Search

Light, essential for photosynthesis, is present in two periodic cycles in nature: seasonal and diel. Although seasonality of light is typically resolved in ocean ecosystem and biogeochemistry models because of its significance for seasonal succession and biogeography of phytoplankton, the diel light cycle is generally not resolved. Here we use a three-dimensional global ocean model and compare high temporal resolution simulations with and without diel light cycles. The model simulates 15 phytoplankton types of different cell size, encompassing two broad ecological strategies: small cells with high nutrient affinity (gleaners) and larger cells with high maximal growth rate (opportunists). Both are grazed by zooplankton and limited by nitrogen, phosphorus and iron. Simulations show that diel cycles of light induce diel cycles in phytoplankton populations and limiting nutrients in the global ocean. Diel nutrient cycles are associated with higher concentration of limiting nutrients by up to 200% at low latitudes (-40° to 40°), a process that increases opportunists’ biomass by up to 50%. Size classes with the highest maximal growth rates from both gleaner and opportunist groups are favored the most by diel light cycles. This mechanism weakens as latitude increases because the effects of the seasonal cycle dominate over those of the diel cycle. The present work shows that resource competition under diel light cycles has a significant impact on phytoplankton biogeography, indicating the necessity of resolving diel processes in global ocean models.

Diverse microbial communities drive biogeochemical cycles in Earth’s ocean, yet studying these organisms and processes is often limited by technological capabilities, especially in the deep ocean. In this study, we used a novel marine microbial incubator instrument capable of in situ experimentation to investigate microbial primary producers at deep-sea hydrothermal vents., Depressurization and sample processing delays may impact the outcome of shipboard microbial incubations of samples collected from the deep sea. To address this knowledge gap, we developed a remotely operated vehicle (ROV)-powered incubator instrument to carry out and compare results from in situ and shipboard RNA stable isotope probing (RNA-SIP) experiments to identify the key chemolithoautotrophic microbes and metabolisms in diffuse, low-temperature venting fluids from Axial Seamount. All the incubations showed microbial uptake of labeled bicarbonate primarily by thermophilic autotrophic Epsilonbacteraeota that oxidized hydrogen coupled with nitrate reduction. However, the in situ seafloor incubations showed higher abundances of transcripts annotated for aerobic processes, suggesting that oxygen was lost from the hydrothermal fluid samples prior to shipboard analysis. Furthermore, transcripts for thermal stress proteins such as heat shock chaperones and proteases were significantly more abundant in the shipboard incubations, suggesting that depressurization induced thermal stress in the metabolically active microbes in these incubations. Together, the results indicate that while the autotrophic microbial communities in the shipboard and seafloor experiments behaved similarly, there were distinct differences that provide new insight into the activities of natural microbial assemblages under nearly native conditions in the ocean. IMPORTANCE Diverse microbial communities drive biogeochemical cycles in Earth’s ocean, yet studying these organisms and processes is often limited by technological capabilities, especially in the deep ocean. In this study, we used a novel marine microbial incubator instrument capable of in situ experimentation to investigate microbial primary producers at deep-sea hydrothermal vents. We carried out identical stable isotope probing experiments coupled to RNA sequencing both on the seafloor and on the ship to examine thermophilic, microbial autotrophs in venting fluids from an active submarine volcano. Our results indicate that microbial communities were significantly impacted by the effects of depressurization and sample processing delays, with shipboard microbial communities being more stressed than seafloor incubations. Differences in metabolism were also apparent and are likely linked to the chemistry of the fluid at the beginning of the experiment. Microbial experimentation in the natural habitat provides new insights into understanding microbial activities in the ocean.

Depressurization and sample processing delays may impact the outcome of shipboard microbial incubations of samples collected from the deep sea. To address this knowledge gap, we developed an ROV-powered incubator instrument to carry out and compare results from in situ and shipboard RNA Stable Isotope Probing (RNA-SIP) experiments to identify the key chemolithoautotrophic microbes and metabolisms in diffuse, low-temperature venting fluids from Axial Seamount. All the incubations showed microbial uptake of labelled bicarbonate primarily by thermophilic autotrophic Epsilonbacteraeota that oxidized hydrogen coupled with nitrate reduction. However, the in situ seafloor incubations showed higher abundances of transcripts annotated for aerobic processes suggesting that oxygen was lost from the hydrothermal fluid samples prior to shipboard analysis. Furthermore, transcripts for thermal stress proteins such as heat shock chaperones and proteases were significantly more abundant in the shipboard incubations suggesting that hydrostatic pressure ameliorated thermal stress in the metabolically active microbes in the seafloor incubations. Together, results indicate that while the autotrophic microbial communities in the shipboard and seafloor experiments behaved similarly, there were distinct differences that provide new insight into the activities of natural microbial assemblages under near-native conditions in the ocean.

In this manuscript we investigate the use of the maximum entropy production (MEP) principle for modeling biogeochemical processes that are catalyzed by living systems. Because of novelties introduced by the MEP approach, many questions need to be answered and techniques developed in the application of MEP to describe biological systems that are responsible for energy and mass transformations on a planetary scale. In previous work we introduce the importance of integrating entropy production over time to distinguish abiotic from biotic processes under transient conditions. Here we investigate the ramifications of modeling biological systems involving one or more spatial dimensions. When modeling systems with spatial dimensions, entropy production can be maximized either locally at each point in space asynchronously or globally over the system domain synchronously. We use a simple two-box model inspired by two-layer ocean models to illustrate the differences in local versus global entropy maximization. Synthesis and oxidation of biological structure is modeled using two autocatalytic reactions that account for changes in community kinetics using a single parameter each. Our results show that entropy production can be increased if maximized over the system domain rather than locally, which has important implications regarding how biological systems organize and supports the hypothesis for multiple levels of selection and cooperation in biology for the dissipation of free energy.

