ÐÓ°ÉÔ­´´

Few creatures live on Antarctica; ice makes most of the continent inaccessible, and the climate of the rest is inhospitable at best. But the snow is a bonus for geologists; the ice cover, present for much of the past 40 million years, has inhibited erosion that would otherwise have worn down the Transantarctic mountain range and its hinterland. Because the bulk of the rocks are still there, these rocks preserve far more evidence of their history than is usually available in mountain chains elsewhere because erosion modifies the surface as fast as it develops.

The Transantarctic Mountains are unusual in that they mark a tectonic boundary where the Earth’s crust is rifting apart, rather than moving together. This and the unusually good preservation of the shape of the mountains makes Antarctica an ideal place to investigate the processes that have built this range. During the past Antarctic summer an expedition set off to run a seismic line across the ice in order to understand why the Transantarctic mountains are there. Although the results will need many months of processing before we know exactly what we have found, the expedition is already a success as a pilot project for future geophysical forays into the Antarctic.

The expedition, under the name of Seismic Experiment Ross Ice Shelf, or SERIS, was a joint venture between researchers from the Geophysics Division, Department of Scientific and Industrial Research, based in Wellington, New Zealand and Stanford University in California. Our aim was to produce an image of the crustal structure across the Transantarctic Mountain front down to perhaps 40 kilometres, by reflecting seismic energy from layers deep below ground. Where the ice of the East Antarctic ice sheet has spilled over the Transantarctic Mountains into the Ross Embayment, outlet glaciers have evolved as large ‘highways’ of ice. We took advantage of these natural highways and travelled part way up one, making seismic measurements on the way. The seismic profile followed a line down the Lowery and Robb Glaciers, just north of the Nimrod Glacier, at a latitude of about 82 degrees south, then continued for another 100 kilometres onto the Ross Ice Shelf. Apart from the expeditions of the explorers Scott and Shackleton at the beginning of this century, the area has seen few humans.

Today’s geological explorers have different challenges in their quest to know what shaped the Earth. The attraction of mountains for geophysical analysis is that they mark places where rocks have been distorted in a measurable way; they can reveal how the lithosphere, the strong elastic outer part of the Earth, about 100 kilometres thick, responds to the forces that built them.

The Earth’s lithosphere consists of a number of plates that move relative to each other, creating earthquakes and volcanoes where they meet. To understand the stress and strain in these boundary areas, geologists need to know some-thing about the mechanical properties of the plates. Engineers who need to understand the properties of beams or othermaterials take samples into a laboratory, load them and observe how they deform.

Without the luxury of a laboratory that can hold the planet, geologists must seek out places where the processes they want to understand are happening now or took place in the past. To discover the properties of the Earth’s plates, they observe their response to natural loads. The distribution and timing of uplift of a chain of mountains, now and through its history, is one such response. Understanding the form, magnitude and variations of crustal structure associated with mountain uplift helps us to learn about the processes that transfer heat and material through the Earth as a whole.

The Transantarctic Mountains, 4500 metres high at their peak and stretching for nearly 3000 kilometres, form a striking boundary between the two continental plates that make up Antarctica. Other mountain chains of the world have similar dimensions, but the Transantarctic Mountains are different because they lie at a plate margin that is extending, making a ‘rift margin’. The world’s major mountain belts form in three different ways. First, where continental plates collide they push up the crust to form the impressively high Himalayan type of chain. Secondly, where one ocean plate slides beneath another, or beneath a continental plate at a subduction zone, mountains such as the Andes result. Fluids and molten rock generated within the subduction zone rise into the overriding plate, elevating it by a mixture of heat and compression. The third form of mountains, which include the Transantarctic range, form in areas where plates move apart.

For ranges formed by compression, the forces that could build the mountains are easy to imagine – collision between continents either crumples or overlaps their edges, and the thicker crust rides higher on the mantle below. But it is much more difficult to understand what shaped rift shoulder mountains, formed where the crust has stretched and thinned.

