Ocean Sciences [OS]

OS36F HCC:HALL 3 Wednesday

Diatom Diazotroph Assemblages II Posters

Presiding:D G Capone, University of Southern California; A E White, Oregon State University

OS36F-01

Do Diatom-Diazotroph Associations Contribute Significantly to C Sequestration Globally?

* Subramaniam, A (ajit@ldeo.columbia.edu) , Lamont Doherty Earth Observatory of Columbia University, 61 Rt 9W, Palisades, NY 10964 United States
Karl, D M (dkarl@hawaii.edu) , School of Ocean and Earth Science and Technology, University of Hawaii, Honolulu, HI 96822 United States
Shipe, R (rshipe@ucla.edu) , UCLA Institute of the Environment, LaKretz Hall, Suite 300, Los Angeles, CA 90095-1496 United States
Villareal, T (tracy@utmsi.utexas.edu) , Marine Science Institute, The University of Texas at Austin, 750 Channel View Dr., Port Aransas, TX 78373 United States

The nitrogen-fixing cyanobacteria {\it Richelia intracellularis} is often found as a symbiont associated with various diatom species such as {\it Hemiaulus} and {\it Rhizosolenia}. Extensive blooms of Diatom-Diazotroph Associations (DDA) have been described in the Western and Eastern tropical Atlantic Ocean, associated with the Amazon and Congo River plumes. Presence of DDA have also been reported in the surface waters of the South China Sea and have been inferred from sediment trap samples in the Bay of Bengal. Variations in elemental stoichiometry have been used in the Mediterranean to relate DDA to freshwater input there. Extensive near surface blooms of DDA that have been tracked through the water column into sediment traps at 4000m have also been described in the subtropical North Pacific Ocean. DDA are potentially a significant source of "new" nitrogen and the sinking of "new" carbon fixed by DDA can be a globally important source of biologically mediated carbon sequestration to the deep ocean. We have calculated that DDA associated with the Amazon River alone could sequester up to 15 TgC/year. We will use the inventory of literature reports of DDA to present the global impact of sequestration of carbon due to sinking DDA.

OS36F-02

Probing the Diazocyte Hypothesis: $^{13}$C and $^{15}$N Uptake in {\it Trichodesmium} IMS-101 Using Secondary Ion Mass Spectrometry

Finzi, J A (finzi@usc.edu) , University of Southern California, 3616 Trousdale Pkwy, AHF 108, Los Angeles, CA 90089 United States
Weber, P K (weber21@llnl.gov) , Lawrence Livermore National Laboratory, 7000 East Avenue, Livermore, CA 94550 United States
Popa, R (rpopa@pdx.edu) , Portland State University, PO Box 751, Portland, OR 97207 United States
Pett-Ridge, J (pettridge2@llnl.gov) , Lawrence Livermore National Laboratory, 7000 East Avenue, Livermore, CA 94550 United States
Fallon, S J (fallon4@llnl.gov) , Lawrence Livermore National Laboratory, 7000 East Avenue, Livermore, CA 94550 United States
Nealson, K H (knealson@usc.edu) , University of Southern California, 3616 Trousdale Pkwy, AHF 108, Los Angeles, CA 90089 United States
* Capone, D G (capone@usc.edu) , University of Southern California, 3616 Trousdale Pkwy, AHF 108, Los Angeles, CA 90089 United States

The diazotrophic cyanobacterium {\it Trichodesmium} is capable of fixing both CO$_{2}$ and N$_{2}$ concurrently throughout the day yet it does not contain heterocysts. In order to avoid oxygen inhibition of nitrogenase, recent theories suggest that {\it Trichodesmium} may spatially and/or temporally segregate these two processes along the trichome. Using nanometer-scale secondary ion mass spectrometry (nanoSIMS), we sought to address these two theories. {\it Trichodesmium} IMS-101 cultures were incubated in media enriched with both $^{15}$N-N$_{2}$ gas and $^{13}$C-NaHCO$_{3}$. We sampled at seven timepoints throughout the light part of the cycle. Through analysis on the nanoSIMS, we were able to measure parallel $^{15}$N and $^{13}$C incorporation into individual cells along the trichome at each discrete timepoint. Our results suggest that there is indeed temporal segregation of the two processes with CO$_{2}$ fixation maximal in the morning and N$_{2}$ fixation maximal at mid-day. In addition, subsamples were also analyzed on an isotope ratio mass spectrometer (IRMS). The IRMS data corroborate the nanoSIMS data and equally suggest that there is temporal segregation of the two fixation processes. The case for spatial segregation was not as clear.