Determining how microbial communities organize and function at the ecosystem level is essential to understanding and predicting how they will respond to environmental change. Mathematical models can be used to describe these communities, but properly representing all the biological interactions in extremely diverse natural microbial ecosystems in a mathematical model is challenging. We examine a complementary approach based on the maximum entropy production (MEP) principle, which proposes that systems with many degrees of freedom will likely organize to maximize the rate of free energy dissipation. In this study, we develop an MEP model to describe biogeochemistry observed in Siders Pond, a phosphate limited meromictic system located in Falmouth, MA that exhibits steep chemical gradients due to density-driven stratification that supports anaerobic photosynthesis as well as microbial communities that catalyze redox cycles involving O, N, S, Fe and Mn. The MEP model uses a metabolic network to represent microbial redox reactions, where biomass allocation and reaction rates are determined by solving an optimization problem that maximizes entropy production over time, and a 1D vertical profile constrained by an advection-dispersion-reaction model. We introduce a new approach for modeling phototrophy and explicitly represent oxygenic photoautotrophs, photoheterotrophs and anoxygenic photoautotrophs. The metabolic network also includes reactions for aerobic organoheterotrophic bacteria, sulfate reducing bacteria, sulfide oxidizing bacteria and aerobic and anaerobic grazers. Model results were compared to observations of biogeochemical constituents collected over a 24 hour period at 8 depths at a single 15 m deep station in Siders Pond. Maximizing entropy production over long (3 day) intervals produced results more similar to field observations than short (0.25 day) interval optimizations, which support the importance of temporal strategies for maximizing entropy production over time. Furthermore, we found that entropy production must be maximized locally instead of globally where energy potentials are degraded quickly by abiotic processes, such as light absorption by water. This combination of field observations and modeling results indicate that natural microbial systems can be modeled by using the maximum entropy production principle applied over time and space using many fewer parameters than conventional models.

Nearly 100 years ago, Alfred Lotka published two short but insightful papers describing how ecosystems may organize. Principally, Lotka argued that ecosystems will grow in size and that their cycles will spin faster via predation and nutrient recycling so as to capture all available energy, and that evolution and natural selection are the mechanisms by which this occurs and progresses. Lotka's ideas have often been associated with the maximum power principle, but they are more consistent with recent developments in nonequilibrium thermodynamics, which assert that complex systems will organize toward maximum entropy production (MEP). In this review, we explore Lotka's hypothesis within the context of the MEP principle, as well as how this principle can be used to improve marine biogeochemistry models. We need to develop the equivalent of a climate model, as opposed to a weather model, to understand marine biogeochemistry on longer timescales, and adoption of the MEP principle can help create such models.

Many attempts have been made to develop a general principle governing how systems develop and organize in ecology. We reviewed the historical developments that led to the conceptualization of several goal-oriented principles in ecosystem ecology. We focused on two prominent principles—the maximum power principle (MPP) and the maximum entropy production principle (MEPP)—and the literature that applies to both. Although these principles have conceptual overlap, we found considerable differences in their historical development, the disciplines that apply these principles, and their adoption in the literature. These principles were more similar than dissimilar, and the maximization of power in ecosystems occurs with maximum entropy production. These principles have great potential to explain how systems develop, organize, and function, but there are no widely agreed-on theoretical derivations for the MEPP and MPP, hindering their broader use in ecological research. We end with recommendations for how ecosystems-level studies may better use these principles.

‘‘Ecosystem’’ turned eighty this July. When Tansley (1935) coined the term, he did so largely as a rejection of the concept of emergence as espoused by Phillips (1935). The Phillips paper actually came out a month after Tansley’s, but Tansley was editor for the Journal of Ecology at the time and he published Phillip’s paper with a footnote on the first page that can only be viewed as a disclaimer. Tansley (1935) left no doubt about his feelings in the title of his response: ‘‘The Use and Abuse of Vegetation Concepts and Terms.’’ However, Tansley’s struggle with the concept of emergence precedes his clash with Phillips and is very clearly presented in a 1932 Lecture to Magdalen Philosophy Club of Oxford University, which was posthumously published by Ecosystems in 2002 (Tansley 2002). Emergence has since reemerged and has again entered the lexicon of ecologists. However, it is not always clear what is meant by an ‘‘emergent property’’; often it seems that the concept is invoked simply to express surprise at some apparently complex system-level property. Our aim here is not to present a comprehensive review and analysis of the various concepts of Emergence; the format is too short and there are others far better qualified (for example, Broad 1925; Birch and Cobb 1981; Blew 1996; Chalmers 2003; Reid 2007; Abel 2009). Our aim is to celebrate Tansley’s (1935, 2002) great contribution to Ecology, to reflect on the irony of the apparent reemergence of a concept rejected by Tansley (1935), and to draw distinctions among various concepts all called ‘‘emergence.’’ The problem, as we see it, is that there are at least three distinct concepts being labeled ‘‘emergence’’: The first of the three concepts is what Bedau (1997) calls ‘‘Strong Emergence.’’ Under this concept, objects at higher levels of organization acquire ‘‘emergent properties’’ that are not explainable in terms of the interactions among their constituent parts. There must therefore be some external causative agent acting toward the emergence of these properties (for example, Morgan 1923) or they arise spontaneously only in the context of the higher level object (Broad 1925). The ideas underlying strong emergence are what inspired Tansley (1935), in a rejection of the concept, to coin the term ‘‘Ecosystem.’’ We will use the term ‘‘Teleological Emergence’’ to describe this concept because it is more descriptive than ‘‘Strong Emergence.’’ The second of the concepts is what Bedau (1997) calls ‘‘Weak Emergence’’ (an unfortunate name for a strong concept). This concept is derived from the study of complex systems and is applied to chaotic phenomena like fractals and strange attractors that arise from nonlinear interactions in a system that are sensitive to initial conditions. We will use the more descriptive term ‘‘Chaotic Emergence’’ for this concept. Some aspects of teleological and chaotic emergence appear superficially similar; however, the underlying foundations differ significantly, and in particular, the attributions of causation explicit in the two concepts are diametrically opposed. The third of the concepts we will call ‘‘Mechanistic Emergence.’’ This concept, we believe, adheres more closely to the common usage of the Received 27 May 2015; accepted 27 May 2015; published online 6 June 2015 Author contributions The manuscript is derived from a longstanding philosophical debate between the authors. Both authors contributed to the writing. *Corresponding author; e-mail: erastett@mbl.edu Ecosystems (2015) 18: 735–739 DOI: 10.1007/s10021-015-9893-6

Childs River and Quashnet River, combined with the absence of such spikes at Sage Lot and Flat Ponds, suggests input by plumes from septic tanks. Mean NI&+ concentration in groundwater varied over a relatively small range across the three sites (Fig. 1, middle panels). NH4+ contributed less than 5% of the mean dissolved inorganic nitrogen (DIN) at Childs and Quashnet Rivers. Most of the inorganic nitrogen, therefore, enters these two estuaries in the form of N03-. Nitrogen from septic tanks and soils enters the aquifer in reduced forms, such as NH.,+ [untreated septic water contains up to 8000 PM NH4+ (J. McClelland, unpub. data)], which are oxidized to N09in the aquifer before reaching the estuary. Mean POd3+ concentrations were relatively low and similar in the three estuaries (Fig. 1, right panels). One would not expect POd3+ to vary across the sites, despite a gradient in residential area, because POd3+ is readily adsorbed by soils. Groundwater data from Waquoit Bay suggest that differences in land use can have a major effect on the concentrations of nitrogen entering a shallow estuary. To estimate the loading from groundwater to receiving estuaries and to estuarine systems, such as Waquoit Bay, concentration data such as those reported in this paper must be converted to flux by estimating actual water and nutrient transport. Calculated values of total system loading will be useful in deciding how to manage coastal estuaries threatened by eutrophication. This work was supported by Waquoit Bay Land Margin Ecosystems Research and a Research Experience for Undergraduates grant. Special thanks to Lori Saucy for operating the autoanalyzer.

