386 nature geoscience | VOL 2 | JUNE 2009 | www.nature.com/naturegeoscience

news & views

whether this is an isolated case, or whether biogenic ice nuclei are a missing link in biosphere–climate interactions, which are of global importance. In a related paper, on page 402 of this issue, Prenni et al. show that biogenic particles may contribute to ice formation in mixed-phase clouds in the Amazon5, indicating that the findings of Pratt and colleagues may be the tip of the iceberg.

The work of Pratt et al.1 strongly suggests that biogenic material can influence cloud

ice formation in the real atmosphere. Further field studies investigating ice nuclei composition in situ, together with a refinement of methodological techniques, are now needed to verify the global significance of these findings. If biological particles do prove to be important in cloud–ice interactions, then we need to find a way of incorporating the source, transport and nucleation efficiency of these particles into global climate and smaller-scale models. ❐

Corinna Hoose is at the University of Oslo, Department of Geosciences, PO Box 1022, Blindern, 0315 Oslo, Norway. e-mail: corinna.hoose@geo.uio.no

References1. Pratt, K. et al. Nature Geosci. 2, 398–401 (2009).2. Möhler, O., DeMott, P. J., Vali, G. & Levin, Z. Biogeosciences

4, 1059–1071 (2007).3. Cziczo, D. J., Murphy, D. M., Hudson, P. K. & Thomson, D. S.

J. Geophys. Res. 109, D04201 (2004).4. Richardson, M. S. et al. J. Geophys. Res. 112, D02209 (2004).5. Prenni, A. J. et al. Nature Geosci. 2, 402–405 (2009).

The phrase ‘Snowball Earth’ conjures images of an icy, desolate planet where life would have been challenging.

But this image may not be correct. Most agree that ice reached equatorial latitudes in Neoproterozoic time, between about 726 million and 635 million years ago, but the actual extent of ice cover and its effects on the biosphere are still hotly debated. New data reported by Nagy and co-authors1 on page 415 of this issue reveal that the phytoplankton extinction once attributed to the effects of Snowball Earth2 actually pre-dates the glaciation by about 16 million years, adding to the growing body of evidence that suggests life was not

as severely affected by Snowball Earth as once thought.

The ambiguity surrounding life during Snowball Earth, when glaciers, ice sheets and sea ice are proposed to have covered almost the entire globe, arises, in part, from the nature of the fossil record itself. Animals with shells, which are more easily preserved in the fossil record, did not evolve until about 90 million years after the final snowball glaciation. Before this evolutionary jump, nearly all life was microscopic and soft-bodied — not a good recipe for creating a fossil record. Special circumstances must conspire to preserve the organic-walled plankton common to

the Neoproterozoic oceans as microfossils: these rare localities are usually rich in organic material, and must not have experienced too much heat or pressure over the ensuing millions of years after deposition. Other microfossil coffers are found in rocks that underwent early silicification, entombing the microfossils in silica before they decayed. Few Neoproterozoic deposits meet these criteria; the common glacial deposits of this era will not preserve organic-walled microfossils very well, if at all.

Unsurprisingly, the record of microfossils in Neoproterozoic glacial deposits is relatively scarce, but the rare deposits that do preserve organic-walled microfossils in shales record assemblages composed almost entirely of Sphaerocongregus or Bavlinella2,3. This is commonly interpreted as a disaster assemblage of bacteria, thought to take over when an earlier, more diverse assemblage of eukaryotic phytoplankton becomes extinct, presumably because of some calamity. Naturally, the disaster was assumed to be global glaciation.

However, Nagy et al. report that the Chuar Group found in the Grand Canyon region of the southwestern US provides a very different snapshot of Neoproterozoic life1. The rocks were deposited just before glaciation (between about 770 million and 742 million years ago), and are exceptionally well preserved. Even better, they are rich in organic material and have never reached temperatures that would have destroyed organic-walled microfossils. Nagy and co-authors show that a similar shift in microfossil assemblage occurs in this unit as well. Older units reveal a diverse

palaeonToloGy

extinction before the snowballA widespread biotic turnover occurred around the time of the Sturtian glaciation. Microfossil analyses show that one regional extinction pre-dates the glacial advance, challenging the more severe models for glacial effects in the Neoproterozoic era.

Frank a. corsetti

a b

Figure 1 | Microfossils from the Chuar Group. Nagy and co-authors1 report a diverse assemblage of fossils from organic-walled plankton including Pterospermopsimorpha insolita (a) was replaced by a low-diversity, bacterial-dominated assemblage including Sphaerocongregus variabilis (b). The shift occurred about 16 million years before the onset of low-latitude glaciation associated with Snowball Earth, ruling out glaciation as a direct cause.

N&V's June run on.indd 386 19/5/09 11:06:26

© 2009 Macmillan Publishers Limited. All rights reserved

nature geoscience | VOL 2 | JUNE 2009 | www.nature.com/naturegeoscience 387

news & views

In the summer of 2008, all eyes were on Beijing. Host to the Olympic and Paralympic Games, China was keen to put on a spectacular show. However, images of athletes arriving in the city wearing face masks and stadiums shrouded in smog cast a shadow over events.

