SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Materials and Methods
  5. 3. Results
  6. 4. Discussion and Conclusions
  7. Acknowledgments
  8. References
  9. Supporting Information

[1] Several mechanisms have been proposed for large CO2 changes at glacial Terminations, including shifting the CaCO3:Corg rain ratio by changing surface water nutrient supply, altering the balance between diatom and coccolithophore production. Diatom Si:N is highest in Fe-stressed high-latitude waters. Southern Ocean Fe enrichment studies suggest diatom Si demands reduced under Fe-replete (glacial) conditions, allowing increased silicic acid to leak northward in subducted intermediate water and upwell at lower latitudes. We test this ‘Silicic Acid Leakage’ hypothesis using relative abundances of phytoplankton-specific biomarkers in Peru margin sediments spanning 0–20 Ka. Results indicate increased coccolithophorid:diatom production from ∼0.5 to 3 between 18.0–15.5 Ka. Temporal correlation with the initial pCO2 rise and early deglacial shift in mode water ventilation implicates a coincidental, possibly causative reorganization of Sub-Antarctic Mode Water formation and reduced Fe abundance. However, coccolithophorid production subsequently declined, suggesting rain ratio changes were only partly responsible for the CO2 deglacial transition.

1. Introduction

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Materials and Methods
  5. 3. Results
  6. 4. Discussion and Conclusions
  7. Acknowledgments
  8. References
  9. Supporting Information

[2] During the Last Glacial Maximum (LGM), the partial pressure of carbon dioxide (pCO2) was approximately 100 ppmv lower than pre-anthropogenic interglacial values [Petit et al., 1999]. Temporal correlation of this change with global ice volume reconstructions suggests it played a primary role in past global climate change [Shackleton, 2000]. Although several hypotheses have been proposed, there is still considerable uncertainty as to the dominant causative mechanism for large CO2 changes at glacial Terminations [e.g., Sigman and Boyle, 2000]. Since the oceans contain ∼60 times more CO2 than the atmosphere, most explanations require change to the ocean C cycle, and several invoke a significant shift in rain ratio of CaCO3:Corg (inorganic to organic carbon) from the mixed layer: increased sinking of CaCO3 exports C, but produces an even greater reduction in surface alkalinity and consequently a net increase in pCO2. In contrast, enhanced export of Corg principally lowers total surface dissolved inorganic CO2 without a significant change in alkalinity, and thereby lowers pCO2.

[3] Reconstructions of glacial pCO2 include a focus on modeling the effect of large shifts in oceanic silica and nitrate inventories in intermediate waters supplying low-latitude upwelling zones. Biogeochemical-General Circulation Models simulate changes in the CaCO3:Corg rain ratio by altering the balance between diatoms and calcite-secreting coccolithophores. The Si:N molar ratio in diatoms at low latitudes is ∼1:1 [Brzezinski, 1985], but is precipitated at ratios ≤4:1 in iron (Fe)-stressed high latitude waters (e.g., Southern Ocean) [Franck et al., 2000; Pondaven et al., 2000]. In contrast, the silicic acid to nitrate molar ratio in southern-sourced intermediate waters (IW) of the upper ocean is <1:1 [Dugdale and Wilkerson, 2001]. Increased silica availability can raise the Si:N ratio and enhance diatom production in both high- and low-latitude waters via this IW conduit, displacing other phytoplankton such as coccolithophores. Mesocosm field experiments suggest that such diatom dominance (≥70%) occurs irrespective of environmental variability (e.g., seasonality, illumination) at silicate concentrations >2 μM [Egge and Asknes, 1992]. Coccolithophores outcompeted diatoms only at silicate values <1 μM. It is thought that diatom success results from high inherent growth rates, a higher photosynthetic capacity due to greater chlorophyll content [Langdon, 1988], and by rapid export of available nutrients from the euphotic zone [Schöllhorn and Granéli, 1996].

