Barbara L. Nowicki

An assessment of the annual mass balance of carbon, nitrogen, and phosphorus in Narragansett Bay*

Narragansett Bay is a relatively well-mixed, high salinity coastal embayment and estuary complex in southern New England (USA). Much of the shoreline is urban and the watershed is densely developed. We have combined our data on C, N, and P inputs to this system, on C, N, and P accumulation in the sediments, and on denitrification with extensive work by others to develop approximate annual mass balances for these elements. The results show that primary production within the bay is the major source of organic carbon (4 times greater than other sources), that land drainage and upstream sewage and fertilizer are the major sources of N, and that landward flowing bottom water from offshore may be a major source of dissolved inorganic phosphorus. Most of the nutrients entering the bay arrive in dissolved inorganic form, though DON is a significant component of the N carried by the rivers. About 40% of the DIN in the rivers is in the form of ammonia. Sedimentation rates are low in most of Narragansett Bay, and it appears that less than 20% of the total annual input of each of these elements is retained within the system. A very small amount of C, N, and P is removed in fisheries landings, denitrification in the sediments removes perhaps 10–25% of the N input, and most of the carbon fixed in the system is respired within it. Stoichiometric calculations suggest that some 10–20% of the organic matter formed in the bay is exported to offshore and that Narragansett Bay is an autotrophic system. Most of the N and P that enters the bay is, however, exported to offshore waters in dissolved inorganic form. This assessment of the overall biogeochemical behavior of C, N, and P in the bay is consistent with more rigorously constrained mass balances obtained using large living models or mesocosms of the bay at the Marine Ecosystem Research Laboratory (MERL).

Benthic nutrient remineralization in a coastal lagoon ecosystem

In situ measurements of the exchange of ammonia, nitrate plus nitrite, phosphate, and dissolved organic phosphorus between sediments and the overlying water column were made in a shallow coastal lagoon on the ocean coast of Rhode Island, U.S.A. The release of ammonia from mud sediments in the dark (20–440 μmol per m2 per h) averaged ten times higher than from a sandy tidal flat (0–60 μmol per m2 per h), and while mud sediments also released nitrate and phosphate, sandy sediments took up these nutrients. Fluxes of nutrients from mud sediments, but not from sandy areas, markedly increased with temperature. Ammonia release rates for mud sediments in the light (0–350 μmol per m2 per h) were lower than those in the dark and it is estimated that some 25% of the ammonia released to the water column on an annual basis may be intercepted by the benthic microfloral community. Estimates of the annual net exchange of nutrients across the sediment-water interface, weighted by sediment type for the lagoon as a whole, showed a release of 450 mmol per m2 of ammonia, 5 mmol per m2 of phosphate, 5 mmol per m2 of dissolved organic phosphorus, and an uptake of 80 mmol per m2 of nitrate. Although rates of ammonia and nitrate exchange were comparable to those described for the deeper heterotrophic bottom communities of nearby Narragansett Bay, rates of benthic phosphate release were significantly lower. On an annual basis the Bay benthos released approximately 20 times more inorganic phosphate per unit area than did the lagoon benthos. As a result., the N/P ratio for the flux from the sediments was 74∶1 in the lagoon, compared with 16∶1 in “average” marine plankton and 8∶1 for the benthic flux from Narragansett Bay. The lack of remineralized phosphate in the lagoon, is reflected in water, column phosphate concentrations (always <1 μm) and water column N/P ratios (annual N/P=27) and suggests that the lagoon may show phosphate limitation rather than the nitrogen limitation commonly associated with marine systems.

Nitrogen losses through sediment denitrification in Boston Harbor and Massachusetts Bay

Sediment denitrification is a microbial process that converts dissolved inorganic nitrogen in sediment porewaters to N2 gas, which is subsequently lost to the atmosphere. In coastal waters, it represents a potentially important loss pathway for fixed nitrogen which might otherwise be available to primary producers. Currently, data are lacking to adequately assess the role of denitrification in reducing or remediating the effects of large anthropogenic nitrogen loads to the coastal zone. This study describes the results of 88 individual measurements of denitrification (as a direct flux of N2 gas) in sediment cores taken over a 3-yr period (1991–1994) from six stations in Boston Harbor, nine stations in Massachusetts Bay, and two stations in Cape Cod Bay. The dataset is unique in its extensive spatial and temporal coverage and includes the first direct measurements of denitrification for North Atlantic shelf sediments. Results showed that rates of denitrification were significantly higher in Boston Harbor (mean=54, range<5–206 μmol N2 m−2 h−1) than in Massachusetts Bay (mean=23, range<5–64 μmol N2 m−2 h−1). Highest rates occurred in areas with organic-rich sediments in the harbor, with slower rates observed for low-organic sandy sediments in the harbor and at shallow shelf stations in the bay. Lowest rates were found at the deepest shelf stations, located in Stellwagen Basin in Massachusetts Bay. Observed rates were correlated with temperature, sediment carbon content, and benthic macrofaunal activity. Seasonally, highest denitrification rates occurred in the summer in Boston Harbor and in the spring and fall in Massachusetts Bay, coincident with peak phytoplankton blooms in the overlying water column. Despite the fact that sediment denitrification rates were high relative to rates reported for other East Coast estuaries, denitrification losses accounted for only 8% of the annual total nitrogen load to Boston Harbor, a consequence perhaps, of the short water-residence times (2–10 d) of the harbor.

