ABSTRACT

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.

A pilot study was carried out to quantify season-salinity-specific phytoplankton reference communities and to develop a phytoplankton index of biotic integrity (P-IBI) for Barnegat Bay–Little Egg Harbor (BB-LEH) estuary, using approaches similar to those for the Chesapeake Bay. Synchronized phytoplankton and water-quality data collected between August 2011 and August 2013 from the Barnegat Bay Water Quality Monitoring Program were used as calibration data set. The results showed that one-fifth of samples were from the least-impaired habitat condition, with low turbidity and low concentrations of dissolved inorganic nitrogen and orthophosphate (PO4). Nearly 60% of samples were from undesirable conditions with poor water clarity and excess nutrients. Phytoplankton reference communities, in comparison with communities in impaired conditions, were characterized with low concentrations of chlorophyll a (Chl a), total nitrogen (TN), and total phosphorus (TP), low Chl a/C ratio, low summer picoplankton biomass, and high spring and summer dissolved oxygen. Thirty-four metrics were evaluated for their ability to discriminate between the least-impaired and impaired habitat conditions. Nine phytoplankton metrics and three physiological and chemical metrics, which showed strong discriminatory ability, were selected, and different combinations of these metrics were used to create phytoplankton indices for spring and summer mesohaline and polyhaline zones in BB-LEH. The current P-IBI was able to correctly classify 64–100% for spring samples and 68–88% for summer samples in the calibration data set. Our work is the first attempt to develop a P-IBI for this region. The calculated reference communities and P-IBI, though constrained because of limited data availability, were region specific and intended to facilitate water-quality assessment and management efforts in BB-LEH. The differences of TN and TP between the least-impaired and impaired conditions in most of season-salinity zones suggested that dual reduction of N and P are necessary to control eutrophication in BB-LEH. Further work on the refinement of the P-IBI is underway as additional phytoplankton and water-quality data are being collected and assessed.

INTRODUCTION

Phytoplankton are known to respond directly to changes of physical and chemical conditions in aquatic ecosystems. They are also the base of the food web and dynamically interact with organisms at higher trophic levels in the ecosystem. The change of phytoplankton assemblages constitutes a good integrated measure of the state of the ecosystem, reflecting both internal interactions within the system and external inputs to the system. These roles make phytoplankton an important group to consider as a valuable bioindicator for water-quality assessment. For these reasons, phytoplankton have been specifically identified by regulatory authorities as a biological water-quality component, its monitoring including such elements as taxonomic composition, abundance, biomass, and blooms (Devlin et al., 2007; European Commission, 2000). The U.S. Clean Water Act mandates the use of biological data to assess the ecological conditions of aquatic ecosystem. Phytoplankton assessment, including chlorophyll a (Chl a) measurements and enumeration of bloom/dominant species, is recommended by the U.S. Environmental Protection Agency's (USEPA) Estuarine and Coastal Technical Guidance (Gibson et al., 2000).

Various studies have been done to develop phytoplankton-based indicators (Kauppila, 2007; Pearl et al., 2003). A bioindicator may be an organism or a set of organisms, whereas an index of biotic integrity is multimetric and therefore able to summarize the features of different elements of the ecosystem by integrating relevant ecological information into an overall expression of condition. The multimetric indices comprised of phytoplankton elements have proven to be more sensitive to environmental change than the individual elements (Johnson and Buchanan, 2014; Martinez-Crego, Alcoverro, and Romero, 2010). The principles of biotic indices were first developed in freshwater systems (Karr, 1981). Later on they were applied to coastal and estuarine ecosystems. Phytoplankton are one of the commonly used groups for developing biotic indices. Multiple phytoplankton metrics have been used in biotic indices in several estuaries including the German Bight (Radach, 1998) and the Chesapeake Bay (Jordan and Vaas, 2000; Lacouture et al., 2006). Reference conditions are established on the basis of phytoplankton elements such as Chl a, productivity, biomass, and cell sizes for several subregions (Buchanan et al., 2005; Devlin et al., 2007).

Recently, bioindicators using benthic macroalgae and eelgrass have been developed to assess eutrophication and nutrient pollution in the Barnegat Bay–Little Egg Harbor system (BB-LEH) (Kennish and Fertig, 2012; Kennish, Fertig, and Sakowicz, 2011). Other bioindicators, including benthic invertebrates and benthic diatoms, have been developed under the New Jersey Department of Environmental Protection (NJDEP)'s Barnegat Bay Comprehensive Plan of Action for Barnegat Bay (Desianti et al., 2017; Taghon et al., 2017). These studies are important for us to better understand the development of different biotic components in response to nutrient loading and water-quality changes in BB-LEH. However, studies on phytoplankton indicators and biotic indices for water quality in Barnegat Bay have been lacking. The major objective of the study was to develop a phytoplankton-based index of biotic integrity and provide insightful information on phytoplankton communities in relation to water-quality conditions to facilitate water-quality assessment and management in BB-LEH.

A 2-year investigation was conducted on the phytoplankton community in BB-LEH from August 2011 to August 2013. Monthly and biweekly samples were collected by the NJDEP Water Quality Monitoring Program for water-quality measurements and the analysis of phytoplankton species composition and cell density, as well as biovolume and carbon biomass (see details in project reports; Ren 2013, 2015). Phytoplankton data, collected simultaneously with water-quality data, provide an ideal calibration data set for the phytoplankton index of biotic integrity (P-IBI) development. We used approaches similar to those used in Chesapeake Bay studies (Buchanan et al., 2005; Lacouture et al., 2006) to quantify phytoplankton reference communities and their supporting habitat (least-impaired) conditions, and to develop a season-salinity-specific P-IBI for the BB-LEH estuary. Phytoplankton habitat conditions were classified for each season-salinity zone on the basis of nutrient (primarily dissolved inorganic nitrogen [DIN] and orthophosphate [PO4]) and light measurements. This work was a pilot study and represents the first attempt to develop a P-IBI for the BB-LEH estuary. The preliminary results from the calibration effort are presented in this paper. The P-IBI and calculated reference conditions, as well as habitat classification criteria, are scientifically based and regionally specific (see Methods below), and are expected, with underway refinements, to facilitate the assessment, restoration, and rehabilitation efforts for water-quality management in the BB-LEH estuary.

METHODS

The general steps for developing indices of biotic integrity are well established (Gibson et al., 2000; National Research Council, 2000). The P-IBI was developed for BB-LEH using the methodology similar to those used for the Chesapeake Bay (Buchanan et al., 2005; Lacouture et al., 2006). The major steps in developing the P-IBI included data compilation and analysis, habitat classification, metrics selection, metrics scoring criteria, metrics scoring, and validation. Phytoplankton reference communities were quantified on the basis of the samples from the least-impaired habitat conditions.

Data Compilation

In total, 205 samples were used as calibration data sets for the P-IBI development. Monthly (October to April) and biweekly (from May to September) samples were collected by the NJDEP Water Quality Monitoring Program for water-quality and phytoplankton analyses from nine sites in the first year, and six of those same sites in the second year (Figure 1, Table 1). The sites BB02, BB05a, and BB14 were excluded in the second year because the phytoplankton communities from those sites were similar to those from BB01, BB04a, and BB12, respectively, on the basis of cluster analysis (Ren, 2015; Ren et al., unpublished data). The phytoplankton data set includes species composition, species abundance (cell density), biovolume, and carbon biomass of the major taxonomic groups. The water-quality data set contains key parameters such as salinity, temperature, Secchi depth, DIN, PO4, dissolved oxygen (DO), dissolved organic carbon (DOC), total organic carbon (TOC), Chl a, total nitrogen (TN), and total phosphorus (TP). Both phytoplankton analysis and water-quality parameter measurements were carried out by the same laboratories using consistent methods throughout the multiple years of investigation.

Figure 1.

Map of sites for phytoplankton sample collection, August 2011–August 2013. Six sites (framed) were collected for both years. Sites BB04a, BB05a, and BB07a were shifted from BB04, BB05, and BB07 (labels in grey) after May 2012.

Figure 1.

Map of sites for phytoplankton sample collection, August 2011–August 2013. Six sites (framed) were collected for both years. Sites BB04a, BB05a, and BB07a were shifted from BB04, BB05, and BB07 (labels in grey) after May 2012.

Table 1.

Location and description of phytoplankton collection sites in Barnegat Bay–Little Egg Harbor from August 2011 to August 2013. Nine sites were analyzed in year 1 (August 2011–September 2012); six sites (with asterisk) were analyzed in year 2 (October 2012–August 2013).

Location and description of phytoplankton collection sites in Barnegat Bay–Little Egg Harbor from August 2011 to August 2013. Nine sites were analyzed in year 1 (August 2011–September 2012); six sites (with asterisk) were analyzed in year 2 (October 2012–August 2013).
Location and description of phytoplankton collection sites in Barnegat Bay–Little Egg Harbor from August 2011 to August 2013. Nine sites were analyzed in year 1 (August 2011–September 2012); six sites (with asterisk) were analyzed in year 2 (October 2012–August 2013).

