The Earth covered with ice

Some 630 million years ago, the Amazon was a desert lying below an icecap

Instead of dense forests with rubber trees, nut trees and other trees of dozens of meters in height, the Amazon of 630 million years ago must have been a vast plain covered by a layer of ice of up to 1 kilometer in thickness, surrounded by seas with ice flows and icebergs. Its location also seems to have been different from what had been imagined. According to the reconstruction of the Earth some thirty years ago by the Canadian geologist Paul Hoffman, from Harvard University, in the United States, the Amazon had at that time been close to the South Pole. But geologists from the states of São Paulo, Amazonas, Pernambuco and Pará are questioning this classical geology model and are revealing that the Amazon had, in truth, been close to the tropic of Capricorn – distant from the previous position by at least 4,500 kilometers, almost one and a half times the distance of Manaus to Sao Paulo in a straight line – where the temperatures must have been much higher.

“The Amazon was lacated in an intertropical region at 22° of latitude south, in a position similar to the current position occupied by the State of São Paulo”, says Afonso Nogueira, a geologist from the Federal University of Amazonas (Ufam), one of the team’s members that was able to count upon researchers from the Geosciences Institute, the Astronomy, Geophysics and Atmospheric Sciences Institute of the University of Sao Paulo (USP), as well as the federal universities of Pernambuco (UFPE) and Para (UFPA). But how could there have been ice so close to the tropics? It was not difficult to find an explanation for this apparent paradox. “Discoveries on Brazilian territory reinforce the idea of a snowball Earth, a theory that postulates that twice, one of them around 630 million years ago, all of the terrestrial surface lay beneath ice, making the Earth an immense snowball”, says he geologist Claudio Riccomini, the team’s coordinator from the Geosciences Institute of USP.

In that era the formation of the super continent known as Gondwana was taking place, which would bring together rocky blocks known as cratons that correspond to the current South America, Africa, Antarctica, Australia and India. One of these the Amazon craton, which moved its position, was made up of the current states of Rondônia, Amazonas, Roraima and Amapá, the northern part of Mato Grosso and the western part of Para, as well as the Guianas, Suriname and part of Bolivia.

The clues that there had been an ice cap also on the Amazon craton were given by diamictic glaciers. Formed by mud and sand, these rocks in general present a gray coloration and include fragments of older rocks, coming from remote regions. Some of these fragments show relatively planar faces, sometimes polished, which keep the striations of glacial abrasion – parallel grooves in the rock caused by moving against other rocks, acquired when these fragments had been at the base of a glacier in movement. There have been well preserved diamictic glaciers in stone of 40 meters in height and 70 meters of extension in the municipality of Mirassol d’Oeste, in the south east of the state of Mato Grosso. The USP geologists, in collaboration with groups led by Alcides Sial, from Pernambuco, and Candido Moura, from Para, have examined the carbon and strontium isotopes of the samples collected and have confirmed: here there were treasures of more than 600 million years.

Magnetic records
With the records of the passage of ice through the region along the surface, the researchers undertook to take the map designed by Hoffman out of the drawer. They did not hesitate in bringing up a doubt: was the Amazon really positioned at the South Pole? The conclusive reply emerged from the use of a technique that had been used in the reconstruction of primitive Earth, paleomagnetism. “When the rocks were forming, the earth’s magnetic field gets recorded on them”, explains Ricardo Trindade, from the Astronomy, Geophysics and Atmospherics Institute of USP. “In the rock from Mirassol d’Oeste, at each 1 or 2 meters, we made use of a special drill to collect five or six rock samples”, points out Eric Font, who then was beginning his doctorate degree under the supervision of Trindade. In a magnetic proof room, the samples – small cylinders of 1-inch diameter by 2 centimeters in height – passed through successive heating in order to eliminate the most recent records of magnetism. The information concerning the original magnetism is then compared with the terrestrial magnetic field, which functions as an enormous magnet buried in the center of the planet “Thus we could discover the block’s latitude and the position in which it had found itself in relation to the Earth’s axis, back at that remote period”, says Trindade.

