Subarctic Ecosystem Response To Iron Enrichment Study Research Articles

Dynamics of particulate and dissolved organic and inorganic phosphorus during the peak and declining phase of an iron-induced phytoplankton bloom in the eastern subarctic Pacific

Abstract Phosphorus (P) is an essential element for all organisms and thus the P cycle plays a key role in determining the dynamics of lower trophic levels in marine ecosystems. P in seawater occurs conceptually in particulate and dissolved organic and inorganic (POP, PIP, DOP, and DIP, respectively) pools and clarification of the dynamics in these P pools is the basis to assess the biogeochemical cycle of P. Despite its importance, behaviors of each P pool with phytoplankton dynamics have not been fully examined. We measured the four operationally defined P pools (POPop, PIPop, DOPop, and SRP) during an iron-induced phytoplankton bloom (as part of the subarctic ecosystem response to iron enrichment study (SERIES)) in the eastern subarctic Pacific in summer 2002. During our observations of the iron-enriched patch from day 15 to day 26 after the iron infusion, chlorophyll-a concentration in the surface layer decreased from 6.3 to 1.2 μg L− 1, indicating the peak through decline phase of the phytoplankton bloom. At the bloom peak, P was partitioned into POPop, PIPop, and DOPop in proportions of 60, 27, and 13%, respectively. While chlorophyll-a and POPop showed similar temporal variations during the declining phase, PIPop showed a different peak timing with a 2 day delay compared to POPop, resulting in a rapid change in the relative proportion of PIPop to total particulate P (TPP = POPop + PIPop) at the peak (25%) and during the declining phase of the bloom (50%). A part of POPop was replaced by PIPop just after slowing down of phytoplankton growth. This process may have a significant role in the subsequent regeneration of P. We conclude that measurement of TPP alone is insufficient to show the interaction between P and phytoplankton dynamics and fractionation of TPP into POPop and PIPop provides useful insights to clarify the biogeochemical cycle of P.

The effect of vertical and horizontal dilution on fertilized patch experiments

A great deal of attention, both negative and positive, has been directed at the potential of large‐scale iron fertilization schemes to sequester carbon by inducing phytoplankton blooms that would, in theory, result in significant export of organic carbon to the deep ocean in high nitrogen ‐ low chlorophyll regions. A suite of iron manipulation or ‘patch’ experiments has been performed over length‐scales of 10s of km. Here, we use a physical‐ecological‐chemical model, with prognostic nitrogen, silica and iron dynamics, to study one of the most successful of these experiments, the Subarctic Ecosystem Response to Iron Enrichment Study (SERIES), focusing on the vertical export of organic material, which is difficult to observe in the field. The implications of large‐scale fertilization, i.e. increasing patch size, are investigated. Our results agree with the general conclusions obtained from the field experiments. Only a modest export of organic carbon occurs (less than 25% of carbon uptake by phytoplankton) at the base of the mixed layer. Furthermore, we show that lateral and vertical supply of silicic acid is necessary to fuel a sustained phytoplankton bloom. Increasing patch size results in less lateral nutrient supply relative to patch area and so a decrease, not only in total production (per unit area), but in the contribution by large phytoplankton due to silica limitation. Most importantly, the export of organic carbon (per unit area) decreases substantially, by nearly an order of magnitude, as scales of 1000 km are approached.

IDENTIFICATION AND ASSESSMENT OF DOMOIC ACID PRODUCTION IN OCEANIC PSEUDO-NITZSCHIA (BACILLARIOPHYCEAE) FROM IRON-LIMITED WATERS IN THE NORTHEAST SUBARCTIC PACIFIC(1).