Phytoplankton samples were collected at a depth of 1 ft from each station. Samples were preserved with glutaraldehyde (0.5–1% v/v) and kept dark and cool (∼4°C) before analysis. Phytoplankton samples were size-fractionated by filtering through 0.2-μm, 3-μm and 8-μm pore-size filters. The latter two fractions were stained with 0.03% proflavine hemisulfate. The 0.2- to 3-μm fraction was counted immediately after filtration. The >8-μm fraction was kept frozen and counted later. Algal identification and enumeration, including soft algae and diatoms, were done under an epifluorescence microscope (Leica DM L 2) with blue and green excitation lights and transmitted light. For 0.2- and 3-μm pore-size filters, observations were done under ×1000 magnification. For each filter, at least five random fields were counted or until at least 100 cells were counted. If the filter was very sparse, then 50 random fields were counted before stopping. For 8-μm pore-size filters, each filter was observed under three magnifications: under ×1000 magnification for phytoplankton <20 μm with the same counting strategy in terms of finishing point; under ×400 magnification for larger (>20 μm) phytoplankton with a maximum of 25 random fields when it was sparse; and under ×100 magnification to catch some large organisms, which might have been missed under higher magnifications because of either their large size or sparse density. The detection limit was approximately 500 to 1000 cells/L, depending on the sample volume filtered (Dortch et al., 1997; Ren et al., 2009). For samples with high abundance and diversity of diatoms, diatom slides were made separately. Diatoms were analyzed to get the percentage of dominant diatoms, especially the small centric diatoms. All phytoplankton were identified to the lowest taxonomic level possible. Biovolumes of common taxa were calculated on the basis of measurements of dimensions and geometric models of phytoplankton (Hillebrand et al., 1999; Olenina et al., 2006). Carbon biomass for diatoms and nondiatoms was calculated on the basis of the biovolume measurements and the cell carbon contents (Eppley, Reid, and Strickland, 1970).

Grab samples were collected for water-quality parameters following the procedures in the NJDEP Field Sampling Procedure Manual (August 2005). Water temperature, pH, conductivity, Secchi depth, and DO and DO saturation were measured in situ, using handheld meters and multiparameter sensors. Salinity was calculated from conductivity. Turbidity and total suspended solids (TSS) were measured from respective filters using methods SM 2130 B-11 and USGS I-3765-85, respectively. Chl a was measured following SM 10200-H 1+2. Measurements for TN, dissolved nitrogen, TP, and dissolved phosphorus followed the method USGS I-4650-03. PO4 was measured using the method EPA 365.5. Dissolved ammonia (NH4) and nitrate + nitrite (NO3 + NO2) were measured following EPA methods 350.1MOD and 353.4, respectively. In addition, TOC and DOC were detected following SM5310 C. More details on methods and water-quality data can be found in the Quality Assurance Sampling Plan for NJDEP Barnegat Bay Long-Term Ambient Water Monitoring (Barnegat Bay LMP QAPP, 2013) and the website http://www.state.nj.us/dep/barnegatbay/bbmapviewer.htm, respectively.

Season and Salinity Classification

The distinction of four seasons was based on the water temperature and the seasonal variabilities of phytoplankton assemblages from the same site and as follows: spring, March–May; summer, June–September; fall, October–November, and winter, December–February. The classification of salinity regimes followed the well-accepted Venice system (Anonymous, 1958).

Phytoplankton Habitat Classification

A combination of water clarity, as Secchi depth, and nutrients, primarily DIN and PO4, was used to classify the habitat condition of each sample for phytoplankton growth. Before the classification, the criteria for Secchi depth, DIN, and PO4 were established for each season-salinity zone, using similar methods and principles outlined in Buchanan et al. (2005). Microcosm experiments showed that phytoplankton growth and production in Barnegat Bay was limited primarily by N and secondly by P in summer and was colimited by N + P in fall (Seitzinger, Styles and Pilling, 2001). The nutrient-limiting thresholds for phytoplankton growth, derived from bioassay experiments (e.g., Fisher and Gustafson, 2003; Ren, 2002;), were used as the criteria to separate the better and poor classes for DIN (0.07 mg/L, Table 2) and PO4 (0.007 mg/L, Table 3). The principles of such settings are as follows: (1) phytoplankton growth is limited when the concentrations of DIN and PO4 are below the thresholds. In this case, the additions of nutrients are expected to significantly stimulate the phytoplankton growth, and such a condition is generally sensitive to excess nutrients, and therefore considered desirable (better); (2) on the other hand, when nutrient concentrations exceed the limiting thresholds, the addition of nutrients do not stimulate the growth at significant levels. Such a condition is generally insensitive to external nutrient input and thus considered undesirable (poor). In addition, the classes of best and worst were added as the extreme subsets of the better and poor classes to further classify DIN and PO4 concentrations. The DIN criteria for the best class was set to <0.03 mg/L (about 40% of the DIN limiting threshold). The PO4 criteria for the best class was set to <0.002 mg/L (about 30% of the PO4 limiting threshold), primarily to accommodate the values under the detection limit, a similar strategy used in the Chesapeake Bay study (Buchanan et al., 2005). The criteria for the worst class of DIN and PO4 were established using the relative status method (Alden and Perry, 1997; Olson, 2002, 2009). All data from samples with their medians of DIN, PO4, Chl a, and TSS in the desirable ranges were extracted, and their quartiles were used to set the threshold criteria for the worst class of DIN and PO4. The relative status method was also used to establish the worst, poor, better, and best classes of Secchi depth in each season-salinity zone (Table 4). The desirable ranges for DIN, PO4, Chl a, and TSS were the lower third of the logistic distribution of the data set, whereas those for Secchi depth were the upper third of the logistic distribution (Buchanan et al., 2005). The classification criteria of DIN, PO4, and Secchi depth are shown in Tables 24.

Table 2.

DIN (mg/L, NO3 + NO2 + ) classification criteria for water-quality classes of worst, poor, better, and best for different seasons and salinity zones. Spring, March–May; summer, June–September; fall, October–November; winter, December–February. Salinity: mesohaline (MH) and polyhaline (PH).

DIN (mg/L, NO3 + NO2 + ) classification criteria for water-quality classes of worst, poor, better, and best for different seasons and salinity zones. Spring, March–May; summer, June–September; fall, October–November; winter, December–February. Salinity: mesohaline (MH) and polyhaline (PH).
DIN (mg/L, NO3 + NO2 + ) classification criteria for water-quality classes of worst, poor, better, and best for different seasons and salinity zones. Spring, March–May; summer, June–September; fall, October–November; winter, December–February. Salinity: mesohaline (MH) and polyhaline (PH).
Table 3.

Orthophosphate (PO4, mg/L) classification criteria for water-quality classes of worst, poor, better, and best for different seasons and salinity zones. Spring, March–May; summer, June–September; fall, October–November; winter, December–February. Salinity: mesohaline (MH) and polyhaline (PH); —, not applicable because of data limitation.

Orthophosphate (PO4, mg/L) classification criteria for water-quality classes of worst, poor, better, and best for different seasons and salinity zones. Spring, March–May; summer, June–September; fall, October–November; winter, December–February. Salinity: mesohaline (MH) and polyhaline (PH); —, not applicable because of data limitation.
Orthophosphate (PO4, mg/L) classification criteria for water-quality classes of worst, poor, better, and best for different seasons and salinity zones. Spring, March–May; summer, June–September; fall, October–November; winter, December–February. Salinity: mesohaline (MH) and polyhaline (PH); —, not applicable because of data limitation.
Table 4.

Secchi depth (m) classification criteria for water-quality classes of worst, poor, better, and best for different seasons and salinity zones. Spring, March–May; summer, June–September; fall, October–November; winter, December–February. Salinity: mesohaline (MH) and polyhaline (PH).

Secchi depth (m) classification criteria for water-quality classes of worst, poor, better, and best for different seasons and salinity zones. Spring, March–May; summer, June–September; fall, October–November; winter, December–February. Salinity: mesohaline (MH) and polyhaline (PH).
Secchi depth (m) classification criteria for water-quality classes of worst, poor, better, and best for different seasons and salinity zones. Spring, March–May; summer, June–September; fall, October–November; winter, December–February. Salinity: mesohaline (MH) and polyhaline (PH).

The Secchi depth, DIN, and PO4 of each sample were then independently classified as worst, poor, better, or best on the basis of the classification criteria. After the classification, each sample was grouped into one of the 10 phytoplankton habitat categories depending on the combination of its class scores of Secchi depth, DIN, and PO4. Six categories of habitat condition were created from these 10 categories, including B (best), BB (better–best), MBL (mixed-better light), MPL (mixed-poor light), PW (poor–worst), and W (Worst) (Table 5). Among the 10 different combinations, the combo 1 (W) and 2 (PW) represent impaired water quality, characterized with light-impoverished condition and excessive DIN and PO4 concentrations in the water column. Combo 1 was the extreme subset of combo 2, with light, DIN, and PO4 values all falling into the class of W. On the other hand, combo 9 (BB) and 10 (B) represented the least-impaired habitat condition, with desirable light conditions and nutrient concentrations below the phytoplankton growth limitation thresholds. Combo 10 was the extreme subset of better, with light, DIN, and PO4 values falling into the class of B. The remaining six combinations were binned into two categories, MPL and MBL, differentiated by the Secchi depth. In MPL, Secchi depth values were low and fell into the criteria of poor, and DIN and PO4 concentrations were either individually (category 3 and 4) or jointly (category 5) below the phytoplankton growth limitation thresholds. In MBL, Secchi depth values met the criteria of better, but one or both nutrients exceeded their limitation thresholds. The grouping reflects the fact that light is one of the most important factors for phytoplankton growth as it provides an energy source for photosynthesis (Cloern, 1999). Such binning categories were assigned to facilitate the identification of various phytoplankton habitat conditions from the standpoint of DIN, PO4, and light (Buchanan et al., 2005). Technically, the binning is necessary to obtain sufficient data for each season-salinity zone and to “lessen the possibility that data from one or two sites dominate a given category” (Buchanan et al., 2005).

