Sequestration efficiency in the iron-limited North Atlantic: implications for iron supply mode to fertilized blooms

Sequestration efficiency in the iron-limited North Atlantic: Implications for iron supply mode to fertilized blooms Frédéric A. C. Le Moigne 1 , C. Mark Moore 2 , Richard J. Sanders 1 , Maria Villa-Alfageme 3 , Sebastian Steigenberger 2 , and Eric P. Achterberg 2,4 1 National Oceanography Centre, Southampton, UK, 2 Univerity of Southampton, Southampton, UK, 3 Departmento Físíca Aplicada II, Universidad de Sevilla, Spain, 4 GEOMAR, Helmholtz Centre for Ocean Research Kiel, Germany Abstract Estimates of the amount of carbon sequestered in the ocean interior per unit iron (Fe) supplied, as quantified by the sequestration efficiency (C effx ), vary widely. Such variability in C effx has frequently been attributed to estimate uncertainty rather than intrinsic variability. Here we derive new estimates of C effx for the subpolar North Atlantic, where Fe stressed conditions have recently been demonstrated. Derived values of C effx from across the region, including areas subject to atypical external Fe fertilization events during the year of sample collection (2010), ranged from 17 to 19 kmol C (mol Fe 1 ). Comparing these estimates with values from other systems, considered in the context of variable bloom durations in the different oceanographic settings, we suggest that apparent variability in C effx may be related to the mode of Fe delivery. 1. Introduction Iron (Fe) availability has been shown to control phytoplankton growth in the so-called high nitrate low chlorophyll (HNLC) regions [Blain et al., 2007; Boyd et al., 2004; de Baar et al., 2005; Pollard et al., 2009; Smetacek et al., 2012] including the Southern Ocean, equatorial Pacific, and subpolar North Pacific. More recently, some oceanic regions characterized by marked seasonal chlorophyll peaks (blooms), including the Irminger and Iceland Basins (hereafter IRB and IB, respectively) of the high latitude North Atlantic (HLNA), [Nielsdottir et al., 2009; Ryan-Keogh et al., 2013; Sanders et al., 2005] have also been suggested to experience a degree of Fe stress. Within all these regions, Fe deficiency potentially contributes to incomplete utilization of surface macronutrients. Subsequent subduction of these unused macronutrients in regions of deep water formation, including the Southern Ocean or the HLNA, represents an inefficiency in the biological carbon pump [Marinov et al., 2008; Nielsdottir et al., 2009; Sarmiento and Orr, 1991]. Consequently, oceanic Fe availability has been invoked as a potential control on atmospheric CO 2 [Martin et al., 1990], with for example, higher aerosol deposition to the glacial Southern Ocean hypothesized to have partly contributed to glacial/interglacial CO 2 cycles [Jickells et al., 2005; Martinez-Garcia et al., 2014; Ridgwell and Watson, 2002]. Quantitative understanding of linkages between variability in external Fe inputs and carbon cycling in different oceanic settings requires an understanding of the relationship between Fe supply and carbon sequestration, as encapsulated in the sequestration efficiency (C effx ). To date, estimates of C effx (defined as the ratio of carbon exported per unit of Fe supplied) from field programs have varied widely (~1.2 to 154 kmol C (mol Fe) 1 )[Boyd et al., 2007; de Baar et al., 2005; Morris and Charette, 2013]. The mode of Fe supply has previously been suggested as a mechanistic driver of such variability in C effx [Boyd et al., 2007; Chever et al., 2010]. However, in the absence of any observed systematic basis for the reported >2 order of magnitude range in C effx , these differences may also reflect uncertainties in calculations [Morris and Charette, 2013], including incomplete accounting for Fe sources or differing approaches for estimating C export. With the exception of values derived for Fe fertilized blooms around (sub-) Antarctic island systems, such as reported for the KErguelen Ocean and Plateau compared Study (KEOPS) and Crozet natural iron bloom and export experiment (CROZEX) studies [Blain et al., 2007; Pollard et al., 2009], no estimates of C effx have yet been derived for highly productive natural oceanic systems. Here we report new estimates for C effx for the HLNA and attempt to synthesize these into a growing understanding of potential controls on this variable in the global ocean. LE MOIGNE ET AL. ©2014. The Authors. 4619 PUBLICATION S Geophysical Research Letters RESEARCH LETTER 10.1002/2014GL060308 Key Points: •Sequestration efficiency from North Atlantic •Sequestration efficiency may be related to the mode Fe supply •Mode of Fe supply needs to be considered in models Supporting Information: •Readme •Le_Moigne_et_al_GRL_sup_June12. docx Correspondence to: F. A. C. Le Moigne, [email protected] Citation: LeMoigne,F.A.C.,C.M.Moore, R. J. Sanders, M. Villa-Alfageme, S. Steigenberger, and E. P. Achterberg (2014), Sequestration efficiency in the iron-limited North Atlantic: Implications for iron supply mode to fertilized blooms, Geophys. Res. Lett.,41, 4619–4627, doi:10.1002/2014GL060308. Received 23 APR 2014 Accepted 18 JUN 2014 Accepted article online 20 JUN 2014 Published online 8 JUL 2014 This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited. 2. Methods for Ancillary Data and Export Fluxes