With the new evidence and the agreement of Hoffman himself, who visited Mirassol d’Oeste in September of 2003, the world-map of primitive Earth is being redesigned. Assuming a new position, related at the end of 2003 in the magazine Terra Nova, the Amazon jumps from the extreme south of the globe to the tropical region. To the north of Laurentia, the block of rock that currently forms North America. To the east one can find two other cratons: the Congo-Sao Francisco, corresponding to the western part of Africa and the Brazilian Northeast; the other is the craton of Rio da Prata, which today forms the Southern region of Brazil and Uruguay. The mosaic completes itself with the cratons of West Africa, Antarctica and India. “As all of the blocks had been very close together, and it would be difficult to imagine that glaciation was a local phenomenon, restricted to the Amazon”, suggests Nogueira. “The most logical situation is that it was in fact a global occurrence.”

Other studies have reinforced this idea. Diamictic glaciers similar to that of Mirassol d’Oeste have been found during the decade of the 1990’s in Canada and Namibia. In May of this year, a team from Oxford University, United Kingdom, described in the magazine Geology rocky deposits with similar characteristics in Oman, in the Middle East, for which the paleomagnetic data also indicates sedimentation in low latitude. Added to other pieces of the puzzle, the Brazilians are helping to consolidate the scenario of the snowball, presented in 1992 by the North American geophysicist Joseph Kirschvink. His idea was born from the unexpected discovery of glacial rocks in the south of Australia, also formed in low latitudes.

Kirschvink then risked a daring suggestion, that little by little is being confirmed: just over 600 million years ago, because of the colossal concentration of continental mass in a single continent, Gondwana, there would be a greater capacity for the Earth to reflect sunlight. As the surface absorbed less heat, the planet’s temperature fell briskly. Another cause of this cooling would have been the decomposition of silicates, the most abundant mineral in the Earth’s crust, in a process that absorbed atmospheric carbon dioxide, one of the gases that is contributing to the greenhouse effect and presently helping to warm up the Earth. These premises had led geophysicist Kirschvink to look upon the planet as a gigantic ball of snow – a snowball.

There is relative consensus of opinion about the existence of glaciation, but there remains a doubt about the length of time the Earth remained covered with snow. Studies carried out by Brazilians in the Amazon show that in only 100,000 years, a geologically short time period, the temperature of the planet’s surface could have passed from 50°C negative to almost 60°C positive. The conclusion comes from the analysis of dolomites, calcium and magnesium carbonate rocks, which covered the diamictic glacier of the Mirassol d’Oeste. Formed in warm sea waters, the dolomites here are as old as the diamictics – this is a sign that the ice that had covered the Earth had already been totally melted when they came about. In another article, published in Geology, the Brazilian team revealed an important detail of these rocks: both the diamictic glaciers and the dolomites have presented deformations that had occurred when they were still soft, not consolidated. “These deformations provide evidence that the changes from a glacial to warm climate were very rapid”, says Riccomini.

Always alive
Even with the ice covering almost all of the terrestrial surface, Thomas Fairchild, a researcher with USP’s Geosciences Institute, and co-author of this work, does not believe that life on the planet earth had disappeared. “It’s difficult to imagine that the evolution of the animals, which had already begun, had been interrupted for later to be taken up again”, he comments. For him, the process that generated the snowball, in spite of being intense, would not have been so radical. Pockets of seas must have resisted, principally in the regions of the equator, with nutrients for their first multicellular beings – microscopic, still with poorly defined outer shape, the parents of current corals and sponges. Only much later, when the temperatures began to rise, would life spread through the continents and new strains of plants and animals would spring up. “The absence of fossils makes the work difficult”, suggests researcher Fairchild, “but we’re learning a lot about remote life on the planet by parallel means”.

In the same stone of Mirassol d’Oeste a type of primitive crude oil has been found, possibly the oldest in Brazil. It is a black bitumen, viscous like honey and with a vitreous appearance. It represents one of the last stages of the maturing of organic material, found in the pores, pockets and fractures in the carbonate rock. According to Nogueira, it is another clear sign of terrestrial life immediately after glaciation, since the hydrocarbons form starting from the decomposition of living beings. This material is being studied in cooperation with researchers from the University of Nancy, in France, in search of information concerning the organisms that formed it. “Probably we’re dealing with primitive bacteria”, believes Nogueira. As it is not a huge bed, this crude oil has no economic importance. Its value is purely scientific.

The Project
Sequence stratigraphy of the High Paraguay group, Neoproterozoic Cambrian of the Paraguay strip, Mato Grosso (nº 00/02903-8); Modality Regular Line of Research Assistance; Coordinator
Claudio Riccomini – IG/USP; Investment R$ 54,392.98 (FAPESP)