2002-10-15

Grant Young
Department of Earth Sciences, University of Western Ontario

Abstract:

Widespread occurrence of both Paleoproterozoic and Neoproterozoic glaciogenic deposits has recently been used as a springboard to launch the speculative snowball Earth hypothesis. Among Paleoproterozoic glacial deposits in North America, northwestern Europe, Western Australia and South Africa, the best preserved and documented are in the Huronian of Ontario, where three glaciogenic formations have been recognized. The youngest is the Gowganda Formation, which is interpreted as a passive margin deposit. On the south side of Lake Superior the oldest Paleoproterozoic succession (ChocolayGroup) begins with glaciogenic diamictites that have been correlated with the Gowganda Formation. The >2.2 Ga passive margin succession (Chocolay Group = upper Huronian) is overlain, with profound unconformity, by a 1.88 Ga foreland basin succession that includies the Superior-type BIF. Thus the glacial deposits of the Gowganda Formation were deposited on a newly formed passive margin and the Superior-type BIF was formed ~300 Ma later in a foreland basin setting, reflecting ocean closure.

In Western Australia, Paleoproterozoic glaciogenic deposits of the Meteorite Bore Member appear to form part of a foreland basin succession that includes BIF. The glaciogenic rocks are separated from underlying BIF by a thick siliciclastic succession. Thus there is no evidence of a genetic association between glaciation and BIF deposition as suggested by some proponents of the snowball Earth hypothesis (SEH). In both regions BIF formed in foreland basin settings but the iron-formations are of greatly different age, suggesting that the most significant control on their formation was not oxygenation of the atmosphere but rather, emplacement of Fe-rich waters (uplifted as a result of ocean floor destruction?) in a siliciclastic-starved environment (transition to a foreland basin setting) where oxidation could take place. Some of the Australian BIF's appear to pre-date the appearance of red beds in North American Paleoproterozoic successions.

Neoproterozoic glaciogenic deposits are widespread on the world's continents. Some are associated with iron-formations. Two theories have emerged to explain these enigmatic BIF's. According to the SEH, frozen oceans would have permitted buildup of dissolved Fe. Release of such iron would have taken place following re-introduction of oxygen to the hydrosphere as the ice cover disappeared. A second theory involves glaciation of Red Sea rift-type basins. Fe-charged brines in such basins would have precipitated on being mixed with "normal" sea water as a result of glacially-driven thermal overturn. Both theories provide an explanation of the hydrothermal imprint on the geochemistry of Neoproterozoic BIF but the restricted development of BIF (relative to glacial deposits), clear evidence of rift activity from dramatic facies and thickness changes, and associated volcanic rocks all favour deposition in a rift environment.

Cap carbonates are one of the cornerstones of the SEH. In Paleoproterozoic glaciogenic successions, with the possible exception of the Espanola Formation in the Huronian Supergroup, such carbonates are absent. Proponents of the SEH suggest that rapid escape from the snowball condition occurred as a result of buildup of high concentrations of CO2 in the atmosphere when the weathering cycle was stopped. Cap carbonates are supposed to represent extreme alakalization of the oceans as a result of weathering in the aftermath of the snowball condition. A corollary is that the first siliciclastic deposits following glaciation, should be extremely weathered, then should show a gradual return to more normal weathering conditions as the CO2 content of the atmosphere "normalized". The degree of weathering of siliciclastic materials can be estimated using a Chemical Index of Alteration (CIA). In the case of the Gowganda Formation and also a Neoproterozoic succession (with a cap carbonate) in Utah, the CIA shows that initial post-glacial weathering was low and gradually increased as climate ameliorated. These findings are the opposite of those predicted by the SEH.

An additional problem with the SEH is the development of thick successions (in some cases several km) of glaciogenic rocks, many of which contain evidence of waterlain deposits, more in keeping with a temperate glacial setting than one of extremely frigid conditions and virtual elimination of the hydrologic cycle. Rapid onset (runaway albedo) and demise (critical levels of atmospheric CO2) of glaciation are predicted by SEH advocates but the rocks below and above the Gowganda Formation contain geochemical evidence of gradual onset and end to frigid conditions. Likewise some Neoproterozoic successions (e.g. in the Sturtian of Australia) terminate with great thicknesses (~ 1 km) of dropstone-bearing mudsones, attesting to a slow disapperance of glacial ice. World-wide distribution of Neoproterozoic glaciogenic deposits is commonly cited in the literature but it should be kept in mind that although the continents are currently scattered across the face of the globe, their contained glacial deposits may have been much more latitudinally restricted if they formed on a single supercontinent (e.g. Rodinia).

There is no doubt that the earth underwent severe climatic perturbations both at the beginning and end of the Proterozoic Eon but whether the Earth attained a totally frozen surface condition (snowball Earth hypothesis) remains speculative.