Sampling took place from 4 July to 10 August 2010, on board the RRS Discovery cruise D354 as part of the Irminger Basin Iron Study (IBIS) program. Inorganic nutrients and particulate organic carbon (POC) and nitrogen (PON) were sampled and analyzed as previously described [Le Moigne et al., 2013b; Sanders and Jickells, 2000]. POC and PON export fluxes were subsequently calculated using the 234 Th “small-volume” technique [Pike et al.,2005].Briefly, vertical profiles of 234 Th activity (integrated to a depth of 100 m) were converted to estimates of downward 234 Th flux using a one dimensional steady-state model [Buesseler et al., 1992]. Extraction efficiencies for 234 Th were 90.6 ± 6.7% [Le Moigne et al., 2013a, 2012]. Observed 234 Th:POC and 234 Th:PON ratios for large (>53 μm) particles collected using in situ Stand Alone Pumping Systems (SAPS) deployed for 1.5 h at a single depth beneath the mixed layer were then used to convert 234 Th fluxes to POC/N fluxes. Approximately 1500–2000 L of seawater was filtered using 53 μmmeshfilters (Nitex), with swimmers manually removed following filtration. Particles were then rinsed off the filters using Th-free seawater as prepared following [Le Moigne et al., 2013b], and the particle suspensions were splitandanalyzedfor 234 Th, POC, and PON as described in Le Moigne et al. [2013b]. Maiti et al. [2012] suggested that high flow rates may lead to particle disintegration. The pump rates we used were about two times larger than those used by Maiti et al. [2012];however,thesurfaceareaofourfilters was around 4 times larger [Le Moigne et al., 2013b]. Therefore, the velocity of seawater through the filter was around half of that used by Maiti et al. [2012]. Dissolved iron was determined following procedures reported by Painter et al. [2014]. 3. Results and Discussion 3.1. The High Latitude North Atlantic in 2010 Biogeochemical cycles in the HLNA were perturbed during 2010 as a result of aerosol Fe deposition to the IB from the eruption of the Icelandic volcano Eyjafjallajökull [Achterberg et al., 2013]. In addition, winter mixed layer depths prior to our cruise were shallower than average in the IRB, while the summer euphotic zone (40 ± 5 m) and mixed layer depths (28 ± 8 m) represented typical conditions for both the IB and IRB [Henson et al., 2013; Painter et al., 2014]. Consequently, inputs of Fe from deep winter convection were also likely up to fourfold higher in the IB (37,500 nmol m 2 ) relative to the IRB (10,000 nmol m 2 )[Painter et al., 2014]. Volcanic Fe inputs during May 2010 to the IB potentially resulted in enhanced macro-nutrient drawdown, with low observed nitrate(<1μM) during summer 2010 in the IB [Achterberg et al.,2013;Ryan-Keogh et al., 2013], while concentrations remained relatively high (3–5μM) and comparable with previous observations in the IRB [Sanders et al., 2005] (Figures 1a and 1c). Silicate was also strongly depleted in the IB (<1μM) relative to IRB (1–4μM) (Figures 1b and 1d). Correspondingly, nutrient enrichment experiments demonstrated clear evidence of Fe limitation in the IRB during summer, while Fe stress was much less severe under the low macronutrient conditions encountered in the IB [Ryan-Keogh et al., 2013]. Furthermore, phytoplankton Fe:C uptake ratios calculated using radiotracer incorporation techniques [Poulton et al., 2010; Twining et al., 2004] averaged 0.9 (±0.6) and 4.1 (±1) μmol mol 1 during summer in the IRB and IB, respectively (C.M.MooreandA.J.Poulton,unpublisheddata),comparable with observations of Fe-limited temperate taxa (~2–10 μmol mol 1 )[Sunda and Huntsman, 1995] and, for the IRB, Fe-limited taxa isolated from chronically low Fe environments [Strzepek et al., 2012]. By the time of our summer sampling, chlorophyll-a (Chl-a) concentrations in the IB had decreased following the spring bloom [Ryan-Keogh et al., 2013]. In contrast, a marked bloom (~1–4μgl 1 ) was still underway in the central IRB, potentially as a result of anomalous hydrographic forcing [Henson et al., 2013]. Conditions in the post-bloom western IRB [Ryan-Keogh et al., 2013] were similar to those within the classical Fe-limited systems, with relatively low Chl-a concentrations (<1μgl 1 ) and residual nitrate (>4μM). Hereafter, we thus restrict discussion of the IRB to this western region [Ryan-Keogh et al., 2013]. Consequently, we compare carbon export measurements for two stations in the IRB (stations 10 and 16, hereafter lower Fe input, Fe) and three stations in the IB (stations 6, 28, and 33, hereafter higher Fe input, +Fe) (Figure 1). 3.2. Carbon Export In order to derive C effx , estimates of both carbon export and Fe input fluxes are required. Following our previous work [Pollard et al., 2009], seasonal nutrient deficits were combined with estimates of the 234 Thbased C and N Geophysical Research Letters 10.1002/2014GL060308 LE MOIGNE ET AL. ©2014. The Authors. 