Fredric Lipschultz

Dinitrogen fixation in the world's oceans

The surface water of themarine environment has traditionally beenviewed as a nitrogen (N) limited habitat, andthis has guided the development of conceptualbiogeochemical models focusing largely on thereservoir of nitrate as the critical source ofN to sustain primary productivity. However,selected groups of Bacteria, includingcyanobacteria, and Archaea canutilize dinitrogen (N2) as an alternativeN source. In the marine environment, thesemicroorganisms can have profound effects on netcommunity production processes and can impactthe coupling of C-N-P cycles as well as the netoceanic sequestration of atmospheric carbondioxide. As one component of an integrated ‘Nitrogen Transport and Transformations’ project, we have begun to re-assess ourunderstanding of (1) the biotic sources andrates of N2 fixation in the worldsoceans, (2) the major controls on rates ofoceanic N2 fixation, (3) the significanceof this N2 fixation for the global carboncycle and (4) the role of human activities inthe alteration of oceanic N2 fixation. Preliminary results indicate that rates ofN2 fixation, especially in subtropical andtropical open ocean habitats, have a major rolein the global marine N budget. Iron (Fe)bioavailability appears to be an importantcontrol and is, therefore, critical inextrapolation to global rates of N2fixation. Anthropogenic perturbations mayalter N2 fixation in coastal environmentsthrough habitat destruction and eutrophication,and open ocean N2 fixation may be enhancedby warming and increased stratification of theupper water column. Global anthropogenic andclimatic changes may also affect N2fixation rates, for example by altering dustinputs (i.e. Fe) or by expansion ofsubtropical boundaries. Some recent estimatesof global ocean N2 fixation are in therange of 100–200 Tg N (1–2 × 1014 g N)yr−1, but have large uncertainties. Theseestimates are nearly an order of magnitudegreater than historical, pre-1980 estimates,but approach modern estimates of oceanicdenitrification.

Inputs, losses and transformations of nitrogen and phosphorus in the pelagic North Atlantic Ocean

The North Atlantic Ocean receives the largest allochthonous supplies of nitrogen of any ocean basin because of the close proximity of industrialized nations. In this paper, we describe the major standing stocks, fluxes and transformations of nitrogen (N) and phosphorus (P) in the pelagic regions of the North Atlantic, as one part of a larger effort to understand the entire N and P budgets in the North Atlantic Ocean, its watersheds and overlying atmosphere. The primary focus is on nitrogen, however, we consider both nitrogen and phosphorus because of the close inter-relationship between the N and P cycles in the ocean. The oceanic standing stocks of N and P are orders of magnitude larger than the annual amount transported off continents or deposited from the atmosphere. Atmospheric deposition can have an impact on oceanic nitrogen cycling at locations near the coasts where atmospheric sources are large, or in the centers of the highly stratified gyres where little nitrate is supplied to the surface by vertical mixing of the ocean. All of the reactive nitrogen transported to the coasts in rivers is denitrified or buried in the estuaries or on the continental shelves and an oceanic source of nitrate of 0.7–0.95 × 1012 moles NO3−1 y−1 is required to supply the remainder of the shelf denitrification (Nixon et al., this volume). The horizontal fluxes of nitrate caused by the ocean circulation are both large and uncertain. Even the sign of the transport across the equator is uncertain and this precludes a conclusion on whether the North Atlantic Ocean as a whole is a net source or sink of nitrate. We identify a source of nitrate of 3.7–6.4 × 1012 moles NO3− y−1 within the main thermocline of the Sargasso Sea that we infer is caused by nitrogen fixation. This nitrate source may explain the nitrate divergence observed by Rintoul & Wunsch (1991) in the mid-latitude gyre. The magnitude of nitrogen fixation inferred from this nitrate source would exceed previous estimates of global nitrogen fixation. Nitrogen fixation requires substantial quantities of iron as a micro-nutrient and the calculated iron requirement is comparable to the rates supplied by the deposition of iron associated with Saharan dust. Interannual variability in dust inputs is large and could cause comparable signals in the nitrogen fixation rate. The balance of the fluxes across the basin boundaries suggest that the total stocks of nitrate and phosphate in the North Atlantic may be increasing on time-scales of centuries. Some of the imbalance is related to the inferred nitrogen fixation in the gyre and the atmospheric deposition of nitrogen, both of which may be influenced by human activities. However, the fluxes of dissolved organic nutrients are almost completely unknown and they have the potential to alter our perception of the overall mass balance of the North Atlantic Ocean.

