JEQ Journal of Natural Resources and Life Sciences Education
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF) Free
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Related articles in JEQ
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via ISI Web of Science (8)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Malecki, L. M.
Right arrow Articles by Reddy, K. R.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Malecki, L. M.
Right arrow Articles by Reddy, K. R.
Agricola
Right arrow Articles by Malecki, L. M.
Right arrow Articles by Reddy, K. R.
Related Collections
Right arrow Biogeochemical Processes
Right arrow Wetlands and Aquatic Processes
Right arrow Nitrogen
Right arrow Phosphorus
Right arrow Soil Biochemistry
Published in J. Environ. Qual. 33:1545-1555 (2004).
© ASA, CSSA, SSSA
677 S. Segoe Rd., Madison, WI 53711 USA

TECHNICAL REPORTS

Wetlands and Aquatic Processes

Nitrogen and Phosphorus Flux Rates from Sediment in the Lower St. Johns River Estuary

Lynette M. Malecki, John R. White* and K. R. Reddy

Wetland Biogeochemistry Laboratory, Soil and Water Science Department, University of Florida, 106 Newell Hall, P.O. Box 110510, Gainesville, FL 32611

* Corresponding author (jrwhite{at}ufl.edu).

Received for publication July 9, 2003.

   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
Internal cycling of nutrients from the sediment and water column can be an important contribution to the total nutrient load of an aquatic ecosystem. Our objective was to estimate the internal nutrient loading of the Lower St. Johns River (LSJR). Dissolved reactive phosphorus (DRP) and ammonium (NH4–N) flux from sediments were measured under aerobic and anaerobic water column conditions using intact cores, to estimate the overall contribution of the sediments to P and N loading to the LSJR. The DRP flux under aerobic water column conditions averaged 0.13 mg m–2 d–1, approximately 37 times lower than that under anaerobic conditions (4.77 mg m–2 d–1). The average NH4–N released from the anaerobic cores (18.03 mg m–2 d–1) was also significantly greater than in the aerobic cores for all sites and seasons, indicating the strong relationship between nutrient fluxes and oxygen availability in the water column. The mean annual internal DRP load was estimated to be 330 metric tons (Mg) yr–1, 21% of the total P load to the river, while the mean annual internal load of NH4–N was determined to be 2066 Mg yr–1, 28% of the total N load to the LSJR estuary. As water resource managers reduce external loading to the LSJR the frequency of anaerobic events should decline, thereby reducing nutrient fluxes from the sediment to the water column, reducing the internal loading of DRP and NH4–N. Results from this study demonstrate that the internal flux of nutrients from sediments may be a significant portion of the total load and should be accounted for in the total nutrient budget of the river for successful restoration.

Abbreviations: BB, Beauclair Bluff • CC, Collee Cove • DL, Doctors Lake • DRP, dissolved reactive phosphorus • LOI, loss on ignition • LSJR, Lower St. Johns River • MBP, microbial biomass phosphorus • RP, Racy Point • SOD, sediment oxygen demand • TMDL, total maximum daily load • TN, total nitrogen • TP, total phosphorus


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
NUTRIENT ENRICHMENT of estuarine waters can result in accelerated accumulation of organic matter in sediments, thus resulting in nutrient flux from the sediment to the overlying water column. This can result in a variety of ecological responses including increased primary productivity (Nixon, 1992) and algal blooms, which tend to occur under low-flow conditions and/or warm water temperatures. Low dissolved oxygen levels can result as the detrital algae and other organic matter undergoes microbial decomposition (Carpenter, 1987; Parker and O'Reilly, 1991). This resulting anoxia in the water column often results in fish kills as well as the death of shellfish and other benthic organisms. Additionally, algal blooms can cause decreased coverage of submerged aquatic vegetation from poor penetration of sunlight through the water column, resulting in a loss of habitat and food source for many fish species (Cadenhead, 1997).

The 1972 Clean Water Act required states to identify impaired water bodies and establish total maximum daily loads (TMDLs) to restore eutrophic aquatic systems. A TMDL is the sum of all point- and nonpoint-source load allocations with an included safety margin and consideration for seasonal variations (Maher, 1997; Florida Administrative Code, 2004). Internal cycling of nutrients between the sediment and water column must also be considered as a contribution to the load. The TMDL calculation is based on the maximum amount of pollutant that the water body can assimilate without exceeding water quality standards. Determining TMDLs is very complex, requiring modelers to develop individualized water quality models for each impaired water body. These models are based on reaction rates of several processes and often require collection of significant amounts of data not available in the literature (Bazel, 1998).