Rapid industrialization has resulted in severe pollution problems in China. Industrial waste spills into rivers, heavy metals weigh down the soils and acid rain falls from the skies. One of the greatest environmental and human health concerns is the level of atmospheric pollutants, a particular problem in China’s cities, where concentrations of ozone and particulate matter often exceed World Health Organization guidelines. Thus when Beijing was chosen as host for the 2008 Olympic Games, concerns — particularly about the health of the athletes — ran high.

In an attempt to rectify the problem and ease some of these concerns, factories were shut down, construction works were stopped and half of the city’s 3.3 million cars were banned from the roads. In total, the Chinese government spent US$17.58 billion on pollution reduction measures. But did they have any effect?

Jan Cermak and Reto Knutti, of the Institute of Atmospheric and Climate Science, Zurich, argue that they did (Geophys. Res. Lett. doi: 10.1029/2009GL038572;

Brief reliefaTmospheRic polluTion

2009). They compared satellite observations of aerosol levels over Beijing during the games, with model predictions of the aerosol burden assuming no emission controls, based on meteorological conditions. Observed aerosol levels were ~15% lower than model projections. Although the reduction is low compared with natural variability, it is noticeable, and suggests that pollution controls were a partial success.

However, despite massive emission controls hundreds of kilometres outside the capital, aerosol reductions were confined to Beijing and its immediate surroundings. Outside this area, control measures had little effect.

The magnitude and scale of the reduction is disappointing, although expected

given that regional transport is a major determinant of atmospheric aerosol levels. But it is encouraging to note just how quickly the benefits of pollution control measures are realized (many of the controls were implemented just months before the games began).

With plans to make the 2012 Olympic Games carbon neutral, London is next in sight. Whether the city can pull off plans to make these games the greenest in history remains to be seen. But, like in Beijing, the event could prove a test bed for the effectiveness of the chosen environmental improvement schemes.

anna aRmsTRonG

assemblage of spiny, organic-walled plankton (Fig. 1), which gives way to a low-diversity assemblage of simpler forms in the younger units, a presumable disaster assemblage.

The turnover itself is not surprising. But Nagy and co-authors reveal that the switch to the low-diversity assemblage occurred some 16 million years before the glaciation, leaving the supposed cause and effect — glaciation and biotic turnover — clearly decoupled. Thus, glaciation need not be invoked to force the phytoplankton turnover from a diverse to an impoverished assemblage.

Although the work by Nagy et al. demonstrates that glaciation is not necessary to cause a phytoplankton extinction in the Neoproterozoic, it does not preclude snowball glaciation triggering later extinctions. But other snapshots of Neoproterozoic life do not necessarily support such biotic upheaval. A microfossil assemblage from Death Valley4, deposited at the time of purported global glaciation, reveals a complex ecosystem

(complex for the Precambrian, at least). The assemblage comprises both photosynthetic and heterotrophic eukaryotes, as well as a host of fossils interpreted as cyanobacteria and other bacteria, which apparently survived during the glaciation. The presence of photosynthesizers also indicates that sunlight reached their environment, placing limits on the presence or thickness of sea ice concurrent with the widespread glaciation.

Organic biomarkers — chemical remains preserved in rocks and sediments that indicate the presence of certain groups of organisms — reveal very little evidence for disaster assemblages during glaciation. Biomarkers from glacial strata in Brazil5 indicate a glacial ecosystem that included eukaryotes and a variety of metabolisms including photosynthesis, once again placing limits on extent and thickness of sea ice. Perhaps even more exciting, sponge biomarkers were found recently in glacial deposits from Oman6, providing some of

the oldest tangible evidence for animals on the Earth, and indicating that even animals survived Snowball Earth.

For me, the work by Nagy and colleagues1 adds to the growing evidence that the conditions for life during Neoproterozoic Snowball Earth were not as severe as some would suggest. ❐

Frank A. Corsetti is at the Department of Earth Sciences, University of Southern California, Los Angeles, California 90089-0740, USA. e-mail: fcorsett@usc.edu

References1. Nagy, R. M., Porter, S. M., Dehler, C. M. & Shen, Y.

Nature Geosci. 2, 415–418 (2009).2. Knoll, A. H., Blick, N. & Awramik, S. M. Am. J. Sci.

281, 247–263 (1981).3. Moczydłowska, M. Precambrian Res. 167, 1–15 (2008).4. Corsetti, F. A., Awramik, S. M. & Pierce, D. Proc. Natl Acad. Sci.

USA 100, 4399–4404 (2003).5. Olcott, A., Sessions, A., Corsetti, F. A., Kaufman, A. J. &

de Oliviera, F. Science 310, 5747–5751 (2005).6. Love, G. et al. Nature 457, 718–721 (2009).

Pa

N&V's June run on.indd 387 19/5/09 11:06:27

© 2009 Macmillan Publishers Limited. All rights reserved