Australia: The Land Where Time Began

A biography of the Australian continent 

Snowball Earth – Cooling Following Algal Rise

Between 720 Ma and 635 Ma the Earth underwent 2 snowball glaciation events. Feulner et al. suggest that preceding these snowball events the eukaryotic algae expansion and the consequent rise in emissions of organic cloud condensation nuclei may have contributed to the dramatic cooling.

The climate of the Earth has been affected by repeated periods of substantial glaciation throughout its almost 4.6 Gyr history. Ice reached tropical latitudes during the largest of these, forming a snowball Earth glaciation (Hoffman & Schrag, 2002).

The earliest of these severe glaciations occurred at a time when the Earth was more prone to cooling than at present because of the low luminosity of the so-called faint young Sun (Feulner, 2012). Rocks from the Archaean, 3.8-2.5 Ga, that are found on most continents, contain evidence of these early glaciations (Tang & Chen, 2013). Huronian glaciations of Archaean age, about 2.4-2.1 Ga, have been linked tentatively to the breakdown of atmospheric methane, a greenhouse gas, in the wake of the Great Oxidation Event (Kopp et al., 2005). At about 2.9 Ga the Pongola Glaciation, an earlier cooling event, occurred (Young et al., 1998), though its extent has remained unclear. Following these early climatic fluctuations there was a period of about 1 Gyr when no evidence has been found of significant glaciation, in spite of a sedimentary record that has been preserved more completely. The lack of significant carbon isotope excursions during this interval is a further indication that was environmental stability during the ‘boring billion’, which Feulner et al. suggest may reflect the evolutionary stasis that has been observed during the Mesoproterozoic (Lyons et al., 2014).

Cryogenian glaciations

A series of major glaciations during the Cryogenian, 720-635 Ma, of the Neoproterozoic Era, 1 Ga to 541 Ma, abruptly ended this long-lasting era of climatic stability. Included in these cold spells, the iconic Sturtian Glaciation, about (717-662 Ma) and the Marinoan Glaciation, about (639-635 Ma) Snowball Earth events (Hoffman et. al., 2014; Lyons et al., 2014). Most of the hypothesised mechanisms to explain these glaciations of the Neoproterozoic focus on the drawdown of atmospheric CO₂ by enhanced weathering of tropical continents (Hoffman & Schrag, 2002) or flood basalts (Rooney et al., 2014; Godderis et al., 2003). Though Feulner et al. suggest similar configurations of continents should have occurred prior to the snowball glaciations. These hypotheses, therefore, cannot explain fully the sudden susceptibility of Earth system to glaciation during the Neoproterozoic, following more than 1 Gyr of stability.

Also, these hypotheses work only if the climate was already cool prior to the onset of continental weathering (Godderis et al., 2003).

Aerosols, clouds and climate

The concentrations of greenhouse gases, and the reflectivity of the Earth’s surface, clouds, as well as solar luminosity, are crucial for determination of the planetary balance. Concentration changes and changes in the nature of atmospheric aerosols can alter the optical properties (Twomey, 1977) and lifetime (Albrecht, 1989) of clouds and therefor the energy budget of the Earth. Specifically more cloud condensation nuclei are provided by increased concentrations of aerosol, the result of which in turn is smaller cloud droplets and therefore clouds that reflect more short wave radiation back to space (Twomey, 1977). A longer atmospheric residence time is exhibited by these smaller droplets before they are rained out and therefore have a long lifetime (Albrecht, 1989). Therefore, increased concentration of cloud condensation nuclei will cool the climate of the Earth.

Eukaryotic algae in the modern ocean are believed to be the principal source of cloud concentration nuclei over the marine realm (Simό, 2001) by their production of dimethylsulfoniopropionate (DMSP): on senescence and cell death, undergoes conversion mediated by bacteria to the more commonly known organic aerosol dimethyl sulphide (DMS). Algal primary producers are implied to be a key feedback in modern climate systems as the marine flux exceeds volcanic sulphur emissions (Charlson et al., 1987).

According to Feulner et al. the rise of eukaryotic algae to ecological prominence can be dated to about 800-750 Ma (Knoll et al., 2006; Knoll, 2014; Parfrey et al., 2011). In this paper Feulner et al. explore the hypothesis that a resulting increase in cloud condensation nuclei could have made the climate system of the Cryogenian much more vulnerable to major glaciations.

