Responses of DMS in the seawater and atmosphere to iron enrichment in the subarctic western North Pacific (SEEDS-II)

Simultaneous measurements of dimethylsulfide (DMS) in the seawater and atmosphere were conducted during SEEDS-II to investigate the responses of DMS to iron (Fe) fertilization in the subarctic North Pacific. No significant increases in the seawater DMS (DMSw) concentration were observed inside the fertilized patch compared to those outside the patch, while particulate dimethylsulfoniopropionate (DMSPp) concentration inside the patch increased 2-fold compared to those outside the patch in the phytoplankton bloom of major DMSP producers such as prasinophytes, cryptophytes, diatoms and prymnesiophytes. In the decline phase of the bloom, maximum DMSw was observed both inside the patch (ca. 6.2 nM) and outside the patch (ca. 9.3 nM). In this period, increases in mesozooplankton and decreases in the DMSP producers (prymnesiophytes and diatoms) were observed both sides of the patch, but larger inside the patch than outside the patch. Large decreases in the DMSPp inside the patch, which was probably related to the large increases in mesozooplankton inside the patch, did not result in increases in the DMSw concentration. Considering biological and nonbiological parameters, we discussed these results, although they could not be completely explained. Unfortunately, the impact of Fe fertilization on the atmospheric DMS (DMSa) concentration was not detected due to no significant changes in DMSw. However, it is noted that DMSa concentrations were dependent on the sea–air DMS flux in the air from higher latitudes and/or the Eurasian continent, though the DMS flux was a minor role to the budget of DMSw. Therefore if DMSw were significantly changed by Fe fertilization, DMSa might be affected through changes in the sea-air flux in this condition.     

Iron-induced alterations of bacterial DMSP metabolism in the western subarctic Pacific during SEEDS-II

The effect of added iron on bacterial cycling of the climate-active gas dimethylsulfide (DMS) and its precursor dimethylsulfoniopropionate (DMSP) was tested during the second Subarctic Pacific Iron Experiment for Ecosystem Dynamics Study (SEEDS II) from 19 July to 21 August 2004 aboard the R/V Hakuho-Maru. The study area in the northwest Pacific Ocean (48°N 165°E) was enriched with Fe and the conservative tracer, SF₆, allowing the fertilized patch to be tracked. Microbial DMSP cycling rates were determined in the surface mixed layer (5 m) during incubations using the ³⁵S-DMSP technique. The addition of iron resulted in a 4-fold increase in concentrations of chlorophyll a (chl a) within the surface mixed layer (5 m depth), and the length of the sampling period allowed the observation of both bloom and post-bloom conditions. Inside the fertilized patch, the alleviation of resource limitation gave rise to the concurrent increase in bacterial abundance and production. Changes in the phytoplankton community within the Fe-enriched patch translated into a sustained decrease in chl a-normalized particulate DMSP (DMSP_p) concentrations, suggesting a preferential stimulation of the growth of DMSP_p-poor phytoplankton species. Despite short-lived peaks of DMSP_p within the Fe-enriched area, concentrations of DMSP_p generally remained stable during the entire sampling period inside and outside the fertilized patch. During the Fe-induced bloom, microbial DMSP-sulfur (DMSP-S) assimilation efficiency increased 2.6-fold inside the Fe-enriched area, which indicated that as bacterial production increased, a greater proportion of DMSP-S was assimilated and possibly diverted away from the bacterial cleavage pathway (i.e. production of DMS). Our results suggest that iron-induced stimulation of weak DMSP_p-producers and DMSP-assimilating bacteria may diminish the potential production of DMS and thus limit its flux towards the atmosphere over the subarctic Pacific Ocean.     

Community structure and photosynthetic physiology of phytoplankton in the northwest subarctic Pacific during an in situ iron fertilization experiment (SEEDS-II)