We present an ecosystem model that describes the biogeochemistry of a sediment nitrate reducing microbial community. In the model, the microbial community is represented as a distributed metabolic network. Biogeochemical pathways are controlled through the synthesis and allocation of biological structure that serves to catalyze each process. Allocation is determined by way of a thermodynamically constrained optimization according to the principle of maximum entropy production (MEP). According to the MEP principle, ecosystems will organize so as to max- imize the dissipation of free energy based upon the available resources (nutrients and electron donors and acceptors). In the model, 3 nitrate reduction pathways, viz. heterotrophic denitrifica- tion, anammox, and dissimilatory nitrate reduction to ammonium, compete for electron acceptors (nitrate and nitrite). The model predicts switches in the dominant nitrate cycling pathways based upon the ratio of carbon to nitrate supply. An advantage of this approach over a traditional organ- ism-centric kinetic approach is that the Monod growth parameters, maximum uptake rate (νmax), and half saturation (kM) constants are determined during the optimization procedure as opposed to parameters specified a priori. Such a model is therefore useful for applications where these kinetic parameters are unknown and difficult to measure, such as the marine subsurface, or when ecosystems undergo large-scale environmental perturbations resulting in shifts in the dominant organisms.

Microbial fuel cells (MFCs) offer the potential for generating electricity, mitigating greenhouse gas emissions, and bioremediating pollutants through utilization of a plentiful renewable resource: soil organic carbon. We analyzed bacterial community structure, MFC performance, and soil characteristics in different microhabitats within MFCs constructed from agricultural or forest soils in order to determine how soil type and bacterial dynamics influence MFC performance. Our results indicated that MFCs constructed from agricultural soil had power output about 17 times that of forest soil-based MFCs and respiration rates about 10 times higher than forest soil MFCs. Agricultural soil MFCs had lower C:N ratios, polyphenol content, and acetate concentrations than forest soil MFCs. Bacterial community profile data indicate that the bacterial communities at the anode of the high power MFCs were less diverse than in low power MFCs and were dominated by Deltaproteobacteria, Geobacter, and to a lesser extent, Clostridia, while low-power MFC anode communities were dominated by Clostridia. These results suggest that the presence of organic carbon substrate (acetate) was not the major limiting factor in selecting for highly electrogenic bacterial communities, while the quality of available organic matter may have played a significant role in supporting high performing bacterial communities.

We used a long-term observation data set (12 years) of fish, insect and primary producer standing stocks in both reference and phosphate-fertilized reaches of the Kuparuk River located on the north slope of Alaska, USA to test a recently developed first-order approximation model. The model employs a flow analysis-type approach, but uses first-order approximations between annual mean compartment stocks and environmental drivers of temperature, discharge and solar radiation. Consequently, the model is more robust and requires fewer observations than standard process-oriented models, and can utilize observations that are difficult to incorporate into process models. Unlike standard inverse models, we show that our model is capable of prediction provided sufficient data are available for model calibration and environmental drivers are known. The results show that the inter-annual variations of several components in the Kuparuk River ecosystem, including dissolve inorganic phosphate, chironomids, black flies and Arctic grayling, can be accurately approximated as a linear function of temperature, discharge and solar radiation. In particular, the model indicates that changes in river habitat brought about by proliferation of the moss Hygrohypnum spp. in the P-fertilized reach caused a temporary shift in flow paths supporting Arctic grayling from primary producers to detrital-based pathways. However, after moss establishment, primary producer flow paths to Arctic grayling returned and detrital-based pathways weakened.

Marine aggregates were evaluated for their potential role in the ecology of aquatic pathogens using underwater video surveys coupled with direct collection of aggregates in modified settling cones. Six locations, two each in New York, Connecticut, and Massachusetts, were surveyed over 8 months to explore differences in the characteristics of aggregates found in habitats populated by clams (Mercenaria mercenaria) and oysters (Cras- sostrea virginica). Microaggregate ( 500 lm) concentrations, but peak concentrations of macroaggregates and microaggregates, mean size of particles, and volume fraction of aggregated material varied among the six shallow-water habitats. Concentrations (colony- forming units per ml) of total heterotrophic bacteria (THB) and total mesophilic pathogenic bacteria (MPB) from samples of aggregates were significantly different among the four locations bordering Long Island Sound (LIS). The highest concentrations and enrichment factors in aggregates were observed in August for THB and in June for MPB. Significant correlations were detected for salinity and the concentrations and enrichment factors of THB in aggregates and for the concentrations and percentages of MPB in seawater samples. Significant correlations were also detected for temperature and the concentrations of MPB in aggregates and the enrichment factors for THB and MPB (marginal significance). Bacterial species identified in association with aggregates included: Vibrio cholerea, V. parahaemolyticus, V. vulnificus, V. alginolyticus, Aeromonas hydrophila, Pseudomonas aeruginosa, Escherichia coli, and Mycobacteria sp. These results have important implications for the way in which aquatic pathogens are collected, quantified, and monitored for risk-based surveillance in shallow-water ecosystems.