[4] The Southern Ocean plays a dominant role in determining the IW Si:N ratio since unused nutrients from polar surface waters are entrained in subducting intermediate and mode waters. Fe enrichment studies in the Southern Ocean suggest that under more Fe-replete (glacial) conditions, diatom Si demands south of the Polar Front were significantly reduced, resulting in elevated remnant silicic acid concentrations in surface waters coincident with enhanced diatom production [Takeda, 1998; Brzezinski et al., 2002, 2003]. Subsequent northward export of subducted IW possessing elevated Si:N from the Southern Ocean has been termed the ‘silicic acid leakage’ (SAL) hypothesis [Matsumoto et al., 2002]. Since diatom export production has only a negligible effect on alkalinity, this rain ratio shift from calcitic coccoliths to Si-rich diatoms could generate close to the glacial pCO2 drawdown [Archer and Maier-Reimer, 1994].

[5] We use phytoplankton class-specific biomarkers to look for evidence of a floristic shift from diatoms to coccolithophores across the deglacial sedimentary record. We focus on organic-rich Peru margin sediments spanning the last 20 Kyr, deposited beneath highly productive waters where upwelling of Equatorial Undercurrent Water delivers nutrients derived from Sub-Antarctic Mode Water originating in the Southern Ocean. The influence of N. Pacific IW cannot be excluded, but significant variations in input to the Peru margin are inconsistent with the SAL hypothesis. We employ organic geochemical indicators for coccolithophore (alkenone) and diatom (brassicasterol, β-sitosterol) production, investigating molecules with demonstrably similar functional group reactivity and preservation within the same organic fraction. Inter-comparison of the relative abundance of these biomarkers allows us to generate an intrinsic index of the relative export by each phytoplankton class. When extracted from the same sediment sample and therefore deposited in synchrony, we can be confident that relative variations in biomarker abundance are primarily related to changes in dominant primary producer. We reconstruct the glacial, deglacial and interglacial signatures required to test the SAL hypothesis.

2. Materials and Methods

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Materials and Methods
  5. 3. Results
  6. 4. Discussion and Conclusions
  7. Acknowledgments
  8. References
  9. Supporting Information

[6] Gravity cores W7706-37K (13°38′S, 76°51′W, 370 m), W7706-40K (11°15′S, 77°58′W, 186 m), and W7706-41K (11°21′S, 78°7′W, 410 m) were sampled every ∼1000-yr from 0–20 Ka and freeze-dried prior to analysis.

[7] Nitrogen abundance (%N) and δ15N measurements were made on ∼25 mg aliquots of homogenized bulk sediment. Details of analytical protocol and instrumentation are published elsewhere [e.g., Higginson and Altabet, 2004]. Total lipid extracts were obtained from each sample by pressurized fluid extraction at 150°C and 2000 psi using small solvent volumes. Separation and analysis of lipid fractions was performed according to published methods [e.g., Canuel and Martens, 1996] (see auxiliary material). Lipid identification and quantification was achieved by co-injection and comparison with authentic standards [Chromadex].

[8] Denitrification and nitrogen fixation are principal (but geographically separate) processes controlling the variability in nitrogen isotopic ratio of subsurface marine nitrate. Water-column denitrification (removal of fixed nitrogen, NO3 to N2) produces large isotopic enrichments, especially in suboxic IW where subsurface δ15N maxima can reach 20‰ [Liu and Kaplan, 1989]. It occurs primarily under suboxic conditions, today largely confined to Oxygen Minimum Zones of the Arabian Sea, Eastern Tropical North Pacific and Peru upwelling zone. Combined water column NO3, trap and surface sediment data confirm the assumption of isotopic balance between NO3 influx to the euphotic zone and loss from sinking particles, and the fidelity of this isotopic signal in sedimentary archives with moderate to excellent organic matter preservation and minimal long-term diagenetic effect [Altabet, 1988]. This fidelity is maintained under conditions of complete annual NO3 utilization found today in all major nearshore upwelling regions and most likely maintained also during the dustier last glacial. We use the effect of denitrification on δ15N, under conditions of complete annual NO3 drawdown by phytoplankton, to elucidate changes in Peru margin water column ventilation and changes in source water formation conditions for SAMW.

