species contributed ≥60% of diatom volume on nearly all dates, although they never exceeded 10% of total diatom cell concentration (Table 2). In 2013, diatoms generally comprised smaller species, and diatom species in the 10–1000 μm3 cell volume range accounted for >95% of the diatom community on all dates (Table 2). At IBSP, diatom species ranging in cell volume from 10 to 1000 μm3 always accounted for the majority of diatom cells (Table 2). Species in the 100–1000 μm3 volume range typically contributed most of the diatom biovolume in 2012, except in late June and early July, when Conscinodiscus concinnus and Odontella aurita (>10,000 μm3) contributed >60% of both diatom and total biovolume. In 2013, cells in the 10–100 μm3 range generally accounted for >70% of all diatom cells (Table 2), and species ranging from 100 to 1000 μm3 contributed >50% of biovolume, except on 5 June, when D. fragilissimus (cell volume = 5173 μm3) comprised 80% of diatom biovolume. During both years, the species making the greatest percentage contribution of both diatom biovolume and cell concentration at IBSP were typically the small forms C. choctawatchea (153 μm3), Phaeodactylum tricornutum (23 μm3), or Minutocellus spp. (∼28 μm3; Table 2).

Dinoflagellate species composition was variable between years but generally comparable between sites. In 2012, the most common species were Gyrodinium estuariale and Prorocentrum minimum (Table 2). In 2013, some species observed at Sedge Is. were absent at IBSP, including Scrippsiella trochoidea, Prorocentrum micans, and Protoperidinium spp. (Table 2). Abundant species present in 2012 but absent in 2013 included P. minimum and Gyrodinium aureolum (Table 2).

Cryptophyte species composition was similar between sites and years, comprising five common species. These species were observed throughout the sampling period, with typical concentrations of 105 cells l−1, the most common being Hemiselmis virescens, Plagioselmis sp., and Leucocryptos marina (Table 2). The chrysophyte Paulinella ovalis (Syn. Calycomonas ovalis) was observed at both sites in late summer 2012 and at all sampling dates in 2013, with higher concentrations at IBSP than at Sedge Is. (Table 2).

Pico-coccoids (≤3 μm diameter) were not identified at the species level. This group occurred at all sampling dates in 2013 and most sampling dates in 2012 (Table 2). The pico-coccoid cell concentration was typically lower at Sedge Is. than at IBSP (Table 2). At IBSP, pico-coccoids contributed most (up to 93%) of total phytoplankton biovolume from late June until late August in both years, whereas at Sedge Is. they contributed <15% of total biovolume in 2012 but up to 56% in 2013 (Figure 4).

Few cyanobacterial species were identified over the study period at either of these two sites. The most common species was Aphanocapsa sp. (Table 2). Mean cyanobacterial contribution to total biovolume ranged from 0.1 to 6% at IBSP and Sedge Is., respectively (Figure 4). Considering the high CHEMTAX estimates of cyanobacterial chl a at IBSP (Figure 3a,b), it is likely that the most common species of this class were unidentified and included in the pico-coccoid group, which frequently accounted for most of the observed phytoplankton biomass at both sites (Figure 4; see “Discussion”).

Phytoplankton species data collected from the NJDEP BB-12 monitoring station in LEH bay, located ∼7 km NE of the Tuckerton Cove study site (Figure 1; Ren, 2013; Ren et al., unpublished data), were used as a proxy for the species composition at Tuckerton in 2012. Generally, the species composition at this site was more like that of Sedge Is. than IBSP. Cyclotella choctawatchea, the most common diatom at the northern sites, was not observed in the 2012 BB-12 samples. Minutocellus scriptus, which were especially abundant at IBSP and Sedge Is. during the latter half of the sampling season, were also not observed at this site. Overall, small diatoms and pico-coccoids were less abundant at BB-12 than at the two northern sites and were observed only in May. The most common dinoflagellate species at IBSP and Sedge Is. were also identified at BB-12, except for S. trochoidea and P. minimum. The highest number of dinoflagellate species, cell, and biovolume concentrations at BB-12 were observed in late August and early September. The euglenoid Eutreptiella sp. was much more abundant at this site than at either IBSP or Sedge Is. during the sampling period, typically occurring at a concentration of 104 l−1. Cryptophyte species composition at BB-12 was similar to that at IBSP and Sedge Is., with Plagioselmis sp. being the most common, but at typical cell concentrations one to two orders of magnitude greater than at the northern sites (Table 2).

Harmful Algal Species

Several dinoflagellate species known to be harmful to bivalve mollusks were found at the two northern BB study sites. Scripsiella trochoidea, known to exert toxic effects on oyster and hard-clam larvae (Tang and Gobler, 2012), was found at both study sites but was more frequent at Sedge Is. than IBSP. It was occasionally the most abundant dinoflagellate by volume (Table 2) but was not associated with a reduction in growth of hard-clam juveniles at this site. The naked dinoflagellate Akashiwo sanguinea, which can clog shellfish gills (Jessup et al., 2009), was found at a concentration of 103 cells l−1 at both Sedge Is. and IBSP on 5 June 2012. Prorocentrum minimum, known to cause mortalities of juvenile bay scallops, A. irradians, and oysters, C. virginica, possibly because of interference with digestive processes (Wikfors, 2005), was the most numerically abundant dinoflagellate species at Sedge Is. and IBSP in 2012, with peak concentrations of ∼105 cells l−1 in early June 2012 (Table 2). Prorocentrum minimum was also found to be the dominant phytoplankton species in winter 2011–12 at northern BB sites, attaining a maximum density of 106 cells l−1 (Ren, 2013). Two Pseudo-nitzschia sp., known producers of domoic acid in other systems (Trainer et al., 2012), P. delicatissima and P. seriata, were observed at IBSP and Sedge Is., attaining a maximum concentration of 105 cells l−1 in September 2012. These species did not occur frequently, however, and their toxicity in BB-LEH is unknown.

The presence of A. anophagefferens was confirmed in the BB-LEH estuary in both years, with peak cell concentrations of 9.2 × 104 cells ml−1 in 2012 at Harvey Cedars (Figure 5a) and 4.4 × 105 cells ml−1 in 2013 at Sedge Is. (Figure 5b). In both years, concentrations peaked in June and remained relatively low through the latter part of the summer, except for a secondary peak at Tuckerton in late July 2012 (Figure 5a,b). Despite high cell densities at Sedge Is. in 2013, A. anophagefferens was not observed at bloom levels known to inhibit feeding rates of hard-clam juveniles (>3.5 × 104 cells ml−1; Bricelj, MacQuarrie, and Schaffner, 2001) at Tuckerton and Harvey Cedars during the same time frame (Figure 5b). The maximum cell density, recorded at Sedge Is. (Figure 5a), coincided with the highest concentration of the pigment 19′but (1.96 μg l−1) measured during the study period, and A. anophagefferens cell concentrations were highly correlated with 19′but (R2 = 0.9688; Figure 5c). At Tuckerton, where A. anophagefferens was detected by immunofluorescence on all sampling dates, at least at low levels (Figure 5a,b), 19′but was never detected by HPLC. While A. anophagefferens was not measured by immunofluorescence at IBSP during this study or at Sedge Is. (2012), the detection of 19′but at these sites likely indicates its presence. At Sedge Is., 19′but was detected at levels corresponding to a maximum of 3.6 × 104 (based on the equation in Figure 5c) cells ml−1 on 15 August 2012. At IBSP, 19′but was detected at levels corresponding to ∼105 cells ml−1 (based on the equation in Figure 5c) on 5–12 June and again on 3 July 2013.

Figure 5.

Aureococcus anophagefferens cell concentrations, as measured by immunofluorescence for Harvey Cedars, Tuckerton, and Sedge Is. during the 2012–13 study periods (a), (b). Cell concentrations were not measured at Sedge Is. (2012) or IBSP (both years). (c) Fitted linear regression of the marker pigment 19′but vs. A. anophagefferens cell concentration for combined 2012–13 data; R2: coefficient of determination; (d) contribution of phytoplankton classes to total chl a, as estimated by CHEMTAX, after the inclusion of A. anophagefferens in the ratio matrix for Sedge Is. 2013. Vertical line separates Trials I and II.

Figure 5.

Aureococcus anophagefferens cell concentrations, as measured by immunofluorescence for Harvey Cedars, Tuckerton, and Sedge Is. during the 2012–13 study periods (a), (b). Cell concentrations were not measured at Sedge Is. (2012) or IBSP (both years). (c) Fitted linear regression of the marker pigment 19′but vs. A. anophagefferens cell concentration for combined 2012–13 data; R2: coefficient of determination; (d) contribution of phytoplankton classes to total chl a, as estimated by CHEMTAX, after the inclusion of A. anophagefferens in the ratio matrix for Sedge Is. 2013. Vertical line separates Trials I and II.

Close modal

To estimate the potential contribution of A. anophagefferens to chl a from cell counts, a constant chl a concentration of 0.03 pg cell−1 was assumed, based on prior studies (Boneillo and Mulholland, 2014; Gobler and Sañudo-Wilhelmy, 2001; Gobler, Renaghan, and Buck, 2002; Mulholland, Gobler, and Lee, 2002). These estimates of chl a, and pigment:chl a ratios (converted from moles to mass units) reported in Alami, Lazar, and Green (2012), were used to estimate this alga's contributions to 19′but and fucoxanthin concentrations. Concentrations of 19′but calculated from A. anophagefferens cell concentrations in this way resulted in a nearly 1:1 relationship with those determined by HPLC (slope = 0.957, R2 = 0.9779; no Tuckerton data were included in the regression [not shown] because of the lack of detectable 19′but at this site). Based on these calculations, A. anophagefferens would have contributed 13.23 μg chl a l−1 (77%) and 49% of the fucoxanthin at Sedge Is. on June 19 (total chl a = 17.15 μg l−1), when maxima of both pigments and a relatively low diatom biovolume concentration was recorded (Figure 4b). When A. anophagefferens concentration peaked at Harvey Cedars in 2012 (Figure 5a), it is estimated that it would have contributed nearly 50% of both chl a and fucoxanthin, although these same calculations may not apply in 2012 (see “Discussion”).