Temporal changes in the abundance, community composition, and photosynthetic physiology of phytoplankton in surface waters were investigated during the second in situ iron (Fe) fertilization experiment in the NW subarctic Pacific (SEEDS-II). Surface chlorophyll a concentration was 0.75 mg m⁻³ on the day before the first Fe enrichment (i.e. Day 0), increased ca. 3-fold until Day 13 after two Fe additions, and thereafter declined with time. The photochemical quantum efficiency (F_v/F_m) and functional absorption cross-section (σPSII) of photosystem II for total phytoplankton in surface waters increased and decreased inside the Fe-enriched patch through Day 13, respectively. These results indicate that the photosynthetic physiological condition of the phytoplankton improved after the Fe infusions. However, the maximum F_v/F_m value of 0.43 and the maximum quantum yield of carbon fixation (φ_max) of 0.041 mol C (mol photon)⁻¹ during the development phase of the bloom were rather low, compared to their theoretical maximum of ca. 0.65 and 0.10 mol C (mol photon)⁻¹, respectively. Diatoms, which were mainly composed of oceanic species, did not bloom, and autotrophic nanoflagellates such as cryptophytes and prasinophytes became predominant in the phytoplankton community inside the Fe-enriched patch. In ferredoxin/flavodoxin assays for micro-sized (20–200 μm in cell length) diatoms, ferredoxin was not detected but flavodoxin expressions consistently occurred with similar levels both inside and outside the Fe-enriched patch, indicating that the large-sized diatoms were stressed by Fe bioavailability inside the Fe-enriched patch even after the Fe enrichments. Our data suggest that the absence of a Fe-induced large-sized diatom bloom could be partly due to their Fe stress throughout SEEDS-II.     

Enrichment of alkanes within a phytoplankton bloom during an in situ iron enrichment experiment in the western subarctic Pacific

During the Subarctic Pacific Iron Experiment for Ecosystem Dynamics Study ΙΙ (SEEDS-II), we monitored variations in the concentrations of non-methane hydrocarbons (NMHCs), CH₃Cl, N₂O, and CH₄ within a phytoplankton bloom. Stable isotopic compositions were also determined to evaluate the sources of the variations. Although there was little variation in either the concentrations or the stable isotopic compositions of alkenes, CH₃Cl, N₂O, and CH₄ during the 23-day observation period, alkane concentrations increased substantially as the phytoplankton bloomed. The column-integrated quantities of alkanes increased to 3 times pre-bloom levels for C₂H₆, 5 times for C₃H₈, and 20 times for n-C₄H₁₀. The δ¹³C values of both C₂H₆ and C₃H₈ remained almost constant while concentrations increased, whereas that of n-C₄H₁₀ increased by about 12‰. To evaluate the sources of the alkanes produced during the bloom, we compared their δ¹³C values with those of alkanes produced in axenic phytoplankton cultures in our laboratory. We concluded that during the SEEDS-ΙΙ experiment the major portions of C₂H₆ and C₃H₈ were produced during the autolysis of diatoms cells, whereas n-C₄H₁₀ was produced during autolysis of other phytoplankton cells such as cryptophytes and dinoflagellates.     

Rare earth elements in seawater during an iron-induced phytoplankton bloom of the western subarctic Pacific (SEEDS-II)

During an iron-enrichment experiment in the western subarctic Pacific (SEEDS-II), concentrations of dissolved and acid-soluble rare earth elements (REEs) were determined in shallow waters. Detailed vertical profiles of dissolved REEs were compared with those of nutrients in a preliminary survey. The results showed good correlation, except for Ce and Y. Along with phytoplankton growth within the iron-enriched patch, dissolved REEs were decreased in surface waters. The changes in REE concentrations were small but systematic: light REEs were adsorbed preferentially over heavy REEs. This systematic trend is consistent with results of earlier studies of REE adsorption onto biogenic particle surfaces. However, we were unable to detect a decrease of acid-soluble REEs in surface waters. During the phytoplankton bloom decline phase, dissolved REEs were elevated to concentration levels at the preliminary survey, although nutrients were decreased. The liberation of REEs from the suspended particles might be related to disaggregation from large particles to colloid particles, desorption from the particulate phase, and remineralization from organic matter. Outside the Fe-patch, time variations of REEs were also observed during SEEDS-II, indicating that both artificially induced phytoplankton blooms and natural variation of biogenic particles can affect the biogeochemical cycles of REEs in shallow waters within a short time.     

Biogeochemical cycling of N and Si during the mesoscale iron-enrichment experiment in the western subarctic Pacific (SEEDS-II)