[9] We use molecular biomarker accumulation rates to identify unambiguously the marine component of total organic matter and to document temporal changes in the dominant phytoplankton class. Long-chain unsaturated ketones (alkenones) are unique to several haptophyte species, including the cosmopolitan coccolithophores E. huxleyi and G. oceanica. Total C37 and C38 alkenone abundance between the LGM and Holocene has been quantified as an exclusive coccolithophore paleo-abundance proxy [e.g., Pelejero et al., 1999]. Diatoms are the dominant source of opal to sediments in all upwelling environments and diatom biomarkers have shown good correlation with opal accumulation rates in areas with good opal preservation [Werne et al., 2000]. We measured a combination of diatom chemotaxonomic molecular biomarkers, brassicasterol and β-sitosterol, with proven utility in similar studies at upwelling locations [Schubert et al., 1998]. Brassicasterol (24-methylcholesta-5, 22-dien-3β-ol) is considered a diatom-derived sterol with sufficient diagenetic resistance to be valuable as a long-term biomarker [e.g., Orcutt and Patterson, 1975; Volkman et al., 1998]. Similarly, β-sitosterol (24-ethylcholesta-5-en-3β-ol) production has been linked with marine diatom production [Volkman, 1986]. Although the source of diatom biomarkers is more equivocal than alkenones (including some Haptophyta algae), we focus on intervals in which the abundance records of diatom biomarkers are similar to each other, but can be differentiated from alkenone abundance patterns. We evaluate the difference in contribution by the two taxa to total productivity through a ratio of coccolithophore:diatom biomarker abundance. Details of age model construction are provided in the auxiliary material.

3. Results

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Materials and Methods
  5. 3. Results
  6. 4. Discussion and Conclusions
  7. Acknowledgments
  8. References
  9. Supporting Information

[10] Substantial changes in intermediate-water ventilation 11–14°S on the Peru margin are recorded by a 5‰ increase in δ15N from 17.6–16.2 Ka, followed by a more gradual 2‰ decline between 16.2–12.4 Ka (Figure 1a). We interpret the early increase in δ15N as the product of increased denitrification in response to reduced remote IW ventilation. We can discount an alternative hypothesis based on enhanced respiration of suddenly increased sinking organic matter driven by stronger upwelling, because the total sedimentary organic content and alkenone-based SST reconstructions increase monotonically from 18–12 Ka (not shown). The IW source is the same SAMW implicated as the conduit for silicic acid leakage out of the Southern Ocean. These data suggest large-scale reorganization of SAMW formation processes 18–16 Ka.

image

Figure 1. Plots of (a) δ15N, (b) brassicasterol and (c) β-sitosterol (diatom biomarker) abundance, (d) ΣC37 alkenone (coccolithophore biomarker) abundance, (e) coccolithophore:diatom proxy ratio, and (f) Antarctic ice-core CO2 and dust abundance, 0–20 Kyr BP. Data from core 37K are shown (squares), core 40K (crosses), and core 41K (diamonds). Grey bars in (e) represent 500-yr moving average calculations from a Stineman Function curve smoothing followed by geometric weighting of each datum and ±10% of the data range. Changes in coccolithophore:diatom proxy ratio are relative, and do not indicate absolute abundance. In lower panel (f), Byrd ice-core CO2 is shown in black, Vostok dust abundance is shown in red (reversed scale).

Download figure to PowerPoint

[11] The concentrations of the two diatom-related sterols (brassicasterol and β-sitosterol) show close agreement and their records are similar in two deglacial cores (Figures 1b and 1c), but are statistically different from relative alkenone abundances (Figure 1d). Neither indicates a significant change in diatom export from 20–15.5 Ka, despite a small increase in northern core 41K at 17.0 Ka. However, both cores exhibit maximum sterol concentrations at 12.4 Ka, before apparently decreasing again in the Holocene. Neither diatom biomarker exhibits a sharp decline from 18–15 Ka, eliminating any association with terrigenous fluxes. In contrast, the concentrations of coccolithophore-derived C37–38 alkenones increase from 18.0 Ka to maxima at 15.5 Ka and remain high to 13.5 Ka before declining again. This indicates an increase in coccolithophorid production at least 2,500 years before an increase in diatoms is recorded.

