The restless earth

A discovery reinforces the idea that deep convection currents move the continents

After studying for more than a decade the movements of the tectonic plates, those immense blocks of rock which form the surface of the earth, the physicist Marcelo Sousa de Assumpção came to an important conclusion. It had already been known that the continents, the visible parts of the tectonic plates, didn’t drift like a small wooden fishing raft without sails. The energy which moves them is certainly the heat contained within the interior of the planet. However, the tectonic plates are going towards or away from each other through convection currents, as in a pan of water being boiled and producing movement on the surface. A professor of the Astronomy and Geophysics Institute (IAG) of the University of São Paulo (USP), Assumpção coordinated a team which discovered that this movement may be much deeper than was at first thought.

According to him, all of the upper mantle, the layer below the crust, moves with the surface plates, at a depth which, at least here in Brazil, can reach to 700 kilometers – not just the 100 or 200 kilometers as it was previously accepted. This movement, now of amplified limits, involves just as much the upper mantle next to the crust, as the lower mantle, much deeper. “However, we can discard one of the previous hypothesis previously accepted which suggested that the convection movement was entirely confined to the upper mantle”, comments Assumpção, who is part of an international group of researchers in this area, also including specialists from the Carnegie Institution of Washington, United States and of the University of Montpellier in France.

This discovery will contribute to the refinement of the theory of the tectonic plates formulated at the beginning of the century by the German geophysicist Alfred Lothar Wegener (1880-1930). Until Wegener published his study Origins of the Continents and Oceans in 1915, it had been taken as certain that the surface of the earth was immovable – and there was no way of explaining in a reasonable and scientific manner, for example, earthquakes, today looked upon as a consequence of the meeting of plates. Killed during the last of three expeditions to Greenland, the German geophysicist proposed that in the geological past there was a unique continent, baptized as Pangaea, surrounded by an ocean called Panthalassa. The breaking up of Pangaea produced smaller blocks and one of them, Gondwana, as it broke up led to the origin of Africa, South America and Antarctica.

The starting point for Wegener to establish his theory of tectonic plates was the similarity between the coasts of Brazil and West Africa, which he saw as pieces which fitted together as in a giant jigsaw puzzle. Wegener thought that the continents drifted aimlessly, floating like enormous rocky boats on a layer of fluid rock which would be the mantle. He substantiated the idea of the drifting of the continents even if the mantle was solid and not fluid (the rocks of the mantle merely behave in a viscous manner when they are viewed on a scale over millions of years). Since that time, researchers from all over the world have mapped the trajectory of various plates, but only during the last few decades has it come about that the convection currents in the interior of the earth, linked directly to the movement of the plates, have begun to be detected by way of seismic tomography, a technique which maps the structures of the interior of the planet through the waves generated by earthquakes.

The team from USP investigated an area which forms a rectangle of 1,700 kilometers of length by 1,000 kilometers of breadth and has a depth of 1,400 kilometers in the Paraná river basin and surrounding area (see map on page 24). The first phase of the work, restricted to a rectangle of 800 kilometers by 400 kilometers, finished in 1995. It was successful enough to the point of justifying the present phase which began in July of 1997 and should be finished in July of this year.

In this project, Structure of the Crust and Upper Mantle in the Southeast of Brazil, with financing of R$173,400 by FAPESP, Assumpção carried out the surveying of the field and the interpretation of the data with the participation of the German geophysicist Martin Schimmel, who took his post doctorate at USP. In Brazil, the data collected during the project has been used as well in other pieces of research by the University of Brasilia, by the Technological Research Institute (IPT) of São Paulo, and by the National Observatory of Rio de Janeiro.

In the beginning, the researchers from USP and Carnegie, most directly involved with this work, used as sources of data the shock waves liberated by earthquakes, the P waves (longitudinal waves which are the first to reach the surface) and S waves (transversal waves) from different regions of the planet. Published in the magazine Nature on the 4th of November 1995, this preliminary research revealed, between the municipalities of São José do Rio Preto, Riberão Preto and Franca, in the interior of São Paulo, the existence of a structure interpreted as a very old volcanic channel in the mantle caused by a plume.

