¿ìè¶ÌÊÓƵ

‘CLUNTON and Clunbury, Clungunford and Clun, are the quietest places
under the Sun . . . ‘. A. E. Housman might have modified his rendering of
the traditional Shropshire jingle on the afternoon of 2 April this year.
People over most of England and Wales felt an earthquake of about magnitude
5.1 on the Richter scale, which had its focus about 14 kilometres below
Clun, a village south of the town of Bishop’s Castle. It caused minor damage
to masonry buildings as far away as Shrewsbury, 30 kilometres to the northeast.

The earthquake was a powerful reminder of two geological phenomena.
First, the Earth’s crust is under constant stress. It seeks continually
to adjust to these forces even in such a seemingly quiescent region as Britain.
Secondly, the old rocks which form Britain’s deep foundations are pervaded
by fractures waiting to give way. This is what happened in Shropshire. Although
tiny by the standards of California or Iran, these earthquakes highlight
the longevity of faults in Britain’s basement, the ancient rocks buried
beneath the cover of younger sediments. Faults that formed and slipped hundreds
of millions of years ago are still active today.

The Welsh Borders suffer frequent earthquakes by British standards.
Over the last century, there have been sizeable shocks near Shrewsbury in
1932, near Ludlow in 1926 and near Hereford in 1896 and 1924. But locating
these old quakes depends on assessing contemporary records of damage and
local people’s responses. The focus of a shock was rarely under the town
where it was most fully reported. The information is frustratingly imprecise,
often pinpointing quakes only to within a circle about 20 kilometres across.

Large earthquakes can now be located more accurately using seismometers
set up across the world. Researchers calculate the time the shock happened,
from the time the earthquake vibrations take to reach any three seismometers,
assuming that they know how fast the waves travel through the Earth (from
experiments and calculations). They also locate the epicentre, the point
on the ground surface above the quake’s focus. Information from many seismometer
stations combines to give more precise locations. The Bishop’s Castle earthquake
was located by Chris Browitt of the British Geological Survey’s Global Seismology
Unit to an accuracy of within 1 kilometre on the ground; its focus was between
13 and 16 kilometres deep.

The bigger the quake, the stronger the signal picked up and the more
precise the location. As recently as 1984, a tiny quake, magnitude 3.6,
was located about 10 kilometres west of the recent shock, but the position
was so inaccurate that it could just as well have happened in the same place
as the one in April.

Earthquakes happen when blocks of rock slip past each other on opposite
sides of a fault plane, a brittle fracture in the crust of the Earth. Because
old fault planes are usually weaker than the surrounding rock which has
not been faulted, they tend to break and slip whenever the crust is sufficiently
stressed . The location of the Bishop’s Castle tremor suggests that the
quakes tend to happen on faults that have existed for hundreds of millions
of years. Geologists have mapped out the pattern of faults with increasing
resolution over the years. Now they may be able to match the larger British
earthquakes with their parent faults.

The picture of basement faults in central Britain has emerged through
three techniques; mapping the rocks at the surface, seismic reflection surveys
and analysis of gravity and magnetic fields. Detailed geological mapping
of fault zones that reach the surface is effective only in the areas of
older rock that dominate the west of central Britain. In these places, such
as Devon and mid-Wales, the fault lines can be pinned down because they
have juxtaposed distinct types of rocks. In the east the old rocks and the
faults are covered by younger sedimentary rocks, and cannot be mapped on
the ground; geologists usually have to rely on the other two methods.

Seismic surveys monitor what happens to waves created artificially at
the surface, usually by an explosion, as they travel through the rocks below.
Layers of rock and faults underground reflect the vibrations back to the
surface, where seismologists record how long they take to arrive. Using
these methods, oil companies have detected gas fields as well as important
faults in eastern England and the Midlands.