We identified and investigated the potential toxicity of oceanic Pseudo-nitzschia species from Ocean Station Papa (OSP), located in a high-nitrate, low-chlorophyll (HNLC) region of the northeast (NE) subarctic Pacific Ocean. Despite their relatively low abundances in the indigenous phytoplankton assemblage, Pseudo-nitzschia species richness is high. The morphometric characteristics of five oceanic Pseudo-nitzschia isolates from at least four species are described using SEM and TEM. The species identified are Pseudo-nitzschia dolorosa Lundholm et Moestrup, P.granii Hasle, P.heimii Manguin, and P.cf. turgidula (Hust.) Hasle. Additional support for the taxonomic classifications based on frustule morphology is provided through the sequencing of the internal transcribed spacer 1 (ITS1) rDNA. Pseudo-nitzschia species identification was also assessed by the construction of ITS1 clone libraries and using automated ribosomal intergenic spacer analysis (ARISA) for environmental samples collected during the Subarctic Ecosystem Response to Iron Enrichment Study (SERIES), conducted in close proximity to OSP in July of 2002. Based on ITS1 sequences, the presence of P.granii, P.heimii, P.cf. turgidula, and at least five other putative, unidentified Pseudo-nitzschia ITS1 variants was confirmed within iron-enriched phytoplankton assemblages at OSP. None of the oceanic isolates produced detectable levels of particulate domoic acid (DA) when in prolonged stationary phase due to silicic acid starvation. The lack of detectable concentrations of DA suggests that either these strains produce very little or no toxin, or that the physiological conditions required to promote particulate DA production were not met and thus differ from their coastal, toxigenic congeners.

Parameter sensitivities in a 1-D model for DMS and sulphur cycling in the upper ocean

We have developed a marine DMS (dimethylsulfide) module and implemented it in a 1-D coupled atmosphere-ocean-biogeochemical model. In developing the marine sulphur model we have found that several parameters used in the model are not known to even an order of magnitude. Our approach is used to test the model's sensitivity to these parameters. A parameter change of ± 25 % is applied to test the respective range of changes in the DMS fluxes. The model is run for a 3-year time period as well as for the time period of the Subarctic Ecosystem Response to Iron Enrichment Study (SERIES) in July 2002. The simulated seasonal cycle is in agreement with available observations: Near surface DMS concentrations vary from 1.5 nmol L - 1 in winter to 13.5 nmol L - 1 in summer. Simulated DMS production is found to be most sensitive to variations of the S:N ratio and the bacterial consumption rate of DMS. Implementing light or UV limited bacterial activity shows a negligible effect in winter and increases DMS concentrations by 0.2– 0.6 nmol L - 1 in summer. Similarly a yield increase under UV stress increases summer values by 1– 2 nmol L - 1 . The simulated diel cycle in surface DMS concentration is no more than 2.5 nmol L - 1 , even when light-dependent changes in bacterial activity are considered. Simulating the DMS response to iron fertilization with the standard run leads to overestimation during an initial bloom of small phytoplankton. While implementing light-dependent bacterial activity has a minor effect, the implementation of yields that depend on nutrient availability significantly improves the results. The model confirms earlier results showing the importance of including atmospheric DMS concentrations in gas flux calculations when there are high surface concentrations and small atmospheric boundary layer heights. Simulated summer concentrations in the upper layer can be underestimated by 2 nmol L - 1 or more if the atmospheric concentration is set to zero. Our study shows that inclusion of mechanistic DMS modules in comprehensive climate models requires better knowledge of the variation of key parameters in the marine sulphur cycle. Even though there are still open questions, the model reasonably reproduces the mean annual cycle; and including variable DMS yield improves the simulation of the DMS response to iron fertilization during SERIES.

Simulating the coupling between atmosphere–ocean processes and the planktonic ecosystem during SERIES

