Polar twins

news & views palaEOcEaNOgraphy polar twins Ice ages in the North Pacific Ocean and the Southern Ocean were marked by low productivity. Accumulating evidence indicates that strong stratification restricted the supply of nutrients from the deep ocean to the algae of the sunlit surface in these regions. gerald h. haug and Daniel m. sigman 1.5 40 Opal concentration (wt %) Biogenic opal (MAR, g cm–2 kyr–1) 6 4 1 20 0.5 0 Warm (interglacial) 8 10 Cold (glacial) 12 14 16 Age (thousand years) 18 0 20 Opal flux (Th-norm MAR, g cm–2 kyr–1) t he North Pacific and Southern oceans, on opposite sides of the Earth, represent remarkably similar polar ocean environments. Both are characterized by net upwelling of deep water, high concentrations of surface nutrients, high productivity and a low-salinity cap that moderates vertical mixing. Furthermore, both regions have the potential to affect the global ocean’s biological pump, which removes carbon dioxide from the atmosphere and stores it in the deep ocean. These two polar twins have also responded similarly to past climate changes1–9 — in both domains productivity was lower during ice ages1,2,5,6,9. This apparently resulted from a reduced supply of waterborne nutrients into the sunlit surface layer, associated with stronger density stratification of the upper water column or a weakening in the physical drivers of vertical mixing 1–6. Recent work by Gebhardt and colleagues1 caps off more than a decade of work on the North Pacific Ocean — the northern twin — with high-resolution data documenting the weakening of stratification at the end of the past four ice ages. The North Pacific is the meeker of the polar twins, protected and constrained by continents on three sides, whereas the Southern Ocean runs free in its open circumpolar-channel. Only modest amounts of intermediate water form in the North Pacific, and winter surface waters do not become dense enough to form deep water. Wind-driven upwelling and downwelling are stronger in the Southern Ocean, and both intermediatedepth and abyssal waters are formed. During the ice ages, stronger stratification in the Antarctic zone of the Southern Ocean may have spread to the coastal regions of Antarctica, reducing the deep- water formation that now occurs there. The reduction in atmospheric carbon dioxide during the ice ages can go some way towards explaining the decrease in upwelling of carbon-dioxide-rich deep water and the decrease in pumping of unused dissolved nutrients back into the deep ocean2. In contrast, the stratification changes in the North Pacific Ocean are probably not 2 Pliocene warm interval 0 0 0.5 1 1.5 2 Age (million years) 2.5 3 3.5 4 Figure 1 | Productivity in the North Pacific Ocean. Ocean productivity at site 882 — shown here by its proxy of opal mass accumulation rate (MAR) in a sediment core — declined dramatically in the North Pacific Ocean 2.7 million years ago, at the onset of Northern Hemisphere glaciation3. This indicates deterioration in the upwelling of nutrient-rich deep water to the surface. Inset: During and immediately after the last deglaciation, about 15,000 years ago, there is a peak in opal mass accumulation, or flux (green) and concentrations (blue) in the North Pacific sediment core6. Gebhardt and colleagues1 attribute this to a better supply of nutrients to the surface, similar to conditions before the past 2.7 million years. important in ventilation of the deep ocean or in glacial–interglacial carbon dioxide changes: because there is no deep-water formation there today, there is nothing that could have been turned off during ice ages. So why study the North Pacific Ocean? There are more tools available for reconstructing the past in the North Pacific than in the Southern Ocean. The Southern Ocean, for all its brawn, is an extremely poor communicator: its sediment records are difficult to date and can be highly variable among sites, buffeted as they are by the strong Southern Ocean currents. Moreover, the extremely cold and nutrient-rich conditions of the Antarctic region discourage the calcite-producing organisms that are required by many of our palaeontological and geochemical tools for reconstructing past conditions. The northern twin, true to its stereotype, is far more articulate. As Gebhardt and colleagues1 demonstrate, the carbonate microfossils in its sediment records clearly identify the major ice ages and interglacials and also allow for reconstruction of past temperature changes. In line with previous work1, Gebhardt and colleagues show that summer surface-temperatures in the open subarctic Pacific Ocean were well above freezing, ruling out sea-ice cover as the cause of lower productivity during the ice ages5,6 (Fig. 1). This finding strengthens the interpretation of North Pacific (and Southern Ocean) stratification during ice ages1–5. Likewise, radiocarbon measurements from the shells of organisms living on the deep-sea floor indicate that North Pacific abyssal waters were ventilated more slowly nature geoscience | VOL 2 | FEBRUARY 2009 | www.nature.com/naturegeoscience © 2009 Macmillan Publishers Limited. All rights reserved. 