The Lower St. Johns River (LSJR) estuary was selected to have TMDLs set for nutrients by 2004. Historically, the LSJR has directly received wastewater from sewage treatment plants and industrial activity along with runoff from agriculture, pasture lands, residential lawns, and streets and highways from the surrounding watershed. Industrial, commercial, and residential development have contributed to a degradation of water quality. Both the urban area of Jacksonville and the row crop agricultural area of the southeastern part of the basin are significant contributors of point- and nonpoint-source pollution (Hendrickson and Konwinski, 1998). For this reason, the NOAA and USEPA (USEPA, 1988) found the LSJR to be among the top four in amount of nutrient pollution. Population densities and species diversity of fish and wildlife inhabiting the LSJR basin have declined as a result of this eutrophication (DeMort, 1991; Keller and Schell, 1993).

The sediments in the LSJR are primarily fine-textured silts and clays; however, their appearance varies widely along the river, ranging from a gelatinous, highly flocculent muck to a coarse shell hash (Keller and Schell, 1993). Floc sediments in the LSJR are fine-grained, and easily resuspended. This resuspension can cause increased biological oxygen demand (BOD), sediment oxygen demand (SOD), and the release of sediment-bound nutrients into the water column.

Quantifying the nutrient flux within a system can aid managers in determining the importance of controlling internal or external nutrient loading. Nutrient loading results not only in an increase in the total nutrient content of the sediment but also increases the concentration of soluble forms that can be released into the overlying water column (Reddy et al., 1998). There are many methods available to quantify the extent of nutrient flux between the sediment–water interface. Simple one-dimensional (depth) diagenetic models such as Fick's first law of diffusion are often used to estimate the vertical flux based on close-interval estimation of pore water concentration gradients (Berner, 1980; Kemp, 1989). One-dimensional diffusion models assume microbial-mediated processes are uniform horizontally (American Society of Limnology and Oceanography, 1998). The nutrient gradients can be measured using pore water equilibration devices or sediment samples (D'Angelo and Reddy, 1994; Moore et al., 1998; Fisher and Reddy, 2001). Another method of determining nutrient flux is achieved by measuring changes in water column concentrations of intact sediment cores over time (Moore et al., 1998; Eckert and Nishri, 2000; Fisher and Reddy, 2001), which was the method used in this study.

Temporal variations in aquatic systems can have direct and indirect effects on factors influencing nutrient fluxes (Thayer, 1971). The rate of microbial processes and structure of the microbial community is largely dependent on environmental factors. Increased temperatures, for example, result in increased process rates (Christian, 1989). Nutrient concentrations and distributions have therefore been documented as having seasonal patterns (Boynton, 1974; Nixon et al., 1976; Baird and Ulanowicz, 1989; Morris, 2000). Seasonal cycles are due to imbalances in the processes of mineralization and consumption (Morris, 2000). Studies have shown there is typically a mid-summer peak in dissolved reactive phosphorus (DRP) and ammonium (NH+4) concentrations in surface waters, attributed to temperature-regulated respiratory regeneration or phytoplankton uptake, as well as changes in sediment redox conditions resulting in benthic regeneration (Kemp, 1989). Variations in phytoplankton populations along with nutrient availability will also affect the sediment oxygen demand (SOD) seasonally.

The objective of this study was to estimate the internal load of dissolved NH4–N and P in the LSJR. Internal cycling of nutrients from the sediment and water column can be an important contribution to the total nutrient loading of the system. Before establishing TMDLs, knowledge of both the external and internal nutrient loads are necessary to better understand the nutrient dynamics of the river. Therefore, specific objectives were to: (i) determine the sediment characteristics seasonally and spatially, (ii) determine the influence of aerobic and anaerobic water column conditions on NH4–N and DRP flux from benthic sediments, and (iii) estimate the annual internal load of P and N from the sediment to the overlying water column.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
Site Description
The St. Johns River is the longest river in Florida, stretching approximately 480 km (DeMort, 1991). It is characterized as an elongated, shallow, blackwater coastal plain river (Keller and Schell, 1993; Dame et al., 2000) with headwaters arising from freshwater marshes in St. Lucie and Indian River Counties. The main source of water storage for the headwaters marsh is the Blue Cypress Lake. The river flows northward to Jacksonville, FL, emptying into the Atlantic Ocean at Mayport, FL. The entire St. Johns River basin drains approximately 22790 km2 of land, nearly one-fifth of the state of Florida (Snell and Anderson, 1970; DeMort and Bowman, 1985).