Given the robust relationship between calculated and measured 19′but concentrations, a second CHEMTAX analysis, including A. anophagefferens in the initial ratio matrix, was conducted for Sedge Is. in 2013 (Supplementary Appendix 1J). Inclusion of this alga in the CHEMTAX analysis altered the final pigment ratios for other classes only slightly when compared to the original analysis (Supplementary Appendix 1E,K). At peak bloom levels observed at Sedge Is. on 19 June, CHEMTAX estimates of A. anophagefferens chl a were 10.4 μg l−1, thus comprising 60% of total phytoplankton chl a (Figure 5d).

Relationship between CHEMTAX-Derived Chl a and Cell Biovolume

Validation of CHEMTAX results and identification of discrepancies between pigment and microscopic analysis was based on a linear relationship between CHEMTAX estimates of class chl a and microscopically determined biovolume, both measures of algal biomass (Table 3). This analysis was conducted only at IBSP and Sedge Is., as parallel samples for microscopy were not taken at the southern sites. The relationship between total chl a and biovolume was only significant at Sedge Is. in 2012 (P < 0.05, Table 3). Slopes of significant regressions between estimated class chl a and biovolume differed by up to an order or magnitude among sites and between years (Table 3), potentially reflecting the somewhat inconsistent relationship between phytoplankton carbon and pigment concentrations. Despite this, relationships were generally significant for cryptophytes (both sites/years), diatoms (Sedge Is. both years and IBSP in 2013), and for species containing chl b (all groups combined, both sites/years) (Table 3). The relationship between dinoflagellate chl a and biovolume was not significant, potentially because of low biovolume and chl a estimates for this group, variations in class species composition, and/or the wide range of pigment profiles characteristic of this class (Wright, 2005; Zapata et al., 2012). Regressions between chl a and biovolume for individual chl b-containing classes were generally not significant, except for euglenophytes (IBSP 2012) and prasinophytes (IBSP 2013), potentially because of the lack of identification of small chlorophytes at the species level or poor resolution between these groups because of shared pigment markers. It was assumed that a large fraction of the group identified as “pico-coccoids” (Figure 4) comprised chlorophytes for regression analyses (see “Discussion”). When pico-coccoids were included in biovolume calculations for chlorophytes, significant relationships were observed at IBSP and Sedge Is. in both years (Table 3). No regressions were fitted for euglenophytes in 2013, as this group was not recorded from microscopic analysis. The relationship between cyanobacterial CHEMTAX chl a estimates and biovolume was generally not significant, except at IBSP in 2013 (Table 3), when Aphanocapsa sp. was present on all sampling dates (Table 2). Much of the cyanobacterial biovolume present may have been included in the pico-coccoid group (see “Discussion”).

Table 3.

Results from linear regressions of phytoplankton class chl a concentration predicted by CHEMTAX vs. class biovolume determined microscopically at Sedge Is. and IBSP study sites. Results are shown only for parameters with significant (slope ≠ 0) relationships for at least one site/sampling year.

Results from linear regressions of phytoplankton class chl a concentration predicted by CHEMTAX vs. class biovolume determined microscopically at Sedge Is. and IBSP study sites. Results are shown only for parameters with significant (slope ≠ 0) relationships for at least one site/sampling year.
Results from linear regressions of phytoplankton class chl a concentration predicted by CHEMTAX vs. class biovolume determined microscopically at Sedge Is. and IBSP study sites. Results are shown only for parameters with significant (slope ≠ 0) relationships for at least one site/sampling year.

In some cases, relationships between CHEMTAX chl a estimates and biovolume were significant only after the removal of putative outliers (Table 3). For example, on 12 June 2013 the highest diatom biovolume (Figure 4), primarily comprising D. fragilissimus, was determined at IBSP (Table 2) despite a relatively low diatom chl a estimate (Figure 3b). At Sedge Is. on 19 June 2013, CHEMTAX estimates of diatom chl a peaked at this site (Figure 3d), despite a relatively low biovolume (Figure 4), likely because of the high concentration of A. anophagefferens present (see “Discussion”).

Tidal Effect on Phytoplankton Community Structure

Weekly sampling in the present study was unable to capture short-term temporal variability in phytoplankton characteristics and, for logistic reasons, was not designed to occur at the same stage of the tidal cycle. Effects of tidal stage on phytoplankton composition, however, were documented at IBSP in 2012, where water-column sampling occurred over a wide range of tidal conditions. Tidal stage clearly affected the relative contribution of diatoms and cyanobacteria to total phytoplankton biomass (Figure 6). These two phytoplankton groups showed a reciprocal trend in their relative contribution to total chl a in 2012, where the fuco:chl a ratio, indicative of the percentage of diatoms, was generally greater at high tide and the zea:chl a ratio, indicative of the percentage of cyanobacteria, was greater at low tide. The same pattern was not observed in 2013 or at the other sampling sites, presumably because consecutive samplings did not represent a wide range of tidal stages and other sites were not located near a major freshwater source.

Figure 6.

Weekly fluctuation in (above) Zeaxanthin:chl a and (below) Fucoxanthin:chl a in relation to tidal height above mean lower low water (MLLW) at Island Beach State Park (2012), calculated from NOAA tide predictions for Seaside Park Marina. Error bars indicate the standard error (SE). Dashed gray line indicates >20-day gap in sampling between trials. Vertical arrows indicate anomalies in the photopigment pattern in relation to tidal height.

Figure 6.

Weekly fluctuation in (above) Zeaxanthin:chl a and (below) Fucoxanthin:chl a in relation to tidal height above mean lower low water (MLLW) at Island Beach State Park (2012), calculated from NOAA tide predictions for Seaside Park Marina. Error bars indicate the standard error (SE). Dashed gray line indicates >20-day gap in sampling between trials. Vertical arrows indicate anomalies in the photopigment pattern in relation to tidal height.

Close modal

Growth of Juvenile Hard Clams

Overall, highly significant differences in juvenile clam growth rates (as measured by the instantaneous growth coefficient, kDW) were found among the four study sites and among weeks (two-way ANOVA, P < 0.001). A highly significant week × site interaction (P < 0.001) also occurred. Significant differences in kDW were found at IBSP in Trial I 2012 (P < 0.001), where the highest kDW (9.5% day−1) of the entire study was observed in early June, then tapering off to relatively low and eventually negative soft-tissue growth (Figure 7a). No significant difference between weekly kDW was observed at Sedge Is. during this same period (Figure 7c). Growth results from Trial I 2012 are not shown for Tuckerton and Harvey Cedars, as clams were deployed in smaller sized mesh bags; results are thus not directly comparable to other trials in this study. In Trial II 2012, significant differences in weekly kDW were found at all sites (Figure 7).

Figure 7.

Weekly growth rates in dry weight (DW) of soft tissues, kDW (mean instantaneous growth coefficient or percentage increase or decrease per day ± standard error, see “Methods”) over the 2-year study period for Trials I and II (data for Trial I 2012 were only available for IBSP and Sedge Is.). Asterisks indicate an overall significant difference in kDW values within each trial/site (one-way ANOVAs); different letters identify significant differences among weeks determined from Tukey's a posteriori multiple comparisons tests. Gray arrow in (h) marks growth cessation attributable to heavy infestation by solitary tunicates (presumably Molgula spp.); horizontal gray bar in (b) marks the reduction in growth attributed to low salinities at this site (see text).

Figure 7.

Weekly growth rates in dry weight (DW) of soft tissues, kDW (mean instantaneous growth coefficient or percentage increase or decrease per day ± standard error, see “Methods”) over the 2-year study period for Trials I and II (data for Trial I 2012 were only available for IBSP and Sedge Is.). Asterisks indicate an overall significant difference in kDW values within each trial/site (one-way ANOVAs); different letters identify significant differences among weeks determined from Tukey's a posteriori multiple comparisons tests. Gray arrow in (h) marks growth cessation attributable to heavy infestation by solitary tunicates (presumably Molgula spp.); horizontal gray bar in (b) marks the reduction in growth attributed to low salinities at this site (see text).

Close modal

In 2013, significant differences in weekly kDW were found at IBSP and Tuckerton during Trial I and among all sites during Trial II (Figure 7). Weekly clam growth rates at Tuckerton were relatively high throughout the 2013 study period, except for 1 week (27 August to 3 September ) during Trial II (Figure 7h), when complete cessation of growth coincided with heavy fouling by solitary tunicates. Removal of fouling from cages at this site resulted in immediate resumption of clam growth, and no recolonization of tunicates was observed on the next sampling date. Low growth rates were also observed at Harvey Cedars during this same period (Figure 7f), although no such fouling was observed at this site. Low growth rates at IBSP during Trial 1 2013 (kDW < 1% day−1 throughout June; Figure 7b) are likely associated with the low salinities (minimum of 16 in 2013) occurring at this time (Table 1).

Ranking of the absolute shell growth rate (in μm day−1) among the study sites during each trial is shown in Table 4, as this parameter can be compared more readily with juvenile M. mercenaria growth data reported in the literature. These growth patterns were generally comparable to those in relative shell growth rates (kSL) (not shown) given the relatively narrow size range within each trial. This integrated shell growth rate per trial ranked the lowest or second lowest for clams at IBSP, with the minimum documented in this study (only 3 μm day−1) in early summer 2013, and was the lowest in two out of three trials at Harvey Cedars (Table 4). In contrast, clams at Tuckerton were characterized by the highest growth during both trials conducted in 2013 and second highest during midsummer 2012 (Table 4). Ranking of growth rates was less consistent among trials at Sedge Is. where clams exhibited the highest growth rate (176 μm day−1) measured in the entire study during Trial II 2012 but the second lowest in Trial I 2013 (Table 4).

Table 4.

Ranking of shell growth rates (in μm day−1) of Mercenaria mercenaria juveniles integrated over each experimental trial among the four study sites. Means of three cages (n = 50 clams per cage for final samples) ± standard error, SE). Overall site differences within each trial were significant at p < 0.001 (ANOVA). Different superscript letters indicate significant differences determined by a posteriori Tukey's multiple comparison tests. Initial mean clam shell length ± SE were 8.99 mm (±0.08, n = 100) and 8.81 mm (±0.09, n = 100) for Trials I and II 2013, respectively, and 9.05 mm (±0.11, n = 50) for Trial II 2012.