4620 19448007, 2014, 13, Downloaded from https://agupubs.onlinelibrary.wiley.com/doi/10.1002/2014GL060308 by Spanish Cochrane National Provision (Ministerio de Sanidad), Wiley Online Library on [28/11/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License export fluxes [Buesseler et al., 1992; Le Moigne et al., 2013b]. Dividing the observed inorganic N deficit (Figure 2a) with thorium-derived estimates of downward N flux, then multiplying the resultant estimates of export duration by the thorium-derived daily organic carbon fluxes (Figure 2b) (see supporting information), we determined annually integrated POC export estimates of 780 and 1330 mmol m 2 in the IRB (Fe) and the IB (+Fe), respectively (Figure 2c, range for both +Fe and Fe given in Table 1). Excess C export between the Fe-replete (IB) and Fe-limited (IRB) area (Figure 2c) was thus 550 mmolm 2 , similar to a previous estimate for the Crozet region [Morris and Charette,2013;Pollard et al., 2009], but lower than that for the KEOPS experiment [Blain et al., 2007]. 3.3. Sources of New Iron In keeping with prior work [Blain et al., 2007; Pollard et al., 2009], we consider new sources of Fe to the system when deriving C effx, while acknowledging that recycled Fe was likely important in supporting additional recycled production [Boyd and Ellwood, 2010]. As summarized in Table 1, we considered five potential sources of new Fe to the IB and IRB during 2010 (see supporting information for detailed derivations), namely: (1) typical (i.e., non volcanic) atmospheric deposition [Achterberg et al., 2013], (2) additional volcanic atmospheric Fe flux to the IB as a result of the Eyjafjallajökull eruption [Achterberg et al., 2013; Painter et al., 2014], (3) upward vertical diffusive flux [Painter et al., 2014], (4) convective winter mixing flux [Painter et al., 2014], and (5) horizontal Fe flux from the adjacent continental shelves. The latter may include material originating from benthic supplies from continental shelves [Elrod et al., 2004] and direct runoff from rivers and glacial melt [Bhatia et al., 2013; Hawkings et al., 2014]. Horizontal fluxes were estimated by considering surface water Fe concentrations away from shelf regions, ridges, or major currents in both the IB and the IRB following previously approaches [Bucciarelli et al., 2001; Planquette et al., 2007; Rijkenberg et al., 2012] (see supporting information), considering two main sources to the IB (Icelandic shelf and Reykjanes ridge) and three main sources to the IRB (Greenland shelf, the Reykjanes ridge, and the Eastern Greenland Current). Overall, elevated dissolved and particulate iron (DFe and PFe) concentrations (Figures S2 and S3) observed close to the Iceland and Greenland shelves did not persist into Figure 1. Nutrients in the high-latitude North Atlantic. (a) Map of surface water nitrate+nitrite concentrations (μM) during summer 2010. Carbon export stations are indicated for reference, 10 and 16 are the high nitrate stations while 06, 28, and 33 are the low nitrate stations. (b) Map of surface water silicate concentrations (μM) during summer 2010. (c) Nitrate + nitrite concentration (μM) profiles at high (blue filled, triangles: st10; circles: st16) and low (green filled, circles: st6; triangles: st28, squares: st33) nutrients stations. (d) Silicate concentration (μM) profiles at, st6, 10, 16, 28, and 33, symbols are in Figure 1c (see Table S1, S2, and text in supporting information for full description). Geophysical Research Letters 10.1002/2014GL060308 LE MOIGNE ET AL. ©2014. The Authors. 4621 19448007, 2014, 13, Downloaded from https://agupubs.onlinelibrary.wiley.com/doi/10.1002/2014GL060308 by Spanish Cochrane National Provision (Ministerio de Sanidad), Wiley Online Library on [28/11/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License the central basins and hence horizontal inputs were minor, albeit still somewhat uncertain (Table 1). Horizontal fluxes and similar minor vertical diffusive fluxes were scaled up by multiplying daily fluxes by 100 days (the estimated bloom duration), and this showed that both were ultimately minor contributors to overall annual new Fe inputs (Table 1), which were dominated by the winter convective supply [Nielsdottir et al.,2009;Painter et al., 2014]. In total, the annual input of new Fe to the IB (0.041–0.044 mmol m 2 ) was estimated to be ~4 times larger than the input to the IRB (0.011 mmol m 2 ), principally due to higher convective and volcanic fluxes (Table 1). Both these Fe input estimates are comparable with reported values from the KEOPS (0.001–0.023 mmol m 2 ) and CROZEX (0.016–0.076 mmol m 2 ) studies [Morris and Charette, 2013] (Figure 3a). 3.4. Sequestration Efficiency Values of C eff in the IB and IRB were ~30 and ~65 kmol C (mol Fe) 1 , respectively (Table 2), further confirming that the IB experienced Fe fertilization (values of C eff were calculated separately for the two basins rather than as the difference between the two basins, see supporting information). Consistently with previous studies [Blain et al., 2007; Pollard et al., 2009], we also calculated the apparent efficiency of the excess export resulting from the enhanced flux of new Fe to the IRB (C effx ) (supporting information), yielding values of 17–19 kmol C (mol Fe) 1 , close to the CROZEX (17 kmol mol 1 ) but lower than KEOPS (154 kmol mol 1 ) values (Table 2, note hereafter we use the corrected values [Chever et al., 2010; Morris and Charette, 2013] for KEOPS which include an additional source of Fe not considered in the seminal study [Blain et al., 2007]). All these estimates are considerably higher those from three artificial Fe fertilizations experiments (1.2–6.5 kmol mol 1 , locations, and references given in Table 2). Previous studies discussing such ranges in estimates for C eff(x) have tended to focus on the comparability and validity of the methods used to construct Fe and C budgets [Blain et al., 2007; Chever et al., 2010; Morris and Charette, 2013; Pollard et al., 2009], presumably based on the inherent assumption that the variability in C eff(x) was too large to be ascribed to natural factors. 