Abstract We have developed a 1-D atmosphere–ocean-biogeochemical model to investigate the coupling between atmosphere–ocean exchanges and the planktonic ecosystem during the Subarctic Ecosystem Response to Iron Enrichment Study (SERIES) in 2002. The atmospheric Single Column Model (SCM) is based on the Canadian Centre for Climate Modelling and Analysis (CCCma) Atmospheric General Circulation Model (AGCM). The ocean component employs the General Ocean Turbulence Model (GOTM). A seven-component ecosystem model is embedded in GOTM, which includes nitrogen, organic and inorganic carbon, silica and oxygen cycling. We use observations from SERIES combined with atmospheric reanalysis data to initiate and force the coupled physical model. We found that atmospheric temperatures and humidities are higher and the stratification more stable if nudged to National Centre of Environmental Prediction (NCEP) rather than to European Centre for Medium-Range Weather Forecasts (ECMWF) 40-yr reanalysis data. Doubling the vertical resolution in the atmosphere improved the representation of mixing and the thermal structure, affecting cloudiness and radiative fluxes at the ocean surface as well as planetary boundary layer heights and gas dispersion in the lower atmosphere. From observed ocean-surface dimethyl sulphide (DMS) concentrations (outside the patch) we simulated DMS dispersion in the atmospheric boundary layer by applying a first-order loss term, with turnover times ranging from 1 to 4 days. During SERIES, shallow boundary-layer heights that occurred when DMS production was highest prevented dispersion into the atmosphere beyond several 100 m. Finally, successive model runs with iron fertilization starting on June 25, July 10 and 25 showed that the general nature of the response to iron enrichment at OSP (SERIES) is robust, but the strength as well as length of the response depend strongly on short-term atmospheric conditions (wind and radiative fluxes).

Carbon distribution and fluxes during the SERIES iron fertilization experiment with special reference to the fugacity of carbon dioxide (fCO2)

Abstract Surface seawater fugacity of carbon dioxide (fCO2) was measured during the Subarctic Ecosystem Response to Iron Enrichment Study (SERIES), July 9–August 5, 2002. Three ships sampled the iron-fertilized waters near Ocean Station P (50°N, 145°W): the Canadian CCGS J.P. Tully (July 9–23, 2002), the chartered Mexican M/V El Puma (July 9–28, 2002), and the Japanese fisheries research ship M/V Kaiyo Maru (July 24–August 5, 2002). Data used here are from the CCGS J.P. Tully and the M/V Kaiyo Maru. From the onset of the experiment to the peak of the iron-induced diatom bloom on day 19, sea-surface fCO2 decreased from 350 to 265 μatm and average DIC concentration in the upper 30 m decreased from 2030 to 1990 μmol kg−1. Changes in fCO2 in and near the iron patch as observed from the CCGS J.P. Tully and later from the M/V Kaiyo Maru were used to estimate CO2 drawdown and air-sea fluxes, and in generating a carbon budget during the growth phase (days 3–19) of the experiment. Without considering patch dilution, sources of dissolved inorganic carbon to the patch (1.6±0.25 mol m−2) were nearly double the sum (0.87±0.34 mol m−2) of the sinks: accumulations of dissolved organic and particulate carbon, and the flux of particulate carbon to sediment traps below the patch. However, the budget is balanced after considerations of the effects of patch expansion on property concentrations within the patch. A comparison with other iron fertilization experiments from 1995 to present was made to assess the CO2 drawdown values.

Iron speciation and dynamics during SERIES, a mesoscale iron enrichment experiment in the NE Pacific

Abstract During the Sub-arctic Ecosystem Response to Iron Enrichment Study (SERIES), the addition of ferrous iron to high-nitrate low-chlorophyll (HNLC) waters near Ocean Station PAPA (OSP: 50°N, 145°W) produced a phytoplankton bloom and CO 2 drawdown, as evidenced by decreasing CO 2 fugacity ( f CO 2 ). We analyzed five fractions or phases of iron: soluble ( 0.22 μm). Here, we describe iron distributions and the evolution of iron phases in the upper ocean during the experiment. We also present an iron budget accounting for horizontal and vertical dilution. At the time of our first sampling eight hours after fertilization was completed, total iron reached 8.6 nmol L −1 and dissolved iron was approximately 3 nmol L −1 . Early in the experiment the dissolved iron phase decreased the most rapidly and by late day 6 the integrated dissolved iron (8.6 μmol m −2 ) represented less than 10% of the initial addition (90–95 μmol m −2 ). However at this same time the total integrated iron at the centre of the patch was still 52 μmol m −2 or almost 60% of the calculated initial addition. By day 12,45% of the added iron (from both injections) could be accounted for in the patch. The half-life of total iron in the patch for the first injection was estimated to be less than 5 days if dilution is not considered, but more than 13 days if dilution is taken into account. The most notable change in iron percentages from one form to another occurred early in the first week of the experiment where the predominant phase shift was from the colloidal portion of dissolved iron to labile particulate iron that could have been biologically induced or simply aggregation of oxyhydroxides. This was immediately followed by a physical event resulting in a reduction in the non-labile particulate iron due to sinking out of the patch. The second infusion did not change the relative concentration of the various pools of iron as might be expected, but this was likely due to the fact that it was a much smaller injection than the first. The most pronounced change after the second infusion was the reduction in the labile particulate pool which coincided with one of the largest decreases in silicate observed during the entire experiment. In general the gradual decrease in the fraction of the 10 m colloidal iron as well as episodic losses of, or shifts in, integrated colloidal iron are thought to be the result of adsorption of colloidal iron to the plankton cell surfaces as well as aggregation of oxyhydroxides but could also be the result of utilization of colloidal iron by mixotrophic phytoplankton.

