ÐÓ°ÉÔ­´´

The atlas that you bought last year is already out of date. Dominating its maps of Asia is the Soviet Union, a country that no longer exists. Turn to the map of the Balkans and you will see the boundaries of Yugoslavia encompassing what are now various independent, warring states. The physical landscape, though somewhat more stable, changes too, as rivers alter course, the sea eats away at coastlines, and volcanic islands are born. Images obtained from satellites are continually supplementing land-based surveys to help the map makers keep track.

Last month the Geological Society published its Atlas of Palaeogeography and Lithofacies, which traces the palaeogeography of Britain, Ireland and their surrounding seas – the evolution of their geography over the past billion years. There is no political dimension to palaeogeography, and no satellite images to conjure up clearer pictures of coastlines that are millions of years old. So why is it that in this atlas, as in others, the picture it paints is so different from that of its predecessor, published in 1951?

Over the past four decades there has been a revolution in geology, with profound effects on all branches of the earth sciences. As recently as 30 years ago, the hypothesis of continental drift was little more than an interesting speculation for academic debate; now it is generally accepted. From continental drift has come plate tectonics, the idea that the Earth’s outer shell consists of discrete plates that are constantly on the move. Plate tectonics, in turn, has produced a unifying theory that explains many of the Earth’s natural phenomena that were largely mysteries a generation ago. Understanding the processes that change the pattern of continents and oceans has helped geologists to comprehend the evolution of palaeogeography. The growth of offshore exploration for oil and gas has also helped to reshape our understanding of Britain’s geological past. Geophysical surveying, and the many boreholes, have produced a wealth of data revealing many surprises about the geology of our offshore regions.

WHERE’S THE BEACH?

So, what exactly is palaeogeography, and how is a map drawn up to show Britain as it was at a particular time in the past? The simplest map of contemporary Britain shows just the coastline, dividing land from sea. More detailed maps can show estuaries, wetlands, lakes and mountains on land, and differentiate between shallow and deep seas. Palaeogeographical maps aim to do exactly the same for past phases in the Earth’s history, based on information preserved in the rocks themselves.

Virtually every exposure of rock can tell you something about the conditions under which it formed; a sedimentary rock, such as a sandstone or a mudstone, may well contain characteristic features, or structures, such as ripples or dunes, that show the direction from which the sediment was derived and whether it was laid down by wind or water. Fossils are especially valuable, because of what we know about where and how particular species lived. They can show whether a rock was deposited under water and, if so, whether it was sea water or fresh; they can indicate the depth of the water and, in some cases, even its temperature.

From the type of rock, or ‘litho-facies’, it is also possible to glean far more detailed information about the environment in which it was formed: a sandy shore, a muddy seabed and so on. This information is important to our understanding of the ancient geography of the area that now forms the British Isles.

THE CORAL SEAS OF DEVON

Investigation of the rocks of the Devonian period in southwestern England and Wales, formed about 380 million years ago, illustrates how to build up a palaeogeographical map. The cliffs of south Devon, around Torquay, are made up of a succession of limestones and mudstones. Closer examination shows that they contain fossils of many types of sea creature; in particular, corals, which show that the limestones formed in a shallow tropical sea.

The next stage is to find out how far this sea extended by finding more rock of the same age. At Ilfracombe, on the north coast of Devon, more Devonian limestones with corals occur, suggesting that conditions there were similar to those in the Torquay area. However, unlike the uniform southern outcrops, parts of the north Devon coast include layers of red sandstone that show no signs of originating under the sea. These non-marine rocks lie both above and below the marine limestones, some of which must therefore have been deposited earlier than the red layers, and some later. Continuing northwards, across the Bristol Channel into the Brecon Beacons of South Wales, reveals more Devonian rocks. Here they consist entirely of nonmarine red sandstones and mudstones, including the Ridgeway conglomerate, in a thick formation known as the Old Red Sandstone.

From this it can be deduced that the coastline of southwest England in the Devonian period must have lain to the south of the Brecon Beacons, because the succession of rocks there is nonmarine, and to the north of Torquay, where all the rocks are marine in origin. As the rocks in north Devon contain interleaved marine and nonmarine sediments, the position of the coastline must have shifted between mid-Devon and the Bristol Channel.

The sedimentary structures in the sandstones and mudstones of the Old Red Sandstone can also supply useful information about the terrain near the coast. The landscape consisted of alluvial plains crossed by rivers carrying sediment south to the sea. With a closer analysis of the lithofacies of older and younger rocks across this area, geologists can draw up a series of palaeogeographies for this corner of Britain, showing positions of the coastline at various times during the Devonian period. In terms of modern geography, the coastline ran roughly from east to west, but moved between 50 and 60 kilometres north and south in the course of about 40 million years.

But this analysis oversimplifies the story, because these rocks have been compressed since they were laid down, into complex folds. The original distance in Devonian times between rocks now found in Torquay and the Brecon Beacons was probably twice what it is now. Palaeogeographers cannot always take this sort of crustal shortening into account, but they should always be aware of it. The new atlas includes some maps that show the positions of the sediments at the time they were deposited.