3.5. Mode of Fe Supply Here we focus on an alternative hypothesis, namely that, as previously suggested [Boyd et al., 2007; Chever et al., 2010], the observed range may instead be driven by differences related to the mode of Fe supply. Three distinct modes of Fe supply characterize the various experiments. (1) In the shallow waters of the Kerguelen plateau (500 m deep [Blain et al., 2007]) C effx was large at intermediate levels of Fe inputs during Figure 2. Annual carbon export in the Iceland and Irminger basins. (a) POC and PON export fluxes integrated at 100 m (mmol m 2 d 1 ), (b) nitrogen budget (nitrogen deficit mmol m 2 , and bloom duration, days, see main text for explanation), and (c) annual carbon export (mmol m 2 ). Errors (standard deviation) are also given in Table 1. Geophysical Research Letters 10.1002/2014GL060308 LE MOIGNE ET AL. ©2014. The Authors. 4622 19448007, 2014, 13, Downloaded from https://agupubs.onlinelibrary.wiley.com/doi/10.1002/2014GL060308 by Spanish Cochrane National Provision (Ministerio de Sanidad), Wiley Online Library on [28/11/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License the KEOPS study, with continued supply Fe throughout the year through a combination of diapycnal mixing across a vertical gradient over the shallow shelf and a northwestward flow from Heard islands past the eastern side of the Kerguelen plateau [Chever et al., 2010]. Hence, the Fe stocks above the plateau did not appear to vary much over the duration of the bloom [Blain et al., 2008; Chever et al., 2010]. (2) Conversely, in the deep waters of the productive HLNA and the fertilized region of the CROZEX bloom, the main supply of Table 1. Fe Inputs and Carbon Budget in Both Iceland and Irminger Basins (IB and IRB) Annual Flux of Iron (nmol m 2 ) IRB (Fe) IB (+Fe) Lower Limit b IB (+Fe) Upper Limit b Reference Carbon Budget IRB (Fe) IB (+Fe) Atmospheric (1168) 3478 6128 [Achterberg et al., 2013; Painter et al., 2014] Nitrate drawdown (mmol m 2 ) 201 ± 25 287 ± 9 Vertical diffusive flux 160 a 250 a 250 a [Painter et al., 2014] Daily POC ex (mmol m 2 d 1 ) 7.0 ± 5.7 10.2 ± 4.1 Winter mixing 10,000 37,500 37,500 [Painter et al., 2014] Daily PON ex (mmol m 2 d 1 ) 0.8 ± 0.6 1.5 ± 0.6 Horizontal Fe 15 a 4.5 a 4.5 a Bloom duration (days) 111 130 Total 11,343 41,232 43,882 Annual POCex (mmol m 2 ) 778 1331 Total (in mmol m 2 ) 0.011 0.041 0.044 Range c (lower and upper limits) 682–769 1266–1449 a This was multiplied by 100 days to reflect the duration of the bloom (see text in supporting information). Values in brackets are measured “background”fluxes [Painter et al., 2014] (see text in supporting information). b The lower and upper limit of Fe flux to the IB reflect the sensitivity analysis on the modeled volcano iron deposition given in Achterberg et al. [2013]. c Calculated as [Pollard et al., 2009]. Figure 3. Sequestration efficiency, annual carbon export, and nutrients drawdown for six artificial and natural Fe fertilization on the ocean. (a) Annual carbon export (mmol m 2 ) versus annual Fe input (mmol m 2 ), fertilized regions are in green circles, and non-fertilized regions are in blue circles. Experiment names and fertilized/non fertilized regions are indicated (I: IBIS; K: KEOPS; C: CROZEX). (b) Nitrate drawdown (mmol m 2 ) (inpatch outpatch concentrations of nitrate multiplied by export integration depth for SERIES, EIFEX, and SOFEX) versus C effx and C eff (mol mol 1 ); red are naturally fertilized regions (circles: C effx , triangles: C eff + Fe regions, squares: C eff Fe regions), and blue circles are artifical fertilizations (C effx ) (see Table 2). The mode of Fe supply is indicated on top of the panel as well as the Fe:C ratio in phytoplankton (reference in the main text). (c) Estimated export duration (days, given in [Pollard et al., 2009] for CROZEX and in Figure 2b for IBIS. KEOPS export duration was estimated in a similar fashion by dividing the daily downward fluxes of POC by the “seasonaly integrated”downward fluxes of POC given in Blain et al. [2007]] (their Table 1). Bloom durations for KEOPS are 198 and 167 days, respectively, for + and Fe regions) versus C eff (mol mol 1 ); red circles are natural fertilized/non-fertilized regions, and blue circles are artificial fertilizations (see Table 2) concentrations (in μM, estimated from nutrients references given in Table 2 and Levitus [1982]) and annual carbon export (mmol m 2 ). (d) Bloom duration in the +Fe region (days) estimated from observed satellite-derived Chl-a time series (KEOPS [Blain et al., 2007]; CROZEX [Pollard et al., 2009]; IBIS [Achterberg et al., 2013], see Table S3 in supporting information) versus C effx . Bloom duration for artificial experiments is assumed to be equal to the length of each experiment [Boyd et al., 2004; Buesseler et al., 2004; Smetacek et al., 2012]. Geophysical Research Letters 10.1002/2014GL060308 LE MOIGNE ET AL. ©2014. The Authors. 