Biogeochemical cycles of N and Si were examined in the surface mixed layer during the mesoscale iron-enrichment (IE) experiment in the high-nutrient low-chlorophyll (HNLC) western subarctic Pacific (SEEDS-II). Although the IEs increased nitrate uptake, silicic acid utilization was not stimulated. The nitrate drawdown in the iron-patch (IN-patch, 140.3 mmol m⁻² in the surface mixed layer, 0–30 m) was only 25% of the initial inventory, which was 1/3–2/5 of the previous IE experiments in the subarctic Pacific. This relatively weak response of nutrient drawdown to IEs was due to the high biomass of mesozooplankton (MZ) dominated by copepod Neocalanus plumchrus. Feeding of MZ (247.2 mmol m⁻² during Day 0–21 from the first IE) in the IN-patch was higher than the nitrate drawdown and prevented further development of the phytoplankton bloom. In the later period of the experiment (Day 14–21), the increase in the feeding activity and resultant decrease in phytoplankton biomass induced the accumulation of dissolved organic nitrogen (DON) and ammonium. Among total growth of MZ (81.6 mmol N m⁻²), 89% (72.8 mmol N m⁻²) was transported to the depth by the ontogenetic downward migration of N. plumchrus. Although silicic acid drawdown was not increased by the IEs, Si export flux increased by 2.7 times. The increase in Si export was also due to the increase in MZ, which egested faecal pellets with higher Si:N ratio and faster sinking speed than diatoms. The export efficiency (78% of new production) and total amount of export flux (143.8 mmol N m⁻², 1392 mmol C m⁻²) were highest records within the IE experiments despite weak responses of nutrient drawdown to the IE. During SEEDS-II, the high biomass of MZ reduced the phytoplankton response and nutrient drawdown to the IEs but via grazing and ontogenetic vertical migration accelerated the export flux as well as accumulations of dissolved forms of N. Results of the present and previous IE experiments indicate that the ecosystem and biogeochemical responses to IEs in the HNLC region are quite sensitive to the ecosystem components, especially for grazers of diatoms such as copepods and heterotrophic dinoflagellates. More attention needs to be paid to the ecosystem components and their biogeochemical functions as well as physical and chemical properties of the ecosystems in order to hindcast or forecast the impacts of changes in atmospheric iron deposition.     

Dynamics and elemental stoichiometry of carbon,nitrogen, and phosphorus in particulate and dissolved organic pools during a phytoplankton bloom induced by in situ iron enrichment in the western subarctic Pacific (SEEDS-II)

The dynamics of organic carbon (C), nitrogen (N), and phosphorus (P) were examined during an in situ mesoscale iron-enrichment experiment in the western North Pacific in the summer of 2004. We separately determined the production of particulate organic matter (POM) and dissolved organic matter (DOM) and their subsequent removal during the bloom decline. As the iron-induced phytoplankton bloom progressed (days 0–14), POM increased in the surface mixed layer, while DOM did not increase significantly. The molar ratios for C:N, C:P, and N:P of the newly produced POM were estimated to be 4.9, 190, and 37 in the surface mixed layer, whereas the dissolved inorganic nitrogen to soluble reactive phosphorus drawdown ratio was 17. Preferential remineralization of P over C and N from the POM was postulated during the developing phytoplankton bloom. During the bloom decline (days 16–25), surface POM decreased with a similar C:N of 5.2. The N:P ratio of surface DOM increased during the bloom decline. Below the surface mixed layer, DOC and DON increased moderately after the peak of the bloom. The time-series variation of DOC and DON was not identical. The C, N, and P dynamics through the accumulation and removal of POM and DOM were complex. Grazing by mesozooplankton during the experiment may have played a significant role in the uncoupling of the dynamics of C, N, and P.     

Population dynamics of phytoplankton, heterotrophic bacteria, and viruses during the spring bloom in the western subarctic Pacific

We characterized the community composition of phytoplankton in the western subarctic Pacific from the pre-bloom to the decline phase of the spring bloom with special reference to decreases in the silicic acid concentration in surface waters as an index for diatom bloom development. Furthermore, responses of heterotrophic bacteria and viruses to the spring bloom were also concomitantly investigated. Under pre-bloom conditions when nutrients were abundant but the surface mixed layer depth was relatively deep, chlorophyll (Chl) a concentrations were consistently low and green algae (chlorophytes and prasinophytes), cryptophytes, and diatoms were predominant in the phytoplankton assemblages as estimated by algal pigment signatures. Together with the shallowing of the mixed layer depth and the decrease in silicic acid concentration, diatoms bloomed remarkably in the Oyashio region, though the magnitude of the bloom in the Kuroshio-Oyashio transition (hereafter Transition) region was relatively small. A total of 77 diatom species were identified, with the bloom-forming diatoms mainly consisting of Thalassiosira, Chaetoceros, and Fragilariopsis species. It has become evident that the carotenoid fucoxanthin can serve as a strong indicator of the diatom carbon biomass during the spring diatom bloom. Differences in the species richness of diatoms among stations generally enabled us to separate the Oyashio bloom stations from the Transition and the Oyashio pre-bloom stations. Relatively high values of the Shannon-Wiener index for the diatom species were also maintained during the Oyashio bloom, indicating that a wide variety of species then shared dominance. In the decline phase of the Oyashio bloom when surface nutrient concentrations decreased, senescent diatom cells increased, as inferred from the levels of chlorophyllide a. Although the cell density of heterotrophic bacteria changed little with the development of the diatom bloom, viral abundance increased toward the end of the bloom, suggesting an increased likelihood of mortality among organisms including diatoms resulting from viral infection. This is the first report on the microbial trophodynamics, including viruses, during the spring diatom bloom in the western subarctic Pacific.     