So much for the basic method of producing these maps. What has happened in recent years to change geologists’ understanding of British palaeogeography? The first big change came with plate tectonics and the subsequent recognition of the distances involved in the movement of the Earth’s plates. For Britain, some of the clearest evidence comes from the Cambrian period, 550 million years ago, when animals first developed hard parts and became more readily fossilised. (Before this time fossils had no hard parts and are, consequently, hard to find.) Among the most typical Cambrian fossils are a group of segmented sea-dwellers, the trilobites. These creatures evolved into distinct forms over relatively short periods of time, so through their fossils, Cambrian rocks can be correlated precisely across the world. Such comparisons make it clear that Cambrian trilobites fall into a number of separate ‘faunal provinces’; in other words, some species and genera are restricted to various parts of the world.

British rocks contain trilobites from two faunal provinces. The early Cambrian trilobite species of England and Wales are not found in rocks of the same age in northwest Scotland; these have their own, quite different, species. The Scottish species are not unique, but also appear in the Appalachian Mountains of Pennsylvania and in northern Newfoundland.

THE OLD NORTH-SOUTH DIVIDE

This strange distribution suggests that there was some sort of barrier to the spread of trilobites between northern and southern Britain in Cambrian times. Forty years ago, the accepted wisdom of the time maintained that trilobites were unable to migrate from England and Wales to northwest Scotland, or vice versa, because there was an area of land separating the two seas. Palaeogeographical maps consequently showed a narrow area of land across southern Scotland.

Now that the notion of continental drift is generally accepted, geologists believe that the barrier to trilobite migration was instead a wide, deep ocean. Before the Atlantic Ocean opened in the Mesozoic Era, about 100 million years ago, North America and Britain were near neighbours. Looking further back in time, some 550 million years ago, there was another ocean in this area, called the Iapetus Ocean. Instead of separating America and Britain, as the Atlantic does, it divided America and northwest Scotland from southern Britain. The two groups of trilobites lived in the shallow seas around the edges of the Iapetus Ocean, which was in Cambrian times probably thousands of kilometres wide. They could not cross the deep ocean and so lived and evolved in isolation on opposing shores, giving rise to the two faunal provinces.

The ocean eventually closed as continental drift drew the far shores together at least 420 million years ago. Until then, northern and southern Britain were too far apart to even be shown on the same map, unless it is drawn to a very small scale. This very early north-south divide means that the new atlas cannot show the British Isles as a whole until after the Iapetus Ocean closed.

Closure of an ocean has major repercussions for palaeogeography. Not only does an ocean disappear from the maps, but mountains rise up, land emerges from the sea in one place and is drowned in another, and volcanoes can build new islands where none were before. One effect of the increasingly sophisticated understanding of plate tectonics of the past few decades has been the realisation that other types of tectonics can have equally wide-ranging consequences.

Thanks to its earthquakes, the San Andreas Fault is probably the best-known fault in the world. It is also the boundary between two of the Earth’s crustal plates. The Coast Ranges of California, which lie on the Pacific plate, are moving slowly but inexorably northwestwards relative to the North American plate. This sideways sliding, termed ‘strike-slip movement’, has left fragments of the Earth’s crust side by side that were thousands of kilometres apart tens of millions of years ago. Because they are on different plates, rocks on one side of the fault have different geological histories to rocks in neighbouring areas on the other. Their rock successions, disparate in the distant past, became similar only around the time when the two areas came together. This phenomenon has brought a new dimension to palaeogeography and, with it, its own terminology. The fragments of plates are called terranes, and when they slide into position next to each other like boats pulling alongside a quay, they are said to ‘dock’.

Over the past decade geologists have recognised that some of the major faults across Britain – especially Scotland – are strike-slip faults that mark the boundaries of terranes. The faults, among them the Great Glen Fault, cross Scotland from northeast to southwest. Adjoining slices of rock that are now part of Scotland were once far away from each other. These terranes, mapped out in the new atlas, docked to produce the foundations of modern Britain.

Having recognised these terranes, geologists have been able to begin solving some long-standing problems in the palaeogeography of Britain. For example, the sequence of rocks in the Grampians is radically different from that of the Midland Valley, which is now adjacent. Geologists previously had to resort to all sorts of complex explanations, but the simple answer is that they were once far apart, and slid together along the Highland Boundary Fault.

Unusual pebbles found in conglomerate rocks in the Southern Uplands of Scotland posed another problem that has been resolved in a similar way. The sedimentary structures in the conglomerates show clearly that the pebbles came from somewhere to the north of the Southern Uplands, but no one has found any rocks of the right type in the expected place. In this case the source remains a mystery, but the theory says that it has slid away along a strike-slip fault and presumably now lies buried under younger rocks, possibly hundreds of kilometres away.