Table 5.

Explanation of phytoplankton habitat category classification. Combo#: combination number in consideration of the classes of light (Secchi depth), DIN, and orthophosphate (PO4). Low = poor and worst light classes; high = better and best light classes; excess = poor and worst nutrient classes; limiting = better and best nutrient classes. The poor habitat category represents impaired conditions; the better category and some MBL represent least-impaired conditions. The worst category is a subset of poor (with worst DIN, PO4, and Secchi depth); the best category is a subset of better (with best DIN, PO4, and Secchi depth).

Explanation of phytoplankton habitat category classification. Combo#: combination number in consideration of the classes of light (Secchi depth), DIN, and orthophosphate (PO4). Low = poor and worst light classes; high = better and best light classes; excess = poor and worst nutrient classes; limiting = better and best nutrient classes. The poor habitat category represents impaired conditions; the better category and some MBL represent least-impaired conditions. The worst category is a subset of poor (with worst DIN, PO4, and Secchi depth); the best category is a subset of better (with best DIN, PO4, and Secchi depth).
Explanation of phytoplankton habitat category classification. Combo#: combination number in consideration of the classes of light (Secchi depth), DIN, and orthophosphate (PO4). Low = poor and worst light classes; high = better and best light classes; excess = poor and worst nutrient classes; limiting = better and best nutrient classes. The poor habitat category represents impaired conditions; the better category and some MBL represent least-impaired conditions. The worst category is a subset of poor (with worst DIN, PO4, and Secchi depth); the best category is a subset of better (with best DIN, PO4, and Secchi depth).

Quantification of Reference Communities

Among the six categories, the categories B and BB represent the least-impaired habitat conditions (Table 5). Phytoplankton populations in the least-impaired habitats were considered the best representatives of the phytoplankton communities that are not impaired by excess nutrients and poor water clarity. Phytoplankton reference communities were, therefore, calculated and quantified from those least-impaired samples. For those season-salinity zones when categories B and BB were very few or not present, samples from MBL were used as the “surrogates” for the least-impaired habitat condition (Buchanan et al., 2005). The significance of differences between the reference communities and impaired populations were tested by analysis of variance (ANOVA). The 25th and 75th percentile values were used as the boundaries of the reference values.

Metric Selection

Thirty-four phytoplankton physiological and chemical metrics were evaluated for their ability to discriminate between least-impaired and impaired conditions. The least-impaired conditions included the BB and B categories. The impaired conditions included PW and W categories. In the cases where data from those categories were lacking or insufficient, data from the MBL or MPL were included to augment the number of data, or used as surrogates of the least-impaired or impaired if BB + B or PW + W were absent. The Kruskal–Wallis test was done on each metric to test the significance level of their discriminatory ability (Lacouture et al., 2006).

Metric Scoring Criteria and P-IBI Scoring

Phytoplankton metrics showing significant discriminatory ability between the least-impaired and impaired conditions were selected to compose the P-IBI. The scoring criteria for each metric were established using similar approaches as described in previous studies (Gibson et al., 2000; Lacouture et al., 2006; Weisberg et al., 1997). Efforts were made primarily on the spring and summer P-IBI because relatively more data were available, 60 for spring and 85 for summer, respectively, compared with fall (25) and winter (35). The scoring criteria for each metric in season-salinity-specific P-IBI are shown in Table 6. Before being incorporated into P-IBI scores, each metric was scored separately but on the same scale of 5, 3, or 1 to avoid unintended biases. We used the same principle and strategies as applied in the Chesapeake Bay study (Lacouture et al., 2006), where 5 is “reference-like”, 3 is “somewhat like reference”, and 1 is “not like reference”. Individual metric scores were then averaged to generate a P-IBI score for each sampling event. The discrimination efficiency (DE) was calculated for each metric to indicate how well an individual metric can correctly identify least-impaired (L-Imp) and impaired (Imp) conditions. The classification efficiency (CE) of the P-IBI was calculated to determined how well the index can discriminate between the least-impaired and impaired conditions (Gibson, 2000; Johnson and Buchanan 2014; Lacouture et al., 2006). The equations for calculating DE and CE are as follows:

Table 6.

Phytoplankton IBI metrics scoring criteria for spring and summer mesohaline and polyhaline zones.

Phytoplankton IBI metrics scoring criteria for spring and summer mesohaline and polyhaline zones.
Phytoplankton IBI metrics scoring criteria for spring and summer mesohaline and polyhaline zones.

DE of individual metric for the least-impaired (DEL-Imp) and impaired (DEImp) conditions:

 
formula
 
formula

where, L-Impmetric score≥3 is the number of the least-impaired samples in which a metric scores 3 or 5. Impmetric score<3 is the number of the impaired samples in which a metric scores less than 3. L-Imptot and Imptot are the total number of the least-impaired and impaired samples, respectively.

CE of P-IBI for the least-impaired (CEL-Imp) and impaired (CEImp) conditions:

 
formula
 
formula

where, L-ImpPIBI score≥3 is the number of the least-impaired samples in which the P-IBI scores 3 or more. ImpPIBI score<3 is the number of the impaired samples in which the P-IBI scores less than 3. L-Imptot and Imptot are the total number of the least-impaired and impaired samples, respectively.

Overall CE of the P-IBI was calculated as follows:

 
formula

Statistical Analyses

A Kruskal–Wallis test was used to determine the discriminatory ability of phytoplankton metrics (Lacouture et al., 2006). The method compared the data sets of each particular metric in reference (least-impaired) and degraded (impaired) communities. In addition, one-way ANOVA was used to test the significance of differences between the reference communities and impaired ones.

RESULTS

Phytoplankton Habitat Conditions

All samples collected from 2011 to 2013 fell into two salinity zones: mesohaline (MH; 5–18 parts per thousand [ppt]) and polyhaline (PH; >18 ppt). The majority (82%) of the total 205 samples were PH. MH samples were collected mostly within the fluvial plume of the Toms River, a narrow segment entering Barnegat Bay (BB04a, Figure 1), from north of the Toms River to Silver Bay (BB02). In addition, samples from BB01 and BB05a fell occasionally into the class of MH. PH samples were collected from sites BB07a, BB09, BB10, BB12, BB14, and frequently from BB01 and BB05 (Figure 1).

Pie plots show the percentages of habitat categories at each individual site (Figure 2a–f) and of all sites (Figure 2g). In addition, the frequency of each category, as the number of samplings, for each site and season are listed in Table 7. For most sites, more than half of the samples were classified as MPL. Overall, the MPL and PW + W accounted for 60% of total samples, representing reduced light, an undesirable condition during those sampling events. Categories BB + B were regarded as the least-impaired conditions, with high water clarity for phytoplankton growth. The percentage of category BB + B varied from 8% to 41% among different sites, except for BB12, where no sample was found to be in BB + B condition. Overall, 20% of the total samples were classified as BB + B condition, most of which were from northern and central Barnegat Bay in spring and summer (Table 7). In contrast to BB + B, categories PW + W were featured with poor light condition and excess nutrients in the system. In total, only 6% of samples were classified in PW + W. Most of PW + W samples were found to be in summer and fall from the sites near the Barnegat Inlet (BB07a, BB09) and in Little Egg Harbor (BB12) (Table 7).

Figure 2.

Percentage of phytoplankton habitat categories at each individual site and of all sites in BB-LEH. Categories: BB + B, Best-better and best; MBL, mixed-better light; MPL, mixed-poor light; W + PW, worst and poor–worst.

Figure 2.

Percentage of phytoplankton habitat categories at each individual site and of all sites in BB-LEH. Categories: BB + B, Best-better and best; MBL, mixed-better light; MPL, mixed-poor light; W + PW, worst and poor–worst.

Table 7.

Frequency (as number of sampling events) of phytoplankton habitat categories in spring and summer from each site, derived from data collected between August 2011 and August 2013. See Table 5 for the explanation of category codes and combination numbers (combo#).

Frequency (as number of sampling events) of phytoplankton habitat categories in spring and summer from each site, derived from data collected between August 2011 and August 2013. See Table 5 for the explanation of category codes and combination numbers (combo#).
Frequency (as number of sampling events) of phytoplankton habitat categories in spring and summer from each site, derived from data collected between August 2011 and August 2013. See Table 5 for the explanation of category codes and combination numbers (combo#).

Phytoplankton Reference Communities

Reference communities were quantified on the basis of the data from the least-impaired (reference) habitat conditions. For the PH zone, 24 samples in the B + BB categories were used to calculate the reference communities. In the case of MH, only six samples were available in the B + BB categories; therefore five samples in the MBL category were included to augment the number of data for the calculation. The representative metrics of phytoplankton communities included Chl a, Chl a/C ratio, nano- and microphytoplankton (NM) abundance, NM biomass, average NM cell size, and summer picoplankton biomass. In addition, parameters that are related to phytoplankton growth and biomass were considered, including DO, TSS, TN, and TP. The median, maximum, and minimum values of each metric in the reference communities for MH and PH zones are listed in the Tables 8 and 9, respectively. Statistical significance of differences, tested by one-way ANOVA, showed that the reference values of Chl a, Chl a/C ratio, TSS, TN, and TP were significantly lower compared with the values in the impaired communities. On the other hand, the concentration of DO and average NM cell size in MH were significantly higher in the reference communities than those in the impaired ones (Table 8). Compared with the MH zone, statistically more metrics showed significant differences between the reference and impaired conditions in the PH zone (Table 9). Lower p-values (<0.01) were obtained for metrics such as Chl a, DO, and TN in the PH zone than those in the MH zone, indicating more statistically significant differences between the least-impaired and impaired phytoplankton communities.