Salt marsh ecosystems have been considered not susceptible to nitrogen overloading because early studies suggested that salt marshes adsorbed excess nutrients in plant growth. However, the possible effect of nutrient loading on species composition, and the combined effects of nutrients and altered species composition on structure and function, was largely ignored. Failure to understand interactions between nutrient loading and species composition may lead to severe underestimates of the impacts of stresses. We altered whole salt marsh ecosystems (;60 000 m 2 /treatment) by addition of nutrients in flooding waters and by reduction of a key predatory fish, the mummichog. We added nutrients (N and P; 15-fold increase over ambient conditions) directly to the flooding tide to mimic the way anthropogenic nutrients are delivered to marsh ecosystems. Despite the high concentrations (70 mmol N/L) achieved in the water column, our annual N loadings (15-60 g Nm � 2 � yr � 1 ) were an order of magnitude less than most plot-level fertilization experiments, yet we detected responses at several trophic levels. Preliminary calculations suggest that 30-40% of the added N was removed by the marsh during each tidal cycle. Creek bank Spartina alterniflora and high marsh S. patens production increased, but not stunted high marsh S. alterniflora. Microbial production increased in the fertilized creek bank S. alterniflora habitat where benthic microalgae also increased. We found top-down control of benthic microalgae by killifish, but only under nutrient addition and in the opposite direction (increase) than that predicted by a fish-invertebrate-microalgae trophic cascade. Surprisingly, infauna declined in abundance during the first season of fertilization and with fish removal. Our results demonstrate ecological effects of both nutrient addition and mummichog reduction at the whole-system level, including evidence for synergistic interactions.

Ecosystems exhibit nonlinear dynamics that are often difficult to capture in models. Consequently, linearization is commonly applied to remove some of the uncertainties associated with the nonlinear terms. However, since the true model is unknown and the operating point to linearize the model about is uncertain, developing linear ecosystems models is non-trivial. To develop a linear ecosystem model, we assume that the annual mean state of an ecosystem is a minor bias from the long-term mean state. A first order approximation inverse model to govern the year-to-year dynamics of ecosystems whose characteristic time scales are less than 1 year is developed, through theoretically formulation, on the basis of steady state analysis, time scale separation and nondimensionalization. The approach is adept at predicting year-to-year variations and to tracking system response to changes in environmental drivers when compared to data generated with a standard nonlinear NPZD model.

We describe a nonlinear inverse technique to estimate gross primary production (GPP), community respiration (CR), and net ecosystem production in the vertically well mixed Parker River, Plum Island Sound estuarine system located in northeastern Massachusetts, USA. The approach uses a calibrated 1D advection–dispersion model to predict oxygen and salt concentrations along the estuarine length, which are compared to oxygen concentrations measured during high speed transects near dawn and dusk over a 2-day period in June 1995. The spatiotemporal shapes of the GPP and CR surfaces are represented with cubic B-splines that are deformed as needed by an algorithm that seeks to minimize error between oxygen observations and model predictions. The spatially resolving solution shows maximum instantaneous GPP of 1030 mmol O2 m −3 d −1 and CR of 340 mmol O2 m −3 d −1 in the upper portions of the estuary; however, the overall estuary is net heterotrophic ( −4.8 Mg C d −1 ). Analysis of the advection–dispersion model reveals that gas invasion and longitudinal dispersion account for up to 21 and 14% of local O2 accumulation relative to CR, respectively. However, an oxygen balance averaged over the 2-day sampling period shows that the negative net ecosystem productivity is balanced by a loss in estuarine oxygen storage. Carbon budget analysis indicates that terrestrial, marine, and marsh allochthonous inputs account for 2, 27 and 71% of the observed net heterotrophy, respectively. © 2005 Elsevier B.V. All rights reserved.

The relationships of land use/land cover (LULC) on major solute concentrations in stream water were investigated for the Ipswich River basin (404 km2) in northeastern Massachusetts. Stream water was sampled seven times during base flow in 43 first-order catchments and four times in 28 second- and third-order catchments. Regression analysis of the first-order catchment data indicates that NO3−, acid neutralizing capacity (ANC), Cl−, SO42−, and the base cations had positive, mostly exponential relationships with the increasing extent of urban + agricultural area (P < 0.05), whereas dissolved organic nitrogen (DON) and dissolved organic carbon (DOC) had positive, exponential relationships with the increasing extent of wetland + open water (P < 0.05). Solute sources responsible for many of these relationships are human-derived constituents found in septic effluent, fertilizers, and road salts. In contrast to more conservative solutes, concentrations of NO3− in the first-order streams were commonly higher than in those of second- and third-order streams with similar proportions of urban + agricultural land use. Using LULC subclasses (e.g., high density residential), as well as the proportions of LULC in 50, 100, and 200 m concentric zones bordering streams, generally decreased the relationships (r2) determined above. Hence, the disturbed area of the entire subbasin was the best descriptor of streamwater solute concentrations. Potassium concentrations in stream water had stronger relationships than any other ion, yet these explained < 60% of the variability, indicating that there are a number of important ancillary factors that affect streamwater solute composition in the Ipswich River basin.

Oceanic dissolved organic carbon (DOC) constitutes one of the largest pools of reduced carbon in the biosphere. Estimated DOC export from the surface ocean represents 20% of total organic carbon flux to the deep ocean1,2,3, which constitutes a primary control on atmospheric carbon dioxide levels4. DOC is the carbon component of dissolved organic matter (DOM) and an accurate quantification of DOM pools, fluxes and their controls is therefore critical to understanding oceanic carbon cycling. DOC export is directly coupled with dissolved organic nitrogen and phosphorus export. However, the C:N:P stoichiometry (by atoms) of DOM dynamics is poorly understood. Here we study the stoichiometry of the DOM pool and of DOM decomposition in continental shelf, continental slope and central ocean gyre environments. We find that DOM is remineralized and produced with a C:N:P stoichiometry of 199:20:1 that is substantially lower than for bulk pools (typically >775:54:1), but greater than for particulate organic matter (106:16:1—the Redfield ratio). Thus for a given mass of new N and P introduced into surface water, more DOC can be exported than would occur at the Redfield ratio. This may contribute to the excess respiration estimated to occur in the interior ocean5. Our results place an explicit constraint on global carbon export and elemental balance via advective pathways.

Increased riverine nitrogen (N) fluxes have been strongly correlated with land use changes and are now one of the largest pollution problems in the coastal region of the United States. In the present study, the Hydrological Simulation Program-FORTRAN (HSPF) is used to simulate transport of N in the Ipswich River basin in Massachusetts and to evaluate the effect of future land use scenarios on the water quality of the river. Model results show that under a land use change scenario constructed with restrictions from environmental protection laws, where 44 percent of the forest in the basin was converted to urban land, stream nitrate concentrations increased by about 30 percent of the present values. When an extreme land use scenario was used, and 100 percent of the forest was converted to urban land, concentrations doubled in comparison to present values. Model simulations also showed that present stream nitrate concentrations might be four times greater than they were prior to urbanization. While pervious lands with high density residential land use generated runoff with the highest N concentrations in HSPF simulations, the results suggested that denitrification in the riparian zone and wetlands coupled with the hydrology of the basin are likely to control the magnitude of nitrate loads to the aquatic system. The simulation results showed that HSPF can predict the general patterns of inorganic N concentrations in the Ipswich River and tributaries. Nevertheless, HSPF has some difficulty simulating the extreme variability of the observed data throughout the main stem and tributaries, probably because of limitations in the representation of wetlands and riparian zones in the model, where N processes such as denitrification seem to play a major role in controlling the transport of N from the terrestrial system to the river reaches.