Palaeobiological clues

Eukaryotes are indicated by organic walled microfossils to have existed since at least 1.5 Ga (Knoll et al., 2006), though for some time they were probably not ecologically significant. During the Early to Middle Neoproterozoic about 800-750 Ma (Knoll et al., 2006) the abundance and diversity of eukaryotic microfossils increased, and a major diversification of eukaryotes is suggested by molecular clocks to have occurred at about 800 Ma (Knoll, 2014; Parfrey et al., 2011). The oldest known unambiguously synergetic common steranes, which are biomarkers of significant eukaryotic abundance, are present in sediments dated to 742 Ma (Summons et al., 1988). Therefore, a shift to more prevalent eukaryotic algae appears to immediately predate the onset of the Sturtian Glaciation.

In alveolate protists and haptophyte algae the production of DMSP is particularly prominent. Though both lineages branched early in the history of the Earth (Parfrey et al., 2011), it is not clear if the early forms of these modern algae had the ability to synthesise DMSP. To date there is insufficient detail to identify the phylogenetic distribution of DMSP synthesis with public sequence databases, though the enzymatic steps of the DMSP biosynthetic pathway are known from Ulvophyceae (Stefels, 2000) and related candidate genes have been identified tentatively (Lyon et al., 2011). Feulner et al. used instead information on the presence of intracellular DMSP (Keller, 1989; Keller, Bellows & Guillard, 1989) that was derived from literature to infer the capacity for the biosynthesis in various extant algae and combined this information with time-calibrated tree of eukaryotes that was published most recently (Parfrey et al., 2011).

Oxyrrhis marina is the only species of all the species of dinoflagellate considered in ref. (Parfrey et al., 2011), whose ancestor diverged early around 670 Ma (Parfrey et al., 2011), lacks DMSP (Keller, Bellows & Guillard, 1989), which suggests that the biosynthesis of DMSP emerged later in the alveolate lineage. This capacity suggested by the production of active DMSP in all the haplotypes used in the time-calibrated tree (Keller, Bellows & Guillard, 1989) that the capacity could have emerged between the early Mesoproterozoic branching of ancestral haplotypes and the divergence of prymnesiophytes and pavlovophytes around 675 Ma (Parfrey et al., 2011). The proto-haptophytes should, therefore, to have acquired the capacity to produce DMSP earlier than the onset of the Sturtian glaciation.

Therefore, it is likely that there was an increase in ocean-atmosphere fluxes of DMS from near zero to higher values during the Early Neoproterozoic, which is in line with the expansion of eukaryotes. Also, the sulphate pool of oceans during the Neoproterozoic was significantly smaller than the present (Kah, Lyons & Frank, 2004), so there would have been only low contributions from sea spray to sulphur aerosols. Therefore, the introduction and intensification of the production DMSP by algae, and ultimately cloud condensation nuclei, would have had a greater relative cooling of the climate than if such a cooling occurred at present.

The climate may have additionally been affected indirectly by the enhanced production of biogenic sulphur aerosols by alteration of the concentration of atmospheric CO₂: at present areas that are not subjected to substantial emissions of anthropogenic particles, the oxidation products of DMS are responsible for up to 40% of total rain acidity (Nguyen, 1992). At the onset of the Cryogenian an increase in the fluxes of atmospheric DMS could have acidified rainwater and enhanced weathering of the continents, promoting the drawdown of CO₂, in particular through the weathering of large basaltic provinces, an effect that would have pushed the climate system closer to the threshold for snowball glaciation (Rooney et al., 2014; Godderis et al., 2003).

Climate model simulations

In order to test the degree of cooling that would have resulted from an increase in cloud condensation nuclei from the expansion of algae could have contributed in the early Cryogenian, Feulner et al. carried out simulation experiments with a coupled climate model that used realistic boundary conditions for 720 Ma (Feulner & Kienert, 2014). In order to quantify possible cooling that was associated with the increase in diversity and abundance of eukaryotic algae, 2 sets of simulations were carried out by Feulner et al.:

1)      The first set used modern cloud properties as an analogue for high eukaryote abundance in the lead up to the glaciations of the Cryogenian;

2)      The second, following ref. Kump & Pollard; 2008, used clouds that were representative of pristine, non-productive ocean regions that were intended to simulate the situation prior to the increase in cloud condensation nuclei that were produced by eukaryotic algae.

Both sets of simulations were run at atmospheric concentrations of CO₂ that ranged from 10 to 1,000 ppm by volume (ppmv).

In the simulations that used cloud characteristics of the present, the model experiment with an atmospheric concentration of CO₂ of 110 ppmv represents the coldest climate state with areas of open ocean that were not frozen (Feulner & Kienert, 2014). This scenario has a global and annual mean surface air temperature of 226 K. In simulations which used lower concentrations of CO₂ and modern clouds, the Earth was in a completely covered snowball state. This value of 110 ppmv for the critical concentration of CO₂ is in good agreement with published values that were derived from more sophisticated climate models (see discussion ref. Feulner & Kienert, 2014).