The role of sediment denitrification in reducing groundwater-derived nitrate inputs to Nauset Marsh estuary, Cape Cod, Massachusetts

The Nauset Marsh estuary is the most extensive (9.45 km2) and least disturbed salt marsh/estuarine system within the Cape Cod National Seashore, even though much of the 19 km2 watershed area of the estuary is developed for residential or commercial purposes. Because all of the Nauset watershed is serviced by on-site individual sewage disposal systems, there is concern over the potential impact of groundwater-derived nutrients passing from these systems to the shallow receiving waters of the estuary. The purpose of this study was to determine whether denitrification (the bacterial conversion of nitrate to gaseous nitrogen) in estuarine sediments could effectively remove the nitrate from contaminated groundwater before it passed from the watershed to the estuary. Rates of denitrification were measured both in situ and in sediment cores, in areas of active groundwater discharge, in relatively pristine locations, and in areas situated down-gradient of moderate to heavily developed regions of the watershed. Denitrification rates for 47 sediment cores taken over an annual cycle at 5 stations ranged from non-detectable to 47 μmol N2 m−2 h−. Mean denitrification rates were positively correlated with sediment organic content, and varied seasonally due to changes in sediment organic content and to the effect of water temperatures on sediment oxygen penetration depths. There was no correlation between observed denitrification rates and corresponding nitrate concentrations in groundwater. A comparison of in situ denitrification rates (supported by groundwater nitrate) with denitrification rates observed in sediment cores (supported by remineralized nitrate) showed that groundwater-driven denitrification rates were small, and not in excess of denitrification rates supported by remineralized nitrate. Most of the denitrification in Nauset sediments was apparently fueled by remineralized nitrate through coupled nitrification/denitrification. Denitrification did not contribute significantly to the direct loss of nitrate from incoming groundwater at Nauset Marsh estuary. Groundwater flow was rapid, and much of it occurred in freshwater springs and seeps through very coarse, sandy, well-oxygenated sediments of limited organic content. There was little opportunity for denitrification to occur during groundwater passage through these sediments. These results have important management implications because they suggest that the majority of nitrogen from contaminated groundwater crosses the sediment/water interface and arrives at Nauset Estuary, where it is available to primary producers. Preliminary budget calculations suggest that while denitrification was not an effective mechanism for the direct removal of nitrate in contaminated groundwater flowing to Nauset Marsh estuary, it may contribute to significant nitrogen losses from the estuary itself.

Denitrification Capacity in a Subterranean Estuary below a Rhode Island Fringing Salt Marsh

Coastal waters are severely threatened by nitrogen (N) loading from direct groundwater discharge. The subterranean estuary, the mixing zone of fresh groundwater and sea water in a coastal aquifer, has a high potential to remove substantial N. A network of piezometers was used to characterize the denitrification capacity and groundwater flow paths in the subterranean estuary below a Rhode Island fringing salt marsh.15N-enriched nitrate was injected into the subterranean estuary (in situ push-pull method) to evaluate the denitrification capacity of the saturated zone at multiple depths (125–300 cm) below different zones (upland-marsh transition zone, high marsh, and low marsh). From the upland to low marsh, the water table became shallower, groundwater dissolved oxygen decreased, and groundwater pH, soil organic carbon, and total root biomass increased. As groundwater approached the high and low marsh, the hydraulic gradient increased and deep groundwater upwelled. In the warm season (groundwater temperature >12 °C), elevated groundwater denitrification capacity within each zone was observed. The warm season low marsh groundwater denitrification capacity was significantly higher than all other zones and depths. In the cool season (groundwater temperature <10.5 °C), elevated groundwater denitrification capacity was only found in the low marsh. Additions of dissolved organic carbon did not alter groundwater denitrification capacity suggesting that an alternative electron donor, possibly transported by tidal inundation from the root zone, may be limiting. Combining flow paths with denitrification capacity and saturated porewater residence time, we estimated that as much as 29–60 mg N could be removed from 11 of water flowing through the subterranean estuary below the low marsh, arguing for the significance of subterranean estuaries in annual watershed scale N budgets.