The LSJR estuary makes up the northern 163-km portion of the St. Johns River from the mouth of the Ocklawaha River in Putnam County to the inlet at the Atlantic Ocean in Duval County. The average gradient of the river is 0.022 m km–1 with an average tidal amplitude of 1.5 m at the ocean inlet. Tidal effects are evident throughout the entire LSJR because of the low gradient and flat topography of the basin. General water quality characteristics of the LSJR include a mean secchi depth of 73 cm, chlorophyll {alpha} of 12.5 mg L–1, total phosphorus (TP) of 0.10 mg L–1, and mean total nitrogen (TN) of 1.1 mg L–1 (Hendrickson, 1994).

The LSJR is divided into three zones based on salinity. The freshwater lacustrine zone in the southern region stretches from south of Palatka to north of Green Cove Springs with an average salinity of 0.5 ppt. The riverbed broadens into the tidal oligohaline lacustrine zone, with an extensive floodplain, which is shallow and slow-moving, spanning from Doctors Lake north to the Fuller Warren Bridge in Jacksonville with a mean salinity of 2.9 ppt. Finally, the northern region from the Fuller Warren Bridge north to the Atlantic Ocean is considered mesohaline riverine, becoming deeper, fast-moving, and well-mixed with a mean salinity of 14.5 ppt (Maher, 1997; Hendrickson and Konwinski, 1998).

Field Sampling and Analysis
Four sampling locations were selected as representative of conditions throughout the length of the river (Fig. 1). Two of the sites were in the northern oligohaline lacustrine portion of the river at Beauclair Bluff (BB) and Doctors Lake (DL), and two were in the southern freshwater lacustrine portion at Collee Cove (CC) and Racy Point (RP). Beauclair Bluff was located near the middle of the river just north of the Fuller Warren Bridge in Jacksonville, having mud sediment. Doctors Lake, south of Jacksonville, had extremely flocculent fluid-like sediment, characteristic of a highly eutrophic lake. Sampling at CC in the southern region of the river occurred near the cove's wide opening to the river and contained sandy mud sediment. Finally, RP was the southernmost site located near the middle of the river with well-consolidated muddy sediment. Samples were collected during the three seasons representing north-central Florida's annual weather variations to determine temporal variability. The first sampling occurred in June 2001 during the warm, dry season, the second in October 2001 during the hot, rainy season, and the third in March 2002 during the cool, moderately dry season.



View larger version (100K):
[in this window]
[in a new window]
  		Fig. 1. Location of temporal study sample sites along the Lower St. Johns River estuary.

 
Triplicate cores (7-cm i.d.) were collected at each station and partitioned into depth increments of 0 to 10 and 10 to 20 cm for sediment characterization. The following physical and chemical parameters were measured on the sediment samples: pH, bulk density (Blake and Hartge, 1986), mass loss on ignition (LOI), microbial biomass phosphorus (MBP), Ca, Mg, Al, and Fe, TP, and TN. Microbial biomass P was determined by a 24-h chloroform fumigation–extraction (CFE) technique after Brookes et al. (1982). To determine the Ca, Mg, Fe, and Al, oven-dried sediment was treated with 25 mL of 1.0 M HCl for 3 h. The supernatant was filtered through 0.45-µm membrane filters and analyzed for Ca, Mg, Al, and Fe (DeBusk et al., 1994; Reddy et al., 1998). Metal analyses were determined by inductively coupled argon plasma spectrometry (Spectro Ciros CCD; Spectro AI, Fitchburg, MA). Total P analysis involved combustion of 0.5-g oven-dried subsamples at 550°C for 4 h in a muffle furnace followed by dissolution of the ash in 6 M HCl on a hot plate (Anderson, 1976). Total P was analyzed using an automated ascorbic acid method (Method 365.4; USEPA, 1993). Ash content was calculated to determine mass LOI, indicating the amount of organic matter content in the estuarine sediment (Luczak et al., 1997). Total N was analyzed on dried ground samples using a Carlo-Erba NA-1500 C-N-S Analyzer (Haak-Buchler Instruments, Saddlebrook, NJ) (DeBusk et al., 1994).