One guess is that heat is at least partly responsible for the uplift, because as the crust becomes thinner, deeper, warmer rock rises closer to the surface but the mechanism for such thermal uplift is not clear. Another force that helps to push up rift margins was first proposed by the Dutch geophysicist Venning Meinesz in his analysis of the East African rift system nearly 60 years ago. He realised that changes in the elevation of the Earth’s surface should lead to unequal forces underground; isostasy is the idea that these variations are compensated by changes in the distribution of rock below ground. For example, the ice that covered Scandinavia in the last ice age weighed down the crust; once the ice melted, the surface rose and is still rising today as ductile rock at depth redistributes itself. The rate at which the crust rose and is still rising gives an estimate of the viscosity of the mantle beneath.

At rift margins, isostatic uplift is a consequence of an inclined fault in the lithosphere, as it floats on the fluid rocks of the asthenosphere beneath. A block of wood floating in a tank of water is a fair analogy. The wood represents the lithosphere of a continent, and the water the underlying, fluid asthenosphere. The top and base of the block are horizontal. If the block is sawn through at an angle and the two pieces left to float independently, the top block tilts towards the cut and the other tilts away. This is a response to a change in the centre of gravity of each block.

As a model for a rift, the break in the block represents a fault, and the blocks of wood are crude representations of the lithosphere. In fact, the blocks are a model of a model, representing the lithosphere as an elastic sheet extending infinitely in both directions away from the fault. The tilting or rotation of the blocks is transformed into a broad uplift and depression on each side of the break, known as lithospheric flexure. This is where modern understanding of the mechanics of the Earth differs from Meinesz’s ideas of isostasy, for the elasticity of the lithosphere provides an extra supporting force. The contrast is like that between a block of wood floating in water and one sitting on an elastic membrane. The membrane bends down at each side of the block, holding it up, but also spreading the effects of the weight of the block over a wider area. The wavelength of this bending providesa measure of the flexural strength of the lithosphere. Lithosphere flexure controls where basins form around mountain chains, and we think that it also contributes to the uplift of the Transantarctic Mountains.

If it does, we have a way of finding the strength of the lithosphere beneath Antarctica. But an accurate assessment demands a precise picture of the gently warped crust that extends for hundreds of kilometres behind the mountains and beneath the East Antarctic ice cap. This is what the Transantarctic Mountains and their associated hinterland provide; compared to mountains in middle latitudes, they have been preserved from strong erosion and sedimentation, which distort the picture. Around the East African rift, for example, huge thicknesses of sediment have built up in the hinterland to the mountains to form the Kalahari basin. This distorts the picture of changing crustal properties that studies of lithospheric flexure can reveal. Through time, erosion of the mountains and deposition of sediments in the hinterland basin has continually redistributed mass across the section. But in Antarctica, the ice has preserved the topography.

We can draw up a cross section of the surface and the structure of the crust across the boundary between East and West Antarctica. Echo sounding with radio waves has revealed the shape of the surface beneath the East Antarctic icecap, but other elements of the cross section are less accurate, particularly the deep crustal structure around the boundary. But other geophysical observations, such as the variations in the acceleration due to gravity, and seismic work in the Ross Embayment, show that the crust becomes between 15 and 20 kilometres thicker from West to East Antarctica.

East of the boundary the lithosphere curls up like a piece of warped plywood; this is balanced by the West Antarctic side sagging down to form a large sedimentary basin next to the Transantarctic Mountains. Other rift structures show similar differential flexure, and the simple model using floating blocks of wood predicts just this asymmetry. But the upliftand downwarp seen elsewhere and predicted from experiments are usually on the order of about 1000 metres up and down, not the 5000 metres relief we found at the front of the Transantarctic Mountains.

This area also differs from the ideal case in the width of the warping. The topographic profile alone shows the vast difference in the wavelength of flexure each side of the boundary; in the simple model, the uplift and depression should stretch across the same width on each side. But on the east side of the Transantarctic Mountains the wavelength of the bulge is more than twice that in the west, in the Ross Sea.

This indicates a difference in the strength of the lithosphere on each side of the rift, perhaps arising from variations in thickness, composition, or heat. Returning to an engineering analogy, if you heat a stiff metal bar with a blow torch, the bar will become weaker and easier to bend. In particular, when loaded the bar will bend over a much shorter length – it has a greatly reduced flexural wavelength. The Earth’s lithosphere appears to behave in the same way. Where it is hotter, we see small flexural wavelengths. So the West Antarctic plate, with a narrow bulge, appears relatively hot. In contrast, the East Antarctic plate has a large flexural wavelength. It appears to be very stiff, and by implication, cold.