4623 19448007, 2014, 13, Downloaded from https://agupubs.onlinelibrary.wiley.com/doi/10.1002/2014GL060308 by Spanish Cochrane National Provision (Ministerio de Sanidad), Wiley Online Library on [28/11/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License Fe appears to be seasonal [Pollard et al., 2009] (Table 1). Hence, in both these regions, Fe might be expected to build up within the mixed layer during winter (a period of about 100 days). Once this mixed layer shoals in spring, the Fe pool can be used during the productive season until exhaustion [Nielsdottir et al., 2009; Planquette et al., 2007]. A similar situation is likely to characterize much of the Southern Ocean [Boyd et al., 2012; Tagliabue et al., 2014]. (3) Finally, during the artificial experiments, several tons of FeSO 4 were supplied to the surface ocean in one or several discrete seedings over a period of several weeks, resulting in a large loss of Fe due to the formation of insoluble Fe oxy-hydroxides [Boyd et al., 2000]. 3.6. Implication for the Mode of Iron Supply in Fertilized Blooms We suggest that it is likely that high C effx values are associated with protracted blooms where Fe availability is maintained due to continuous Fe supply and perhaps also Fe complexation by organic ligands, sustaining continued C export over a prolonged period, rather than being rapidly lost through precipitation and scavenging. We test the association of high C effx with prolonged blooms by comparing Fe supply modes to in situ estimates of nitrate uptake, duration of the export phase, and bloom duration (derived from satellite Chl-a time series, see supporting information Table S3). We acknowledge that estimates of the duration of export for the HLNA and CROZEX are not independent of the calculated seasonally integrated C export (and thus C eff(x) ), as both are based on observed nitrate drawdown. However, nitrate drawdown and C effx were independent variables in KEOPS and in the artificial Fe release experiments. Moreover, bloom durations derived from satellite Chl-a time series are independent of C effx calculations in all cases and scaled well with estimated export durations (see Table S3 in supporting information). Overall, both C effx and/or C eff , were correlated with bloom (or experiment) nitrate uptake (Figure 3b), export duration (Figure 3c), and bloom duration (Figure 3d). Moreover, the maximum estimates of C effx and/or C eff were comparable to the range of C:Fe ratios measured in Fe-replete phytoplankton cultures [Sunda and Huntsman, 1995], which could conceivably set the upper bound on C effx . All the markedly lower estimates of C eff to date thus appear to relate to situations where Fe supply was discontinuous (Figure 3). Blooms characterized by a more continuous supply of Fe (KEOPS) thus appear to export carbon more efficiently per unit of Fe added over a complete seasonal cycle. Such an effect would be consistent with a low but continuous supply of Fe allowing efficient retention of Fe within the dissolved phase in the euphotic zone, likely through full complexation by organic ligands [Gledhill and Buck, 2012], or within the ecosystem itself. Indeed, biological uptake within such blooms could be envisaged to generate the gradients in bioavailable Fe at the periphery (either below or to the edges of the bloom), which would likely dictate the magnitude of Fe flux into the bloom region. Such a scenario is in marked contrast to the situation in Table 2. Carbon Sequestration Efficiency and Nutrient Drawdown Area Type of Fertilization Nitrate Drawdown (mmol m 2 ) d Sequestration Efficiency (Ceffx and Ceff, kmol mol 1 ) Reference C effx HLNA Natural 287 17.2–19.0 This study Southern Ocean Natural 687 b 154.0 a [Blain et al., 2007] Southern Ocean Natural 404 c 17.2 a [Pollard et al., 2009] Southern Ocean Artificial 75 e 6.5 [Smetacek et al., 2012] Southern Ocean Artificial 75 e 3.3 [Buesseler et al., 2004] Subarctic PacificArtificial 200 e 1.2 [Boyd et al., 2004] C eff HLNA +Fe (IB) given above 30.2–32.5 This study Fe (IRB) 201 65.5 Southern Ocean (KEOPS Kerguelen) +Fe given above 227.3 a [Blain et al., 2007] Fe 332 b 2,883.3 a Southern Ocean (CROZEX Crozet) +Fe given above 18/8 a [Pollard et al., 2009] Fe 161 c 25.0 a a Recalculated in [Morris and Charette, 2013]. b Nutrient data from KEOPS are presented in Mosseri et al. [2008]. c Nutrient data from CROZEX are presented in Sanders et al. [2007] (reference in supporting information). d Nitrate drawdown in the upper panel of the table is given for the fertilized area or patch (+Fe). e Nutrient drawdown was calculated as surface nutrient in-patch surface nutrient out-patch multiplied by export integration depth. Geophysical Research Letters 10.1002/2014GL060308 LE MOIGNE ET AL. ©2014. The Authors. 