Modern palaeogeographical maps of Scotland look very different from those of 40 years ago. For Lower Palaeozoic times, this rearrangement produces a series of totally new palaeogeographies for the whole of northern Britain. Keith Ingham and Brian Bluck of the University of Glasgow have been the brains behind this part of the atlas. They have selected the data that show where the terranes may have been at various points in the past, before they docked into roughly their present positions in the Devonian period. The maps that illustrate this in the atlas are a first attempt at a series of such reconstructions. Although these interpretations could well be proved wrong, they will stimulate debate on the pieces of the jigsaw puzzle that became the British Isles.

One of the major tools used to sort out how the Earth’s plates have moved is palaeomagnetism. Magnetic minerals in rocks hold traces of the magnetic field that prevailed when they formed. The orientation of the Earth’s magnetic field varies from pole to equator; in particular, its tilt varies with latitude. The magnetism remaining in a rock can be used to find the rock’s palaeolatitude – the latitude at which it formed. Comparing samples from opposite sides of terrane boundaries shows how far apart the rocks once were. The magnetic information about the north-south separation, combined with the orientation of the boundary strike-slip fault, indicates where these terranes once were. In this way, palaeomagnetism has confirmed some educated guesses, showing, for instance, that the swampy forests of Carboniferous times, responsible for Britain’s coal, were formed in equatorial regions 300 million years ago. There is scope for a great deal more palaeomagnetic work in Britain. It could help to resolve the history of the northern terranes.

THE LOST LAND OF WALES

If plate tectonics, terrane boundaries and continental drift have drastically altered ideas about the early geography of Britain, offshore exploration has had an equally major impact on interpretations of the more recent past. In the 1950s, the geology of the offshore areas hardly qualified as a subject for research. What little was known about Britain’s undersea rocks came almost entirely by extrapolation from the rocks to be found onshore. But increasingly sophisticated exploration techniques and investment by industry in the technology needed to apply them have brought surprises even on land.

One of the most significant upsets came from a borehole at Mochras Farm near Llanbedr on the shores of Cardigan Bay in North Wales. Geophysical exploration with seismic sections and measurements of gravity and magnetic characteristics of the area suggested that there were a thick series of sediments underground, forming the edge of a basin that extended into the bay. The general picture was thought to be like that in the Cheshire basin to the north. Geologists expected that the rocks would be Triassic mudstones around 230 million years old, like those of Cheshire, with the edge of the basin marking the western limit of an area of land known as the Welsh landmass.

The borehole disproved this idea. Beneath young gravelly conglomerates some 25 million years old, the drill cut through the thickest succession of marine Lower Jurassic rocks (between 170 and 200 million years old) known in northwest Europe. None of the sediments showed any sign of having been deposited close to land. The nearest marine Jurassic rocks are to be found over the Welsh border with Shropshire and Cheshire, around the village of Prees. All previous reconstructions had suggested that, apart from its southern coastal fringe, Wales remained land throughout the Jurassic Period. At a stroke, the findings from the Mochras borehole removed the Welsh landmass from the maps. The thickness of lower Jurassic rocks proved that the whole of Wales must have been covered by Jurassic seas.

Offshore boreholes, many drilled in the course of exploration for hydrocarbons, held similar surprises. One in the North Sea, off northeastern Scotland, revealed a totally unexpected 1000 metres of basaltic lavas in Middle Jurassic rocks. This was the first evidence of significant volcanic activity in Britain in Jurassic times. Older rocks in the North Sea had other surprises in store. One reshaped the Middle Devonian shoreline, defined in southwest England by the boundary between marine limestones and Old Red Sandstone. The coast was thought to continue more or less eastward across southern Britain; it divided Old Red Sandstone beneath the younger rocks of southeast England from the marine limestones, with fossil corals, found around Boulogne. Researchers assumed that this trend continued eastward into the European continent. But marine limestones with corals of mid-Devonian age were found in a borehole around the Argyll oilfield in the North Sea, at about the same latitude as St Andrews. The only palaeogeographical explanation is that the coastline swept round in a huge bay reaching some 600 kilometres to the north.

The major effect that these offshore boreholes have on palaeogeography is a perpetual headache to compilers of a palaeogeographical atlas. Peter Rawson of University College London, who was responsible for editing the younger Mesozoic and Cainozoic parts of the atlas (dealing with rocks younger than about 250 million years), has had his work cut out trying to keep up with new information released by exploration companies in order to ensure that the maps include the most recent data. But the position is more difficult for earlier periods, because much of the latest exploration is penetrating the more deeply buried Upper Palaeozoic rocks, particularly those from the Devonian and Carboniferous ages. It seems likely that discoveries of this sort will force amendments to even the latest palaeogeographical maps long before another 40 years pass.

John Cope is a reader in geology at the University of Wales in Cardiff, and coordinating editor of the Atlas of Palaeogeography and Lithofacies published by the Geological Society. Price £295.