Upward transport of oceanic nitrate by migrating diatom mats

The oligotrophic gyres of the open sea are home to a flora that includes the largest known phytoplankton. These rare species migrate as solitary cells or aggregations (mats) between deep nutrient pools (below 80–100 m) and the surface. This migration contributes to new production because of the concomitant upward transport of nitrate. But just how significant this contribution is remains uncertain because of the difficulty of making quantitative measurements of these rare cells. Here we report remote video observations of a previously undersampled class of diatom (Rhizosolenia) mats throughout the upper 150 m of the central North Pacific Ocean. These mats are virtually invisible to divers, and their presence increases the calculated phytoplankton-mediated nitrate transport into the surface ocean by up to a factor of eight. Cruise averages indicate that Rhizosolenia mats transport 18–97 µmol N m−2 d−1; however, this value reached 171 μmol N m−2 d−1 at individual stations, a value equivalent to 59% of the export production. Although considerable temporal and spatial variability occurs, this means of upward nutrient transport appears to be an important source of new nitrogen to the surface ocean, and may contribute to other regional elemental cycles as well.

An assessment of nitrogen fixation as a source of nitrogen to the North Atlantic Ocean

The role of nitrogen fixation in the nitrogen cycle of the North Atlantic basin was re-evaluated because recent estimates had indicated a far higher rate than previous reports. Examination of the available data on nitrogen fixation rates and abundance ofTrichodesmium, the major nitrogen fixing organism, leads to the conclusion that rates might be as high as 1.09 × 1012 mol N yr−1. Several geochemical arguments are reviewed that each require a large nitrogen source that is consistent with nitrogen fixation, but the current data, although limited, do not support a sufficiently high rate. However, recent measurements of the fixation rates per colony are higher than the historical average, suggesting that improved methodology may require a re-evaluation through further measurements. The paucity of temporally resolved data on both rates and abundance for the major areal extent of the tropical Atlantic, where aeolian inputs of iron may foster high fixation rates, represents another major gap.

New production in the Sargasso Sea: History and current status

[1] The Sargasso Sea has been, and continues to be, the focus for research on new production in the open ocean. The history of the concept and the evolution of understanding of the mechanisms is reviewed from its inception in the early 1960s through a controversial period in the 1980s to the current status of a plethora of sources of new nitrogen. Rather than viewing all processes supplying new nutrients as uniformly distributed over the Sargasso Sea, it is now clear that new production in the northern or subtropical area is primarily sustained by nitrogen injection via mesoscale eddies and winter convection. In the tropical area, where permanent stratification precludes deep winter mixing and eddy kinetic energy is low, nitrogen fixation is potentially the dominant source along with diapycnal mixing and atmospheric deposition. The timescale of new production measurements has lengthened to an annual basis using time series measurements and satellite imagery but, in the context of climate change, should be lengthened further to greater than decadal scales. INDEX TERMS: 4845 Oceanography: Biological and Chemical: Nutrients and nutrient cycling, 4806 Oceanography: Biological and Chemical: Carbon cycling, 9325 Information Related to Geographic Region: Atlantic Ocean, 1615 Global Change: Biogeochemical processes (4805)