Simulations that used low levels of cloud condensation nuclei, which was representative of the period prior to the rise of eukaryotic algae about 800 Ma, show a warmer state that was warmer than simulations that used cloud characteristics at similar atmospheric CO₂ concentrations. Using this scenario there is a global and annual mean surface air temperature of 266 K. The Earth plunges into a state of completely ice covered snowball conditions in simulations with lower concentrations of co₂ and modern clouds. This value of 110 ppmv for the critical concentration of CO₂ is in good agreement with values that have been published that are derived from more sophisticated climate models that have been published (see discussion in ref. Feulner & Kienert, 2014).

A warmer climate state is shown for simulations that use lower levels of cloud condensation nuclei, which are representative of the period prior to the rise of the eukaryotic algae about 800 Ma, than simulations using cloud characteristics of the present at similar atmospheric CO₂ concentrations. The scenario that used 110 ppmv concentrations of CO₂ and low amounts of cloud condensation nuclei yields a global temperature of 276 K – i.e., 10 K higher than cloud of modern characteristic. According to Feulner et al. this difference in temperature will be model-dependent to some degree, as there is a considerable degree of variation in simulated cloud cover and indirect effects of aerosol between different climate models. However, the increase in temperature due to the changes in cloud properties that are reported here agrees with results that have been obtained with other climate models for different time periods, and with a simple estimate that is based on planetary energy balance.

With atmospheric concentrations of CO₂ as low as 15 ppmv, a global glaciation will not occur, if concentrations of cloud condensation nuclei are low. Feulner et al. therefore suggest that it is exceedingly unlikely that the Earth would enter a snowball regime at Neoproterozoic boundary conditions prior to the rise of eukaryotic algae, that produce DMSP, to global ecological significance. The assumption that before the pulse of algal diversification and abundance, areas of ocean were similar to pristine, marine regions that are not productive at the present seem to be reasonable, though it is not clear whether aerosol (and therefore cloud condensation nuclei) concentrations reached the modern levels that were used in the simulations. A significant contribution of to the cooling is, nevertheless, likely for 2 reasons:

1)    The magnitude of the diversity and abundance of eukaryotic algae rise that is observed in the microfossil record (Knoll et al., 2006) makes a marked increase in cloud condensation nuclei concentrations in the early Cryogenian plausible.

2)    The sensitivity of the climate experiments for the 110 ppmv CO₂ simulation show that even for an increase to 25%, 50% and 75% of modern concentrations of cloud condensation nuclei, the cooling amounts to 2 K, 4K and 6 K, respectively, with modern levels resulting in a 10 K cooling.

Therefore, any cooling that is associated with the rise of eukaryotic algae was probably sufficient to play a role in the subsequent glaciations of the Neoproterozoic, even with conservative estimates.

Implications for initiation of snowball conditions

The point of initiation of snowball has recently been found to coincide with comparatively low critical CO₂ levels, even for modern concentrations of cloud condensation nuclei, and this could possibly be a challenge for scenarios of the initiation of the Snowball Earth events of the Cryogenian, as such a strong, rapid drawdown of CO₂ may be hard to achieve. If cloud condensation nuclei did not reach modern levels in the early Neoproterozoic this would be even more difficult. The cooling effects of increasing cloud condensation nuclei that is reported in this paper therefore do not mitigate the need for the drawdown of CO₂ to trigger the onset of a Snowball Earth event, i.e., the weathering of fresh basalts that were emplaced by igneous provinces (Rooney et al., 2014; Godderis et al., 2003). Though prior cooling does reduce the amount of the CO₂ drawdown that is required, and it may be required to make feasible the drawdown mechanism that was proposed previously.

The identification by Feulner et al. of the climate consequences of cloud condensation nuclei from the ecological rise of eukaryotic algae that produce DMSP adds further to the known role of evolution in the shaping of the Earth system, as well as offering an explanation for the climate stability that lasted 1 Gyr preceding the glaciations of the Cryogenian Snowball Earth. There are still some fundamental pieces missing of the Neoproterozoic climate puzzle, but Feulner et al. are expecting that a stronger focus on biological and environmental coevolution throughout the Neoproterozoic should continue to shed light on the glaciations of Snowball Earth which have remained one of the true enigmas in Palaeoclimatology.

Sources & Further reading

Feulner, G., et al. (2015). "Snowball cooling after algal rise." Nature Geosci 8(9): 659-662.