STABLE ISOTOPIC COMPOSITION OF SUSPENDED ORGANIC MATTER IN THE MULLICA RIVER/GREAT BAY ESTUARY (SOUTHERN NEW JERSEY, U.S.A.).
KEY WORDS: [[delta].sup.13]C [[delta].sup.15]N, Mullica River, biogeochemistry
The Mullica River/Great Bay estuary was recently designated as the Jacques Y. Cousteau National Estuarine Research Reserve. It is a relatively pristine estuary, with much of the surrounding watershed residing in state, federal or local preserves (Psuty et al., 1993; DeLuca et al., 1997).
The site has been the focus of investigations of primary and secondary production, principally of fishes (e.g., Sogard and Able, 1991; Rountree and Able, 1992; Able et al., 1996; Able and Fahay, 1998). Recent research has focused on the trophic linkages between juvenile fishes, their immediate prey, and ultimately, the primary producers at the base of the food web (Litvin, 1996; Wainright, unpublished data). The latter studies have used a stable isotopic approach to follow the pathways of C and N through the food web. Primary producers thought to contribute to fish production include extensive stands of macrophytes via their detritus, in situ production by benthic and planktonic algae, and offshore phytoplankton production that is imported to the estuary tidally. However, there is no published information on the isotopic composition of these primary producers in this estuarine system. Other investigators have demonstrated large isotopic variation (including seasonal variation) both among and within estuarine systems (e.g., Mariotti et al., 1984; Peterson et al., 1985; Cifuentes et al., 1988; Wainright and Fry, 1994; Deegan and Garritt, 1997; McLelland et al., 1997; Stribling and Cornwell, 1997; Bird et al., 1998).
Chmura and Aharon (1995) and Deegan and Garritt (1997) found that the isotopic composition of suspended POM and emergent vegetation varies spatially within an estuary, such that upstream sites typically have more negative [[delta].sup.13]C values, coincident with a change in emergent vegetation from [C.sub.4] to [C.sub.3] as one moves up-estuary. A recent study in African rivers demonstrated that river discharge can influence the [[delta].sup.13]C values of seston by transporting different types of organic matter into the river from different parts of the floodplain (Bird et al., 1998).
The nitrogen isotopic composition of the primary producers may vary according to the concentration of DIN and uptake rate (Cifuentes et al., 1988; Montoya et al., 1990), and the source of DIN to the estuary (Litvin, 1996; McLelland et al., 1997).
Our goals were three-fold: (1) to investigate seasonal variation during an 18-month period in both [[delta].sup.13]C and [[delta].sup.15]N values of suspended particulate organic matter (POM) in the Mullica River and Great Bay Estuary (These data would help to constrain the endmember isotope values of Suspended (POM) for biogeochemical and food web studies.); (2) to investigate the reasons for any seasonal variation, with particular attention to the differences in C pools at the oligohaline and marine ends of the estuary; (3) to provide potential users of the Jacques Y. Cousteau National Estuarine Research Reserve with information on the seasonal variation of a number of water chemistry parameters.
MATERIALS AND METHODS
The Great Bay-Mullica River estuary is a highly productive estuarine system for shellfish and finfish, and is perhaps one of the cleanest on the east coast of the U.S. Previous studies of nutrients, productivity and biota are summarized in a series of technical reports (Durand and Nadeau, 1972; Durand, 1979; Durand 1990). Unlike most estuaries in the northeastern U.S., the surrounding area, including most of the Pine Barrens watershed (Fig. 1), is protected from large-scale human disturbance. The Kirkwood-Cohansey aquifer, which is the subject of a major program of study through the U.S. Geological Survey, feeds the estuarine system. Almost the entire Mullica River drainage basin is part of the New Jersey Pinelands National Reserve; in 1983 it was named a UNESCO Biosphere Reserve, and in 1994 a National Estuarine Research Reserve. The downstream portions of the river are further buffered from anthropogenic effects by federal (Edwin B. Forsythe National Wildlife Refuge) and state (Great Bay Wildlife Management Area) wildlife refuges.
The study was conducted at two sites within the Great Bay/Mullica River estuarine system (A and B in Fig. 1). The downstream site (here-after referred to as the Marine site) was located at the Rutgers University Marine Field Station (RUMFS), in Tuckerton, New Jersey, at the meteorological tower adjacent to the inlet to Great Bay from the Atlantic Ocean. Benthic macroalgae (principally Ulva spp.) and the seagrass, Zostera marina, are common in Great Bay, especially in Little Egg Harbor. The upstream site (hereafter referred to as the Oligohaline site) was located at NJ Highway 563, Green Bank, in the oligohaline region of the river. Vegetation at that site was a mixture of Typha, Phragmites, and freshwater emergent species. Water color was deep brown. Strong mixing is promoted by tidal currents and shallow water depths at both sites and by river discharge at the Oligohaline site.
Samples were collected for the following analyses: water temperature and salinity, dissolved oxygen (DO), dissolved inorganic carbon concentration ([DIC]), pH, dissolved organic carbon concentration ([DOC]), total suspended solids, chlorophyll concentration, size fractionated suspended particulate organic matter (POM) for determination of particulate organic carbon (POC) and particulate nitrogen (PN). Stable isotopic compositions of POM ([[delta].sup.13]C and [[delta].sup.15]N) and DIC ([DI.sup.13]C) were determined.