Deduced from studies on the computer and in a reduced laboratory scale, the plumes are columns of hot rock which originate at a profound depth in the mantle, and rise to the surface and cause extensive volcanic activity. In the last few years the plumes are finally being detected, such as that of the island of Iceland in the North Atlantic. According to Assumpção, this matter coming from the interior of the earth can perforate the crust with enormous projectiles and had an important role in the process of the rupture of the supercontinent Gondwana.

A piece of the fossil plume – however no longer active – of the interior of the state of São Paulo is situated at a depth of between 200 and 700 kilometers and has a width of approximately 300 kilometers. The temperature of this structure, according to the data obtained, is in the region of 1,700 degrees Celsius, at 200 degrees hotter than the region of its surroundings. According to Assumpção, this structure was at one point in the middle of the Atlantic ocean associated with the archipelago of Tristan da Cunha, a territory of the United Kingdom, of volcanic origin in the middle of the South Atlantic.

Afterwards, the rupture of Gondwana, which originated the present territories of Africa and South America, gave birth to the Atlantic Ocean around 130 million years ago. Evidently the tectonic plates have not quietened down. Even today, on the sea floor, where the crust is least thick, the movement of the plates produces fractures through which springs hot material from the mantle. This is the mechanism responsible for the Mid-Atlantic Ridge, of which the archipelago Tristão da Cunha is part. We are speaking of an area in which the plumes are still active – the so-called hot spots.

The data obtained in the Paraná Basin indicate that a piece of the plume, a column of hot material of the mantle, diverged to the west and, in this manner, accompanied the movement of the South American plate, it being today inactive and consequently a fossil. In the evaluation of the researchers, merely this discovery already brings an important contribution to the study of the dynamics of plates. Brazil is right in the middle of the South American plate which is moving west at a speed of 1.5 centimeters per year. Since the African plate is itself moving to the east at the same speed, the two continents are distancing themselves at a total velocity of three centimeters per year.

Moving towards the west, as a consequence of the movement of the mantle, the plume of the interior of the State of São Paulo is keeping itself under the Paraná basin until today. A little before the rupture of Gondwana, it was responsible for spilling of volcanic rock (basalt) which covered the surface for more than a thousand kilometers, starting from the point of origin. In this manner the researchers explain a geological characteristic of the Paraná Basin, the outcrops of basalt. Basalt, a volcanic rock of dark color, transformed itself into soil over millions of years and gave origin to the red and fertile soils of Paraná and part of the interior of São Paulo, which helped to maintain one of the most agriculturally productive areas of the country.

In the most recent study, Assumpção and Schimmel demarcate in a more exact way the column of hot material, the piece of the fossil plume, by way of some 38 seismographic stations distributed along the area of study, which extends from Brasilia until part of the state of Paraná. For this reason they had to travel close to 20,000 kilometers during the three years which the study has already taken. Every two months the researchers or the technicians traveled to the stations to collect the data which were stored by the computerized seismographs. Installed in small cabins normally on a farm, the sensors registered the soil vibrations, detecting on a daily basis, earthquakes which occurred in the whole world. At times, they also recorded the small low intensity earth tremors which take here in Brazil.

The speed which the shock waves of earthquakes reach the stations depends on the temperature of the rocks through which they pass. Thus, the researchers trace a profile of the lithosphere (the crust and a small rigid part of the upper mantle) and of the mantle up to a depth of 1,400 kilometers. Images made by computer also permit the visualization of the plume, contained within the mantle (see above).
One of the consequences of the plates dynamics is mountain formation as it has happened in the west of South America where we have the Andes.

This mountain chain, which runs from Chile up to Colombia, originated from the shock of the South American plate with another tectonic plate, that of Nazca, located under the Pacific. The shock ended up with the heavier Nazca plate diving under the lighter South American plate, drifting in the opposite direction. The shock of the plates amasses and thickens the border of the lighter plate. In the case of the South America and Nazca plates, the thickening of the western border of the South American plate resulted in the structure of the Andes, a still active process, and responsible for the frequent earthquakes and volcanoes of the Andean countries such as Ecuador and Colombia.