Why might the Earth be hotter in West Antarctica than in East Antarctica? Where the lithosphere has stretched, the hotter asthenosphere beneath will be closer to the surface than elsewhere. The lithosphere then heats up more, softens and can bend with a shorter wavelength. Geological observations and reconstructions tell us that the lithosphere of the West Antarctic has been stretched and thinned in the last 100 million years, making it hotter than East Antarctica, one of the world’s oldest and most stable continents. Where the two plates are juxtaposed, there is a zone where heat from the upper mantle beneath West Antarctica conducts into the upper mantle of East Antarctica. This process of lateral heat transfer takes millions of years. One result is that parts of the crust and mantle on the east side of the boundary will expand as they warm. As they become less dense, the Transantarctic Mountains will float higher in the asthenosphere, giving the extra uplift we noticed, over and above the isostatic uplift arising solely from the rifting of the continental plates.

We will have to wait to see the results of our endeavors in the Antarctic; deep seismic profiles take a lot of computing time before they produce the cross section that we want to see. But we hope for surprises, too. Although this project is directed towards understanding mountain uplift at a major rift margin, other deep seismic exploration experiments have shown that we may discover something unexpected (see ‘Slicing through a continent’, New ÐÓ°ÉÔ­´´, 18 May). This is the essence of the current excitement about this type of exploration among geophysicists worldwide.

For example, will there be something about the long-lived presence of a major icecap that changes how the crust reflects seismic energy? And how is it that the lithosphere can deform so impressively in Antarctica without producing significant earthquakes? What, if any, is the generic link between icecap growth and stability and mountain uplift? These are just three questions that our results may address. Answers to any one of these could change the way geophysicists interpret data from similar experiments around the world.

Tim Stern works in the Geology and Geophysics division of the Department of Scientific and Industrial Research in Wellington. Uri ten Brink is in the Department of Geophysics at Stanford University and has recently moved to the US Geological Survey at Woods Hole.

* * *

A long stretch: seismic surveys across the ice

Antarctica presents both advantages and challenges for acquiring seismic data. The main advantage is that the ice provides a regular and evenly graded surface on which to work; motorised toboggans made the process of laying out and retrieving the cables and geophones (the instruments that pick up the returning seismic signals) quick and efficient. We could also make the 20-metre-deep holes for our small dynamite charges extremely quickly with a ‘drill’ that uses hot water. The charges were 200 metres apart along the 134 kilometres of the profile, so speed mattered.

But we also had to consider the damage our experiments could do to the life of the Antarctic. Each seismic charge produces a discernible ‘thud’ at the surface, and a fountain of ice and snow. Although this is not large enough to make the surface ice and snow crack or collapse, the explosions could disrupt animals, so we were careful to site our survey in an uninhabited region.

A technical problem of seismic work in the Antarctic is that it is so cold; the electronics needed to collect seismic information work far better at temperatures above zero than at -30 °C. We overcame this problem by operating out of a heated caravan mounted on a sledge. Explosives, too, have to perform at low temperatures. This meant that we had to use ammonium-nitrate explosives in the main, rather than a cheaper and safer alternative that does not detonate well at less than -25 °C.

Probably the biggest obstacle to work in Antarctica is the isolation. Typically we would operate with a team of between 10 and 14 people and have two or three heavy pulling vehicles and a number of smaller light-weight skidoos. Keeping all the machinery operating and supplying the team with food and fuel requires careful planning, for if we ran out of a type of food or needed a part for our seismograph, we might have to wait weeks or even months for assistance. In an attempt to avoid the sort of delay that could halt work, we carried extensive backup equipment.

The weather was another factor. The glaciers we used to cross the mountains were sheltered from the severe winds that sweep down from the polar plateau. Even so, when wind speeds exceeded 6 knots (3 metres per second), we found extra noise on the data collected with a particular type of equipment.

The results of this experiment are now being processed; it will take many months before we know exactly what we have found. But the experience we have gained with a variety of geophysical techniques will stand us in good stead for future seismic exploration of Antarctica.