Table 8.

Phytoplankton reference communities and the habitat conditions that support them for mesohaline zone (5–18 ppt).

Phytoplankton reference communities and the habitat conditions that support them for mesohaline zone (5–18 ppt).
Phytoplankton reference communities and the habitat conditions that support them for mesohaline zone (5–18 ppt).
Table 9.

Phytoplankton reference communities and the habitat conditions that support them for polyhaline zone (>18 ppt).

Phytoplankton reference communities and the habitat conditions that support them for polyhaline zone (>18 ppt).
Phytoplankton reference communities and the habitat conditions that support them for polyhaline zone (>18 ppt).

The median values and interquartile ranges, and 25th and 75th percentiles of selected physiological and chemical metrics and phytoplankton metrics in reference and impaired conditions in six season-salinity zones are shown in Figures 4 and 5. Fall and winter MH zones are not shown because fewer data were available for such comparison. The concentrations of TSS and DOC, particularly in summer, were higher in the impaired condition compared with the reference. In contrast, the concentration of DO in the impaired condition was often lower than that in the reference. The concentrations of TN and TP in the reference condition were significantly lower than those in the impaired, except for spring-MH, when there were few data available for statistical comparison (Figure 4). Our results show significant differences of several phytoplankton metrics between the reference and impaired conditions (Figure 5). Chl a in the reference condition was lower and its variation was smaller compared with those in the impaired condition. In most season-salinity zones, the average cell size of the NM phytoplankton tended to be smaller in the degraded environment than that in the reference condition. Summer picoplankton biomass (as carbon) and its percentage of total phytoplankton biomass were generally higher in the impaired communities than in the reference ones.

Figure 4.

Comparison of some physiochemical metrics between the least-impaired and impaired conditions. L-Imp, least-impaired; Imp, impaired. Sample sizes (as L-Imp–Imp): spring MH, 4–2; spring PH, 13–30; summer MH, 4–12; summer PH, 12–38; fall PH, 9–9; winter PH, 13–14.

Figure 4.

Comparison of some physiochemical metrics between the least-impaired and impaired conditions. L-Imp, least-impaired; Imp, impaired. Sample sizes (as L-Imp–Imp): spring MH, 4–2; spring PH, 13–30; summer MH, 4–12; summer PH, 12–38; fall PH, 9–9; winter PH, 13–14.

Figure 5.

Comparison of some phytoplankton metrics between the least-impaired and impaired conditions. L-Imp, least-impaired; Imp, impaired. Sample sizes (as L-Imp–Imp): spring MH, 4–2; spring PH, 13–30; summer MH, 4–12; summer PH, 12–38; fall PH, 9–9; winter PH, 13–14. For picoplankton: sample sizes (as L-Imp–Imp): summer MH, 4–12; summer PH, 5–23; fall PH, 3–3.

Figure 5.

Comparison of some phytoplankton metrics between the least-impaired and impaired conditions. L-Imp, least-impaired; Imp, impaired. Sample sizes (as L-Imp–Imp): spring MH, 4–2; spring PH, 13–30; summer MH, 4–12; summer PH, 12–38; fall PH, 9–9; winter PH, 13–14. For picoplankton: sample sizes (as L-Imp–Imp): summer MH, 4–12; summer PH, 5–23; fall PH, 3–3.

P-IBI Metrics

Table 10 lists the 34 phytoplankton and physicochemical metrics evaluated for their discriminatory ability between the least-impaired and impaired habitat conditions. In spring and summer PH zones, up to half of the metrics showed significant distinction between the least-impaired and impaired categories (Kruskal–Wallis test). MH had fewer metrics showing strong discriminatory power, mainly due to (1) MH in BB-LEH is a relatively small area but receiving high discharge of freshwater and nutrients; and (2) few data points were available in the current data set for such comparison to reach statistical significance. However, several metrics showed good discriminatory ability in more than one season-salinity category; therefore, we selected them to form a P-IBI. These metrics included Chl a, Chl a/C ratio, total abundance of NM phytoplankton and average NM phytoplankton cell size in summer, percent diatoms, percent dinoflagellate biomass, percent cryptophyte biomass, summer percent picoplankton, and percent cyanobacteria biomass. In addition, physiological and chemical parameters including DO, DOC/TOC, and TSS, were selected as well for their strong discriminatory ability (Table 10). P-IBI metrics for spring and summer MH and PH zones, together with scoring criteria of each metric, are summarized in Table 6.

Table 10.

Phytoplankton metrics and their discriminatory ability for significant differences between least-impaired and impaired communities examined by Kruskal–Wallis test.

Phytoplankton metrics and their discriminatory ability for significant differences between least-impaired and impaired communities examined by Kruskal–Wallis test.
Phytoplankton metrics and their discriminatory ability for significant differences between least-impaired and impaired communities examined by Kruskal–Wallis test.

Classification Efficiency of P-IBI

The DE of each major metric in P-IBI and the CEs of P-IBI in the least-impaired, impaired, and overall conditions are shown in Table 11. The DEs of the metrics percent DT (percentage of diatom biomass) and Chl a are generally higher than 60% for the impaired and the least-impaired samples, indicating that the two metrics have a high probability for correctly identifying both least-impaired and impaired habitat conditions. TOC showed good discrimination between the least-impaired and impaired in the summer MH samples, whereas DOC and Chl a/C ratio had good DE for summer PH samples, especially for the least-impaired ones (Table 11). Picoplankton metrics, including its abundance, biomass, and percent picoplankton biomass, seemed very efficient in discriminating summer MH samples, especially for the least-impaired ones on the basis of the present data pool. The current P-IBI was able to correctly classify 64–100% of the least-impaired samples and 53–100% of the impaired samples from the calibration data set. Higher CE was achieved for the impaired samples in summer than spring for the MH and PH zones. Overall, the P-IBI correctly classified 64–100% of spring samples and 68–88% of summer samples in the calibration data set (Table 11).

Table 11.

Discrimination efficiencies of individual metrics and classification efficiencies of the P-IBI in least-impaired (L-Imp) and impaired (Imp) conditions for spring and summer mesohaline (MH) and polyhaline (PH) zones.

Discrimination efficiencies of individual metrics and classification efficiencies of the P-IBI in least-impaired (L-Imp) and impaired (Imp) conditions for spring and summer mesohaline (MH) and polyhaline (PH) zones.
Discrimination efficiencies of individual metrics and classification efficiencies of the P-IBI in least-impaired (L-Imp) and impaired (Imp) conditions for spring and summer mesohaline (MH) and polyhaline (PH) zones.

Actual separation of the P-IBI score for the least-impaired and impaired communities is shown in Figure 3. Except for the summer PH, the 25th percentile of P-IBI scores in the least-impaired were higher than or equal to the 75th percentile of P-IBI scores in the impaired distributions. The high degree of separation demonstrates the discriminatory power of the current P-IBI for spring and summer communities in BB-LEH.

Figure 3.

Distribution of P-IBI scores for the least-impaired (reference) and impaired (degraded) communities. The interquartile range, median value (lines within the bars), and 5th and 95th percentiles (lines below and above each bar) are displayed for summer and spring mesohaline (MH) and polyhaline (PH) zones.

Figure 3.

Distribution of P-IBI scores for the least-impaired (reference) and impaired (degraded) communities. The interquartile range, median value (lines within the bars), and 5th and 95th percentiles (lines below and above each bar) are displayed for summer and spring mesohaline (MH) and polyhaline (PH) zones.

DISCUSSION

Phytoplankton Habitat Conditions

The salinity gradient in Barnegat Bay is mainly affected by freshwater discharge, tides and water circulation, and exchange with the adjacent oceanic water (Kennish, 2001). Barnegat Bay, particularly in the north, receives a majority of freshwater discharges. Toms River is the largest source of inflowing freshwater and nutrients to the northern segment of Barnegat Bay. As a result, the MH condition was often detected near the mouth and within the plume of Toms River, spreading to southern Silver Bay. Barnegat Inlet in the center and Little Egg Inlet at the southern end directly connect the estuary to the ocean. A recent hydrological study demonstrated a pronounced northward subtidal flow from the Little Egg Inlet (Defne and Ganju, 2015), transporting marine water to the north. The majority of BB-LEH is PH, covering areas south of the Toms River and around Barnegat Inlet, Manahawkin Bay, and Little Egg Harbor. The northern end of Barnegat Bay is connected to the ocean through the Point Pleasant Canal and Manasquan River. Site BB01 near Mantoloking was often found to be PH because of the influences of both the indirect connection with the oceanic water and the subtidal force from the south. It's worth mentioning, despite being in the class of PH, that salinity in the northern end of the Bay (BB01) and south of Toms River (BB05) generally ranges from 18 to 25 ppt, indicating the effect of freshwater discharge, whereas salinity in southern Barnegat Bay and Little Egg Harbor area (sites BB07, 09, 10, 12 and 14) is generally higher than 25 ppt and up to 33 ppt. Salinity is one of the most controlling factors for phytoplankton species composition. Cluster analysis showed that phytoplankton communities at BB01 and BB05 were significantly distinct from those in southern Barnegat Bay and Little Egg Harbor (see project report for details; Ren, 2015). We combined both regions into the same salinity zone in the study following the traditional salinity classification and methods for Chesapeake Bay (Buchanan et al., 2005). However, it may be necessary for future refinement purposes, when more data are available, to distinguish the current PH into two salinity zones, 18–25 ppt and >25 ppt, in consideration of the differences in phytoplankton communities.