Responses of pico- and nanophytoplankton to artificial iron infusions observed during the second iron enrichment experiment in the western subarctic Pacific (SEEDS II)

How pico- and nanophytoplankton responded to artificial iron infusions was investigated using flow cytometry during SEEDS II, the second mesoscale in situ iron enrichment experiment in the western subarctic North Pacific. Two iron infusions on days 0 and 7 caused a remarkable increase in cellular chlorophyll fluorescence and cell size of all the four phytoplankton groups investigated: Synechococcus, cryptophytes, picoeucaryotes and nanoeucaryotes other than cryptophytes. After the second infusion, the abundance of three phytoplankton groups, excluding Synechococcus, also started to increase. After surface dissolved iron concentration decreased to <0.2 nM on day 11, chlorophyll fluorescence of all the four groups returned to a level observed before the iron infusions, suggesting that pico- and nanophytoplankton were physiologically stressed by iron deficiency. Cell concentrations of pico- and nanoeucaryotes decreased to the pre-infusion level by day 23, while that of cryptophytes remained high until day 25. Flow cytometric diagnosis showed that cryptophytes were physiologically limited during this period and effective iron uptake from suspended particles, as reported for freshwater cryptophytes, was not observed. Thus their prosperity may have been due to alleviation from grazing. Cell concentration of Synechococcus started to increase at a net specific growth rate of 0.13 d⁻¹ after day 12, reaching more than 6 times that of the pre-infusion level on day 24. This may have been due to the elevation of surface water temperature observed during the survey period, together with trophic cascading effects of increased copepod grazing.     

Changes in iron concentrations and bio-availability during an open-ocean mesoscale iron enrichment in the western subarctic Pacific,SEEDS II

A patch of water in the western subarctic gyre (low iron concentration, <0.02 nM) was fertilized twice with 322 and 159 kg of iron to induce a phytoplankton bloom. In order to understand the changes in iron distribution and bio-availability throughout the evolution and termination phase of the iron-induced bloom, iron concentrations were measured at stations inside and outside of the iron-fertilized patch, and shipboard culture experiments using iron and desferrioxamine B (DFB) inoculation to regulate iron availability were conducted 5 times with water collected from the center of the iron-fertilized patch on D2, D7, D11, D17 and D23.After the iron fertilization, we observed a significant increase in dissolved iron (1.38 nM at 5 m depth) at the center of the patch (D1). Dissolved iron concentrations subsequently decreased to an ambient level (~0.08 nM) on D16–D17, despite the second iron fertilization made on D6. During the 4-day incubations of the shipboard culture experiments, excess DFB-inoculated treatment inhibited the phytoplankton growth compared to the controls for D2, D7 and D11 patch water. This indicated that available iron existed in the iron-fertilized patch at least until D11. Moreover, iron-inoculated treatments induced growth of large-sized phytoplankton with an accompanying silicate decrease for D7, D11 and D17 patch water, but not for D23 patch water. These results indicated that large diatoms, which can respond to additional iron inoculation, existed in the iron-fertilized patch in evolution and early termination phase of the iron-induced bloom (at least until D17); however, there was no significant amount of large diatoms, which could rapidly respond to iron, in late termination phase (D23) of the iron-induced phytoplankton bloom.     

Effect of iron enrichment on the dynamics of transparent exopolymer particles in the western subarctic Pacific

Dynamics of transparent exopolymer particles (TEP) was studied during the first in situ iron-enrichment experiment conducted in the western subarctic Pacific in July–August 2001, with the goal of evaluating the contribution of TEP to vertical flux as a result of increased primary production following iron enrichment in open ocean ecosystems. Subsequent to the enhancement of phytoplankton production, we observed increase in TEP concentration in the surface layer and sedimentation of organic matter beneath it. Vertical profiles of TEP, chlorophyll a (Chl a) and particulate organic carbon (POC) were obtained from six depths between 5 and 70 m, from a station each located inside and outside the enriched patch. TEP and total mass flux were estimated from the floating sediment traps deployed at 200 m depth. Chl a and TEP concentrations outside the patch varied from 0.2 to 1.9 μg L⁻¹ and 40–60 μg XG equiv. L⁻¹, respectively. Inside the patch, Chl a increased drastically from day 7 reaching the peak of 19.2 μg L⁻¹ on day 13, which coincided with the TEP peak of 189 μg XG equiv. L⁻¹. TEP flux in the sediment trap increased from 41 to 88 mg XG equiv. m⁻² d⁻¹, with 8–14% contribution of TEP to total mass flux. This forms the basic data set on ambient concentrations of TEP in the western subarctic Pacific, and evaluation of the effect of iron enrichment on TEP.     