Another set of triplicate cores (7-cm i.d.) was collected to determine the SOD at each site. Using the procedure described by Fisher and Reddy (2001), the floodwater in each core was replaced with filtered, oxygen-saturated site water. The water was filtered through no. 4 Whatman (Maidstone, UK) filter paper (20-µm) to remove any algae present and bubbled with room air for 24 h to achieve 95 to 100% O2 saturation. After water was added to achieve a 15-cm water column, the cores were immediately sealed and incubated in the dark for eight or more hours. The dissolved oxygen (DO) concentrations were recorded at designated intervals using a YSI Model 58 DO meter (Yellow Springs Instrument Company, Yellow Springs, OH) equipped with a YSI Model 5905 BOD stirring probe. The probe was inserted through a pre-drilled port in each stopper to collect readings over a minimum 8-h period. The SOD rates for the initial fast phase of consumption and slower consumption phase that followed were calculated as:

where SOD is sediment oxygen demand (mg cm–2 h–1), DOi is initial DO reading (mg L–1), DOf is final DO reading (mg L–1), T is time (h), and SA = {pi}r2, surface area of core (cm2).

Nutrient flux rates were determined by measuring changes in water column concentrations of intact sediment cores over time (Moore et al., 1998; Fisher and Reddy, 2001). Two sets of triplicate cores (7-cm i.d.) were collected at each site to study the nutrient flux, one set each under anaerobic and aerobic water column conditions. Floodwater was replaced with filtered (no. 4 Whatman filter paper) site water to maintain a 20-cm water column to initiate the intact core flux study. Cores were allowed to equilibrate overnight with water columns purged with either O2–free N2 gas (anaerobic treatment) or aerated with room air using aquarium pumps (aerobic treatment). Redox probes were installed in the water column at mid-depth and at a 5-cm depth in sediment of two cores selected randomly from each treatment to confirm anaerobic and aerobic conditions throughout the study. The cores were covered with an opaque foil shroud to exclude light and incubated at an average temperature of 21°C. Water samples were taken at designated intervals (0, 2 h, 4 h, 8 h, 0.5 d, 1 d, 2 d, 5 d, 10 d, 15 d, 20 d, 25 d) over a 25-d period, filtered through 0.45-µm syringe filters, and analyzed for DRP and NH4–N. The volume of water taken for each sampling from the cores (15 mL) was replaced with filtered site water.

Flux calculations were based on the change in water column concentrations of DRP and NH4–N over time. These gradients were used to estimate the diffusive P and N fluxes. Initial and average flux rates were calculated using the linear portion of the concentration versus time curves for the DRP under anaerobic and aerobic water column conditions and NH4–N under anaerobic water column conditions. Nitrification rates were calculated in the cores with aerobic water columns using the linear portion of the NH4–N concentration versus time curves.

Statistical Analysis
The mean and standard deviation of each parameter used for sediment characterization, the nutrient fluxes, and SOD were calculated with Excel (Microsoft, 2000). Additionally, linear regression analysis was used to determine the relationship between HCl-extractable Fe and SOD consumption rates (Microsoft, 2000). Data normality was determined using the Kolmogorov–Smirnov test. One-way ANOVAs were performed to determine significant differences between sediment characteristics of the northern and southern regions of the river (Minitab, 2000). One-way ANOVAs and multiple comparisons by Fisher's least significant difference (LSD) were used to determine significant differences (p < 0.05) in nutrient fluxes and the SOD among stations and seasons (Minitab, 2000). Parameters that did not follow a normal distribution were analyzed nonparametrically using the Kruskal–Wallis ANOVA on ranks test, and differences between medians were determined using the Wilcoxon rank sum test (Minitab, 2000). Data was log-transformed to fit a normal distribution where possible (Microsoft, 2000).