4624 19448007, 2014, 13, Downloaded from https://agupubs.onlinelibrary.wiley.com/doi/10.1002/2014GL060308 by Spanish Cochrane National Provision (Ministerio de Sanidad), Wiley Online Library on [28/11/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License purposeful FeSO 4 release experiments, where a large amount of Fe is supplied to a region with low preexisting biomass. Orders of magnitude lower C effx in such situations might be expected to result from loss of much of the pulsed excess Fe inputs [Boyd et al., 2004; Buesseler et al., 2004; Smetacek et al., 2012] due to saturation of the available Fe complexing ligands sites and rapid transfer of Fe into colloidal (>200 kDa) phases with subsequent particle scavenging alongside a limited capacity for the extant microbial community to take up the sudden large Fe excess [Bowie et al., 2001; Boyd et al., 2007]. Finally, in systems where the supply of Fe is seasonally dominated (CROZEX, IBIS), the bloom likely terminates when the Fe stock is used up [Boyd et al., 2012; Nielsdottir et al., 2009]. Consistent with such arguments, both the excess carbon export (+Fe minus Fe carbon export) and excess Fe input (+Fe minus Fe Fe input) for IBIS were about half that of CROZEX (552 versus 1041 mmol C m 2 and 0.031 versus 0.062 mmol Fe m 2 , respectively). As outlined above, our estimates of C effx are, by definition, based on new Fe inputs to the upper ocean. However, a large fraction of the total Fe used by phytoplankton is likely derived from regeneration, for example through grazing and viral lysis [Boyd et al., 2012, 2005; Strzepek et al., 2005]. Our simple explanation for the apparent relationship between C eff(x) and bloom duration (Figure 3), as dictated by the period over which Fe is retained and/or resupplied to the system will, in reality, likely reflect a range of complex microbial recycling processes in the mixed layer [Boyd and Ellwood, 2010; Strzepek et al., 2005]. Subsequent consequences for C effx ,asdefined, will depend on the extent to which Fe or C/macronutrients are preferential remineralized within a system [Frew et al., 2006; Twining et al., 2014]. For example, if C is effectively remineralized at shallower depth than Fe, potentially due to (re-)scavenging of the latter [Frew et al., 2006; Twining et al., 2014], C effx might be further depressed as the ratio between new and recycled Fe supply (termed the “fe-ratio”[Boyd et al., 2005]) decreases, as might be expected in chronically low Fe systems or the latter stages of the seasonal cycle [Tagliabue et al., 2014]. Overall, on the basis of available data (Figure 3), it thus appears that C eff exhibits systematic variability which is not a simple function of the magnitude of new Fe inputs (Figure 3a) but is instead related to the mode of new Fe supply and the duration of the bloom/export phase (Figures 3b, 3c, and 3d), potentially moderated further by the intensity of subsequent recycling. Consequently, differences in the natural modes of new Fe supply likely result in spatio-temporal gradients in C effx within both the modern and paleo-oceans. For example, relatively low but continuous benthic Fe inputs to coastal upwelling regions would be expected to result in high C effx ; however, this efficiency would be expected to decrease as inputs increase, particularly if it exceeds biological demand or the complexing capacity of natural organic Fe binding ligands [Elrod et al., 2004]. In contrast, aerosols inputs to Fe-limited regions, representing localized and seasonally variable sources of Fe deposited over timescales of weeks to months [Jickells et al., 2005; Mahowald et al., 2009], might be expected to be associated with mid range values of C effx , which could vary as a function of deposition intensity, alongside other factors which may alter overall biological demand, such as the availability of macronutrients. A more complete mechanistic understanding of linkages between atmospheric CO 2 and past [Martinez-Garcia et al., 2014; Ridgwell and Watson, 2002] or future [Jickells et al., 2005] natural Fe fertilization from dust deposition may thus need to consider such processes. More broadly, any attempts to quantitatively link the oceanic Fe and C cycles will need to consider the potential for delivery mode to control the amount of carbon sequestered per unit Fe added. References Achterberg, E. P., et al. (2013), Natural iron fertilisation by the Eyafjallojokull eruption, Geophys. Res. Lett.,40, 921–926, doi:10.1002/grl.50221. Bhatia, M. P., E. B. Kujawinski, S. B. Das, C. F. Breier, P. B. Henderson, and M. A. Charette (2013), Greenland meltwater as a significant and potentially bioavailable source of iron to the ocean, Nat. Geosci.,6, 274–278. Blain, S., et al. (2007), Effect of natural iron fertilization on carbon sequestration in the Southern Ocean, Nature,446(7139), 1070–1074. Blain, S., G. Sarthou, and P. Laan (2008), Distribution of dissolved iron during the natural iron-fertilization experiment KEOPS (Kerguelen Plateau, Southern Ocean), Deep Sea Res., Part II,55(5–7), 594–605. Bowie, A. R., M. T. Maldonado, R. D. Frew, P. L. Croot, E. P. Achterberg, R. F. C. Mantoura, P. J. Worsfold, C. S. Law, and P. W. Boyd (2001), The fate of added iron during a mesoscale fertilisation experiment in the Southern Ocean, Deep Sea Res., Part II,48(11–12), 2703–2743. Boyd, P. W., and M. J. Ellwood (2010), The biogeochemical cycle of iron in the ocean, Nat. Geosci.,3, 675–682. Boyd, P. W., et al. (2000), A mesoscale phytoplankton bloom in the polar Southern Ocean stimulated by iron fertilization, Nature,407(6805), 695–702. Boyd, P. W., et al. (2004), The decline and fate of an iron-induced subarctic phytoplankton bloom, Nature,428(6982), 549–553. Boyd, P. W., et al. (2005), FeCycle: Attempting an iron biogeochemical budget from a mesoscale SF(6) tracer experiment in unperturbed low iron waters, Global Biogeochem. Cycles,19, GB4S20, doi:10.1029/2005GB002494. Boyd, P. W., et