Isotopically labeled nitrate ( 15 NO ) was added continuously to the Rowley estuary, Massachusetts, for 22 d to 2 3 assess the transport, uptake, and cycling of terrestrially derived nitrogen during a period of high river discharge and low phytoplankton activity. Isotopic enrichment of the 3.5-km tidal prism (150,000 m 3 ) was achieved for the 3 weeks and allowed us to construct a nitrogen mass balance model for the upper estuary. Mean d 15 NO in the 2 3 estuary ranged from 300‰ to 600‰, and approximately 75%‐80% of the 15 N was exported conservatively as 15 NO to the coastal ocean. Essentially all of the 20%‐25% of the 15 N processed in the estuary occurred in the 2 3 benthos and was evenly split between direct denitrification and autotrophic assimilation. The lack of water-column 15 N uptake was attributed to low phytoplankton stocks and short water residence times (1.2‐1.4 d). Uptake of watercolumn NO by benthic autotrophs (enriched in excess of 100‰) was a function of NO concentration and satisfied 2 2 3 3 up to 15% and 25% of the total nitrogen demand for benthic microalgae and macroalgae, respectively. Approximately 10% of tracer assimilated by benthic autotrophs was mineralized and released back to the water column as 15 NH . By the end of the study, 15 N storage in sediments and marsh macrophytes accounted for 50%‐70% of the 1 4 15N assimilated in the estuary. These compartments may sequester watershed-derived nitrogen in the estuary for time scales of months to years.

Decomposition of dissolved organic carbon, nitrogen and phosphorus (DOC, DON, DOP) was measured for surface and bottom waters of the middle Atlantic bight (MAB) and deep slope water adjacent to the MAB on two occasions in March and August 1996. We used standard bottle incubation techniques to measure the decrease in dissolved organic matter (DOM) concentrations over a 180-day interval. Generally DOM concentrations in the MAB were elevated (125mM DOC, 10.2mM DON and 0.30mM DOP) relative to the surface ocean and deep slope water (46.7mM DOC, 2.76mM DON, 0.03mM DOP). On average the C:N:P ratio of shelf DOM (431:36:1) was substantially higher than the Redfield ratio, but not nearly as high for that of deep slope water (2700:215:1). Decomposition time course data were fit to a three-pool (very labile, labile, and recalcitrant pools) multi-G model using a Marquardt fitting routine. The threepool model was superior to a simple exponential decay model assuming a single pool of DOM. We observed no significant changes in concentration of DOM in deep-water samples, attesting to the old age of this material, its recalcitrant nature, and the cleanliness of our technique for measuring decomposition. There were major differences in the relative amount of very labile, labile and recalcitrant fractions of shelf-water DOC, DON and DOP as a result of preferential remineralization of P over N and N over C. Averaged over stations, the decomposable portion of the bulk DOC, DON and DOP pools increased from 30% to 40% to 81% for C, N and P. There was a wide range in decay coefficients for the very labile and labile DOM pools: average decay coefficient for the very labile pool was 0.219 d � 1 , and 0.018 d � 1 for the labile pool. Average half-lives calculated from the decay coefficients were 4, 12 and 8 days for the very labile DOC, DON and DOP pools, and 54, 113 and 90 days for the labile DOC, DON and DOP pools. On the basis of pool turnover times relative to shelf-water residence time (B100 days) we conclude that autochthonous algal production is the source of the very labile DOM pools. Its rate of production is sufficient to sustain estimated rates of bacteria C demand in continental margins. Our results for the MAB indicate that while substantial amounts of DOM are remineralized in the same time frame as shelf-water residence time, there is substantial DOM remaining that is depleted in N and P relative to C. Strong concentration gradients in DOM occur between shelf and ocean waters and between surface and deeper waters. Coupled with appropriate vertical and horizontal advective and eddy diffusive transports, DOM export from the MAB and other shelf systems may be a significant component of ocean C dynamics. r 2002 Elsevier Science Ltd. All rights reserved.

Our inability to accurately model marine food webs severely limits the prognostic capabilities of current generation marine biogeochemistry models. To address this problem we examine the use of data assimilation and mesocosm experiments to facilitate the development of food web models. The components of the data assimilation demonstrated include the construction of measurement models, the adjoint technique to obtain gradient information on the objective function, the use of parameter constraints, incorporation of discrete measurements and assessing parameter observability. We also examine the effectiveness of classic and contemporary optimization routines used in data assimilation. A standard compartment-type food web model is employed with an emphasis on organic matter production and consumption. Mesocosm experiments designed to examine the interaction of inorganic nitrogen with organic matter provide the data used to constrain the model.Although we are able to obtain reasonable e ts between the mesocosm data and food web model, the model lacks the robustness to be applicable across trophic gradients, such as those occurring in coastal environments. The robustness problem is due to inherent structural problems that render the model extremely sensitive to parameter values. Furthermore, parameters governing actual ecosystems are not constants, but rather vary as a function of environmental conditions and species abundance, which increases the sensitivity problem. We conclude by briee y discussing possible improvements in food web models and the need for rigorous comparisons between models and data (a modeling workbench) so that performance of competing models can be assessed. Such a workbench should facilitate systematic improvements in prognostic marine food web models.