Microbial response to a mesoscale iron enrichment in the NE subarctic Pacific: Heterotrophic bacterial processes

The response of heterotrophic bacteria to an in situ mesoscale Fe-addition was characterized during the Subarctic Ecosystem Response to Iron Enrichment Study (SERIES), in the high nutrient low chlorophyll region of the Northeast subarctic Pacific, during July 2002. Samples were collected from inside and outside the Fe-enriched patch for the determination of bacterial biomass, and rates of production and growth, and community respiration. The addition of Fe significantly changed the dynamics of the mixed layer heterotrophic bacterial community compared to unfertilized waters. Outside the patch, bacterial dynamics remained relatively constant. Inside the Fe-enriched patch, depth-integrated bacterial biomass decreased 5-fold during the first 12 days after fertilization, after which biomass increased more than 10-fold, to a maximum of 23.3 mg C m−3. Similarly, bacterial production decreased 3-fold over the first 8 days, followed by a 15-fold increase to 5.7 mg C m−3 d−1. Bacterial specific growth rates remained constant for 8 days after the initial Fe-addition and close to values initially observed outside the patch. After day 8, mixed layer specific growth rates inside the patch increased more than 10-fold to a maximum of 1.24 d−1 by day 12, then steadily decreased to 0.22 d−1 by day 16 and remained relatively constant thereafter. Temporal changes in growth were not significantly different inside and outside the patch, suggesting that bacterial growth was not directly limited by Fe availability. The temporal uncoupling of bacterial biomass and production inside the patch, combined with the lack of evidence for direct iron limitation, suggest that inside the patch, bacteria were initially controlled by a combination of moderate bottom-up control, due to the effects of organic substrate limitation of bacterial growth, and strong top-down control, by processes such as microzooplankton bacterivory or viral lysis. Release of bacteria from grazing pressure (around day 12), coupled with an increase in specific growth rate (day 8), resulted in the rapid increase in bacterial biomass observed towards the end of the observation period. Mixed layer bacterial carbon demand ranged from 1.5 to 22.9 mg C m−3 inside the patch and accounted for an average (±st dev) of 25% (±11%) of primary production. Consequently, a high proportion of the Fe-enhanced primary production in the mixed layer during SERIES was channelled through the microbial food web, thus reducing the amount of organic carbon available for export.

Microbial response to a mesoscale iron enrichment in the NE Subarctic Pacific: Bacterial community composition

Changes in microbial community composition were determined during the subarctic ecosystem response to iron enrichment study (SERIES), a mesoscale Fe enrichment conducted in a high-nutrient low-chlorophyll (HNLC) region of the Northeast Subarctic Pacific, in July 2002. Phylogenetic composition using fluorescence in situ hybridization (FISH), relative DNA content using flow cytometry (FCM), and cellular morphometrics (shape and volume) of heterotrophic bacteria were used to characterize community composition from samples collected within and below the mixed layer, inside and outside the Fe-patch. The proportion of total cells detected as members of the Cytophaga-Flavobacterium cluster increased in a log-linear manner from 16 (±1.0)% to 47 (±1.9)% in samples within the mixed layer, inside the Fe-enriched patch, while outside the patch, the proportion remained ⩽21 (±2.2)%. Temporal changes in the proportion of cells in the mixed layer with high DNA content (% HDNA) were significantly different inside and outside the Fe-enriched patch, where inside the patch % HDNA increased 2-fold after a week, reaching 93% towards the end of the observation period. Coupling in situ observations with the results of manipulation experiments allowed us to determine the relative contributions of bottom-up (nutrient limitation) and top-down (grazing) processes on heterotrophic bacterial abundance and community composition. Shifts in heterotrophic bacterial community composition inside the Fe-enriched patch were mainly controlled by top-down processes and moderately controlled by bottom-up controls (organic substrate limitation).