Surface water was collected by gently immersing a bucket. Water temperature was measured with an NBS-calibrated Hg thermometer. The pH was determined in the field with a Fisher Accumet pH meter. Salinity was determined with a refractometer which was calibrated using distilled water, and verified occasionally with an Autosal salinometer (Guildline Instruments, LTD, Orlando, Florida, model 8400B) or a flame atomic absorption spectrophotometer (Perkin Elmer, model 4000). Three replicate samples were taken for determination of DO, and two replicate 500-ml samples were preserved with 2-ml saturated [Hgcl.sub.2] solution for determination of [DIC] and [DI.sup.13]C. Two replicate 20-ml samples were also preserved with saturated [HgCl.sub.2] solution for determination of DOC concentration. Water samples were filtered at [less than or equal to]5 psi onto precombusted (500[degrees]C) glass fiber filters (Whatman GF/F) for particulate organic matter (POC, PN) and chlorophyll analyses. Chlorophyll a and phaeopigment con centrations were determined fluorometrically (Parsons et al., 1984).
Suspended POM was size-fractionated by passing up to 90 liters of bulk estuarine water through 114, 60, and 20 [micro]m Nitex screens sequentially by gravity filtration. The material retained on the screens was rinsed gently from each mesh onto precombusted (500[degrees]C) glass fiber filters (Whatman GF/F) with a stream of ambient water. Particulate material in these size fractions and the [less than]20 [micro]m fraction were analyzed separately for POC and PN concentrations, and stable isotopic compositions. Chlorophyll concentrations were determined on bulk estuarine water and the [less than]20 [micro]m fraction.
Dissolved oxygen (DO) concentrations were determined with an automated Winkler titrator (Friederich et al., 1991). Analytical error was less than 0.3% (CV).
DIC concentrations were determined coulometrically using a modification of ASTM (1983) and Johnson et al. (1985), on a model 5012 [CO.sub.2] coulometer (UIC, Inc. Joliet, IL). Briefly, a 10-ml water sample was injected into a reaction vessel through a septum, acidified with 2 ml of 8.5% phosphoric acid, and sparged with [CO.sub.2]-free [N.sub.2] gas (Ultra High purity, Matheson) at a flow rate of 75 ml [min.sup.-1] for 15 min. The carrier gas was passed through Ascarite and magnesium perchlorate traps prior to sparging. Concentrations of DIC were calibrated against solutions of [Na.sub.2][CO.sub.3] (J.T. Baker, 99.5% purity) following the methods of Goyet and Hacker (1992). Analytical precision (CV) was approx. 0.5%. Separate preserved samples were sent to the Stable Isotope Laboratory at Boston University for determination of the stable carbon isotopic composition of DIC.
DOC was determined by first sparging the DIC from an acidified 10-ml subsample (50 [micro]l of 50% (w/w) [H.sub.3][PO.sub.4] was added to Marine samples; 10 [micro]l to Oligohaline samples). Fifty-[micro]l of the acidified sample was then combusted in a vertical tube furnace containing quartz chips at 800[degrees]C. The carrier gas (ultra high purity oxygen, Matheson) was scrubbed through a water trap (molecular sieve, dehydrite) and a hydrocarbon trap (molecular sieve). The combustion gases were scrubbed through a chloride trap (1 ml of stock solution [(5 g [Ag([NO.sub.3]).sub.2] + 95 g [H.sub.3][PO.sub.4]) diluted 1:10 with deionized water (DI)] a mercury trap (5 mls of the following: 10g [SnCl.sub.2] 2 ml conc. HCl, 8 ml DI water) and a second chloride trap before passing through a U-tube condenser (in ice water) and a magnesium perchlorate trap to remove water vapor. The purified [CO.sub.2] was quantified with an infrared [CO.sub.2] analyzer (Model 6252, Li-Cor, Lincoln, NE). This procedure was modified from Peltzer and Brewer (1993). The standard was potassium hydrogen phthalate (EM Science, 99.95% purity) in DI water or seawater, as appropriate.
[CO.sub.2(aq)] and [H.sub.2][CO.sub.3] concentrations were calculated from DIC concentrations and pH using constants taken from UNESCO (1987). pH is expressed on the seawater scale.
Concentrations of POC and PN were determined from POM filters treated with 1-1.5 ml of 1% Pt[Cl.sub.2] in 1 N HCl in glass petri dishes, dried without rinsing ([less than]45[degrees]C overnight), and powdered with approximately 1 g of precombusted (800[degrees]C, 3 h) [Cu.sup.o] with a Wig-L-Bug amalgamating machine (Crescent Dental Manufacturing Co., Lyons, IL). The powder and 0.5 g [Cu.sup.o] were sealed in precombusted (900[degrees]C, 1 h) 9 mm quartz (Vycor) tubes under vacuum, and combusted at 750[degrees]C for 3 hours. [CO.sub.2] and [N.sub.2] gases liberated by combustion were purified cryogenically on a vacuum line. [CO.sub.2] content was determined manometrically, and [N.sub.2] content was determined from the mass 28 signal voltage on a mass spectrometer, using a regression based on standards.
Stable isotopic compositions of C and N were determined on a Finnigan Delta-S stable isotope mass spectrometer, and are reported as [[delta].sup.13]C and [[delta].sup.15]N values, in parts per thousand (%[degrees]) deviations relative to the PDB standard or nitrogen in air, respectively. Replicates carried through this procedure usually differed by less than 0.2%[degrees]. Filter blanks were routinely determined and sample concentrations and isotopic ratios corrected accordingly. These blank values were 1.61 [+ or -] 0.72 (s.d.) [micro]moles C, and -21.53 [+ or -] 2.38%[degrees] [[delta].sup.13]C (n=8). Nitrogen blanks were undetectable.