Salinity distributions, dye release studies, and parameter estimation techniques are used to determine the longitudinal (or tidal) dispersion coefficient in the Plum Island Sound estuary in north-eastern Massachusetts, U.S.A. A one-dimensional, intertidal, advection–dispersion model is used to test the validity of the steady-state assumption employed to estimate tidal dispersion from salinity. It is found that salinity distributions in the upper Parker River can take months to relax to steady state under low freshwater discharge conditions, so that dispersion cannot be reliably estimated using the steady-state assumption. To overcome this problem, dye release studies and parameter estimation techniques are used to determine the functionality of the dispersion coefficient that produces the best fit between the observed salinity profiles, dye plumes and model output under transient conditions for the period between 28 April 1992 and 12 April 1995. The estimated dispersion coefficient is found to be a hyperbolic function of distance from the head of the estuary and varies from an average of 3·6 m 2 s −1 in the upper Parker River (0–5·2 km) to an average of 670 m 2 s −1 in Plum Island Sound (14·3–24 km). The advection–dispersion model is also used to investigate three characteristic mixing-time scales involving average age, average residence time and average transit time (also called turnover or flushing time) for freshwater and saltwater sources in defined subsections of the estuary. It is found that these time scales can vary from a few hours to 3 months depending on location and freshwater discharge. Transport in the upper Parker River can be either advection or dispersion dominated depending on discharge, while Plum Island Sound is always dominated by dispersive terms. Tidal channel morphometry leads to a very dynamic system in the upper Parker River that significantly impacts ecological processes in the region. We discuss the relevance and appropriate application of the three characteristic mixing-time scales to ecological processes.

Dissolved and particulate organic matter (DOM and POM) collected from rivers or groundwater feeding five estuaries along the east and west coasts of the USA were characterized with a variety of biogeochemical techniques and related to bioavailability to estuarine microbes. Surface water was sampled from the Columbia, Satilla, Susquehanna and Parker Rivers and groundwater was sampled from the Childs River. Several geochemical descriptors (percent organic matter of suspended particulate matter, C/N, lignin phenol content, ratio of vanillic acid to vanillin) suggested an ordering of the systems with respect to POM lability: Satilla 2000 pM for the Satilla River. Elemental analysis of DOM concentrates (>I000 D) was used to predict organic matter composition and to calculate degree of substrate reduction using two different modeling approaches. Models predicted aliphatic carbon ranging between 43 and 60% and aromatic carbon between 26 and 36%, with aliphatic content lowest in the Satilla and highest in the Columbia River. The degree of substrate reduction of the organic matter concentrates followed a pattern similar to that for aliphatic C, being lowest in the Satilla (3.5) and highest in the Columbia (4.0). Extracellular enzyme activity varied broadly across the systems, but again ordered sites in the same way as did aliphatic content and degree of substrate reduction. Bacterial growth rates ranged from 1.3 ug mg-' d-' DOC in the Satilla to 1.7 ug mg-' d-' DOC in the Parker River. Bioassays confirmed patterns of dissolved organic matter lability predicted by the chemical models. Between 67% to 75% of the variation in bacterial growth could be explained by differences in organic matter composition.

The maximum entropy production (MaxEP) principle has been applied to steady state systems, but biogeochemical problems of interest are typically transient in nature. To apply MaxEP to biogeochemical reaction networks, we propose that living systems maximum entropy production over appropriate time horizons based on strategic information stored in their genomes, which differentiates them from inanimate chemistry, such as fire, that maximizes entropy production instantaneously. We develop a receding horizon optimal control procedure that maximizes internal entropy production over different intervals of time. This procedure involves optimizing the stoichiometry of a reaction network to determine how biological structure is partitioned to reactions over an interval of time. The modeling work is compared to a methanotrophic microcosm experiment that is being conducted to examine how microbial systems integrate entropy production over time when subject to time varying energy input attained by periodically cycling feed-gas composition. The MaxEP-based model agrees well with experimental results, and model analysis shows that increasing the optimization time horizon increases internal entropy production.

Simulation modeling has been an integral, albeit ad hoc, component of the field of aquatic microbial ecology for the past two decades. One of the most critical steps in simulation modeling is the initial formulation of a clear set of questions and goals. It is doubtful that a single generic model could be constructed to address adequately all questions of interest concerning the microbial loop because of the tremendous range in time scales that define these questions. Progress in the field of aquatic microbial ecology will benefit from an integrated research program including experimental and modeling approaches. A submodel of bacterial utilization of various qualities of organic matter that we have under construction is presented. This submodel will be a component of a larger model to evaluate the effects of quality and quantity of organic matter and inorganic nutrient inputs on estuarine food web structure and efficiency. The overall model will be general enough in its structure that it should be applicable to a wide range of questions concerning the microbial loop, with time scales ranging from hours to days.

A bioenergetic model has been developed to examine growth kinetics associated with bacterial utilization of dissolved organic matter (DOM), NH4+, and N03-. A set of 11 metabolic reactions are used to govern the incorporation, oxidation, and N remineralization of DOM and dissolved inorganic N associated with bacterial growth. For each reaction, free energies and electron transfer requirements are calculated based on the C, H, 0, and N composition of the substrates and their concentration in the environment. From these reactions, an optimization problem is constructed in which bacterial growth rate is maximized subject to constraints on energetics, electron balances, substrate uptake kinetics, and bacterial C : N ratio. The optimization approach provides more information on bacterial growth kinetics than do the Monod-type models that are typically used to describe bacterial growth. Simulations are run to examine bacterial C yield and growth rate, N remineralization or immobilization, and substrate preferences as resource concentrations and compositions are varied. Results from the model agree well with observations in the literature, which indicate that the premise of the model, that bacteria allocate resources to maximize growth rate, may be an accurate overall description of bacterial growth. Simulations indicate that bacterial growth rate and yield are strongly correlated to the oxidation state of the labile DOM, as determined from its bulk elemental composition. Furthermore, the model demonstrates that bacterial growth cannot always be explained by a single constraint (such as the C : N ratio of substrate), since several constraints are often active simultaneously and continuously change with environmental conditions. Dissolved organic matter (DOM) represents an important component of aquatic ecosystems in terms of both the size of the compartment relative to particulate organic matter (POM) and its impact on the functioning of microbial food webs. Consequently, many descriptive and experimental studies have been conducted to characterize the composition of DOM, examine its production and utilization, understand its impact on microbial dynamics, and determine the extent to which it supports higher trophic levels. The results of these studies have been incorporated to various extents in both simple bacterial growth models and more complete food web models that examine DOM utilization. These models are based almost

In order to achieve the theoretical maximum yield of lysine from glucose (75% molar), flux partitioning at principal branch points (glucose 6-phosphate, phosphoenolpyruvate, and pyruvate) in the primary metabolic network of Corynebacterium glutamicum must be significantly altered from that observed during balanced growth, so that lysine precursors are synthesized in optimal stoichiometric ratios. In this article, we employ flux analysis, which provides a snapshot of metabolic flux distributions from extracellular measurements, in conjunction with two metabolic perturbations of the pyruvate dehydrogenase complex (PDC) to examine the possibility that flux partitioning at the pyruvate (Pyr) branch point (i. e., pyruvate availability) limits lysine yield. The two perturbation experiments involved (1) isolation and flux analysis of a PDC-attenuated mutant of C. glutamicum and (2) flux analysis following the addition of fluoropyruvate (FP) to a standard fermentation during the initial period of lysine overproduction. No significant alteration in flux partition was observed at any of the three principal nodes in the PDC-attenuated strain. However, the PDC mutation did cause a 70% uniform decrease in carbon flow throughout the network, which indicates that the lysine yield is not limited solely by a weakly rigid Pyr branch point. The addition of FP did not affect the lysine synthesis rate, but did temporarily redirect carbon flow away from the TCA cycle toward pyruvate excretion, which further confirms that the lysine yield is not pyruvate-limited and that the Pyr branch point is not weakly rigid.