Light and nutrient availability are the primary factors for phytoplankton growth and species development (Cloern, 1999). The majority of our samples (73%) from 2011–13 fell into two categories: MBL and MPL. There were six combinations in MBL and MPL, distinguished primarily by water clarity and secondarily by nutrient availability (Table 5). In particular, combo #6 in category MBL presented high water clarity and excess DIN and PO4, a favorable condition for possible phytoplankton growth and biomass increase. Such conditions can easily change to category MPL or even to category poor if provided with continuous nutrient replenishment from external sources. Combo #6 was found more often in summer near Barnegat Inlet and Little Egg Harbor. More than 52% of the samples were in MPL, which was over 2.5 times more than MBL (Figure 2), indicating MPL a common condition in the BB-LEH estuary.

BB01 in spring and BB07a in summer had high frequency of the BB (and B) category compared with other sites (Table 7, Figure 2). Both sites were in the PH zone most of the time, the same salinity class as southern BB and LEH, despite salinity at BB01 often being lower than BB07 and other sites in the south. Water clarity at BB01 is often improved with less-suspended solids because of relatively long residence time in comparison with the southern sites (Defne and Ganju, 2015). Similarly, at site BB07a, located near the Intracoastal Waterway, the water column is deeper than other sites. Both sites may have elevated light condition compared with other PH sites because of their specific hydrophysical features. The classification criteria for light conditions, however, were calculated on the basis of all the data points from PH samples, of which approximately one-third were from BB01 and BB07 and more than two-thirds were from sites BB09, 10, 12, and 14. It is possible that the current light criterion for the classes of better/B may slightly favor sites BB01 and BB07 over other PH sites.

Phytoplankton Reference Communities

In general, more metrics showed significant difference between the least-impaired (reference) and impaired conditions, and higher significance of difference (p-value) was obtained for the PH zone than for the MH. This is likely because (1) very few data points were available for MH; and (2) the MBL samples were included to augment the data pool for the quantification of reference communities for MH.

Phytoplankton reference communities in both MH and PH zones, compared with the impaired ones, were characterized with lower Chl a, TSS, lower TN and TP, lower Chl a/C ratio, lower summer picoplankton, and higher DO concentrations. These results are in agreement with studies on the relationship between nutrient enrichment, phytoplankton growth, and water quality in various estuarine and coastal ecosystems (Hoyer et al., 2002; Nielsen et al., 2002; Rabalais and Nixon, 2002; Smith and Schindler, 2009; Turner et al., 2007). These results are also coincident with those from Chesapeake Bay. The reference values for some parameters, such as Chl a, Chl a/C, average NM cell size, and DO, were comparable with those in the same salinity zones in Chesapeake Bay (Buchanan et al., 2005). TSS was generally higher than that in Chesapeake Bay both in MH and PH areas because of shallow water depth and high suspended matter in BB-LEH. For picoplankton biomass, the boundaries of reference biomass in summer were comparable with those in Chesapeake Bay. However, the median value of picoplankton biomass in MH was only one-third of that in Chesapeake Bay. One possible reason was that most of the BB + B samples in this study were collected in June when picoplankton abundance was usually one or two magnitudes lower compared with its peak in August/September (Ren, 2013, 2015). The calculated Chl a concentration in MH reference condition was generally higher than that in PH (Table 8), indicating elevated phytoplankton biomass due to riverine nutrient inputs, mainly from Toms River. The boundaries of the summer reference Chl a in PH were 1.3–7.4 μg/L (Table 9). The values were comparable but relatively lower than those from a 1969–70 investigation that showed the concentration of Chl a varied from 5 to 12 μg/L in summer in middle and lower Barnegat Bay (Mountford, 1984). Some of the collection sites in the 1969–70 survey were located on the west side of the Bay near the discharges of Stouts Creek, Forked River, and Oyster Creek. Our sites (BB07a and BB09), however, are located in the middle of the Bay. Phytoplankton growth and biomass may have been affected more by the freshwater and nutrient input from those streams. In the 1970s, the septic system and waste dumping had been a point source for organic matter pollution to the Barnegat Bay and had caused low-level or acute degradation of the system. Although the pollution had been briefly controlled by the removal of the large sewage outfalls and installation of regional treatment plants, an increase in both low-level and acute degradation was shown from the late 1980s to early 1990 (Carter, 2001). Average summer Chl a in the northern end (Mantoloking) and southern Barnegat Bay ranged from 12 to 18 μg/L during 1987–98 (Olsen and Mahoney, 2001). Those values exceeded the upper boundary of Chl a in the reference communities. They were also higher than the median values of Chl a in the PW + W + MPL categories (9.3 μg/L, range 3–20 μg/L), suggesting nutrient overenrichment in the 1980s–90s in BB-LEH. More historic Chl a data before the 1970s could help better understand the ecological evolution of nutrient enrichment in BB-LEH.

The BB-LEH estuary is a shallow, poorly flushed system bordered by a highly developed watershed. The ecological health of the estuary has deteriorated over the last few decades, and the estuary has been classified as a highly eutrophic system (Kennish, Haag, and Sakowicz, 2010; Nixon, 1995). The calculated reference communities and their supporting conditions represented the present-day least-impaired water quality in BB-LEH. They may not be the same as “historical” or “pristine” condition/sites as defined and described by other approaches (Gibson et al., 2000; NRC, 2000). These populations, however, were considered as the present-day minimally affected communities in each specific season-salinity habitat. Reference conditions are important for water-quality assessment and restoration efforts for any specific water body. The calculated median values and boundaries of the metrics in reference communities, though preliminary, can be helpful for water-quality evaluation in the BB-LEH system.

TN and TP are considered primary nutrient causal factors for estuarine and coastal eutrophication (USEPA, 2001). A positive relationship between TN and TP and Chl a in estuarine and coastal ecosystems has been revealed in several other studies (Guildford and Hecky, 2000; Smith, Tilman, and Nekola, 1999). In BB-LEH, nutrient loading, both TN and TP, has been repeatedly cited as a primary cause of ecosystem eutrophication (Kennish, Fertig, and Lathrop, 2010). Multivariate analysis showed that TN and TP were the major factors significantly affecting phytoplankton composition in BB-LEH, and the dominance of picoplankton and cyanobacteria in summer was significantly related to TN and DOC (Ren, 2015; Ren et al., unpublished data). This study showed that the concentrations of TN and TP in the least-impaired communities were significantly lower than those in the impaired ones in most season-salinity zones (Figure 4), except for spring MH when limited data are available for statistical comparison. This result is in agreement with earlier findings that related the water-quality degradation in BB-LEH to the nutrient (TN and TP) loading (Kennish, Fertig, and Lathrop, 2010). The comparison of the TN and TP values in the least-impaired and impaired habitat conditions from this study provided evidence for dual reduction of N and P loading in BB-LEH, as recommended for some other estuarine and coastal regions (Pearl, 2009; Smith and Schindler, 2009). The results from this study can be useful for the development of nutrient criteria for the region and guidance for the control of external nutrient loading for BB-LEH.

P-IBI Metrics

Several metrics of phytoplankton community composition turned out to be important for P-IBI, e.g., NM phytoplankton, summer picoplankton, and taxonomic groups such as diatoms, cryptophytes, chrysophytes, and cyanobacteria (Table 10). NM phytoplankton is defined as species and specimens with cell size in the range of 2 to 200 μm. Phytoplankton in this size category include most species in major taxonomic groups except for the picoplankton (mainly 1–4 μm in diameter) and cyanobacteria (mostly <1 μm). The abundance of NM phytoplankton is a metric indicative of phytoplankton abundance in exclusion of picoplankton and small cyanobacteria. During the investigation, over 140 taxa were recorded from both years; more than half were diatoms. The size of diatom cells ranged from nearly 100 μm long, such as some Rhizosolenia species, to some small species of Cyclotella and Thalassiosira only 3–5 μm in diameter. Other major groups included dinoflagellates and cryptophytes with about 25 and 8 common taxa, respectively (Ren, 2013; Ren et al., unpublished data). The cell size in dinoflagellates varied from >50 μm long, such as some Ceratium spp., to <10 μm of some Gyrodinium species. The cell size of major species in cryptophytes ranged from ∼25 to 5 μm. A list of recorded taxa and the biovolume of major species can be found in Ren (2013). The average NM phytoplankton cell size was calculated from total NM biovolume divided by total NM abundance, and is an indicator of food quality for planktonic grazers (Buchanan et al., 2005). Taxonomic groups such as diatoms, dinoflagellates, chrysophytes, and cryptophytes and their abundance and biomass percentage were important components of the index. These metrics reflect the biodiversity of phytoplankton in the system, which is an essential indicator for a heathy ecosystem. Calculated metric scoring criteria showed that in class 1 (as impaired community), the biomass percentage of each major taxonomic group could be either very low or very high (Table 6). For example, in summer when the percentage of diatom biomass was >54% in MH and >67.9% in PH conditions, the percentage of some other taxonomic group, such as cryptophytes, became very low (∼1%). A least-impaired community, however, was more balanced in community composition as shown in the criteria for class 5 (Table 6). Picoplankton and cyanobacteria are less favorable food sources for shellfish and clams (Bricelj, 1999). In summer, the biomass and percentage of picoplankton in the impaired condition were significantly higher than that in the least-impaired condition (Figure 5). High abundance or percentage of picoplankton (>1.2 × 108 cells/L, or >43% in Table 6) indicates impaired habitat in MH and PH areas. Microscopic analysis on samples showed that overdominance of picoplankton in summer was often associated with very low percentage of other taxonomic groups. Multivariate analysis showed that summer phytoplankton abundant with picoplankton, cryptophytes, and dinoflagellates could be an indication of silica limitation in the water column, especially in northern Barnegat Bay (Ren et al., unpublished data). A high percentage of picoplankton, including cyanobacteria in summer, is an indicator for nutrient overenrichment in estuarine and coastal ecosystems (Olson and Mahoney, 2001; Ren et al., unpublished data).