INTERNAL NITRATE CONCENTRATIONS IN SINGLE CELLS OF LARGE PHYTOPLANKTON FROM THE SARGASSO SEA1

Nitrate concentrations within individual cells of Ethmodiscus, Pyrocystis, and Halosphaera and chains of Rhizosolenia were determined from samples collected in the Sargasso Sea. In all cases, field populations exhibited a wide range of internal nitrate concentrations (INCs) within a single sampling date. Halosphaera INCs reached 100 mM, in contrast to diatom and dinoflagellate INCs, which did not exceed 22 mM. Sinking Rhizosolenia, Ethmodiscus and Pyrocystis had significantly lower internal NO3‐ pools than did floating cells (P< 0.05). Ethmodiscus incubations in surface seawater resulted in a dramatic reduction in the proportion of high INC cells concurrent with decreases in average INCs and an increased proportion of sinking cells. Population buoyancy was inversely related to INC, and negatively buoyant cells rarely exceeded I mM INC, suggesting that a critical INC threshold may exist. The photosynthetic parameters Pmax and α decreased with time as internal NO3‐‐ Pools were depleted. Internal nitrate depletion rates were consistent with oxygen production rates during this time. Based on the known characteristics of Pyrocystis and Ethmodiscus, we conclude that virtually all of the > 100 μm‐sized phytoplankton present in the Sargasso Sea can vertically migrate. However, the appropriate time scale for migrators such as Halosphaera that reproduce by swarmer formation is unclear and may be significantly different than the other taxa studied. Changes in the frequency distributions, buoyancy‐internal pool relationships, and general P‐1 photosynthesis‐irradiance time series data in Ethmodiscus suggest that nutrient limitation is related to these migrations. High INC appears to be a fundamental property of the largest microalgal cells present in oligotrophic seas and suggests that nitrate transport by these nonmotile cells is widespread.

BIOLOGICAL AND CHEMICAL CHARACTERISTICS OF THE GIANT DIATOM ETHMODISCUS (BACILLARIOPHYCEAE) IN THE CENTRAL NORTH PACIFIC GYRE

Cells of the giant diatom Ethmodiscus Castr. gathered from the upper 15 m were examined for O2 evolution, nitrate reductase activity (NRA), C and N composition, internal NO concentrations, , and 15NO, 15NH , and 32Si uptake in a series of cruises in the central N. Pacific gyre. The δ15N (2.56–5.09 ‰), internal NO concentrations (0.0– 11.5 mM NO−), and NRA (6.7 ± 4.7 × 10−4μM NO cell −1·h−1) were consistent with recent exposure to elevated nitrate concentrations and utilization of deep NO as a primary N source. These results are similar to other diatoms that migrate vertically to the nutricline as part of their life cycle. Rate measures (Si[OH]4 uptake, NRA, and O2 evolution) indicated surface doubling times from 45 h to 75 h. Both NO and NH uptake in surface waters were low and inadequate to supply N needs at surface NO and NH concentrations. Our results suggest a partitioning in nutrient acquisition, with N acquired at depth and C and Si acquired at the surface. Doubling rates were two to three times higher than predicted from cell volume and C content models. These data are consistent with the observed elemental content being lower than expected because of the dominance of cell volume by the vacuole. Our calculations suggest that Ethmodiscus contributes little to the biogeochemistry of the upper water column via upward nutrient transport. Although reported as a paleo‐upwelling indicator, thisevidence suggests that Ethmodiscus has adapted to the nutrient‐poor open ocean by a vertical migration strategy and has biological characteristics inconsistent with a upwelling indicator.