Ranking of shell growth rates (in μm day−1) of Mercenaria mercenaria juveniles integrated over each experimental trial among the four study sites. Means of three cages (n = 50 clams per cage for final samples) ± standard error, SE). Overall site differences within each trial were significant at p < 0.001 (ANOVA). Different superscript letters indicate significant differences determined by a posteriori Tukey's multiple comparison tests. Initial mean clam shell length ± SE were 8.99 mm (±0.08, n = 100) and 8.81 mm (±0.09, n = 100) for Trials I and II 2013, respectively, and 9.05 mm (±0.11, n = 50) for Trial II 2012.
Ranking of shell growth rates (in μm day−1) of Mercenaria mercenaria juveniles integrated over each experimental trial among the four study sites. Means of three cages (n = 50 clams per cage for final samples) ± standard error, SE). Overall site differences within each trial were significant at p < 0.001 (ANOVA). Different superscript letters indicate significant differences determined by a posteriori Tukey's multiple comparison tests. Initial mean clam shell length ± SE were 8.99 mm (±0.08, n = 100) and 8.81 mm (±0.09, n = 100) for Trials I and II 2013, respectively, and 9.05 mm (±0.11, n = 50) for Trial II 2012.

Relationship between Clam Growth and Phytoplankton Metrics

Weekly growth rates (kDW) of juvenile M. mercenaria in the two study years were generally greatest at the two sites (Sedge Is. and Tuckerton) (Figure 7), where phytoplankton was characterized by a greater contribution of diatoms to chl a relative to chlorophytes and cyanobacteria (Figure 3). Integrated shell growth rate of clams ranked the lowest or third lowest at IBSP in all trials conducted over the 2-year study (Table 5), a site characterized by high picoplankton concentrations and episodic low salinity.

Table 5.

Results from linear regressions of juvenile clam-growth coefficients based on soft tissue dry weight (kDW) vs. CHEMTAX results (μg chl a l−1 by class) and diagnostic photopigments (μg pigment l−1). Picoplankton indicates combined estimates of chlorophytes and cyanobacteria. Results are shown only for parameters with significant relationships for at least one site/sampling year.

Results from linear regressions of juvenile clam-growth coefficients based on soft tissue dry weight (kDW) vs. CHEMTAX results (μg chl a l−1 by class) and diagnostic photopigments (μg pigment l−1). Picoplankton indicates combined estimates of chlorophytes and cyanobacteria. Results are shown only for parameters with significant relationships for at least one site/sampling year.
Results from linear regressions of juvenile clam-growth coefficients based on soft tissue dry weight (kDW) vs. CHEMTAX results (μg chl a l−1 by class) and diagnostic photopigments (μg pigment l−1). Picoplankton indicates combined estimates of chlorophytes and cyanobacteria. Results are shown only for parameters with significant relationships for at least one site/sampling year.

Results of linear regression analyses based on juvenile growth rates and photopigment-derived measures of phytoplankton community structure are summarized in Table 5. Generally, significant positive relationships were found between kDW and indicators of diatoms (Diatom chl a, percentage Diatom chl a, fuco, and fuco:chl a). In turn, significant negative relationships were found between kDW and indicators of chlorophytes (chl a, percentage chl a, chl b, and chl b:chl a) and cyanobacteria (chl a, percentage chl a, zea, and zea:chl a; Table 5).

At IBSP, indicators of the relative contribution of diatoms (fuco:chl a and percentage diatoms) showed a positive relationship with kDW, while indicators of chlorophytes and cyanobacteria (picoplankton chl a, percentage picoplankton, and zea) showed a negative relationship with kDW in 2012 (Table 5a). At Sedge Is., positive relationships were determined between kDW and percentage of diatoms (2012 and combined), diatom chl a (2013 and combined), and fuco (2013) (Table 5b). Negative relationships were found between kDW and percentage of cyanobacteria, chl b:chl a, and zea:chl a (2012 and combined; Table 5b). At Tuckerton, positive relationships were detected between percentage of diatoms (2012) and fuco:chl a (2012, 2013, and combined) (Table 5d). The only sites that showed any significant relationship between kDW and chl a were Sedge Is. (2013, Table 5b) and Harvey Cedars (combined, Table 5c), sites generally dominated by diatoms.

A strong negative relationship between kDW and chlorophytes (chlorophytes, picoplankton, percentage chlorophytes, percentage picoplankton; Table 5c) was observed at Harvey Cedars in 2012. It is likely that some other factor affected growth, however, as chlorophyte chl a was very low during the period where minimum growth was observed (30 August–11 September; Figure 7). It is unclear what contributed to low growth during this period. Positive relationships between kDW and cyanobacteria, chl b, and zea were also observed when both years were combined (Table 5c). This is likely attributable to the higher concentrations of chl b and zeaxanthin observed in 2013, when growth rates were generally higher at this site (Figure 7). Similarly, at Tuckerton, positive relationships between kDW and cyanobacteria and zeaxanthin were found when both years were combined, despite the low contribution of this phytoplankton group to chl a (Table 5). This is likely attributable to slightly higher mean cyanobacterial concentrations in 2013 (Figure 3h), when kDW was generally higher (Figure 7). Positive relationships between kDW and cyanobacteria, picoplankton, chl b, and zea were observed at Sedge Is. in 2013 (Table 5b). High concentrations of these classes/pigments were found primarily in Trial I (Figure 3d), when size-specific mortality may have positively skewed growth rates (see “Discussion”).

Despite the occurrence of brown tide levels that can negatively affect growth of clams in both years (Figure 5a,b), no relationship was detected between A. anophagefferens cell concentration and hard-clam growth (Table 4). Yet kDW at Harvey Cedars and Tuckerton approached zero or was negative from 6–19 June 2012 (not shown), following peak levels of A. anophagefferens. In 2013, no growth inhibition was observed during the period of peak A. anophagefferens concentrations at Sedge Is. (Figure 7d). High mortality rates (up to 43%; unpublished data) were observed during this period, although it is unclear whether they can be attributed to A. anophagefferens (see “Discussion”). Clam growth was near zero at IBSP from 4–26 June (Figure 7b) during the period when the highest 19′but concentrations were observed, but the presence of A. anophagefferens was not determined by immunofluorescence at this site, and growth inhibition also coincided with the period of low salinities during early summer 2013 at this site (Table 2).

Phytoplankton Characterization by CHEMTAX Analysis including Brown Tide

Analysis of photopigments by CHMETAX was useful in identifying major temporal trends and differences in community structure between the sites selected in this study. Based on comparison with microscopy, where available, CHEMTAX analysis was also able to distinguish between groups with shared photopigment markers, as well as inform the taxonomy of small forms difficult to distinguish by microscopy. In future studies, using microscopy to ground-truth photopigment analysis at a wider variety of sites may improve estimates, as well as using local isolates to better inform photopigment matrices for CHEMTAX analysis.

As the present study involved only the summer phytoplankton composition in BB-LEH and, based on other studies (Ren, 2013, 2015), the seasonal shifts in dominant phytoplankton classes had already occurred, the photopigment analysis of fall–spring the phytoplankton community in BB may require a different set of ratio matrices, depending on the species present.

Dinoflagellate species without peridinin as the main accessory pigment may not be represented in the CHEMTAX analysis conducted in the present study, as indicated by the poor correlation of dinoflagellate chl a with its microscopically determined biovolume. The most abundant and frequently occurring dinoflagellates (Table 2), however, were peridinin-containing species (Zapata et al., 2012). Inclusion of pigment ratios for dinoflagellates containing fucoxanthin derivatives did not improve overall CHEMTAX residuals or linear regressions with biovolume (not shown). It is more likely that the poor agreement in results obtained between methods is attributable to their low abundance relative to other classes and not attributable to misrepresentation in the ratio matrices. It is possible that some dinoflagellate species present are represented in the diatom or Aureococcus chl a fractions because of the presence of fucoxanthin in some species.

The pigment 19′but, indicative of A. anophagefferens, was not detected at Tuckerton in either study year, despite the detection of the pigment at both Harvey Cedars and Sedge Is. on dates when A. anophagefferens concentrations were lower (minimum = 8926 cells ml−1) than those at Tuckerton. Because of the relatively low concentrations of A. anophagefferens at this site and lower volumes filtered (attributable to high seston levels at this site), 19′but concentrations may have been below the detection limit of the HPLC method used (∼0.02 μg l−1). Alternatively, it is possible that cellular concentrations of 19′but were down regulated in response to low light (Alami, Lazar, and Green, 2012) or by greater utilization of dissolved organic carbon by A. anophagefferens (Gobler et al., 2011; Gobler and Sunda, 2012) and a lower overall requirement for light-harvesting pigments.

Pigment ratios for A. anophagefferens may have differed between sites and study years. Despite the robust relationship between 19′but calculated from A. anophagefferens cell counts and that determined by HPLC, 19′but was detected concomitant with A. anophagefferens on only one date in 2012, at Harvey Cedars, and at a lower pigment:cell ratio than was typically measured in 2013. Pigment ratios in A. anophagefferens are known to vary based on light conditions (Alami, Lazar, and Green, 2012) and between ecosystems. Trice et al. (2004) also described a positive linear relationship between A. anophagefferens cell counts and 19′but concentrations for coastal bays of Maryland and Virginia, resulting in a slope an order of magnitude greater than that obtained in the present study (Figure 5c). Note that the absence of 19′but based on HPLC analysis may not necessarily indicate the absence of A. anophagefferens cells. The present study shows, however, that 19′but concentrations are very useful to screen out samples that merit further confirmatory analysis of A. anophagefferens by immunofluorescence. Additional work is necessary to accurately predict A. anophagefferens cell concentrations from water-column concentrations of 19′but in the BB-LEH estuary.