One of our goals was to determine the underlying causes for the stable isotopic variation in the [less than]20 [micro]m and 20-114 [micro]m size fractions by relating the isotopic data (dependent variables) with the other data collected during this study. The independent variables were reduced to a number less than the number of observations by eliminating variables judged a priori to contain redundant information. The data were transformed where necessary to achieve a normal distribution and Principle Components Analysis (PCA) was applied to the correlation matrix to reveal the correlation structure. A separate analysis was performed for the Marine and the Oligohaline sites, and in both cases, PCA explained over 90% of the variation with the first 4 or 5 principle components (PCs). The PC scores from the first 4 PCs were then used as independent variables in a multiple regression analysis to determine which of them best explained the variation in the [[delta].sup.13]C and [[delta].sup.15]N values of the part iculate material in the [less than] 20[micro]m and 20-114 [micro]m size fractions. We applied a multiple regression procedure in this exploratory analysis, to examine the ability of the PCs to explain the isotopic data. Analyses were performed using JMP software for Macintosh (SAS Institute, Inc. Cary, NC).
RESULTS AND DISCUSSION
Water temperatures were similar at both sites and varied seasonally over a 26[degrees]C range (Fig. 2A). Salinity ranged from 25 to 34%[degrees] at the Marine site and from 0 to 3.5%[degrees] at the Oligohaline site (Fig. 2B). In the winter, when water temperatures were less than 10[degrees]C, DO concentrations were 300 to 450 [micro]M for both the Oligohaline and the Marine sites (Fig. 3). In the summer, DO concentrations reached a minimum of 180 to 200 [micro]M at the Oligohaline site. Concentrations of DO were near saturation at the Marine site (Fig. 3B), while at the Oligohaline site they were generally undersaturated (by 21%). Wave action may have enhanced oxygen saturation at the Marine site. The undersaturated DO concentrations at the Oligohaline site indicate that respiration outweighs production (as was found by Durand and Nadeau, 1972), although it is possible that groundwater of low DO contributed to this undersaturation.
The Mullica River discharge rate (gauged at station 01409400 at Lat. 39[degrees]40'28", Longt. 74[degrees]39'55", approx. 3 mi upstream of the Oligohaline site) for the specific days sampled reached peak flows of 9.8 and 7.3 [m.sup.3] [sec.sup.-1] in March, 1993 and January, 1994, respectively (Fig. 2C). Discharge rates were less than 1.4 [m.sup.3] [sec.sup.-1] between June and October.
At the Oligohaline site, phytoplankton blooms (up to approx. 23 mg/[m.sup.3] chlorophyll a) occurred in June through October, 1993 and in June, 1994 (Fig. 4A). The first part of the 1993 bloom was associated with the 20-114 [micro]m size fraction, while the second half was dominated by smaller material. At the Marine site, a summer bloom of lesser magnitude (approximately 5 to 10 mg/[m.sup.3]) occurred during April to August; a more intense bloom (approximately 14 mg/[m.sup.3]) occurred in February, 1994. Similar winter blooms were reported by Durand and Nadeau (1972). In general, at both sites, over 65% of the chlorophyll was associated with the [less than]20 [micro]m size fraction (Fig. 4B). However, during the two major bloom periods of June, 1993 (Oligohaline site) and January to March, 1994 (Marine site) the 20--114 [micro]m chlorophyll fraction dominated over the smaller fraction.
The chlorophyll/phaeophytin ratio (Chlor/Phaeo; Fig. 4C) is interpreted as a measure of the "freshness" of the organic matter, and varied with chlorophyll concentration. The ratio of POC/Chlorophyll provided information similar to the Chlor/Phaeo ratio, but was dropped from the analysis to avoid redundancy. At the Oligohaline site, the presence of fresh phytoplankton is indicated by a Chlor/Phaeo ratio greater than 1 in the bulk particulate material in June, July, September and October in 1993. The Chlor/Phaeo ratio less than 1 for the [less than]20 [micro]m size fraction (Fig. 4D) corroborates that the June, 1993 bloom occurred in the [greater than]20 [micro]m size class, while the October bloom was associated with the [less than]20 [micro]m size class. At the Marine site, the Chlor/Phaeo ratio was greater than 1 from May to August, 1993 and from February to April, 1994. The Chlor/Phaeo ratio of the [less than]20 [micro]m size fraction (Fig. 4D) followed a similar pattern as that for the bulk phytoplankton except during the February, 1994 bloom, when phytoplankton [greater than]20 [micro]m dominated the bloom.
The particulate organic carbon (POC) concentrations were up to five times higher and were more variable at the Oligohaline site than at the Marine site (Fig. 5A). The POC concentrations at the Oligohaline site peaked in May through July, 1993, Dec. '93 to Jan. '94, and and in May and June, 1994. Two peaks coincided with chlorophyll maxima (Fig. 4A), as did the POC peak in February 1994 at the Marine site. At both sites, the [less than]20 [micro]m size fraction contained most of the POC.
Particulate nitrogen (PN) concentrations followed patterns that were similar, but not identical, to the POC concentrations. PN concentrations ranged from 5 to 140 [micro]M at the Oligohaline site, and from 5 to 39 [micro]M at the Marine site (Figs 5C and D). In a previous study conducted near the Marine site, Denmark (1975) showed that the 0.8- 64 [micro]m size fraction of POM contained over 90% of the total PN.
Carbon/nitrogen ratios (C/N) of the POM at both sites were between 3 and 10 (indicative of relatively undegraded material), except during June, 1993 at the Oligohaline site where the C/N ratio reached a high of 22 (Fig. 6A). At that time, the [less than]20 [micro]m size fraction had a C/N ratio of 25 (Fig. 6B), and the chlorophyll concentration in that size fraction was low. This suggests that the bulk of the POC at that time was fine detritus of relatively high C/N ratio. Estuarine macrophytes, including Spartina spp, and Phragmites australis, from comparable locations along Delaware Bay commonly have C:N ratios of 20-50 (Wainright, unpublished data). During the April through August bloom at the Marine site, the C/N ratio of the [less than]20 [micro]m size fraction reached a high of 11.5 in June, 1993, suggesting that macrophyte detritus was not its major source.