DENITRIFICATION IN FRINGING SALT MARSHES OF NARRAGANSETT BAY, RHODE ISLAND, USA

In the past century, loading of terrestrial inorganic nitrogen to coastal receiving waters has increased dramatically. Salt marshes, because of their location between upland regions and coastal waters and their recognized role as nutrient transformers, have the potential to ameliorate some of this loading. In the current study, we used core incubations in the laboratory to investigate denitrification rates in high marsh soils from five fringing salt marshes in Narragansett Bay, Rhode Island, USA. The marshes showed a wide variety of terrestrial N loading, with rates ranging from 2 to 6037 kg N ha−1 yr−1. Field-collected cores were selected to include both vegetated and bare soils at each marsh, and the six-hour incubations were designed to approximate natural tidal rhythms. Total dissolved nitrogen flux in these marshes ranged between +1255 and −710 μmol N m−2 hr−1, with N2 gas accounting for the majority of the total N flux (average 76%). Nitrogen gas flux ranged between −375 (nitrogen fixation) and +420 (denitrification) μmol N2 m−2 hr−1. While N2 gas fluxes were significantly correlated (r=+0.64, p<0.05) with marsh organic carbon content, we also detected a significant inverse relationship (r=−0.91, P<0.05) between average N2 gas fluxes and terrestrial nitrogen loads. Comparison of N2 gas fluxes in vegetated vs. bare soils indicated a significant (p <0.05) but variable effect of vegetation on N2 flux. This field survey shows the potential of New England fringe salt marshes to intercept and transform land-derived nitrogen loads; however, sediment characteristics (e.g., percent of labile organic matter) and plant community structure can significantly affect the capacity of the marsh to process inorganic nitrogen loads. In order to understand the role of salt marshes in buffering coastal N loading, we need a better understanding of the natural and anthropogenic factors controlling denitrification and net N losses.

Net system production in coastal waters as a function of eutrophication, seasonality and benthic macrofaunal abundance

Net system production ranged from 13% to 29% of apparent system production in enclosures modelling coastal marine waters. Net production was measured by direct and indirect methods along with factors which impact its magnitude and fate. The direct measures of carbon, phosphorus, and nitrogen content of accumulating flocculent material in enclosures without sediment agreed with indirect measures by net system metabolism and by net sediment storage from nutrient mass balances. Increased nutrient supply, increased the absolute, but not relative, net system production ultimately stored in the sediment. Net production as dry weight of floc did not agree with that calculated from oxygen metabolism owing to a high silicon content of the organic matter. The presence of a benthic macrofauna decreased net system production storage by about 28–54%.

Groundwater Nitrogen Transport and Input along the Narragansett Bay Coastal Margin

The transport of dissolved nitrogen (N) in groundwater has been historically difficult to monitor, model, and predict. Quantitative assessments of its significance to Narragansett Bay and its sub-estuaries suffer from a paucity of information and a lack of direct studies. Two factors exerting the greatest impact on groundwater N delivery to Narragansett Bay are the characteristic surficial geology of the area, and the presence of densely populated un-sewered development in the coastal zone. Local differences in soils, geology, and hydrology are critical in determining the degree of N migration from nonpoint sources to Narragansett Bay. Ultimately, the transport of N via groundwater to Narragansett Bay is governed by the interaction of groundwater flow paths with local soils, geology, and land use. Rough mass balance calculations suggest that groundwater makes up less than 10% of the direct freshwater input to Narragansett Bay. Nevertheless, localized groundwater seepage to numerous coves and embayments lining the Narragansett Bay shoreline has had a significant impact on these smaller nearshore ecosystems. In areas with individual sewage disposal systems (ISDS or septic systems), groundwater flows provide a direct connection between human waste disposal and nearby marshes, rivers and sub-estuaries. Deteriorating habitat and water quality observed in many of the smaller coves and embayments of Narragansett Bay over the past 20 years have influenced the public’s perception of the general health of the coastal waters of ‘‘the Ocean State,’’ and have cast a pall over otherwise significant improvements in wastewater and coastal zone management for the bay as a whole. In this chapter, the interplay of local soils, coastal geomorphology, and land use are described, which in concert with the unique biogeochemistry of nitrogen, act to mediate groundwater N transformation and transport in the coastal