Overall, A. anophagefferens, was present in the BB-LEH ecosystem during 2012 and 2013 at levels that, based on laboratory (Bricelj, MacQuarrrie, and Smolowitz, 2004) and field studies in other coastal lagoons (Wazniak and Glibert, 2004), can inhibit production of hard-clam larvae and juveniles. Surprisingly, the highest concentration, ∼440,000 cells ml−1, was measured at Sedge Is., within the MCZ, an area highly influenced by oceanic exchange via the BB Inlet and characterized by low turbidity (<10 mg l−1 total suspended solids [TSS]). In contrast, bloom levels of A. anophagefferens were not detected in Tuckerton where it historically attained the highest concentrations in earlier studies (Mahoney, Olsen, and Jeffress, 2006; Pecchioli, Lathrop, and Haag, 2006). The present study thus provides the first confirmation of brown tide within the MCZ, as early studies in the late 1990s and in 2011–13 did not collect samples in this body of water. Although the combined spatial coverage of sampling in 2012–13 from this study and Ren (2013, 2015) was limited, brown tide events were reported in both studies, indicating that A. anophagefferens continues to pose a risk within this ecosystem.

Taxonomy of the Pico-occoid Group

In the present study, pico-coccoids (<3-μm diameter) made a large contribution to phytoplankton biovolume, primarily at IBSP (Figure 5), but were not characterized taxonomically. Based on a comparison between photopigment and immunofluorescence data, however, it is likely that this group primarily comprised chlorophytes, cyanobacteria, and A. anophagefferens.

At Sedge Is., the peak in pico-coccoid and A. anophagefferens cell concentrations co-occurred on 19 June 2013; at IBSP, where peak 19′-but levels were observed on 12 June 2013, pico-coccoids were detected at cell concentrations comparable to Sedge Is., in parallel with relatively low levels of zeaxanthin and chl b, indicative of chlorophyte and cyanobacterial biomass. A peak in pico-coccoid concentration at BB12 (adjacent to the Tuckerton site) was observed in mid-late May 2012, just prior to the beginning of this study, and was comparable to A. anophagefferens cell counts determined by immunofluorescence in early June (when corresponding microscopy data for BB12 were not available). These associations suggest a dominance of A. anophagefferens in the picoplankton during early summer at these sites.

Chlorophytes were rarely identified microscopically at IBSP and Sedge Is., but pico-coccoid biovolume and chl b exhibited a strong positive correlation. Ratios of lutein:chl b at both sites also paralleled patterns in pico-coccoid biovolume and generally exceeded those typical for euglenoids or prasinophytes. These associations strongly suggest that a large portion of this group comprises small chlorophytes, such as N. atomus, previously shown to be dominant in the BB-LEH estuary in the summer (Olsen and Mahoney, 2001). A strong positive correlation was also observed between the zeaxanthin concentration and that of pico-coccoids at Sedge Is. and IBSP, and high pico-coccoid biovolumes occasionally co-occurred with peak zeaxanthin levels and relatively low concentrations of chl b, suggesting some contribution of cyanobacteria to the group as well.

Potential Physico-chemical and Hydrographic Effects on Phytoplankton Composition

A characteristic, reproducible, and very different summer–early fall phytoplankton community was found at the two northern study sites, IBSP and Sedge Is., that also differed relative to Harvey Cedars (central BB) and Tuckerton Cove (LEH). Most noteworthy, the IBSP site, characterized by low salinities and more eutrophic conditions, exhibited a high contribution of cyanobacteria and pico-coccoids to total algal biomass, whereas Sedge Is. and Tuckerton generally exhibited a summer phytoplankton community dominated by diatoms. These differences were maintained over the two study years despite the impact of Hurricane Sandy. Differences among sites in physical parameters (temperature and salinity) were also relatively consistent between 2012 and 2013, pre- and post-Hurricane Sandy years, respectively. Thus no evidence exists that Hurricane Sandy exerted major effects on the phytoplankton community or on juvenile hard-clam growth by the summer following its impact on BB-LEH in fall 2012. This agrees with results of Ren (2015), who found the most pronounced post-Sandy differences in phytoplankton assemblages in the northernmost sector of BB (north of Toms River), presumably attributable to the incursion of more saline waters, but these differences had largely dissipated by the summer.

The BB-LEH estuary experiences long residence times, particularly in the summer and in the northern sectors of the bay (Defne and Ganju, 2015; Guo et al., 2004). Residence time varies greatly within the system, depending on proximity to inlets, depth, and local topography such as shoals, marsh islands, and channels (Defne and Ganju, 2015). The IBSP site lies within one of the least energetic regions of the bay, characterized by long residence times, while the Tuckerton Cove site, despite its proximity to Little Egg Inlet, lies within a marsh region, where residence times are relatively high because of sheltering from winds and tides and relatively deep water (Defne and Ganju, 2015). Both of these sites exhibited the highest chl a concentrations yet differed markedly in their phytoplankton community composition, most notably the high contribution of diatoms and relatively low concentrations of cyanobacteria and chlorophytes at Tuckerton, compared to IBSP.

The above differences in phytoplankton community may be partly explained by nutrient speciation and differences in coastal development. Unlike nitrogen (N) and phosphorus (P), there is no substantial anthropogenic source of silica (Si) (Sferratore et al., 2006), and salt marshes have been shown to effectively recycle Si, acting as point sources in estuaries (Struyf et al., 2005; Vieillard et al., 2011). While Tuckerton's proximity to Little Egg Inlet could account for the seeding of diatoms from oceanic inputs, the extensive marshes in this area likely provide a supply of recycled dissolved Si, thus affording diatoms a competitive advantage over pico-planktonic chlorophytes, cyanobacteria, and flagellates. As no comparable source of regenerated Si at IBSP occurs, attributable to the highly developed coastline in this sector of the bay, N and P enrichment, together with transient Si limitation, can potentially shift nutrient ratios to values that favor picoplankton and flagellate communities over diatoms at this site (Officer and Ryther, 1980; Rocha, Galvão, and Barbosa, 2002). Additionally, Seitzinger, Sanders, and Styles (2002) found that a higher fraction of DON from urban/suburban stormwater runoff was bioavailable to plankton collected from BB relative to agricultural and naturally forested sources. Therefore, a greater proportion of DON is likely bioavailable to phytoplankton in the northern part of BB, either directly or via bacterial remineralization. Cyanobacteria and chlorophytes, generally with a wide salinity tolerance, have been associated with freshwater inputs as well as riverine sources of DON and high residence times in estuaries (Paerl et al., 2003; Pinckney et al., 1998). In northern BB, low salinities, organic N enrichment from anthropogenic sources (Hunchak-Kariouk and Nicholson, 2001; Wienben and Baker, 2009), silica limitation, and high residence times all likely play a role in sustaining the high summer concentrations of pico-chlorophytes and cyanobacteria characteristic of the IBSP study site. These differences in nutrient sources may account for the contrasting phytoplankton characteristics between IBSP and Tuckerton.

An inverse relationship between tidal height and zeaxanthin:chl a concentrations detected in 2012 (Figure 7) suggested that tidal transport from the Toms River may provide an influx of cyanobacteria to the IBSP site. To further examine this hypothesis, microscopy data from IBSP were compared to those obtained at the BB-05 NJDEP phytoplankton monitoring site (Figure 1). Pico-coccoid and cyanobacterial concentrations and biovolumes were generally lower or equal at BB-05 (not shown) to those at IBSP, but this may reflect greater mixing and deeper water than at the IBSP site, rather than disproving the hypothesis that the Toms River provides a source of these algae.

Proliferation of A. anophagefferens and a competitive advantage of this species relative to other algae in coastal lagoons occurs under conditions of high salinities (LaRoche et al., 1997), high turbidities (the species is tolerant of high light attenuation), and a low DIN/DOM ratio because of its capacity to take up organic nutrients (Gobler et al., 2011; Lomas et al., 2001). During the 2012–13 study period (present study; Ren, 2013), detection of brown tide was only episodic, and highest concentrations were found in central BB sites (Harvey Cedars and especially Sedge Is.) rather than in southern sectors of BB-LEH where the highest intensities were historically documented (Mahoney, Olsen, and Jeffress, 2006; Pecchioli, Lathrop, Haag, 2006). It is notable that, although moderate concentrations of A. anophagefferens (estimated from pigment analysis) were recorded in June 2013 at IBSP, these levels did not persist or become widespread, and 19′but was absent at IBSP on 19 June after a large pulse of rainfall and minimum salinity of 16 were recorded, yet returned on 3 July (∼105 cells ml−1) when the salinity had increased to 27. While only speculative, high precipitation events, associated with increased riverine input, and lower salinity may have hindered A. anophagefferens bloom development and kept bloom levels localized and transient. Riverine input was previously identified as a controlling factor in brown tide development in BB-LEH (Pecchioli, Lathrop, and Haag, 2006).

Comparison with Other Phytoplankton Studies in BB-LEH

Historically, studies of phytoplankton community structure in the BB-LEH estuary have been sparse. Early studies (Martin, 1929; Mountford, 1969, 1971, 1984) were focused primarily on large forms readily identified by microscopy and were conducted decades before the next comprehensive study (Olsen and Mahoney, 2001) was undertaken. Aside from the most recent studies on the phytoplankton community conducted between August 2011 and August 2013 (Ren, 2013, 2015; Ren et al., 2017), routine phytoplankton monitoring in BB-LEH by the state has focused solely on measurements of chl a and A. anophagefferens (Olsen and Mahoney, 2001; Pecchioli, Lathrop, and Haag, 2006). Both earlier taxonomic phytoplankton studies in BB-LEH (Mountford, 1984; Olsen and Mahoney, 2001) and more recent ones (Ren, 2013, 2015) have identified a diverse and dynamic phytoplankton community structure, with diatoms and dinoflagellates dominating from fall to spring and a greater contribution of picoplankton, flagellates, and small diatom species in the summer.

Maximum chl a detected at NJDEP study sites in 2012–13 (Ren, 2013, 2015) was lower than the maximum reported by Olsen and Mahoney (2001) and more comparable to those reported in the 1967–70 surveys (Mountford, 1984). Differences in sampling sites must be taken into account; however, as most NJDEP phytoplankton sites were located in open water and sites in the present study as well as the 1987 study and subsequent NMFS surveys (Olsen and Mahoney, 2001) were near shorelines, piers, and bridges. The maximum chl a values in the present study (∼30 and ∼23 μg l−1 at Tuckerton and IBSP, respectively) were higher than those observed at the corresponding open-water NJDEP sites.