OS36F-03

Station ALOHA Blooms: Biogeochemical Characteristics of a Large Plankton Bloom in the Oligotrophic North Pacific Ocean

* Sadler, D W (sadler@hawaii.edu) , School of Ocean and Earth Science and Technology, Department of Oceanography, University of Hawaii, 1000 Pope Rd., Honolulu, HI 96816 United States
Bjorkman, K M (bjorkman@hawaii.edu) , School of Ocean and Earth Science and Technology, Department of Oceanography, University of Hawaii, 1000 Pope Rd., Honolulu, HI 96816 United States
Church, M J (mjchurch@hawaii.edu) , School of Ocean and Earth Science and Technology, Department of Oceanography, University of Hawaii, 1000 Pope Rd., Honolulu, HI 96816 United States
Clemente, T M (tclement@hawaii.edu) , School of Ocean and Earth Science and Technology, Department of Oceanography, University of Hawaii, 1000 Pope Rd., Honolulu, HI 96816 United States
Curless, S E (curless@hawaii.edu) , School of Ocean and Earth Science and Technology, Department of Oceanography, University of Hawaii, 1000 Pope Rd., Honolulu, HI 96816 United States
Foley, D G (Dave.Foley@noaa.gov) , Joint Institute for Marine and Atmospheric Research, University of Hawaii, 1352 Lighthouse Ave., Pacific Grove, CA 93950 United States
Fong, A A (fonga@hawaii.edu) , School of Ocean and Earth Science and Technology, Department of Oceanography, University of Hawaii, 1000 Pope Rd., Honolulu, HI 96816 United States
Fujieki, L A (fujieki@hawaii.edu) , School of Ocean and Earth Science and Technology, Department of Oceanography, University of Hawaii, 1000 Pope Rd., Honolulu, HI 96816 United States
Grabowski, E M (grabowsk@hawaii.edu) , School of Ocean and Earth Science and Technology, Department of Oceanography, University of Hawaii, 1000 Pope Rd., Honolulu, HI 96816 United States
Gregory, T K (thomasg@hawaii.edu) , School of Ocean and Earth Science and Technology, Department of Oceanography, University of Hawaii, 1000 Pope Rd., Honolulu, HI 96816 United States
Karl, D M (dkarl@hawaii.edu) , School of Ocean and Earth Science and Technology, Department of Oceanography, University of Hawaii, 1000 Pope Rd., Honolulu, HI 96816 United States
Letelier, R M (letelier@coas.oregonstate.edu) , College of Oceanic and Atmospheric Sciences, Oregon State University, 104 Oceanography Administration Building, Corvallis, OR 97331-5503 United States
Lethaby, P J (lethaby@hawaii.edu) , School of Ocean and Earth Science and Technology, Department of Oceanography, University of Hawaii, 1000 Pope Rd., Honolulu, HI 96816 United States
Lukas, R (rlukas@hawaii.edu) , School of Ocean and Earth Science and Technology, Department of Oceanography, University of Hawaii, 1000 Pope Rd., Honolulu, HI 96816 United States
Mahaffey, C (mahaffey@soest.hawaii.edu) , School of Ocean and Earth Science and Technology, Department of Oceanography, University of Hawaii, 1000 Pope Rd., Honolulu, HI 96816 United States
Maenner, S (Stacy.Maenner@noaa.gov) , Pacific Marine Environmental Laboratory, National Oceanic and Atmospheric Administration, 7600 Sandpoint Way NE, Seattle, WA 98115 United States
McAndrew, P M (pmacandre@hawaii.edu) , School of Ocean and Earth Science and Technology, Department of Oceanography, University of Hawaii, 1000 Pope Rd., Honolulu, HI 96816 United States
Sabine, C L (Chris.Sabine@noaa.gov) , Pacific Marine Environmental Laboratory, National Oceanic and Atmospheric Administration, 7600 Sandpoint Way NE, Seattle, WA 98115 United States
Santiago-Mandujano, F (santiago@hawaii.edu) , School of Ocean and Earth Science and Technology, Department of Oceanography, University of Hawaii, 1000 Pope Rd., Honolulu, HI 96816 United States
Watkins, B V (blakewat@hawaii.edu) , School of Ocean and Earth Science and Technology, Department of Oceanography, University of Hawaii, 1000 Pope Rd., Honolulu, HI 96816 United States