At the Marine site, the pH was relatively constant at around 8 (Fig. 6C), while at the Oligohaline site the pH ranged from 4 to 6.5. The lowest pH values occurred during the winter months, followed by a gradual increase to a maximum in October, followed by a steep decline. This seasonal shift may reflect a seasonal upstream-downstream movement of the salt water interface with consequent effects on water chemistry, or a change in the amount or chemistry of the freshwater flowing into the system. pH was negatively related to the concentration of carbonic acid (as would be expected), and was related to river discharge, i.e., low pH during high river discharge. The pH at the Oligohaline site was weakly (p = 0.083) related to [DOC]. The DOC concentration was much greater and more variable at the Oligohaline site (500 to 1100 [micro]M) than at the Marine site (200 to 600 [micro]M) (Fig. 6D). The [DOC] at the Oligohaline site was similar to that reported by Fox (1983), while [DOC] at the Marine site was somewhat hi gher than expected, perhaps because salinities were not fully marine or because of inputs of high-DOC water from adjacent marshes (J.E. Hughes and J. Lurman, unpublished data).
Concentrations of DIC ranged from 20 to 233 [micro]M [kg.sup.-1] at the Oligohaline site (Fig. 7A). At the Marine site, [DIC] was 1624-1960 [micro]M [kg.sup.-1], an order-of-magnitude higher than the Oligohaline site. Both sites followed similar temporal trends with higher concentrations beginning in June at the Marine site and in July at the Oligohaline site and remained elevated through October. In 1994, the [DIC] at the Marine site began to increase in May.
Calculated [pCO.sub.2] (Fig. 7C) was much lower at the Marine site, and was correlated with the [DIC] at both sites (p=0.0l2 and 0.005 at the Oligohaline and Marine sites, respectively). A number of studies have found that the [[delta].sup.13]C of phytoplankton depends on [pCO.sub.2] (e.g., Rau et al., 1989; Laws et al., 1995; Fry, 1996; Rau et al., 1996). [H.sub.2][CO.sub.3] constituted the majority of the DIC at the Oligohaline site, but only a small fraction of the total DIC at the Marine site.
Values for [DI.sup.13]C fluctuated between approx. -2 and -18 at the Oligohaline site (Fig. 7B), but were comparatively constant at the Marine site, averaging about 0%[degrees]. The negative [DI.sup.13]C values at the Oligohaline site are not unlike those reported for other riverine systems, and may reflect inputs of respiratory [CO.sub.2] (Keough et al., 1998). Although the [[delta].sup.13]C values of phytoplankton, and hence SPOM, were expected to reflect the [[delta].sup.13]C values of the source [CO.sub.2], correlations between these parameters were not significant at either site.
The [[delta].sup.13]C values of SPOM (Figs. 8A and B) were more negative at the Oligohaline site (-25 to -28%[degrees]) than at the Marine site (-18 to -23%[degrees]). For comparison, the [[delta].sup.13]C values of the seagrass Zostera marina, the macroalga Ulva lactuca, and benthic sediments collected near the Marine site were approx. -11%[degrees], -16 to -l9%[degrees], and -l7%[degrees], respectively (Litvin, 1996). There were no obvious differences between [[delta].sup.13]C values of the other size classes examined, therefore, only the [less than]20 and 20-114 [micro]m fractions are shown. The rather negative [[delta].sup.13]C values at the Oligohaline site are consistent with either an input of detrital material from the surrounding watershed or in situ production of phytoplankton using isotopically negative DIC. At the Marine site, the [[delta].sup.13]C values are consistent with a phytoplankton origin (Wainright and Fry, 1994).
The [[delta].sup.15]N values of POM ranged between 1 and 6.5%[degrees] at the Oligohaline site and 5 to 8.5%[degrees] at the Marine site (Figs. 8C and D). The [[delta].sup.15]N values of Z. marina, U. lactuca, and benthic sediments collected near the Marine site were approx. 4%[degrees], 8-11%[degrees], and 4.5%[degrees], respectively (Litvin, 1996). The [less than]20 [micro]m size fraction was more variable than the 20-114 [micro]m fraction at the Marine site, while at the Oligohaline site both size fractions were variable. Inputs of fertilizer-derived [[NO.sub.3].sup.-] (Durand, 1984), or changes in isotopic fractionation of dissolved inorganic nitrogen during uptake (see below), may account for the lower [[delta].sup.15]N values at the Oligohaline site.
Sources of Stable Isotopic Variation
The Principal Components (PC) analysis was performed on a subset of 11 of the original variables (Table I). [DI.sup.13]C did not play a significant role in determining the [[delta].sup.13]C values of any of the POM size fractions (based on pair-wise correlations); therefore, it was dropped from the PC analysis. We used [[H.sub.2][CO.sub.3]], rather than [pCO.sub.2], as an indicator of the free [CO.sub.2] available to phytoplankton because more observations of [[H.sub.2][CO.sub.3]] were available. Five principle components (PCs) accounted for 90.5% of the total variation of the original 11 variables at the Marine site. Table I presents the proportion of the original variance explained by each of the first 4 PCs, and the loadings (eigenvectors) of each of the original variables on those PCs. For example, pH was correlated strongly with PC 1. An eigenvector greater than 0.4 was arbitrarily chosen to represent a strong correlation; other variables including river discharge, and --[[H.sub.2][CO.sub.3]] were more weakly correlated with PC 1. Thus, PC 1 is an abstract variable that captures variation of the carbonate system and river discharge. Similarly, PC 2 was associated with fresh phytoplankton (chlorophyll concentration and Chlor/Phase ratio) and seasonal change (temperature). PC 3 was a more complex variable, influenced by [DOC], the [less than]20 [micro]m fraction of the POC, and carbonic acid. PC 4 was related to the C:N ratio and salinity. PC 5 accounted for less than 5% of the variation of the data at both sites, and is not considered further.