Mesozooplankton grazing manipulations during in vitro iron enrichment studies in the NE subarctic Pacific

IronEx I demonstrated a rapid and marked response by grazers to Fe-induced increases in phytoplankton stocks, which was thought to be due, in part, to arrested vertical migration by mesozooplankton. These observations prompted an investigation of the relative roles of Fe enrichment and grazing pressure in controlling the magnitude of phytoplankton stocks in the NE subarctic Pacific. The grazing impact of increased mesozooplankton abundance in response to a localised Fe-induced enhancement of algal biomass was simulated by performing in vitro (6 d) grazer perturbation experiments in May 1994 and September 1995 at Ocean Station Papa (OSP), when pelagic mesozooplankton stocks are usually at their annual maximum and submaximal, respectively. Manipulations were designed to increase mesozooplankton stocks in 25L carboys after various lag-times corresponding to grazing pressure greater or equal to that in situ, and to monitor changes in chlorophyll a levels as a proxy for grazing pressure. At the onset of the experiments, in vitro mesozooplankton abundances were comparable to those in situ. Despite the addition of mesozooplankton to selected Fe-enriched carboys in May after 24, 48 and 72 h, corresponding to ca. two-fold increases in their abundances, chlorophyll a increased to ca. 2 μg l⁻¹ in all treatments. In September, chlorophyll a levels increased five-fold to 2 μg l⁻¹ after 4 days – but little thereafter – in the presence of up to ten-fold higher animal abundances (added at t=0) than observed in situ. Thus, Fe-induced increases in diatom growth rates were sufficiently high to escape both initial and additional grazing pressure. If and when Fe is supplied to this region, it is unlikely that mesozooplankton can respond and graze down the resulting elevated algal abundance. Theoretical calculations, based on algal growth and grazing rate data from May in this study, suggested that a greater than five-fold increase in mesozooplankton abundance, after a 48-h lag, is required to exert sufficient grazing pressure to prevent Fe-mediated increases in algal biomass. These findings are discussed in relation to the scale dependency of such events, and the pelagic ecology of other High Nitrate Low Chlorophyll regions.     

Temporal variation in the structure of autotrophic and heterotrophic communities in the subarctic Pacific

Autotrophic and microheterotrophic plankton populations were monitored in the euphotic zone of the eastern subarctic Pacific during 6 one-month cruises in spring and summer, 1984, 1987 and 1988. Transmitted light, epifluorescence, and electron microscopy were used to identify, enumerate and estimate the biomass of size-populations of species. The 2–10μm size class dominated the biomass of both autotrophs and heterotrophs. The autotrophic flagellate, Phaeocystis pouchetii, was frequently observed in its non-colonial phase. Temporal variation in all the stocks was evident and could be explained only partially by the physical, chemical or biological factors investigated here. The general structure of the autotrophic community was similar to that in the North Atlantic, but major, unexplained variations between cruises occurred. Variation in mixed-layer depth and day length (but not variation in daily insolation) explained 25% of the variation in autotrophic doubling rate. Heterotrophic biomass comprised, in decreasing order of importance, non-pigmented flagellates, dinoflagellates, and ciliates. Ciliates rarely contributed more than 40% to the total. Microheterotrophic biomass rarely exceeded 30μg C 1⁻¹ (avg 15μg C 1⁻¹, 0–60m) whereas autotrophic biomass averaged 20μg C 1⁻¹, 0–60m, and reached 74μg C 1⁻¹ on one occasion, yet the grazing capacity of these microheterotrophs averaged 100% of primary production.     