Analyses indicate that the lysine yield in Corynebacterium glutamicum is limited by suboptimal flux partitioning at either the glucose 6-phosphate (Glc6P), phosphoenolpyruvate (PEP), or pyruvate (Pyr) branch points (or a combination thereof), which results in disproportionate production of the required lysine precursors. Suboptimal flux partitioning at a metabolic branch point can result from the inadequate enzyme activity of a subordinate branch (a weakly rigid branch point) or from the active feedback regulation of the flux partitioning (strongly rigid branch point). In order to examine branch-point characteristics, we have utilized metabolite balances, constrained by biochemistry, to estimate flux distributions in the primary metabolic network of C. glutamicum from measured, extracellular, metabolite accumulation rates. These flux distributions, in combination with local metabolic perturbations, were used to infer branch-point characteristics. This study examines the flexibility of the Glc6P branch point, which could limit lysine yield via inadequate NADPH production, by perturbations induced from attenuation of Glc6P isomerase activity (first branch-point enzyme of glycolysis) and by fermentation on gluconate, which effectively bypasses the Glc6P branch point. Results from the analyses of these perturbations, as well as the flux distributions at the Glc6P branch point during the different phases of a control lysine fermentation, indicate that the Glc6P branch point is neither weakly nor strongly rigid, so that limitations in lysine yield must result from rigidity at either the PEP or Pyr branch point.

Background The increasing availability of time series microbial community data from metagenomics and other molecular biological studies has enabled the analysis of large-scale microbial co-occurrence and association networks. Among the many analytical techniques available, the Local Similarity Analysis (LSA) method is unique in that it captures local and potentially time-delayed co-occurrence and association patterns in time series data that cannot otherwise be identified by ordinary correlation analysis. However LSA, as originally developed, does not consider time series data with replicates, which hinders the full exploitation of available information. With replicates, it is possible to understand the variability of local similarity (LS) score and to obtain its confidence interval. Results We extended our LSA technique to time series data with replicates and termed it extended LSA, or eLSA. Simulations showed the capability of eLSA to capture subinterval and time-delayed associations. We implemented the eLSA technique into an easy-to-use analytic software package. The software pipeline integrates data normalization, statistical correlation calculation, statistical significance evaluation, and association network construction steps. We applied the eLSA technique to microbial community and gene expression datasets, where unique time-dependent associations were identified. Conclusions The extended LSA analysis technique was demonstrated to reveal statistically significant local and potentially time-delayed association patterns in replicated time series data beyond that of ordinary correlation analysis. These statistically significant associations can provide insights to the real dynamics of biological systems. The newly designed eLSA software efficiently streamlines the analysis and is freely available from the eLSA homepage, which can be accessed at http://meta.usc.edu/softs/lsa.

The incorporation of 32P-phosphorus into marine bacteriophage nucleic acid was studied in culture experiments to investigate the source of nucleotides used by the phage. We consistently found that the 32P-specific activity in the phage genome increased during the 11 h incubation and was low relative to the specific activity in the medium, averaging 21% (±SD 5.9) for the three phage isolates. This was in accordance with a mathematical model where most of the nucleotides for phage DNA synthesis were derived from the host cell nucleic acid rather than de novo synthesis. We propose that this metabolic strategy may be common among marine phages, as an adaptation to a nutrient poor environment. Consequently, the contribution of free DNA to the dissolved fraction through phage lysis of bacteria, may be less that previously thought. Also during radiolabelling of bacteriophages in natural water samples, isotope dilution may be dependent on the specific growth rate of the bacterial host.

The two main contributions of this are the solidification of Corynebacterium glutamicum biochemistry guided by bioreaction network analysis, and the determination of bansal metabolic flux distributions during growth and lysine synthesis. Employed methodology makes use of stoichiometrically based mass balances to determine flux distributions in the C. glutamicum metabolic network. Presented are a brief description of the methodology, a through literature review of glutamic acid bacteria biochemistry, and specific results obtained through a combination of fermentation studies and analysis-directed intracellular assays. The latter include the findings of the lack of activity of glyoxylate shunt, and that phosphoenolpyruvate carboxylase (PPC) is the only anaplerotic reaction expressed in C. glutamicum cultivated on glucose minimal media. Network simplifications afforded by the above findings facilitated the determination of metabolic flux distributions under a variety of culture conditions and led to the following conclusions. Both the pentose phosphate pathway and PPC support fluxes during growth and lysine overproduction branch point does not appear to limit lysine synthesis.

Received 27 December 2009; revised 2 June 2010; accepted 17 June 2010; published 16 October 2010. [1] Wetland‐estuarine‐shelf interaction processes in the Plum Island Sound and Merrimack River system in the Massachusetts coast are examined using the high‐resolution unstructured grid, finite volume, primitive equations, coastal ocean model. The computational domain covers the estuarine and entire intertidal area with a horizontal resolution of 10–200 m. Driven by five tidal constituents forcing at the open boundary on the inner shelf of the eastern coast of the Gulf of Maine, the model has successfully simulated the 3‐D flooding/drying process, temporal variability, and spatial distribution of salinity as well as the water exchange flux through the water passage between the Plum Island Sound and Merrimack River. The model predicts a complex recirculation loop around the Merrimack River, shelf, and Plum Island Sound. During the ebb tide, salt water in the Plum Island Sound is injected into the Merrimack River, while during flood tide, a significant amount of the freshwater in the Merrimack River is forced into Plum Island Sound. This water exchange varies with the magnitude of freshwater discharge and wind conditions, with a maximum contribution of ∼30%–40% variability in salinity over tidal cycles in the mouth of the Merrimack River. Nonlinear tidal rectification results in a complex clockwise residual recirculation loop around the Merrimack River, shelf, and Plum Island Sound. The net water flux from Plum Island Sound to the Merrimack River varies with the interaction between tide, river discharge, and wind forcing. This interaction, in turn, affects the salt transport from this system to the shelf. Since the resulting water transport into the shelf significantly varies with the variability of the wind, models that fail to resolve this complex estuarine and shelf system could either overestimate or underestimate the salt content over the shelf.