The multiple regression analysis (Table II) revealed that the [[delta].sup.13]C of the 20-114 [micro]m fraction of POC ([PO.sup.13][C.sub.(20-114[micro]m)]) was explained by PC 2 (p=0.0004), and to a much lesser degree, PC 1 (p= 0.1121). Therefore, the [PO.sup.13][C.sub.(20-114[micro]m)] is influenced most strongly by a combination of variables associated with those PCs, i.e., chlorophyll a concentration, Chlor/Phaeo ratio, water temperature (PC 2), and pH (PC 1).
The [PO.sup.13][C.sub.([less than]20[micro]m)] was not well-explained by any PC (Table II). This may be a consequence of the potentially heterogeneous sources of materials, the alternation between dominant size fractions associated with chlorophyll during blooms (Fig. 4), and the dynamic spatial and temporal nature of estuaries. For example, pools measured today may be the product of processes in the past occurring upstream or down-stream from the point of collection.
The [P.sup.15][N.sub.(20-114[micro]m)] was well-explained ([r.sup.2] = 0.74) by PC 4(p=0.0028), and PC 1 (p=0.0039). Addition of other PCs to the analysis did not improve the adjusted [r.sup.2] (Table II). Variables associated with these PCs were the C/N ratio and salinity (PC 4), and pH (PC 1). We also found a weak negative relationship with the concentration of nitrate (Fig. 9), based on a partial data set (Sept. 1993 to June 1994; see Oligohaline Site, below, for further discussion). The positive relationship between [[delta].sup.15]N with C/N suggests that as organic matter becomes recalcitrant its [[delta].sup.15]N value increases.
The [P.sup.15][N.sub.([less than]20[micro]m)] was well-explained ([r.sup.2] = 0.71) by PC 4 (p=0.0103), PC 3 (p=0.0l38) and PC 2 (p=0.0295). Variables associated with PC 4 and PC 2 were mentioned above. Variables associated with PC 3 were DOC, the [less than]20 [micro]m size fraction of POC, and [[H.sub.2][CO.sub.3]]. Thus, the [P.sup.15][N.sub.([less than]20[micro]m)] is influenced by a more heterogeneous group of variables than is the larger PN fraction.
In summary, the [[delta].sup.13]C value of suspended organic matter at the Marine site varied seasonally, and was associated with fresh phytoplankton biomass. The [[delta].sup.15]N values were correlated with the C/N ratio (detrital quality), salinity, and pH. The relationship with pH may be indirect, and related to the coincidence of low pH and high [[[NO.sub.3].sup.-]] upstream in the estuary (see below).
The PCA was performed on a subset of 11 of the original variables. [DI.sup.13]C was excluded from the analysis, and [[H.sub.2][CO.sub.3]] was used instead of [pCO.sub.2] for the same reasons as at the Marine site. Four PCs accounted for 89% of the total variation of the independent variables (Table I). Variables that were strongly correlated (eigenvectors [greater than or equal to]0.4) with PC 1 included chlorophyll, river discharge, and water temperature. PC 1 is, therefore, a composite factor associated with these variables. The association with water temperature and river discharge indicates a component of seasonality, and indicates that associated variables also exhibit seasonal variation. Similarly, PC 2 was associated with the 20-114 [micro]m size fraction of the POC, [DOC], and the C:N ratio of POM (negative association with the quality of the organic matter). PC 3 was influenced, inversely, by salinity, and [[H.sub.2][CO.sub.3]]. PC 4 was related to [DOC], the [less than]20 [micro]m size fraction of t he POC, and, inversely, to the C/N ratio of the POM.
The Multiple Regression analysis revealed that the [PO.sup.13][C.sub.(20-114 [micro]m)] and the [P.sup.15][N.sub.([less than]20 [micro]m)] were not well-explained by any of the PCs (Table III). Potential explanations were given above.
The [PO.sup.13][C.sub.([less than]20 [micro]m)] was explained ([r.sup.2] = 0.66) by PC 1 (p=0.003), and, to a lesser degree, PC 2 (p=0.04), indicating seasonal variation, and a relationship with the amount and quality of organic matter. The latter is in agreement with previous findings that refractory POC is isotopically heavier than fresh POC (Benner et al., 1987).
The [p.sup.15][N.sub.(20-114 [micro]m)] was weakly explained ([r.sup.2]=0.34) by PC 1 (p=0.0161), which, in turn, was associated with seasonally variable physical and biotic factors (Table I). We also found a highly significant negative correlation with the concentration of nitrate (Fig. 9), based on a pair-wise correlation with a partial nitrate data set (Sept. 1993 to June 1994). Lowest [[delta].sup.15]N values were found when [[NO.sub.3].sup.-] was highest (up to 39 [micro]M), i.e., during the winter. The latter is consistent with reduced isotopic fractionation during uptake when nitrate concentrations are high (Mariotti et al., 1981; Waser et al., 1998). Nitrate from fertilizers may wash into the Mullica River from surrounding agricultural soils during the non-growing season (Durand, 1984).
In summary, [[delta].sup.13]C values of suspended matter at the Oligohaline site were seasonally variable, and related to the amount and quality of POC and DOC. The [[delta].sup.15]N values of suspended matter were seasonally variable, and potentially linked with inputs of nitrate to the system during winter months.