   RESULTS AND DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
CONCLUSIONS
 REFERENCES
 
Sediment Characterization
There were two distinct patterns apparent in the sediment characteristics (0–10 cm) separating the LSJR in the northern (BB and DL) and southern regions (CC and RP) of the river (Table 1). The log-transformed organic matter content (LOI) was significantly greater (p < 0.01) in the north, while bulk density was greater (p < 0.01) in the south. The organic matter allows the fine-textured sediment to flocculate and remain porous, typically decreasing the bulk density (Brady and Weil, 1999). Sediment TP and TN were significantly greater (p < 0.01) in the north than in the south. There was no significant difference in Fe content (log-transformed) between the northern and southern region of the river. However, the mean Al and Mg sediment concentrations were significantly greater (p < 0.01) in the north than the south, and the log-transformed Ca concentrations were significantly greater (p < 0.01) in the south than in the north. The spatial differences between parameters can most likely be attributed to differences in land use in the surrounding watershed. The northern sites were influenced by point- and nonpoint-source pollution from metropolitan Jacksonville, while the area proximal to the southern sites was dominated by agricultural practices.


View this table:
[in this window]
[in a new window]
  		Table 1. Mean sediment characterization data for the 0- to 10-cm depth at the two northern (Beauclair Bluff [BB] and Doctors Lake [DL]) and southern (Collee Cove [CC] and Racy Point [RP]) sampling locations.{dagger}

 
Sediment Oxygen Demand
The dissolved oxygen levels in the sealed cores rapidly decreased initially, and then decreased at a much slower rate for the remainder of the studies (Fig. 2). Calculations were therefore divided into fast (initial) and slow oxygen consumption rates. This two-phase behavior can be related to separate processes occurring in the cores (Reddy et al., 1980). Chemical oxidation generally occurs faster than biochemical reactions (Ross and Potos, 1968). Therefore, initial O2 consumption was attributed primarily to the chemical oxidation, while the second phase of O2 consumption was attributed to both chemical and microbial-mediated oxidation (Reddy et al., 1980).



View larger version (19K):
[in this window]
[in a new window]
  		Fig. 2. Mean oxygen consumption in sediment cores collected in March 2002 at Beauclair Bluff (BB), Doctors Lake (DL), Collee Cove (CC), and Racy Point (RP) (n = 3 per site).

 
The initial fast oxygen consumption rates at DL, CC, and RP all followed the same seasonal trend decreasing from October to March (Table 2). The initial SOD consumption rates did not correlate with the HCl-extractable Fe concentrations at most sites (r2 ranged from 0.93 at CC to 0.08 at BB) indicating that the initial phase may not be solely attributed to Fe oxidation in these estuarine sediments. Using the combined monthly data (n = 8 or 9) from each site, a one-way ANOVA was used to determine that DL had a significantly greater fast oxygen consumption rate than the other three sites, consistent with the significantly higher organic matter present in DL.


View this table:
[in this window]
[in a new window]
  		Table 2. Mean intact core sediment oxygen demand over time for the four sampling locations.{dagger}

 
The slow O2 consumption phase was attributed to microbial respiration as well as continued chemical oxidation. The decrease at three of the four sites from June to October may be attributed to decreased microbial biomass (indicated by MBP) and activity in the surface layer of sediment from BB, CC, and RP (Table 2). There were no significant differences in the log-transformed slow SOD reaction rates between sites, suggesting no discernable difference in microbial activity among the sites.

One-way ANOVAs were used to determine seasonal differences in the mean oxygen consumption rates across sites. There were no significant seasonal differences in the initial SOD rates; however, the second-phase oxygen consumption rates were significantly greater in June 2001, averaging 0.001 mg cm–2 h–1. Greater consumption in June is indicative of the increased microbial activity during central Florida's warm months due to the increased water column productivity resulting in more bioavailable material than in the winter months.

Dissolved Reactive Phosphorus Flux
Dissolved reactive P concentrations released from the sediment under an anaerobic water column were significantly (p < 0.01) greater than the flux under the aerobic water column for all sites and seasons (Fig. 3). This release of P was also seen in Lake Okeechobee sediments under similar anaerobic conditions (Olila and Reddy, 1997; Moore et al., 1998). This release is likely attributed to Fe reduction due to the lower redox conditions in the sediment under anaerobic conditions (Mortimer, 1971). Phosphorus removal under oxic conditions is usually attributed to its binding with ferric iron (Fe3+), forming insoluble complexes (Upchurch et al., 1974). However, under anoxic conditions Fe3+ is reduced to soluble ferrous iron (Fe2+) leading to liberation of P to the overlying water column (Lee et al., 1977). The DRP concentrations in the water column continued to increase over time in the anaerobic water column while staying relatively low and stable over time in the aerobic cores (Fig. 3).