DOC and TOC can be metrics indicative of the physiological status of phytoplankton. Principal component analysis (PCA) showed that Chl a was significantly correlated with DOC and TOC in both years, with p < 0.001 in year 1 and p < 0.01 in year 2 (Ren et al., unpublished data). The sources of DOC and TOC include phytoplankton extracellular and cellular excretion and lysis due to zooplankton grazing and other stresses, and particulate parts of phytoplankton/zooplankton (as TOC). DOC/TOC is also an important parameter related to microbial metabolism and DO level in the water column (Lacouture et al., 2006). DO is a metric reflecting biological metabolism of an aquatic system including phytoplankton photosynthesis, respiration, and microbial decomposition. It is an important indicator for water-quality assessment. Hypoxia, a condition when DO in water is less than 2 mg/L, is one of the major concerns in some coastal ecosystems, especially where summer stratification takes place (Rabalais and Nixon, 2002; Rabalais and Turner, 2001). BB-LEH is a relatively shallow system with an average depth of 1.5 m. The water column is generally well mixed and homogeneous because of the effect of winds, tides, and water exchange and circulation. The concentration of DO was generally higher than 2 mg/L in the water column. Our results showed a significantly higher concentration of DO in the least-impaired condition compared with that in the impaired condition, especially in spring and summer (Figure 4). The differences of DO in fall and winter were not as significant as spring and summer, likely reflecting low biological metabolism in the water column because of decreasing temperature. The DE of DO for the least-impaired condition was considerably good (96–100%, Table 11). However, the performance of DO for the impaired condition was relatively poor (Table 11), suggesting that low DO may have been driven more by higher bacterial consumption than lower phytoplankton production in degraded habitat condition. TSS is not usually dominated with phytoplankton biomass in many estuaries. The correlation between Chl a and TSS is less significant compared with DOC and TOC. However, the significance level p was still about 0.08 on the basis of PCA results from year 1 data. No significant correlation was found from year 2 data, most likely due to the disruption of Hurricane Sandy (Ren et al., unpublished data). In current P-IBI, TSS was included as phytoplankton metric in consideration of its detrimental effect of phytoplankton self-shading on, e.g., submerged aquatic vegetation in Barnegat Bay, especially in summer when picoplankton reaches high abundance (Kennish, Fertig, and Sakowicz, 2011; Kennish, Haag, and Sakowicz, 2010; Kennish et al., 2007).

Chl a and Chl a/C ratio are two metrics indicating phytoplankton photosynthesis activity. Chl a is one of the key parameters in routine water-quality monitoring as phytoplankton biomass because it is relatively easy to measure. Cellular Chl a contents in the same species may vary with light, temperature, and nutrient composition and availability in the water body. The total concentration of Chl a is considered a good indicator for total phytoplankton biomass. Different from Chl a, carbon biomass is usually calculated by multiplying the cell biovolume by carbon/biovolume ratios. The carbon/biovolume ratios vary with different species or taxonomic groups (Mullin, Sloan, and Eppley, 1966; Stramski, 1999). Decoupling between Chl a biomass and biovolume-based biomass may occur because of the effects of light, temperature, nutrients, and taxonomic composition on Chl a content (Felip and Catalan, 2000; Geider, 1987). It is suggested that having both Chl a and Chl a/C in the P-IBI contributes to a stronger overall index (Lacouture et al., 2006).

Limitation and Uncertainty of the P-IBI

The developed P-IBI showed considerable power in discriminating the least-impaired and impaired communities (Figure 3). However, there was some limitation and uncertainty in the development of the P-IBI due to the limited data availability.

Indicator species analyses showed that several individual taxa can serve as metrics for P-IBI for different nutrient regimes and season-salinity categories (see details in project report; Ren and Belton, 2015). However, the current database does not provide sufficient data for most of the indicator species to effectively discriminate different habitat conditions. Brown tide, the bloom of Aureococcus anophagefferens, has been a major concern in the BB-LEH system. It has occurred episodically since the first confirmed detection in 1995. The study showed that the abundance of A. anophagefferens is positively associated with warmer water temperature and higher salinity, and may also be related to extended drought conditions (Gastrich et al., 2004; Mahoney, Olsen, and Jeffress, 2006). We detected relatively low density of A. anophagefferens (105–106 cells/L) in southern Barnegat Bay. An incidence of A. anophagefferens bloom was detected near Sedge Island on 19 June 2013 (4.5 × 108 cells/L; Fantasia, Bricelj, and Ren, 2017). In addition, two harmful dinoflagellates, Prorocentrum minimum and Heterocapsa rotundata, commonly detected in BB-LEH, were previously found to be associated with different forms of N nutrients and water-quality conditions (Rothenberger, Burkholder, and Wentworth, 2009). However, none of these species showed significant discriminatory ability between the least-impaired and impaired conditions. In addition, pheophytin, degradation product of Chl a, is considered an effective metric for P-IBI (Lacouture et al., 2006). However, this parameter is not available in the water-quality data set.

Furthermore, Hurricane Sandy made landfall on New Jersey on 29 October 2012 and caused unprecedented disruption to the BB-LEH system. Study showed that phytoplankton assemblages after the hurricane (in winter 2012 and spring 2013) were significantly different from those from the same time in the previous year, especially in the northern area of the Bay (Ren, 2015). Significant interannual changes of phytoplankton communities have also been observed in southern Barnegat Bay and Little Egg Harbor. As a result, the calculated phytoplankton metrics may exhibit considerable uncertainty. Metrics such as cryptophyte biomass and abundance, percent cryptophyte biomass, and percent diatoms to total biomass showed strong discriminatory ability between the least-impaired and impaired communities, especially in PH zones (Table 10). However, the values of those metrics in the least-impaired communities often showed large overlap with those in the impaired communities (figures not shown). The DEs of individual metrics such as Chl a and %DT biomass varied from 75% to 83%, indicating a good probability of correctly identifying impaired habitat condition and least-impaired condition in summer MH (Table 11). However, the DEs of most metrics varied between 45% and 63%, suggesting moderate effectiveness in discriminating least-impaired and impaired conditions. Except for spring MH with 100% CE, the CE of the P-IBI, 64–88% for spring PH and summer MH and PH, are slightly lower compared with those in the Chesapeake Bay (70–84.4%, Lacouture et al., 2006). More data collections on phytoplankton community together with the water quality monitoring can certainly help understand the development of the BB-LEH system after Hurricane Sandy, and are essential to refine and strengthen the calculated P-IBI.

CONCLUSIONS

A multimetric P-IBI can be an effective functional tool for water quality assessment and management in the estuarine and coastal ecosystems, as shown by the studies done a decade ago for Chesapeake Bay (Buchanan et al., 2005; Lacouture et al., 2006) and its recent validation (Johnson and Buchanan, 2014). We followed the same procedures and made the first attempt to develop a P-IBI specifically for BB–LEH. During the study we established light and nutrient (DIN and PO4) criteria for the classification of habitat conditions, quantified phytoplankton reference communities, tested 34 phytoplankton physiological and chemical metrics for their discriminatory abilities, and developed a P-IBI for four season-salinity zones (spring MH and PH, and summer MH and PH) for BB-LEH. About 10 to 12 phytoplankton metrics and four to five physiological and chemical metrics have shown significant differences between the least-impaired and impaired conditions. The current P-IBI was able to correctly classify 64–100% for spring samples and 68–88% for summer samples in the calibration data set. The high degree of separation of the P-IBI scores between the least-impaired and impaired demonstrates the discriminatory power of the current P-IBI for the spring and summer communities in BB-LEH. In addition, our results showed that the concentrations of TN and TP were significantly lower in the reference conditions than in the impaired conditions, suggesting that a dual reduction for TN and TP is necessary to control eutrophication in BB-LEH. These results are useful in assessing and evaluating phytoplankton habitat conditions, as well as developing nutrient criteria for water-quality management in BB-LEH. This study, though preliminary, has demonstrated that the multimetric P-IBI method, which has been used successfully in several ecosystems, is working for BB-LEH. Our work provided the basic and insightful information for further developing and refining P-IBI for BB-LEH.