[1] We assessed the effects of historical (1931–1998) changes in both land use and climate on the water budget of a rapidly urbanizing watershed, Ipswich River basin (IRB), in northeastern Massachusetts. Water diversions and extremely low flow during summer are major issues in the IRB. Our study centers on a detailed analysis of diversions and a combined empirical/modeling treatment of evapotranspiration (ET) response to changes in climate and land use. A detailed accounting of diversions showed that net diversions increased due to increases in water withdrawals (primarily groundwater pumping) and export of sewage. Net diversions constitute a major component of runoff (20% of streamflow). Using a combination of empirical analysis and physically based modeling, we related an increase in precipitation (2.7 mm/yr) and changes in other climate variables to an increase in ET (1.7 mm/yr). Simulations with a physically based water-balance model showed that the increase in ET could be attributed entirely to a change in climate, while the effect of land use change was negligible. The land use change effect was different from ET and runoff trends commonly associated with urbanization. We generalized these and other findings to predict future streamflow using climate change scenarios. Our study could serve as a framework for studying suburban watersheds, being the first study of a suburban watershed that addresses long-term effects of changes in both land use and climate, and accounts for diversions and other unique aspects of suburban hydrology.

[1] We calculated N budgets and conducted nutrient uptake experiments to evaluate the fate of N in the aquatic environment of the Ipswich River basin, northeastern Massachusetts. A mass balance indicates that the basin retains about 50% of gross N inputs, mostly in terrestrial components of the landscape, and the loss and retention of total nitrogen (TN) in the aquatic environment was about 9% of stream loading. Uptake lengths of PO4 and NH4 were measurable in headwater streams, but NO3 uptake was below detection (minimum detection limit = 0.05 μM). Retention or loss of NO3 was observed in a main stem reach bordered by wetland habitat. Nitrate removal in urban headwater tributaries was because of water withdrawals and denitrification during hypoxic events and in ponded wetlands with long water residence times. A mass balance using an entire river network indicates that basin-wide losses due to aquatic denitrification are considerably lower than estimates from several recent studies and range from 4 to 16% of TDN in stream loading. Withdrawals for domestic use restrict the runoff of headwater catchments from reaching the main stem during low base flow periods, thereby contributing to the spatial and temporal regulation of N export from headwater tributaries.

Biogeochemistry is the study of how living systems in combination with abiotic reactions process and cycle mass and energy on local, regional, and global scales (Schlesinger, 1997). Understanding how these biogeochemical cycles function and respond to perturbations has become increasingly important, as anthropogenic impacts have significantly altered many of these cycles (Galloway and Cowling, 2002; Houghton et al., 2002). Biogeochemistry is strongly governed by microbial processes, and it appears to closely follow thermodynamic constraints in that electron acceptor (O(2), NO(3)(-), SO(4)(2-), etc.) utilization closely follows a priori expectations based on energetics (Vallino et al., 1996; Hoehler et al., 1998; Jakobsen and Postma, 1999; Amend and Shock, 2001). Consortiums of microorganisms seem to have evolved to exploit chemical potentials wherever they exist in the environment, as manifested by the recent discovery of anaerobic methane oxidation by sulfate (Boetius et al., 2000) or sulfide oxidation by nitrate (Schulz et al., 1999). Three and a half billion years of natural selection have produced living systems capable of degrading most chemical potentials. We may therefore ask: If all ecosystem niche space is filled, is the biogeochemistry we observe in the environment dependent on the organisms that occupy that environment, or is the biogeochemistry determined by fundamental forces, with the evolution of living systems being the implementation of those forces? Recent developments in nonequilibrium thermodynamics (NET) are beginning to support the latter alternative, and advances in genomics are allowing us to explore microbial consortiums in detail. Taking advantage of ideas being suggested by NET, we have developed a modeling framework that views microbial consortiums as an inter-species distributed metabolic network. When combined with experimental observations, the model should help us test hypotheses that govern how living systems function.

(Valiela et al., unpub. data) and possess different amounts of surrounding salt marsh. Ten sites at each estuary were sampled with an Eckman dredge (0.15 m*) at the beginning of each month in all estuaries and at intervening 2-week intervals in Jehu and Hamblin Ponds. Samples were sorted by species, dried, and weighed. Nitrogen loading and salt marsh area showed no relationship (P for regression >0.05). We therefore regressed nutrient loading versus biomass, and salt marsh area versus biomass. Eelgrass biomass was inversely related to nitrogen loading (Fig. 1, top left; P = 0.000 I), in contrast to macroalgal biomass, which increased with nutrient loading (top right; P = 0.0041). The shapes of the best-fit curves suggest that eelgrass is particularly sensitive to even low rates of additional nutrient loading (Fig. 1, top left), but macroalgae respond linearly over a broader range of nutrient loadings. The correlations (Fig. 1, top), however, have considerable variation; for instance, Jehu Pond has substantially more eelgrass and less algae than would be expected based on nutrient loadings alone. Part of this variability may be associated with the amount of salt marsh in the estuaries. As salt marsh area increases, eelgrass biomass increases (Fig. 1, bottom left; P = 0.000 I) and macroalgal biomass decreases (Fig. 1, bottom right; P = 0.0011). Correlation coefficients for best-fit lines improved for both eelgrass and macroalgal biomass when plotted against salt marsh area (compare r* values in top and bottom panels, Fig. I). This result could be caused by lowered nitrogen loading owing to denitrification in salt marshes. Seagrass growth is inhibited under increased nitrogen loading and macroalgal growth is enhanced. Eelgrass apparently undergoes rapid, near-exponential reduction in biomass over a narrow range of nutrient loadings, whereas macroalgae respond in a linear fashion over a broader range. The relationship between nutrient loading to coastal watersheds and estuarine macrophytes may, however, be strongly influenced by salt marsh fringing the estuaries. Fringing salt marsh might act as a buffer against nutrient loading, and thus might allow eelgrass beds to be maintained even when they are surrounded by relatively urbanized watersheds. This work was supported by REU-NSF (OCE 9300490), NOAA (NA 170R2 110 1), and The Waquoit Bay Fellowship awarded to J. McClelland.