The isotopic composition of suspended POM varies seasonally and spatially within the estuary, however, the seasonal changes are different in different parts of the estuary. The [[delta].sup.13]C and [[delta].sup.15]N values of POM suggest a marine phytoplankton origin at the Marine site. The source of POM is not as clear at the Oligohaline site, but may include terrestrial vegetation, riverine phytoplankton or a mixture. The relative sizes and isotopic compositions of the various C pools at the Marine and Oligohaline sites showed seasonal patterns. We did not find strong relationships between the [[delta].sup.13]C values of POC ([PO.sup.13]C) and parameters related to the carbonate system or [DI.sup.13]C. Our multivariate analysis revealed that [PO.sup.13]C was associated with amount and quality of POC, and in particular, with freshly produced phytoplankton biomass. The [P.sup.15]N was associated with the amount and quality of POM (C/N ratio), salinity and pH. The latter appears to be related to concomitant variations in nitrate concentration.
We are grateful to J. Hughes for his efforts in developing the methodology for DOC analysis in our laboratory. D. Scala assisted with field collections and DO analyses, W. Wang, J. Ross and P. Field assisted with nutrient analyses and salinometer data. Data on the Mullica River discharge rate was provided by Robert Reiser of the U.S. Geological Survey, Water Resources Division, West Trenton, New Jersey. We thank 3 anonymous reviewers for improving the manuscript. This is Contribution # 1999-02 of the Institute of Marine and Coastal Sciences, Rutgers University, and Contribution # 1999-01 of the Jacques Y. Cousteau National Estuarine Research Reserve.
ABLE, K.W., M.P. FAHAY. 1998. The first year in the life of estuarine fishes in the Middle Atlantic Bight. Rutgers University Press, New Brunswick, New Jersey.
_____, D.A. Witting, R.S. McBride, R.A. Rountree, K.J. Smith, 1996. Fishes of polyhaline estuarine shores in Great Bay-Little Egg Harbor, New Jersey: a case study of seasonal and habitat influences. Pp. 335-353 in Nordstrom, K.F., C.T. Roman (eds.). Estuarine shores: evolution, environments and human alterations. John Wiley and Sons, Chichester, England.
ASTM. 1983. Standard test methods for: carbon dioxide and bicarbonate and carbonate ions in water. American Society for Testing and Materials D 513-82:1-21.
BENNER, R., M.L. FOGEL, E.K. SPRAGUE, R.E. HODSON. 1987. Depletion of [C.sup.13] in lignin and its implications for stable carbon isotope studies. Nature 329:708-710.
BIRD, M.I., P. GIRESSE, S. NGOS. 1998. A seasonal cycle in the carbon-isotope composition of organic carbon in the Sanaga River, Cameroon. Limnol. Oceanogr. 43:143- 146.
CHMURA, G.L., P. AHARON. 1995. Stable carbon isotope signatures of sedimentary carbon in coastal wetlands as indicators of salinity regimes. J. Coastal Res. 11:124- 135.
CIFUENTES, L.A., J.H. SHARP, M.L. FOGEL. 1988. Stable carbon and nitrogen isotope biogeochemistry in the Delaware estuary. Limnol. Oceanogr. 33:1102-1115.
DEEGAN, L.A., R.H. GARRITT. 1997. Evidence for spatial variability in estuarine food webs. Mar. Ecol. Prog. Ser. 147:31-47.
DELUCA, M., J. MCDONNELL, R. LATHROP JR., K. ABLE, S. WHITNEY, N. PSUTY, E. HAYNES, C. CAFAGNA. 1997. Mullica River-Great Bay (MRGB) National Estuarine Research Reserve in New Jersey: Final Management Plan. Institute of Marine and Coastal Sciences, Rutgers University. IMCS Technical Report 97-25
DENMARK, R.E. 1975. Particulate organic detritus in a New Jersey salt marsh-estuarine ecosystem. M.S. thesis, 111 pp. Rutgers University, New Brunswick, NJ.
DURAND, J.B. 1979. Nutrient and hydrologic effects of the Pine Barrens on neighboring estuaries. Pp. 195-211 in Forman, R.T.F. (eds.). Pine Barrens: Ecosystems and Landscape. Academic Press, New York.
_____ 1984. Nitrogen distribution in New Jersey coastal bays. Pp. 29-51 in Kennish, M.J., R.A. Lutz (eds.). Ecology of Barnegat Bay, New Jersey. Springer-Verlag, New York.
_____ 1990. Field studies in the Mullica River-Great Bay estuarine system. Center for Coastal and Environmental Studies, Rutgers University, New Brunswick, NJ. Tech. Rep.
_____, R.J. NADEAU. 1972. Water Resources development in the Mullica River basin. part 1. Biological evaluation of the Mullica River-Great Bay estuary. New Jersey Water Resources Research Institute, Rutgers University, New Brunswick, NJ.
FOX, L.E. 1983. The removal of dissolved humic acid during estuarine mixing. Estuar. Coastal Shelf Sci. 16:431-440.
FRIEDERICH, G.E., L.A. CODISPOTI, C.M. SAKAMOTO. 1991. An easy-to-construct automated Winkler titration system. Monterey Bay Aquarium Research Institute Technical Report 91-6
FRY, B. 1996. [C.sup.13]/[C.sup.12] fractionation by marine diatoms. Mar. Ecol. Prog. Ser. 134:283- 294.
GOYET, C., S.D. HACKER. 1992. Procedure for calibration of a coulometric system used for total inorganic carbon measurements in seawater. Mar. Chem. 38:37-51.
JOHNSON, K.M., A.E. KING, J.M. SJEBURTH. 1985. Coulometric [TCO.sup.2] analyses for marine studies; and introduction. Mar. Chem. 16:61-82.
KEOUGH, J.R., C.A. HAGLEY, E. RUZYCKI, M. SIERSZEN. 1998. [[delta].sup.13]C composition of primary producers and role of detritus in a freshwater coastal ecosystem. Limnol. Oceanogr, 43:734-740.