Although the P-IBI provides an effective way to interpret large amounts of monitoring data and insightful information on phytoplankton communities in relation to water quality conditions, the application and interpretation of phytoplankton reference communities and the P-IBI for BB-LEH is constrained by limited data availability. More data were available for the PH zone, particularly for spring and summer; however, it is still modest compared with the 18 years of long-term data sets for P-IBI development in the Chesapeake Bay. Large interannual variability in phytoplankton community has been observed between the 2 years in our study due not only to natural variability but also the disruption of Hurricane Sandy coming ashore in New Jersey at BB-LEH (Ren, 2015; Ren et al., unpublished data). As a result, the calculated phytoplankton reference communities and developed P-IBI in this study may exhibit considerable uncertainty. More simultaneously collected phytoplankton and water-quality data are necessary to reduce the uncertainty and deviation caused by disturbance from the large storm event. In addition, it is necessary to further refine the salinity classifications as well as light and nutrient criteria. It is also important to re-examine the phytoplankton metrics and test additional ones for their discriminatory ability as a refinement effort for the calculated P-IBI as more data become available. It is important to note that this study is part of the ongoing research initiative on phytoplankton communities in BB-LEH. Two more years of data, April 2014 to April 2015 and May 2016 to May 2017, will become available and be used for the refinement of the P-IBI. We believe that the P-IBI, once the refinement effort is complete, will be an effective management tool facilitating water quality monitoring and assessment in BB-LEH, which is essential in practical restoration efforts based on empirical data.

ACKNOWLEDGMENTS

We sincerely thank Bill Heddendorf and the field crew from the Bureau of Marine Water Monitoring, NJDEP) for their dedicated efforts on sample collections. We thank Patricia Ingelido and Helen Pang for their assistance on water quality data. We thank Elena Colon for her laboratory assistance and Dr. Donald F. Charles for his support during the work. We are grateful to three anonymous reviewers for their instructive comments on the manuscript. This work was funded by NJDEP through an agreement with New Jersey Sea Grant Consortium (project no. 4904-0032, NJDEP no. SR14-009).