91 news & views during the last ice age relative to today 8,10. Decreased deep-ocean ventilation rules out North Pacific deep-water formation during the last ice age, and also suggests that Antarctic deep-water formation was reduced. Looking further back in time, we find that the onset of major ice-age cycles approximately 2.7 million years ago coincided with a sharp drop in the flux of biogenic detritus in both the North Pacific (Fig. 1) and the Antarctic3,4. Thus, the apparent link between cooling and stratification indicated by data from the most recent glacial cycles, also applies to other climate transitions. Early work in the North Pacific indicated that the biogenic flux during interglacials, although higher than in the ice ages, did not reach pre-2.7 million year levels3 (Fig. 1). However, North Pacific sediment records show an enigmatic peak in biogenic flux at each of the large deglaciations during the past 2.7 million years (Fig. 1, inset). In their detailed study of these events, Gebhardt and colleagues1 argue that the peaks represent a brief reversion to the conditions before the past 2.7 million years of extremely weak stratification and high biogenic flux. If this interpretation of the early deglacial intervals is correct, then the North Pacific not only resembled its more potent state three million years ago, but also came closer to the modern behaviour of its more muscular southern twin. The past changes in the polar twins may have much to tell us about their future. Climate models of anthropogenic warming predict an increase in the stratification of these regions, with consequences for the productivity and fluxes of energy and carbon dioxide11. Yet the accumulated palaeoclimate data from the polar twins seems to argue for the opposite response, with stronger stratification in cold climates, not warm ones. One would tend to put more faith in the model results, as they are the outcome of much work and thought. Moreover, the sensitivities of the models can be rationalized in terms of relatively simple dynamics: for example, warming tends to strengthen the poleward transport of water vapour through the atmosphere, which should work to stratify the two polar twins11. In contrast, the postulated tendency for polar-ocean stratification to increase as global temperatures fall is much harder to explain, and has never been adequately simulated. However, ongoing changes in the Southern Ocean seem to contradict the anthropogenic warming simulations, and instead fit with the responses found in the palaeoclimate data12. Accumulating palaeoceanographic evidence, including this latest work by Gebhardt and colleagues, indicate that, in the past, the polar twins have responded similarly to climate change. Is it just a matter of time before the North Pacific follows its braver twin and contradicts our modelbased expectations? ❐ Gerald H. Haug 1,2 and Daniel M. Sigman3 are at the 1Geological Institute, Department of Earth Sciences, ETH-Zentrum, 8092 Zürich, Switzerland; 2 Leibniz Center for Earth Surface and Climate Studies, Institute for Geosciences Potsdam University, 14476 Potsdam, Germany; 3Department of Geosciences, Princeton University, Princeton, New Jersey, 08544, USA. e-mail: gerald.haug@erdw.ethz.ch references 1. Gebhardt, H. et al. Paleoceanography 23, doi: 10.1029/2007PA001513 (2008). 2. Francois, R. et al. Nature 389, 929–935 (1997). 3. Haug, G. H., Sigman, D. M., Tiedemann, R., Pedersen, T. F. & Sarnthein, M. Nature 401, 779–782 (1999). 4. Sigman, D. M., Jaccard, S. L. & Haug, G. H. Nature 428, 59–63 (2004). 5. Jaccard, S. L. et al. Science 308, 1003–1006 (2005). 6. Jaccard, S. L. et al. Earth Planet. Sci. Lett. 277, 156–165 (2009). 7. Brunelle, B. G. et al. Paleoceanography 22, doi: 10.1029/2005PA001205 (2007). 8. Galbraith, E. D. et al. Nature 449, 890–894 (2007). 9. Keigwin, L. D. Paleoceanography 13, 323–339 (1998). 10. Sarnthein, M., Grootes, P. M., Kennett, J. P. & Nadeau, M. J. Geophys. Monograph Series 173, 175–196 (2007). 11. Sarmiento, J. L., Hughes, T. M. C., Stouffer, R. J. & Manabe, S. Nature 393, 245 (1998). 12. Toggweiler, J. R. & Russell, J. Nature 451, 286 (2008). BIOgEOchEmIstry mercury methylation made easy The exact mechanism used by microorganisms to produce the neurotoxin methyl mercury is unclear. The latest laboratory studies point to the amino acid cysteine as an important aid for the uptake of inorganic mercury and its transformation to methyl mercury in Geobacter sulfurreducens. richard sparling t he cycling of mercury in aquatic environments is important because it is extremely toxic in organic form. The most potent organic mercury compound is methyl mercury — a neurotoxin that accumulates in aquatic food chains. The transformation of inorganic mercury to methyl mercury is primarily mediated by aquatic microorganisms. How these microorganisms take up mercury, and how they convert it to methyl mercury — a process known as mercury methylation — is one of the last remaining uncertainties in the biogeochemical mercury cycle. On page 105 of this issue, Schaefer and Morel1 show that complexation of inorganic 92 mercury with the amino acid cysteine significantly enhances mercury uptake and the rate of methylation in the iron-reducing bacterium Geobacter sulfurreducens (Fig. 1). It has been known for a long time that microorganisms are crucial for the production of methyl mercury, and several strains of bacteria have been isolated that are capable of methylating mercury when exposed to environmentally relevant inorganic mercury concentrations2. Most of these organisms fall into two distinct, but related groups: sulphate reducing bacteria3 and iron-reducing bacteria4. It was originally thought that inorganic mercury entered these bacteria in the form of an uncharged, hydrophobic molecule (such as mercury dichloride or mercury sulphide) that passively diffused across the cell membrane2. However, since then it has been shown that microbial mercury uptake is enhanced at low pH and in the presence of certain amino acids; conditions in which passive diffusion is less probable as uncharged molecules containing mercury are not expected to dominate5,6. In addition, the influence of growth conditions on uptake rates in these experiments indicates that microbial physiology (also influenced by the growth environment) may exert more control over uptake than previously thought. nature geoscience | VOL 2 | FEBRUARY 2009 | www.nature.com/naturegeoscience © 2009 Macmillan Publishers Limited. All rights reserved.