Geophysics goes underground

But seismic surveys are costly; they need a lot of equipment and people,
and, if the results are commercially valuable, they tend to be confidential.
Perhaps the most attractive technique for locating faults in the basement
across an area such as Britain uses the variations in the strength of gravity
and the Earth’s magnetic field. Both these parameters change as the the
underlying rocks vary. The pull of gravity is greater over denser rocks.
The local magnetic field depends on the magnetic minerals in rocks. It is
more variable over igneous rocks, and some sedimentary rocks that contain
magnetic minerals. Major basement faults may juxtapose rocks of different
density and magnetic character, and therefore show up as marked linear features
on maps of the gravity or magnetic field. Conventionally, geologists have
picked out faults using contoured maps of gravity or magnetic field strength.
Recently, Michael Lee and colleagues at the British Geological Survey have
processed those data to produce colour and shaded-relief images, that reveal
the otherwise hidden fault pattern beneath south Britain. The structure
of the basement is complex, but three contrasting areas stand out; a central
triangular ‘Midland Platform’ and the rocks around it to the north and south.
The Midland Platform is a part of the Earth’s crust which has remained relatively
stable for 550 million years, since its last episode of strong deformation
and distortion. It is cut by faults that run from north to south in its
eastern part, and from northeast to southwest in the west. K K South of
the Midland Platform, there is a belt of faults that run from east to west.
These formed about 290 million years ago, when the Earth’s crust was crumpled
up to make mountains in southern Europe, a time known as the Variscan orogeny.
The same faults in southern England slipped again around 35 million years
ago, during the collision between continents that formed the Alps. The faults
on the northwest side of the Midland Platform triangle run roughly from
northeast to southwest. The Bishop’s Castle earthquake occurred directly
below the surface trace of one of this suite of faults, although on a segment
with a more northerly trend. The greatest slip on these faults happened
around 400 million years ago, during the building of much of the Caledonian
mountain chain, which stretches from the Appalachians in the US, through
Britain into Scandanavia. But many of these fault zones have an important
earlier history. Some, in the Welsh Borders, were active in late Ordovician
times, about 450 million years ago. Marks left on old fault planes by slip
and rocks ground down by abrasion and recrystallisation record the history
of movement of the fault. The slip can be dated using the decay of radioactive
minerals associated with the fault rocks, or by stratigraphic means. The
latter method brackets the age of movement by finding the oldest rocks cut
by the fault and the youngest rocks that are unaffected. Using these techniques,
Wes Gibbons, from the University of Wales in Cardiff, has shown that faults
near the Menai Strait – in the area of the last sizeable Welsh earthquake
in 1984 – moved about 550 million years ago. Many of these faults with a
northeast trend may have a similarly ancient origin, even though later slip
has obliterated the evidence for the older movements.

Most of the faults on the northeast side of the Midland Platform are
easterly or southeasterly in direction. These faults are mostly buried beneath
younger sedimentary rocks, that lie almost horizontally. However Tim Pharaoh
and colleagues in the Deep Geology Unit of the British Geological Survey
have analysed borehole data, gravitational and magnetic fields from this
area. They suggest a similar history to that of the northwest side of the
platform, dominated by Acadian deformation about 400 million years ago.
The buried faults have controlled the later Variscan structures of 290 million
years ago, which in turn controlled where coal formed in Yorkshire and the
East Midlands.

A picture now emerges of the rocks of central Britain involving a triangle
of crust that has remained stable for more than 550 million years. It is
surrounded by belts of rock that were deformed later, mainly 400 million
years ago to the north and about 290 million years ago to the south. But
all these areas have a history made up of several phases of deformation.
Later faults grow from older lines of weakness or follow the existing patterns,
so it is usually impossible to establish how and when any of the old fault
systems began.

This complex and ancient pattern of faults in the basement is the template
which controls earthquakes in Britain now and in the future. But why are
these faults moving at all, given that Britain is far from areas suffering
major crustal shortening or extension? The clue lies in the distribution
of stress in the rocks today, revealed by measurements of distortion in
deep boreholes. The crust in Britain is being compressed from northwest
to southeast, as is much of northwest Europe. This is probably due to the
horizontal push exerted by the spreading ridge in the middle of the Atlantic,
where ocean crust is currently forming. Although this stress is too low
to cause new fractures to develop, it is enough to reactivate existing faults.
So the old lines of weakness continue to slip.

What is still unclear is why some old faults slip more frequently than
others. Why do the faults along the Welsh borders on the west of the Midland
Platform generate more earthquakes than those further east? Is the important
factor their different orientation or the different age of the last major
deformation of their host crust? These questions remain to be answered.
British earthquakes may be small and infrequent, but each one will provide
vital information about these inherited faults in Britain’s basement.

* * *

OLD FAULTS SLIP MORE EASILY

IN the upper part of the Earth’s crust, rock is elastic; it can deform
in the same way as a rubber band. The relationship between the stress –
the pull you exert on the rubber band, and the resulting strain – how much
the rubber stretches, plots as a straight line on a graph until the band
snaps.

In the Earth, stress and strain energy will accumulate until the rock
reaches the limit of its brittle strength, and the rock fractures. The fault
that forms is weaker than the rock that surrounds it. With increasing slip,
irregularities along the surface of the fault are ground down, and the debris
that results has a lower frictional resistance than intact rock. In addition,
water passing through the fractured rock may break chemical bonds within
minerals in the rock and further weaken it.

As stress builds up again, further elastic strain accumulates. But the
energy that can be stored is now limited by the weakness of the fault zone.
Before the stress reaches a level that can fracture intact rock, the old
fault slips again at some lower stress determined by its reduced cohesion
and friction.

Nigel Woodcock is a lecturer in earth sciences at the University of
Cambridge and a fellow of Clare College.