LITERATURE CITED

LITERATURE CITED
Alden,
R.W.
and
Perry,
E.S.,
1997
.
Presenting measurements of status: Report to the Chesapeake Bay Program monitoring subcommittee's Data Analysis Workgroup
.
Annapolis, Maryland
:
Chesapeake Bay Program
.
Anonymous
,
1958
.
The Venice system of the classification of marine waters according to salinity
.
Limnology and Oceanography
,
3
(
3
),
346
347
.
Barnegat Bay LMP QAPP
,
2013
.
Barnegat Bay Long Term Ambient Monitoring Program
.
Trenton
:
New Jersey DEP Water Monitoring and Standards
.
Bricelj,
V.M.,
1999
.
Perspectives on possible factors influencing the abundance of hard clams
.
In:
Schlenk,
C.G.
(
ed.
),
Workshop on Hard Clam Population Dynamics Research Priorities for the South Shore of Long Island. Port Jefferson, New York
.
Stony Brook
:
NY Sea Grant
.
Buchanan,
C.;
Lacouture,
R.V.;
Marshall,
H.G.;
Olson,
M.,
and
Johnson,
J.M.,
2005
.
Phytoplankton reference communities for Chesapeake Bay and its tidal tributaries
.
Estuaries
,
28
(
1
),
138
159
.
Carter,
G.P.,
2001
.
Eight characterizing indicators in the Barnegat Bay watershed, Ocean County, New Jersey
.
In:
Kenish, M.J. (ed.),
Characterization of the Barnegat Bay – Little Egg Harbor, New Jersey, Estuarine System and Watershed Assessment
.
Journal of Coastal Research
,
Special Issue No.
32
, pp.
82
101
.
Cloern,
J.E.,
1999
.
The relative importance of light and nutrient limitation of phytoplankton growth: A simple index of coastal ecosystem sensitivity to nutrient enrichment
.
Aquatic Ecology
,
3
(
1
),
3
15
.
Defne,
Z.
and
Ganju,
N.K.,
2015
.
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
. doi:
Desianti,
N.;
Potapova,
M.;
Enache,
M.;
Belton,
T.J.;
Velinsky,
D.J.;
Thomas,
R.,
and
Mead,
J.,
2017
.
Sediment diatoms as environmental indicators in New Jersey coastal lagoons
.
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
.
127–140
.
Devlin,
M.;
Best,
M.;
Bresnan,
E.;
O'Boyle,
S.,
and
Park,
R.,
2007
.
Establishing boundary classes for the classification of UN marine waters using phytoplankton communities
.
Marine Pollution Bulletin
,
55
(
1–6
),
91
103
.
Dortch,
Q.;
Robichaux,
R.;
Pool,
S.;
Milsted,
D.;
Mire,
G.;
Rabalais,
N.N.;
Soniat,
T.M.;
Fryxell,
G.A.;
Turner,
R.E.,
and
Parsons,
M.L.,
1997
.
Abundance and vertical flux of Pseudo-nitzchia in the northern Gulf of Mexico
.
Marine Ecology Progress Series
,
146
,
249
264
.
Eppley,
R.W.;
Reid,
F.M.H.,
and
Strickland,
J.D.H.,
1970
.
Estimates of phytoplankton crop size, growth rate, and primary production
.
Bulletin of the Scripps Institute of Oceanography
,
17
(
1
),
33
42
.
European Commission
,
2000
.
Directive 2000/60/EC of the European Parliament and Council of 23 October 2000 establishing a framework for community action in the field of water policy
.
Official Journal of the European Community
,
L327
.
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
.
Journal of Coastal Research
, Special Issue No.
sp1
,
pp
.
106–126
.
Felip,
M.
and
Catalan,
J.,
2000
.
The relationship between phytoplankton biovolume and chlorophyll in a deep oligotrophic lake: Decoupling in their spatial and temporal maxima
.
Journal of Plankton Research
,
22
(
1
),
91
105
.
Fisher,
T.R.
and
Gustafson,
A.B.,
2003
.
Nutrient-Addition Bioassays in Chesapeake Bay to Assess Resources Limiting Algal Growth. Progress report: August 1990–December 2002
.
Prepared for Maryland Department of Natural Resources
,
Chesapeake Bay Water Quality Monitoring Program
,
by University of Maryland Horn Pint Laboratory, Cambridge, Maryland
.
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
.
Geider,
R.J.,
1987
.
Light and temperature dependence of the carbon to chlorophyll a ratio in microalgae and cyanobacteria: Implications for physiology and growth of phytoplankton
.
New Phytologist
,
106
(
1
),
1
34
.
Gibson,
G.R.;
Bowman,
A.B.;
Gerritsen,
J.,
and
Snyder,
B.D.,
2000
.
Estuarine and coastal marine waters: Bioassessment and biocriteria technical guidance. EPA 822-B-00-024
.
Washington, D.C
.:
Office of Water, U.S. Environmental Protection Agency
.
Guildford,
S.J.
and
Hecky,
R.E.,
2000
.
Total nitrogen, total phosphorus, and nutrient limitation in lakes and oceans: Is there a common relationship?
Limnology and Oceanography
,
45
(
6
),
1213
1223
.
Hillebrand,
H.;
Dürselen,
C.D.;
Kirschtel,
D.;
Pollingher,
U.,
and
Zohary,
T.,
1999
.
Biovolume calculation for pelagic and benthic microalgae
.
Journal of Phycology
,
35
(
2
),
403
424
.
Hoyer,
M.V.;
Frazer,
T.K.;
Notestein,
S.K.,
and
Canfield,
D.E.,
Jr.,
2002
.
Nutrient, chlorophyll, and water clarity relationships in Florida's nearshore coastal waters with comparisons to freshwater lakes
.
Canadian Journal of Fisheries and Aquatic Sciences
,
59
(
6
),
1024
1031
.
Johnson,
J.M.
and
Buchanan,
C.,
2014
.
Revisiting the Chesapeake Bay phytoplankton index of biotic integrity
.
Environmental Monitoring and Assessment
,
186
(
3
),
1431
1451
.
Jordan,
S.J.
and
Vaas,
P.A.,
2000
.
An index of ecosystem integrity for northern Chesapeake Bay
.
Environmental Science and Policy
,
3
(
S1
),
59
88
.
Karr,
J.R.,
1981
.
Assessment of biotic integrity using fish communities
.
Fisheries
,
6
(
6
),
21
27
.
Kauppila,
P.,
2007
.
Phytoplankton quantity as an indicator of eutrophication in Finnish coastal waters. Applications within the Water Framework Directive
.
Monographs of the Boreal Environmental Research
,
31
.
Helsinki
:
Finnish Environment Institute
,
58
p
.
Kennish,
M.J.,
2001
.
Physical description of the Barnegat Bay–Little Egg Harbor estuarine system
.
In:
Kenish, M.J. (ed.),
Characterization of the Barnegat Bay – Little Egg Harbor, New Jersey, Estuarine System and Watershed Assessment
.
Journal of Coastal Research
,
Special Issue No.
32
, pp.
13
27
.
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
(
S5
),
3
16
.
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
,
23
30
.
Kennish,
M.J.;
Fertig,
B.M.,
and
Lathrop,
R.G.,
2010
.
Assessment of Nutrient Loading and Eutrophication in Barnegat Bay–Little Egg Harbor, New Jersey in Support of Nutrient Management Planning
.
Report to New England Interstate Water Pollution Cotrol Commission
.
Kennish,
M.J.;
Fertig,
B.,
and
Sakowicz,
G.P.,
2011
.
Benthic macroalgal blooms as an indicator of system eutrophy in the Barnegat Bay–Little Egg Harbor estuary
.
Bulletin of the New Jersey Academy of Sciences
,
56
(
1
),
1
5
.
Kennish,
M.J.;
Haag,
S.M.,
and
Sakowicz,
G.P.,
2010
.
Seagrass decline in New Jersey coastal lagoons: A response to increasing eutrophication
.
In:
Kennish,
M.J.,
and
Paerl,
H.W.
(
eds.
),
Coastal Lagoons: Critical Habitats of Environmental Change
.
Boca Raton, Florida
:
CRC Press
,
pp
.
167
201
.
Lacouture,
R.V.;
Johnson,
J.M.;
Buchanan,
C.,
and
Marshall,
H.G.,
2006
.
Phytoplankton index of biotic integrity for Chesapeake Bay and its tidal tributaries
.
Estuaries and Coasts
,
29
(
4
),
598
616
.
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
.
Martinez-Crego,
B.;
Alcoverro,
T.,
and
Romero,
J.,
2010
.
Biotic indices for assessing the status of coastal waters: A review of strengths and weaknesses
.
Journal of Environmental Monitoring
,
12
(
5
),
1013
1028
.
Mountford,
K.,
1984
.
Phytoplankton
.
In:
Kennish,
M.J.
and
Lutz,
R.A.
(
eds
),
Ecology of Barnegat Bay, New Jersey
.
New York
:
Springer-Verlag
. doi:
Mullin,
M.M.;
Sloan,
P.R.,
and
Eppley,
R.W.,
1966
.
Relationship between carbon content, cell volume, and area in phytoplankton
.
Limnology and Oceanography
,
11
(
2
),
307
311
.
National Research Council
,
2000
.
Clean Coastal Waters: Understanding and Reducing the Effects of Nutrient Pollution
.
Washington, D.C
.:
National Academies
,
428
p
.
Nielsen,
S.L.;
Sand-Jensen,
K.;
Borum,
J.,
and
Geertz-Hansen,
O.,
2002
.
Phytoplankton, nutrients and transparency in Danish coastal waters
.
Estuaries
,
25
(
5
),
930
937
.
Nixon,
S.W.,
1995
.
Coastal eutrophication: A definition, social causes, and future concerns
.
Ophelia
,
41
(
1
),
199
220
.
Olenina,
I.;
Hajdu,
S.;
Edler,
L.;
Andersson,
A.;
Wasmund,
N.;
Busch,
S.;
Goebel,
J.;
Gromisz,
S.;
Huseby,
S.;
Httunen,
M.;
Jaanus,
A.;
Kokkonen,
P.;
Ledaine,
I.,
and
Niemkiewicz,
E.,
2006
.
Biovolumes and size-classes of phytoplankton in the Baltic Sea
.
HELCOM Baltic Sea Environment Proceedings
,
106
,
144
p
.
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:
Kenish, M.J. (ed.),
Characterization of the Barnegat Bay – Little Egg Harbor, New Jersey, Estuarine System and Watershed Assessment
.
Journal of Coastal Research
,
Special Issue No.
32
, pp.
115
143
.
Olson,
M.,
2002
.
Benchmarks for nitrogen, phosphorus, chlorophyll and suspended solids in Chesapeake Bay
.
Annapolis, Maryland
:
Chesapeake Bay Program Technical Report Series
.
Olson,
M.,
2009
.
Relative status indicator. Development and evolution of a relative measure of condition for assessing the status of water quality and biological parameters tracked in the US/EPA Chesapeake Bay Program long-term monitoring programs
.
Washington, D.C.
:
ICPRB Report 09-4
.
Pearl,
H.W.,
2009
.
Controlling eutrophication along the freshwater–marine continuum: Dual nutrient (N and P) reductions are essential
.
Estuaries and Coasts
,
32
(
4
),
593
601
. doi:
Pearl,
H.W.;
Valdes,
L.M.;
Pinchney,
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
965
.
Rabalais,
N.N.
and
Nixon,
S.W.
(
eds.
),
2002
.
Nutrient over-enrichment in coastal waters: Global patterns of cause and effect
.
Estuaries
,
25
(
4B
),
639
900
.
Rabalais,
N.N.
and
Turner,
R.E.
(
eds.
),
2001
.
Coastal hypoxia, consequences for living resources and ecosystems
.
In
:
Coastal and Estuarine Studies, Volume 58
.
Washington, D.C
.:
American Geophysical Union, 463p
.
Radach,
G.,
1998
.
Quantification of long-term changes in the German Bight using an ecological development index
.
ICES Journal of Marine Sciences
,
55
(
4
),
587
599
.
Ren,
L.,
2002
.
Biogeochemical Conversion of Nitrogen in Enclosed Pelagic Coastal Ecosystems of the German Bight: Mesocosm and Modelling Studies
.
Hamburg, Germany
:
Hamburg University
,
Ph.D. dissertation
.
Ren,
L.,
2013
.
Baseline characteristics of phytoplankton and harmful algal blooms in Barnegat Bay–Little Egg Harbor estuary (year one)
.
Technical Report to New Jersey Sea Grant and New Jersey DEP
.
Philadelphia
:
The Academy of Natural Sciences of Drexel University.http://nj.gov/dep/dsr/barnegat/finalreport-year1/phytoplankton-year1.pdf
.
Ren,
L.
2015
.
Baseline characteristics of phytoplankton and harmful algal blooms in Barnegat Bay–Little Egg Harbor estuary (year two)
.
Philadelphia
:
The Academy of Natural Sciences of Drexel University
.
Technical Report to New Jersey Sea Grant and New Jersey DEP. http://nj.gov/dep/dsr/barnegat/final-reports/phytoplankton-year2.pdf
.
Ren,
L.
and
Belton,
T.,
2015
.
Barnegat Bay phytoplankton year 3: Phytoplankton reference communities and index of biotic integrity
.
Technical Report to New Jersey Sea Grant and New Jersey DEP
. .
Ren,
L.;
Rabalais,
N.N.;
Turner,
R.E.;
Morrison,
W.,
and
Mendenhall,
W.,
2009
.
Nutrient limitation on phytoplankton growth in Upper Barataria Basin, Louisiana: Microcosm bioassays
.
Estuaries and Coasts
,
32
,
958
974
.
Rothenberger,
M.B.;
Burkholder,
J.M.,
and
Wentworth,
T.R.,
2009
.
Use of long-term data and multivariate ordination techniques to identify environmental factors governing estuarine phytoplankton species dynamics
.
Limnology and Oceanography
,
54
(
6
),
2107
2127
.
Seitzinger,
S.P.;
Styles,
R.M.,
and
Pilling,
I.E.,
2001
.
Benthic microalgal and phytoplankton production in Barnegat Bay, New Jersey (USA): Microscosm experiments and data synthesis
.
In:
Kenish, M.J. (ed.),
Characterization of the Barnegat Bay – Little Egg Harbor, New Jersey, Estuarine System and Watershed Assessment
.
Journal of Coastal Research
,
Special Issue No.
32
, pp.
144
162
.
Smith,
V.H.
and
Schindler,
D.W.,
2009
.
Eutrophication science: Where do we go from here?
Trends in Ecology and Evolution
,
24
(
4
),
201
207
.
Smith,
V.H.;
Tilman,
G.D.,
and
Nekola,
J.C.,
1999
.
Eutrophication: impacts of excess nutrient inputs on freshwater, marine, and terrestrial ecosystems
.
Environmental Pollution
,
100
(
1–3
),
179
196
.
Stramski,
D.,
1999
.
Refractive index of planktonic cells as a measure of cellular carbon and chlorophyll a content
.
Deep-Sea Research I
,
46
(
2
),
335
351
.
Taghon,
G.L.;
Ramey,
P.A.;
Fuller,
C.M.;
Petrecca,
R.F.;
Grassle,
J.P.;
and
Belton,
T.J.,
2017
.
Benthic invertebrate community composition and sediment properties in Barnegat Bay–Little Egg Harbor from 1969 to 2014
.
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
.
169–183
.
Turner,
R.E.;
Rabalais,
N.N.;
Alexander,
R.B.;
Mcisaac,
G.,
and
Howarth,
R.W.,
2007
.
Characterization of nutrient, organic carbon, and sediment loads and concentrations from the Mississippi River into the Northern Gulf of Mexico
.
Estuaries and Coasts
,
30
(
5
),
773
790
.
USEPA (U.S. Environmental Protection Agency)
,
2001
.
Nutrient Criteria Technical Guidance Manual: Estuarine and Coastal Marine Waters
.
Washington, D.C
.:
U.S. Environmental Protection Agency, EPA-822-B-01-003
.
Weisberg,
S.B.;
Ranasinghe,
J.A.
Dauer,
D.M.;
Schaffner,
L.C.;
Diaz,
R.J.,
and
Frithsen,
J.B.,
1997
.
An estuarine benthic index of biotic integrity (B-IBI) for Chesapeake Bay
.
Estuaries
,
20
(
1
),
149
158
.