View larger version (41K):
[in this window]
[in a new window]
  		Fig. 3. Changes in dissolved reactive phosphorus concentration of the water column under (a) anaerobic and (b) aerobic conditions from sediment cores of Beauclair Bluff (BB), Doctors Lake (DL), Collee Cove (CC), and Racy Point (RP) (n = 3 per site).

 
The Kruskal–Wallis test (n = 12) was used to determine seasonal and spatial differences in the anaerobic DRP flux rates (Table 3). The initial P release from the sediment was significantly (p < 0.01) greater in June 2001 under anaerobic conditions when compared with the October 2001 or March 2002 samples. This difference may be attributed to increased microbial biomass and activity in the surface layer of sediment during the summer months due to more bioavailable organic matter present on the sediment surface. There was no significant difference in anaerobic initial P flux rates between sites (Table 4). Beauclair Bluff, the northernmost site closest to urban Jacksonville, did have a significantly greater average anaerobic DRP flux (6.72 mg m–2 d–1) than RP, which is located in the southernmost sample location. The anaerobic mean average DRP flux was 4.55 mg m–2 d–1 for all sites.


View this table:
[in this window]
[in a new window]
  		Table 3. Mean temporal variations in phosphorus and ammonium N flux rates from Lower St. Johns River (LSJR) sediment under anaerobic and aerobic water column conditions.{dagger}

 

View this table:
[in this window]
[in a new window]
  		Table 4. Spatial and temporal variations in dissolved reactive phosphorus flux from sediment under anaerobic and aerobic water column conditions.{dagger}

 
Sediment cores treated aerobically from BB, DL, and CC had negative initial P fluxes as the sediment adsorbed P in June 2001 and October 2001 (Table 4). Therefore, in March 2002, there was a significantly (p < 0.01) greater initial P flux out of the sediment into the water column than June 2001 and October 2001 (Table 3), which may have been due to fresh material deposited into the river. However, the exact opposite was true for the average P fluxes. June and October average P fluxes out of the sediment were both significantly greater (p < 0.05) than the March average P flux into the sediment. There were no significant differences between sites for the initial nor average P fluxes in the aerobic cores. The aerobic mean DRP average flux was 0.13 mg m–2 d–1 for all sites.

Several recent studies have been performed on P fluxes in a variety of aquatic systems using intact core incubation (Table 5). When comparing flux rates from fine-grained muddy sediment in this study to similar intact core studies, the mean DRP average anaerobic flux rates in the LSJR (8.13 to 3.22 mg m–2 d–1) range from two to five times greater than the maximum fluxes from the sand and mud of the Indian River Lagoon and Lake Okeechobee (1.54 mg m–2 d–1) (Moore et al., 1998; Reddy et al., 2001) as well as from the peat of South Bay (0.77 mg m–2 d–1) (Moore et al., 1998). The greater flux rates in this study are related to the accretion of nutrient-rich material in the LSJR, which could potentially continue to drive eutrophic conditions under low oxygen conditions. The maximum release of P from the LSJR sediments under aerobic conditions (0.60 mg m–2 d–1) was much lower than maxima seen in the Indian River Lagoon, FL, and the Swan–Canning Estuary, Australia. This suggests that as the water quality in the LSJR improves and low oxygen events decrease due to decreases in the external load, the sediments will release less P over time, decreasing internal eutrophication.


View this table:
[in this window]
[in a new window]
  		Table 5. Ranges of sediment–water dissolved reactive phosphorus fluxes reported in previous studies.