al. (2007), Mesoscale iron enrichment experiments 1993–2005: Synthesis and future directions, Science,315(5812), 612–617. Acknowledgments We thank all the scientists and crew members on IBIS cruises D350/354 (R.R.S. Discovery, 1962–2012, NERC). We acknowledge M. Stinchcombe, B. Head and K. Pabortsava, A. Milton, and C. Marsay for technical support. This work was funded by NERC (NE/E006833/1, E.P.A. & C.M.M.) and the EU (CalMarO, 215157, R.J.S., F.A.C.L.M). Data are held at the British Oceanographic Data Centre, http://www.bodc.ac.uk/. We would like to thank two anonymous reviewers and the editor for providing constructive comments. The Editor thanks Robert Strzepek and an anonymous reviewer for their assistance in evaluating this paper. Geophysical Research Letters 10.1002/2014GL060308 LE MOIGNE ET AL. ©2014. The Authors. 4625 19448007, 2014, 13, Downloaded from https://agupubs.onlinelibrary.wiley.com/doi/10.1002/2014GL060308 by Spanish Cochrane National Provision (Ministerio de Sanidad), Wiley Online Library on [28/11/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License Boyd, P. W., et al. (2012), Microbial control of diatom bloom dynamics in the open ocean, Geophys. Res. Lett.,39, L18601, doi:10.1029/ 2012GL053448. Bucciarelli, E., S. Blain, and P. Treguer (2001), Iron and manganese in the wake of the Kerguelen Islands (Southern Ocean), Mar. Chem.,73(1), 21–36. Buesseler, K. O., M. P. Bacon, J. K. Cochran, and H. D. Livingston (1992), Carbon and nitrogen export during the JGOFS North Atlantic Bloom Experiment estimated from 234 Th: 238 U disequilibria, Deep Sea Res., Part I,39(7–8), 1115–1137. Buesseler, K. O., J. E. Andrews, S. M. Pike, and M. A. Charette (2004), The effects of iron fertilization on carbon sequestration in the Southern Ocean, Science,304(5669), 414–417. Chever, F., G. Sarthou, E. Bucciarelli, S. Blain, and A. R. Bowie (2010), An iron budget during the natural iron fertilisation experiement KEOPS (Kerguelen Islands, Southern Ocean), Biogeosciences,7, 455–468. de Baar, H. J. W., et al. (2005), Synthesis of iron fertilization experiments: From the iron age in the age of enlightenment, J. Geophys. Res.,110, C09S16, doi:10.1029/2004JC002601. Elrod, V. A., W. M. Berelson, K. H. Coale, and K. S. Johnson (2004), The flux of iron from continental shelf sediments: A missing source for global budgets, Geophys. Res. Lett.,31, L12307, doi:10.1029/2004GL020216. Frew, R. D., D. A. Hutchins, S. Nodder, S. A. Sanudo-Wilhelmy, A. Tovar-Sanchez, K. Leblanc, C. E. Hare, and P. W. Boyd (2006), Particulate iron dynamics during FeCycle in subantarctic waters southeast of New Zealand, Global Biogeochem. Cycles,20, GB1S93, doi:10.1029/ 2005GB002558. Gledhill, M., and K. N. Buck (2012), The organic complexation of iron in the marine environment: A review, Front. Microb.,28,3–69. Hawkings, J. R., J. L. Wadham, M. Tranter, R. Raiswell, L. G. Benning, P. J. Statham, A. Tedstone, P. Nienow, K. Lee, and J. Telling (2014), Ice sheets as a significant source of highly reactive nanoparticulate iron to the oceans, Nat. Commun.,5, 3929, doi:10.1038/ncomms4929. Henson, S., S. C. Painter, N. P. Holliday, M. Stinchcombe, and S. L. C. Giering (2013), Unusual subpolar North Atlantic phytoplankton bloom in 2010: volcanic fertilisation or North Atlantic Oscillation?, J. Geophys. Res. Oceans,118,1–10, doi:10.1002/jgrc.20363. Jickells, T. D., et al. (2005), Global iron connections between desert dust, ocean biogeochemistry, and climate, Science,308(67), 67–71. Le Moigne, F. A. C., R. J. Sanders, M. Villa-Alfageme, A. P. Martin, K. Pabortsava, H. Planquette, P. J. Morris, and S. J. Thomalla (2012), On the proportion of ballast versus non-ballast associated carbon export in the surface ocean, Geophys. Res. Lett.,39, L15610, doi:10.1029/ 2012GL052980. Le Moigne, F. A. C., S. A. Henson, R. J. Sanders, and E. Madsen (2013a), Global database of surface ocean particulate organic carbon export fluxes from the 234 Th technique, Earth Syst. Sci. Data Discuss.,6, 163–187, doi:10.5194/essdd-6-163-2013. Le Moigne, F. A. C., M. Villa-Alfageme, R. J. Sanders, C. M. Marsay, S. Henson, and R. Garcia-Tenorio (2013b), Export of organic carbon and biominerals derived from 234 Th and 210 Po at the Porcupine Abyssal Plain, Deep Sea Res., Part I,72,88–101. Levitus, S. E. (1982), Climatological atlas of the world ocean, NOAA Professional Paper, US Government Printing Office, Washington, D. C. Mahowald, N. M., et al. (2009), Atmospheric Iron Deposition: Global Distribution, Variability, and Human Perturbations, Annu. Rev. Mar. Sci.,1, 245–278. Maiti, K., et al. (2012), Intercalibration studies of short-lived thorium-234 in the water column and marine particles, Limnol. Oceanogr. Methods,10, 631–644. Marinov, I., A. Gnanadesikan, J. L. Sarmiento, J. R. Toggweiler, M. Follows, and B. K. Mignone (2008), Impact of oceanic circulation on biological carbon storage in the ocean and atmospheric pCO 2 ,Global Biogeochem. Cycles,22, GB3007, doi:10.1029/2007GB002958. Martin, J. H., R. M. Gordon, and S. E. Fitzwater (1990), Iron in Antarctic waters, Nature,345(6271), 156–158. Martinez-Garcia, A., D. M. Sigman, H. Ren, R. F. Anderson, M. Straub, D. A. Hodell, S. L. Jaccard, T. I. Eglinton, and G. H. Haug (2014), Iron fertilization of the Subantarctic Ocean during the last ice age, Science,343, 