[12] The ratio of coccolithophorid to diatom production in the Peru upwelling zone was low and unchanged from 20–17.6 Ka, increased sharply from 17.6–15.5 Ka, but then decreased again to near-glacial levels by 13 Ka (Figure 1e). Measurements of late Holocene sediments at site 40K suggest a relatively unchanged ratio, despite evidence for significant increases in concentration of all biomarkers, perhaps indicative of enhanced preservation in modern core-top sediments. The interval of more dominant coccolithophorid production (17.6–15.5 Ka) encompasses the timing of the sharp increase in δ15N (17.6–16.2 Ka), with small discrepancies likely due to differences in sampling resolution. This coincidence is consistent with an explanation relating the change in both proxies to a significant shift in both Si:N and dissolved O2 profile of upwelled water, and corresponding changes in properties of source SAMW.

4. Discussion and Conclusions

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Materials and Methods
  5. 3. Results
  6. 4. Discussion and Conclusions
  7. Acknowledgments
  8. References
  9. Supporting Information

[13] There is good correspondence between the timing of the early rise in Peru margin δ15N, the interval of more dominant coccolithophorid export production, the dramatic decline in dust abundance and early deglacial rise in CO2 (Figure 1f). We temporally correlate this shift in dominant phytoplankton class and denitrification with a rise in pCO2 of around 45 ppmv, approximately half of the deglacial CO2 change.

[14] Consistent with the SALH, one explanation for this correlation is that reduced dust abundance recorded in Vostok ice signifies an abrupt increase in Fe-limitation and the Si:N uptake ratio of diatoms in the Southern Ocean from 17.5–15.5 Ka. This would have created a depleted pool of exported silicic acid, coincident with increased sea-ice cover/stratification that reduced relative ventilation of these waters in the Southern Ocean. Export of water to the eastern Pacific boundary current may have simultaneously effected a shift in ventilation and reduction in equatorial water Si:N availability, thus favoring coccolithophore over diatom production.

[15] The temporal correlation of these events during early deglaciation supports the SALH but does not constitute proof of causation. If the relationship between production and sedimentary deposition of coccolithophore and diatom biomarkers is linear, then the maximum shift in alkenone:sterol ratio from 0.5 to 3.0 relative to nominal unity is the first evidence for a large change in marine primary production and shift in inorganic:organic C rain ratio during deglaciation. However, although the shift in alkenone abundance occurs coincidentally with the start of the pCO2 rise (within the precision of the age models described), it is relatively short-lived. If the post- 13.5 Ka decline in alkenone concentration indicates a postglacial decline in relative coccolithophore export, then this must implicate other mechanisms to explain the remainder of the deglacial pCO2 increase [Calvo et al., 2004]. Further, we do not find biomarker evidence for increased low-latitude glacial diatom productivity relative to the Holocene, as predicated by the SALH for lower glacial pCO2. This suggests that rain ratio changes were only partly responsible for the CO2 deglacial transition.

[16] These data raise the question of why subducted waters from the Southern Ocean containing apparently increasingly less Si during the onset of deglaciation also rapidly became less well ventilated at a time of pre-Bølling cooling and iceberg discharge in the N. Atlantic. Changes in source water dissolved O2 levels may result from significant changes in sea-ice cover, overturning rates or SST in the Southern Ocean between 17.6–15.5 Ka. One implication is that reduced ventilation coincided with a collapse of Atlantic meridional overturning during Heinrich event 1 [McManus et al., 2004], consistent with the bipolar seesaw. We speculate that a reduction in global ice volume and aridification continued outside the N. Atlantic basin, reducing dust (Fe) availability and triggering an early shift in ocean carbon cycling and pCO2.