SERIES (subarctic ecosystem response to iron enrichment study): A Canadian–Japanese contribution to our understanding of the iron–ocean–climate connection

SERIES (subarctic ecosystem response to iron enrichment study): A Canadian–Japanese contribution to our understanding of the iron–ocean–climate connection

The effects of patch expansion on budgets of C, N and Si for the Subarctic Ecosystem Response to Iron Enrichment Study (SERIES)

Abstract Iron enrichment was carried out near Ocean Station Papa (OSP; 50°N, 145°W) in the subarctic northeast Pacific Ocean in July 2002, to assess the biogeochemical response to iron availability in these high-nitrate, low-chlorophyll (HNLC) waters. The phytoplankton responded immediately to iron addition. At day 0, flagellates were predominant and reached maximum abundance (days 6–9) before diatoms (days 16–19). At days 15–16, the iron-enriched patch shifted from a northward to an eastward direction and, from day 19.5 onward, mixing increased salinity and nutrient concentrations within the patch. Also at day 19.5, the iron-induced diatom bloom began to decline. We present budgets of C, N and Si for the periods of bloom development (days 3.5–19.5) and decline (days 19.5–26.5). A model that corrects for the dilution caused by patch expansion improves budgets for bloom development, but it does not account for the mixing that occurred during bloom decline. However, the model generates realistic rates of nutrient uptake during bloom decline if waters at the pycnocline were the source of salt and nutrients into the patch and if estimates of patch expansion were artificially low towards the end of the experiment, a possibility due to weakened tracer signals at patch boundaries.

Assessing microbial responses to iron enrichment in the Subarctic Northeast Pacific: Do microcosms reproduce the in situ condition?

Abstract A microcosm experiment was conducted in the NE Pacific in July 2002 to compare the microbial response between microcosms and the Subarctic Ecosystem Response to Iron-Enrichment Study (SERIES) in situ iron-enrichment experiment. Seawater microcosms (20 L) were incubated aboard ship under natural light using three treatments: (1) low-iron seawater amended with 4 nmol l−1 FeSO4 (+Fe); (2) low-iron seawater amended with 4 nmol l−1 FeSO4 and 86 nmol l−1 GeO2 (+Fe+Ge); (3) seawater collected from the in situ Fe-enriched patch (PW). The +Fe+Ge treatment used germanium to control diatom growth to assess the role of diatoms in dimethylsulfoniopropionate (DMSP) production. The following variables were measured in the microcosms and in situ: chlorophyll a (chl a), nitrate ( NO 3 - ), silicic acid (Si(OH)4), phytoplankton abundance and species identification, bacterial abundance (including estimates of low- and high-DNA bacteria), bacterial production, bacterial specific growth rate, particulate and dissolved DMSP and dimethylsulfide (DMS) concentrations. There was little or no significant difference (ANCOVA) in the response of most variables between the +Fe and PW microcosms, but large differences were observed between both these treatments and the in situ data from the enriched patch. Chl a in all microcosms increased from ambient levels (approx. 0.5–1 μg l−1) to approx. 4.5–6.2 μg l−1 after 11 d incubation, when NO 3 - was fully depleted from all microcosms. During this same period, in situ chl a increased more slowly to a maximum of 2.9 μg l−1 on day 11. Nanophytoplankton and picophytoplankton were more abundant in the microcosms relative to the in situ community, which became dominated by large diatoms. Bacterial abundance was similar in the microcosms and in situ, but bacterial production was significantly higher in the microcosms. While neither DMSPd nor DMS accumulation showed significant differences between the microcosms and in situ, particulate DMSP concentrations increased significantly faster in the +Fe and PW treatments. These differences represent bottle effects resulting from the containment of the microcosms, which suppresses grazing, alters community and food web structure, enhances iron and nutrient regeneration, and isolates the community from physical transport and export processes including sinking. Thus during this experiment, the microcosms were not a good model for the in situ system in terms of the effects of iron on the phytoplankton biomass, nutrient uptake, bacterial dynamics and DMSPp production. In the germanium-amended treatment, the inhibition of diatom growth resulted in enhanced growth of other taxa and a suppression of bacterial production, leading to increased production of DMSP and DMS and strong correlations between DMSP, DMS and non-diatom phytoplankton taxa. Diatoms did not contribute significantly to particulate DMSP concentrations.