Although the phytoplankton community composition in this study was similar to that reported in earlier (Olsen and Mahoney, 2001) and recent (Ren, 2013, 2015) studies, where they covered similar sectors of the estuary, some abundant species were observed that had not been previously recorded. Because of differences in methodologies and changes in algal taxonomic classification, however, it is often difficult to compare species composition with studies conducted decades earlier. Prominent species recorded in the present study, which were not present or uncommon in the studies conducted during and prior to the 1990s, include Aphanocapsa sp., Chaetoceros cf. tenuissimus, and Chlamydomonas sp. (Table 2).

In the present study, both microscopy (Figure 4, Table 2) and chemotaxonomy (Figure 3a,b) confirmed that at IBSP, the lowest salinity site, pico-coccoids were numerically dominant and made a major contribution to phytoplankton biomass in summer. This generally supports the results of Ren (2013, 2015), who found that summer picoplankton dominance was most pronounced at northern sites, i.e. north of Barnegat Inlet, and not found baywide. In contrast, Olsen and Mahoney (2001) described a positive relationship between picoplankton abundance and salinity, with greater concentrations found in the southern and central regions of the BB-LEH than in the northern regions. They reported the lowest picoplankton cell concentrations (<2 × 105 cells ml−1) at upper-central regions of the BB-LEH estuary, just south of Toms River, corresponding to the IBSP site in the present study. While the latter did not examine phytoplankton community in the BB sector north of Toms River, the opposite trend was observed within the two northern sites where microscopy data was available. Higher cell concentrations of pico-coccoids were found at IBSP than at Sedge Is. and, based on photopigment analysis, likely the two southern sites as well. In the present study, picoplankton cell concentrations never reached 106 ml−1, the peak picoplankton concentration typically observed during 1987–98 bloom periods (Olsen and Mahoney, 2001). The wider cell-size range used in their study (1–5 μm diameter) relative to the pico-coccoids identified in this study (1–3 μm), however, precludes making direct comparisons. The positive relationship between pico-coccoid concentrations and salinity described by Olsen and Mahoney (2001) may be at least partly attributable to the high levels of A. anophagefferens during several of their study years because this pico-coccoid alga thrives at higher salinities (Cosper et al., 1989; Gastrich et al., 2004; LaRoche et al., 1997).

Olsen and Mahoney (2001) noted association of certain species with increased picoplankton abundance during the summer, including the chrysophyte P. ovalis (taxonomic synonym of C. ovalis), a “Nitzschoid” diatom complex comprising Nitzschia spp. and Minutocellus polymorphus, cyanobacteria, and various microflagellates. These same associations were observed in the present study, as P. ovalis and Minutocellus spp. were most abundant at or approximately at the time of the midsummer pico-coccoid maximum, especially at IBSP (Figure 4, Table 2). Previously described reductions in species richness during the pico-coccoid maximum, especially large dinoflagellate and diatoms, were also observed in the present study (Figure 4, Table 2). The most abundant cyanobacterial species reported in BB-LEH by Olsen and Mahoney (2001), tentatively identified as Synechococcus sp., were not observed in the present study but may have been included as part of the pico-coccoid group. Aphanocapsa sp., the most common cyanobacterial species observed in this study, and by Ren (2013, 2015), was not recorded in BB-LEH in earlier studies.

Growth of M. mercenaria Juveniles in Relation to Environmental Factors

Maximum shell growth rates of juvenile M. mercenaria in the BB-LEH Estuary based on the four selected study sites were 176 and 144 μm day−1 (Table 5) in 2012 and 2013, respectively, approaching but not attaining maxima reported in other mid-Atlantic estuaries (∼200 μm day−1) (reviewed by Grizzle, Bricelj, and Shumway, 2001). Overall, however, significant spatial and temporal variability in juvenile clam growth rates occurred in BB-LEH. Clam growth rates at Sedge Is. were generally higher than at IBSP despite consistently lower mean daily temperatures, high early summer temperature fluctuations (up to 10°C and 16°C day−1 in 2012 and 2013, respectively), and lower chl a concentrations at Sedge Is. The growth rates are attributed to the high food quality at this site (high diatom contribution) relative to IBSP, where cyanobacteria and chlorophytes were more abundant. The IBSP site was also characterized by a consistently high percentage of POM and PON relative to total seston and high detrital contribution (low chl a/PON) (Bricelj et al., unpublished data), indicative of poor food quality despite high chl a concentrations relative to Sedge Is. Lower clam growth rates at IBSP than at Tuckerton and Sedge Is. are also attributed to transient, low-salinity events during early summer at IBSP (Table 1), which result in suboptimal salinities for hard-clam growth. It is thus clear from the present study that poor growth of hard-clam juveniles at IBSP could be attributed to the combined effects of poor food quality and episodic low salinities. At Harvey Cedars, episodic low growth was not clearly related to any environmental factors measured. Finally, clams at Tuckerton, in LEH, which historically supported high production of hard clams (reviewed by Bricelj, Kraeuter, and Flimlin, 2017), showed the highest or second highest overall shell growth rates (in μm day−1); these rates are attributed to both high food quantity and quality.

Generally, low hard-clam growth rates were associated in the present study with measures of cyanobacteria and chlorophyte abundance (Table 5). Laboratory studies have shown that species of cyanobacteria and chlorophytes, including Chlorella spp., Nannochloris spp., and Synechococcus sp., provide an inadequate food source for juvenile hard clams because of low retention and absorption efficiencies (Bass, Malouf, and Shumway, 1990; Bricelj, Bass, and Lopez, 1984).

Despite the generally positive association between the abundance of diatoms and kDW and the negative association between chlorophytes/cyanobacteria and kDW in the present study, in many cases the linear relationship between juvenile kDW and phytoplankton metrics was relatively weak (Table 5). This may be partly attributable to the sampling frequency used, as the mean of phytoplankton concentrations measured at the beginning and end of each week was taken to approximate weekly values for regression analysis. This type of analysis may not be appropriate for phytoplankton groups that contribute minimally to chl a, as an increase in these groups proportional to that of kDW can yield a significant relationship despite playing a minor role in determining clam growth (as observed at Tuckerton and Harvey Cedars; Table 5c,d). Multivariate analysis, including salinity, temperature, algal quantity and quality, outside the scope of the present study, however, may further contribute to explaining overall growth patterns.

In 2013, a positive relationship between CHEMTAX estimates of diatom concentration and hard-clam growth was determined at Sedge Is. (Table 5), but further analysis, in which chl a estimates included brown tide, indicated that most of the diatom peak observed during Trial I was likely contributed by A. anophagefferens (Figure 5d). In this case, the bloom could have been transient, or high, size-specific clam mortalities recorded at this site (see below) may have skewed measurements of growth rates. When CHEMTAX included A. anophagefferens, a significant positive relationship between diatom chl a and hard-clam growth (kDW) was still detected (slope = 0.660; R2 = 0.323, P < 0.05), but the relationship between kDW and the percentage of diatom contribution to chl a, however, remained nonsignificant.

It is noteworthy that while brown tide bloom categories have been classified based on cell concentrations that inhibited feeding and growth of bivalves in laboratory studies (Gastrich and Wazniak, 2002), the adverse effects of A. anophagefferens are dependent on strain toxicity as well as the physiological state of the alga. A clear concentration threshold that must be exceeded to elicit negative impacts in hard clams or other bivalves has not been established for the BB-LEH estuary, although the only bioassay conducted to date using an isolate from this system indicated that it was less toxic that two isolates from Long Island waters (Bricelj and MacQuarrie, unpublished data). An apparent negative effect on clam growth was observed in early 2012 due to relatively low background levels of A. anophagefferens at Harvey Cedars and Tuckerton but was not found in 2013, when much higher concentrations were observed. This suggests that strain differences in toxicity may occur, even within this system, and/or that other mitigating factors may be modulating bloom toxicity.

Overall cumulative juvenile hard-clam mortality was generally low (<18%) during both study years, but a relatively high mortality event (43%) occurred during June 2013 at Sedge Is. It is unclear whether high mortality at this site relative to other sites was attributable to A. anophagefferens, low temperatures, or a combination of factors. Juvenile hard clams were exposed to A. anophagefferens for a relatively short time period (∼1 week) before the high mortality event was recorded, and laboratory studies of juvenile exposure to A. anophagefferens at comparable cell concentrations did not induce mortalities over a 3-week period (Bricelj, MacQuarrie, and Smolowitz, 2004). High mortalities (60%) have been reported only in the field after sustained, 3-week exposure of smaller juveniles (2–3 mm SL) to high A. anophagefferens concentrations (1.5 × 106 cells ml−1) (Greenfield and Lonsdale, 2002). It cannot be ruled out that the high contribution of A. anophagefferens to the phytoplankton assemblage contributed to increased clam mortalities in 2013, although high mortalities may also reflect conditions at this site at the time of cage deployment, e.g., intrusion of cold waters via Barnegat Inlet at the time of planting, given that this site is subject to high fluctuations in temperature.

Ren et al. (2017) developed tentative indices of water quality for BB-LEH and associated least-impaired (reference) conditions with lower chl a, TSS, and chl/C and lower picoplankton biomass. Yet the Tuckerton site exhibited the highest chl a, chl a/N, chl a/C, and TSS concentrations (although a low picoplankton biomass) and typically supported the highest clam-shell growth rates in 2013 and second highest in 2012 (Bricelj et al., unpublished data; results of the present study). Therefore, the indices that are traditionally used to classify water quality in the BB-LEH and other estuaries (including reduction of SAV) (e.g., Ren et al., 2017; Wazniak et al., 2007) may not necessarily correspond to those that provide optimum conditions for hard-clam production. Mercenaria mercenaria can grow well in eutrophic coastal lagoon ecosystems such as Great South Bay, New York (Bricelj, Kraeuter, and Flimlin, 2017; Carmichael, Shriver, and Valiela, 2004), and nutrient ratios rather than bulk eutrophication (N loading) that affect algal composition (and salinity) appear to be more important to characterize suitable conditions for hard-clam somatic and reproductive production.