[17] Since the data we present are limited to shallow locations on the Peru margin, it is unclear whether these results are representative of the deglacial of the whole Peru margin or Equatorial Pacific. Further work is needed to discover whether a similar pattern is observable in other marginal sedimentary deposits below upwelled intermediate waters principally fed by SAMW. Candidate locations include the Benguela and Cap Blanc upwelling regions off W Africa (following the oceanic distribution of Si(OH)4 deficits [Sarmiento et al., 2004]). Studies of similar upwelling environments are required to more thoroughly test the applicability of the SALH and conclusively extrapolate these results as part of a global marine biogeochemical explanation of the last deglaciation.

Acknowledgments

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Materials and Methods
  5. 3. Results
  6. 4. Discussion and Conclusions
  7. Acknowledgments
  8. References
  9. Supporting Information

[18] We thank A. Mix and B. Conard (Oregon State Univ.) for supplying samples and supporting data. The OSU core repository is supported by the U.S. National Science Foundation (NSF), grant OCE-0081247. We thank Z. Liu, P. Feng and R. Singh who assisted with sample preparation. We thank D. Montluçon, WHOI for supplying sterol standards. Funding for this research was provided by grants from the U.S. NSF, (OCE-0214365, OCE-0318371). This manuscript is UMD SMAST contribution 04-0602.