Estimation of the air/sea exchange of ammonia for the North Atlantic Basin

As gas phase atmospheric ammonia reacts with acidic aerosol particles it affects the chemical, physical, and optical properties of the particles. A knowledge of the source strengths of NH3 is useful in determining the effect of NH3 on aerosol properties on a regional basis. Here, an attempt is made to determine the direction and magnitude of the air/sea flux of ammonia for the North Atlantic Basin from both measured and modeled seawater and atmospheric ammonia concentrations. Previously reported measured seawater concentrations range from less than 30 to 4600 nM with the highest concentrations reported for the Caribbean Sea, the North Sea, and the Belgium coast. Measured atmospheric ammonia concentrations range from 2 to 500 nmol m−3 with the largest values occurring over the Sargasso Sea, the Caribbean Sea, and the North Sea. For comparison to the measurements, seawater ammonia concentrations were calculated by the Hamburg Model of the Ocean Carbon Cycle (HAMOCC3). HAMOCC3 open ocean values agree well with the limited number of reported measured concentrations. Calculated coastal values are lower than those measured, however, due to the coarse resolution of the model. Atmospheric ammonia concentrations were calculated by the Acid Deposition Model of the Meteorological Synthesizing Center (MSC-W) and by the global 3-dimensional model Moguntia. The two models predict similar annually averaged values but are about an order of magnitude lower than the measured concentrations. Over the North Sea and the NE Atlantic, the direction and magnitude of the air/sea ammonia flux calculated from MSC-W and Moguntia agree within the uncertainty of the calculations. Flux estimates derived from measured data are larger in both the positive and negative direction than the model derived values. The discrepancies between the measured and modeled concentrations and fluxes may be a result of sampling artifacts, inadequate chemistry and transport schemes in the models, or the difficulty in comparing point measurements to time-averaged model values. Sensitivity tests were performed which indicate that, over the range of values expected for the North Atlantic, the accuracy of the calculated flux depends strongly on seawater and atmospheric ammonia concentrations. Clearly, simultaneous and accurate measurements of seawater and atmospheric ammonia concentrations are needed to reduce the uncertainty of the flux calculations, validate the model results, and characterize the role of oceanic ammonia emissions in aerosol processing and nitrogen cycling for the North Atlantic.

Seasonal fluctuations of nitrite concentrations in the deep oligotrophic ocean

Abstract Concentrations of nitrite in the Sargasso Sea near Bermuda were measured monthly for 3 years using a chemiluminescent analysis capable of precise determination of concentrations as low as 1 nM. Ammonium and dissolved primary amine concentrations were also determined occasionally. The mean nitrite concentration over the entire period declined exponentially with depth between 300 and 1000 m ( r 2 > 0.99) and then declined slightly from 1.9 nM to 1.3 nM at 2600 m. From 150 to 250 m, the data had a bimodal distribution, with more than one-third of the observations comprising higher concentrations (mean ∼ 147 nM) during February-April that are characteristic of the “classic” primary nitrite maximum (PNM) and almost two-thirds comprising low concentration values (mean ∼ 17 nM) that would not have been detected by colorimetric analysis. A rapid oscillation between the two modes was observed. Combining the exponential [N0 2 − ]-depth relationship with Redfield ratio assumptions and apparent oxygen utilization (AOU) rates for the region (Jenkins, 1982), it was calculated that at steady state, nitrite turnover rates over the 150–1000 m depth interval range from 3 to 7 days. The depth integrated nitrite inventory reaches a maximum in the spring and is correlated with peaks in primary productivity and sediment flux. Ammonium concentrations were similar to, or higher than, nitrite concentrations and also increased dramatically during winter mixing with values of 50–100 nM in the 100–300 m depth interval before decreasing to 5–20 nM values at greater depths. During the remainder of the year, concentrations were relatively constant with depth compared to the nitrite concentration profile.

Diode Array Spectrometer for Nitrogen Isotopic Analysis

The analytical precision for measurements of the isotopic composition of dinitrogen gas by emission spectrometry has primarily been limited by variability of the spectral background. Imaging the spectral region surrounding the isotopic bandheads with a diode array permitted a correction to be made for the background spectra and increased the statistical precision of the raw data. With the use of a multiple regression algorithm to process the spectral data, the standard deviation for an individual determination at the natural abundance level was 0.002 atom %. After an initial calibration, the standard deviation for measurements of the same sample, repeated over several years, remained at that level. In contrast to the case for commercial instruments, samples of widely varying mass can be readily analyzed. Further increases in precision are limited by an unidentified source of variance.