Summer chl a concentrations at all four study sites were generally ≤15 μg l−1 and thus below levels (>25 μg l−1) that are considered indicative of estuarine eutrophic conditions. Dinoflagellates made a very small contribution to chl a, typically <6% at all sites. Species present that are known to adversely affect bivalves included the dinoflagellates P. minimum and S. trochoidea and the pelagophyte A. anophagefferens.

The present study confirms the presence of A. anophagefferens in the BB-LEH estuary at levels occasionally exceeding those known to negatively affect hard-clam larvae and juveniles, thus indicating for the continued need for routine monitoring of this HAB species. A maximum density of 4.4 × 105 cells ml−1 was found at Sedge Is. within the MCZ where brown tides were not previously reported, which supports seeding efforts and a recreational clam fishery. A strong relationship was found between the concentration of A. anophagefferens and that of the 19′but pigment in BB-LEH; therefore, the latter offers a powerful tool to predict A. anophagefferens concentrations in this ecosystem. The present study further demonstrates that the occurrence of A. anophagefferens at bloom levels (>105 cells ml−1) can have major confounding effects on the results of CHEMTAX analysis and needs to be considered in areas such as BB-LEH that have been historically affected by brown tides and also where brown tide persists. These effects are attributable to the fact that A. anophagefferens contains a relatively high fucoxanthin to chl a ratio and that fucoxanthin is generally used as a diagnostic photopigment for diatoms.

Microalgal photopigment analysis provided a powerful tool to characterize the contribution of various phytoplankton taxonomic groups to total phytoplankton biomass in the BB-LEH. This study provides the first application of this analysis to determine temporal and site-specific characterization of the phytoplankton in this estuary, thus allowing more synoptic analysis of the phytoplankton within this large estuary at reduced cost. Additional ground-truthing using microscopic analysis, however, is required to further validate photopigment analysis for the BB-LEH estuary. Photopigment analysis also proved useful to detect short-term temporal variability in phytoplankton composition over the tidal cycle.

Growth of juvenile hard clams was typically positively related to the absolute and/or relative concentration of diatoms and negatively related to that of chlorophytes and cyanobacteria. A significant relationship between clam growth and chl a was found only at Sedge Is., where total phytoplankton biomass appeared to be limiting clam growth, especially during Trial I 2012, yet was compensated by the high food quality (prevalence of diatoms) at this site. Low food quality, despite relatively high chl a levels and episodic low salinities, explained poor clam growth at IBSP in the norther sector of BB, where nutrient loading is highest and is exacerbated by high residence times. Therefore, food quality appears to be more important than food quantity, as measured by total chl a, in predicting M. mercenaria growth. Overall, the present study indicates that environmental conditions during the summer–early fall growing season at four representative sites in BB-LEH can support moderate growth rates of juvenile M. mercenaria, although these were highly variable on a site-specific and weekly basis.

This study was supported by the NJDEP as part of the New Jersey Governor's 10 Point Action Plan to improve water quality in the BB-LEH ecosystem and by the Barnegat Bay Partnership. We thank Carola Noji, Lisa Izzo, Ashley Andrews, and DMCS/Rutgers University (RU) for their technical assistance in field sampling, sample processing, and data analysis and John Kraeuter (HSRL/RU) and Gef Flimlin for their collaboration in the overall clam project as co-investigators of our NJDEP-sponsored research. All HPLC analyses were conducted in Hans Paerl's analytical laboratory at the Institute of Marine Sciences, University of North Carolina at Chapel Hill, North Carolina, and we thank Karen Rossignol for her technical advice. We especially thank Chris Gobler, School of Marine and Atmospheric Science, Stony Brook University, for analysis of A. anophagefferens. We also thank Conor MacDonnell, supported by a National Science Foundation Research Internship in Ocean Sciences at DMCS/RU, for participation in this study in 2013. Jeffrey Silady, ReClam the Bay, provided boat access to Sedge Is., and Larry Murphy provided waterfront access at Harvey Cedars; staff at the IBSP Forked River Interpretive Center and the Sedge Is. Education Center provided use of their facilities. Juvenile clams were obtained from the following New Jersey commercial hatcheries: Mathis Clam Farm, Egg Harbor (Trials I, 2012, 2013), Pritchard Hammock Cove Clam Co. (Trial II, 2012), and Bill Avery's Quality Bay Clams, Atlantic City (Trial II, 2013).