References

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Materials and Methods
  5. 3. Results
  6. 4. Discussion and Conclusions
  7. Acknowledgments
  8. References
  9. Supporting Information
  • Altabet, M. A. (1988), Variations in nitrogen isotopic composition between sinking and suspended particles: Implications for nitrogen cycling and particle transformation in the open ocean, Deep Sea Res., Part A, 35, 535554.
  • Archer, D., and E. Maier-Reimer (1994), Effect of deep-sea sedimentary calcite preservation on atmospheric CO2 concentration, Nature, 367, 26263.
  • Brzezinski, M. A. (1985), The Si:C:N ratio of marine diatoms: Interspecific variability and the effect of some environmental variable, J. Phycol., 21, 347357.
  • Brzezinski, M. A., C. J. Pride, V. M. Franck, D. M. Sigman, J. L. Sarmiento, K. Matsumoto, N. Gruber, G. H. Rau, and K. H. Coale (2002), A switch from Si(OH)4 to NO3 depletion in the glacial Southern Ocean, Geophys. Res. Lett., 29(12), 1564, doi:10.1029/2001GL014349.
  • Brzezinski, M. A., M.-L. Dickson, D. M. Nelson, and R. Sambrotto (2003), Ratios of Si, C and N uptake by microplankton in the Southern Ocean, Deep Sea Res., Part II, 50, 619633.
  • Calvo, E., C. Pelejero, G. A. Logan, and P. De Deckker (2004), Dust-induced changes in phytoplankton composition in the Tasman Sea during the last four glacial cycles, Paleoceanography, 19, PA2020, doi:10.1029/2003PA000992.
  • Canuel, E. A., and C. S. Martens (1996), Reactivity of recently deposited organic matter: Degradation of lipid compounds near the sediment-water interface, Geochim. Cosmochim. Acta, 60, 17931806.
  • Dugdale, R. C., and F. P. Wilkerson (2001), Sources and fates of silicon in the ocean: The role of diatoms in the climate and glacial cycles, Sci. Mar., 65, 141152.
  • Egge, J. K., and D. L. Asknes (1992), Silicate as regulating nutrient in phytoplankton competition, Mar. Ecol. Prog. Ser., 83, 281289.
  • Franck, V. M., M. A. Brzezinski, K. H. Coale, and D. M. Nelson (2000), Iron and silicic acid concentrations regulate Si uptake north and south of the Polar Frontal Zone in the Pacific Sector of the Southern Ocean, Deep Sea Res., Part II, 47, 33153338.
  • Higginson, M. J., and M. A. Altabet (2004), Comment on “Sedimentary phosphorus record from the Oman margin: New evidence of high productivity during glacial periods” by F. Tamburini et al. Paleoceanography, 19, PA2024, doi:10.1029/2004PA001017.
  • Langdon, C. (1988), On the causes of interspecific differences in the growth-irradiance relationship for phytoplankton. II. A general review, J. Plankton. Res., 10, 12911312.
  • Liu, K.-K., and I. R. Kaplan (1989), The eastern tropical Pacific as a source of 15N-enriched nitrate in seawater off southern California, Limnol. Oceanogr., 34, 820830.
  • Matsumoto, K., J. L. Sarmiento, and M. A. Brzezinski (2002), Silicic acid leakage from the Southern Ocean: A possible explanation for glacial atmospheric pCO2, Global Biogeochem. Cycles, 16, 3, 1031, doi:10.1029/2001GB001442.
  • McManus, J. F., R. Francois, J.-M. Gherardi, L. D. Keigwin, and S. Brown-Leger (2004), Collapse and rapid resumption of Atlantic meridional circulation linked to deglacial climate changes, Nature, 428, 834837.
  • Orcutt, D. M., and G. W. Patterson (1975), Sterol, fatty acid, and elemental composition of diatoms grown in chemically-defined media, Comp. Biochem. Physiol. B Comp. Biochem., 50, 579583.
  • Pelejero, C., J. O. Grimalt, M. Sarnthein, L. Wang, and J.-A. Flores (1999), Molecular biomarker record of sea surface temperature and climatic change in the South China Sea during the last 140,000 years, Mar. Geol., 156, 109121.
  • Petit, J. R., et al. (1999), Climate and atmospheric history of the past 420,000 years from the Vostok ice core, Antarctica, Nature, 399, 429436.
  • Pondaven, P., O. Ragueneau, P. Treguer, A. Hauvespre, L. Dezileau, and J. L. Reyss (2000), Resolving the “opal paradox” in the Southern Ocean, Nature, 405, 168172.
  • Sarmiento, J. L., N. Gruber, M. A. Brzezinski, and J. P. Dunne (2004), High-latitude controls of thermocline nutrients and low latitude productivity, Nature, 427, 5660.
  • Schöllhorn, E., and E. Granéli (1996), Influence of different nitrogen to silica ratios and artificial mixing on the structure of a summer phytoplankton community from the Swedish west coast (Gullmar fjord), J. Sea Res., 35, 159167.
  • Schubert, C. J., J. Villanueva, S. E. Calvert, G. L. Cowie, U. von Rad, H. Schulz, U. Berner, and H. Erlenkeuser (1998), Stable phytoplankton community structure in the Arabian Sea over the past 200,000 years, Nature, 394, 563566.
  • Shackleton, N. J. (2000), The 100,000-year ice age cycle identified and found to lag temperature, carbon dioxide, and orbital eccentricity, Nature, 289, 18971901.
  • Sigman, D. M., and E. A. Boyle (2000), Glacial/interglacial variations in atmospheric carbon dioxide, Nature, 407, 859869.
  • Takeda, S. (1998), Influence of iron availability on nutrient consumption ratio of diatoms in oceanic waters, Nature, 393, 774777.
  • Volkman, J. K. (1986), Review of sterol markers for marine and terrigenous organic matter, Org. Geochem., 9, 8399.
  • Volkman, J. K., S. M. Barrett, S. I. Blackburn, M. P. Mansour, E. L. Sikes, and F. Gelin (1998), Microalgal biomarkers: A review of recent research developments, Org. Geochem., 29, 11631179.
  • Werne, J. P., D. J. Hollander, T. W. Lyons, and L. C. Peterson (2000), Climate-induced variation in productivity and planktonic ecosystem structure from the Younger Dryas to Holocene in the Cariaco Basin, Venezuela, Paleoceanography, 15, 1929.

Supporting Information

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Materials and Methods
  5. 3. Results
  6. 4. Discussion and Conclusions
  7. Acknowledgments
  8. References
  9. Supporting Information

Auxiliary material for this article contains one text file describing additional details for lipid analysis and one text file containing age/depth data for all cores.

FilenameFormatSizeDescription
grl18611-sup-0001-README.txtplain text document2KREADME.txt
grl18611-sup-0002-METHODS.txtplain text document2KLipid Analysis
grl18611-sup-0003-C14.txtplain text document1KAge/Depth Data

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.