A central objective of the Hawaii Ocean Time-series (HOT) program is to understand how physical processes influence ocean biogeochemistry. Integration of satellite remote sensing technologies and moorings with monthly shipboard measurements is beginning to provide insight into both spatial and temporal dynamics in upper ocean processes. Using satellite derived ocean color we observed and then sampled a $\sim$200 km wide, spatially coherent region of enhanced chlorophyll concentration in the immediate vicinity of Station ALOHA. Analyses of satellite derived sea surface altimetry and surface currents suggested the feature was associated with an eddy traveling south-west towards Station ALOHA. To assess the influence of such events on upper ocean biogeochemistry we conducted a transect from the approximate center of the region of enhanced chlorophyll towards its western edge. Samples were collected from the upper ocean (0-125 m) for determination of soluble reactive phosphorus (SRP), nitrate+nitrite (N+N), silicic acid (Si) and dissolved inorganic carbon concentrations, as well as measurements of plankton community structure based on flow cytometry. Generally these measurements revealed that depth-integrated inventories of SRP and N+N were elevated relative to their respective seasonal climatologies at Station ALOHA, driven by an apparent shoaling of the nutricline associated with the eddy feature. A moored pCO$_{2}$ sensor clearly showed the bloom passing through Station ALOHA and revealed this feature to be an unseasonable sink for CO$_{2}$. Based on microscopic analyses, the bloom consisted of large diatoms supported in part by increased abundances of nitrogen-fixing bacteria. Similar diatom-diazotroph blooms appear to be a regular occurrence during the summertime months throughout the subtropical North Pacific. As a result, understanding the spatial characteristics of this eddy may provide insight into the physical and biogeochemical conditions that favor the regular, summertime accumulation of similar plankton assemblages in the subtropical North Pacific Ocean.

OS36F-04

Oceanographic Conditions Related to Diatom Only and Diatom-Diazotroph Assemblages in the Western Tropical Atlantic Ocean, Spring 2003

* Shipe, R F (rshipe@ucla.edu) , UCLA Department of Ecology and Evolutionary Biology, 621 Charles Young Drive South, Los Angeles, CA 90095 United States
Curtaz, J (jcurtaz@ucla.edu) , UCLA Department of Ecology and Evolutionary Biology, 621 Charles Young Drive South, Los Angeles, CA 90095 United States
Subramaniam, A (ajit@ldeo.columbia.edu) , Lamont Doherty Earth Observatory at Columbia University, 61 Rt 9W, Palisades, NY 10964 United States
Carpenter, E (ecarpent@sfsu.edu) , Romberg Tiburon Center, San Francisco State University, 3152 Paradise Drive, Tiburon, CA 94920 United States
Capone, D G (capone@usc.edu) , University of Southern California, Wrigley Institute for Environmental Studies & Department of Biological Sciences, 3616 Trousdale Parkway, AHF 108, Los Angeles, CA 90089 United States

The introduction of nutrient rich river plume water into the Western Tropical Atlantic Ocean supports the growth of extensive phytoplankton blooms that are often rich in diatoms. During high discharge seasons, the blooms extend hundreds of kilometers offshore and assemblages can be categorized as those dominated by diatoms with or without associations with diazotrophic (nitrogen-fixing) cyanobacteria. However, few measurements have addressed (i.) the environmental conditions that support the growth of diatom-diazotroph assemblages (DDAs) versus diatom only assemblages (DAs) nor (ii.) the impacts of DDAs and DAs on ecosystem processes and biogeochemical cycles. We begin to address these concerns with chemical and biological measurements throughout the euphotic zone in plume waters (salinities 24-35) sampled during a research expedition in the Western Tropical Atlantic in April-May 2003. DAs were observed at 8 stations to the northwest of the Amazon river mouth in relatively shallow mixed layers (6-17 m) whereas DDAs were observed farther to the north at 8 stations characterized by deeper mixed layers (13-32 m) and higher salinities. One major finding was that DA stations had higher dissolved silicon concentrations compared to DDA stations, whereas nitrate was depleted to undetectable concentrations at the surface at all stations. A surprising result is that DDAs, although associated with low dissolved silicon waters, were characterized by higher biogenic silica and chlorophyll a content per cell. As vertically integrated primary productivity rates were much higher at DDA stations, we suggest that areas dominated by DDA have a much larger impact on vertical particulate fluxes relative to DAs.