Alami,
M.;
Lazar,
D.,
and
Green,
B.R.,
2012
.
The harmful alga Aureococcus anophagefferens utilizes 19′-butanoyloxyfucoxanthin as well as xanthophyll cycle carotenoids in acclimating to higher light intensities
.
Biochimica et Biophysica Acta (BBA)—Bioenergetics
,
1817
(
9
),
1557
1564
.
Bass,
A.E.;
Malouf,
R.E.,
and
Shumway,
S.E.,
1990
.
Growth of northern quahogs (Mercenaria mercenaria (Linnaeus, 1758)) fed on picoplankton
.
Journal of Shellfish Research
,
9
(
2
),
299
307
.
Boneillo,
G.E.
and
Mulholland,
M.R.,
2014
.
Interannual variability influences brown tide (Aureococcus anophagefferens) blooms in coastal embayments
.
Estuaries and Coasts
,
37
(
S1
),
147
163
.
Bricelj,
V.M.,
2009
.
The Hard Clam Research Initiative: Factors Controlling Mercenaria mercenaria Populations in South Shore Bays of Long Island, NY. New York Sea Grant Report NYSGI-T-09-001
,
43
p
.
Bricelj,
V.M.;
Bass,
A.E.,
and
Lopez,
G.R.,
1984
.
Absorption and gut passage time of microalgae in a suspension feeder: An evaluation of the 51Cr:14C twin tracer technique
.
Marine Ecology Progress Series
,
17
,
57
63
.
Bricelj,
V.M.;
Kraeuter,
J.N.,
Flimlin,
G.,
2017
.
Status and trends of hard clam, Mercenaria mercenaria, populations in a coastal lagoon ecosystem, Barnegat Bay-Little Egg Harbor, New Jersey
.
In:
Buchanan,
G.A.;
Belton,
T.J.,
and
Paudel,
B.
(
eds.
),
A Comprehensive Assessment of Barnegat Bay–Little Egg Harbor, New Jersey. Journal of Coastal Research
,
Special Issue No
.
sp1
,
pp
.
205
253
.
Bricelj,
V.M.
and
MacQuarrie,
S.P.,
2007
.
Effects of brown tide (Aureococcus anophagefferens) on hard clam, Mercenaria mercenaria, larvae and implications for benthic recruitment
.
Marine Ecology Progress Series
,
331
,
147
159
.
Bricelj,
V.M.;
MacQuarrie,
S.P.,
and
Schaffner,
R.,
2001
.
Differential effects of Aureococcus anophagefferens isolates (“brown tide”) in unialgal and mixed suspensions on bivalve feeding
.
Marine Biology
,
139
(
4
),
605
616
.
Bricelj,
V.M.;
MacQuarrie,
S.P.,
and
Smolowitz,
R.,
2004
.
Concentration-dependent effects of toxic and non-toxic isolates of the brown tide alga Aureococcus anophagefferens on growth of juvenile bivalves
.
Marine Ecology Progress Series
,
282
,
101
114
.
Bricker,
S.B.;
Longstaff,
B.;
Dennison,
W.;
Jones,
A.B.;
Boicourt,
K.;
Wicks,
E.C.,
and
Woerner,
J.L.,
2007
.
Effects of Nutrient Enrichment in the Nation's Estuaries: A Decade of Change. NOAA, National Ocean Service, Special Projects Office and National Centers for Coastal Ocean Science
,
Silver Spring, Maryland, USA.
Carmichael,
R.H.;
Shriver,
A.C.,
and
Valiela,
I.,
2004
.
Changes in shell and soft tissue growth, tissue composition, and survival of quahogs, Mercenaria mercenaria, and softshell clams, Mya arenaria, in response to eutrophic-driven changes in food supply and habitat
.
Journal of Experimental Marine Biology and Ecology
,
313
(
1
),
75
104
.
Cosper,
E.M.;
Dennison,
W.;
Milligan,
A.;
Carpenter,
E.J.;
Lee,
C.;
Hozapfel,
J.,
and
Milanese,
L.,
1989
.
An examination of the environmental factors important to initiating and sustaining “brown tide” blooms
.
In:
Cosper,
E.M.;
Bricelj,
V.M.,
and
Carpenter,
E.J.
(
eds.
),
Novel Phytoplankton Blooms: Causes and Impacts of Recurrent Brown Tides and Other Unusual Blooms
.
New York
:
Springer-Verlag
,
pp
.
317
340
.
Defne,
Z.
and
Ganju,
N.K.,
2014
.
Quantifying the residence time and flushing characteristics of a shallow, back-barrier estuary: Application of hydrodynamic and particle tracking models
.
Estuaries and Coasts
,
38
(
5
),
1719
1734
.
DeYoe,
H.R.;
Stockwell,
D.A.;
Bidigare,
R.R.;
Latasa,
M.;
Johnson,
P.W.;
Hargraves,
C.E.,
and
Suttle,
C.A.,
1997
.
Description and characterization of the algal species Aureoumbra and Aureococcus to the pelagophyceae
.
Journal of Phycology
,
33
,
1042
1048
.
Gainey,
L.F.,
Jr.,
and
Shumway,
S.E.,
1991
.
The physiological effect of Aureococcus anophagefferens (“brown tide”) on the lateral cilia of bivalve mollusks
.
Biological Bulletin
,
181
(
2
),
298
306
.
Gastrich,
M.D.;
Lathrop,
R.;
Haag,
S.;
Weinstein,
M.P.;
Danko,
M.;
Caron,
D.A.,
and
Schaffner,
R.,
2004
.
Assessment of brown tide blooms, caused by Aureococcous anophagefferens, and contributing factors in New Jersey coastal bays: 2000–2002
.
Harmful Algae
,
3
,
305
320
.
Gastrich,
M.D.
and
Wazniak,
C.E.,
2002
.
A brown tide bloom index based on the potential harmful effects of the brown tide alga, Aureococcus anophagefferens
.
Aquatic Ecosystem Health and Management
,
5
(
4
),
435
441
.
Gobler,
C.J.;
Berry,
D.L.;
Dyhrman,
S.T.;
Wilhelm,
S.W.;
Salamov,
A.;
Lobanov,
A.V.;
Zhang,
Y.;
Collier,
J.L.;
Wurch,
L.L.;
Kustka,
A.B.;
Bill,
B.D.;
Shah,
M.;
VerBerkmoes,
N.C.;
Kuo,
A.;
Terry,
A.;
Pangilinan,
J.;
Lindquist,
E.A.;
Lucas,
S.;
Paulsen,
I.T.;
Hattenrath-Lehmann,
T.K.;
Talmage,
S.C.;
Walker,
E.A.;
Koch,
F.;
Burson,
A.M.;
Marcoval,
M.A.;
Tang,
Y.-Z.;
LeCleir,
G.R.;
Coyne,
K.J.;
Berg,
G.M.;
Bertand,
E.M.;
Saito,
M.A.;
Gladyshev,
V.N.,
and
Grigoriev,
I.V.,
2011
.
Niche of harmful alga Aureococcus anophagefferens revealed through ecogenomics
.
Proceedings of the National Academy of Sciences
,
108
(
11
),
4352
4357
.
Gobler,
C.J.;
Lonsdale,
D.J.,
and
Boyer,
G.L.,
2005
.
A review of the causes, effects, and potential management of harmful brown tide blooms caused by Aureococcus anophagefferens (Hargraves et Sieburth)
.
Estuaries
,
28
(
5
),
726
749
.
Gobler,
C.J.;
Renaghan,
M.J.,
and
Buck,
N.J.,
2002
.
Impacts of nutrients and grazing mortality on the abundance of Aureococcus anophagefferens during a New York brown tide bloom
.
Limnology and Oceanography
,
47
(
1
),
129
141
.
Gobler,
C.J.
and
Sañudo-Wilhelmy,
S.A.,
2001
.
Effects of organic carbon, organic nitrogen, inorganic nutrients, and iron additions on the growth of phytoplankton and bacteria during a brown tide bloom
.
Marine Ecology Progress Series
,
209
,
19
34
.
Gobler,
C.J.
and
Sunda,
W.G.,
2012
.
Ecosystem disruptive algal blooms of the brown tide species Aureococcus anophagefferens and Aureoumbra lagunensis
.
Harmful Algae
,
14
,
36
45
.
Greenfield,
D.I.
and
Lonsdale,
D.J.,
2002
.
Mortality and growth of hard clams, Mercenaria mercenaria, during brown tide
.
Marine Biology
,
141
,
1045
1050
.
Grizzle,
R.;
Bricelj,
V.M.,
and
Shumway,
S.E.,
2001
.
Physiological ecology of Mercenaria mercenaria
.
In:
Kraeuter,
J.N.
and
Castagna,
M.
(
eds.
),
Biology of the Hard Clam
.
New York
:
Elsevier
,
pp
.
305
382
.
Guo,
Q.;
Psuty,
N.P.;
Lordi,
G.P.;
Glenn,
S.;
Mund,
M.R.,
and
Gastrich,
M.D.,
2004
.
Hydrographic Study of Barnegat Bay. New Jersey Department of Environmental Protection. Division of Science, Research and Technology Technical Report
.
4
p
.
Hillebrand,
H.;
Dürselen,
C.;
Kirschtel,
D.;
Pollingher,
U.,
and
Zohary,
T.,
1999
.
Biovolume calculation for pelagic and benthic microalgae
.
Journal of Phycology
,
35
(
2
),
403
424
.
Hunchak-Kariouk,
K.
and
Nicholson,
R.S.,
2001
.
Watershed contributions of nutrients and other nonpoint source contaminants to the Barnegat Bay-Little Egg Harbor Estuary
.
In:
Kennish,
M.J.
(
ed.
),
Barnegat Bay-Little Egg Harbor, New Jersey: Estuary and Watershed Assessment. Journal of Coastal Research, Special Issue No. 32
,
pp
.
28
81
.
Jessup,
D.A.;
Miller,
M.A.;
Ryan,
J.P.;
Nevins,
H.M.;
Kerkering,
H.A.;
Mekebri,
A.;
Crane,
D.B.;
Johnson
T.A.,
and
Kudela,
R.M.,
2009
.
Mass stranding of marine birds caused by a surfactant-producing red tide
.
PLoS ONE
,
4
(
2
),
e4550
.
Kennish,
M.J.;
Bricker,
S.B.;
Dennison,
W.C.;
Glibert,
P.M.;
Livingston,
R.J.;
Moore,
K.A.;
Noble,
R.T.;
Paerl,
H.W.;
Ramstack,
J.M.;
Seitzinger,
S.;
Tomasko,
D.A.,
and
Valiela,
I.,
2007
.
Barnegat Bay–Little Egg Harbor Estuary: Case study of a highly eutrophic coastal bay system
.
Ecological Applications
,
17(SP5), S3–S16.
Kennish,
M.J.
and
Fertig,
B.,
2012
.
Application and assessment of a nutrient pollution indicator using eelgrass (Zostera marina L.) in Barnegat Bay–Little Egg Harbor Estuary, New Jersey
.
Aquatic Botany
,
96
(
1
),
23
30
.
LaRoche,
J.;
Nuzzi,
R.;
Waters,
R.;
Wyman,
K.;
Falkowski,
P.,
and
Wallace,
D.,
1997
.
Brown tide blooms in Long Island's coastal waters linked to interannual variability in groundwater flow
.
Global Change Biology
,
3
(
5
),
397
410
.
Laza-Martinez,
A.;
Seoane,
S.;
Zapata,
M.,
and
Orive,
E.,
2007
.
Phytoplankton pigment patterns in a temperate estuary: From unialgal cultures to natural assemblages
.
Journal of Plankton Research
,
29
(
11
),
913
929
.
Lewitus,
A.J.;
White,
D.L.;
Tymowski,
R.G.;
Geesey,
M.E.;
Hymel,
S.N.,
and
Noble,
P.A.,
2005
.
Adapting the CHEMTAX method for assessing phytoplankton taxonomic composition in southeastern U.S. estuaries
.
Estuaries
,
28
(
1
),
160
172
.
Li,
Y.;
Olsen,
C.B.;
Lyman,
M.J.;
Van Heukelem,
L.;
WIlkfors,
G.;
Miranda,
L.,
and
Sengie,
L.,
2004
.
Monitoring phytoplankton in long island sound with HPLC photopigment profiles
.
Proceedings of the Seventh Biennial Long Island Sound Research Conference (Stony Brook, NY)
,
pp
.
45
54
.
Lomas,
M.W.;
Gilbert,
P.M.;
Clougherty,
D.A.;
Huber,
D.R.;
Jones,
J.;
Alexander,
J.,
and
Haramoto,
E.,
2001
.
Elevated organic nutrient ratios associated with brown tide algal blooms of Aureococcus anophagefferens (Pelagophyceae)
.
Journal of Plankton Research
,
23
(
12
),
1339
1344
.
Mackey,
M.D.;
Higgins,
H.W.;
Mackey,
D.J.,
and
Wright,
S.W.,
1997
.
CHEMTAX User's Manual: A Program for Estimating Class Abundances from Chemical Markers—Application to HPLC Measurements of Phytoplankton Pigments
.
Hobart, Australia: CSIRO Marine Laboratories Report 229,
47
p
.
Mackey,
M.D.;
Mackey,