LAWS, E.A., B.N. BOPP, R.R. BIDIGARE, M.C. KENNICUTT, S.A. MACKO. 1995. Dependence of phytoplankton carbon isotopic composition on growth rate and [[CO2].sub.aq]: theoretical considerations and experimental results. Geochim. Cosmochim. Acta 59:1131-1138.
LITVIN, S.Y. 1996. Trophic linkages between juvenile fishes and organic sources in three northeastern estuaries as indicated by stable isotope ratios. M.S. thesis. Rutgers University, New Brunswick, NJ.
MARIOTTI, A., J.C. GERMON, P. HUBERT, P. KAISER, R. LETOLLE, A. TARDIEUX, P. TARDIEUX. 1981. Experimental determination of nitrogen kinetic isotope fractionation: some principles; illustration for the denitrification and nitrification processes. Plant and Soil 62:413-430.
_____, C. LANCELOT, G. BILLEN. 1984. Natural isotopic composition of nitrogen as a tracer of origin for suspended organic matter in the Scheldt estuary. Geochim. Cosmochim. Acta 48:549-555.
MCLELLAND, J.W., I. VALIELA, R.H. MICHENER. 1997. Nitrogen-stable isotope signatures in estuarine food webs: a record of increasing urbanization in coastal watersheds. Limnol. Oceanogr. 42:930-937.
MONTOYA, J.P., S.G. HORRIGAN, J.J. McCARTHY. 1990. Natural abundance of [N.sup.15] in particulate nitrogen and zooplankton in the Chesapeake Bay. Mar. Ecol. Prog. Ser. 65:35-61.
PARSONS, T.R., Y. MAITA, C.M. LALLI. 1984. A manual of chemical and biological methods for seawater analysis. Pergamon Press, New York. 173
PELTZER, E.T., P.G. BREWER. 1993. Some practical aspects of measuring DOC-sampling artifacts and analytical problems with marine samples. Mar. Chem. 41:243-252.
PETERSON, B.J., R.W. HOWARTH, R.W. GARRITT. 1985. Multiple stable isotopes used to trace the flow of organic matter in estuarine food webs. Science 227:1361-1363.
PSUTY, N.P., M.P. DE LUCA, R. LATHROP, K.W. ABLE, S. WHITNEY, J.F. GRASSLE 1993. The Mullica River-Great Bay National Estuarine Research Reserve: A unique opportunity for research, preservation and management. Coastal Zone '93, New Orleans.
RAU, G.H., U. RIEBESELL, D. WOLF-GLADROW. 1996. A model of photosynthetic [C.sup.13] fractionation by marine phytoplankton based on diffusive molecular [CO.sub.2] uptake. Mar. Ecol. Prog. Ser. 133:275-285,
_____. T. TAKAHASHI, D.J. DES MARAIS. 1989. Latitudinal variations in plankton [[delta].sup.13]C implications for [CO.sub.2] and productivity in past oceans. Nature 341:516-518.
ROUNTREE, R.A., K.W. ABLE, 1992. Fauna of polyhaline subtidal marsh creeks in southern New Jersey: composition, abundance and biomass. Estuaries 15:171-185.
SOGARD, S.M., K.W. ABLE. 1991. A comparison of eelgrass, sea lettuce, macroalgae, and marsh creeks as habitats for epibenthic fishes and decapods. Estuar. Coastal Shelf Sci. 33:501-519.
STRIBLING, J.M. J.C. CORNWELL, 1997. Identification of important primary producers in a Chesapeake Bay tidal creek system using stable isotopes of carbon and sulfur. Estuaries 20:77-85.
UNESCO. 1987. Thermodynamics of the carbon dioxide system in seawater. UNESCO Tech. Pap. Mar. Sci. 51.
WAINRIGHT S.C., B. FRY. 1994. Seasonal variation of the stable isotope compositions of plankton from Woods Hole, Massachusetts, USA, and Georges Bank. Estuaries 17:552-560.
WASER, NA., K. YIN, Z. Yu, K. TADA, P.J. HARRISON D.H. TURPIN, S.E. CALVERT. 1998. Nitrogen isotope fractionation during nitrate, ammonium and urea uptake by marine diatoms and coccolithophores under various conditions of N availability. Mar. Ecol. Prog. Ser. 169:29-41.
Principle Component (PC) analysis of independent variables used to explain isotopic variation in the stable isotopic composition of suspended matter. Abbreviations: Chlor = chlorophyll a; Chlor/Phae = Chlorophyll/ Phaeopigments; ln(POC[20-114]) = natural log of the POC concentration in the 20-114 [micro]m size fraction. The Eigenvalue is the percent of the variance explained by a given PC, Cum. Percent is the cumulative percentage of variance explained. The Eigenvectors reveal the strength of the association of the original variables with the principle components; larger values (+ or -) represent stronger association. Eigenvectors greater than 0.4 are shown in bold italics to emphasize those that are the strongest.