 
Ammonium Flux
Ammonium N concentrations released from the sediment under an anaerobic water column were significantly (p < 0.01) greater than under aerobic conditions for all sites and seasons. Under anaerobic conditions, NH4–N was released from the sediment continuously, increasing water column concentrations over time (Fig. 4). However, under aerobic conditions, NH4–N concentrations rapidly decreased and then leveled off in the water column over time. Recall that nitrification is an obligatory aerobic process where NH+4 is oxidized to nitrate (NO–3) via chemoautotrophs, which use reduced, inorganic N as their energy source through oxidation to fix inorganic carbon (Christian, 1989). Therefore, NH4–N was released in the anaerobic treatment by microbial decomposition of organic matter and built up in the water column because nitrification was inhibited due to lack of O2. However, under aerobic conditions, NH4–N was released from the sediment but the concentration did not build up in the water column due to the immediate onset of nitrification. Once the nitrifying microbial population was well established (less than one day), NH4–N concentrations dropped to 0.02 mg L–1 and remained low for the remainder of the incubation (Fig. 4). The aerobic nitrification rates ranged from 9.80 to 242 mg m–2 d–1 with mean site nitrification rates exceeding mineralization rates (Table 5), hence leading to low NH4–N concentrations over the length of the aerobic incubation.



View larger version (43K):
[in this window]
[in a new window]
  		Fig. 4. Changes in ammonium concentration of the water column under (a) anaerobic and (b) aerobic conditions from sediment cores of Beauclair Bluff (BB), Doctors Lake (DL), Collee Cove (CC), and Racy Point (RP) (n = 3 per site).

 
The average NH4–N flux from the sediment into the anaerobic water column was significantly (p < 0.01) greater in June 2001 than October 2001 and March 2002 (Table 3). This difference may be due to decreased microbial biomass (indicated by MBP) that followed the same trend in the surface layer at these sites, as well as the decreased water column productivity in the winter months resulting in less bioavailable material, decreasing microbial activity. Sediments from BB had the greatest mean initial NH4–N flux (40.2 mg m–2 d–1) followed by those from CC, DL, and RP respectively, similar to the trend found in the initial SRP flux rates (Table 6). There were no significant differences between sites for the initial nor average NH4–N flux rates under anaerobic conditions. There were also no significant differences between the aerobic nitrification rates (log-transformed) at each site. However, the nitrification rates (log-transformed) were significantly different seasonally (Table 3). The nitrification rates were significantly (p < 0.05) less in March 2002, averaging 20.7 mg m–2 d–1, than in June 2001 and October 2001, most likely due to the decreased microbial activity during the winter months.


View this table:
[in this window]
[in a new window]
  		Table 6. Spatial and temporal variations in ammonium flux and nitrification rates from sediment under anaerobic and aerobic water column conditions.{dagger}

 
Several studies have been performed on NH4–N fluxes in a variety of aquatic systems using intact core incubation (Table 7). When comparing flux rates from the fine-grained muddy sediment of the LSJR (13.1–23.9 mg m–2 d–1) to similar intact core studies, the mean average anaerobic NH4–N flux rates in the LSJR are within the range of flux rates from mud sediment of Morlaix Bay, France (21.0 mg m–2 d–1) (Lerat et al., 1990) and higher than fluxes from the sand sediment of the Neuse River estuary (9.50 mg m–2 d–1) (Rizzo et al., 1992), while being comparable or slightly greater than fluxes reported in the Indian River Lagoon (17.6–36.3 mg m–2 d–1) (Reddy et al., 2001).


View this table:
[in this window]
[in a new window]
  		Table 7. Ranges of sediment–water ammonium N fluxes reported in previous studies.

 
Annual Internal Loads
The seasonal and annual internal loads of P and N were calculated using the mean DRP and the anaerobic NH4–N flux rates (Table 8). A shortcoming of the annual internal load calculation is that detailed oxygen profiles are not available for the river, and consequently there was no way to weigh the aerobic versus anaerobic average annual internal loads calculated. Therefore, to determine at least an order of magnitude estimation of the internal loading rates, the mean of the annual anaerobic and aerobic fluxes was used to calculate the total DRP load. Under completely anaerobic conditions, 628 more Mg of DRP would be released per year than under completely aerobic conditions over the 365-km2 area that the LSJR covers (Table 8). In the aerobic cores Fe3+ was available to bind P; however, under anaerobic conditions this reaction no longer occurred, allowing significantly more P to be released into the water column.


View this table:
[in this window]
[in a new window]
  		Table 8. Mean seasonal and annual internal loading rates of dissolved reactive phosphorus and ammonium using flux rates from the intact core studies (n = 4).