Phytoplankton processes during a mesoscale iron enrichment in the NE subarctic Pacific: Part I—Biomass and assemblage

We report results from the Subarctic Ecosystem Response to Iron Enrichment Study (SERIES) experiment in waters of the NE subarctic Pacific in which a large scale iron (Fe) enrichment lead to a shift in the phytoplankton assemblage from pico- and nanophytoplankton to one dominated by large diatoms. The phytoplankton response to the added Fe was monitored for 26 days following two infusions into a 77 km 2 patch of seawater. During the course of the experiment, the resulting algal bloom was constrained within the upper 30 m and spread to a region measuring over 1000 km 2 . Phytoplankton chlorophyll a (chl a) increased from 0.3 mg m � 3 to a peak of 6.3 mg m � 3 18 days after the initial addition of Fe. Water-column integrated chl a was enhanced 8-fold, reaching a maximum of 114 mg m � 2 on day 17. The resulting bloom is described in two ecological phases based on dominant phytoplankton groups. In Phase I, which encompassed the initial infusion up to day 10, all size-fractions (0.2–2, 2–20 and 420mm) increased in biomass as indicated by chl a, contributing to a surface standing stock of 2 mg m � 3 .I n Phase II, from days 10 to 18, the bloom was dominated by microphytoplankton (420mm), with a concomitant decrease in phytoplankton o20mm. Microphytoplankton, which initially accounted for 25% of the phytoplankton biomass and increased by a factor of 50, consisted primarily of the pennate diatom genera, Pseudo-nitzschia, Neodenticula and Thalassiothrix and the centric diatom genera, Chaetoceros, Rhizosolenia, and Proboscia. Particulate carbon-to-chl a (PC: chl a) ratios for large cells (X5mm) decreased 5-fold by day 18, indicative of enhanced cellular chl a content and increased phytoplankton contributions to PC. Pennate diatoms were most abundant in the patch, although when converted to biovolume, centric diatoms contributed larger amounts of algal carbon (C) to the bloom. A rapid decline in chl a on day 19 marked the onset of bloom decline. The magnitude, duration and composition of the phytoplankton response to the Fe enrichment clearly depicted a major shift in the structure of the algal assemblage and increased C export potential. r 2006 Elsevier Ltd. All rights reserved.

Stable nitrogen isotope dynamics of a mesoscale iron enrichment experiment in the NE Subarctic Pacific

We report the response in the natural abundance of the stable isotopes of nitrogen (δ15N) during the Fe-enrichment experiment SERIES (Subarctic Ecosystem Response to Iron Enrichment Study) in the NE Pacific. Samples were collected for isotope analysis of nitrate, particulate material (including size fractionated samples), and particles trapped in the water column from beneath the Fe-enriched patch. In all sample types, δ15N changed in response to increased phytoplankton productivity after the Fe enrichment. The nitrate concentration and δ15N of nitrate were inversely related, the result of the opposing effects of isotope fractionation during nitrate assimilation and the addition of new nitrate by periodic mixing of water from outside the Fe patch. During the growth period a decrease in the difference of the δ15N of particulate nitrogen and nitrate occurred that was attributed to physical mixing, shifts in growth from regenerated nitrogen sources to nitrate, and the change in the community assemblage from<5-μm phytoplankton cells to a larger assemblage dominated by diatoms. The surface-tethered sediment trap δ15N samples indicate that the nitrate isotope fractionation signal in surface waters was not transported below the permanent mixed layer until the end of the phytoplankton growth period, and therefore only the highest values associated with the isotope fractionation process were recorded in the sinking material from the patch. An important conclusion from this study is that mesoscale physical mixing effects and nitrogen remineralization can reduce the expression of isotope fractionation during phytoplankton growth, explaining why the high fractionation values measured in laboratory studies are not commonly observed in the natural environment.