OS36F-05

Diazotroph Abundance and Diversity During a Large Bloom in the Oligotrophic North Pacific Gyre

* Fong, A A (fonga@hawaii.edu) , University of Hawaii, 1000 Pope Road, Honolulu, HI 96822 United States
Church, M J (mjchurch@hawaii.edu) , University of Hawaii, 1000 Pope Road, Honolulu, HI 96822 United States
Karl, D M (dkarl@hawaii.edu) , University of Hawaii, 1000 Pope Road, Honolulu, HI 96822 United States
Lukas, R (rlukas@hawaii.edu) , University of Hawaii, 1000 Pope Road, Honolulu, HI 96822 United States
Mahaffey, C (clairema@hawaii.edu) , University of Hawaii, 1000 Pope Road, Honolulu, HI 96822 United States
Zehr, J P (zehr@pmc.ucsc.edu) , University of California Santa Cruz, 1156 High Street, Santa Cruz, CA 95064 United States

The Hawaii Ocean Time-series (HOT) program has monitored biogeochemical inventories and fluxes at Station ALOHA since October 1988. Resolution of unique phenomena such as bloom events are relatively rare due in part to the monthly sampling regime. In July of 2005, a localized region of enhanced chlorophyll in the vicinity of Station ALOHA was observed by ocean color satellite and sampled for a suite of biogeochemical properties. Microscopic analyses revealed the bloom was dominated by a diverse assemblage of diatoms, and unicellular and heterocystous cyanobacteria. Samples were collected for DNA and RNA analyses of nifH gene abundances, expression, and diversity to assess the distributions and activities of nitrogen fixers (diazotrophs) along a transect from the observed center of the bloom. In addition, surface water rates of nitrogen fixation were measured by 15N assimilation into plankton biomass. The nifH gene abundances and diversity of six specific groups of diazotrophs were quantified using quantitative PCR (QPCR). The QPCR analyses suggest significant spatial variability in nifH gene abundances and diazotroph diversity, with unicellular cyanobacteria typically dominating the nifH gene abundances within and outside the bloom feature. Surface water rates of nitrogen fixation were on average 3-8 times greater than rates typically measured at Sta. ALOHA. By merging rate measurements of nitrogen fixation with molecular determinations of the dominant diazotroph assemblages we hope to better constrain the contribution of specific diazotrophs to new production in oligotrophic ocean ecosystems. Interpretation of these data will provide insight into the response and ecological functionality of diazotroph supported blooms in the oligotrophic North Pacific Ocean.

OS36F-06

In situ Optical Characterization of Summer Blooms in the North Pacific Subtropical Gyre

* Nahorniak, J (jasmine@coas.oregonstate.edu) , College of Ocaenic and Atmospheric Sciences, Oregon State University, 104 COAS Admin. Bldg., Corvallis, OR 97331 United States
Letelier, R M (letelier@coas.oregonstate.edu) , College of Ocaenic and Atmospheric Sciences, Oregon State University, 104 COAS Admin. Bldg., Corvallis, OR 97331 United States
Amanda, A (aashe@coas.oregonstate.edu) , College of Ocaenic and Atmospheric Sciences, Oregon State University, 104 COAS Admin. Bldg., Corvallis, OR 97331 United States
Karl, D M (dkarl@hawaii.edu) , School of Ocean and Earth Science and Technology, University of Hawaii, 1000 Pope Road, Honolulu, HI 96816 United States
Church, M J (mjchurch@hawaii.edu) , School of Ocean and Earth Science and Technology, University of Hawaii, 1000 Pope Road, Honolulu, HI 96816 United States
Lance, F (lfujieki@hawaii.edu) , School of Ocean and Earth Science and Technology, University of Hawaii, 1000 Pope Road, Honolulu, HI 96816 United States