Responses of phytoplankton and heterotrophic bacteria in the northwest subarctic Pacific to in situ iron fertilization as estimated by HPLC pigment analysis and flow cytometry

To verify the hypothesis that the growth of phytoplankton in the Western Subarctic Gyre (WSG), which is located in the northwest subarctic Pacific, is suppressed by low iron (Fe) availability, an in situ Fe fertilization experiment was carried out in the summer of 2001. Changes over time in the abundance and community structure of phytoplankton were examined inside and outside an Fe patch using phytoplankton pigment markers analyzed by high-performance liquid chromatography (HPLC) and flow cytometry (FCM). In addition, the abundance of heterotrophic bacteria was also investigated by FCM. The chlorophyll a concentration was initially ca. 0.9 μg l⁻¹ in the surface mixed layer where diatoms and chlorophyll b-containing green algae (prasinophytes and chlorophytes) were predominant in the chlorophyll biomass. After the iron enrichment, the chlorophyll a concentration increased up to 9.1 μg l⁻¹ in the upper 10 m inside the Fe patch on Day 13. At the same time, the concentration of fucoxanthin (a diatom marker) increased 45-fold in the Fe patch, and diatoms accounted for a maximum 69% of the chlorophyll biomass. This result was consistent with a microscopic observation showing that the diatom Chaetoceros debilis had bloomed inside the Fe patch. However, chlorophyllide a concentrations also increased in the Fe patch with time, and reached a maximum of 2.2 μg l⁻¹ at 5 m depth on Day 13, suggesting that a marked abundance of senescent algal cells existed at the end of the experiment. The concentration of peridinin (a dinoflagellate marker) also reached a maximum 24-fold, and dinoflagellates had contributed significantly (>15%) to the chlorophyll biomass inside the Fe patch by the end of the experiment. Concentrations of 19′-hexanoyloxyfucoxanthin (a prymnesiophyte marker), 19′-butanoyloxyfucoxanthin (a pelagophyte marker), and alloxanthin (a cryptophyte marker) were only incremented a few-fold increment inside the Fe patch. On the contrary, chlorophyll b concentration reduced to almost half of the initial level in the upper 10 m water column inside the Fe patch at the end of the experiment. A decrease with time in the abundance of eukaryotic ultraphytoplankton (<ca. 5 μm in size), in which chlorophyll b-containing green algae were possibly included was also observed by FCM. Overall, our results indicate that Fe supply can dramatically alter the abundance and community structure of phytoplankton in the WSG. On the other hand, cell density of heterotrophic bacteria inside the Fe patch was maximum at only ca. 1.5-fold higher than that outside the Fe patch. This indicates that heterotrophic bacteria abundance was little respondent to the Fe enrichment.     

An in situ iron-enrichment experiment in the western subarctic Pacific (SEEDS): Introduction and summary

To test the iron hypothesis in the subarctic Pacific Ocean, an in situ iron-enrichment experiment (SEEDS) was performed in the western subarctic gyre in July–August 2001. About 350 kg of iron (as acidic iron sulfate) and 0.48 mol of the inert chemical tracer sulfur hexafluoride were introduced into a 10-m deep surface mixed layer over an 80 km² area. This single iron infusion raised dissolved iron levels to 2.9 nM initially. Dissolved iron concentrations rapidly decreased after the infusion, but levels remained close to 0.15 nM even at the end of the 14-day experimental period. During SEEDS there were iron-mediated increases in chlorophyll a concentrations (up to 20 μg l⁻¹), primary production rates, biomass and photosynthetic energy conversion efficiency relative to waters outside the iron-enriched patch. The rapid and very high accumulation of phytoplankton biomass in response to the iron addition appeared to be partly attributable to shallow mixed-layer depth and moderate water temperature in the western subarctic Pacific. However, the main reason was a floristic shift to fast-growing centric diatom Chaetoceros debilis, unlike the previous iron-enrichment experiments in the equatorial Pacific and the Southern Ocean, in both of which iron stimulated the growth of pennate diatoms. The iron-mediated blooming of diatoms resulted in a marked consumption of macronutrients and drawdown of pCO₂. Biological and physiological measurements indicate that phytoplankton growth in the patch became both light- and iron-limited, making phytoplankton biomass relatively constant after day 9. The increase in microzooplankton grazing rate after day 9 also influenced the net growth rate of phytoplankton. There was no significant increase in the export flux of carbon to depth during the 14-day occupation of the experimental site. The export flux between day 4 and day 13 was estimated to be only 13% of the integrated primary production in the iron-enriched patch. The major part of the carbon fixed by the diatom bloom remained in the surface mixed layer as biogenic particulate matter. Our findings support the hypothesis that iron limits phytoplankton growth and biomass in a ‘bottom up’ manner in this area, but the fate of algal carbon remains unknown.     