Principle Component 1 2 3 4 MARINE SITE Eigenvalue 3.6144 2.5281 2.0756 1.1890 Percent of Variance 32.8585 22.9824 18.8695 10.8088 Cum. Percent 32.8585 55.8409 74.7104 85.5192 Eigenvectors: Temperature -0.28933 0.46711 -0.10362 -0.11529 Salinity -0.24709 0.30528 -0.24691 0.46972 pH 0.47268 -0.09563 -0.21340 0.13271 In(Chlor) 0.25023 0.50083 0.16291 0.06381 In(Chlor/Phae) 0.29197 0.42788 -0.10629 -0.19992 ln(POC[20-114]) 0.33677 0.08724 0.31196 -0.11742 ln(POC[[less than]20]) -0.21197 -0.09565 0.51126 0.29204 C:N 0.18834 0.15761 0.14306 0.73424 [H.sub.2]C[O.sub.3] -0.36738 -0.05500 0.41102 -0.03818 DOC 0.13285 0.27671 0.52170 -0.22816 In(Discharge) 0.37084 -0.35267 0.15697 0.11281 OLIGOHALINE SITE Eigenvalue 4.5996 2.8681 1.7353 0.5866 Percent of Variance 41.8143 26.0735 15.7758 5.3326 Cum. Percent 41.8143 67.8878 83.6635 88.9961 Eigenvectors: Temperature 0.42389 0.03552 0.22899 -0.04058 Salinity 0.00543 -0.02381 -0.68922 0.18706 pH 0.37999 -0.27280 -0.16839 -0.00341 In(Chlor) 0.45605 0.02802 0.01792 0.06508 ln(Chlor/Phae) 0.38399 -0.18290 -0.13086 0.05356 In(POC[20-114]) 0.18613 0.48990 0.00291 -0.23306 In(POC[[less than]20]) 0.31343 0.35281 0.05951 0.41582 In(In(C:N)) 0.06573 0.46259 -0.17682 -0.67271 [H.sub.2]C[O.sub.3] 0.05828 -0.27083 0.57514 -0.20683 DCC -0.04071 0.47236 0.24336 0.47194 In(Discharge) -0.42215 0.12074 0.04370 0.10476
Multiple regression (MR) analysis of data from the Marine site. The isotopic data, i.e., [[delta].sup.13]C and [[delta].sup.15]N in the 20-114 [micro]m and [less than]20 [micro]m size fractions, respectively, are the dependent variables. Independent variables are the top 4 PCs, listed in order of their probability of significance (Prob) in the MR. PC 5 explained [less than]5% of the variance of the independent variables and was excluded from the analyses. [r.sup.2] and adjusted [r.sup.2] are the coefficient of determination and that adjusted for the proper degrees of freedom as the PCs are entered in the order listed.
Parameter Prob [r.sup.2] Adjusted [r.sup.2] DEPENDENT VARIABLE: [[delta].sup.13]C OF THE 20-114 [micro]M SIZE FRACTION PC 2 0.0004 0.6571 0.6571 PC 1 0.1121 0.7299 0.6849 PC 4 0.3683 0.7512 0.6833 PC 3 0.5524 0.7602 0.6643 DEPENDENT VARIABLE: [[delta].sup.15]N OF THE 20-114 [micro]M SIZE FRACTION PC 4 0.0028 0.3872 0.3401 PC 1 0.0039 0.7356 0.6915 PC 2 0.5563 0.7443 0.6746 PC 3 0.6227 0.7508 0.6511 DEPENDENT VARIABLE: [[delta].sup.13]C OF THE [less than]20 [micro]M SIZE FRACTION PC 4 0.1570 0.1597 0.0951 PC 2 0.2683 0.2535 0.1290 PC 3 0.3874 0.3092 0.1208 PC 1 0.7309 0.3177 0.0448 DEPENDENT VARIABLE: [[delta].sup.15]N OF THE [less than]20 [micro]M SIZE FRACTION PC 4 0.0103 0.2792 0.2237 PC 3 0.0138 0.5289 0.4504 PC 2 0.0293 0.7105 0.6316 PC 1 0.5993 0.7188 0.6063
Multiple regression (MR) analysis of data from the Oligohaline site. The isotopic data, i.e., [[delta].sup.13]C and [[delta].sup.15]N in the 20-114 [micro]m and [less than]20 [micro]m size fractions, respectively, are the dependent variables. Independent variables are the PCs, listed in order of their probability of their significance (Prob) in the MR. PC 5 explained [less than]5% of the variance of the independent variables and was excluded from the analyses. [r.sup.2] and adjusted [r.sup.2] are the coefficient of determination and that adjusted for the proper degrees of freedom as the PCs are entered in the order listed.
Parameter Prob [r.sup.2] Adjusted [r.sup.2] DEPENDENT VARIABLE: [[delta].sup.13]C OF THE 20-114 [micro]M SIZE FRACTION PC 3 0.1906 0.1477 0.0822 PC 1 0.4509 0.1939 0.0595 PC 2 0.4656 0.2370 0.0289 PC 4 0.6809 0.2504 -0.049 DEPENDENT VARIABLE: [[delta].sup.15]N OF THE 20-114 [micro]M SIZE FRACTION PC 1 0.0161 0.3440 0.2935 PC 4 0.1087 0.4717 0.3836 PC 2 0.1376 0.5789 0.4641 PC 3 0.6454 0.5882 0.4235 DEPENDENT VARIABLE: [[delta].sup.13]C OF THE [less than]20 [micro]M SIZE FRACTION PC 1 0.0030 0.4892 0.4499 PC 2 0.0440 0.6605 0.6040 PC 4 0.5304 0.6742 0.5853 PC 3 0.7690 0.6771 0.5480 DEPENDENT VARIABLE: [[delta].sup.15]N OF THE [less than]20 [micro]M SIZE FRACTION PC 2 0.3756 0.0739 0.0026 PC 3 0.5875 0.1009 -0.049 PC 4 0.6383 0.1211 -0.119 PC 1 0.6421 0.1408 -0.203
|Printer friendly Cite/link Email Feedback|
|Author:||WAINRIGHT, SAM C.; FULLER, CHARLOTTE M.; MCGUINNESS, LORA R.|
|Publication:||Bulletin of the New Jersey Academy of Science|
|Date:||Mar 22, 1999|
|Previous Article:||ASYMMETRICAL EMERGENCE AND DELAY IN EMERGENCE OF RIGHT FORELIMB IN THYROXINE-INDUCED METAMORPHOSIS OF THE BULLFROG, RANA CATESBEIANA.|
|Next Article:||A MODEL OF THE EFFECTS OF LIMITED PREY AVAILABILITY ON THE SURVIVAL OF WEAKFISH (CYNOCION REGALIS) LARVAE.|