The technical advance of the detectors throughout the nineties permitted a qualitative leap in the gathering of data and the refining of models of the tectonic plates proposed by Wegener. With investigations deeper and deeper into the mantle it was possible to learn, for example, that in the processes of subduction – as are called the diving of one plate under another – the destruction of the plate which is diving only occurs at great depths, at times more than 1,000 or 2,000 kilometers.

“We have superseded the previous model which supposed that the diving plate dissolved itself away at a depth of only 700 kilometers”, comments Schimmel “Instead of this, on diving it remains intact without dissolving itself away.” Though constituted from the same material of the mantle, the plate is colder, more rigid and heavier – consequently it goes down more easily. The new technological resources allow us also to know that a plate can change its inclination, and from an almost vertical dive, can change to a horizontal movement. “What we know today, for certain, is that the deep earthquakes which occur up to 700 kilometers below the South American plate, are linked to the diving of the Nazca plate”, explains Assumpção “And it is also probable that the plate will continue diving to greater depths.” The tomography indicates a block of colder rock at approximately 1,300 kilometers below the region of the Southeast of Brazil in a range which extends from Brasilia to Curitiba, which seems to be a piece of the Nazca plate.

The discovery of a plume under the Paraná Basin was surprising. The researchers intend to study the region in order to make a comparison with other areas considered geologically to be very old, such as some regions in Canada. The interest was to discover how the earth’s lithosphere was formed in these areas known as cratons, a term which defines the geological provinces which have suffered little or no deformation since  pre-Cambrian time, more or less 600 million years ago. Another objective of the work will be to determine the thickness of the crust and of the lithosphere plate in the Southeast, until now unknown, and to investigate the possibility that they might contain blocks just as old as those of the São Francisco craton in the basin of the São Francisco river in Minas Gerais.

The São Francisco craton, dating from around 3 billion years, is one of the oldest structures on the face of the planet. Now, the work of USP has defined the thickness of the craton which extends from the surface to a depth of around 300 kilometers. It also revealed the existence of a fossil plume giving evidence to the drag of the crust by the upper mantle. For Assumpção it is too early to evaluate if the discovery of the fossil plume might help in things as to forecast earthquakes, especially on the borders of the clashing of plates.

The immediate contribution to be seen, is to show that the upper mantle moves with the crust and in this way we will understand the forces to which a plate is submitted even in its internal part. In these areas also frequent earthquakes may occur. “One reason for earthquakes could be the concentration of tensions due to the variation of the thickness of the lithosphere”, he explains. This is what could be happening in a region with constant earth tremors which cut across the State of Goiás in the southwest to northeast direction. The region presents behavior similar to the central region of the United States where there has been satisfactorily established a relationship between earthquakes and a lower thickness of the lithosphere.

The USP researchers believe that the discovery of the structure of the plumes, associated with the understanding of great depths of diving of the plates, can lead to a greater interest in the investigation of the region of contact between the base of the mantle and the fluid core of the earth (the liquid sphere at the center of the earth has, in its interior a solid core). This region is layer D ’’ (read as D two lines) situated at approximately 2.700 kilometers from the surface. It is an area of physical and chemical reactions, very complex, between the fluid core and the pasty mantle whose structure may remind one of the irregular teeth of a spherical saw.

The geophysicists estimate that in this region the basic mechanism for the convection currents are located, the same type of force which makes water, which is boiling, circulate from bottom to top in the pan. “All that goes down has to come up”, argues Schimmel to establish the relationship between the diving of plates formed from cold material and the ascending currents of material at elevated temperatures. According to the theory in place, the heat from the core of the earth, a result of enormous gravitational energy which gave birth to the plates some 4.6 billion years ago along with the important addition contribution from natural radiation. When the earth was formed, the heat was so much that it melted everything. Iron, heavier, went into the center where it is until today. It could be that over the next few years that these consecrated ideas will be revised, from what we now know with respect to the movements of the tectonic plates and of the convection currents of the mantle.

Marcelo Sousa de Assumpção is 49 years of age. He graduated in physics from the Physics Institute of the University of São Paulo (USP) and is a doctor in Geophysics from the University of Edinburgh in Scotland. He has been a professor of the Astronomy and Geophysics Institute since 1974.

The Project
The Structure of the Crust and Upper Mantle in the South East of Brazil (nº 97/03640-6); Investment: R$ 173.478,14