Mesozooplankton responses to iron-fertilization in the western subarctic Pacific (SEEDS2001)

A mesoscale iron-fertilization experiment was carried out in the western subarctic Pacific during summer 2001. The iron-patch was traced for 14 days after the fertilization, and the abundance and behavior of mesozooplankton were compared with those outside of the patch. The phytoplankton biomass in the patch rapidly increased to over 15 times the initial level by the later half of the observation period, and was composed of large-sized (>10 mm), centric diatoms. Dominant zooplankton species in the upper 200-m depth were large copepods: Neocalanus plumchrus, Neocalanus cristatus, Eucalanus bungii and Metridia pacifica. Mesozoplankton biomass as well as species composition did not change significantly in the patch over the observation period. Furthermore, no changes of vertical distribution or diel vertical migration were observed for any species or stages of mesozooplankton throughout the observation period. However, the abundance of the first copepodite stages of N. plumchrus and E. bungii increased several fold in the patch after the diatom bloom formation compared to the densities outside the patch. The increases of both species are considered to be due to lowered mortality during the egg and nauplius stages. Spawning of N. plumchrus takes place at depth using lipid storage, while spawning of E. bungii takes place in the surface layer supported by grazing. These facts suggest that the relative importance of nauplii in the diets of the large copepods was decreased in the patch by the diatom bloom. Gut-pigment contents of dominant copepods in the patch increased 4–18 times, and the maximum values were observed during the bloom peak. However, the grazing impact on phytoplankton was low throughout the experiment, especially during the bloom period (<6% of the primary production).     

Persistence of iron limitation in the western subarctic Pacific SEEDS II mesoscale fertilization experiment

The cumulative evidence from more than a dozen mesoscale iron-enrichment studies in high nitrate low chlorophyll (HNLC) waters demonstrates that iron limitation is widespread and very likely affects atmospheric carbon dioxide and thus global climate. However, the responses of microphytoplankton (>20 μm), predominantly diatoms, vary greatly among these mesoscale experiments even though similar amounts of iron were added, making it difficult to quantitatively incorporate iron effects into global climate models. Nowhere is this difference more dramatic than between the massive bloom observed during Subarctic Pacific Iron Experiment for Ecosystem Dynamics Study (SEEDS) I and the order of magnitude smaller ecosystem response in SEEDS II; two mesocale experiments performed in the same HNLC region of the western subarctic Pacific in different years. Deckboard incubation experiments initiated during the early, middle, and late stages of the 32-day SEEDS II experiment show that while the two iron infusions increased phytoplankton growth, diatoms remained significantly limited by iron availability, despite total dissolved Fe concentrations in the patch being well above the diffusion-limited threshold for rapid diatom growth. This iron limitation was apparent <6 days after the initial iron infusion and was not alleviated by the second, smaller iron infusion. In contrast, smaller phytoplankton (<20 μm) showed a more restricted response to further iron amendments, indicating that their iron nutrition was near optimal. Iron complexed to desferrioximine B, a commonly available siderophore produced by at least one marine bacterium, was poorly available to diatoms throughout the patch evolution, indicating that these diatoms lacked the ability to induce high-affinity iron uptake systems. These results suggest that the strong organic complexation of Fe(III) observed in the SEEDS II-fertilized patch was not compatible with rapid diatom growth. In contrast, iron associated with protoporphyrin IX, a weaker iron complexing ligand of a class hypothesized to be representative of recycled iron species, was readily available to diatoms. Our findings demonstrate that a persistence of iron limitation was the primary factor underlying the comparatively small diatom response during SEEDS II. This continued growth limitation would have increased the importance of mesozooplankton grazing as a controlling factor in the SEEDS II ecosystem response.     

Seasonal changes in the mesozooplankton biomass and community structure in subarctic and subtropical time-series stations in the western North Pacific

Seasonal changes in mesozooplankton biomass and their community structures were observed at time-series stations K2 (subarctic) and S1 (subtropical) in the western North Pacific Ocean. At K2, the maximum biomass was observed during the spring when primary productivity was still low. The annual mean biomasses in the euphotic and 200- to 1000-m layers were 1.39 (day) and 2.49 (night) g C m^?2 and 4.00 (day) and 3.63 (night) g C m^?2, respectively. Mesozooplankton vertical distribution was bimodal and mesopelagic peak was observed in a 200- to 300-m layer; it mainly comprised dormant copepods. Copepods predominated in most sampling layers, but euphausiids were dominant at the surface during the night. At S1, the maximum biomass was observed during the spring and the peak timing of biomass followed those of chlorophyll a and primary productivity. The annual mean biomasses in the euphotic and 200- to 1000-m layers were 0.10 (day) and 0.21 (night) g C m^?2 and 0.47 (day) and 0.26 (night) g C m^?2, respectively. Copepods were dominant in most sampling layers, but their mean proportion was lower than that in K2. Mesozooplankton community characteristics at both sites were compared with those at other time-series stations in the North Pacific and with each other. The annual mean primary productivities and sinking POC fluxes were equivalent at both sites; however, mesozooplankton biomasses were higher at K2 than at S1. The difference of biomasses was probably caused by differences of individual carbon losses, population turnover rates, and trophic structures of communities between the two sites.     