DMSP and DMS dynamics during a mesoscale iron fertilization experiment in the Northeast Pacific—Part I: Temporal and vertical distributions

This paper reports on the influence of the Fe fertilization conducted during the subarctic ecosystem response to iron enrichment study (SERIES) on the distribution of the biogenic sulfur compounds dimethylsulfoniopropionate (DMSP) and dimethylsulfide (DMS) in the context of changes in plankton composition. The Fe enrichment resulted in a rapid increase in the abundance of a nanoplankton assemblage dominated by Prymnesiophyceae, Prasinophyceae, small diatoms (<5 μm), heterotrophic dinoflagellates, and zooflagellates. This first assemblage persisted for 8 days before collapsing abruptly due to an increase in microzooplankton herbivory. The abundance of large diatoms started to increase shortly after the initial Fe fertilization but peaked 1–2 days after the crash of the nanoplankton bloom. Inside the Fe patch, particulate DMSP (DMSPp) increased from 100 to 285 nmol L−1 during the nanoplankton bloom, decreased rapidly back to initial level as this bloom collapsed, and remained low during the bloom of large diatoms. Outside the patch, phytoplankton and protists abundance and DMSPp concentrations remained low and relatively stable throughout the experiment. DMS concentrations were elevated at the onset of the experiment outside the patch (maximum of 15.7 nmol L−1 on day 1), increased up to 26.5 nmol L−1 10 days after the enrichment, and decreased to ca. 6 nmol L−1 by the end of the experiment. This large natural pulse in DMS coincided with conditions of high irradiance and decreasing wind speed. Inside the Fe patch, DMS concentrations exhibited the same general pattern, but with distinctive features related to the Fe fertilization. First, DMS concentrations tended to increase more rapidly inside the patch during the initial nanoplankton bloom, leading to DMS concentrations ca. 2 times higher inside the patch than outside on day 6. Second, DMS concentrations became consistently lower inside the patch (often below our limit of quantification of 0.03 nmol L−1) than outside (ca. 6 nmol L−1) during the peak of the diatom bloom. Our results thus confirm the rapid increase in nanoplankton and DMSPp reported during all previous Fe-fertilization experiments. On the other hand, the decrease in DMS concentrations measured inside the Fe patch during SERIES is unique and shows that adding Fe to HNLC waters may not always lead to conditions that could mitigate climate warming.

Modeling analysis of the effect of iron enrichment on dimethyl sulfide dynamics in the NE Pacific (SERIES experiment)

The large‐scale iron enrichment conducted in the NE Pacific during the Subarctic Ecosystem Response to Iron Enrichment Study (SERIES) triggered a phytoplankton bloom dominated successively by nanophytoplankton and large diatoms. During the first 14 days, surface dimethyl sulfide (DMS) levels increased both inside (up to 22 nmol L−1) and outside (up to 19 nmol L−1) the patch, with no consistent Fe effect. Later, DMS concentrations became sixfold lower inside the patch than outside. In this study, we used a DMS budget module embedded in a one‐dimensional ocean turbulence model to investigate the contribution of the interacting physical, photochemical, and biological processes to this particular DMS response. Temporal variations in biological net DMS production were reconstructed using an inverse modeling approach. Our results show that short‐term (days) variations in both the physical processes (i.e., turbulent mixing and ventilation) and the biological cycling of DMS are needed to explain the time evolution of DMS concentrations both outside and inside the Fe‐enriched patch. The biological net DMS production was generally high (up to 0.35 nmol L−1 h−1) and comparable outside and inside the patch during the first 10 days, corresponding to the observed accumulation of DMS inside and outside the patch. Later, it became negative (net DMS biological consumption) inside the patch, suggesting a change in dimethylsulfoniopropionate bacterial metabolism. This study stresses the importance of short‐term variations in biological processes and their sensitivity to the physical environment in shaping the DMS response to iron enrichment.