1347–1350. Morris, P. J., and M. A. Charette (2013), A synthesis of upper ocean carbon and dissolved iron budgets for the Southern Ocean natural iron fertilization, Deep Sea Res., Part II,90, 147–157. Mosseri, J., B. Queguiner, L. Armand, and V. Cornet-Barthaux (2008), Impact of iron on silicon utilization by diatoms in the Southern Ocean: A case study of Si/N cycle decoupling in a naturally iron-enriched area, Deep Sea Res., Part II,55(5–7), 801–819. Nielsdottir, M., C. M. Moore, E. P. Achterberg, R. J. Sanders, and D. J. Hinz (2009), Iron limitation of ht epostbloom phytoplankton communities in the Iceland Basin, Global Biogeochem. Cycles,23, GB3001, doi:10.1029/2008GB0034. Painter, S. C., S. A. Henson, A. Forryan, S. Steigenberger, J. Klar, M. Stinchcombe, N. Rogan, A. Baker, E. P. Achterberg, and C. M. Moore (2014), An assessment of the vertical diffusive flux of iron and other nutrients to the surface waters of the subpolar North Atlantic Ocean, Biogeosciences,11, 2113–2130. Pike, S. M., K. O. Buesseler, J. Andrews, and N. Savoye (2005), Quantification of Th-234 recovery in small volume sea water samples by inductively coupled plasma-mass spectrometry, J. Radioanal. Nucl. Chem.,263(2), 355–360. Planquette, H., et al. (2007), Dissolved iron in the vicinity of the Crozet Islands, Southern Ocean, Deep Sea Res., Part II,57, 1999–2019. Pollard, R. T., et al. (2009), Southern Ocean deep-water carbon export enhanced by natural iron fertilization, Nature,457(7229), 577–U581. Poulton, A., A. Charalampopoulou, J. R. Young, G. A. Tarran, M. I. Lucas, and G. D. Quartly (2010), Coccolithophore dynamics in non-bloom conditions during late summer in the central Iceland Basin (July-August 2007), Limnol. Oceanogr.,4(55), 1601–1613. Ridgwell, A. J., and A. J. Watson (2002), Feedback between aeolian dust, climate and atmospheric CO2 in glacial time, Paleoceanography,17(4, 1059), doi:10.1029/2001PA000729. Rijkenberg, M. J. A., S. Steigenberger, C. F. Powell, H. van Haren, M. D. Patey, A. R. Baker, and E. P. Achterberg (2012), Fluxes and distribution of dissolved iron in the eastern (sub-) tropical North Atlantic Ocean, Global Biogeochem. Cycles,26, GB3004, doi:10.1029/2011GB004264. Ryan-Keogh, T. J., A. I. Macey, M. C. Nielsdo´ttir, M. I. Lucas, S. S. Steingenberger, M. C. Stinchcombe, E. P. Achterberg, T. S. Bibby, and C. M. Moore (2013), Spatial and temporal development of phytoplankton iron stress in relation to bloom dynamics in the high latitude North Atlantic Ocean, Limnol. Oceanogr.,58(2), 533–545. Sanders, R., and T. D. Jickells (2000), Total organic nutrients in Drake passage, Deep Sea Res., Part I,47, 997–1014. Sanders, R., L. Brown, S. Henson, and M. Lucas (2005), New production in the Irminger Basin during 2002, J. Mar. Syst.,55(3–4), 291–310. Sanders, R., P. J. Morris, M. Stinchcombe, S. Seeyave, H. Venables, and M. Lucas (2007), New production and the f ratio around the Crozet Plateau in austral summer 2004–2005 diagnosed from seasonal changes in inorganic nutrient levels, Deep Sea Res., Part II,54(18–20), 2191–2207. Sarmiento, J. L., and J. C. Orr (1991), 3-dimensional simulations of the impact of Southern Ocean nutrient depletion on atmospheric CO 2 and ocean chemistry, Limnol. Oceanogr.,36(8), 1928–1950. Smetacek, V., et al. (2012), Deep carbon export from the Southern Ocean iron-fertilized diatom bloom, Nature,487, 313–319. Strzepek, R. F., M. T. Maldonado, J. L. Higgin, J. Hall, K. Safi,S.W.Wilhelm,andP.W.Boyd(2005),Spinningthe"FerrousWheel": The importance of the microbial community in an iron budget during the FeCycle experiment, Global Biogeochem. Cycles,19, GB4S26, doi:10.1029/2005GB002490. Geophysical Research Letters 10.1002/2014GL060308 LE MOIGNE ET AL. ©2014. The Authors. 4626 19448007, 2014, 13, Downloaded from https://agupubs.onlinelibrary.wiley.com/doi/10.1002/2014GL060308 by Spanish Cochrane National Provision (Ministerio de Sanidad), Wiley Online Library on [28/11/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License Strzepek, R. F., K. A. Hunter, R. D. Frew, P. J. Harrison, and P. W. Boyd (2012), Iron-light interactions differ in Southern Ocean phytoplankton, Limnol. Oceanogr.,57(4), 1182–1200, doi:10.1029/2005GB002490. Sunda, W. G., and S. A. Huntsman (1995), Iron uptake and growth limitation in oceanic and coastal phytoplankton, Mar. Chem.,50, 189–206. Tagliabue, A., J. B. Sallee, A. R. Bowie, M. Levy, S. Swart, and P. D. Boyd (2014), Surface-water iron supplies in the Southern Ocean sustained by deep winter mixing, Nat. Geosci.,7, 314–320. Twining, B. S., S. B. Baines, N. S. Fisher, and M. R. Landry (2004), Cellular iron contents of plankton during the Southern Ocean Iron Experiment (SOFeX), Deep Sea Res., Part I,51, 1827–1850. Twining, B. S., S. D. Nodder, A. L. King, D. A. Hutchins, G. R. LeCleir, J. M. De Bruyn, E. W. Maas, S. Vogt, S. W. Wilhelm, and P. W. Boyd (2014), Differential remineralization of major and trace elements in sinking diatoms, Limnol. Oceanogr.,59, 689–704. Geophysical Research Letters 10.1002/2014GL060308 LE MOIGNE ET AL. ©2014. The Authors. 4627 19448007, 2014, 13, Downloaded from https://agupubs.onlinelibrary.wiley.com/doi/10.1002/2014GL060308 by Spanish Cochrane National Provision (Ministerio de Sanidad), Wiley Online Library on [28/11/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License