Meso- and microzooplankton responses to an in situ iron fertilization experiment (SEEDS II) in the northwest subarctic Pacific

A mesoscale iron fertilization experiment was carried out in the western subarctic Pacific during summer 2004. The iron-patch was traced for 26 days after the enrichment, and the abundance and behavior of meso- and microzooplankton was compared with those outside of the patch. The surface chlorophyll-a concentration in the patch was high between days 10 and 13 (2.5 mg m⁻³) and decreased to the initial level after day 20. Microzooplankton grazing rates, estimated by a dilution method, was mostly balanced with phytoplankton growth rates throughout the observed period. Dominant mesozooplankton species in the upper 200 m were copepods: dominated by Eucalanus bungii, Neocalanus plumchrus and Metridia pacifica. Species composition did not change in the patch over the observation period. The copepod biomass was 3–5 times higher than in Subarctic Pacific Iron Experiment for Ecosystem Dynamics Study (SEEDS), the previous iron-enrichment experiment in the same area, before the bloom, and exponentially increased both inside and outside the patch, which was mainly brought by the development of N. plumchrus. The development rates of N. plumchrus were not significantly different between inside and outside the patch. Estimated grazing rate suggest that the copepod grazing was main cause of the low accumulation of phytoplankton biomass, and dominance of grazing-resistant organisms such as large ciliates, large diatoms and diatoms with extremely long setae. “Arrested migration” for M. pacifica and upward shift of vertical distribution by E. bungii were observed during the bloom period, even if the accumulation of phytoplankton biomass was very low compared to other iron-enrichment experiments. These results indicate that the copepod grazing shaped the food-web structure of the lower trophic levels (biomass and species composition) in SEEDS II.     

Grazing impact of the copepod community in the Oyashio region of the western subarctic Pacific Ocean

The role of copepod grazing on the ecosystem dynamics in the Oyashio region, western subarctic Pacific was investigated during six cruises from June 2001 to June 2002. In situ grazing rates of the copepod community (CGR) were measured by the gut fluorescence method in respect to developmental stages of dominant species. In terms of biomass, more than 80% of the copepod community was dominated by six large calanoid species (Neocalanus cristatus, Neocalanus flemingeri, Neocalanus plumchrus, Eucalanus bungii, Metridia pacifica and Metridia okhotensis) throughout the year. Resulting from the observed pattern of the interzonal migrating copepods, the CGR in the Oyashio region was divided into three phases, i.e. spring (bloom), summer (post-bloom) and autumn-winter phase. During the spring bloom, late copepodites of the interzonal migrating species, N. cristatus, N. flemingeri and E. bungii appeared in the surface layer (0-50 m) to consume the production of the bloom, resulting in a high grazing rate of the copepod community (7.9 mg Chl m⁻² d⁻¹), though its impact on phytoplankton community was low due to the high primary productivity. During the post-bloom period, although the copepod community which was dominated by N. cristatus, N. plumchrus, M. pacifica and newly recruited E. bungii still maintained a high biomass, the CGR was generally lower (1.8-2.6 mg Chl m⁻² d⁻¹ for June and August 2001), probably due to the lower availability of phytoplankton. Nevertheless, the highest CGR was also observed during this period (10.5 mg Chl m⁻² d⁻¹ in June 2002). The high CGR on autotrophic carbon accounted for 69% of the primary production, suggesting that the copepod community in the Oyashio region potentially terminates the phytoplankton bloom. Abundant occurrence of young E. bungii, which is a characteristic phenomenon in the Oyashio region, was largely responsible for the high grazing pressure in June 2002 suggesting that success of reproduction, growth, and survival in E. bungii during the spring bloom is an important factor in controlling phytoplankton abundance during the post-bloom season. During autumn and winter, CGR was the lowest in the year (0.29-0.38 mg Chl. m⁻² d⁻¹) due to the disappearance of the interzonal migrating copepods from the surface layer. Diel migrant M. pacifica was the most important grazer during this period. The annual ingestion of the copepod community is estimated as 37.7 gC m⁻² on autotrophic carbon (converted using C:Chl ratio of 30) or 137.9 gC m⁻² on suspended particles (using C:Chl ratio of in situ value, 58-191), accounting for 13% and 46% of annual primary production, respectively. This study confirms that copepod grazing is an important pathway in carbon flow in the Oyashio region and in particular their role in the phytoplankton dynamics is significant for the termination of the spring bloom.