The North Pacific subtropical gyre has been characterized historically as an oligotrophic pelagic ecosystem with low carrying capacity. However, over the past 15 years, the Hawaii Ocean Time-series (HOT) program has recorded several summer phytoplankton blooms where the dominant photoautotrophic taxa can be filamentous cyanobacteria or diatoms. These events are ecologically and biogeochemically important because they can trigger large exports of organic matter into the benthic environment. Because the depth distribution of these blooms is characterized by a surface or subsurface biomass maximum located at the base of the mixed-layer, making it difficult to quantify their magnitude from above water remote sensing observations. As part of the HOT program, nearly monthly deployments of a Profiling Reflectance Radiometer (PRR) in the upper water column (0-200m) have been made since February 1998. In addition, in October 2002, an optical package including a 9 wavelength absorption and attenuation meter (ac-9), a fast repetition rate fluorometer (FRRf), and a CTD was added to our monitoring program in order to assess the vertical and temporal variability of inherent optical properties and phytoplankton photosynthetic parameters in the euphotic zone. The PRR records display a distinct signature for summer blooms during the years 1998, 2000, and 2005. This signature consists of a significant increase in the attenuation coefficient at the base of the mixed-layer, a concomitant decrease in the PRR derived fluorescence quantum yield in this layer, and a shoaling of isolumes below it. In addition, although no significant blooms were observed in the PRR data between 2002 and 2004, the ac-9 measurements in the upper euphotic zone recorded a significant increases in scattering and changes in the particulate absorption spectrum that are consistent with the presence of filamentous cyanobacteria during autumn 2002 and 2003. This change in the absorption spectrum was not observed in a diatom dominated bloom during summer 2005. Our results suggest that it is possible to characterize the nature and magnitude of summer blooms and to quantify seasonal and interannual changes in the abundance of filamentous cyanobacteria in the North Pacific Subtropical gyre using vertical profiling moorings equipped with optical instruments similar to those used in this study.

http://picasso.coas.oregonstate.edu/ORSOO

OS36F-07

Nitrogen Fixation in {\it Teredinibacter turnerae}, a Cultivated Symbiont of the Marine Shipworm {\it Lyrodus pedicellatus}

* Horak, R E (gtg284z@mail.gatech.edu) , Department of Biology, Georgia Institute of Technology, 311 Ferst Dr., Atlanta, GA 30332 United States
Distel, D L (distel@OGLF.org) , Ocean Genome Legacy, New England Biolabs, Inc., 32 Tozer Rd., Beverly, MA 01915 United States
Montoya, J P (j.montoya@biology.gatech.edu) , Department of Biology, Georgia Institute of Technology, 311 Ferst Dr., Atlanta, GA 30332 United States

Marine wood-boring bivalves, or shipworms, harbor a polysymbiosis of cellulolytic and diazotrophic bacteria in gill bacteriocytes that allows the host to live on a nitrogen-poor diet composed only of wood. As the principal consumer of wood in marine ecosystems, shipworms can affect both the carbon and nitrogen cycles in marine environments by degrading cellulosic material and converting a portion into biomass. {\it Teredinibacter turnerae}, a symbiont isolated from the shallow water marine shipworm {\it Lyrodus pedicellatus}, is capable of both cellulose digestion and nitrogen fixation. The major goals of our research were to quantify the contribution of {\it T. turnerae} to host nutrition and to determine how nitrogen fixation in {\it T. turnerae} is affected by ambient conditions, such as oxygen, ammonium, and nitrate. {\it Teredinibacter turnerae} cultures grown in nitrogen-free media are relatively depleted in Δ$ $^{15}$N (Δ$ $^{15}$N = 2.2$\permil$) while cultures grown with combined nitrogen in the form of ammonium are relatively enriched in Δ$ $^{15}$N (Δ$ $^{15}$N = 16.65$\permil$). Acetylene reduction assays of nitrogenase activity showed that {\it T. turnerae} had a nitrogen specific growth rate of 5.6 x 10$^{-4}$ hr$^{-1}$ and a nitrogen turnover time of 75 days.

OS36F-08

Nitrogenase Gene Expression in the North Pacific Subtropical Gyre

* Zehr, J P (zehrj@pmc.ucsc.edu) , University of California, Department of Ocean Sciences, E&MS 438, Santa Cruz, CA 95064 United States
Montoya, J P (joseph.montoya@biology.gatech.edu) , Georgia Institute of Technology, School of Biology, Atlanta, GA 30332 United States
Short, C (Cynthia.Short@uchsc.edu) , University of California, Department of Ocean Sciences, E&MS 438, Santa Cruz, CA 95064 United States
Hansen, A (ahansen71@hotmail.com) , University of Hawaii, Dept. of Oceanography, Honolulu, HI 96822 United States
Jenkins, B D (bjenkins@uri.edu) , University of California, Department of Ocean Sciences, E&MS 438, Santa Cruz, CA 95064 United States
Church, M J (mjchurch@hawaii.edu) , University of Hawaii, Dept. of Oceanography, Honolulu, HI 96822 United States
Karl, D M (dkarl@hawaii.edu) , University of Hawaii, Dept. of Oceanography, Honolulu, HI 96822 United States