D.J.;
Higgins,
H.W.,
and
Wright,
S.W.,
1996
.
CHEMTAX—A program for estimating class abundances from chemical markers: Application to HPLC measurements of phytoplankton
.
Marine Ecology Progress Series
,
144
,
265
283
.
Mahoney,
J.B.;
Olsen,
P.S.,
and
Jeffress,
D.,
2006
.
Bloom history of picoplankter Aureococcus anophagefferens in the New Jersey Barnegat Bay-Little Egg Harbor system and Great Bay, 1995–1999
.
United States Department of Commerce, Northeast Fisheries Science Center Reference Document 06-07
,
54
p
.
Martin,
G.W.,
1929
.
Dinoflagellates from marine and brackish waters in New Jersey
.
Ames, Iowa: University of Iowa Studies in Natural History
,
XII
,
32p
.
Mountford,
K.,
1969
.
A Seasonal Plankton Cycle for Barnegat Bay, New Jersey
.
New Brunswick, New Jersey
:
Rutgers University, Master's thesis
,
62
p
.
Mountford,
K.,
1971
.
Plankton Studies in Barnegat Bay
.
New Brunswick, New Jersey: Rutgers University, Ph.D. dissertation
,
147
p
.
Mountford,
K.,
1984
.
Phytoplankton
.
In:
Kennish,
M.J.
and
Lutz,
R.A.
(
eds.
),
Ecology of Barnegat Bay, New Jersey
.
New York
:
Springer-Verlag
,
pp
.
52
77
.
Mulholland,
M. R.;
Gobler,
C.J.;
and
Lee,
C.,
2002
.
Peptide hydrolysis, amino acid oxidation, and nitrogen uptake in communities seasonally dominated by Aureococcus anophagefferens
.
Limnology and Oceanography
,
47
(
4
),
1094
1108
.
National Oceanic and Atmospheric Administration (NOAA)
,
2017
.
Tides & Currents
. .
Newell,
R.I.E.;
Tettlebach,
S.T.;
Gobler,
C.J.,
and
Kimmel,
D.G.,
2009
.
Relationships between reproduction in suspension feeding hard clams Mercenaria mercenaria and phytoplankton community structure
.
Marine Ecology Progress Series
,
387
,
179
196
.
NJ Weather & Climate, 2017
.
Data Viewer
. .
Officer,
C.B.
and
Ryther,
J.H.,
1980
.
The possible importance of silicon in marine eutrophication
.
Marine Ecology Progress Series
,
3
,
83
91
.
Olsen,
P.S.,
1989
.
Development and distribution of a brown-water algal bloom in Barnegat Bay, New Jersey, with perspective on resources and other red tides in the region
.
In:
Cosper,
E.M.;
Carpenter,
E.J.,
and
Bricelj,
V.M.
(
eds.
),
Novel Phytoplankton Blooms: Causes and Impacts of Recurrent Brown Tides and Other Unusual Blooms. Lecture Notes on Coastal and Estuarine Studies
.
Berlin
:
Springer-Verlag
,
pp
.
189
212
.
Olsen,
P.S.
and
Mahoney,
J.B.,
2001
.
Phytoplankton in the Barnegat Bay-Little Egg Harbor Estuarine System: Species composition and picoplankton bloom development
.
In:
Kennish,
M.J.
(
ed.
),
Barnegat Bay-Little Egg Harbor, New Jersey: Estuary and Watershed Assessment. Journal of Coastal Research, Special Issue No. 32,
pp
.
115
143
.
Paerl,
H.W.;
Hall,
N.S.;
Peierls,
B.L.;
Rossignol,
K.K.,
and
Joyner,
A.R.,
2014
.
Hydrologic variability and its control of phytoplankton community structure and function in two shallow, coastal, lagoonal ecosystems: The Neuse and New River Estuaries, North Carolina, USA
.
Estuaries and Coasts
,
37
(
S1
),
S31
S45
.
Paerl,
H.W.;
Valdes,
L.M.;
Pinckney,
J.L.;
Piehler,
M.F.;
Dyble,
J.,
and
Moisander,
P.H.,
2003
.
Phytoplankton photopigments as indicators of estuarine and coastal eutrophication
.
BioScience
,
53
(
10
),
953
964
.
Paerl,
H.W.;
Valdes-Weaver,
L.M.;
Joyner,
A.R.,
and
Winkelmann,
V.,
2007
.
Phytoplankton indicators of ecological change in the eutrophying Pamlico Sound System, North Carolina
.
Ecological Applications
,
17
(
Sp5
),
88
101
.
Pecchioli,
J.A.;
Lathrop,
R.,
and
Haag,
S.,
2006
.
Brown Tide Assessment Project in NJ Coastal Waters: A Comparison of Three Bloom Years (2000–2002) with Two Non-Bloom Years (2003–2004). Division of Science, Research and Technology, NJDEP Research Project Summary
,
6
p
.
Pinckney,
J.L.;
Paerl,
H.W.;
Harrington,
M.B.,
and
Howe,
K.E.,
1998
.
Annual cycles of phytoplankton community-structure and bloom dynamics in the Neuse River Estuary, North Carolina
.
Marine Biology
,
131
(
2
),
371
381
.
Prasad,
A.K.;
Nienow,
J.A.,
and
Livingston,
R.J.,
1990
.
The genus Cyclotella (Bacillariophyta) in Choctawhatchee Bay, Florida, with special reference to C. striata and C. choctawhatcheeana sp. nov
.
Phycologia
,
29
,
418
436
.
Ren,
L.,
2013
.
Baseline Characterization of Phytoplankton and Harmful Algal Blooms in Barnegat Bay-Little Egg Harbor, New Jersey (Year One)
.
Division of Science and Research
,
NJDEP
,
71
p
.
Ren,
L.,
2015
.
Baseline Characterization of Phytoplankton and Harmful Algal Blooms in Barnegat Bay-Little Egg Harbor Estuary, New Jersey (Year Two)
.
Division of Science and Research
,
NJDEP
,
54
p
.
Ren,
L.;
Belton,
T.J.;
Schuster,
R., and
Enache,
M
.,
2017
.
Phytoplankton index of biotic integrity and reference communities for Barnegat Bay-Little Egg Harbor, New Jersey: A pilot study
.
In:
Buchanan,
G.A.;
Belton,
T.J.,
and
Paudel,
B.
(
eds.
),
A Comprehensive Assessment of Barnegat Bay–Little Egg Harbor, New Jersey. Journal of Coastal Research
,
Special Issue No
.
sp1
,
pp
.
89
105
.
Rocha,
C.;
Galvão,
H.,
and
Barbosa,
A.,
2002
.
Role of transient silicon limitation in the development of cyanobacteria blooms in the Guadiana estuary, south-western Iberia
.
Marine Ecology Progress Series
,
228
,
35
45
.
Seitzinger,
S.P.
and
Pilling,
I.E.,
1993
.
Eutrophication and Nutrient Loading in Barnegat Bay: Initial Studies of the Importance of Sediment-Water Nutrient Interactions
.
Philadelphia, Pennsylvania
:
Academy of Natural Sciences, Report No. 92-24F
.
Seitzinger,
S.P.;
Sanders,
R.W.,
and
Styles,
R.,
2002
.
Bioavailability of DON from natural and anthropogenic sources to estuarine plankton
.
Limnology and Oceanography
,
47
(
2
),
353
366
.
Sferratore,
A.;
Garnier,
J.;
Billen,
G.;
Conley,
D.I.,
and
Pinault,
S.,
2006
.
Diffuse and point sources of silica in the Seine river watershed
.
Environmental Science and Technology
,
40
(
21
),
6630
6635
.
Stauffer,
B.A.;
Schaffner,
R.A.;
Wazniak,
C.,
and
Caron,
D.A.,
2008
.
Immunofluorescence flow cytometry technique for enumeration of the brown-tide alga, Aureococcus anophagefferens
.
Applied and Environmental Microbiology
,
74
(
22
),
6931
6940
.
Struyf,
E.;
Damme,
S.V.;
Gribsholt,
B.,
and
Meire,
P.,
2005
.
Freshwater marshes as dissolved silica recyclers in an estuarine environment (Schelde estuary, Belgium)
.
Hydrobiologia
,
540
(
1–3
),
69
77
.
Tang,
Y.
and
Gobler,
C.,
2012
.
Lethal effects of Northwest Atlantic Ocean isolates of the dinoflagellate, Scrippsiella trochoidea, on Eastern Oyster (Crassostrea virginica) and Northern Quahog (Mercenaria mercenaria) larvae
.
Marine Biology
,
159
(
1
),
199
210
.
Trainer,
V.L.;
Bates,
S.S.;
Lundholm,
N.;
Thessen,
A.E.;
Cochlan,
W.P.;
Adams,
N.G.,
and
Trick,
C.G.,
2012
.
Pseudo-nitzschia physiological ecology, phylogeny, toxicity, monitoring, and impacts on ecosystem health
.
Harmful Algae
,
14
,
271
300
.
Trice,
T.;
Glibert,
P.M.;
Lea,
C.,
and
Heukelem,
L.V.,
2004
.
HPLC pigment records provide evidence of past blooms of Aureococcus anophagefferens in the Coastal Bays of Maryland and Virginia, USA
.
Harmful Algae
,
3
(
4
),
295
304
.
Vieillard,
A.M.;
Fulweiler,
R.W.;
Hughes,
Z.J.,
and
Carey,
J.C.,
2011
.
The ebb and flood of Silica: Quantifying dissolved and biogenic silica fluxes from a temperate salt marsh
.
Estuarine, Coastal and Shelf Science
,
95
(
4
),
415
423
.
Wazniak,
C.E.
and
P.M.
Glibert,
2004
.
Potential impacts of brown tide, Aurecoccus anophagefferens, on juvenile hard clams, Mercenaria mercenaria, in the coastal bays of Maryland, USA
.
Harmful Algae
,
3
,
321
329
.
Wazniak,
C.E.;
Hall,
M.R.;
Carruthers,
J.B.;
Sturgis,
B.;
Dennison,
W.C.,
and
Orth,
R.J.,
2007
.
Linking water quality to living resources in a mid-Atlantic lagoon system, USA
.
Ecological Applications, 17(5) Supplement, S64–S78
.
Wienben,
C.M.
and
Baker,
R.J.,
2009
.
Contributions of Nitrogen to the Barnegat Bay-Little Egg Harbor Estuary: Updated Loading Estimates
.
West Trenton, New Jersey, U.S. Geological Survey Technical Report,
25
p
.
Wikfors,
G.H.,
2005
.
A review and analysis of a trophic interactions between Prorocentrum minimum and clams, oysters and scallops
.
Harmful Algae
,
4
,
585
592
.
Wright,
S.W.,
2005
.
Analysis of Phytoplankton Populations Using Pigment Markers. Course Notes for a Workshop “Pigment Analysis of Antarctic Microorganisms.”University of Malaya, 29 June–1 July 2005, 55p
.
Wright,
S.W.;
Ishikawa,
A.;
Marchant,
H.J.;
Davidson,
A.T.;
van den Enden,
R.L.,
and
Nash,
G.V.,
2009
.
Composition and significance of picophytoplankton in Antarctic waters
.
Polar Biology
,
32
(
5
),
797
808
.
Zapata,
M.;
Fraga,
S.;
Rodríguez,
F.,
and
Garrido,
J.,
2012
.
Pigment-based chloroplast types in dinoflagellates
.
Marine Ecology Progress Series
,
465
,
33
52
.

APPENDIX

Table A1.

Input (A) and final pigment ratio matrices used in CHEMTAX analysis for IBSP (B–C), Sedge Is. (D–E), Harvey Cedars (F–G), and Tuckerton (H–I). Input (I) and final (J) ratio matrices for Sedge Is. CHEMTAX run including Aureococcus anophagefferens (see text for pigment abbreviations).

Input (A) and final pigment ratio matrices used in CHEMTAX analysis for IBSP (B–C), Sedge Is. (D–E), Harvey Cedars (F–G), and Tuckerton (H–I). Input (I) and final (J) ratio matrices for Sedge Is. CHEMTAX run including Aureococcus anophagefferens (see text for pigment abbreviations).
Input (A) and final pigment ratio matrices used in CHEMTAX analysis for IBSP (B–C), Sedge Is. (D–E), Harvey Cedars (F–G), and Tuckerton (H–I). Input (I) and final (J) ratio matrices for Sedge Is. CHEMTAX run including Aureococcus anophagefferens (see text for pigment abbreviations).

Supplementary data