 
The calculated annual mean internal load of DRP from the sediment to the river was 330 Mg yr–1 and 2066 Mg yr–1 of NH4–N (Table 8). The internal load estimated from this study made up approximately 21% of the total P loading to the LSJR and 28% of the total N loading for the LSJR when compared with the 1195 Mg yr–1 external load of total P and 5414 Mg yr–1 external load of total N (Hendrickson and Konwinski, 1998). The contribution of the internal load may be even more significant when considering the internal load calculated in this study consists of nutrients entirely in the bioavailable inorganic form. The total N and P external loads entering the LSJR contain both bioavailable as well as more recalcitrant particulate forms.


   CONCLUSIONS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
REFERENCES
 
There was a spatial relationship in sediment characteristics between the northern and southern region of the river. The LOI, TP, TN, Al, and Mg concentrations were all significantly greater in the north (BB and DL), while bulk density and Ca concentrations were significantly greater in the south (CC and RP). Spatial differences may be attributed to the difference in land use. The northern sites were influenced by point- and nonpoint-source pollution from metropolitan Jacksonville, while the southern sites were influenced by agricultural practices. Nutrient loading from the northern industrial and urban land use seems to be contributing more toward the poor water quality persisting in the LSJR than the southern agricultural land use.

The average flux (all four sites) from the sediment to the water column using the anaerobic sediment cores was 4.77 and 18.0 mg m–2 d–1 for DRP and NH4–N, respectively, while for the aerobic cores DRP fluxes averaged 0.13 mg m–2 d–1. The DRP released under anaerobic conditions was forty times greater than the DRP released from the aerobic cores for all sites and seasons, indicating the strong relationship between nutrient fluxes and oxygen availability in the water column. This relationship suggests that periods of anoxia due to algal blooms will lead to greater releases of DRP from the sediment.

The nitrification rates in the LSJR are sufficient to convert almost all NH4–N released from decomposition of organic matter in the sediment to NO–3, maintaining low NH4–N concentrations in the aerobic surface waters. Therefore, the controlling mechanism for the mediation of inorganic N concentrations in the water column for this system is denitrification.

When results are extrapolated river-wide, the internal loading of P and N to the LSJR was found to be less than the external loading. However, this supply of N and P to the water column does represent a significant internal load that managers should take into account when determining the TMDLs for this system. The frequency of anaerobic events should decline as water resource managers reduce external loading to the LSJR. The total internal load of nutrients should, therefore, decrease as the dissolved oxygen concentrations are maintained higher throughout the year in the LSJR estuary.


   ACKNOWLEDGMENTS
 
This research was supported, in part, by the Florida Agricultural Experiment Station and is approved for publication as Journal Series no. R-09945. The study was funded by the St. Johns River Water Management District (SJRWMD). Special thanks to John Hendrickson of the SJRWMD for project coordination and Yu Wang of the Wetland Biogeochemistry Laboratory at the University of Florida for assistance with laboratory techniques.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 


Related articles in JEQ:

This Issue in Journal of Environmental Quality

JEQ 2004 33: 1177-1182. [Full Text]  



This article has been cited by other articles:


Home page
 J. Environ. Qual.Home page
L. M. Malecki-Brown, J. R. White, and K. R. Reddy
Soil Biogeochemical Characteristics Influenced by Alum Application in a Municipal Wastewater Treatment Wetland
J. Environ. Qual., October 24, 2007; 36(6): 1904 - 1913.
[Abstract] [Full Text] [PDF]

Home page
 Soil Sci.Home page
E. M. Bostic and J. R. White
Soil Phosphorus and Vegetation Influence on Wetland Phosphorus Release after Simulated Drought
Soil Sci. Soc. Am. J., January 1, 2007; 71(1): 238 - 244.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF) Free
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Related articles in JEQ
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via ISI Web of Science (8)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Malecki, L. M.
Right arrow Articles by Reddy, K. R.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Malecki, L. M.
Right arrow Articles by Reddy, K. R.
Agricola
Right arrow Articles by Malecki, L. M.
Right arrow Articles by Reddy, K. R.
Related Collections
Right arrow Biogeochemical Processes
Right arrow Wetlands and Aquatic Processes
Right arrow Nitrogen
Right arrow Phosphorus
Right arrow Soil Biochemistry

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
The SCI Journals Agronomy Journal Crop Science
Vadose Zone Journal Journal of Plant Registrations
Journal of Natural Resources
and Life Sciences Education		 Soil Science Society of America Journal