We performed phosphorus addition experiments with natural assemblages of oceanic phytoplankton in the Pacific Ocean to determine the effect of nutrient enrichment on diazotrophic population dynamics and nitrogenase gene expression. Cloning and sequencing of PCR and RT-PCR {\it nifH} amplification products from experiments with water collected from Station ALOHA and from Kaneohe Bay, Hawai'i, yielded the same groups of nifH phylotypes previously reported: cyanobacteria belonging to the unicellular Groups A and B, the genus Trichodesmium, heterocystous groups as well as gamma Proteobacteria. QPCR and QRT-PCR showed that the abundance of the bacteria, including cyanobacteria, did not change significantly during experimental incubations, but all had distinct diel patterns of gene expression. The Group B cyanobacterial {\it nifH} gene abundance was correlated with the presence of 3-8 micrometer diameter phycoerythrin-containing cells, whereas the Group A cyanobacterial abundance was not, suggesting that Group A may be smaller cells or do not contain phycoerythrin. The Group A cyanobacteria, which have not yet been successfully maintained in culture, express nitrogenase primarily during the day, which is unusual for a unicellular cyanobacterium. Patterns of gene expression in the experiments were similar to those documented for natural populations, and abundances of individual populations were stable over the time period of the experiment (36 hours), indicating that the experimental design simulated the natural environment. Phosphorus additions had no short-term (36 hr) effect on either abundance or {\it nifH} gene expression. The QPCR and QRT-PCR methods, coupled with experimental manipulations, provide a powerful approach for examining the factors controlling the distribution and activity of diazotrophs in the sea.

OS36F-09

Size-Fractionated Rates and Patterns of Nitrogen and Carbon Fixation in the North Pacific Subtropical Gyre

* Montoya, J P (j.montoya@biology.gatech.edu) , School of Biology, Georgia Institute of Technology 311 Ferst Drive, Atlanta, GA 30332 United States
Hansen, A (ahansen@hawaii.edu) , Department of Oceanography, University of Hawaii 1000 Pope Road, Honolulu, HI 96822 United States
Karl, D M (dkarl@hawaii.edu) , Department of Oceanography, University of Hawaii 1000 Pope Road, Honolulu, HI 96822 United States
Zehr, J P (zehrj@cats.ucsc.edu) , Department of Ocean Sciences, University of California at Santa Cruz 110 High Street, Santa Cruz, CA 95064 United States

We measured the rates of nitrogen and carbon fixation in seawater collected at Station ALOHA and in coastal waters just outside Kaneohe Bay, Oahu, Hawaii. The rate of nitrogen fixation by particles passing a 10 $\mu$m filter varied between 0.014 and 0.095 nmol N L$^{-1}$ h$^{-1}$ with little or no diel variation in natural samples. We found similar rates in most samples amended with phosphate, though two of our five experiments showed signs of enhanced nitrogen fixation in the presence of added P. The rate of carbon fixation by small cells ranged from 3.5 to 16.2 nmol C L$^{-1}$ h$^{-1}$ and showed no effect of added P in any of our experiments. The ratio of biomass specific fixation rates indicated that diazotrophs comprised between roughly 0.2 and 2% of the biomass of photosynthetically active plankton in our incubations. In combination with molecular assays of diazotroph community structure, our results indicate that diazotroph abundance is a major determinant of nitrogen fixation rate, and that different diazotroph groups have very different cell-specific rates of nitrogen fixation. Our data imply a rate of nitrogen fixation by small diazotrophs comparable to geochemically derived estimates of the rate of nitrogen fixation at Station ALOHA. On short time scales (weeks to months), N$_{2}$ fixed by small diazotrophs is more likely to be retained in the upper water column, with very different implications for food web structure and vertical export of organic matter than nitrogen fixed by larger, often episodically occurring diazotrophs like {\it Trichodesmium} and chain/mat forming diatoms containing endosymbiotic cyanobacteria.